Split (1)

train · ~9.11M rows (showing the first 93.5k)

identifier stringlengths 11 32	pdf_url stringlengths 17 4.62k ⌀	lang stringclasses 120 values	error stringclasses 1 value	title stringlengths 2 500 ⌀	source_name stringlengths 1 435 ⌀	publication_year float64 1.9k 2.02k	license stringclasses 3 values	word_count int64 0 1.64M	text stringlengths 1 9.75M
https://openalex.org/W2082211146	https://europepmc.org/articles/pmc4118362?pdf=render	English	null	The Theory and Fundamentals of Bioimpedance Analysis in Clinical Status Monitoring and Diagnosis of Diseases	Sensors	2,014	cc-by	15,857	Sensors 2014, 14, 10895-10928; doi:10.3390/s140610895 sensors ISSN 1424-8220 www.mdpi.com/journal/sensors Review The Theory and Fundamentals of Bioimpedance Analysis in Clinical Status Monitoring and Diagnosis of Diseases Sami F. Khalil 1,2,3, Mas S. Mohktar 1,2 and Fatimah Ibrahim 1,2,* 1 Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia; E-Mails: samifathi@siswa.um.edu.my (S.F.K.); mas_dayana@um.edu.my (M.S.M.) 2 Centre for Innovation in Medical Engineering (CIME), Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia 3 Department of Biomedical Engineering, College of Engineering, Sudan University of Science and Technology, 407, Khartoum, Sudan * Author to whom correspondence should be addressed; E-Mail: fatimah@um.edu.my; Tel.: +60-37-967-4486; Fax: +60-37-697-6878. Received: 8 April 2014; in revised form: 3 June 2014 / Accepted: 4 June 2014 / Published: 19 June 2014 OPEN ACCESS Sensors 2014, 14, 10895-10928; doi:10.3390/s140610895 sensors ISSN 1424-8220 www.mdpi.com/journal/sensors OPEN ACCESS sensors ISSN 1424-8220 www.mdpi.com/journal/sensors OPEN ACCESS 1. Introduction Bioimpedance analysis is a broadly applied approach used in body composition measurements and healthcare assessment systems. The essential fundamentals of bioimpedance measurement in the human body and a variety of methods are used to interpret the obtained information. In addition there is a wide spectrum of utilization of bioimpedance in healthcare facilities such as disease prognosis and monitoring of body vital status. Thus, with such a broad utilization, we feel that this warrants a review of the most fundamental aspects and healthcare applications of bioimpedance analysis. Studies on the electrical properties of biological tissues have been going on since the late 18th century [1]. Thomasset [2] explored the utilization of bioimpedance measurement in total body water estimation using needle electrodes. Nyboer [3] applied quad surface electrode readings for bioimpedance measurements to estimate the fat free mass of the human body. Hoffer [4] introduced the association between total body impedance and total body water content in reference to tritium dilution techniques. The electrical properties of biological tissues are currently categorized based on the source of the electricity, i.e., active and passive response. Active response (bioelectricity) occurs when biological tissue provokes electricity from ionic activities inside cells, as in electrocardiograph (ECG) signals from the heart and electroencephalograph (EEG) signals from the brain. Passive response occurs when biological tissues are simulated through an external electrical current source [5]. Bioimpedance or biological impedance is defined as the ability of biological tissue to impede electric current [6]. Due to the noninvasiveness, the low cost and the portability of bioimpedance analysis systems, numerous researchers have conducted studies on bioimpedance analysis and its applications in body composition estimation and evaluation of clinical conditions. Recently, Mialich et al. [7] reviewed the applications of bioimpedance analysis in body composition assessment and monitoring of chronic diseases with a comprehensive listing of the most used equations, however, recent techniques such as real time multi-sine bioimpedance analysis and bioimpedance vector analysis methods were not discussed. Lukaski [8] has revised the conceptual modules of bioimpedance analysis for physiological activities assessment and diseases prognosis. The study states that the applied multiple regression approaches and physical modules in bioimpedance analysis have limited utilization in individuals’ measurement. This paper is a review of the basic fundamentals and the applications of bioimpedance analysis. Sami F. Khalil 1,2,3, Mas S. Mohktar 1,2 and Fatimah Ibrahim 1,2,* Sami F. Khalil 1,2,3, Mas S. Mohktar 1,2 and Fatimah Ibrahim 1,2, 1 Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia; E-Mails: samifathi@siswa.um.edu.my (S.F.K.); mas_dayana@um.edu.my (M.S.M.) 1 Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia; E-Mails: samifathi@siswa.um.edu.my (S.F.K.); mas_dayana@um.edu.my (M.S.M.) 2 Centre for Innovation in Medical Engineering (CIME), Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia 3 Department of Biomedical Engineering, College of Engineering, Sudan University of Science and Technology, 407, Khartoum, Sudan * Author to whom correspondence should be addressed; E-Mail: fatimah@um.edu.my; Tel.: +60-37-967-4486; Fax: +60-37-697-6878. * Author to whom correspondence should be addressed; E-Mail: fatimah@um.edu.my; Tel.: +60-37-967-4486; Fax: +60-37-697-6878. Received: 8 April 2014; in revised form: 3 June 2014 / Accepted: 4 June 2014 / Published: 19 June 2014 Abstract: Bioimpedance analysis is a noninvasive, low cost and a commonly used approach for body composition measurements and assessment of clinical condition. There are a variety of methods applied for interpretation of measured bioimpedance data and a wide range of utilizations of bioimpedance in body composition estimation and evaluation of clinical status. This paper reviews the main concepts of bioimpedance measurement techniques including the frequency based, the allocation based, bioimpedance vector analysis and the real time bioimpedance analysis systems. Commonly used prediction equations for body composition assessment and influence of anthropometric measurements, gender, ethnic groups, postures, measurements protocols and electrode artifacts in estimated values are also discussed. In addition, this paper also contributes to the deliberations of bioimpedance analysis assessment of abnormal loss in lean body mass and unbalanced shift in body fluids and to the summary of diagnostic usage in different kinds of conditions such as cardiac, pulmonary, renal, and neural and infection diseases. Keywords: bioimpedance analysis; body composition; clinical status monitoring; diseases diagnostic and prediction Sensors 2014, 14 Sensors 2014, 14 10896 1. Introduction The first section highlights the main bioimpedance measurement approaches using single frequency, multiple frequencies and broadband frequency spectrum signals, in addition to applied bioimpedance measurements method across the whole body, through body segments and other alternative analysis method such as vector bioimpedance analysis and real time bioimpedance methods. Body composition parameters, which include lean mass and fluid volumes estimation using bioimpedance measurements, are discussed in the second section. Basic factors in bioimpedance measurements, including anthropometric measurements, age, race, protocols and postures, and shape and artifacts of electrode are discussed in the third section. Finally, applications of bioimpedance analysis in diseases prognosis and clinical monitoring systems are outlined in the fourth section. Sensors 2014, 14 Sensors 2014, 14 10897 2. Fundamentals of Bioimpedance Measurement Techniques Impedance (Z), from an electrical point of view, is the obstruction to the flow of an alternating current and, hence, is dependent on the frequency of the applied current, defined in impedance magnitude (\|Z\|) and phase angle (φ) as shown in Equations (1)–(3) [9]. Bioimpedance is a complex quantity composed of resistance (R) which is caused by total body water and reactance (Xc) that is caused by the capacitance of the cell membrane [5]: (1) (2) (3) (1) (2) (3) Resistance of an object is determined by a shape, that is described as length (L) and surface area (A), and material type, that is described by resistivity (ρ), as shown in Equation (4), [9]. Reactance (Xc) of an object as shown in Equation (5), is defined as resistance to voltage variation across the object and is inversely related with signal frequency (f) and capacitance (C) [9]. In biological systems resistance is caused by total water across the body, and reactance occurs due to the capacitance of the cell membrane [5,10]: (4) (5) (4) (5) Capacitance (C) is defined as the ability of the non-conducting object to save electrical charges, that is equal to the ratio between differentiation in voltage across object (dV/dt) and current that is passed through the object (I(t)), as shown in Equation (7). In the parallel capacitor module, capacitance is in direct proportion to the surface area (A) in meters square and inversely proportional to distance (d) in meters between the charged plates, and is dependent on the permittivity constant of vacuum (ε0 ≈ 8.854 × 10−12 F∙m−1) and the relative dielectric permittivity constant (εr) that is defined based on the material between the plates (for a vacuum space, εr = 1), as shown in Equation (6) [9]: (6) (7) (6) (7) Body composition estimation using bioimpedance measurements is based on determination of body volume (Vb) through the basic means of resistance measurement. 2. Fundamentals of Bioimpedance Measurement Techniques From Equation (4) that gives the relation between resistance and ratio of length (L) to surface area (A), body volume (Vb) can be obtained by substituting the surface area (A) with the numerator and denominator of the length (L), as in Equation (8): Sensors 2014, 14 10898 (8) (8) The human body as a volume is composed generally of fat mass (FM) which is considered as a non-conductor of electric charge and is equal to the difference between body weight (WtBody) and fat free mass (FFM), as shown in Equation (9); and FFM, which is considered as the conducting volume that helps the passing of electric current due to conductivity of electrolytes dissolved in body water. Studies show that water, known as total body water (TBW) is the major compound of FFM and is equal to 73.2% in normal hydration subjects, as in Equation (10) [11]: (9) (9) (10) (9) (10) (10) In bioimpedance measurements, the human body is divided into five inhomogeneous segments, two for upper limbs, two for lower limbs and one for the trunk. In the five compartment module, the human body is composed of FM and FFM which consists of bone minerals and body cell mass (BCM) that include protein and total body water that consists of extracellular fluid (ECF) and intracellular fluid (ICF) [5]. Figure 1, shows the five segments and compartments of human body. Figure 1. Main body segments and compartments. Figure 1. Main body segments and compartments. Most of the known prediction methods rely on the relation between water volume and the ratio between square length to resistance (L2/R) [12], however the alternation in anatomical and anthropometric features of the whole human body and segments cause variations in estimated volumes. Right Arm Left Arm Right Leg Left Leg Trunk Right Arm Right Leg Most of the known prediction methods rely on the relation between water volume and the ratio between square length to resistance (L2/R) [12], however the alternation in anatomical and anthropometric features of the whole human body and segments cause variations in estimated volumes. Sensors 2014, 14 10899 Jaffrin and Morel reviewed that most TBW estimation equations between 1985 and 1994 were based on values predicted using the H2/R50 that was introduced by Kyle et al. [13,14] and Houtkouper et al. [15]. Measurement of bioimpedance is obtained from the whole body and body segments separately, using single frequency, multiple frequencies and bioimpedance spectroscopy analysis. In addition to several alternative assessments method such as bioimpedance vector analysis and real time bioimpedance analysis. 2.1. Single Frequency Bioimpedance Analysis (SF-BIA) Analysis of bioimpedance information obtained at 50 KHz electric current is known as single-frequency bioimpedance analysis (SF-BIA). SF-BIA is the most used and is one of the earliest proposed methods for the estimation of body compartments, It is based on the inverse proportion between assessed impedance and TBW, that represents the conductive path of the electric current [5,16]. SF-BIA predicts the volume of TBW that is composed of fluctuating percentages of extra cellular fluid (ECF) which is almost equal to 75% of TBW, and ICF that represent the rest [5]. SF-BIA instruments have been used to assess TBW and FFM using the derived Equations (2) and (3), respectively, for normal hydrated subjects, although SF-BIA is not valid for body conditions with significantly altered hydration [17]. Studies by Hanai [18] on mixture theory report that body tissue conductivity is diverse [5], and SF-BIA shows limitations in ICF variance prediction, however many of studies show an acceptable correlation in ICF estimation [19]. Sensors 2014, 14 10900 Sensors 2014, 14 2.3. Bioimpedance Spectroscopy (BIS) Analysis of bioimpedance data obtained using a broad band of frequencies is known as bioimpedance spectroscopy (BIS). The BIS method is based on the determination of resistance at zero frequency (R0) and resistance at infinity frequency (Rinf) that is then used to predict ECF and TBW, respectively. The use of 100 and 1 kHz, respectively, was earlier proposed by Thomasset [25] who applied the basics of Hanai’s mixture theory [18] and Cole’s module [26,27] as explained by the Cole-Cole plot (Figure 2), however it is complicated to directly measure these values because of the relaxation phenomena of living tissue [20]. Figure 2. Cole-Cole module plot and Cole module parameters. Figure 2. Cole Cole module plot and Cole module parameters. Reference methods for estimating TBW are based on radioisotopic dilution of deuterium, and for ECF estimation they are based on the dilution of bromide [28] and for ICF they are based on the radioactive potassium isotope, 40K, both elements which are readily diffused in the human body [29,30]. Reference techniques are invasive, expensive and complicated when compared to bioimpedance methods, although the precision is dependent on the electrical module and body parameter variation [21]. Estimation of TBW, ECF and ICF using BIS techniques can be performed using either an equation modules approach [10,31–33] or an analytically derived equations approach [27]. Hanai’s mixture theory shows limitations in some studies [14,15,34], however it showed advantages in other studies [35,36]. Ward et al. [37] stated that the differences in biological construction among subjects may limit mixture theory as noted in some studies [38,39]. Scharfetter et al. [40] report that an accurate module for body fluid allocation and trusted fitting methods are most crucial factors in the BIS method. The determination of Cole module parameters (R0 Ri f α F ) in Figure 2 is done using the BIS Reference methods for estimating TBW are based on radioisotopic dilution of deuterium, and for ECF estimation they are based on the dilution of bromide [28] and for ICF they are based on the radioactive potassium isotope, 40K, both elements which are readily diffused in the human body [29,30]. Reference techniques are invasive, expensive and complicated when compared to bioimpedance methods, although the precision is dependent on the electrical module and body parameter variation [21]. 2.2. Multiple Frequency Bioimpedance Analysis (MF-BIA) Analysis of bioimpedance that is obtained at more than two frequencies is known as multiple-frequency bioimpedance analysis (MF-BIA). MF-BIA is based on the finding that the ECF and TBW can be assessed by exposing it to low and high frequency electric currents, respectively. Thomasset [2] has proposed TBW and ECF estimation using 100 and 1 kHz based on the Cole model [20]. However, in later years, Jaffrin et al. [21] stated that technically a bioimpedance analyzer should use frequency range between 5–1000 kHz. Simpson et al. [22] state that low frequency in MF-BIA is generally less than 20 KHz and high frequency is more than 50 KHz. Hannan et al. [23] report that parameters measured using a frequency of less than 5 KHz and more than 200 KHz fluctuate around the actual value and conclude that estimated TBW is more accurate using the MF-BIA than the BIS method with the same predicted values of ECF for both methods. Patel et al. [24] reported that in diseased subjects, TBW prediction using SF-BIA gave more precise results than MF-BIA. In general, the MF-BIA method predicts ECF more precisely than the SF-BIA method; however in elderly diseased subjects the MF-BIA method shows less sensitivity in detecting fluid shifts between ECF and ICF [19]. Sensors 2014, 14 Sensors 2014, 14 Sensors 2014, 14 10901 In Equation (4), the reference method is based on the assumption that the measured resistance (R) represents the total conducting volume of the lean body mass. However in the BIS method, the measured resistance represents the total conducting and non-conducting part of the lean body mass, so that the non-conducting part is included by multiplying the obtained resistance by body shape factor (Kb) and substituting the surface area (A) by body volume (Vb). Ayllon et al. [41] reports that the estimation of Cole module parameters (R0, Rinf, α, Fc) that is obtained by using only resistance achieves slightly better results and there is less standard error based on the Non-Linear Least Squares technique as compared to the capacitive and impedance complex components. Ward et al. [42] concludes that the Cole parameters can be obtained by using four selected frequencies and substituting a fitting technique based on amplitude impedance values at these frequencies: (11) (11) where, R is resistance, ρ is resistivity, Ht is the human height, Vb is the body volume and Kb is a dimensionless shape factor calculated from the length and perimeters of the upper and lower limbs, and the trunk, taken into consideration the body shape composed of the five cylinders.; Van Loan et al. [43] calculated the shape factor (Kb) from statistical anatomical measurements in adults to be equal to 4.3. 2.3. Bioimpedance Spectroscopy (BIS) Estimation of TBW, ECF and ICF using BIS techniques can be performed using either an equation modules approach [10,31–33] or an analytically derived equations approach [27]. Hanai’s mixture theory shows limitations in some studies [14,15,34], however it showed advantages in other studies [35,36]. Ward et al. [37] stated that the differences in biological construction among subjects may limit mixture theory as noted in some studies [38,39]. Scharfetter et al. [40] report that an accurate module for body fluid allocation and trusted fitting methods are most crucial factors in the BIS method. Reference methods for estimating TBW are based on radioisotopic dilution of deuterium, and for ECF estimation they are based on the dilution of bromide [28] and for ICF they are based on the radioactive potassium isotope, 40K, both elements which are readily diffused in the human body [29,30]. Reference techniques are invasive, expensive and complicated when compared to bioimpedance methods, although the precision is dependent on the electrical module and body parameter variation [21]. Reference methods for estimating TBW are based on radioisotopic dilution of deuterium, and for ECF estimation they are based on the dilution of bromide [28] and for ICF they are based on the radioactive potassium isotope, 40K, both elements which are readily diffused in the human body [29,30]. Reference techniques are invasive, expensive and complicated when compared to bioimpedance methods, although the precision is dependent on the electrical module and body parameter variation [21]. Estimation of TBW, ECF and ICF using BIS techniques can be performed using either an equation modules approach [10,31–33] or an analytically derived equations approach [27]. Hanai’s mixture theory shows limitations in some studies [14,15,34], however it showed advantages in other studies [35,36]. Ward et al. [37] stated that the differences in biological construction among subjects may limit mixture theory as noted in some studies [38,39]. Scharfetter et al. [40] report that an accurate module for body fluid allocation and trusted fitting methods are most crucial factors in the BIS method. The determination of Cole module parameters (R0, Rinf, α, Fc), in Figure 2 is done using the BIS method which is based on the argument that the human body is composed of a mixture containing conducting and non-conducting compartments [18]. 2.4. Whole Body Bioimpedance Measurement Measurement of total body bioimpedance is the most commonly used method for estimating whole body compartments. Many of the whole body bioimpedance instruments apply three approaches for impedance measurement: hand to foot method [14,17], foot to foot [44–46] method and hand to hand method [47,48]. The hand to foot (Figure 3a) one is the most commonly used method. It was introduced by Hoofer [4] and later revised by Nyboer [3] to decrease the contact impedance between skin and electrodes, and validated by Lukaski [17] in 140 normal adults. Tetrapolar hand to foot measurements are performed on a supine subject for 15 min, placing electrodes filled with gel to minimize gap impedance on the dorsal surfaces of the right hand and foot, distal (current) ones being respectively proximal to the metacarpal and metatarsal phalangeal joints, in accordance with standard tetrapolar electrode placement [49]. Foot to foot measurements (Figure 3b) were introduced by Nuñez et al. [50] through the use of a pressure-contact foot-pad electrode. In leg to leg bioimpedance measurements, the subject stands vertically, with uncovered feet, on four stainless steel footpads electrodes and divided for each foot into frontal and back portion for current injecting and voltage measurement [46]. Hand to hand bioimpedance measurements were introduced by Ghosh et al. [48] by performing body composition analyses using a handheld impedance meter in subjects with malnutrition. The device was held while both arms were stretched out horizontally in front of the body. Deurenberg et al. [47] validated the hand to hand method on 298 Singaporean subjects and reported that readings obtained using a handheld impedance meter were significantly acceptable for those subjects. Sensors 2014, 14 10902 Figure 3. Whole body bioimpedance measurement techniques, (a) hand to foot and (b) foot to foot electrodes positioning. 2.5. Body Segment Bioimpedance Measurement Current Electrodes Voltage Electrodes (a) (b) (b) Current Electrodes Voltage Electrodes 2.5. Body Segment Bioimpedance Measurement Segmental bioimpedance analysis achieves better estimation of skeletal muscle mass (SMM) than whole body bioimpedance analysis, with a reported standard error of 6.1% in reference to MRI measurements among 30 male subjects [51]. Baumgartner et al. [52] stated that multi-frequency segmental bioimpedance analysis enhances and elucidates the relationship between bioimpedance analysis and body compartment estimation after examining the impact of phase angle on body composition prediction among 116 normal subjects. 2.4. Whole Body Bioimpedance Measurement Segmental bioimpedance analysis detects the fluctuation in ECF due to differences in posture and is more precise than the ankle foot method [53], and gives a better estimation of TBW than total body measurements with reference to dilution method [54]. Segmental or perpendicular bioimpedance analysis defines the measurement method of body segments that is mostly treated as five cylinders as in Figure 1 [5], and was introduced to overcome the disagreement between trunk resistance to upper limbs ratio and trunk resistance to lower limbs ratio of 0.72 and 0.66 respectively [52]; Earthman et al. stated that the trunk represents 50% of the body mass [55]. Kyle et al. pointed out that total bioimpedance measurement assesses mainly the upper and lower limb compartments, and shows some limitation to predict water compartments of the trunk [13]. Measurement of segmental bioimpedance can be achieved through four types of protocols. The first approach, as suggested by Scheltinga et al. [56], uses dual current injection electrodes on the proximal 10903 Sensors 2014, 14 Sensors 2014, 14 bioimpedance method [44]. However, Kyle et al. [13] concluded that enhancement can be achieved through applying the MF-BIA method and further studies on electrode types and allocation. Sensors 2014, 14 Sensors 2014, 14 area of the right forearm and lower leg, and quad voltage electrodes placed on the right proximal forearm, shoulder, upper thigh and lower leg (Figure 4a). The second approach is suggested by Zhu et al. [57], through the sum of segments technique, that uses dual current injection electrodes on the right wrist and foot, and quad voltage electrodes placed on the right wrist, shoulder, upper iliac spine and foot (Figure 4b). A third approach was presented by Organ et al. [58], who suggested the use of dual current injection electrodes on the right wrist and foot, and quad voltage electrodes, two placed on the right wrist and foot, and two on the left wrist and foot (Figure 4c). The fourth approach as suggested by Jaffrin et al. [16,59,60], is through the use of quad current injection electrodes located on the right and left wrist and foot, and quad voltage electrodes located at the same place (Figure 4d). Figure 4. Segmental bioimpedance analysis techniques, (a) right side dual current and quad voltage electrodes, (b) right side dual current and quad voltage electrodes, (c) double sides dual current and quad voltage electrodes and (d) double sides quad current and quad voltage electrodes. (a) (b) (c) (d) Voltage Electrodes Current Electrodes (d) (a) Voltage Electrodes Limitations of whole body bioimpedance measurement in evaluating body segment compartments have given rise to the demand for segment localized bioimpedance analysis applications. Scharfetter et al. [40], reported that using segmental (across the waist) localized bioimpedance analysis can significantly estimate abdominal fat with a correlation coefficient of R2 = 0.99; furthermore Seward et al. [61], introduced localized bioimpedance analysis as a trending diagnostic tool for neuromuscular disorders. The study was applied on 25 neuromuscular patients and 45 normal subjects for control. Studies report that the segmental bioimpedance analysis method shows some limitations in the estimation of FFM [62,63], with estimation power not significantly different from whole body 10904 2.6. Alternative Bioimpedance Analysis Method Bioimpedance analysis, as an independent method for the assessment of the human health status from absolute bioimpedance measurements, has triggered a new path of data analysis and interpretation. The bioimpedance vector analysis method (BIVA) is a novel approach established essentially by Piccoli et al. [64,65] to estimate the hydration status using height indexed resistance and reactance data (R-Xc graph) from bioimpedance measurements. Using 8,022 normal subjects (3796 female and 4226 male) Piccoli et al. [66] formulated 50%, 75%, and 95% tolerance ellipses that determine increasing and decreasing body mass if the minor vector falls in the left and right half of the 50% ellipse, along with increasing and decreasing hydration ratio if the major vector falls in the lower and upper half of the 50% ellipse (Figure 5). Figure 5. Bioimpedance vector analysis (BIVA) and tolerance ellipses. Figure 5. Bioimpedance vector analysis (BIVA) and tolerance ellipses. Evaluation study of the BIVA method by Cox-Reijven et al. [67], on 70 diseased subjects with gastrointestinal disorders, conclude the high specificity and low sensitivity of BIVA method in classifying patients with extraordinarily rates of body fluids. Low values (Xc/H < 27.7 O/m and R/H < 563.6 O/m) in the BIVA method can be considered as predictors of severity among diseased children, as shown in a study conducted on 332 precarious pediatric patients with multiple organ dysfunction (MODS), acute respiratory distress syndrome (ARDS) and acute lung injury (ALI) [68]. In [69] the BIVA method successfully monitored rapid increases in ECF during short term recovery (3 weeks) and a dramatic increase in BCM during long term recovery (3 months) among 47% of 57 diseased women with anorexia nervosa [5]. The BIVA method is also considered as a valid tool for the estimation of dry weight in 24 haemodialysis patients’ with reference to the Bilbrey Index based on different allocation of values before and after obtrusion [70]. Sensors 2014, 14 10905 Kyle et al. reported that the BIVA method is affected by differences in biological factors and measurement artifacts [5]. Ward and Heithmann state that BIVA is affected by body size and influenced by the cross-sectional area of the body [71]. Kyle et al. reported that the BIVA method is affected by differences in biological factors and measurement artifacts [5]. Ward and Heithmann state that BIVA is affected by body size and influenced by the cross-sectional area of the body [71]. 2.6. Alternative Bioimpedance Analysis Method A specific BIVA method has been proposed by Marini et al. [72] to neutralize the bias due to body size. The specific BIVA method used a resistivity-reactivity graph that is constructed using information and results collected from multiplication of resistance and reactance by ratio of cross section area and length (L/A) from Ohm’s law (Equation (1)). The cross section area (A) and length (L) were estimated as follows: A = (0.45(arm area) + 0.45(calf area) + 0.10(waist area)) in square meter [73,74], where segment area = c2/4π and (c) is the circumference in meter of the arm, waist and calf, respectively; L = 1.1 (Ht), where Ht is body height in meters. Another alternative method for analysis is real time processing of bioimpedance data which is currently introduced as a key feature for body health monitoring applications. A logarithmic analysis carried out between 0.01 and 10 Hz with five frequencies needs 276 s to be completed, this includes the calculation time [75]. Sanchez et al. [76] stated that real time processing, accuracy and the ability of data retrieval and throughput of a BIS system were the most important features to be applied in health monitoring systems, and Sanchez et al. [77] introduced a local polynomial based method for impedance-frequency-response estimation. Comparison studies between four different multi-sine periodic broadband excitations broadband for EIS measurements in term of accuracy and speed in frequency and time domain concluded that multi-sine and discrete interval binary sequences (DIBS) enhance SNRZ and have better accuracy than chirp and maximum length binary sequences (MLBS) [75]. Use of multi-sine excitation signals in bioimpedance measurements that is proposed in [78,79] helped increase the accuracy of the measured bioimpedance parameters. It has been validated using a set of optimal multi-sine measurements on 2R-1C equivalent electrical circuits, then applied on healthy myocardium tissue. The multi-sine excitation method was introduced as a parametric-in-time identification method for electrical bioimpedance measurements with inclusion of harmonic impedance spectra (HIS). HIS directly identified from noisy current and voltage myocardium measurements at the multi-sine measurement frequencies to express periodic changes of impedance, rather than the commonly used method that assumed the measurement changing over time [80]. 3.1. Fat Mass (FM) and Fat Free Mass (FFM) FM and FFM estimations are considered one of the main objectives of body composition assessment techniques. Variations in FM among the reference population are due to several factors, but are believed to follow aging factors in addition to gradual changes in lifestyle [81]. Anthropometric and skin fold thickness measurements are traditional, simple and inexpensive methods for body fat estimation to assess the size of specific subcutaneous fat depots [82] compared with other methods such as underwater weighing, dilution method and dual-energy x-ray absorptiometry [DXA] that requires a trained practitioner to perform it. Bioimpedance analysis has been shown in recent studies to be more precise for determining lean or fat mass in humans [83]. In comparison with BMI, anthropometric and skin fold methods, BIA offers trustable results in the estimation of fatness across human tissues [84]. Several studies conducted to establish reference values for FFM are based on bioimpedance measurements. Kyle et al. [13] developed a single Equation (12) for the prediction of FFM, using 343 normal subjects aged from 22 to 94 years old, with body mass indexes between 17.0 and 33.8 kg/m2 in reference to DXA method: (12) (12) (12) (12) where (Ht) is body height, (R50) and (XC, 50) is resistance and reactance at 50 KHz, and (Wt) is body weight. The developed equation achieved a correlation coefficient (R) that is equal to 0.986, standard error of the estimate (SEE) is equal to 1.72 kg and technical error is 1.74 kg. In [81,85], FFM was assessed in a population of 5,225 white subjects aged from 15 to 98 years old using bioimpedance measurements and it was concluded that mean FFM was 8.9 kg or 14.8% lower in men older than 85 years than in men 35 to 44 years old and 6.2 kg or 14.3% lower in women older than 85 years than in women 45 to 54 years old. Sun et al. [86], used a multi-component model based on densitometry, isotope dilution, and dual-energy X-ray absorptiometry to build Equations (13) and (14) for FFM estimation: (13) (14) (13) (14) The mean FFM prediction equations achieved a correlation coefficient R2 = 0.90 and 0.83 and root mean square errors of 3.9 and 2.9 kg for males and females, respectively. The mean FFM prediction equations achieved a correlation coefficient R2 = 0.90 and 0.83 and root mean square errors of 3.9 and 2.9 kg for males and females, respectively. Sensors 2014, 14 Sensors 2014, 14 10906 3. Body Composition Prediction Using Bioimpedance Analysis Body composition assessment is considered a key factor for the evaluation of general health status of humans. Several methods use different assumptions to estimate body composition based on the number of compartments. This review considers that the human body is composed of two main compartments, FM and body lean mass or FFM. FFM is composed of bone minerals and body cell mass (BCM) that includes skeletal muscle mass (SMM). BCM contains proteins and TBW that represents 73% of lean mass in normal hydrated subjects. TBW is composed of ICF and ECF as illustrated in Figure 1. In this section, several predictive equations for both lean and fat mass, in addition to body fluids, will be discussed. Sensors 2014, 14 Sensors 2014, 14 Sensors 2014, 14 10907 Pichard et al. [88], assessed FFM and FM in a 3,393 white subject population aged from 15 to 64 years old using bioimpedance measurements and performed a comparison of %FM as determined by BIA with %FM determined by calculations using BMI developed by Deurenberg et al. [89], and concluded that the mean FFM ranged of 59.1–61.0 kg for men and 43.3–44.1 kg for women which is 38% greater in men. Heitmann [90] compared three body composition methods (BMI, skin folds and BIA) using 139 healthy subjects aged from 35 to 65 years old: ( (16) The multiple regression Equation (16) for impedance had a higher correlation coefficient (R2 = 0.89) and lower standard estimation error (SEE = 3.32 kg) than the multiple regression equations for skin fold (R2 = 0.81, SEE = 3.91 kg) or body mass index (R2 = 0.85, SEE = 3.94 kg). Heitmann [88] assessed FFM and FM in 2987 out of a 3608 subject Danish population aged from 35 to 65 years old. The obtained data, which are estimated from measurements of electrical impedance, concluded that men have a FM of 4.5 kg, an increase by 30%, when compared to women that have a 6.9 kg increase of 36% for evaluated sample. Recently, Pichler et al. [91] assessed FM in 116 subjects (32 healthy subjects and 84 patients) and concluded that the following prediction equation overestimated FM by 6.55 ± 3.86 kg: (17) (18) (17) (18) where Recf and Rtbw represents resistance of extracellular fluids and total body water extracted using the Cole module [26]. In conclusion, all studies state that the men have higher estimated FM as compared to women. Moreover, FFM for both genders decreases progressively with increasing age [81,88]. where Recf and Rtbw represents resistance of extracellular fluids and total body water extracted using the Cole module [26]. In conclusion, all studies state that the men have higher estimated FM as compared to women. Moreover, FFM for both genders decreases progressively with increasing age [81,88]. 3.2. Body Fluids 3.1. Fat Mass (FM) and Fat Free Mass (FFM) Deurenberg et al. [87], used densitometry, anthropometry and bioelectrical impedance to formulate FFM prediction Equation (15) using 661 normal adult subjects aged from 7 to 83 years old: (15) (15) The FFM prediction equations achieved a correlation coefficient R2 = 0.93 and standard estimation error (SEE) = 2.63 kg. 3.2. Body Fluids Body fluid is the total volume of fluids inside a human body that represents the majority of the FFM volume percentage. TBW includes the fluids inside the cellular mass that is known as ICF; and the fluid located outside the cell body which is composed of plasma and interstitial fluid which is known as ECF. ECF and ICF fluids that are incorporated under TBW, contain several ion types with different concentrations, however the main ions in ECF are Na+ and Cl−, and for ICF are K+ and PO−4 [92]. Body fluids estimation using bioimpedance measurements are based on the inversely proportional between body resistance and the total amount of body water [93]. There are varieties of methods for estimating body fluid volumes based on bioimpedance analysis approach. Sun et al. [86] developed prediction Equations (19) and (20) of the TBW reference to dilution method using SF-BIA from a multi-ethnic pool of 1830 people aged from 12 to 94 years old: Sun et al. [86] developed prediction Equations (19) and (20) of the TBW reference to dilution method using SF-BIA from a multi-ethnic pool of 1830 people aged from 12 to 94 years old: (19) (19) (19) Sensors 2014, 14 10908 (20) (20) (20) The developed equation achieved a correlation coefficient (R2) and mean square error equal to 0.84 and 3.8 L in men, and 0.79 and 2.6 L in women. The developed equation achieved a correlation coefficient (R2) and mean square error equal to 0.84 and 3.8 L in men, and 0.79 and 2.6 L in women. For ECF and ICF estimation using SF-BIA, a few studies performed were based on measurement of bioimpedance in 50 KHz frequency, Sergi et al. [94], predict ECF using two frequencies (1 and 50 KHz): (21) ; (22) (21) ; (22) (21) (22) After measurements performed using bioimpedance and bromide dilution methods on 40 subjects (19 males and 21 females) aged 21–81 years, of which 22 were healthy subjects, 12 were affected by chronic heart failure and 6 by chronic renal failure, the best estimation results at 1 KHz achieved a correlation coefficient (R2) and standard estimation error equal to 0.89 and 1.7 L. Due to incomplete conduction of the intracellular fluid at 50 kHz [2], MF-BIA was proposed to increase accuracy of estimation of TBW, ECF and ICF. Deurenberg et al. 3.2. Body Fluids [95] used MF-BIA (1, 5, 50, 100 KHz) to predict TBW using Z100KHz and Z50KHz; and ECF using Z1KHz and Z5KHz among 139 normal adult subjects with reference to deuterium oxide dilution and bromide dilution: (23) (24) (23) (24) (24) The prediction equation of TBW achieved a correlation coefficient (R2) and standard error of estimate (SEE) equal to 0.95 and 1.73 L using Z100KHz, and 0.95 and 1.74 L using Z50KHz: The prediction equation of TBW achieved a correlation coefficient (R2) and ti t (SEE) l t 0 95 d 1 73 L i Z d 0 95 d 1 74 L i Z The prediction equation of TBW achieved a correlation coefficient (R2) and standa (25) (26) (25) (26) The prediction equation of ECF achieved a correlation coefficient (R2) and standard error of estimate (SEE) equal to 0.87 and 0.98 L using Z1KHz, and 0.86 and 1.02 L using Z5KHz. The prediction equation of ECF achieved a correlation coefficient (R2) and standard error of estimate (SEE) equal to 0.87 and 0.98 L using Z1KHz, and 0.86 and 1.02 L using Z5KHz. Prediction of body fluids using the BIS method in three steps involves firstly determination using the values of Re from R0 and Rinf, secondly, inclusion of the body shape factor Kb due to the variation of body segments, and thirdly, inclusion of apparent resistivity ρa instead of the general resistivity ρ as stated by Hanai in mixture theory [18]: (27) (27) where (c) is volume fraction of non-conducting tissue. Based on Hanai’s mixture method [18], tissue resistance (R) is measured based on conductive tissue, so it should exclude non-conducting tissue. where (c) is volume fraction of non-conducting tissue. Based on Hanai’s mixture method [18], tissue resistance (R) is measured based on conductive tissue, so it should exclude non-conducting tissue. Sensors 2014, 14 10909 Thus, by substituting Equation (27) in Equation (11), the apparent resistance (Ra) can be calculated using the following Equation (28): Thus, by substituting Equation (27) in Equation (11), the apparent resistance (Ra) can be calculated using the following Equation (28): (28) (28) At low frequencies the current will pass through extracellular fluids only without intracellular fluid due to the high capacitance of cell membranes [96]. In that case the conducting volume is equal to the ratio between ECF volume (Vecf) and TBW volume (Vb). 3.2. Body Fluids The volume fraction of non-conducting tissues at low frequencies calculated as in Equation (29): (29) (29) Based on the mixture theory [18], apparent resistivity (ρa) at low frequency represents the extracellular fluid resistivity (ρAecf), thus the resistance of ECF (Recf) can be recalculated in Equation (31), by substituting Equation (29) in Equation (28) and including the outcome of apparent resistivity (ρaecf) from Equation (30): (30) (31) (30) (31) Hanai [18], calculated ρecf to be equal to 40.3 Ω∙cm for men and 42.3 Ω∙cm for women, which is close to that achieved by saline, and is about 40 Ω∙cm t for the ECF composed of plasma and interstitial water [49]. To reform the equation to evaluate the variance in ECF volume (Vecf) caused by changes in estimated ECF resistance (Recf), that is achieved by replacing body volume (Vb), that is equal to the ratio between body weight (Wt) in Kg and body density (Db) in Kg/L from Equation (32) in Equation (33): (32) (32) (33) (33) Body factor (Kb), extracellular fluid resistivity (ρaecf) and body density (Db) are constant values that can be included in one factor defined as extracellular fluid factor (Ke) as in Equation (34), and for extracellular fluid volume (Vecf) as in Equation (35): (34) (34) Sensors 2014, 14 10910 (35) (35) Van Loan et al. [43], calculated Ke using the bromide dilution method to be equal 0.306 for men and 0.316 for women; and the ratio between ρicf and ρecf to be equal to 3.82 for men and 3.40 for women. De Lorenzo et al. [10] calculated Ke to be equal to 0.229 in women; and ρecf to be equal to 40.5 Ω∙cm and 39.0 Ω∙cm for men and women, respectively; and the ratio between ρicf and ρecf to be equal to 6.76 for men and 6.79 for women. Ellis and Wong [30], analyzed the BIS method as introduced by Van Loan et al. [43], with reference to the H2O and Br dilution technique in 469 multi-ethnic healthy subjects. The study suggested that the ratio between ρicf and ρecf is equal to 3.032 for men and 2.694 for women, due to underestimation of TBW caused by misprediction of ICF measurements. Biasing factors and different regression module approaches caused slight differences in the ratios obtained by these researchers [30]. Moissl et al. 3.2. Body Fluids [32] introduced a second generation mixture theory to overcome the limitations of the first generation in predicting intracellular fluid volume (Vicf) using a new assumption for TBW resistivity (ρtbw), as in Equation (32): (41) (41) In the second version of mixture theory, total body water volume is considered to be equal to the summation of ECF and ICF, for ECF estimation, the relation in Equation (35) is considered as a valid method, and for ICF estimation, the method uses Equation (42); note that the ratio (Rtbw/Recf) is opposite and proportional to (Vtbw/Vecf): In the second version of mixture theory, total body water volume is considered to be equal to the summation of ECF and ICF, for ECF estimation, the relation in Equation (35) is considered as a valid method, and for ICF estimation, the method uses Equation (42); note that the ratio (Rtbw/Recf) is opposite and proportional to (Vtbw/Vecf): (42) (42) Moissl et al. [97] calculated ρicf to be equal to 273.9 Ω∙cm and ρecf = 40.5 Ω∙cm in men and 264.9 Ω∙cm and 39.0 Ω∙cm, respectively in women. De Lorenzo et al. [10] suggest the formula in Equation (34) to determine intracellular volume (Vicf): Moissl et al. [97] calculated ρicf to be equal to 273.9 Ω∙cm and ρecf = 40.5 Ω∙cm in men and 264.9 Ω∙cm and 39.0 Ω∙cm, respectively in women. De Lorenzo et al. [10] suggest the formula in Equation (34) to determine intracellular volume (Vicf): (43) (43) Jaffrin and Morel [21] claim that the prediction of ECF by Hanai [18] mixture theory is valid and direct, however the ICF prediction by De Lorenzo et al. [10], who state that the determination of Ri is less accurate than for Re in parallel module because it sums up the errors on Re and Rinf, is not. Jaffrin and Morel [21] claim that the prediction of ECF by Hanai [18] mixture theory is valid and direct, however the ICF prediction by De Lorenzo et al. [10], who state that the determination of Ri is less accurate than for Re in parallel module because it sums up the errors on Re and Rinf, is not. Moissl et al. 3.2. Body Fluids [97], suggested a body composition spectroscopy method through recalculating Kecf , using different assumptions through inclusion of body mass index (BMI) and taking the module of non-conducting tissue factor (c) in Equation (14) as a valid assumption, as in Equation (36), and then determining the (Vecf) using the same equation as Equation (20): (36) (36) From [97], (a) and (b) were calculated to be equal to 0.188 and 0.2883 based on measurements using the Br dilution method as a reference method on dialyzed patients and 120 healthy subjects. At high frequencies, the current will pass through the whole TBW which is composed of ECF and ICF [96], so the conducting volume is equal to the ratio between TBW and total body volume. Jaffrin et al. [31] suggested calculating the TBW directly from Rinf using the same assumption of mixture theory [96], and assuming uniformity of water compartments inside human body. Thus, using the same assumption as in Equation (29), the volume fraction of non-conducting tissue (c) at high frequencies can be calculated using Equation (37): (37) (37) To determine the apparent resistivity of total body water (ρa_tbw) from actual total body water resistivity (ρtbw), the parameters in (c) from Equation (37), was included into Equation (38): To determine the apparent resistivity of total body water (ρa_tbw) from actual total body water resistivity (ρtbw), the parameters in (c) from Equation (37), was included into Equation (38): (38) (38) By replacing the actual resistivity by apparent resistivity for total body water in Equation (11), and restoring the value of (Vb) from Equation (32), Equation (40) to determine the total body water factor (Ktbw) and total body water volume (Vtbw) is recalculated by using Equation (39): (39) (39) Sensors 2014, 14 10911 (40) (40) Considering that total body water is equal to the accumulation of ECF and ICF, Jaffrin et al. [31] calculated ρtbw to be equal to 104.3 Ω∙cm in men and 100.5 Ω∙cm. A validation study conducted in 28 dialysed patients [31], concluded that ρtbw was equal to 108.1 Ω∙cm in men and 100.2 Ω∙cm, which predicted 91% of mean water loss when compared with 39% for Cole method [43], but overestimated TBW compared to the original BIS method in 21 healthy subjects with the same ρtbw and hydration rate values. For ICF prediction using a BIS method, Matthie et al. 3.2. Body Fluids [97] introduced a different method for calculation of intracellular fluid volume (Vicf), taking into consideration that the non-conducting tissue factor (c) is as given in Equation (44): (44) (44) Then the recalculated intracellular fluid factor (Kicf) and intracellular fluid volume (Vicf) are added as in Equations (45) and (46), respectively, and it is concluded that total body water factor (Ktbw) and total body water volume (Vtbw) is equal to the summation of ECF and ICF volumes as in Equation (47) and recalculated (Vtbw) using different assumption of (Ktbw) and (ρtbw) from Jaffrin et al. [31], and Matthie et al. [32], as given in Equations (48) and (49): Sensors 2014, 14 Sensors 2014, 14 10912 (45) (46) (45) (45) (46) (46) (47) (47) (48) (49) (48) (49) where (c) and (d) are calculated to be equal to 5.8758 and 0.4194 in [97], when using the 40K isotope [98] as a reference method on dialyzed patients and 120 healthy subjects. Fenech and Jaffrin [2] state that ECF prediction using segmental bioimpedance analysis in supine position (0.79 liter) is less than Watson anthropomorphic method [3] (1.12 liter) and for ICF is reduced by 3.4% for segmental bioimpedance and 3.8% for the Watson anthropomorphic method [3]: (50) (50) (50) Pichler et al. [91] examined the BIS method using an Impedimed device (SFB7) in TBW, ECF and FFM with reference to the deuterium space method, sodium bromide space method and DXA method, respectively. The study was applied on 32 healthy subjects and 84 patients with different types of diseases (congestive heart failure, coronary heart disease, essential hypertension, atherosclerosis, kidney disease, chronic renal failure, gastrointestinal diseases, type II diabetes, morbid obesity, osteoporosis, cancer, chronic polyarthritis and anorexia nervosa): (51) (52) (53) (54) (51) (52) (53) (54) Pichler’s equations for TBW achieved a correlation coefficient 0.91 and 0.89 for men and women, respectively, as in Equations (51) and (52). For ECF it achieved 0.87 and 0.89 for men and women, respectively, as in Equations (53) and (54) [91]. Hanai mixture equations [18], when applied in SFB7 give ECF measurements higher than the sodium bromide space method by mean ± SD (0.93 ± 2.62 Liter) however it is noted that the Hanai mixture equations applied in SFB7 detect ECF Pichler’s equations for TBW achieved a correlation coefficient 0.91 and 0.89 for men and women, respectively, as in Equations (51) and (52). 4.1. Anthropometric Measurements Anthropometric measurements such as weight, height, skin fold thickness, lengths, diameters and circumferences that involves mathematical modules are the main contributors in the estimation of body compartments [5,99]. Bioimpedance parameters only without body dimension measurements are considered poor estimators for body composition [91,100]. Diaz et al. [101] concluded that in FM and FFM prediction, resistance and capacitance measurements contribute by 0%–20%. In contrast, the percentages increase to 11%–53% after height inclusion, and 22%–68% after inclusion of Ht2/R ratio. Ward et al. [102] presented a validation study to predict BCM and ECF as a portion of TBW without measuring height and using BIA device, the Soft Tissue Analyzer STATM (Akern Sri, Florence, Italy) with a correlation coefficient referenced to the total body potassium counting method is equal to 0.91, 0.82 and 0.89, and a standard estimation error equal of 5.6 kg, 6.3 kg and 1.3 kg for FFM, BCM and ECF, respectively. Sensors 2014, 14 10913 Sensors 2014, 14 Sensors 2014, 14 excess in 9 patients, and TBW measurements higher than the deuterium space method by mean ± SD (3.82 ± 3.37 Liter), and FFM measurements lower than the DXA method by mean ± SD (6.55 ± 3.86 kg). 4.2. Gender Variations in body composition between male and female were proven in several studies [103]. In body composition prediction, methods based on bioimpedance analysis, and most equations tend to include gender as one of the main determining factors for body compartment assessment [13,86,87]. FFM or lean mass studies show that males have greater FFM than females with different ranges. Kyle et al. [81] state that mean FFM for male is 8.9 kg and 6.2 kg for female and fat mass index FMI increases based on age, in females from 5.6 to 9.4 and from 3.7 to 7.4 in males. In a recent study [104] on 1649 healthy children-adults (6–18 years) and 925 adult-elders (19–92 years) using BIA and DXA it was concluded that for all age ranges, males have less fat mass and more fat free mass than females. TBW averaged 73.2% of fat free mass in the healthy population; however several studies show that males have less TBW than females [11]. Sun et al. [86], stated that in a mixed ethnic groups prediction equation, TBW volume for males start from 1.2 L compared with 3.75 L for females. Jaffrin et al. [31] state that determined TBW resistivity (ρ) is on average 104.3 ± 7.9 Ω∙cm for men and 100.5 ± 7.8 Ω∙cm for women. The values are smaller in men are due to their larger limb cross section. Due to the different body composition between males and females, gender considerations have a strong impact in estimating body compartments. 4.3. Age 3.2. Body Fluids For ECF it achieved 0.87 and 0.89 for men and women, respectively, as in Equations (53) and (54) [91]. Hanai mixture equations [18], when applied in SFB7 give ECF measurements higher than the sodium bromide space method by mean ± SD (0.93 ± 2.62 Liter) however it is noted that the Hanai mixture equations applied in SFB7 detect ECF Sensors 2014, 14 Sensors 2014, 14 10914 weight is seen among the elder population compared with adults associated with increment in fat mass [81]. In some studies [58], the general body composition prediction equations were unsatisfactory in elderly men over 75 years of age, especially in TBW estimation. Several studies were conducted using the BIA method on children [68,105] adults [13], and elders [106,107]. In children, the BIA method using the Deurenberg equation [87], underestimates body fat as determined by DXA. It however achieved a better correlation than the skin fold method [108]. Muscle mass loss among the elderly reduces the fat free mass at a certain age, followed by decreases in total body water and bone mass [109]. Marja et al. [107] reported that in 75-year-old Swedes, average fat free mass index was 15.6 and 18.3; and body fat index was 11.0 and 8.6 for women and men, respectively, compared to the DXA method. 4.4. Ethnic Groups Body composition varies among different races and ethnic groups due to the environment, nutrition factors, culture and anthropometric measurements that include body conformation [110]. There is also difference in limb length [111], body structure [112], body size [89] and that lead to variation in body fat percentages among different ethnic groups which may lead to prediction errors (3%) [111]. The majority of bioimpedance measurement studies have been done on Caucasian subjects [5], Kotler et al. [113] and Sun et al. [86] have included African American and Hispanic subjects in their studies. Kim et al. assessed the segmental lean mass among Koreans [106], Schulz et al. assessed the fat free mass among Germans and compared it to the American and Swiss population [114]. Siváková et al. studied the clinical applications of BIVA on Slovaks [115]. Nigam et al. had performed a comparative study among two different Indian races [116], whereas Saragat et al. obtained specific BIVA reference values for the Italian healthy elderly population in order to construct the specific tolerance ellipses to be used for reference purposes for assessing body composition in gerontological practice and for epidemiological purposes [117]. Validation of bioimpedance measurements among different ethnicities is thus needed due to differences in body composition among certain populations. 4.3. Age Aging is defined as a multi-factor changing in the physical and biological activities of the human body that leads to differences in body composition among age groups. When the human body becomes older it leads to a gradual increase in fat mass and spontaneous decrease in lean mass. Fat free mass to fat mass ratio increases gradually in response to increase of age, and a noticeable increment in average 4.6. Electrode Shape and Measurement Error In bioimpedance analysis, the geometrical structure of electrode has a strong impact on elementary data retrieved during the measurement process. In bioimpedance analysis electrodes are defined as isoelectric materials with a negligible voltage drop along the connectors. The minimum numbers of electrodes required to perform the bioimpedance measurements are two, one for current injection with the assumption of zero potential difference and the other for collecting the voltage drop with a negligible current flow and is more affected by position. The tetrapolar electrode approach become widely used for whole bioimpedance measurements because of the uniformity of current distribution compared to monopolar electrodes [6], and the usage of more than two potential collecting electrodes or octapolar electrode method were used for segmental bioimpedance studies to assess compartments in different body segments [73]. Ag-AgCl electrodes are now used in most bioimpedance measurements because it has a well-defined DC potential with electrolyte gel to minimize the gap impedance between skin and electrodes. Circular and rectangular electrode shapes with a contact area greater than 4 cm2 are the most commonly used shapes [1]. Buendía et al. investigated the impact of electrode discrepancy on BIS measurements and concluded that mismatched potential electrode causes 4% overestimated measurements in resistance at zero and infinite frequency because of an imbalanced electrical field distribution [123]. Shiffman [124] addressed the artifacts caused by inaccurate distance between electrodes in four electrode measurement methods performed on a 17.5 cm segment of the thigh area. That study reported that the values of resistance and reactance were four times larger when the current injecting electrodes were placed 2.5 cm from the sensing electrodes. Scharfetter et al. stated that capacitance between different body segments and earth, and capacitance between the signal ground of the device and earth cause a significant false dispersion in the measured impedance spectra at frequencies >500 kHz [125]. Errors in bioimpedance measurements are caused by many factors such as motion, miss-positioning, connector length and fabrication errors. Moreover, the diversity of the commercially available bioimpedance analyzers cause a wide range of fluctuations in measurements between the devices. Thus the calibration of the components inside a bioimpedance analyzer such as signal generator, sensing apparatus, scales of weight and height and electrical interference should be conducted to ensure the reliability of the bioimpedance analyzers [1]. Sensors 2014, 14 10915 Sensors 2014, 14 Sensors 2014, 14 studies skin temperature should be counted [84,120]. Subjects under test should not perform any exercise activities before measurements that could lead to errors in assessed resistance and reactance equal to 3% and 8% respectively [121]. Roos et al. concluded that the error in total body water prediction range from 1 to 1.5 L figured out after laying at rest for one hour [122]. 4.5. Measurements Protocols and Posture Simplicity and the economic acceptance of bioimpedance analysis method for body composition estimation have increased the need to unify the protocols and procedures of bioimpedance measurements in order to retrieve robust data. For the foot to ankle measurement method, bioimpedance measurements performed in a supine position with abduction of the upper limbs to 30 degrees and lower limbs to 45 degrees for 5 to 10 min. studies show that when the posture changes from a standing to a supine body position, the ECV decreased in the arms by 2.51% and legs by 3.02%, but increased in the trunk by 3.2% [118]. Fasting for at least 8 hours and bladder voiding before measurements are recommended as consumption of food and beverages may decrease impedance by 4–15 O over a 2–4 h period after meals and that causes an error (<3%) [84,119,120]. Body anthropometric measurements should be retrieved prior of the test and for scale or foot to foot bioimpedance analyzer weight retrieved automatically [1]. Electrodes should be placed on the pre-cleaned metacarpal and metatarsal phalangeal joints with a distance in between of at least 5 cm without skin lesions at the location of the electrodes. In some Sensors 2014, 14 10916 Sensors 2014, 14 Sensors 2014, 14 method and underwater weight measurements is used to estimate fat mass and fat free mass, however bioimpedance analysis can estimate FM and FFM in addition to total and particular body fluids which is very helpful for disease prognosis [127]. The National Health and Nutrition Examination Survey program in United States included bioimpedance analysis in the third NHANES program between 1999 and 2004 to assess the health and nutritional status of adults and children because of a general frustration with the dependability of the skin fold thickness method to estimate FM and FFM, especially in subjects with higher amount of segmented fat [128]. Observation of body compartment fluctuations like fat free mass, fat mass and total body water from normal limits are considered as key factors to be used in bioimpedance analysis in healthcare applications. Abnormal loss in lean body mass and unbalanced shifts in body fluids are the most measured parameters to be used to assess the healthiness of the human body. Analysis of bioimpedance parameters has bern used in several studies to estimate and analyze the changes in disorders of different kind of diseases. Norman et al. [70] stated that phase angle is an essential predictor of clinical status. Pichler et al. [91] stated that estimation of body fluids using BIS was slightly better than anthropometric methods among healthy and diseased. Table 1 contains some of the applications of bioimpedance analysis in disease diagnosis that are organized according to the organ systems of human body, diseases or abnormalities diagnosed based on bioimpedance parameters, and comments on how these factors are applied to determine the health condition. Bioimpedance analysis is a common method used for estimating body composition among healthy and diseased subjects in research and clinical trials. This review has focused on the theoretical and the fundamentals of bioimpedance analysis. Thus it may have some limitations, where possible important studies on the applications of bioimpedance analysis in diagnostic of diseases and the related shifts in bioimpedance parameters may have been missed. Table 1. Applications of bioimpedance analysis in clinical status monitoring and diagnosis of diseases. Organ Systems Diseases BIA Parameters Remarks Authors Pulmonary system Lung cancer, stages IIIB and IV R and Xc (BIVA) Reactance components decrease in patients (phase angle <4.5). Clinical Study. 5. Applications of Bioimpedance Analysis in Clinical Status Monitoring and Diagnosis of Diseases Bioimpedance analysis in healthcare practice contributes to the estimation of body compartments to assess the regular change in nutrition status in in-patients and to monitor nutritional risk in out-patients [126]. Most of the body composition assessment methods like BMI techniques, skin fold Sensors 2014, 14 Toso et al., 2000 [129] Pulmonary edema monitoring R (SFBIA) Mean resistivity for left and right lung (1205 ± 163, 1200 ± 165 Ω∙cm) and system reproducibility (2%). Research Study. Zlochiver et al., 2007 [130] Cardio- vascular system Fluid accumulation after cardiac surgery. Ht2/Z (MFBIA) Significant increase in segmental trunk bioimpedance after surgery due to fluid accumulation. Clinical Study. Bracco et al., 1998 [131] ble 1. Applications of bioimpedance analysis in clinical status monitoring and diagnosis ble 1. Applications of bioimpedance analysis in clinical status monitoring and diagnosis diseases. 10917 Sensors 2014, 14 Table 1. Cont. Organ Systems Diseases BIA Parameters Remarks Authors Circulatory system Volaemic status and hyponatraemia TBW (SFBIA) In elderly hyponatraemic patients, TBW assessment using BIA method was correlated with dilution of deuterium oxide (R = 0.68). Clinical Study. Hoyle et al., 2011 [132] Hydration status and hyponatraemia in elderly TBW (SFBIA) Assessment of hydration status in elderly hyponatraemic patients using BIA method was more accurate than clinical procedures (Cohen’s kappa coefficient = 0.52). Clinical Study. Cumming et al., 2014 [133] Renal system Chronic hemodialysis ECF (BIS) ECF to weight ratio of hypertensive patient’s increase from that of normal patients (24.29 ± 3.56% vs. 21.50 ± 2.38). Clinical Study. Chen et al., 2002 [134] Dry weight in kidney failure. ECF (BIS) ECF/Wt is 0.239 and 0.214 L/kg for male and female healthy subjects. Clinical Study. Chamney et al., 2002 [135] Hydration states monitoring in hemodialysis patients Calf-BIS (BIS) Normalized resistivity (μ = ρ /BMI) increased from 17.9 ± 3 to 19.1 ± 2.3 × 10−2 Ω3∙Kg−1, and weight was reduced from 78.3 ± 28 to 77.1 ± 27 kg in Post-dialysis. Research Study. Zhu et al., 2007 [136], 2008 [137] Dry weight assessment hemodialysis patients Calf-BIS (BIS) Dry weight assessed by cBIS underestimate left ventricular mass and blood pressure while antihypertensive medication remains unchanged. Clinical Study. Seibert et al., 2013 [138] Body fluids estimation in hemodialysis patients ECF, ICF and TBW (BIS) Correlation between proposed equation corrected for BMI and the references (mean ± SD) was −0.4 ± 1.4 L for ECF, 0.2 ± 2.0 L for ICF and −0.2 ± 2.3 L for TBW. Clinical Study. Moissl et al., 2006 [97] Dry weight assessment HD patients R and Xc (BIVA) BIVA method shows significant different in vectors in post dialysed patients. Clinical Study. Sensors 2014, 14 Atilano et al., 2012 [139] Neural system Alzheimer’s disease R and Xc (BIVA) BCM decreased in patients for men, T2 (Hotelling's statistic) = 12.8 and for women, T2 = 34.9. Clinical Study. Buffa et al., 2010 [140] Anorexia nervosa (eating disorder) FM, FFM, TBW and ECF (BIS) The BCM to Ht2 ratio was found to be significantly changed between diseased and controls subjects. Clinical Study. Moreno et al., 2008 [141] Anorexia nervosa (eating disorder) R and Xc (BIVA) Gradually increasing in BCM and decreasing in ECF during treatments. Clinical Study. Haas et al., 2012 [69] Sensors 2014, 14 10918 Table 1. Cont. Organ Systems Diseases BIA Parameters Remarks Authors Muscular system Body composition changes monitoring during exercise training FFM and FM (MFBIA) BIA method underestimates FM (−3.42 kg) and overestimated FFM (3.18 kg); and undetected small shift in body composition due to exercise training. Clinical Study. Sillanpää et al., 2013 [142] Immunology system Comparison between SFBIA and MFBIA in HIV patients ECF and TBW (BIS) Insignificant differences in TBW and ECF estimation using SFBIA, MFBIA and BIS methods. Clinical Study. Paton et al., 1998 [143] Dengue haemorrhagic fever estimation in children ECF and ICF (BIS) (ECF/ICF) increase with increasing dengue virus infections severity in children. Clinical Study. Libraty et al., 2002 [144] Cancer patients TBW (SFBIA) Change in TBW using BIA method (Ht2/R50) correlate with deuterium dilution in underweight and normal-weight cancer patients (R2 = 0.43 and SEE = 1.22 L). Clinical Study. Simons et al., 1999 [145] Early diagnosis and risk analysis of dengue R, C, φ and Xc (SFBIA) Reactance variations among dengue patients during defervescence of feverintervalis an indicator for classifying risk category in the DHF patients. Clinical Study. Ibrahim et al., [146] Other diseases Critically ill subjects FM, TBW and ECF (BIS) Body composition using BIS method show slightly more significant in estimation of FM, TBW and ECF among healthy and diseased subjects. Clinical Study. Pichler et al., 2013 [91] Gastrointestinal disease R, Xc, Fc, FFM, TBW, ECF and ICF (BIS) In critically diseased subjects, Fc and ECF increased, Xc decreased, and TBW and ICF remain the same. Clinical Study. Cox-Reijven et al., 2003 [67] Sensors 2014, 14 Sensors 2014, 14 Sensors 2014, 14 10919 suitable bioimpedance measurements techniques, procedures and population, age, ethnic groups or disease-dedicated bioimpedance analysis equations. Further studies are needed to evaluate the correlations between variations in bioimpedance parameters, especially in ECF and ICF, and the deviation from health to disease. Acknowledgments This research is supported by UM High Impact Research Grant UM-MOHE UM.C/625/1/HIR/MOHE/05 from the Ministry of Higher Education Malaysia, Fundamental Research Grant Scheme (FRGS: FP042-2013B) and University of Malaya Research Grant (UMRG: RP009C-13AET). Author Contributions All authors contributed extensively to the work presented in this paper. Conflict of Interest The authors declare no conflict of interest. The authors declare no conflict of interest. 9. Kasap, S.O. Principles of Electrical Engineering Materials and Devices; McGraw-Hill: New York City, NY, USA, 1997. 6. Conclusions Increasing demands for accurate, cost effective and non-invasive systems for clinical status monitoring and diagnosis of diseases in healthcare, has accelerated the research endeavors to provide new methods and technologies to evaluate the health condition of human body. Body composition assessment tools has been considered a promising approach for the quantitative measurement of tissues characteristic over time, in addition to direct relativity between fluctuations in body composition equivalences and survival rate, clinical condition, illness and quality of life. Bioimpedance analysis is a growing method for body compartments estimation in nutrition studies, sport medicine and evaluation of hydration rate, fat mass and fat free mass between healthy and diseased populations. Fat mass, fat free mass including skeletal muscle mass, bone minerals, and total body water, which is composed of intercellular fluid and extracellular fluid, are compartments that can be predicted and analyzed using Sensors 2014, 14 Sensors 2014, 14 10920 10. De Lorenzo, A.; Andreoli, A.; Matthie, J.; Withers, P. Predicting body cell mass with bioimpedance by using theoretical methods: A technological review. J. Appl. Physiol. 1997, 82, 1542–1558. 10. De Lorenzo, A.; Andreoli, A.; Matthie, J.; Withers, P. Predicting body cell mass with bioimpedance by using theoretical methods: A technological review. J. Appl. Physiol. 1997, 82, 1542–1558. 11. Genton, L.; Hans, D.; Kyle, U.G.; Pichard, C. Dual-energy x-ray absorptiometry and body composition: Differences between devices and comparison with reference methods. Nutrition 2002, 18, 66–70. 12. Thomas, B.; Ward, L.; Cornish, B. Bioimpedance spectrometry in the determination of body water compartments: Accuracy and clinical significance. Appl. Radiat. Isot. 1998, 49, 447–455. 13. Kyle, U.G.; Genton, L.; Karsegard, L.; Slosman, D.O.; Pichard, C. Single prediction equation for bioelectrical impedance analysis in adults aged 20–94 years. Nutrition 2001, 17, 248–253. 14. Gudivaka, R.; Schoeller, D.; Kushner, R.; Bolt, M. Single-and multifrequency models for bioelectrical impedance analysis of body water compartments. J. Appl. Physiol. 1999, 87, 1087–1096. 15. Chertow, G.M.; Lazarus, J.M.; Lew, N.L.; Ma, L.; Lowrie, E.G. Development of a population-specific regression equation to estimate total body water in hemodialysis patients. Kidney Int. 1997, 51, 1578–1582. 16. Ward, L.C.; Dyer, J.M.; Byrne, N.M.; Sharpe, K.K.; Hills, A.P. Validation of a three-frequency bioimpedance spectroscopic method for body composition analysis. Nutrition 2007, 23, 657–664. 17. Lukaski, H.C.; Bolonchuk, W.W.; Hall, C.B.; Siders, W.A. Validation of tetrapolar bioelectrical impedance method to assess human body composition. J. Appl. Physiol. 1986, 60, 1327–1332. 18. Hanai, T. Electrical properties of emulsions. Kolloid-Zeitschrift 1961, 177, 57–61. 19. Olde, R.M.; Deurenberg, P.; Jansen, R.; Van’t Hof, M.; Hoefnagels, W. Validation of multi-frequency bioelectrical impedance analysis in detecting changes in fluid balance of geriatric patients. J. Am. Geriatr. Soc. 1997, 45, 1345–1351. 20. Woodrow, G.; Oldroyd, B.; Turney, J.; Davies, P.; Day, J.; Smith, M. Measurement of total body water by bioelectrical impedance in chronic renal failure. Eur. J. Clin. Nutr. 1996, 50, 676–681. 21. Jaffrin, M.Y.; Morel, H. Body fluid volumes measurements by impedance: A review of bioimpedance spectroscopy (bis) and bioimpedance analysis (bia) methods. Med. Eng. Phys. 2008, 30, 1257–1269. 22. Simpson, J.; Lobo, D.; Anderson, J.; Macdonald, I.; Perkins, A.; Neal, K.; Allison, S.; Rowlands, B. Body water compartment measurements: A comparison of bioelectrical impedance analysis with tritium and sodium bromide dilution techniques. Clin. Nutr. 2001, 20, 339–343. 23. References 1. Kyle, U.G.; Bosaeus, I.; De Lorenzo, A.D.; Deurenberg, P.; Elia, M.; Manuel Gómez, J.; Lilienthal Heitmann, B.; Kent-Smith, L.; Melchior, J.-C.; Pirlich, M. Bioelectrical impedance analysis—Part ii: Utilization in clinical practice. Clin. Nutr. 2004, 23, 1430–1453. 1. Kyle, U.G.; Bosaeus, I.; De Lorenzo, A.D.; Deurenberg, P.; Elia, M.; Manuel Gómez, J.; Lilienthal Heitmann, B.; Kent-Smith, L.; Melchior, J.-C.; Pirlich, M. Bioelectrical impedance analysis—Part ii: Utilization in clinical practice. Clin. Nutr. 2004, 23, 1430–1453. 2. Thomasset, A. Bio-electrical properties of tissue impedance measurements. Lyon Med. 1962, 207, 107–118. 2. Thomasset, A. Bio-electrical properties of tissue impedance measurements. Lyon Med. 1962, 207, 107–118. 3. Nyboer, J. Electrical Impedance Plethysmograph, 2nd ed.; Thomas, C., Ed.; Thomas publishers: Springfield, IL, USA, 1970. 3. Nyboer, J. Electrical Impedance Plethysmograph, 2nd ed.; Thomas, C., Ed.; Thomas publishers: Springfield, IL, USA, 1970. 4. Hoffer, E.C.; Meador, C.K.; Simpson, D.C. Correlation of whole-body impedance with total body water volume. J. Appl. Physiol. 1969, 27, 531–534. 4. Hoffer, E.C.; Meador, C.K.; Simpson, D.C. Correlation of whole-body impedance with total body water volume. J. Appl. Physiol. 1969, 27, 531–534. 5. Kyle, U.G.; Bosaeus, I.; De Lorenzo, A.D.; Deurenberg, P.; Elia, M.; Gómez, J.M.; Heitmann, B.L.; Kent-Smith, L.; Melchior, J.-C.; Pirlich, M. Bioelectrical impedance analysis—Part i: Review of principles and methods. Clin. Nutr. 2004, 23, 1226–1243. 5. Kyle, U.G.; Bosaeus, I.; De Lorenzo, A.D.; Deurenberg, P.; Elia, M.; Gómez, J.M.; Heitmann, B.L.; Kent-Smith, L.; Melchior, J.-C.; Pirlich, M. Bioelectrical impedance analysis—Part i: Review of principles and methods. Clin. Nutr. 2004, 23, 1226–1243. 6. Martinsen, O.G.; Grimnes, S. Bioimpedance and Bioelectricity Basics; Academic Press: Waltham, MA, USA, 2011. 6. Martinsen, O.G.; Grimnes, S. Bioimpedance and Bioelectricity Basics; Academic Press: Waltham, MA, USA, 2011. 7. Mialich, M.S.; Sicchieri, J.M.F.; Junior, A.A.J. Analysis of body composition: A critical review of the use of bioelectrical impedance analysis. Int. J. Clin. Nutr. 2014, 2, 1–10. 8. Lukaski, H. Evolution of bioimpedance: A circuitous journey from estimation of physiological function to assessment of body composition and a return to clinical research. Eur. J. Clin. Nutr. 2013, 67, S2–S9. 9. Kasap, S.O. Principles of Electrical Engineering Materials and Devices; McGraw-Hill: New York City, NY, USA, 1997. Sensors 2014, 14 Sensors 2014, 14 10921 Sensors 2014, 14 27. Cornish, B.; Ward, L.; Thomas, B.; Jebb, S.; Elia, M. Evaluation of multiple frequency bioelectrical impedance and cole-cole analysis for the assessment of body water volumes in healthy humans. Eur. J. Clin. Nutr. 1996, 50, 159–164. 27. Cornish, B.; Ward, L.; Thomas, B.; Jebb, S.; Elia, M. Evaluation of multiple frequency bioelectrical impedance and cole-cole analysis for the assessment of body water volumes in healthy humans. Eur. J. Clin. Nutr. 1996, 50, 159–164. 28. Pastan, S.; Gassensmith, C. Total body water measured by bioelectrical impedance in patients after hemodialysis: Comparison with urea kinetics. ASAIO J. 1992, 38, M186–M189. 29. Scanferla, F.; Landini, S.; Fracasso, A.; Morachiello, P.; Righetto, F.; Toffoletto, P.; Bazzato, G. On-line bioelectric impedance during haemodialysis: Monitoring of body fluids and cell membrane status. Nephrol. Dial Transplant. 1990, 5, 167–170. 30. Ellis, K.J.; Wong, W.W. Human hydrometry: Comparison of multifrequency bioelectrical impedance with 2H2O and bromine dilution. J. Appl. Physiol. 1998, 85, 1056–1062. 31. Jaffrin, M.Y.; Fenech, M.; Moreno, M.V.; Kieffer, R. Total body water measurement by a modification of the bioimpedance spectroscopy method. Med. Biol. Eng. Comput. 2006, 44, 873–882. 32. Matthie, J.R. Second generation mixture theory equation for estimating intracellular water using bioimpedance spectroscopy. J. Appl. Physiol. 2005, 99, 780–781. 33. Matthie, J.; Zarowitz, B.; De Lorenzo, A.; Andreoli, A.; Katzarski, K.; Pan, G.; Withers, P. Analytic assessment of the various bioimpedance methods used to estimate body water. J. Appl.Physiol. 1998, 84, 1801–1816. 34. Baarends, E.; Van Marken Lichtenbelt, W.; Wouters, E.; Schols, A. Body-water compartments measured by bio-electrical impedance spectroscopy in patients with chronic obstructive pulmonary disease. Clin. Nutr. 1998, 17, 15–22. 35. Cox-Reijven, P.; Soeters, P. Validation of bio-impedance spectroscopy: Effects of degree of obesity and ways of calculating volumes from measured resistance values. Int. J. Obes. 2000, 24, 271–280. 36. Earthman, C.P.; Matthie, J.R.; Reid, P.M.; Harper, I.T.; Ravussin, E.; Howell, W.H. A comparison of bioimpedance methods for detection of body cell mass change in hiv infection. J. Appl. Physiol. 2000, 88, 944–956. 37. Ward, L.; Elia, M.; Cornish, B. Potential errors in the application of mixture theory to multifrequency bioelectrical impedance analysis. Physiol. Meas. 1998, 19, 53–60. 38. Hannan, W.; Cowen, S.; Plester, C.; Fearon, K.; DeBeau, A. Comparison of bieimpedance spectroscopy and mu it if requency biei m pedance analysis for the assessment of extracellular and total body water in surgical patients. Clin. Sci. Sensors 2014, 14 Pierson, R.; Wang, J.; Colt, E.; Neumann, P. Body composition measurements in normal man: The potassium, sodium, sulfate and tritium spaces in 58 adults. J. Chronic Dis. 1982, 35, 419–428. 24. Patel, R.V.; Peterson, E.L.; Silverman, N.; Zarowitz, B.J. Estimation of total body and extracellular water in post-coronary artery bypass graft surgical patients using single and multiple frequency bioimpedance. Crit. Care Med. 1996, 24, 1824–1828. 25. Thomasset, A. Bio-electrical properties of tissues. Lyon Med. 1963, 209, 1325–135 26. Cole, K.S.; Cole, R.H. Dispersion and absorption in dielectrics i. Alternating current characteristics. J. Chem. Phys. 1941, 9, 341–351. Sensors 2014, 14 Sensors 2014, 14 10922 42. Ward, L.C.; Essex, T.; Cornish, B.H. Determination of cole parameters in multiple frequency bioelectrical impedance analysis using only the measurement of impedances. Physiol. Meas. 2006, 27, 839–950. 43. Cole, K.S. Membranes, Ions, and Impulses: A Chapter of Classical Biophysics; University of California Press: Oakland, CA, USA, 1968; Volume 1. 44. Xie, X.; Kolthoff, N.; Bärenholt, O.; Nielsen, S. Validation of a leg-to-leg bioimpedance analysis system in assessing body composition in postmenopausal women. Int. J. Obes. 1999, 23, 1079–1084. 45. Jebb, S.A.; Cole, T.J.; Doman, D.; Murgatroyd, P.R.; Prentice, A.M. Evaluation of the novel tanita body-fat analyser to measure body composition by comparison with a four-compartment model. Br. J. Nutr. 2000, 83, 115–122. 46. Utter, A.C.; Nieman, D.C.; Ward, A.N.; Butterworth, D.E. Use of the leg-to-leg bioelectrical impedance method in assessing body-composition change in obese women. Am. J. Clin. Nutr. 1999, 69, 603–607. 47. Deurenberg, P.; Deurenberg‐Yap, M. Validation of skinfold thickness and hand‐held impedance measurements for estimation of body fat percentage among singaporean chinese, malay and indian subjects. Asia Pac. J. Clin. Nutr. 2002, 11, 1–7. 48. Ghosh, S.; Meister, D.; Cowen, S.; Hannan, J.W.; Ferguson, A. Body composition at the bedside. Eur. J. Gastroenterol. Hepatol. 1997, 9, 783–788. 49. Buchholz, A.C.; Bartok, C.; Schoeller, D.A. The validity of bioelectrical impedance models in clinical populations. Nutr. Clin. Pract. 2004, 19, 433–446. 50. Nuñez, C.; Gallagher, D.; Visser, M.; Pi-Sunyer, F.X.; Wang, Z.; Heymsfield, S.B. Bioimpedance analysis: Evaluation of leg-to-leg system based on pressure contact footpad electrodes. Med. Sci. Sports Exerc. 1997, 29, 524–531. 51. Tanaka, N.I.; Miyatani, M.; Masuo, Y.; Fukunaga, T.; Kanehisa, H. Applicability of a segmental bioelectrical impedance analysis for predicting the whole body skeletal muscle volume. J. Appl. Physiol. 2007, 103, 1688–1695. 52. Baumgartner, R.N.; Chumlea, W.C.; Roche, A.F. Bioelectric impedance phase angle and body composition. Am. J. Clin. Nutr. 1988, 48, 16–23. 53. Thomas, B.; Cornish, B.; Ward, L.; Patterson, M. A comparison of segmental and wrist-to-ankle methodologies of bioimpedance analysis. Appl. Radiat. Isot. 1998, 49, 477–478. 54. Thomas, B.; Cornish, B.; Pattemore, M.; Jacobs, M.; Ward, L. A comparison of the whole-body and segmental methodologies of bioimpedance analysis. Acta Diabetol. 2003, 40, s236–s237. 55. Earthman, C.; Traughber, D.; Dobratz, J.; Howell, W. Bioimpedance spectroscopy for clinical assessment of fluid distribution and body cell mass. Nutr. Clin. Pract. 2007, 22, 389–405. 56. Moon, J. Sensors 2014, 14 1995, 89, 651–658. 39. Deurenberg, P.; Andreoli, A.; De Lorenzo, A. Multi-frequency bioelectrical impedance: A comparison between the cole-cole modelling and hanai equations with the classical impedance index approach. Ann. Hum. Biol. 1996, 23, 31–40. 40. Scharfetter, H.; Monif, M.; Laszlo, Z.; Lambauer, T.; Hutten, H.; Hinghofer-Szalkay, H. Effect of postural changes on the reliability of volume estimations from bioimpedance spectroscopy data. Kidney Int. 1997, 51, 1078–1087. 41. Ayllon, D.; Seoane, F.; Gil-Pita, R. Cole equation and parameter estimation from electrical bioimpedance spectroscopy measurements-a comparative study. In Proceedings of Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2009), Minneapolis, MN, USA, 3–6 September 2009; pp. 3779–3782. Sensors 2014, 14 Body composition in athletes and sports nutrition: An examination of the bioimpedance analysis technique. Eur. J. Clin. Nutr. 2013, 67, S54–S59. 57. Fogelholm, M.; van Marken, L.W. Comparison of body composition methods: A literature analysis. Eur. J. Clin. Nutr. 1997, 51, 495–503. 58. Fuller, N.; Sawyer, M.; Laskey, M.; Paxton, P.; Elia, M. Prediction of body composition in elderly men over 75 years of age. Ann. Hum. Biol. 1996, 23, 127–147. 10923 Sensors 2014, 14 59. Chumlea, W.; Schubert, C.; Sun, S.; Demerath, E.; Towne, B.; Siervogel, R. A review of body water status and the effects of age and body fatness in children and adults. J. Nutr. Health Aging 2007, 11, 111–118. 60. Coppini, L.Z.; Waitzberg, D.L.; Campos, A.C.L. Limitations and validation of bioelectrical impedance analysis in morbidly obese patients. Curr. Opin. Clin. Nutr. Metab. Care 2005, 8, 329–332. 61. Rutkove, S.B.; Aaron, R.; Shiffman, C.A. Localized bioimpedance analysis in the evaluation of neuromuscular disease. Muscle Nerve 2002, 25, 390–397. 62. Jaffrin, M.Y. Body composition determination by bioimpedance: An update. Curr. Opin. Clin. Nutr. Metab. Care 2009, 12, 482–486. 63. Thomas, B.; Cornish, B.; Ward, L. Bioelectrical impedance analysis for measurement of body fluid volumes: A review. J. Clin. Eng. 1992, 17, 505–510. 64. Piccoli, A.; Piazza, P.; Noventa, D.; Pillon, L.; Zaccaria, M. A new method for monitoring hydration at high altitude by bioimpedance analysis. Med. Sci. Sports Exerc. 1996, 28, 1517–1522. 65. Piccoli, A.; Rossi, B.; Pillon, L.; Bucciante, G. A new method for monitoring body fluid variation by bioimpedance analysis: The rxc graph. Kidney Int. 1994, 46, 534–539. 66. Piccoli, A.; Pillon, L.; Dumler, F. Impedance vector distribution by sex, race, body mass index, and age in the united states: Standard reference intervals as bivariate scores. Nutrition 2002, 18, 153–167. 67. Cox-Reijven, P.L.; van Kreel, B.; Soeters, P.B. Bioelectrical impedance measurements in patients with gastrointestinal disease: Validation of the spectrum approach and a comparison of different methods for screening for nutritional depletion. Am. J. Clin. Nutr. 2003, 78, 1111–1119. 68. Azevedo, Z.M.A.; Moore, D.C.B.C.; de Matos, F.A.A.; Fonseca, V.M.; Peixoto, M.V.M.; Gaspar-Elsas, M.I.C.; Santinoni, E.; dos Anjos, L.A.; Ramos, E.G. Bioelectrical impedance parameters in critically ill children: Importance of reactance and resistance. Clin. Nutr. 2013, 32, 824–829. 69. Haas, V.; Riedl, A.; Hofmann, T.; Nischan, A.; Burghardt, R.; Boschmann, M.; Klapp, B. Bioimpedance and bioimpedance vector analysis in patients with anorexia nervosa. Eur. Eat. Disord. Rev. Sensors 2014, 14 Sensors 2014, 14 10924 74. Yanovski, S.Z.; Hubbard, V.S.; Heymsfield, S.B.; Lukaski, H.C. Bioelectrical impedance analysis in body composition measurement: National institutes of health technology assessment conference statement. Am. J. Clin. Nutr. 1996, 64, 524S–532S. 74. Yanovski, S.Z.; Hubbard, V.S.; Heymsfield, S.B.; Lukaski, H.C. Bioelectrical impedance analysis in body composition measurement: National institutes of health technology assessment conference statement. Am. J. Clin. Nutr. 1996, 64, 524S–532S. 75. Sanchez, B.; Vandersteen, G.; Bragos, R.; Schoukens, J. Basics of broadband impedance spectroscopy measurements using periodic excitations. Meas. Sci. Technol. 2012, 23, doi:10.1088/0957-0233/23/10/105501. 76. Sanchez, B.; Bandarenka, A.S.; Vandersteen, G.; Schoukens, J.; Bragos, R. Novel approach of processing electrical bioimpedance data using differential impedance analysis. Med. Eng. Phys. 2013, 35, 1349–1357. 77. Sanchez, B.; Schoukens, J.; Bragos, R.; Vandersteen, G. Novel estimation of the electrical bioimpedance using the local polynomial method. Application to in vivo real-time myocardium tissue impedance characterization during the cardiac cycle. IEEE Trans. Biomed. Eng. 2011, 58, 3376–3385. 78. Sanchez, B.; Vandersteen, G.; Bragos, R.; Schoukens, J. Optimal multisine excitation design for broadband electrical impedance spectroscopy. Med. Eng. Phys. 2011, 22, doi:10.1088/ 0957-0233/22/11/115601. 79. Sanchez, B.; Rojas, C.R.; Vandersteen, G.; Bragos, R.; Schoukens, J. On the calculation of the d-optimal multisine excitation power spectrum for broadband impedance spectroscopy measurements. Med. Eng. Phys. 2012, 23, doi:10.1088/0957-0233/23/8/085702. 80. Sanchez, B.; Louarroudi, E.; Jorge, E.; Cinca, J.; Bragos, R.; Pintelon, R. A new measuring and identification approach for time-varying bioimpedance using multisine electrical impedance spectroscopy. Physiol. Meas. 2013, 34, 339–357. 81. Kyle, U.G.; Genton, L.; Slosman, D.O.; Pichard, C. Fat-free and fat mass percentiles in 5225 healthy subjects aged 15 to 98 years. Nutrition 2001, 17, 534–541. 82. Roubenoff, R.; Dallal, G.E.; Wilson, P. Predicting body fatness: The body mass index vs. estimation by bioelectrical impedance. Am. J. Public Health 1995, 85, 726–728. 83. Kyle, U.G.; Pichard, C. Dynamic assessment of fat-free mass during catabolism and recovery. Curr. Opin. Clin. Nutr. Metab. Care 2000, 3, 317–322. 84. Heitmann, B. Impedance: A valid method in assessment of body composition? Eur. J. Clin. Nutr. 1994, 48, 228–248. 85. Kyle, U.G.; Schutz, Y.; Dupertuis, Y.M.; Pichard, C. Body composition interpretation: Contributions of the fat-free mass index and the body fat mass index. Nutrition 2003, 19, 597–604. 86. Sun, S.S.; Chumlea, W.C.; Heymsfield, S.B.; Lukaski, H.C.; Schoeller, D.; Friedl, K.; Kuczmarski, R.J.; Flegal, K.M.; Johnson, C.L.; Hubbard, V.S. Sensors 2014, 14 2012, 20, 400–405. 70. Norman, K.; Stobäus, N.; Pirlich, M.; Bosy-Westphal, A. Bioelectrical phase angle and impedance vector analysis–clinical relevance and applicability of impedance parameters. Clin. Nutr. 2012, 31, 854–861. 71. Ward, L.C.; Heitmann, B.L. Re: ―Electrical maturation trajectory of human tissues identified by bioelectrical impedance vector analysis‖. Nutrition 2000, 16, 319–320. 72. Marini, E.; Sergi, G.; Succa, V.; Saragat, B.; Sarti, S.; Coin, A.; Manzato, E.; Buffa, R. Efficacy of specific bioelectrical impedance vector analysis (biva) for assessing body composition in the elderly. J. Nutr. Health Aging 2013, 17, 515–521. 73. Bracco, D.; Thiébaud, D.; Chioléro, R.L.; Landry, M.; Burckhardt, P.; Schutz, Y. Segmental body composition assessed by bioelectrical impedance analysis and dexa in humans. J. Appl. Physiol. 1996, 81, 2580–2587. Sensors 2014, 14 Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. Am. J. Clin. Nutr. 2003, 77, 331–340. 87. Deurenberg, P.; Van der Kooy, K.; Leenen, R.; Weststrate, J.; Seidell, J. Sex and age specific prediction formulas for estimating body composition from bioelectrical impedance: A cross-validation study. Int. J. Obes. 1991, 15, 17–25. Sensors 2014, 14 10925 88. Pichard, C.; Kyle, U.G.; Bracco, D.; Slosman, D.O.; Morabia, A.; Schutz, Y. Reference values of fat-free and fat masses by bioelectrical impedance analysis in 3393 healthy subjects. Nutrition 2000, 16, 245–254. 88. Pichard, C.; Kyle, U.G.; Bracco, D.; Slosman, D.O.; Morabia, A.; Schutz, Y. Reference values of fat-free and fat masses by bioelectrical impedance analysis in 3393 healthy subjects. Nutrition 2000, 16, 245–254. 89. Deurenberg, P.; Weststrate, J.A.; Seidell, J.C. Body mass index as a measure of body fatness: Age-and sex-specific prediction formulas. Br. J. Nutr. 1991, 65, 105–114. 90. Heitmann, B.L. Evaluation of body fat estimated from body mass index, skinfolds and impedance. A comparative study. Eur. J. Clin. Nutr. 1990, 44, 831–837. 91. Pichler, G.P.; Amouzadeh-Ghadikolai, O.; Leis, A.; Skrabal, F. A critical analysis of whole body bioimpedance spectroscopy (BIS) for the estimation of body compartments in health and disease. Med. Eng. Phys. 2013, 35, 616–625. 92. Sargent, J.A.; Gotch, F.A. Principles and biophysics of dialysis. In Replacement of Renal Function by Dialysis; Springer: Dordrecht, The Netherlands,1989; pp. 87–143. 93. Lukaski, H.C.; Johnson, P.E.; Bolonchuk, W.; Lykken, G. Assessment of fat-free mass using bioelectrical impedance measurements of the human body. Am. J. Clin. Nutr. 1985, 41, 810–817. 94. Sergi, G.; Bussolotto, M.; Perini, P.; Calliari, I.; Giantin, V.; Ceccon, A.; Scanferla, F.; Bressan, M.; Moschini, G.; Enzi, G. Accuracy of bioelectrical impedance analysis in estimation of extracellular space in healthy subjects and in fluid retention states. Ann. Nutr. Metab. 1994, 38, 158–165. 95. Deurenberg, P.; Tagliabue, A.; Schouten, F.J. Multi-frequency impedance for the prediction of extracellular water and total body water. Br. J. Nutr. 1995, 73, 349–358. 96. Schloerb, P.R.; Friis-Hansen, B.J.; Edelman, I.S.; Solomon, A.; Moore, F.D. The measurement of total body water in the human subject by deuterium oxide dilution: With a consideration of the dynamics of deuterium distribution 1. J. Clin. Investig. 1950, 29, 1296–1310. 97. Moissl, U.M.; Wabel, P.; Chamney, P.W.; Bosaeus, I.; Levin, N.W.; Bosy-Westphal, A.; Korth, O.; Müller, M.J.; Ellegård, L.; Malmros, V. Sensors 2014, 14 Body fluid volume determination via body composition spectroscopy in health and disease. Physiol. Meas. 2006, 27, 921–933. 98. Wagner, D.R.; Heyward, V.H. Techniques of body composition assessment: A review of laboratory and field methods. Res. Q. Exerc. Sport 1999, 70, 135–149. 99. Serrano, M.D.M.; de Espinosa, M.G.-M.; Zamorano, E.M. Relationship between physical measures of anthropometry and bioimpedance measures. In Handbook of Anthropometry; Springer: New York City, NY, USA, 2012; pp. 459–473. 100. Jaffrin, M.Y.; Bousbiat, S.; Dongmo, E. A comparison between two methods for measuring limb resistances with wrist and ankle electrodes. Med. Eng. Phys. 2011, 33, 943–949. 101. Diaz, E.; Villar, J.; Immink, M.; Gonzales, T. Bioimpedance or anthropometry? Eur. J. Clin. Nutr. 1989, 43, 129–137. 102. Ward, L.; Heitmann, B. Assessment of body composition by bioelectrical impedance analysis without the need for measurement of height. Clin. Nutr. 2001, 20, 21–26. 103. Mridha, S. A comparative study on body composition of male and female national level sub-junior volleyball players. Br. J. Sports Med. 2010, 44, i37–i38. 104. Kirchengast, S. Gender differences in body composition from childhood to old age: An evolutionary point of view. J. Life Sci. 2010, 2, 1–10. Sensors 2014, 14 10926 105. Fomon, S.J.; Haschke, F.; Ziegler, E.E.; Nelson, S.E. Body composition of reference children from birth to age 10 years. Am. J. Clin. Nutr. 1982, 35, 1169–1175. 106. Kim, J.H.; Choi, S.H.; Lim, S.; Kim, K.W.; Lim, J.Y.; Cho, N.H.; Park, K.S.; Jang, H.C. Assessment of appendicular skeletal muscle mass by bioimpedance in older community-dwelling korean adults. Arch. Gerontol. Geriatr. 2014, 58, 303–307. 107. Tengvall, M.; Ellegård, L.; Malmros, V.; Bosaeus, N.; Lissner, L.; Bosaeus, I. Body composition in the elderly: Reference values and bioelectrical impedance spectroscopy to predict total body skeletal muscle mass. Clin. Nutr. 2009, 28, 52–58. 108. Eisenmann, J.C.; Heelan, K.A.; Welk, G.J. Assessing body composition among 3‐ to 8‐year‐old children: Anthropometry, bia, and dxa. Obes. Res. 2004, 12, 1633–1640. 109. Buffa, R.; Floris, G.U.; Putzu, P.F.; Marini, E. Body composition variations in ageing. Coll. Antropol. 2011, 35, 259–265. 110. Deurenberg, P.; Deurenberg-Yap, M.; Schouten, F. Validity of total and segmental impedance measurements for prediction of body composition across ethnic population groups. Eur. J. Clin. Nutr. 2002, 56, 214–220. 111. Deurenberg, P.; Deurenberg-Yap, M. Validity of body composition methods across ethnic population groups. Acta Diabetol. 2003, 40, s246–s249. 112. Deurenberg, P.; Wolde-Gebriel, Z.; Schouten, F. Sensors 2014, 14 Sensors 2014, 14 Sensors 2014, 14 10927 121. Liang, M.; Norris, S. Effects of skin blood flow and temperature on bioelectric impedance after exercise. Med. Sci. Sports Exerc. 1993, 25, 1231–1239. 122. Roos, A.; Westendorp, R.; Frölich, M.; Meinders, A. Tetrapolar body impedance is influenced by body posture and plasma sodium concentration. Eur. J. Clin. Nutr. 1992, 46, 53–60. 123. Buendía, R.; Bogónez-Franco, P.; Nescolarde, L.; Seoane, F. Influence of electrode mismatch on cole parameter estimation from total right side electrical bioimpedance spectroscopy measurements. Med. Eng. Phys. 2012, 34, 1024–1028. 124. Shiffman, C. Adverse effects of near current-electrode placement in non-invasive bio-impedance measurements. Physiol. Meas. 2013, 34, 1513–1545. 125. Scharfetter, H.; Hartinger, P.; Hinghofer-Szalkay, H.; Hutten, H. A model of artefacts produced by stray capacitance during whole body or segmental bioimpedance spectroscopy. Physiol. Meas. 1998, 19, 247–261. 126. Kondrup, J.; Allison, S.; Elia, M.; Vellas, B.; Plauth, M. Espen guidelines for nutrition screening 2002. Clin. Nutr. 2003, 22, 415–421. 127. Thibault, R.; Genton, L.; Pichard, C. Body composition: Why, when and for who? Clin. Nutr. 2012, 31, 435–447. 128. Kuczmarski, R.J. Bioelectrical impedance analysis measurements as part of a national nutrition survey. Am. J. Clin. Nutr. 1996, 64, 453S–458S. 129. Toso, S.; Piccoli, A.; Gusella, M.; Menon, D.; Bononi, A.; Crepaldi, G.; Ferrazzi, E. Altered tissue electric properties in lung cancer patients as detected by bioelectric impedance vector analysis. Nutrition 2000, 16, 120–124. 130. Zlochiver, S.; Arad, M.; Radai, M.; Barak-Shinar, D.; Krief, H.; Engelman, T.; Ben-Yehuda, R.; Adunsky, A.; Abboud, S. A portable bio-impedance system for monitoring lung resistivity. Med. Eng. Phys. 2007, 29, 93–100. 131. Bracco, D.; Revelly, J.-P.; Berger, M.M.; Chiolero, R.L. Bedside determination of fluid accumulation after cardiac surgery using segmental bioelectrical impedance. Crit. Care Med. 1998, 26, 1065–1070. 132. Hoyle, G.; Chua, M.; Soiza, R. Volaemic assessment of the elderly hyponatraemic patient: Reliability of clinical assessment and validation of bioelectrical impedance analysis. QJM 2011, 104, 35–39. 133. Cumming, K.; Hoyle, G.; Hutchison, J.; Soiza, R.L. Bioelectrical impedance analysis is more accurate than clinical examination in determining the volaemic status of elderly patients with fragility fracture and hyponatraemia. J. Nutr. Health Aging in press. 134. Chen, Y.-C.; Chen, H.-H.; Yeh, J.-C.; Chen, S.-Y. Adjusting dry weight by extracellular volume and body composition in hemodialysis patients. Nephron 2002, 92, 91–96. 135. Chamney, P.W.; Krämer, M.; Rode, C.; Kleinekofort, W.; Wizemann, V. Sensors 2014, 14 Validity of predicted total body water and extracellular water using multifrequency bioelectrical impedance in an ethiopian population. Ann. Nutr. Metab. 1995, 39, 234–241. 113. Kotler, D.P.; Burastero, S.; Wang, J.; Pierson, R. Prediction of body cell mass, fat-free mass, and total body water with bioelectrical impedance analysis: Effects of race, sex, and disease. Am. J. Clin. Nutr. 1996, 64, 489S–497S. 114. Schulz, H.; Teske, D.; Penven, D.; Tomczak, J. Fat-free mass from two prediction equations for bioelectrical impedance analysis in a large german population compared with values in swiss and american adults: Reasons for a biadata project. Nutrition 2006, 22, 973–975. 115. Siváková, D.; Vondrová, D.; Valkovič, P.; Cvíčelová, M.; Danková, Z.; Luptáková, L. Bioelectrical impedance vector analysis (biva) in slovak population: Application in a clinical sample. Cent. Eur. J. Biol. 2013, 8, 1094–1101. 116. Nigam, P.; Misra, A.; Colles, S.L. Comparison of dexa-derived body fat measurement to two race-specific bioelectrical impedance equations in healthy indians. Diabetes Metab. Syndr. 2013, 7, 72–77. 117. Saragat, B.; Buffa, R.; Mereu, E.; De Rui, M.; Coin, A.; Sergi, G.; Marini, E. Specific bioelectrical impedance vector reference values for assessing body composition in the italian elderly. Exp. Gerontol. 2014, 50, 52–56. 118. Zhu, F.; Schneditz, D.; Wang, E.; Levin, N.W. Dynamics of segmental extracellular volumes during changes in body position by bioimpedance analysis. J. Appl. Physiol. 1998, 85, 497–504. 119. Schols, A.; Dingemans, A.; Soeters, P.; Wouters, E. Within-day variation of bioelectrical resistance measurements in patients with chronic obstructive pulmonary disease. Clin. Nutr. 1990, 9, 266–271. 120. Kushner, R.F.; Gudivaka, R.; Schoeller, D.A. Clinical characteristics influencing bioelectrical impedance analysis measurements. Am. J. Clin. Nutr. 1996, 64, 423S–427S. Sensors 2014, 14 A new technique for establishing dry weight in hemodialysis patients via whole body bioimpedance. Kidney Int. 2002, 61, 2250–2258. Sensors 2014, 14 Sensors 2014, 14 10928 136. Zhu, F.; Kuhlman, M.; Kotanko, P.; Handelman, G.; Leonard, E.; Levin, N. A device for monitoring hydration state in hemodialysis patients using a calf bioimpedance technique. In Proceedings of the 13th International Conference on Electrical Bioimpedance and the 8th Conference on Electrical Impedance Tomography, Graz, Austria, 29 August–2 September 2007; pp. 775–778. 137. Zhu, F.; Kuhlmann, M.; Kotanko, P.; Seibert, E.; Leonard, E.; Levin, N. A method for the estimation of hydration state during hemodialysis using a calf bioimpedance technique. Physiol. Meas. 2008, 29, S503–S516. 138. Seibert, E.; Mueller, S.G.; Fries, P.; Pattmoeller, J.; Kuss, O.; Heine, G.H.; Girndt, M.; Schneider, G.; Kotanko, P.; Zhu, F. Calf bioimpedance spectroscopy for determination of dry weight in hemodialysis patients: Effects on hypertension and left ventricular hypertrophy. Kidney Blood Press. Res. 2013, 37, 58–67. 139. Atilano, X.; Luis Miguel, J.; Martínez, J.; Sánchez, R.; Selgas, R. Bioimpedance vector analysis as a tool for determination and adjustment of dry weight in hemodialysis patients. Kidney Res. Clin. Prac. 2012, 31, A17–A18. 140. Buffa, R.; Mereu, R.; Putzu, P.; Floris, G.; Marini, E. Bioelectrical impedance vector analysis detects low body cell mass and dehydration in patients with alzheimer’s disease. J. Nutr. Health Aging 2010, 14, 823–827. 141. Moreno, M.V.; Djeddi, D.-D.; Jaffrin, M.Y. Assessment of body composition in adolescent subjects with anorexia nervosa by bioimpedance. Med. Eng. Phys. 2008, 30, 783–791. 142. Sillanpää, E.; Häkkinen, A.; Häkkinen, K. Body composition changes by dxa, bia and skinfolds during exercise training in women. Eur. J. Appl. Physiol. 2013, 113, 2331–2341. 143. Paton, N.I.; Elia, M.; Jennings, G.; Ward, L.C.; Griffin, G.E. Bioelectrical impedance analysis in human immunodeficiency virus-infected patients: Comparison of single frequency with multifrequency, spectroscopy, and other novel approaches. Nutrition 1998, 14, 658–666. 144. Libraty, D.H.; Endy, T.P.; Kalayanarooj, S.; Chansiriwongs, W.; Nisalak, A.; Green, S.; Ennis, F.A.; Rothman, A.L. Assessment of body fluid compartment volumes by multifrequency bioelectrical impedance spectroscopy in children with dengue. Trans. R. Soc. Trop. Med. Hyg. 2002, 96, 295–299. 145. Simons, J.; Schols, A.; Westerterp, K.; Ten Velde, G.; Woliters, E. Bioelectrical impedance analysis to assess changes intotal body water in patients with cancer. Clin. Nutr. 1999, 18, 35–39. 145. Simons, J.; Schols, A.; Westerterp, K.; Ten Velde, G.; Woliters, E. Bioelectrical impedance analysis to assess changes intotal body water in patients with cancer. Clin. Nutr. 1999, 18, 35–39. 146. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/). Sensors 2014, 14 Ibrahim, F.; Taib, M.N.; Abas, W.A.B.W.; Guan, C.C.; Sulaiman, S. A novel approach to classify risk in dengue hemorrhagic fever (dhf) using bioelectrical impedance analysis (bia). IEEE Trans. Instrum. Meas. 2005, 54, 237–244. 146. Ibrahim, F.; Taib, M.N.; Abas, W.A.B.W.; Guan, C.C.; Sulaiman, S. A novel approach to classify risk in dengue hemorrhagic fever (dhf) using bioelectrical impedance analysis (bia). IEEE Trans. Instrum. Meas. 2005, 54, 237–244. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/). © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).
https://openalex.org/W3028776652	http://old.scielo.br/pdf/rbepid/v23/en_1980-5497-rbepid-23-e200046.pdf	English	null	Síndrome metabólica: inatividade física e desigualdades socioeconômicas entre idosos brasileiros não institucionalizados	Revista brasileira de epidemiologia	2,020	cc-by	8,772	ORIGINAL ARTICLE / ARTIGO ORIGINAL Ana Cristina de Oliveira CostaI , Yeda Aparecida de Oliveira DuarteII , Fabíola Bof de AndradeIII ABSTRACT: Objective: Evaluate the association between Metabolic Syndrome (MetS), physical activity and socioeconomic conditions among non-institutionalized elderly individuals. Methodology: Cross-sectional study with, elderly individuals (≥ 60) living in the city of São Paulo. MetS was evaluated by means of the National Cholesterol Education Program criteria, the Adult Treatment Panel III. Descriptive and bivariate analyses were performed, followed by multiple logistic regression with a 5% significance level. An attributable fraction (AF) and a proportional attributable fraction (PAF) were calculated in relation to physical activity. The magnitude of the socioeconomic inequalities was evaluated using the Slope Index of Inequality (SII) and the Relative Index of Inequality (RII). Results: The prevalence of MetS was 40.1%, and 23.3% of the individuals had at least one MetS’ component. Physically inactive elderly had higher chances of having MetS. The prevalence of MetS was higher among those with lower education levels in both absolute and relative terms. AF and PAF were significant among the inactive individuals and for the total population. Conclusion: This study demonstrated that physical activity and schooling are significantly associated with MetS, highlighting the importance of these factors for the control of this syndrome. Keywords: Metabolic syndrome. Physical activity. Socioeconomic Factors. Elderly. IPost -graduate Program in Collecteive Health. Instituto de Pesquisa René Rachou, Study Center in Public Health and Aging, Fundação Oswaldo Cruz – Belo Horizonte (MG), Brazil. IISchool of Nursing, Universidade de São Paulo – São Paulo (SP), Brazil. IIIInstituto de Pesquisa René Rachou, Fundação Oswaldo Cruz – Belo Horizonte (MG), Brazil. Corresponding author: Fabíola Bof de Andrade. Avenida Augusto de Lima, 1.715, Barro Preto, CEP: 30.190-002, Belo Horizonte, MG, Brazil. E-mail: fabiola.bof@fiocruz.br Conflict of interests: nothing to declare – Financial support: none. REV BRAS EPIDEMIOL 2020; 23: E200046 1 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. RESUMO: Objetivo: Avaliar a associação da síndrome metabólica (SM) com a atividade física e as condições socioeconômicas entre idosos não institucionalizados. Metodologia: Estudo transversal com idosos (≥ 60) não institucionalizados e residentes na cidade de São Paulo (SP). A SM foi classificada com base nos critérios da National Cholesterol Education Program Adult Treatment Panel III. Realizou-se analise descritiva e bivariada seguida por regressão logística múltipla com nível de significância de 5%. Calcularam-se a fração atribuível (FA) e a fração atribuível proporcional (FAP) e determinou-se a magnitude das desigualdades por meio do índice absoluto de desigualdade e pelo índice relativo de desigualdade. Resultados: A prevalência de SM foi de 40,1%, e 23,3% dos idosos apresentavam pelo menos um componente da síndrome. A chance de SM foi maior entre os idosos fisicamente inativos. Idosos menos escolarizados apresentaram prevalências de SM significativamente maiores em termos absolutos e relativos. As FA e FAP entre os inativos e na população foram significativos. Conclusão: Este estudo demonstrou que a prática de atividade física e a escolaridade são fatores significativamente associados à SM, reforçando a importância desses fatores para o controle dessa síndrome. Palavras-chave: Síndrome metabólica. Atividade física. Fatores socioeconômicos. Idoso. REV BRAS EPIDEMIOL 2020; 23: E200046 INTRODUCTION Metabolic syndrome (MetS) is a challenging clinical condition for public health1. It is characterized by a set of pathophysiological changes that simultaneously act to increase the risk of developing cardiovascular diseases (CVD), type 2 diabetes mellitus (DM)2 and death (NCEP-ATP III 2001). Individuals affected by this syndrome are twice as likely to develop CVD and five times more likely to develop DM compared to those without MetS1. The elderly is the group with the highest prevalence of unfavorable cardiovascu- lar outcomes3-5. Recent studies show an increasing evolution of MetS, with prevalence ranging from 25% in countries in the Middle East6 to 50.2% in India7. In Brazil, the review presented by Vidigal et al.8 showed a prevalence of 29.6% (14.9–65.3%). Recently, Vieira et al.9 found a prevalence of 32% among adults and elderly residents in the city of São Paulo (SP).fi MetS is a widely studied condition, but there are different criteria that define it1,10,11. Among them, the most widely used is the National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATP III) 200111, as it is more accessible for clinical practice and epidemiological research12,13 and because it is considered to be a better risk predictor of car- diovascular events when compared to other criteria4,14. In Brazil, this is the criterion pro- posed by the I Brazilian Guideline for the Diagnosis and Treatment of Metabolic Syndrome15. In this criterion, the syndrome is defined as the presence of three or more of the following criteria: central obesity, arterial hypertension, insulin resistance, low high-density lipopro- tein (HDL) and high triglyceride (TG). These components come from modifiable risk fac- tors (ex., sedentary behavior, unhealthy eating, physical inactivity) that result in the accu- mulation of adipose tissue and the progression of obesity16. 2 REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Regular activity can reduce the chance of developing MetS by 31%17. However, an increase in physical inactivity from the age of 4018,19 and the lower adherence to programs to encourage physical activity among the elderly17,20 contribute to the expansion of the prevalence of MetS in this group21. Regular activity can reduce the chance of developing MetS by 31%17. METHODOLOGY A cross-sectional study was carried out based on data from the Health, Well-Being and Aging (Saúde, Bem-Estar e Envelhecimento - SABE) 2010 study. g g y The SABE study was started in 2000 as a multi-center study developed in seven countries in Latin America (Argentina, Cuba, Uruguay, Barbados, Mexico, Chile and Brazil) and the Caribbean, with the aim of evaluating and comparing the health condition of people 60 years old or older at different stages of aging. In Brazil, the study was conducted in the city of São Paulo, with a repre- sentative sample of community-dwelling people aged 60 or over. In 2006, only in Brazil, SABE was transformed into a longitudinal study of multiple cohorts with follow-up at intervals of approx- imately five years. Information about the study and sampling has been previously published32. In 2010, 1,333 elderly people were interviewed and underwent physical examinations. This study included 1,201 individuals and 143 (9.9%) were excluded due to the lack of infor- mation on any of the variables of interest. There was no significant difference between those included in this study and those excluded in relation to sex, age, education level and number of self-reported diseases. The SABE study was started in 2000 as a multi-center study developed in seven countries in Latin America (Argentina, Cuba, Uruguay, Barbados, Mexico, Chile and Brazil) and the Caribbean, with the aim of evaluating and comparing the health condition of people 60 years old or older at different stages of aging. In Brazil, the study was conducted in the city of São Paulo, with a repre- sentative sample of community-dwelling people aged 60 or over. In 2006, only in Brazil, SABE was transformed into a longitudinal study of multiple cohorts with follow-up at intervals of approx- imately five years. Information about the study and sampling has been previously published32. In 2010, 1,333 elderly people were interviewed and underwent physical examinations. This study included 1,201 individuals and 143 (9.9%) were excluded due to the lack of infor- mation on any of the variables of interest. There was no significant difference between those included in this study and those excluded in relation to sex, age, education level and number of self-reported diseases. INTRODUCTION However, an increase in physical inactivity from the age of 4018,19 and the lower adherence to programs to encourage physical activity among the elderly17,20 contribute to the expansion of the prevalence of MetS in this group21.f In addition to factors related to lifestyle, different studies have shown that the occurrence of MetS is associated with socioeconomic inequalities8,22-27. These are characterized by the higher prevalence of the syndrome and its components among individuals belonging to the groups in the worst socioeconomic conditions when compared to those in the best con- ditions8,25. When these differences are unjust and avoidable, they are defined as inequities28. fi The most recent report on health conditions issued by the World Health Organization (WHO) in 2018 highlights that the chronic noncommunicable diseases (NCDs) that make up MetS are increasing, and it highlights the need for studies on the risk factors associated with these conditions in order for control measures to be adopted29. Although the association between MetS, physical activity30 and socioeconomic conditions8 is quite often described, an analysis of the literature reveals that there is a gap in the assessment of factors associated with MetS in developing countries, especially among the elderly, who are in the fastest grow- ing group in the world population and have the highest incidence of the syndrome1,8,25,31. Thus, this study was carried out with the objective of evaluating the association of MetS with physical activity and socioeconomic conditions among non-institutionalized Brazilian elderly. REV BRAS EPIDEMIOL 2020; 23: E200046 INDEPENDENT VARIABLES The variables considered in the study were: sociodemographic factors (age [60 to 64, 65 to 69, 70 to 74 and 75 or more], sex, marital status [with marital relationship including married or cohabited individuals and no marital relationship including divorced, separated, and sin- gle individuals] and education [0 to 3 years (insufficient education), 4 to 7 years (incomplete elementary school) and 8 years or more (completed elementary school or more)])33,34 , health conditions (number of NCDs, depressive symptoms and self-rated health status (good (good and very good) and bad (normal, bad and very bad))) and health behavior (physical activity). The practice of physical activity was assessed using the reduced version of the International Physical Activity Questionnaire (IPAQ)35. This instrument aims to verify what types of phys- ical activity people do as part of their daily lives. The questions relate to their time spent doing physical activity in the last week, including physical activity carried out at work, com- muting from place to place, physical activity for leisure, for sport, for exercise or as part of their activities at home or in the garden. Elderly people who performed at least 150 min- utes a week of moderate physical activity or 75 minutes a week of vigorous activity were considered to be active, according to WHO recommendations36. The number of NCDs was assessed based on the answer to the following question: Has a doctor or nurse ever told you that you have (name of the disease)? The diseases ques- tioned were: heart disease (congestive disease, coronary heart disease or infarction), joint disease (arthritis, rheumatism or arthrosis), stroke, diabetes, high blood pressure and chronic obstructive pulmonary disease. Depressive symptoms were assessed using the Geriatric Depression Scale in an abbre- viated format. Individuals with six or more points on the scale were considered to have depressive symptoms37. • abdominal obesity measured by waist circumference (≥ 102 cm for men; ≥ 88 cm for women); • TG (≥ 150 mg/dL); • reduction of HDL-cholesterol (<40 mg/dL for men; <50 mg/dL for women); • increased systolic blood pressure (SBP) (≥ 130 mmHg) and/or diastolic blood pressure (≥ 85 mmHg); • increased systolic blood pressure (SBP) (≥ 130 mmHg) and/or diastolic blood pressure (≥ 85 mmHg); • fasting blood glucose (≥ 110 mg/dL). REV BRAS EPIDEMIOL 2020; 23: E200046 DEPENDENT VARIABLE MetS was classified based on the criteria adopted by the NCEP ATP III11. According to this criterion, three or more of the following characteristics are required to determine if it’s present: 3 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. • abdominal obesity measured by waist circumference (≥ 102 cm for men; ≥ 88 cm for women); • TG (≥ 150 mg/dL); ANALYSIS OF THE DATA A descriptive analysis of the sample was performed according to the outcome and all independent variables, followed by a bivariate analysis between MetS and the independent variables. The association between categorical variables was tested using the chi-square test with a Rao-Scott correction38. 4 REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly All variables that presented p < 0.20 in the bivariate analysis were included in the multiple logistic regression model. The results were presented using odds ratios and respective 95% con- fidence intervals (CI). Based on the results of the multiple regression model, the attributable fraction (AF) and proportional attributable fraction (PAF) relating to the practice of physical activity were calculated as previously proposed39-41. The first, when expressed as a percentage, indicates the proportion of cases that can be attributed to exposure in exposed individuals and in the population. The second represents the proportion of cases of MetS that could have been avoided in those exposed and in the population, if the exposure were eliminated or changed. The magnitude of inequalities was assessed by means of absolute and relative measures using, respectively, the absolute index of inequality (slope index of inequality - SII) and the relative index of inequality (RII)42,43, with education used as a measure of socioeconomic position. Each socioeconomic group was assigned a value that corresponds to the midpoint of the cumulative distribution of the socioeconomic position measure43. The SII and RII indices were obtained by regressing the health variable over a relative position score, which was obtained by measuring the socioeconomic position in increasing order, from the worst (score equal to zero) to the best socioeconomic situation (score equal to one). The SII is the absolute difference in the prevalence of MetS among the elderly with the highest socioeco- nomic position (higher level of education) and those with the lowest socioeconomic sta- tus (lower level of education). The RII is the prevalence ratio (relative inequality) between groups in the higher and lower socioeconomic position. An SII value below zero and an RII value less than one indicate that the prevalence of MetS is higher among the group in the lowest socioeconomic position. All analyses were performed using the Stata 13.0 program (Stata Corporation, College Station, TX, United States) using the command survey, which allows for the consideration of the complex structure of the sampling process. ANALYSIS OF THE DATA All analyses were performed considering the sample weights. ETHICAL ASPECTS The SABE study was approved by the Research Ethics Committee of the School of Public Health of the University of Sao Paulo. REV BRAS EPIDEMIOL 2020; 23: E200046 RESULTS The sample consisted of elderly people aged 60 or over, with a mean age of 70.32 years (95%CI 69.03 - 71.62), with the majority of them being female, in a marital relationship, with 0 to 3 years of study, and regular practice of physical activity. The prevalence of MetS was 40.1% (95%CI 37.1 - 43.2). Regarding the components associ- ated with the diagnosis of MetS, 23.3% (95%CI 20.8 - 25.9) of the elderly had at least one con- dition, with the increase in SBP being the most prevalent (66.1%; 95%CI 62.8 - 69.2) (Table 1). 5 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. Through the bivariate analysis, it was found that MetS was associated with two sociode- mographic conditions (age and sex). The prevalence was higher among the elderly who did not perform physical activity and who had a greater number of chronic diseases (Table 2). Based on the multiple regression model, it was found that the chance of having MetS was significantly lower among older individuals and among those who were physically active. Physically active individuals were 33% less likely to have MetS when compared to those who were not active. The results showed that more educated individuals had significantly less chances of MetS when compared to those with 0–3 years of study (Table 3). There was no interaction between physical activity and education. Through the analysis of inequalities using the absolute and relative inequality indices, it was observed that less educated individuals had significantly higher prevalence of MetS in absolute and relative terms [SII = -0.12 (95%CI -0.231; -0.017) and RII = 0.73 (95%CI 0.535 - 0.931)]. Figure 1 shows the AFs and PAFs for the population and among those exposed. From the AF analysis, it can be seen that the difference in the prevalence of MetS between the active and inactive elderly in the population is 3.71% (95%CI 1.53 - 5.88). Among the inactive, the AF was 8.75% (95%CI 3.59 - 13.87). With regard to PAF, 9.27% (95%CI 3.56 - 14.65) of cases of MetS could have been prevented if all of the elderly in the population were active. Among the inactive, 19.9% (95%CI 8.33 - 28.94) of the cases would have been prevented. 6 Table 1. Prevalence of the components of metabolic syndrome (MetS) and the number of components. RESULTS % (95%CI) Components of MetS Altered waist circumference 50.4 (47.1 - 53.7) Elevated triglyceride 30.1 (28.5 - 33.8) Altered blood glucose 28.6 (25.7 - 31.6) Altered HDL 42.9 (39.6 - 46.3) Altered DBP 34.0 (30.7 - 37.5) Altered SBP 66.1 (62.8 - 69.2) Number of METS components 0 9.7 (7.8 – 12.0) 1 23.3 (20.8 – 25.9) 2 26.9 (24.1 – 29.9) 3 24.7 (22.1 – 27.5) 4 10.6 (8.9 – 12.5) 5 4.8 (3.7 – 6.1) 95%CI: 95% confidence interval; HDL: high density lipoprotein; DBP: diastolic blood pressure; SBP: systolic blood pressure. REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Table 2. Descriptive and bivariate analysis according to the independent variables. Total Metabolic syndrome % (95%CI) % (95%CI) Sociodemographic Sex Male 39.3 (36.3 – 42.3) 35.7 (31.4 – 40.3)a Female 60.7 (57.7 – 63.7) 42.9 (38.4 – 47.6) Age 60–64 31.6 (23.9 – 40.4) 50.0 (44.9 – 55.1)c 65–69 22.8 (15.8 – 31.8) 40.6 (33.9 – 47.7) 70–74 18.2 (14.9 – 22.1) 34.0 (28.7 – 39.7) 75 + 27.4 (21.7 – 33.8) 32.3 (27.9 – 36.9) Marital status No marital relationship 45.5 (41.6 – 49.4) 37.0 (31.6 – 42.8) Marital relationship 54.5 (50.6 – 58.4) 42.7 (38.3 – 47.1) Education level 0-3 years 35.9 (31.2 – 41.0) 43.0 (38.2 – 47.9) 4–7 years 37.0 (33.6 – 40.5) 38.5 (34.2 – 42.9) 8 + years 27.1 (22.2 – 32.6) 38.5 (31.7 – 45.8) Health behaviors Physical activity Active 58.3 (53.7 – 62.8) 36.3 (32.3 – 40.5)c Not active 41.7 (37.2 – 46.3) 45.5 (41.5 – 49.5) Health conditions Depressive symptoms No 81.4 (78.3 – 84.2) 37.8 (34.6 – 41.1)a Yes 18.6 (15.8 – 21.7) 47.1 (39.9 – 54.4) Number of diseases (average) 1.63 (1.54 – 1.72) 1.94 (1.80 – 2.08)c Health self-assessment Good 49.4 (45.4 – 53.4) 35.6 (31.5 – 40.0)b Poor 50.6 (46.6 – 54.6) 44.6 (40.3 – 49.0) 95%CI: 95% confidence interval; ap <0.05; bp<0.01; cp<0.001. Table 2. Descriptive and bivariate analysis according to the independent variables. 95%CI: 95% confidence interval; ap <0.05; bp<0.01; cp<0.001. COSTA, A.C.O. ET AL. DISCUSSION This study used a representative sample of non-institutionalized elderly to evaluate the prevalence of MetS based on the criteria of the NCEP ATP III11 and to evaluate its asso- ciation with physical activity and socioeconomic conditions. It was observed that 40% of the population met the criterion for MetS. The main findings demonstrate the existence of socioeconomic inequalities related to MetS and the significance of physical activity as a preventive factor, as demonstrated by the AFs, data hitherto unavailable in the Brazilian lit- erature for the elderly population. The prevalence found in this study is high, but lower than that observed among elderly people from countries that, like Brazil, are in the process of social and economic devel- opment. In Mexico the prevalence was 72.9%, using the criteria of the American Heart Association/National Heart, Lung and Blood Institute44. In Ecuador, it was observed that 66.0% of women and 47.1% of men presented the syndrome, according to the criteria of the Joint Interim Statement45. On the other hand, in China, a prevalence of 22.8% was demonstrated according to the NCEP ATP III46 criterion. When compared to the preva- lence in developed countries, it appears that the prevalence observed in this study was lower than that observed among women and men aged 60–69 years in the United States, 55.4 and 59.3%, respectively47. However, as previously mentioned, it should be noted that the differ- ences between the estimates are also associated with the different criteria used in the studies. 8 REV BRAS EPIDEMIOL 2020; 23: E200046 Table 3. Multiple logistic regression model for factors associated with metabolic syndrome. Metabolic syndrome OR (95%CI) p Sex (male) * Female 1.25 (0.90 – 1.74) 0.162 Age (60-64 years old) * 65–69 0.52 (0.35 – 0.77) 0.001 70–74 0.37 (0.26 – 0.52) < 0.001 75 + 0.29 (0.21 – 0.40) < 0.001 Education level (0-3 years)* 4-7years 0.74 (0.56 – 0.98) 0.036 8 + years 0.69 (0.48 – 0.96) 0.033 Physical activity (no)* Active 0.67 (0.53 – 0.84) 0.001 Number of diseases 1.54 (1.37 – 1.73) < 0.001 95%CI: 95% confidence interval; OR: odds ratio; *reference category. e 3. Multiple logistic regression model for factors associated with metabolic syndrome. REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly There is no national estimate of the prevalence of MetS in the literature, especially for the elderly population. REV BRAS EPIDEMIOL 2020; 23: E200046 DISCUSSION The available studies report prevalence rates between 3048 and 69.8%49. The Brazilian population survey that investigated the prevalence of MetS (14.2%) did not estimate the specific prevalence for the elderly population and defined the presence of two or more conditions as MetS, which is different from this study50. Physical inactivity is the third major risk factor for mortality among the factors proposed by the WHO for the control of CVD51. Wu et al.17 verified that regular activity can reduce the chance of developing METS by 31%. Corroborating these findings, in this study, active elderly presented a 33% less chance of having MetS when compared to non-active elderly individuals. In addition, this study has made an advancement by demonstrating the impact of physical activity as a preventive measure. Attributable fraction (AF) Proportional attributable fraction (PAF) Percentage of MetS avoided if all of the elderly were active (95%CI) Elderly 0 5 10 15 20 25 30 All Inactive people All Inactive people Attributable fraction (AF) Elderly Proportional attributable fraction (PAF) Inactive people Percentage of MetS avoided if all of the elderly were active (95%CI) CI95%: 95% Confidence Interval CI95%: 95% Confidence Interval Figure 1. Attributable fraction and proportional attributable fraction to physical inactivity in the prevalence of metabolic syndrome (MetS). Figure 1. Attributable fraction and proportional attributable fraction to physical inactivity in the prevalence of metabolic syndrome (MetS). 9 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. Estimates of proportional AFs demonstrate that prevention strategies aimed at the popula- tion and inactive individuals could prevent around 9 and 20% of MetS cases, respectively, if these people were active. Physical activity as a way of preventing NCDs is extremely important, and for this reason, the reduction of physical inactivity was included as one of the goals of the Strategic Action Plan for Coping with Chronic Non-Communicable Diseases in Brazil, which was imple- mented by the Ministry of Health, and aims to reduce this risk factor by 10% by the year 202552,53. With regard to socioeconomic conditions, results from different studies are consistent in showing a higher prevalence of MetS among people with lower income54 and educational level55 and worse occupations22. Similarly, in this study, it was observed that elderly people with fewer years of schooling were more likely to have MetS. DISCUSSION Additionally, this research con- tributes to the literature by demonstrating that the magnitude of socioeconomic inequalities was significant from an absolute and relative point of view. Thus, this finding reinforces the proposal of other authors who suggest that socioeconomic condition should be included as a modifiable risk factor in local and global health strategies and policies51 and reaffirms the theory that there is a socioeconomic and cultural gradient in determining chronic diseases52,53. According to Stringhini et al.51, the impact of socioeconomic conditions on health is sim- ilar to that of the six key risk factors adopted by the WHO for the prevention of chronic diseases. It suggests the importance of addressing not only proximal risk factors, but also structural solutions, which involve investment in early childhood education and work incen- tive programs, which is an economic mechanism to reduce health inequalities. Literature analysis shows a positive association between MetS and age in the adult popu- lation8,56-59. However, among the elderly, studies show that there is no association with age60 or there is a reduction in prevalence with increasing age61,62. Longitudinal studies have found that the prevalence of MetS increased with age up to the age range of 60 to 69 years, with a reduction in the older age groups and absence of differences between these age groups62. Khosravi-Boroujeni et al.61 observed an increase in the prevalence of MetS with increasing age up to 75 years, when a decline was observed. In addition, longitudinal analyses of the Japanese population have shown that there is a cohort effect related to the prevalence of MetS in which the younger cohorts have higher prevalences62. As such, the results of this study are in accordance with the literature. The age groups above 65 years are less likely to have MetS when compared to the elderly aged 60 to 64 years old, and there was no gradient between higher age groups. Among the reasons for the lower chance of MetS in the older age groups, the reduction in appetite and abdominal obesity associated with age have been proposed61,63,64. In addition, premature death due to condi- tions associated with MetS would lead to the maintenance of a healthier elderly cohort54,61. REV BRAS EPIDEMIOL 2020; 23: E200046 REFERENCES 4. Athyros VG, Ganotakis ES, Elisaf METS, Liberopoulos EN, Goudevenos IA, Karagiannis A. Prevalence of vascular disease in metabolic syndrome using three proposed definitions. Int J Cardiol [Internet] 2007 [acessado em 6 jan. 2017]; 117(2): 204-10. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/16854482 https://doi.org/10.1016/j.ijcard.2006.04.078 5. He Y, Jiang B, Wang J, Feng K, Chang Q, Fan L, et al. Prevalence of the metabolic syndrome and its relation to cardiovascular disease in an elderly Chinese population. J Am Coll Cardiol [Internet]. 2006 [acessado em 6 jan. 2017]; 47(8): 1588-94. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/16630995 https://doi. org/10.1016/j.jacc.2005.11.074 1. Alberti KGMM, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the Metabolic Syndrome A Joint Interim Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation [Internet] 2009 [acessado em 20 out. 2017]; 120(16): 1640- 5. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/19805654 https://doi.org/10.1161/ CIRCULATIONAHA.109.192644 1. Alberti KGMM, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the Metabolic Syndrome A Joint Interim Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation [Internet] 2009 [acessado em 20 out. 2017]; 120(16): 1640- 5. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/19805654 https://doi.org/10.1161/ CIRCULATIONAHA.109.192644 He Y, Jiang B, Wang J, Feng K, Chang Q, Fan L, et al. Prevalence of the metabolic syndrome and its relation to cardiovascular disease in an elderly Chinese population. J Am Coll Cardiol [Internet]. 2006 [acessado em 6 jan. 2017]; 47(8): 1588-94. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/16630995 https://doi. org/10.1016/j.jacc.2005.11.074 2. Raposo L, Severo M, Barros H, Santos AC. The prevalence of the metabolic syndrome in Portugal: the PORMETS study. BMC Public Health [Internet] 2017 [acessado em: 20 out. 2017]; 17. Disponível em: https://bmcpublichealth.biomedcentral.com/ articles/10.1186/s12889-017-4471-9 https://doi. org/10.1186/s12889-017-4471-9 Ansarimoghaddam A, Adineh HA, Zareban I, Iranpour S, HosseinZadeh A, Kh F. Prevalence of metabolic syndrome in Middle-East countries: Meta-analysis of cross-sectional studies. Diabetes Metab Syndr [Internet] 2018 [acessado em 21 jan. 2019]; 12(2): 195-201. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/29203060 https://doi.org/10.1016/j.dsx.2017.11.004 Schmidt MI, Duncan BB, Azevedo e Silva G, Menezes AM, Monteiro CA, Barreto SM, et al. Chronic non-communicable diseases in Brazil: burden and current challenges. Lancet [Internet] 2011 [acessado em 1º jan. 2017]; 377(9781): 1949-61. DISCUSSION Among the strengths of this study is the use of a representative sample of non-institutional- ized elderly people in Brazil’s largest city, using criteria that is comparable to other international studies in order to assess the prevalence and factors associated with MetS. In addition, this study was innovative by demonstrating the impact of the practice of activity as a preventive measure. The cross-sectional design and the impossibility of inferring the direction of the associations are among the limitations of the study. The self-reporting of diseases can also be considered one of 10 REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly the limitations, but this measure is considered to be valid65. In addition, it is worth noting that in this study only education level was used as a measure of socioeconomic status. Thus, studies using other measures may find different results regarding the magnitude of inequalities, given that each socioeconomic measurement has a way to explain these differences66, 67. CONCLUSION This study demonstrated that the practice of physical activity and education level are factors that are significantly associated with MetS. This association reinforces the role of physical activity as a first-line non-pharmacological treatment to control this condition68,11 due to its direct relationship with all of the syndrome’s components11. In addition, the exis- tence of socioeconomic inequalities related to MetS calls attention to the need for prevention policies to be based on approaches that consider these differences, and go beyond the other factors that are commonly explored, since both the practice of physical activity69 and the other modifiable factors, are disproportionately inaccessible to the nation’s poorest people70. REFERENCES 2018]. Disponível em: http:// www.aefa.es/wp-content/uploads/2014/04/NECP- guidelines-.pdf 11. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) [Internet]. 2001 [acessado em 12 abr. 2018]. Disponível em: http:// www.aefa.es/wp-content/uploads/2014/04/NECP- guidelines-.pdf . Oguoma VM, Nwose EU, Skinner TC, Richards RS, Digban KA, Onyia IC. Association of physical activity with metabolic syndrome in a predominantly rural Nigerian population. Diabetes Metab Syndr [Internet] 2016 [acessado em 20 nov. 2017]; 10(1): 13-8. Disponível em: https://www.sciencedirect.com/science/article/ abs/pii/S1871402115000776 https://doi.org/10.1016/j. dsx.2015.08.010 12. Márquez-Sandoval F, Macedo-Ojeda G, Viramontes- Hörner D, Fernández Ballart JD, Salas Salvadó J, Vizmanos B. The prevalence of metabolic syndrome in Latin America: a systematic review. Public Health Nutr [Internet] 2011 [acessado em 12 abr. 2018]; 14(10): 1702-13. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/21486521 https://doi.org/10.1017/ S1368980010003320 20. Lee H, Kim B. Physical activity disparities by socioeconomic status among metabolic syndrome patients: The Fifth Korea National Health and Nutrition Examination Survey. J Exerc Rehabil [Internet]. 2016 [acessado em 20 nov. 2017]; 12(1): 10-4. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/26933654 https://doi.org/10.12965/jer.150269 20. Lee H, Kim B. Physical activity disparities by socioeconomic status among metabolic syndrome patients: The Fifth Korea National Health and Nutrition Examination Survey. J Exerc Rehabil [Internet]. 2016 [acessado em 20 nov. 2017]; 12(1): 10-4. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/26933654 https://doi.org/10.12965/jer.150269 21. Bergström G, Behre C, Schmidt C. Increased Leisure- Time Physical Activity is Associated With Lower Prevalence of the Metabolic Syndrome in 64-Year Old Women With Impaired Glucose Tolerance. Angiology [Internet] 2012 [acessado em 20 nov. 2017]; 63(4): 297-301. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/21873352 https://doi. org/10.1177/0003319711414867 13. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet [Internet] 2005 [acessado em 10 out. 2017]; 365(9468): 1415-28. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/15836891 https:// doi.org/10.1016/S0140-6736(05)66378-7 13. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet [Internet] 2005 [acessado em 10 out. 2017]; 365(9468): 1415-28. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/15836891 https:// doi.org/10.1016/S0140-6736(05)66378-7 14. Assmann G, Guerra R, Fox G, Cullen P, Schulte H, Willett D, et al. The metabolic syndrome. Am J Cardiol [Internet] 2007 [acessado em 10 out. 2017]; 99(4): 541-8. Disponível em: https://www.ncbi.nlm. nih.gov/pubmed/17293200 https://doi.org/10.1016/j. amjcard.2006.08.045 22. Cho KI, Kim BH, Je HG, Jang JS, Park YH. Gender- Specific Associations between Socioeconomic Status and Psychological Factors and Metabolic Syndrome in the Korean Population: Findings from the 2013 Korean National Health and Nutrition Examination Survey. Biomed Res Int [Internet] 2016 [acessado em: 20 nov. 2017]. REFERENCES Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/21561658 https://doi.org/10.1016/ S0140-6736(11)60135-9 Schmidt MI, Duncan BB, Azevedo e Silva G, Menezes AM, Monteiro CA, Barreto SM, et al. Chronic non-communicable diseases in Brazil: burden and current challenges. Lancet [Internet] 2011 [acessado em 1º jan. 2017]; 377(9781): 1949-61. Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/21561658 https://doi.org/10.1016/ S0140-6736(11)60135-9 Subramani SK, Mahajan S, Chauhan P, Yadav D, Mishra M, Pakkirisamy U, et al. Prevalence of metabolic syndrome in Gwalior region of Central India: A comparative study using NCEP ATP III, IDF and Harmonized criteria. Diabetes Metab Syndr [Internet]. 2019 [acessado em 21 jan. 2019]; 13(1): 816-21. Disponível em: https://www. sciencedirect.com/science/article/pii/S1871402118305666 https://doi.org/10.1016/j.dsx.2018.12.003 11 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. 16. Ahima RS. Adipose tissue as an endocrine organ. Obesity [Internet] 2006 [acessado em 10 out. 2017]; 14(Supl. 5): 242S-249S. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/17021375 https://doi. org/10.1038/oby.2006.317 8. Vidigal FC, Bressan J, Babio N, Salas-Salvadó J. Prevalence of metabolic syndrome in Brazilian adults: a systematic review. BMC Public Health [Internet] 2013 [acessado em 12 abr. 2018]; 13: 1198. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/24350922 https://doi.org/10.1186/1471-2458-13-1198 16. Ahima RS. Adipose tissue as an endocrine organ. Obesity [Internet] 2006 [acessado em 10 out. 2017]; 14(Supl. 5): 242S-249S. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/17021375 https://doi. org/10.1038/oby.2006.317 17. Wu S, Fisher-Hoch SP, Reininger B, McCormick JB. Recommended Levels of Physical Activity Are Associated with Reduced Risk of the Metabolic Syndrome in Mexican-Americans. PLoS One [Internet]. 2016 [acessado em 10 out. 2017]; 11(4): e0152896. Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/27054324 https://doi.org/10.1371/journal. pone.0152896 9. Vieira DAS, Hermes Sales C, Galvão Cesar CL, Marchioni DM, Fisberg RM. Influence of Haem, Non-Haem, and Total Iron Intake on Metabolic Syndrome and Its Components: A Population-Based Study. Nutrients [Internet] 2018 [acessado em 20 jun. 2018]; 10(3). Disponível em: https://www.ncbi.nlm. nih.gov/pubmed/29518910 https://doi.org/10.3390/ nu10030314 8. Althoff T, Sosic˘ R, Hicks JL, King AC, Delp SL, Leskovec J. Recommended Levels of Physical Activity Are Associated with Reduced Risk of the Metabolic Syndrome in Mexican-Americans. Nature [Internet] 2017 [acessado em 10 out. 2017]; 547(7663): 336- 9. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/28693034 https://doi.org/10.1038/ nature23018 10. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med [Internet] 2006 [acessado em 20 jun. 2017];3(11): e442. Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/17132052 https://doi.org/10.1371/journal. pmed.0030442 11. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) [Internet]. 2001 [acessado em 12 abr. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly 31. Roos V, Elmståhl S, Ingelsson E, Sundström J, Ärnlöv J, Lind L. Metabolic Syndrome Development During Aging with Special Reference to Obesity Without the Metabolic Syndrome. Metab Syndr Relat Disord [Internet] 2017 [acessado em 21 jan. 2018]; 15(1): 36-43. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/27754771 https://doi.org/10.1089/met.2016.0082 31. Roos V, Elmståhl S, Ingelsson E, Sundström J, Ärnlöv J, Lind L. Metabolic Syndrome Development During Aging with Special Reference to Obesity Without the Metabolic Syndrome. Metab Syndr Relat Disord [Internet] 2017 [acessado em 21 jan. 2018]; 15(1): 36-43. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/27754771 https://doi.org/10.1089/met.2016.0082 23. Yang JJ, Yoon HS, Lee SA, Choi JY, Song M, Han S, et al. Metabolic syndrome and sex-specific socio- economic disparities in childhood and adulthood: the Korea National Health and Nutrition Examination Surveys Diabet Med [Internet] 2014 [acessado em 20 nov. 2017]; 31(11): 1399-409. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/24925795 https:// doi.org/10.1111/dme.12525 32. Moura RF, Andrade FB, Duarte YAO, Lebrão ML, Antunes JLF. Factors associated with adherence to influenza vaccination among non-institutionalized elderly in São Paulo, Brazil. Cad Saúde Pública [Internet] 2015 [acessado em 21 jan. 2018]; 31(10): 1-11. Disponível em: http://www.scielo.br/pdf/csp/ v31n10/en_0102-311X-csp-31-10-2157.pdf https:// doi.org/10.1590/0102-311X00065414 32. Moura RF, Andrade FB, Duarte YAO, Lebrão ML, Antunes JLF. Factors associated with adherence to influenza vaccination among non-institutionalized elderly in São Paulo, Brazil. Cad Saúde Pública [Internet] 2015 [acessado em 21 jan. 2018]; 31(10): 1-11. Disponível em: http://www.scielo.br/pdf/csp/ v31n10/en_0102-311X-csp-31-10-2157.pdf https:// doi.org/10.1590/0102-311X00065414 24. Dutra ES, Carvalho KMB, Miyazaki E, Merchán- Hamann E, Ito MK. Metabolic syndrome in central Brazil: prevalence and correlates in the adult population. Diabetol Metab Syndr [Internet] 2012 [acessado em 20 nov. 2017]; 4: 20. Disponível em: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3457864/ https:// dx.doi.org/10.1186%2F1758-5996-4-20 33. Andrade JM, Duarte YAO, Alves LC, Andrade FCD, Souza Junior PRB, Lima-Costa MF, et al. Frailty profile in Brazilian older adults: ELSI-Brazil [Internet] 2018 [acessado em 21 jan. 2019]; 52(Supl. 2): 17s. Disponível em: http://www.scielo.br/pdf/rsp/v52s2/0034-8910- rsp-52-s2-S1518-87872018052000616.pdf https://doi. org/10.11606/S1518-8787.2018052000616 25. Leitão MPC, Martins IS. Prevalência e fatores associados à síndrome metabólica em usuários de unidades básicas de saúde em São Paulo – SP. Rev Assoc Med Bras [Internet] 2012 [acessado em: 20 nov. 2017]; 58(1): 60-9. Disponível em: http://www.scielo.br/pdf/ramb/v58n1/v58n1a16. pdf https://doi.org/10.1590/S0104-42302012000100016 26. Gronner M, Bosi PL, Carvalho AM, Casale G, Contrera D, Pereira MA, et al. Prevalence of metabolic syndrome and its association with educational inequalities among Brazilian adults: a population-based study. Braz J Med Biol Res [Internet] 2011 [acessado em: 20 nov. 2017]; 44(7): 713-9. Disponível em: http://www.scielo.br/ pdf/bjmbr/v44n7/997.pdf https://doi.org/10.1590/ S0100-879X2011007500087 34. REFERENCES Disponível em: https://www.hindawi. com/journals/bmri/2016/3973197/ https://doi. org/10.1155/2016/3973197 15. Sociedade Brasileira de Hipertensão. I Diretriz Brasileira de Diagnóstico e Tratamento da Síndrome Metabólica [Internet]. Sociedade Brasileira de Hipertensão; 2005 [acessado em 10 out. 2017]. Disponível em: http://publicacoes.cardiol.br/consenso/2005/ sindromemetabolica.pdf 15. Sociedade Brasileira de Hipertensão. I Diretriz Brasileira de Diagnóstico e Tratamento da Síndrome Metabólica [Internet]. Sociedade Brasileira de Hipertensão; 2005 [acessado em 10 out. 2017]. Disponível em: http://publicacoes.cardiol.br/consenso/2005/ sindromemetabolica.pdf 12 REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Andrade FB, Lebrão ML, Santos JL, Duarte YA. Relationship Between Oral Health and Frailty in Community‐Dwelling Elderly Individuals in Brazil. J Am Geriatr Soc [Internet] 2013 [acessado em 21 jan. 2019]; 61(5): 809-14. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/23647172 https://doi. org/10.1111/jgs.12221 35. International Physical Activity Questionnaire. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ) 27. Santos AC, Ebrahim S, Barros H. Gender, socio- economic status and metabolic syndrome in middle- aged and old adults. BMC Public Health [Internet] 2008 [acessado em 20 nov. 2017]; 8: 62. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/18282285 https://doi.org/10.1186/1471-2458-8-62 – Short and Long Forms [Internet]. 2005 [acessado em 21 jan. 2019]. Disponível em: https://sites.google. com/site/theipaq/ 28. Whitehead M. The Concepts and Principles of Equity and Health. Int J Health Serv [Internet] 1992 [acessado em 20 nov. 2017]; 22(3): 429-45. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/1644507 https:// doi.org/10.2190/986L-LHQ6-2VTE-YRRN 28. Whitehead M. The Concepts and Principles of Equity and Health. Int J Health Serv [Internet] 1992 [acessado em 20 nov. 2017]; 22(3): 429-45. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/1644507 https:// doi.org/10.2190/986L-LHQ6-2VTE-YRRN 36. World Health Organization. Recomendaciones mundiales sobre actividad física para la salud [Internet]. Genebra: World Health Organization; 2010 [acessado em 10 out. 2017]. Disponível em: https://apps.who.int/ iris/bitstream/handle/10665/44441/9789243599977_ spa.pdf?sequence=1 29. World Health Organization. World health statistics 2018: monitoring health for the SDGs, sustainable development goals [Internet]. Genebra: World Health Organization; 2018 [acessado em 21 jan. 2019]. Disponível em: https://apps.who.int/iris/bitstream/ handle/10665/272596/9789241565585-eng.pdf?ua=1 30. Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel JP, et al. The World report on ageing and health: a policy framework for healthy ageing. Lancet [Internet] 2016 [acessado em 21 jan. 2019]; 387(10033): 2145-54. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/26520231 https://doi.org/10.1016/ S0140-6736(15)00516-4 29. World Health Organization. World health statistics 2018: monitoring health for the SDGs, sustainable development goals [Internet]. Genebra: World Health Organization; 2018 [acessado em 21 jan. 2019]. Disponível em: https://apps.who.int/iris/bitstream/ handle/10665/272596/9789241565585-eng.pdf?ua=1 37. Almeida OP, Almeida SA. Confiabilidade da versão brasileira da Escala de Depressão em Geriatria (GDS) versão reduzida. Arq Neuropsiquiatr [Internet] 1999 [acessado em 10 out. 2017]; 57(2-B): 421-6. Disponível em: http://www.scielo.br/pdf/anp/v57n2B/1446.pdf https://doi.org/10.1590/S0004-282X1999000300013 38. Rao JNK, Scott AJ. On Chi-Squared Tests for Multiway Contingency Tables with Cell Proportions Estimated from Survey Data. Ann Statist [Internet] 1984 [acessado em 10 out. 2017]; 12(1): 46-60. Disponível em: https:// projecteuclid.org/euclid.aos/1176346391 https://doi. org/10.1214/aos/1176346391 37. Almeida OP, Almeida SA. Confiabilidade da versão brasileira da Escala de Depressão em Geriatria (GDS) versão reduzida. Arq Neuropsiquiatr [Internet] 1999 [acessado em 10 out. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly 2017]; 57(2-B): 421-6. Disponível em: http://www.scielo.br/pdf/anp/v57n2B/1446.pdf https://doi.org/10.1590/S0004-282X1999000300013 37. Almeida OP, Almeida SA. Confiabilidade da versão brasileira da Escala de Depressão em Geriatria (GDS) versão reduzida. Arq Neuropsiquiatr [Internet] 1999 [acessado em 10 out. 2017]; 57(2-B): 421-6. Disponível em: http://www.scielo.br/pdf/anp/v57n2B/1446.pdf https://doi.org/10.1590/S0004-282X1999000300013 30. Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel JP, et al. The World report on ageing and health: a policy framework for healthy ageing. Lancet [Internet] 2016 [acessado em 21 jan. 2019]; 387(10033): 2145-54. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/26520231 https://doi.org/10.1016/ S0140-6736(15)00516-4 38. Rao JNK, Scott AJ. On Chi-Squared Tests for Multiway Contingency Tables with Cell Proportions Estimated from Survey Data. Ann Statist [Internet] 1984 [acessado em 10 out. 2017]; 12(1): 46-60. Disponível em: https:// projecteuclid.org/euclid.aos/1176346391 https://doi. org/10.1214/aos/1176346391 13 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. 47. Ford CS, Li C, Zhao G. Prevalence and correlates of metabolic syndrome based on a harmonious deﬁnition among adults in the US [Internet] 2010 [acessado em 20 out. 2017]; 2(3): 180-93. Disponível em: https://onlinelibrary.wiley.com/doi/ full/10.1111/j.1753-0407.2010.00078.x https://doi. org/10.1111/j.1753-0407.2010.00078.x 39. Piola TS, Bacil EDA, Silva MP, Pacífico AB, Camargo EM, Campos W. Impact of physical activity correlates in the isolated and combined presence of insufficient level of physical activity and high screen time among adolescents. Rev Paul Pedriatr [Internet] 2019 [acessado em 21 jan. 2019]; 37(2): 194-201. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/30672979 https:// doi.org/10.1590/1984-0462/;2019;37;2;00011 47. Ford CS, Li C, Zhao G. Prevalence and correlates of metabolic syndrome based on a harmonious deﬁnition among adults in the US [Internet] 2010 [acessado em 20 out. 2017]; 2(3): 180-93. Disponível em: https://onlinelibrary.wiley.com/doi/ full/10.1111/j.1753-0407.2010.00078.x https://doi. org/10.1111/j.1753-0407.2010.00078.x 48. Paula JAT, Moreira OC, da Silva CD, Silva DS, dos Santos Amorim PR. Metabolic syndrome prevalence in elderly of urban and rural communities participants in the HIPERDIA in the city of Coimbra/MG, Brazil. Invest Educ Enferm [Internet] 2015 [acessado em 20 out. 2017]; 33(2): 325-33. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/26535853 https:// doi.org/10.1590/S0120-53072015000200015 40. Tanamas SK, Permatahati V, Ng WL, Backholer K, Wolfe R, Shaw JE, et al. Estimating the proportion of metabolic health outcomes attributable to obesity: a cross-sectional exploration of body mass index and waist circumference combinations. BMC Obes [Internet] 2016 [acessado em 21 jan. 2019]; 3: 4. Disponível em: https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC4734864/ https://dx.doi.org/10.1186%2Fs40608-016-0085-5 49. Santos BMP, Daltro C, Farias MCAD, Abreu LC, Valenti VE, Bezerra IMP, et al. Metabolic syndrome in elderly from a northeastern Brazilian city. Int Arch Med [Internet] 2015 [acessado em 20 out. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly 2017]; 8(20). Disponível em: https://imed.pub/ojs/ index.php/iam/article/view/1009/683 https://doi. org/10.3823/1619 41. Newson RB. Attributable and unattributable risks and fractions and other scenario comparisons. Stata J [Internet] 2013 [acessado em 21 jan. 2019]; 13(4): 672-98. Disponível em: https://ideas.repec.org/a/ tsj/stataj/v13y2013i4p672-698.html 42. Moreno-Betancur M, Latouche A, Menvielle G, Kunst AE, Rey G. Relative Index of Inequality and Slope Index of Inequality: A Structured Regression Framework for Estimation. Epidemiology [Internet] 2015 [acessado em 20 out. 2017]; 26(4): 518-27. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/26000548 https:// doi.org/10.1097/EDE.0000000000000311 50. Sá NNB, Moura EC. Fatores associados à carga de doenças da síndrome metabólica entre adultos brasileiros. Cad Saúde Pública [Internet] 2010 [acessado em 20 out. 2017]; 26(9): 1853-62. Disponível em: http:// www.scielo.br/pdf/csp/v26n9/18.pdf https://doi. org/10.1590/S0102-311X2010000900018 51. Stringhini S, Carmeli C, Jokela M, Avendaño M, Muennig P, Guida F, et al. Socioeconomic status and the 25 × 25 risk factors as determinants of premature mortality: a multicohort study and meta-analysis of 1·7 million men and women. Lancet [Internet] 2017 [acessado em 12 abr. 2018]; 389(10075): 1229-37. Disponível em: https://www.thelancet.com/journals/ lancet/article/PIIS0140-6736(16)32380-7/fulltext https://doi.org/10.1016/S0140-6736(16)32380-7 43. Mackenbach JP, et al. Socioeconomic Inequalities in Health in 22 European Countries. N Engl J Med [Internet] 2008 [acessado em 20 out. 2017]; 358: 2468-81. Disponível em: https://www.nejm.org/doi/full/10.1056/NEJMsa0707519 https://doi.org/10.1056/NEJMsa0707519 44. Ortiz-Rodríguez MA, Yáñez-Velasco L, Carnevale A, Romero-Hidalgo S, Bernal D, Aguilar-Salinas C, et al. Prevalence of metabolic syndrome among elderly Mexicans. Arch Gerontol Geriatr [Internet] 2017 [acessado em 12 abr. 2018]; 73: 288-93. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/28910752 https://doi.org/10.1016/j.archger.2017.09.001 52. Lübs L, Peplies J, Drell C, Bammann K. Cross-sectional and longitudinal factors influencing physical activity of 65 to 75-year-olds: a pan European cohort study based on the survey of health, ageing and retirement in Europe (SHARE). BMC Geriatr [Internet] 2018 [acessado em 21 maio 2018]; 18(1): 94. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/29661154 https://doi.org/10.1186/s12877-018-0781-8 52. Lübs L, Peplies J, Drell C, Bammann K. Cross-sectional and longitudinal factors influencing physical activity of 65 to 75-year-olds: a pan European cohort study based on the survey of health, ageing and retirement in Europe (SHARE). BMC Geriatr [Internet] 2018 [acessado em 21 maio 2018]; 18(1): 94. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/29661154 https://doi.org/10.1186/s12877-018-0781-8 45. Orces CH, Gavilanez EL. The prevalence of metabolic syndrome among older adults in Ecuador: Results of the SABE survey. Diabetes Metab Syndr [Internet] 2017 [acessado em 12 abr. 2018]; 11(Supl. 2): S555-S560. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/28427839 https://doi.org/10.1016/j.dsx.2017.04.004 53. Wu HF, Tam T, Jin L, Lao XQ, Chung RY, Su XF, et al. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly 54. Brasil. Ministério da Saúde. Relatório do III Fórum de Monitoramento do Plano de Ações Estratégicas para o Enfrentamento das Doenças Crônicas não Transmissíveis no Brasil [Internet]. Brasil: Ministério da Saúde; 2018 [acessado em 21 jan. 2019]. Disponível em: http://bvsms.saude. gov.br/bvs/publicacoes/relatorio_iii_forum_ monitoramento_plano.pdf 61. Khosravi-boroujeni H, Sarrafzadegan N, Sadeghi M, Roohafza H, Talaei M, Ng SK, et al. Secular Trend of Metabolic Syndrome and Its Components in a Cohort of Iranian Adults from 2001 to 2013. Metab Syndr Relat Disord [Internet] 2017 [Acessado em: 12 abr. 2018]; 15(3): 137-44. Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/28135122 https://doi.org/10.1089/met.2016.0073 61. Khosravi-boroujeni H, Sarrafzadegan N, Sadeghi M, Roohafza H, Talaei M, Ng SK, et al. Secular Trend of Metabolic Syndrome and Its Components in a Cohort of Iranian Adults from 2001 to 2013. Metab Syndr Relat Disord [Internet] 2017 [Acessado em: 12 abr. 2018]; 15(3): 137-44. Disponível em: https://www.ncbi.nlm.nih.gov/ pubmed/28135122 https://doi.org/10.1089/met.2016.0073 62. Kuzuya M, Ando F, Iguchi A, Shimokata H. Age-specific change of prevalence of metabolic syndrome: longitudinal observation of large Japanese cohort. Atherosclerosis [Internet] 2007 [acessado em 12 abr. 2018]; 191(2): 305-12. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/16828779 https://doi.org/10.1016/j.atherosclerosis.2006.05.043 55. Malta DC, Stopa SR, Iser BPM, Bernal RTI, Claro RM, Nardi ACF, et al. Risk and protective factors for chronic diseases by telephone survey in capitals of Brazil, Vigitel 2014. Rev Bras Epidemiol [Internet] 2015 [acessado em 12 abr. 2018]; 18(Supl. 2): 238- 55. Disponível em: http://www.scielo.br/pdf/ rbepid/v18s2/en_1980-5497-rbepid-18-s2-00238.pdf 10.1590/1980-5497201500060021 63. Costa CS, Schneider BC, Cesar JA. Obesidade geral e abdominal em idosos do Sul do Brasil: resultados do estudo COMO VAI. Ciênc Saúde Coletiva [Internet] 2016 [acessado em 21 jan. 2019]; 21(11): 3585-96. Disponível em: http:// www.scielo.br/pdf/csc/v21n11/1413-8123-csc-21-11-3585. pdf https://doi.org/10.1590/1413-812320152111.02492016 63. Costa CS, Schneider BC, Cesar JA. Obesidade geral e abdominal em idosos do Sul do Brasil: resultados do estudo COMO VAI. Ciênc Saúde Coletiva [Internet] 2016 [acessado em 21 jan. 2019]; 21(11): 3585-96. Disponível em: http:// www.scielo.br/pdf/csc/v21n11/1413-8123-csc-21-11-3585. pdf https://doi.org/10.1590/1413-812320152111.02492016 56. França SL, Lima SS, Vieira JRS. Metabolic Syndrome and Associated Factors in Adults of the Amazon Region. PLoS One [Internet] 2016 [acessado em 12 abr. 2018]; 11(12): e0167320. Disponível em: https:// journals.plos.org/plosone/article?id=10.1371/ journal.pone.0167320 https://doi.org/10.1371/ journal.pone.0167320 56. França SL, Lima SS, Vieira JRS. Metabolic Syndrome and Associated Factors in Adults of the Amazon Region. PLoS One [Internet] 2016 [acessado em 12 abr. 2018]; 11(12): e0167320. Disponível em: https:// journals.plos.org/plosone/article?id=10.1371/ journal.pone.0167320 https://doi.org/10.1371/ journal.pone.0167320 64. Assumpção D, Borim FSA, Francisco PMSB, Neri AL. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Age, gender, and socioeconomic gradients in metabolic syndrome: biomarker evidence from a large sample in Taiwan, 2005–2013. Ann Epidemiol [Internet] 2017 [acessado em 12 abr. 2018]; 27(5): 315-22. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/28595736 https://doi.org/10.1016/j. annepidem.2017.04.003 46. Li W, Song F, Wang X, Wang L, Wang D, Yin X, et al. Prevalence of metabolic syndrome among middle- aged and elderly adults in China: current status and temporal trends. Ann Med [Internet] 2018 [acessado em 21 jan. 2019]; 50(4): 345-53. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/29637787 https:// doi.org/10.1080/07853890.2018.1464202 14 REV BRAS EPIDEMIOL 2020; 23: E200046 Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Fatores associados ao baixo peso em idosos comunitários de sete cidades brasileiras: Estudo FIBRA. Ciênc Saúde Coletiva [Internet] 2018 [acessado em 21 jan. 2019]; 23(4): 1143-50. Disponível em: http://www. scielo.br/pdf/csc/v23n4/1413-8123-csc-23-04-1143.pdf https://doi.org/10.1590/1413-81232018234.17422016 57. Moreira GC, Cipullo JP, Ciorlia LA, Cesarino CB, Vilela- Martin JF. Prevalence of Metabolic Syndrome: Association with Risk Factors and Cardiovascular Complications in an Urban Population. PLoS One [Internet] 2014 [acessado em 31 maio 2018]; 9(9): e105056. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/25180496 https://doi.org/10.1371/ journal.pone.0105056 57. Moreira GC, Cipullo JP, Ciorlia LA, Cesarino CB, Vilela- Martin JF. Prevalence of Metabolic Syndrome: Association with Risk Factors and Cardiovascular Complications in an Urban Population. PLoS One [Internet] 2014 [acessado em 31 maio 2018]; 9(9): e105056. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/25180496 https://doi.org/10.1371/ journal.pone.0105056 65. Haapanen N, Miilunpalo S, Pasanen M, Oja P, Vuori I. Agreement between questionnaire data and medical records of chronic diseases in middle-aged and elderly Finnish men and women. Am J Epidemiol [Internet] 1997 [acessado em 21 jan. 2019]; 145(8): 762-9. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/9126003 https://doi.org/10.1093/aje/145.8.762 65. Haapanen N, Miilunpalo S, Pasanen M, Oja P, Vuori I. Agreement between questionnaire data and medical records of chronic diseases in middle-aged and elderly Finnish men and women. Am J Epidemiol [Internet] 1997 [acessado em 21 jan. 2019]; 145(8): 762-9. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/9126003 https://doi.org/10.1093/aje/145.8.762 58. Lim H, Nguyen T, Choue R Wang Y. Sociodemographic disparities in the composition of metabolic syndrome components among adults in South Korea. Diabetes Care [Internet] 2012 [acessado em 31 maio 2018]; 35(10): 2028-35. Disponível em: https://www. ncbi.nlm.nih.gov/pubmed/22837361 https://doi. org/10.2337/dc11-1841 66. Grundy E, Holt G. The socioeconomic status of older adults: How should we measure it in studies of health inequalities. J Epidemiol Community Health [Internet] 2001 [acessado em 21 jan. 2019]; 55(12): 895-904. Disponível em: https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC1731799/ https://dx.doi.org/10.1136%2Fjech.55.12.895 66. Grundy E, Holt G. The socioeconomic status of older adults: How should we measure it in studies of health inequalities. J Epidemiol Community Health [Internet] 2001 [acessado em 21 jan. 2019]; 55(12): 895-904. Disponível em: https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC1731799/ https://dx.doi.org/10.1136%2Fjech.55.12.895 59. Azizi F, Salehi P, Etemadi A, Zahedi-Asl S. Prevalence of metabolic syndrome in an urban population: Tehran Lipid and Glucose Study. Diabetes Res Clin Pract [Internet] 2003 [acessado em 31 maio 2018]; 61(1): 29-37. Disponível em: https://www.ncbi.nlm. nih.gov/pubmed/12849921 https://doi.org/10.1016/ s0168-8227(03)00066-4 67. Avlund K, Holstein BE, Osler M, Damsgaard MT, Holm- Pedersen P, Rasmussen NK. Social position and health in old age: the relevance of different indicators of social position. Metabolic syndrome: physical inactivity and socioeconomic inequalities among non-institutionalized Brazilian elderly Scand J Public Health [Internet] 2003 [acessado em 21 jan. 2019]; 31(2): 126-36. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/12745763 https://doi.org/10.1080/14034940210134130 67. Avlund K, Holstein BE, Osler M, Damsgaard MT, Holm- Pedersen P, Rasmussen NK. Social position and health in old age: the relevance of different indicators of social position. Scand J Public Health [Internet] 2003 [acessado em 21 jan. 2019]; 31(2): 126-36. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/12745763 https://doi.org/10.1080/14034940210134130 60. Khanam MA, Qiu C, Lindeboom W, Streatfield PK, Kabir ZN, Wahlin Å. The metabolic syndrome: prevalence, associated factors, and impact on survival among older persons in rural Bangladesh. PLoS One [Internet] 2011 [acessado em 12 abr. 2018]; 6(6): e20259. Disponível em: https://www.ncbi.nlm.nih. gov/pubmed/21697988 https://doi.org/10.1371/ journal.pone.0020259 68. Oh SH, Son SH, Kang SH, Kim DK, Seo KM, Lee SY. Relationship Between Types of Exercise and Quality of Life in a Korean Metabolic Syndrome Population: A Cross- Sectional Study. Metab Syndr Relat Disord [Internet] 2017 [acessado em 21 jan. 2019]; 15(4): 199-205. Disponível em: https://www.ncbi.nlm.nih.gov/pubmed/28287907 https://doi.org/10.1089/met.2016.0151 15 REV BRAS EPIDEMIOL 2020; 23: E200046 COSTA, A.C.O. ET AL. COSTA, A.C.O. ET AL. Received on: 08/31/2018 Revised on: 01/30/2019 Accepted on: 03/11/2019 Received on: 08/31/2018 Revised on: 01/30/2019 Accepted on: 03/11/2019 Received on: 08/31/2018 Revised on: 01/30/2019 Accepted on: 03/11/2019 69. O’Donoghue G, Kennedy A, Puggina A, Aleksovska K, Buck C, Burns C, et al. Socio-economic determinants of physical activity across the life course: A “DEterminants of DIet and Physical ACtivity” (DEDIPAC) umbrella literature review. PLoS One [Internet] 2018 [acessado em: 21 jan. 2019]; 13(1): e0190737. Disponível em: https:// www.ncbi.nlm.nih.gov/pubmed/29351286 https:// doi.org/10.1371/journal.pone.0190737 Authors’ contributions: ACOC: Design, interpretation of the results, preparation and writing of the manuscript, and a critical review of the content. YAOD: Preparation and writing of the manuscript, and a critical review of the content. FBA: Design, analysis and interpretation of the results, preparation and writing of the manuscript, and a critical review of the content. All of the authors approved the final version of the manuscript. 70. Cesare MD, Khang Y-H, Asaria P, Blakely T, Cowan MJ, Farzadfar F, et al. Inequalities in non-communicable diseases and effective responses. Lancet [Internet] 2013 [acessado em 31 maio 2018]; 381(9866): 585-97. Disponível em: https://www.thelancet.com/journals/ lancet/article/PIIS0140-6736(12)61851-0/fulltext https://doi.org/10.1016/S0140-6736(12)61851-0 16 REV BRAS EPIDEMIOL 2020; 23: E200046 © 2020 Associação Brasileira de Saúde Coletiva This is an open access article distributed under the terms of the Creative Commons license.
W4283023079.txt	https://www.mdpi.com/2077-0383/11/12/3466/pdf?version=1655695762	en		Hepatitis B Virus and Tuberculosis Are Associated with Increased Noncommunicable Disease Risk among Treatment-Naïve People with HIV: Opportunities for Prevention, Early Detection and Management of Comorbidities in Sierra Leone	Journal of clinical medicine	2,022	cc-by	8,705	Journal of Clinical Medicine Article Hepatitis B Virus and Tuberculosis Are Associated with Increased Noncommunicable Disease Risk among Treatment-Naïve People with HIV: Opportunities for Prevention, Early Detection and Management of Comorbidities in Sierra Leone George A. Yendewa 1,2,3, * , Sulaiman Lakoh 4,5 , Darlinda F. Jiba 5 , Sahr A. Yendewa 5 , Umu Barrie 6 , Gibrilla F. Deen 4,5 , Mohamed Samai 4,5 , Jeffrey M. Jacobson 1,2 , Foday Sahr 4,5 and Robert A. Salata 1,2 1 2 3 4 5 Citation: Yendewa, G.A.; Lakoh, S.; 6 Jiba, D.F.; Yendewa, S.A.; Barrie, U.; * Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, OH 44106, USA; jxj573@case.edu (J.M.J.); robert.salata@uhhospitals.org (R.A.S.) Division of Infectious Diseases and HIV Medicine, University Hospitals Cleveland Medical Center, Cleveland, OH 44106, USA Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA Department of Medicine, College of Medicine and Allied Health Sciences, University of Sierra Leone, Freetown, Sierra Leone; lakoh2009@gmail.com (S.L.); gibrilladeen1960@yahoo.com (G.F.D.); dhmsamai@yahoo.co.uk (M.S.); fsahr65@gmail.com (F.S.) Ministry of Health and Sanitation, Freetown, Sierra Leone; darlindajiba.dj@gmail.com (D.F.J.); syendewa@gmail.com (S.A.Y.) Infectious Disease Research Network, Freetown, Sierra Leone; barrieumu1993@gmail.com Correspondence: gay7@case.edu Deen, G.F.; Samai, M.; Jacobson, J.M.; Sahr, F.; Salata, R.A. Hepatitis B Virus and Tuberculosis Are Associated with Increased Noncommunicable Disease Risk among Treatment-Naïve People with HIV: Opportunities for Prevention, Early Detection and Management of Comorbidities in Sierra Leone. J. Clin. Med. 2022, 11, 3466. https://doi.org/10.3390/ jcm11123466 Academic Editor: Loredana Sarmati Received: 30 March 2022 Accepted: 15 June 2022 Published: 16 June 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- Abstract: Noncommunicable diseases (NCDs) are a growing public health concern in low- and middle-income countries and disproportionately affect people living with HIV (PWH). Hepatitis B virus (HBV) and tuberculosis (TB) coinfection are presumed risk factors in endemic settings; however, supporting evidence is conflicting. We analyzed baseline data of newly diagnosed PWH prospectively enrolled in the Sierra Leone HIV Cohort Study in Freetown, Sierra Leone, from March to September 2021. Logistic regression was used to identify associations between NCDs, HBV and TB. A total of 275 PWH aged ≥18 years were studied (55% female, median age 33 years, median CD4 307 cells/mm3 , 15.3% HIV/HBV, 8.7% HIV/TB). NCDs were bimodally distributed, with 1 in 4 PWH clustered around liver disease (fibrosis/cirrhosis), diabetes/prediabetes and obesity/preobesity, while 1 in 8 had renal impairment or hypertension (HTN). Overall, 41.5% had ≥1 NCD, while 17.5% were multimorbid (≥2 NCDs). After adjusting for age, sex, sociodemographic factors and CD4 count, liver fibrosis/cirrhosis was strongly associated with HBV (aOR 8.80, 95% CI [2.46–31.45]; p < 0.001) and diabetes/prediabetes (aOR 9.89, 95% CI [1.14–85.67]; p < 0.037). TB independently predicted diabetes/prediabetes (aOR 7.34, 95% CI [1.87–28.74]; p < 0.004), while renal impairment was associated with proteinuria (aOR 9.34, 95% CI [2.01–43.78]; p < 0.004) and HTN (aOR 6.00, 95% CI [1.10–35.39]; p < 0.049). Our findings warrant the implementation of NCD-aware HIV programs for the prevention, early detection and management of comorbidities. iations. Keywords: HIV; HBV; tuberculosis; noncommunicable diseases; Sierra Leone Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1. Introduction Globally, an estimated 38 million people were reported to be infected with the human immunodeficiency virus (HIV) in 2021, with two-thirds residing in sub-Saharan Africa (SSA) [1]. Simultaneously, SSA accounts for 18% and 29% of the global burden of hepatitis B virus (HBV) and tuberculosis (TB), respectively [2,3]. Due to the overlap of all three epidemics in SSA, coinfections are common and serve as important determinants of morbidity J. Clin. Med. 2022, 11, 3466. https://doi.org/10.3390/jcm11123466 https://www.mdpi.com/journal/jcm J. Clin. Med. 2022, 11, 3466 2 of 15 and mortality in this region [2,3]. HIV/HBV coinfected individuals experience high rates of immune reconstitution upon initiating antiretroviral therapy (ART) [4] and an accelerated pace of progression to acquired immunodeficiency syndrome (AIDS) [5], liver cirrhosis and hepatocellular carcinoma [6]. Additionally, HIV infection is a well-recognized risk factor for the activation of latent TB infection, increasing the risk of progression to active TB disease 18-fold [3]. The widespread availability of ART, in addition to improved screening and management strategies for coinfections have substantially reduced HIV-related morbidity and prolonged life expectancy [1]. Paradoxically, this epidemiologic transition has been accompanied by an increase in the burden of noncommunicable diseases (NCDs) among people living with HIV (PWH) [7]. NCDs account for an estimated 41 million deaths annually, with 77% of these occurring in low- and middle-income countries (LMICs) [8]. According to a recent meta-analysis by Patel et al. [9], the most common NCDs among PWH in LMICs include cardiovascular disease (CVD), cervical cancer, depression and diabetes (DM). Studies from high-income countries have identified HIV-induced inflammation [10], ART-related toxicities [11,12] and a high prevalence of lifestyle-associated risk factors such as excessive alcohol use, tobacco smoking, unhealthy dietary habits and reduced physical activity as contributing to NCDs in an aging cohort of PWH [10–12]. However, there is limited research on the prevalence, correlates and pathogenesis of NCDs among PWH in LMICs. While a heightened proinflammatory state and immune dysregulation induced by coinfections such as viral hepatitis [13] and TB [14] have been suggested as possible risk factors in endemic settings, their relative contributions to the pathogenesis of NCDs have not been quantified. Sierra Leone is a West African country with high HBV endemicity (estimated population prevalence of 8–10%) [15–17] and one of 30 high TB burden countries [18] in the setting of a generalized HIV epidemic [19]. Although the HIV epidemic in Sierra Leone has not been fully characterized, preliminary studies have revealed that the majority of PWH (up to 75%) present with late-stage disease (defined as CD4 < 350 cells/mm3 ) [20]. Other prominent features of the HIV epidemic include high rates of HIV drug resistance and virologic failure [21–23] and high rates of coinfections with HBV and TB [24–27]. Improving HIV care and clinical outcomes for PWH are key priorities of the national HIV control program, with the introduction of dolutegravir (DTG)-based ART in Sierra Leone in 2020 [23]. With NCD rates among PWH expected to continue increasing in Sierra Leone and other LMICs in the coming decades, especially given the well-known association between integrase strand inhibitor (INSTI) use and excess weight gain and metabolic complications [12], it is imperative that HIV control programs move towards NCD-aware and integrative approaches to facilitate the early detection and management of comorbidities. The aim of this study was to assess the prevalence and associated factors of NCDs among newly diagnosed PWH prior to initiating ART at the largest HIV treatment center in Freetown, Sierra Leone. Given the high rates of coinfections in Sierra Leone, we further aimed to explore the role of HBV and TB as potential modifiers of NCD risk among PWH in this setting. 2. Materials and Methods 2.1. Study Setting, Design and Population We analyzed sociodemographic and baseline clinical data of adults enrolled in the Sierra Leone HIV Cohort Study at the HIV Clinic at Connaught Hospital in Freetown, Sierra Leone, from March to September 2021. The HIV Clinic at Connaught Hospital is the largest HIV treatment center in Sierra Leone and is affiliated with the College of Medicine and Allied Health Sciences of the University of Sierra Leone. The Sierra Leone HIV Cohort Study is a prospective study that was commenced in March 2021, with the primary objective of evaluating immunologic and virologic outcomes of PWH initiated on DTG-based ART in Sierra Leone. All patients who were aged J. Clin. Med. 2022, 11, 3466 3 of 15 ≥18 years, ART-naïve, received screening for HBV and TB and provided informed consent were eligible for inclusion into the cohort study. 2.2. Routine Laboratory Assessments HIV status was determined using the rapid test by SD Bioline HIV-1/2 3.0 (Standard Diagnostics, Inc., Suwon, Korea) while HBsAg status was determined using the testing kit CitestTM Diagnostics Inc. (Vancouver, BC, Canada) according to the manufacturers’ instructions. TB was defined as having at least one sputum test with a positive acid-fast bacillus (AFB) and/or positive Xpert MTB RIF test. Complete blood count (leukocytes, hemoglobin, platelets), electrolytes, serum creatinine, liver function tests (alkaline phosphatase, ALP; aspartate transaminase, AST; alanine transaminase, ALT; and gamma-glutamyl transpeptidase, GGT), bilirubin (total and direct), total protein and albumin were determined using automated analyzers by Cypress Diagnostics Inc. (Vancouver, BC, Canada). CD4 count was determined using the Alere Pima Analyzer (Abbott, Jena, Germany). 2.3. Assessment of Liver Disease We used three validated non-invasive scores to screen patients for the presence of liver fibrosis and cirrhosis. The Aspartate Transaminase to Platelet Ratio (APRI) score was calculated using the formula [AST (IU/L)/AST (Upper Limit of Normal) (IU/L)]/[Platelet Count (109/L)] × 100. The following APRI thresholds were used to stage liver disease, as previously described by others [28–31]: APRI < 0.5, normal liver; APRI 0.5–1.5, significant fibrosis; and APRI > 1.5, cirrhosis. The FIB-4 score was calculated using the formula [Age (years) × AST (IU/L)]/[Platelet Count (109/L) × [ALT (IU/L)]1/2]. A FIB-4 score < 1.45 was interpreted as having normal liver, FIB-4 score of 1.45–3.25 as significant fibrosis, while FIB-4 > 3.25 as cirrhosis [28–31]. Finally, the GGT-to-platelet ratio (GPR) was calculated using the formula GGT (IU/L)/Platelet count (109/L), with threshold of GPR > 0.32 for significant fibrosis or cirrhosis as previously described by Lemoine et al. [32]. Finally, alcoholic hepatitis was defined as AST/ALT ratio > 2, as previously described [33]. We used reference values for LFTs previously reported in studies from West Africa [32,34]. 2.4. Assessment of Renal Impairment Glomerular filtration rate (eGFR) was estimated and staged (stages 1 to 5) using the Modification of Diet in Renal Disease (MDRD) Study equation, as follows [35]: eGFR = 175 × (serum creatinine) − 1.154 × (age) − 0.203 × (0.742 if female) × (1.212 if black) where GFR is expressed as mL/min/1.73 m2 of body surface area and serum creatinine is expressed in mg/dL. Renal impairment was defined as eGFR < 60 mL/min/1.73 m2 (i.e., stages 3 to 5) [36]. 2.5. Assessment of Impaired Glucose Metabolism We screened for impaired glucose metabolism in accordance with the American Diabetic Association (ADA)’s criteria, defined as follows: (1) prediabetes, defined as fasting blood glucose level of 5.6–6.9 mmol/L and/or HbA1c 5.7–6.4%; (2) DM, fasting blood glucose level ≥ 7.0 mmol/L, HbA1c ≥ 6.5% or being on antidiabetic medications, regardless of biomarker levels [37]. 2.6. Other Assessments In accordance with the American College of Cardiology/ American Heart Association (ACC/AHA) [38], hypertension was assessed using the average of two readings and/or being on treatment with antihypertensives, and was classified as follows: (1) normal: < 120/80 mmHg; (2) elevated: systolic blood pressure (SBP) > 120–129 mmHg and diastolic blood pressure (DBP) < 80 mmHg; Stage 1: SBP > 130–139 mmHg or DBP ≥ 80–89; (2) Stage 2: SBP ≥ 140 mmHg or DBP ≥ 90 mmHg. Anemia was defined as hemoglobin concentration of < 11.0 g/dL for women and < 13.0 g/dL for men aged ≥ 18 years. J. Clin. Med. 2022, 11, 3466 4 of 15 2.7. Statistical Analyses Statistical analyses were performed using the SPSS Version 28.0 (IBM Corp; Armonk, NY, USA). Categorical variables were reported as frequencies (percentages) and associations assessed using Pearson’s chi-square or Fisher’s exact tests. Continuous variables were presented as medians (interquartile ranges, IQR) and associations assessed using the nonparametric independent samples Mann–Whitney U-test. A logistic regression model was used to identify associations between NCDs, HBV and TB. For predicting significant liver fibrosis, we used APRI > 0.7 in accordance with Lin et al. [31] who reported a sensitivity of 77% and specificity of 72% using this threshold. Risk factors associated with NCDs were accessed in the univariate analysis. Variables that attained a p-value of < 0.2 in the univariate analysis were included in the multivariate regression model. Associations were reported as crude (OR) and adjusted odds ratios (aOR) with 95% confidence intervals (CI), with statistical significance set at p < 0.05. 2.8. Ethical Considerations Ethical approval was obtained from the Sierra Leone Ethics and Scientific Review Committee (approved 10 February 2020). Written informed consent was obtained from all study participants prior to enrolment. 3. Results 3.1. Sociodemographic and Clinical Characteristics of Study Participants Baseline data were analyzed on 275 adults aged ≥18 years newly diagnosed with HIV prior to initiating DTG-based ART (Table 1). The median age was 33 years (IQR 27–42) and the median BMI was 22 kg/m2 (IQR 19.8–24.3). The majority were female (56.0%, 154/275) and employed in the informal sector (69.1%, 190/275). About 23.6% (65/275) endorsed alcohol use and 20.7% (57/275) were smokers. About 15.3% (26/174) had HIV/HBV coinfection, and 8.7% (24/275) had tuberculosis. The prevalence of opportunistic conditions was low. The median CD4 cell count was 307 cells/mm3 (149–502), with 57.2% (119/259) meeting the criteria for late-stage HIV presentation (i.e., CD4 < 350 cells/mm3 ). Furthermore, 30.9% (80/259) had AIDS (i.e., CD4 < 200 cells/mm3 ), while 15.1% (39/259) had severe immunosuppression (i.e., CD4 < 100 cells/mm3 ). Table 1. Sociodemographic and clinical characteristics of study participants. Variables N (%) Gender, n (%) Male 121/275 (44.0) Female 154/275 (56.0) Age, years, n (%) Median (IQR) 33 (27–42) <25 42/275 (15.3) 25–34 110/275 (40.0) 35–44 70/275 (25.5) 45–54 42/275 (15.3) ≥55 11/275 (4.0) Highest education attained, n (%) None 52/275 (18.9) Primary 40/275 (14.5) Secondary 128/275 (46.5) Tertiary 55/275 (20.0) J. Clin. Med. 2022, 11, 3466 5 of 15 Table 1. Cont. Variables N (%) Employment status, n (%) Unemployed 57/275 (20.7) Informal 190/275 (69.1) Formal 28/275 (10.2) Monthly earning, n (%) <USD 100 217/275 (78.9) ≥USD 100 58/275 (21.1) Body mass index, kg/m2 , n (%) Median (IQR) 22.0 (19.8–24.3) <18.5 48/275 (17.5) 18.5–24.9 164/275 (59.6) 25.0–29.9 43/275 (15.6) ≥30.0 20/275 (7.3) Lifestyle-associated risk factors, n (%) Smoking 57/275 (20.7) Alcohol use 65/275 (23.6) Drug use 28/275 (10.2) Coinfections, n (%) HBV 26/174 (15.3) Tuberculosis 24/275 (8.7) CD4 count, cells/mm3 , n (%) Median (IQR) 307 (149–502) 0–99 39/259 (15.1) 100–199 41/259 (15.8) 200–349 68/259 (26.3) ≥350 111/259 (42.9) Abbreviations: HBV, hepatitis B virus; IQR, interquartile range; USD, United States dollars. 3.2. Baseline Prevalence of NCDs Prior to Initiation of ART Table 2 displays the baseline laboratory findings of study participants by coinfection status, while Table 3 provides a summary of the prevalence of individual NCDs. Table 2. Baseline laboratory parameters of study participants. p-Value Laboratory Parameters All HIV Only HIV/HBV HIV/TB N 275 225 26 24 5.1 (4.0–6.4) 5.1 (4.1–6.4) 5.3 (4.4–6.1) 4.8 (3.0–6.5) 0.745 Median (IQR) 11.2 (9.5–12.4) 11.1 (9.5–12.7) 11.7 (10.9–12.3) 10.1 (8.4–11.9) 0.204 Anemia 125/170 (73.5) 94/131 (71.8) 17/23 (73.9) 12/12 (85.7) 0.532 Leukocytes, ×109 /L, n (%) Median (IQR) Hemoglobin, g/dL, n (%) J. Clin. Med. 2022, 11, 3466 6 of 15 Table 2. Cont. All HIV Only HIV/HBV HIV/TB p-Value Median (IQR) 268 (209–344) 272 (198–347) 253 (225–306) 254 (223–410) 0.821 Thrombocytopenia 16/169 (9.5) 13/131 (9.9) 1/23 (4.3) 1/13 (7.7) 0.880 Median (IQR) 96 (79–143) 95 (77–135) 96 (78–168) 104 (87–135) 0.467 Elevated 52/161 (32.3) 36/124 (29.0) 10/23 (43.5) 4/12 (33.3) 0.379 Median (IQR) 33 (23–47) 32 (22–44) 34 (25–59) 44 (30–65) 0.276 Elevated 96/161 (59.6) 73/124 (58.9) 16/23 (69.6) 5/12 (41.7) 0.279 Median (IQR) 18 (13–27) 18 (13–27) 25 (18–47) 17 (13–20) 0.016 Elevated 35/161 (21.7) 25/124 (20.2) 10/23 (43.5) - 0.007 Median (IQR) 38 (27–59) 37 (28–53) 40 (23–112) 44 (30–65) 0.707 Elevated 71/127 (55.9) 54/99 (54.5) 9/18 (50.0) 7/9 (77.8) 0.356 Median (IQR) 1.8 (1.2–2.5) 1.8 (1.2–2.5) 1.3 (1.0–1.9) 2.0 (1.6–2.8) 0.064 >2.0 62/161 (38.5) 49/124 (39.5) 5/23 (21.7) 6/12 (50.0) 0.167 Median (IQR) 1.1 (0.8–1.8) 1.1 (0.7–1.8) 1.4 (1.0–2.2) 0.9 (0.6–1.0) 0.030 Elevated 70/161 (43.5) 55/124 (44.4) 13/23 (56.5) 2/12 (16.7) 0.078 Median (IQR) 0.4 (0.2–1.0) 0.4 (0.2–1.0) 0.5 (0.3–1.4) 0.3 (0.2–0.7) 0.233 Elevated 90/161 (55.9) 68/124 (54.8) 16/23 (69.6) 6/12 (50.0) 0.378 7.1 (6.8–7.7) 7.1 (6.8–7.7) 7.0 (6.7–7.7) 7.3 (6.9–8.4) 0.712 Median (IQR) 3.8 (3.7–3.9) 3.8 (3.7–3.9) 3.8 (3.6–4.0) 3.8 (3.6–4.2) 0.644 Hypoalbuminemia 16/153 (10.5) 12/121 (9.9) 3/20 (15.0) 1/11 (9.1) 0.781 Median (IQR) 4.7 (4.0–5.5) 4.6 (4.0–5.4) 4.8 (4.0–5.5) 6.1 (5.0–6.5) 0.063 <5.5 (normal) 102/137 (74.5) 82/106 (77.4) 16/20 (80.0) 2/9 (22.2) <0.001 5.6–6.9 (prediabetes) 28/137 (20.4) 18/106 (17.0) 3/20 (15.0) 7/9 (77.8) ≥7.0 (diabetes) 7/137 (5.1) 6/106 (5.7) 1/20 (5.0) - 0.9 (0.8–1.2) 0.9 (0.8–1.2) 1.0 (0.8–1.2) 1.0 (0.7–1.3) 0.884 Median (IQR) 99 (77–119) 97 (76–118) 109 (78–121) 100 (87–128) 0.569 ≥90 (Stage 1) 106/171 (62.0) 83/134 (61.9) 14/23 (60.9) 8/12 (66.7) 0.724 60–89 (Stage 2) 41/171 (24.0) 30/134 (22.4) 8/23 (34.8) 2/12 (16.7) 45–59 (Stage 3a) 6/171 (2.5) 4/134 (3.0) 1/23 (4.3) 1/12 (8.3) Laboratory Parameters Platelets, ×109 /L, n (%) ALP, U/L, n (%) AST, U/L, n (%) ALT, U/L, n (%) GGT, U/L, n (%) AST/ALT, n (%) Total bilirubin, mg/dL, n (%) Direct bilirubin, mg/dL, n (%) Total protein, mg/dL, n (%) Median (IQR) Albumin, mg/dL, n (%) Fasting blood glucose, mmol/L, n (%) Serum creatinine, mg/dL, n (%) Median (IQR) eGFR, mL/min/1.73 m2 , n (%) J. Clin. Med. 2022, 11, 3466 7 of 15 Table 2. Cont. Laboratory Parameters p-Value All HIV Only HIV/HBV HIV/TB 30–44 (Stage 3b) 3/171 (1.8) 3/134 (2.2) - - 29–15 (Stage 4) 2/171 (0.7) 2/134 (1.5) - - <15 (Stage 5) 13/171 (7.6) 12/134 (9.0) - 1/12 (8.3) Proteinuria, n (%) 27/162 (16.7) 20/126 (15.9) 4/22 (18.2) 3/12 (25.0) 0.763 Median (IQR) 0.4 (0.2–0.5) 0.3 (0.2–0.5) 0.4 (0.3–0.9) 0.4 (0.2–0.6) 0.296 <0.5 (normal liver) 113/159 (71.1) 90/122 (73.8) 14/23 (60.9) 9/12 (75.0) 0.421 0.5–1.5 (fibrosis) 46/159 (29.0) 31/122 (25.4) 8/23 (34.8) 3/12 (25.0) >1.5 (cirrhosis) 2/159 (1.3) 1/22 (0.8) 1/23 (4.3) - Median (IQR) 1.0 (0.6–1.4) 1.0 (0.6–1.4) 0.8 (0.6–1.4) 0.9 (0.6–1.3) 0.930 <1.45 (normal liver) 121/159 (76.1) 96/122 (78.7) 15/23 (65.2) 10/12 (83.3) 0.481 1.45–3.25 (fibrosis) 38/159 (23.9) 23/122 (18.9) 8/23 (34.8) 2/12 (16.7) >3.25 (cirrhosis) 4/159 (1.5) 3/122 (2.5) - - Median (IQR) 0.45 (0.29–0.72) 0.41 (0.29–0.65) 0.49 (0.25–1.22) 0.51 (0.45–0.69) 0.430 <0.32 (normal liver) 92/124 (74.2) 72/97 (74.2) 12/18 (66.7) 7/8 (87.5) 0.532 ≥0.32 (fibrosis) 32/124 (25.8) 25/97 (25.8) 6/18 (33.3) 1/8 (12.5) eGFR, mL/min/1.73 m2 , n (%) APRI, n (%) FIB-4 score, n (%) GPR, n (%) Abbreviations: ALP, alkaline phosphatase; ALT, alanine transaminase; APRI, aspartate transaminase to platelet index; AST, aspartate transaminase; AST/ALT, aspartate transaminase to alanine transaminase ratio; BMI, body mass index; eGFR, estimated glomerular filtration rate; FIB-4, fibrosis-4 score; GGT, gamma-glutamyl transferase; GPR, gamma-glutamyl transferase to platelet ratio; HBV, hepatitis B virus, HIV, human immunodeficiency virus; IQR, interquartile range. Table 3. Prevalence of NCDs by type. Type of NCD N (%) ≥1 NCD 113/275 (41.5) ≥2 NCDs 48/275 (17.5) Liver fibrosis/cirrhosis (APRI > 0.5) 48/159 (30.3) Diabetes/prediabetes (FBG > 5.5 mmol/L) 35/137 (25.5) Preobesity/obesity (BMI ≥ 25 kg/m2 ) Underweight/malnutrition (BMI < 18.5 63/275 (22.9) kg/m2 ) 48/275 (17.5) m2 ) 24/171 (12.6) Renal impairment (eGFR < 60 mL/min/1.72 Hypertension * Elevated blood pressure 27/275 (9.8) Stage 1 7/275 (2.5) Stage 2 28/275 (10.2) Chronic lung disease 5/275 (1.8) Abbreviations: APRI, aspartate transaminase to platelet index; BMI, body mass index; eGFR, estimated glomerular filtration rate; FBG, fasting blood glucose; * hypertension classification was based on the AHA 2019 guidelines [38]. J. Clin. Med. 2022, 11, 3466 8 of 15 3.3. Malnutrition, Preobesity and Obesity Based on BMI measurements, 17.5% (48/275) were underweight (malnourished), while 22.9% (63/275) were classified as having obesity/preobesity (Table 2). 3.4. Hypertension and Chronic Lung Disease About 9.8% (27/275) had elevated BP, 2.5% (7/275) had stage 1 HTN and 10.2% (28/275) had stage 2 HTN. The overall prevalence of chronic lung diseases (i.e., asthma and chronic obstructive lung disease) was low, at 1.8% (5/275) (Table 2). 3.5. Anemia and Thrombocytopenia Overall, 73.5% (125/170) had anemia, while a relatively smaller proportion (9.5%, 16/169) had thrombocytopenia. There were no differences in median values based on coinfection status (Table 2). 3.6. Derangements in Liver Function Tests and Alcoholic Liver Disease A large proportion of patients had deranged LFTs at baseline (Table 2). Overall, 32.3% (52/161) had elevated ALP, while 59.6% (96/161), 21.7% (35/161) and 55.9% (71/161) had elevated AST, ALT and GGT, respectively. Similarly, many participants had elevated total bilirubin (43.5%, 70/161) and direct bilirubin (55.9%, 90/161), while 10.5% (16/153) had hypoalbuminemia. There was no difference in LFT levels based on coinfection status, except for ALT (p < 0.016) and total bilirubin (p = 0.030). About 23.6% (65/275) reported regular alcohol use, while 38.5% (62/161) had AST/ALT > 2.0, suggesting a higher prevalence of alcoholic liver disease than was self-reported. 3.7. Liver Fibrosis and Cirrhosis The median APRI score was 0.4 (IQR 0.2–0.5) (Table 2). The majority (71.1%, 113/159) had mild to significant fibrosis, while 1.3% (4/159) had cirrhosis. Across groups, HIV/HBV coinfected individuals tended to have a higher prevalence of liver fibrosis (34.8% vs. 29.0%) and cirrhosis (4.3% vs. 1.3%) without achieving statistical significance (p = 0.421). Similar results were observed for the overall prevalence of fibrosis using the FIB-4 score (23.9%, 38/159) and the GPR (25.8%, 32/124). 3.8. Kidney Disease and Proteinuria The median serum creatinine and eGFR were 0.9 mg/L and 99 mL/min/1.73 m2 , respectively, with no statistically significant differences observed across groups based on coinfection status (Table 2). About 12.6% (24/171) were classified as having renal impairment, with 7.6% (13/171) having end-stage renal disease (ESRD). Additionally, 16.7% (27/162) had proteinuria (urine dipstick ≥ +1). 3.9. Prediabetes and Diabetes Mellitus The median fasting blood glucose was 4.7 mmol/L. The prevalence of DM was 5.1% (7/137) and the prevalence of prediabetes was 20.4% (28/137). Individuals with TB had significantly higher rates of prediabetes (77.8% vs. 20.4%, p < 0.001) (Table 2). 3.10. Predictors of Liver Fibrosis, Renal Impairment and Impaired Glucose Metabolism Table 4 displays the results of univariate and multivariate regression analysis of factors associated with NCDs. In logistic regression analysis, liver fibrosis/cirrhosis was independently associated with HBV infection (aOR 8.80, 95% [2.45–31.45]; p < 0.001) and DM/prediabetes (aOR 9.89, 95% CI [1.14–85.67]; p = 0.037). DM/prediabetes was predicted by age > 35 years (aOR 2.43, 95% CI [1.01–5.84]; p = 0.047) and TB infection (aOR 7.34, 95% CI [1.87–28.74]; p = 0.004]. Lastly, renal impairment was independently associated with proteinuria (aOR 9.37, 95% CI [2.01–43.78]; p = 0.004) and SBP > 140 mmHg (aOR 6.00, 95% CI [1.01–35.39]; p = 0.049). The full univariate and multivariate analysis for associations between HBV, TB and NCDs are presented in the Supplementary Materials (Tables S1–S3). J. Clin. Med. 2022, 11, 3466 9 of 15 Table 4. Factors associated with NCDs. n (%) Type of NCD Univariate Analysis Multivariate Analysis Yes No Crude Odds Ratio (95% CI) p-Value Adjusted Odds Ratio (95% CI) p-Value Yes 10 (43.5) 15 (11.4) 6.00 (2.24–16.05) <0.001 8.80 (2.46–31.45) <0.001 No 13 (56.5) 117 (88.6) Ref Yes 20 (95.2) 82 (71.3) 8.05 (1.04–62.44) No 1 (4.8) 33 (28.7) Ref Yes 22 (62.9) 44 (43.1) 2.23 (1.01–4.91) No 13 (37.1) 58 (56.9) Ref Yes 7 (20.0) 4 (3.9) 6.13 (1.67–22.44) No 28 (80.0) 98 (96.1) Ref Yes 16 (66.7) 61 (41.5) 2.82 (1.14–7.00) No 8 (33.3) 86 (58.5) Ref Yes 6 (25.0) 12 (8.2) 3.72 (1.24–11.11) No 18 (75.0) 134 (91.8) Ref Risk Factors HBV Liver fibrosis (APRI > 0.7) Ref Diabetes/prediabetes 0.020 9.89 (1.14–85.67) 0.037 Ref Age > 35 years Diabetes/prediabetes (FBG > 5.5 mmol/L) 0.044 2.43 (1.01–5.84) 0.047 Ref Tuberculosis 0.003 7.34 (1.87–28.74) 0.004 Ref Age > 35 years Renal impairment (eGFR < 60 mL/min/1.72m2 ) 0.022 5.90 (1.12–31.10) 0.036 Ref SBP > 140 mmHg 0.013 5.22 (1.17–23.33) 0.031 Ref Abbreviations: APRI, aspartate transaminase to platelet index; eGFR, estimated glomerular filtration rate; FBG, fasting blood glucose; Ref, reference category; SBP, systolic blood pressure. 4. Discussion In this study, we estimated the burden and associated risk factors of common NCDs in a cross-section of newly diagnosed PWH prior to the initiation of ART in Freetown, Sierra Leone. The primary aim of the study was to identify opportunities for early intervention in the management of comorbidities, in a bid to foster greater integration of HIV and NCD care. Despite being a relatively young cohort (median age 33 years), the prevalence of NCDs was substantial and bimodally distributed, with about 1 in every 4 PWH screened clustered around liver disease (fibrosis/cirrhosis), impaired glucose intolerance (DM/prediabetes) and obesity/preobesity, while about 1 in 8 had renal impairment or HTN. The prevalence of chronic lung disease was low, affecting only 1.8%. Overall, 41.5% had at least one NCD, while 17.5% were multimorbid (≥2 NCDs). Previous studies from SSA have revealed similarly high rates of NCDs among PWH. A recent study from Uganda by Kansiime et al. [39] found that 20.3% of PWH initiating ART had at least one NCD. Similarly, Ekrikpo et al. [40] recently reported an NCD prevalence rate of about 25% among ART-naïve PWH in Nigeria. In terms of individual NCDs, liver morbidity was the most significant baseline finding in our cohort. This was reflected in the high proportion of patients showing evidence of hepatic necroinflammatory activity, with 21.7% to 59.6% having elevated AST, ALT and GGT in the setting of a high prevalence of HIV/HBV coinfection (i.e., 15.3%). Similarly, the prevalence of liver fibrosis/cirrhosis was high (range, 23.9% to 30.3%) and was strongly predicted by HIV/HBV coinfection, which increased the risk of liver fibrosis almost 9-fold. As previously discussed, HBV is hyperendemic in Sierra Leone, with studies suggesting J. Clin. Med. 2022, 11, 3466 10 of 15 a 1.5- to 2-fold higher risk of HBV infection among PWH compared with the general population [24,25]. The high prevalence of HIV/HBV coinfection and liver morbidity among our patients has two major implications. Firstly, the elevated LFTs (i.e., ALT, AST and GGT) at baseline present challenges for the selection of regimens for the treatment of HIV, HBV and TB, given the high risk of idiosyncratic drug-related hepatoxicities commonly encountered with regimens used in the treatment of all three infections [41,42]. Secondly, given the lack of advanced diagnostic facilities in Sierra Leone for the assessment of liver disease such as liver biopsy and transient elastography, major investments are urgently needed in this area to facilitate early diagnosis and linkage to care. About 5.1% of our cohort had DM, while 20.4% had prediabetes. DM affects 422 million people globally and is responsible for 1.5 million deaths annually [43]. The incidence of DM and related conditions such as obesity, the metabolic syndrome and CVD have been rising in the last three decades, with the impact most dramatically felt in LMICs [43]. Consequently, many LMICs have been prioritizing early screening for prediabetes, which offers opportunities for halting progression to DM through risk reduction interventions such as dietary modifications and increased physical activity. However, many LMICs lack data to help inform policy. To date, only three studies have addressed the prevalence of DM in the general population in Sierra Leone, which have ranged from 2.4% to 7.0% [44–46]. To the best of our knowledge, our study is the first to document the prevalence of DM among PWH in the country. Considerable controversy exists around the association between DM/prediabetes and HIV, with some studies reporting HIV infection as an independent risk factor for the development of DM/prediabetes [47,48], while other studies have failed to demonstrate any differential effect [49,50]. To complicate the picture further, HIV-induced inflammation and ART-related toxicities (e.g., lipodystrophy seen with protease inhibitor use and INSTIrelated weight gain with its resultant cardiometabolic complications) have been implicated in promoting hyperglycemia and increased insulin resistance [10,51]. Additionally, many of the risk factors traditionally associated with the development of DM/prediabetes in the general population (e.g., older age, increased BMI, genetics) also tend to be highly prevalent among PWH. However, in our study, we did not detect statistically significant associations among these factors, despite the well-described link among increased BMI, DM/prediabetes and CVD [52,53]. Studies have shown that waist circumference alone or combined with BMI is a better predictor of cardiovascular and all-cause morbidity and mortality risk than BMI alone, especially at a BMI < 35 kg/m2 [54,55], which could partially explain our findings. We also found that patients with DM/prediabetes were at 10 times higher risk of having liver fibrosis/cirrhosis, compared with their normoglycemic counterparts. Our findings align with several studies that have correlated poor glycemic control with liver disease, especially in cirrhotic patients [56,57]. The mechanism underlying this phenomenon appears to be tied to the central role of the liver in regulating glucose metabolism through glycogenesis and glycogenolysis [58]. In liver disease, glucose metabolism shifts from the liver to muscle and adipose tissue, which stimulates mitochondrial oxidative stress and drives the production of proinflammatory adipokines, e.g., leptin, tumor necrosis factors-alpha and interleukin-6 [59]. Adipokine release leads to the activation of hepatic stellate cells and the excess production and deposition of collagen and extracellular matrix, culminating in liver fibrosis [60,61]. Another important finding in our study was that PWH with TB had a 7.34-fold higher risk of DM/prediabetes compared with patients without TB. DM is a growing global health concern among people with TB, with a recent meta-analysis by Noubiap et al. [62] estimating that 15.3% of TB patients worldwide had DM. The association between TB and DM is well known; however, data are conflicting on the direction of the relationship. Diabetics are 1.5 to 3 times more likely to develop active TB [63,64]. Conversely, a meta-analysis by Menon et al. [65] showed that up to 50% of people diagnosed with TB had hyperglycemia J. Clin. Med. 2022, 11, 3466 11 of 15 at baseline, which remained unresolved in over 10% of cases even after 6 months of effective anti-TB treatment. Although the pathophysiologic basis of hyperglycemia/DM in TB has not been fully elucidated, a role has been postulated for adipokine-mediated alterations in the microenvironment, resulting in increased susceptibility to hyperglycemia through the release of counterregulatory stress hormones (e.g., glucagon, cortisol and growth hormone) [66,67]. Others have suggested that TB-induced impaired cell-mediated immune responses have a direct effect on pancreatic islet cells, resulting in endocrine hypofunction, including a reduction in insulin production [68]. In response to the rising rates of DM in TB-endemic regions, the World Health Organization (WHO) has proposed a collaborative framework, with the aim of achieving greater integration of DM and TB care in LMICs [69]. Renal impairment and HTN (stages 1 and 2) occurred at equal rates in our cohort, affecting about 12.6% (1 in every 8) of PWH. While a proportion of patients will see an improvement in renal function with treatment, many are likely to progress to chronic kidney disease (CKD). CKD is common in HIV infection and is estimated to affect 6.4% of PWH globally, with countries in West Africa reporting the highest prevalence (i.e., 14.6%) [70]. Proteinuria was present in 16.7% of our patients and conferred a 9.34-fold higher risk of kidney disease, suggesting HIV-associated nephropathy (HIVAN) as a major contributor to kidney disease in this setting. HIVAN has a predilection for African populations and has been linked with high-risk variants of the APOL1 gene (i.e., G1 and G2), which have been associated with several nondiabetic glomerular disorders including HIVAN [70,71]. Studies in West African populations, notably Ghana and Nigeria, have found APOL1 allele frequencies of up to 40% [72,73]. Notwithstanding this, CKD was also strongly predicted by HTN (SBP > 140 mmHg) by a factor of 6, in line with the observation that renal disease tends to be multifactorial. However, we did not detect statistically significant associations with other factors traditionally linked with kidney disease such as age, DM or HBV status. Our findings have implications for the use of tenofovir disoproxil fumarate (TDF) in the treatment of HIV and HBV, given that less nephrotoxic treatment options such as tenofovir alafenamide fumarate (TAF) and emtricitabine (FTC) are currently not available in Sierra Leone and many other LMICs. Our study had a few limitations worthy of discussion. Firstly, the study was restricted to a single treatment center and may not be generalizable to other geographic locations in Sierra Leone. Secondly, we used non-invasive scores to assess liver disease due to the unavailability of more precise diagnostic tools such as liver biopsy and transient elastography in Sierra Leone. To improve the precision of our estimates for the prevalence of liver disease, we used three non-invasive scores that have been well-validated in West Africa [28–32]. Thirdly, we were unable to fully assess CVD risk, given that the lipid profile and waist circumference, which are better indicators of health risk, were not routinely measured in this relatively young cohort. Finally, we were unable to assess the impact of HIV viremia on NCDs since, at baseline, HIV viral load measurement was not routinely available prior to ART initiation. However, we do not believe that this is a major limitation. As most patients presented with late-stage HIV disease (60% with CD4 < 350 cells/mm3 ), we assumed that most would have had a high viral load at diagnosis, given the inverse relationship of CD4 count and HIV viremia in untreated HIV infection [74]. This would have likely made it difficult to discern any differential effect between NCDs and HIV viremia. Despite these limitations, our study is the first to characterize NCDs among PWH in Sierra Leone and offers important insights into the link between coexisting endemic infections to NCDs in this setting. This warrants the implementation of HIV programs emphasizing greater integrative care in LMICs, with a focus on the early detection and management of comorbidities. 5. Conclusions In summary, our study found a high burden of NCDs among PWH prior to the initiation of ART in Freetown, Sierra Leone, with 1 in every 4 who were screened cluster- J. Clin. Med. 2022, 11, 3466 12 of 15 ing around liver fibrosis/cirrhosis, DM/prediabetes and obesity/preobesity, while about 1 in 8 had CKD and/or HTN. The prevalence of chronic lung disease was low, affecting only 1.8%. Overall, 41.5% had at least one NCD, while 17.5% were multimorbid (≥2 NCDs). HBV coinfection increased the risk of liver disease (fibrosis/cirrhosis) 9-fold, while TB coinfection conferred a 7.3-fold increase in the risk of DM/prediabetes, which was in turn associated with a 10-fold increase in the risk of liver disease. Larger studies are needed to confirm these findings. Additionally, given the substantially high burden of NCDs among this relatively young cohort of PWH and their link to HBV and TB coinfections, this warrants the implementation of HIV programs emphasizing greater integrative care in Sierra Leone and other LMICs, with a renewed focus on the prevention, early detection and management of comorbidities. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcm11123466/s1, Table S1: Factors associated with liver fibrosis. Table S2: Factors associated with renal impairment. Table S3: Factors associated with impaired glucose metabolism. Author Contributions: Conceptualization, G.A.Y., S.L. and R.A.S.; methodology, G.A.Y., S.L. and R.A.S.; software, G.A.Y.; validation, G.A.Y.; formal analysis, G.A.Y.; investigation, S.L., D.F.J., S.A.Y., U.B., G.F.D., M.S., F.S. and G.A.Y.; resources, G.A.Y., S.L., J.M.J. and R.A.S.; writing—original draft preparation, G.A.Y.; writing—review and editing, all authors.; visualization, G.A.Y.; supervision, G.A.Y. and S.L.; project administration, G.A.Y. and S.L.; funding acquisition, G.A.Y. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by grants to G.A.Y. from the National Institutes of Health (NIH)/AIDS Clinical Trials Group (ACTG) under Award Numbers 5UM1AI068636-15 and 5UM1AI069501-09, the Roe Green Center for Travel Medicine and Global Health/University Hospitals Cleveland Medical Center Award Number J0713 and the University Hospitals Minority Faculty Career Development Award/University Hospitals Cleveland Medical Center Award Number P0603. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki and ethical approval was obtained from the Sierra Leone Ethics and Scientific Review Committee (approval date 9 February 2020). Written informed consent was obtained from all study participants prior to enrolment. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Conflicts of Interest: The authors declare no conflict of interest. References 1. 2. 3. 4. 5. 6. 7. World Health Organization. HIV/AIDS. 2022. Available online: https://www.who.int/news-room/fact-sheets/detail/hiv-aids (accessed on 10 February 2022). World Health Organization. Hepatitis, B. 2022. Available online: https://www.who.int/news-room/fact-sheets/detail/hepatitisb (accessed on 10 February 2022). World Health Organization. Tuberculosis. 2022. Available online: https://www.who.int/health-topics/tuberculosis#tab=tab_1 (accessed on 10 February 2022). Hoffmann, C.J.; Charalambous, S.; Thio, C.L.; Martin, D.J.; Pemba, L.; Fielding, K.L.; Churchyard, G.J.; E Chaisson, R.; Grant, A.D. Hepatotoxicity in an African antiretroviral therapy cohort: The effect of tuberculosis and hepatitis B. AIDS 2007, 21, 1301–1308. [CrossRef] [PubMed] Eskild, A.; Magnus, P.; Petersen, G.; Sohlberg, C.; Jensen, F.; Kittelsen, P.; Skaugt, K. Hepatitis B antibodies in HIV-infected homosexual men are associated with more rapid progression to AIDS. AIDS 1992, 6, 571–574. [CrossRef] Klein, M.B.; Althoff, K.N.; Jing, Y.; Lau, B.; Kitahata, M.M.; Freeman, A.M.; Kirk, G.D.; Hull, M.; Kim, H.N.; Sebastiani, G.; et al. Risk of End-Stage Liver Disease in HIV-Viral Hepatitis Coinfected Persons in North America From the Early to Modern Antiretroviral Therapy Eras. Clin. Infect. Dis. 2016, 63, 1160–1167. [CrossRef] [PubMed] Jespersen, N.A.; Axelsen, F.; Dollerup, J.; Nørgaard, M.; Larsen, C.S. The burden of non-communicable diseases and mortality in people living with HIV (PLHIV) in the pre-, early- and late-HAART era. HIV Med. 2021, 22, 478–490. [CrossRef] [PubMed] J. Clin. Med. 2022, 11, 3466 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 13 of 15 World Health Organization. Noncommunicable Diseases. 2022. Available online: https://www.who.int/news-room/fact-sheets/ detail/noncommunicable-diseases (accessed on 10 February 2022). Patel, P.; Rose, C.E.; Collins, P.Y.; Nuche-Berenguer, B.; Sahasrabuddhe, V.V.; Peprah, E.; Vorkoper, S.; Pastakia, S.D.; Rausch, D.; Levitt, N.S. Noncommunicable diseases among HIV-infected persons in low-income and mid-dle-income countries: A systematic review and meta-analysis. AIDS 2018, 32 (Suppl. 1), S5–S20. [CrossRef] Currier, J.S.; Lundgren, J.D.; Carr, A.; Klein, D.; Sabin, C.A.; Sax, P.E.; Schouten, J.T.; Smieja, M.; Working Group. Epidemiological evidence for cardiovascular disease in HIV-infected patients and relationship to highly active antiretroviral therapy. Circulation 2008, 118, e29–e35, Erratum in Circulation 2008, 118, e108. [CrossRef] Cohen, S.D.; Kopp, J.B.; Kimmel, P.L. Kidney Diseases Associated with Human Immunodeficiency Virus Infection. N. Engl. J. Med. 2017, 377, 2363–2374, Erratum in N. Engl. J. Med. 2018, 378, 1657. [CrossRef] Sax, P.E.; Erlandson, K.M.; Lake, J.E.; McComsey, G.A.; Orkin, C.; Esser, S.; Brown, T.T.; Rockstroh, J.K.; Wei, X.; Carter, C.C.; et al. Weight Gain Following Initiation of Antiretroviral Therapy: Risk Factors in Randomized Comparative Clinical Trials. Clin. Infect. Dis. 2020, 71, 1379–1389. [CrossRef] Coates, M.M.; Kintu, A.; Gupta, N.; Wroe, E.B.; Adler, A.J.; Kwan, G.F.; Park, P.H.; Rajbhandari, R.; Byrne, A.L.; Casey, D.C.; et al. Burden of non-communicable diseases from infectious causes in 2017: A modelling study. Lancet Glob. Health 2020, 8, e1489–e1498. [CrossRef] Stubbs, B.; Siddiqi, K.; Elsey, H.; Siddiqi, N.; Ma, R.; Romano, E.; Siddiqi, S.; Koyanagi, A. Tuberculosis and Non-Communicable Disease Multimorbidity: An Analysis of the World Health Survey in 48 Low- and Middle-Income Countries. Int. J. Environ. Res. Public Health 2021, 18, 2439. [CrossRef] Lakoh, S.; García-Tardón, N.; Adekanmbi, O.; van der Valk, M.; Smith, S.J.; Grobusch, M.P. Prevalence of viral hepatitis B and C in Sierra Leone—current knowledge and knowledge gaps: A narrative review. Trans. R. Soc. Trop. Med. Hyg. 2021, 115, 1106–1113. [CrossRef] [PubMed] Tognon, F.; Sevalie, S.; Gassimu, J.; Sesay, J.; Hann, K.; Sheku, M.; Bearse, E.; Di Gennaro, F.; Marotta, C.; Pellizzer, G.; et al. Seroprevalence of hepatitis B and hepatitis C among blood donors in Sierra Leone: A multi-year retrospective study. Int. J. Infect. Dis. 2020, 99, 102–107. [CrossRef] [PubMed] Massaquoi, T.A.; Burke, R.M.; Yang, G.; Lakoh, S.; Sevalie, S.; Li, B.; Jia, H.; Huang, L.; Deen, G.F.; Beynon, F.; et al. Cross sectional study of chronic hepatitis B prevalence among healthcare workers in an urban setting, Sierra Leone. PLoS ONE 2018, 13, e0201820. [CrossRef] [PubMed] World Health Organization. Global Tuberculosis Report 2021. 2022. Available online: https://www.who.int/publications/i/ item/9789240037021 (accessed on 10 February 2022). Yendewa, G.A.; Poveda, E.; Yendewa, S.A.; Sahr, F.; Quiñones-Mateu, M.E.; Salata, R.A. HIV/AIDS in Sierra Leone: Characterizing the Hidden Epidemic. Aids Rev. 2018, 20, 104–113. [CrossRef] [PubMed] Yendewa, G.A.; Poveda, E.; Lakoh, S.; Yendewa, S.A.; Jiba, D.F.; Salgado-Barreira, A.; Sahr, F.; Salata, R.A. High prevalence of late-stage disease in newly diagnosed HIV patients in Sierra Leone. Open Forum Infect. Dis. 2018, 5, ofy208. [CrossRef] A Yendewa, G.; Sahr, F.; Lakoh, S.; Ruiz, M.; Patiño, L.; Tabernilla, A.; Deen, G.F.; Sesay, M.; A Salata, R.; Poveda, E. Prevalence of drug resistance mutations among ART-naive and -experienced HIV-infected patients in Sierra Leone. J. Antimicrob. Chemother. 2019, 74, 2024–2029. [CrossRef] Yendewa, G.A.; Lakoh, S.; Yendewa, S.A.; Bangura, K.; Tabernilla, A.; Patiño, L.; Jiba, D.F.; Vandy, A.O.; Massaquoi, S.P.; Osório, N.S.; et al. Characterizing HIV-1 Genetic Subtypes and Drug Resistance Mutations among Children, Adolescents and Pregnant Women in Sierra Leone. Genes 2021, 12, 1314. [CrossRef] Lakoh, S.; Jiba, D.F.; Vandy, A.O.; Poveda, E.; Adekanmbi, O.; Murray, M.J.; Deen, G.F.; Sahr, F.; Hoffmann, C.J.; Jacobson, J.M. Assessing eligibility for differentiated service delivery, HIV services utilization and vi-rologic outcomes of adult HIV-infected patients in Sierra Leone: A pre-implementation analysis. Glob. Health Action 2021, 14, 1947566. [CrossRef] Yendewa, G.A.; Sahr, F.; Aguilera, A.; Lakoh, S.; Sesay, M.; Deen, G.F.; Patiño, L.; Poveda, E.; Salata, R.A. Seroprevalence of hepatitis B, hepatitis C and human T-cell lymphotropic virus infections in human immunodeficiency virus-infected patients in Sierra Leone. Am. J. Trop. Med. Hyg. 2019, 100, 1521–1524. [CrossRef] Yendewa, G.A.; Lakoh, S.; Yendewa, S.A.; Bangura, K.; Lawrence, H.; Patiño, L.; Jiba, D.F.; Vandy, A.O.; Murray, M.J.; Massaquoi, S.P.; et al. Prevalence of hepatitis B surface antigen and serological markers of other endemic infections in HIVinfected children, adolescents and pregnant women in Sierra Leone: A cross-sectional study. Int. J. Infect. Dis. 2020, 102, 45–52. [CrossRef] Lakoh, S.; Jiba, D.F.; Kanu, J.E.; Poveda, E.; Salgado-Barreira, A.; Sahr, F.; Sesay, M.; Deen, G.F.; Sesay, T.; Gashau, W.; et al. Causes of hospitalization and predictors of HIV-associated mortality at the main referral hospital in Sierra Leone: A prospective study. BMC Public Health 2019, 19, 1320. [CrossRef] [PubMed] Lakoh, S.; Jiba, D.F.; Adekanmbi, O.; Poveda, E.; Sahr, F.; Deen, G.F.; Foray, L.M.; Gashau, W.; Hoffmann, C.J.; Salata, R.A.; et al. Diagnosis and treatment outcomes of adult tuberculosis in an urban setting with high HIV prevalence in Sierra Leone: A retrospective study. Int. J. Infect. Dis. 2020, 96, 112–118. [CrossRef] [PubMed] Terrault, N.A.; Lok, A.S.; McMahon, B.J.; Chang, K.M.; Hwang, J.P.; Jonas, M.M.; Brown Jr, R.S.; Bzowej, N.H.; Wong, J.B. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology 2018, 67, 1560–1599. [CrossRef] [PubMed] J. Clin. Med. 2022, 11, 3466 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 14 of 15 European Association for the Study of the Liver. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J. Hepatol. 2017, 67, 370–398. [CrossRef] [PubMed] Hui, A.Y.; Chan, H.L.-Y.; Wong, V.W.-S.; Liew, C.-T.; Chim, A.M.-L.; Chan, F.K.-L.; Sung, J.J.-Y. Identification of Chronic Hepatitis B Patients without Significant Liver Fibrosis by a Simple Noninvasive Predictive Model. Am. J. Gastroenterol. 2005, 100, 616–623. [CrossRef] [PubMed] Lin, Z.-H.; Xin, Y.-N.; Dong, Q.-J.; Wang, Q.; Jiang, X.-J.; Zhan, S.-H.; Sun, Y.; Xuan, S.-Y. Performance of the aspartate aminotransferase-to-platelet ratio index for the staging of hepatitis C-related fibrosis: An updated meta-analysis. Hepatology 2011, 53, 726–736. [CrossRef] Lemoine, M.; Shimakawa, Y.; Nayagam, S.; Khalil, M.; Suso, P.; Lloyd, J.; Goldin, R.; Njai, H.-F.; Ndow, G.; Taal, M.; et al. The gamma-glutamyl transpeptidase to platelet ratio (GPR) predicts significant liver fibrosis and cirrhosis in patients with chronic HBV infection in West Africa. Gut 2016, 65, 1369–1376. [CrossRef] Limdi, J.K.; Hyde, G.M. Evaluation of abnormal liver function tests. Postgrad Med. J. 2003, 79, 307–312. [CrossRef] Bonnard, P.; Sombié, R.; Lescure, F.X.; Bougouma, A.; Guiard-Schmid, J.B.; Poynard, T.; Calès, P.; Housset, C.; Callard, P.; Le Pendeven, C.; et al. Comparison of Elastography, Serum Marker Scores, and Histology for the Assessment of Liver Fibrosis in Hepatitis B Virus (HBV)-Infected Patients in Burkina Faso. Am. J. Trop. Med. Hyg. 2010, 82, 454–458. [CrossRef] Levey, A.S.; Coresh, J.; Greene, T.; Stevens, L.A.; Zhang, Y.L.; Hendriksen, S.; Kusek, J.W.; Van Lente, F.; Chronic Kidney Disease Epidemiology Collaboration. Using Standardized Serum Creatinine Values in the Modification of Diet in Renal Disease Study Equation for Estimating Glomerular Filtration Rate. Ann. Intern. Med. 2006, 145, 247–254. [CrossRef] Stevens, L.A.; Coresh, J.; Feldman, H.I.; Greene, T.; Lash, J.P.; Nelson, R.G.; Rahman, M.; Deysher, A.E.; Zhang, Y.L.; Schmid, C.; et al. Evaluation of the Modification of Diet in Renal Disease Study Equation in a Large Diverse Population. J. Am. Soc. Nephrol. 2007, 18, 2749–2757. [CrossRef] [PubMed] American Diabetes Association Diagnosis and classification of diabetes mellitus. Diabetes Care 2014, 37 (Suppl. 1), S81–S90. [CrossRef] [PubMed] Whelton, P.K.; Carey, R.M.; Aronow, W.S.; Casey, D.E., Jr.; Collins, K.J.; Himmelfarb, C.D.; DePalma, S.M.; Gidding, S.; Jamerson, K.A.; Jones, D.W.; et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension 2018, 71, e13–e115. [CrossRef] [PubMed] Kansiime, S.; Mwesigire, D.; Mugerwa, H. Prevalence of non-communicable diseases among HIV positive patients on antiretroviral therapy at joint clinical research centre, Lubowa, Uganda. PLoS ONE 2019, 14, e0221022. [CrossRef] [PubMed] Ekrikpo, U.E.; Akpan, E.E.; Ekott, J.U.; Bello, A.K.; Okpechi, I.G.; Kengne, A.P. Prevalence and correlates of traditional risk factors for cardiovascular disease in a Nigerian ART-naive HIV population: A cross-sectional study. BMJ Open 2018, 8, e019664. [CrossRef] [PubMed] Otto, A.; Rivera, C.; Zeuli, J.; Temesgen, Z. Hepatotoxicity of Contemporary Antiretroviral Drugs: A Review and Evaluation of Published Clinical Data. Cells 2021, 10, 1263. [CrossRef] Ramappa, V.; Aithal, G.P. Hepatotoxicity Related to Anti-tuberculosis Drugs: Mechanisms and Management. J. Clin. Exp. Hepatol. 2013, 3, 37–49. [CrossRef] World Health Organization. Diabetes. 2022. Available online: https://www.who.int/health-topics/diabetes#tab=tab_1 (accessed on 25 February 2022). Ceesay, M.M.; Morgan, M.W.; Kamanda, M.O.; Willoughby, V.R.; Lisk, D.R. Prevalence of diabetes in rural and urban populations in southern Sierra Leone: A preliminary survey. Trop. Med. Int. Health 1997, 2, 272–277. [CrossRef] Odland, M.L.; Bockarie, T.; Wurie, H.; Ansumana, R.; Lamin, J.; Nugent, R.; Bakolis, I.; Witham, M.; Davies, J. Prevalence and access to care for cardiovascular risk factors in older people in Sierra Leone: A cross-sectional survey. BMJ Open 2020, 10, e038520. [CrossRef] Sundufu, A.J.; Bockarie, C.N.; Jacobsen, K.H. The prevalence of type 2 diabetes in urban Bo, Sierra Leone, and in the 16 countries of the West Africa region. Diabetes Metab. Res. Rev. 2017, 33, e2904. [CrossRef] Brown, T.T.; Cole, S.R.; Li, X.; Kingsley, L.A.; Palella, F.J.; Riddler, S.A.; Visscher, B.R.; Margolick, J.B.; Dobs, A.S. Antiretroviral therapy and the prevalence and incidence of diabetes mellitus in the multicenter AIDS cohort study. Arch. Intern. Med. 2005, 165, 1179–1184. [CrossRef] [PubMed] Galli, L.; Salpietro, S.; Pellicciotta, G.; Galliani, A.; Piatti, P.; Hasson, H.; Guffanti, M.; Gianotti, N.; Bigoloni, A.; Lazzarin, A.; et al. Risk of type 2 diabetes among HIV-infected and healthy subjects in Italy. Eur. J. Epidemiol. 2012, 27, 657–665. [CrossRef] [PubMed] Prioreschi, A.; Munthali, R.; Soepnel, L.; A Goldstein, J.; Micklesfield, L.; Aronoff, D.; Norris, S. Incidence and prevalence of type 2 diabetes mellitus with HIV infection in Africa: A systematic review and meta-analysis. BMJ Open 2017, 7, e013953. [CrossRef] [PubMed] Butt, A.A.; McGinnis, K.; Rodriguez-Barradas, M.C.; Crystal, S.; Simberkoff, M.; Goetz, M.; Leaf, D.; Justice, A.C. HIV infection and the risk of diabetes mellitus. AIDS 2009, 23, 1227–1234. [CrossRef] [PubMed] Samaras, K. The Burden of Diabetes and Hyperlipidemia in Treated HIV Infection and Approaches for Cardiometabolic Care. Curr. HIV/AIDS Rep. 2012, 9, 206–217. [CrossRef] J. Clin. Med. 2022, 11, 3466 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 15 of 15 Wilson, P.W.; D’Agostino, R.B.; Sullivan, L.; Parise, H.; Kannel, W.B. Overweight and obesity as determinants of cardiovascular risk: The Framingham experience. Arch. Intern. Med. 2002, 162, 1867–1872. [CrossRef] Gray, L.; Lee, I.-M.; Sesso, H.D.; Batty, G.D. Body weight in early and mid-adulthood in relation to subsequent coronary heart disease mortality: 80-year follow-up in the Harvard Alumni Study. Arch. Intern. Med. 2011, 171, 1768–1770. [CrossRef] Janssen, I.; Katzmarzyk, P.; Ross, R. Waist circumference and not body mass index explains obesity-related health risk. Am. J. Clin. Nutr. 2004, 79, 379–384. [CrossRef] Jacobs, E.J.; Newton, C.C.; Wang, Y.; Patel, A.V.; McCullough, M.L.; Campbell, P.T.; Thun, M.J.; Gapstur, S.M. Waist Circumference and All-Cause Mortality in a Large US Cohort. Arch. Intern. Med. 2010, 170, 1293–1301. [CrossRef] Li, X.; Jiao, Y.; Xing, Y.; Gao, P. Diabetes Mellitus and Risk of Hepatic Fibrosis/Cirrhosis. BioMed Res. Int. 2019, 2019, 5308308. [CrossRef] Alexopoulos, A.; Crowley, M.J.; Wang, Y.; Moylan, C.A.; Guy, C.D.; Henao, R.; Piercy, D.L.; Seymour, K.A.; Sudan, R.; Portenier, D.D.; et al. Glycemic Control Predicts Severity of Hepatocyte Ballooning and Hepatic Fibrosis in Nonalcoholic Fatty Liver Disease. Hepatology 2021, 74, 1220–1233. [CrossRef] [PubMed] Tappy, L.; Minehira, K. New data and new concepts on the role of the liver in glucose homeostasis. Curr. Opin. Clin. Nutr. Metab. Care 2001, 4, 273–277. [CrossRef] [PubMed] Bertolani, C.; Marra, F. The role of adipokines in liver fibrosis. Pathophysiology 2008, 15, 91–101. [CrossRef] [PubMed] Crespo, J.; Cayón, A.; Fernández-Gil, P.; Hernández-Guerra, M.; Mayorga, M.; Domínguez-Díez, A.; Fernández-Escalante, J.C.; Pons-Romero, F. Gene expression of tumor necrosis factor alpha and TNF-receptors, p55 and p75, in nonalcoholic steato-hepatitis patients. Hepatology 2001, 34, 1158–1163. [CrossRef] [PubMed] Lee, U.E.; Friedman, S.L. Mechanisms of hepatic fibrogenesis. Best Pract. Res. Clin. Gastroenterol. 2011, 25, 195–206. [CrossRef] Noubiap, J.J.; Nansseu, J.R.; Nyaga, U.F.; Nkeck, J.R.; Endomba, F.T.; Kaze, A.D.; Agbor, V.N.; Bigna, J.J. Global prevalence of diabetes in active tuberculosis: A systematic review and meta-analysis of data from 2·3 million patients with tuberculosis. Lancet Glob. Health 2019, 7, e448–e460. [CrossRef] Hayashi, S.; Chandramohan, D. Risk of active tuberculosis among people with diabetes mellitus: Systematic review and meta-analysis. Trop. Med. Int. Health 2018, 23, 1058–1070. [CrossRef] Menon, S.; Rossi, R.; Dusabimana, A.; Zdraveska, N.; Bhattacharyya, S.; Francis, J. The epidemiology of tuberculosis-associated hyperglycemia in individuals newly screened for type 2 diabetes mellitus: Systematic review and meta-analysis. BMC Infect. Dis. 2020, 20, 937. [CrossRef] Moideen, K.; Nair, D.; Banurekha, V.V.; Kumar, N.P.; Babu, S. Altered Systemic Adipokine Levels in Pulmonary Tuberculosis and Changes following Treatment. Am. J. Trop. Med. Hyg. 2018, 99, 875–880. [CrossRef] Yüksel, I.; Şencan, M.; Dökmetaş, H.S.; Dökmetaş, I.; Ataseven, H.; Yönem, Ö. The Relation Between Serum Leptin Levels and Body Fat Mass in Patients with Active Lung Tuberculosis. Endocr. Res. 2003, 29, 257–264. [CrossRef] Segura-Cerda, C.A.; López-Romero, W.; Flores-Valdez, M.A. Changes in Host Response to Mycobacterium tuberculosis Infec-tion Associated with Type 2 Diabetes: Beyond Hyperglycemia. Front. Cell Infect. Microbiol. 2019, 9, 342. [CrossRef] [PubMed] Moutschen, M.P.; Scheen, A.J.; Lefebvre, P.J. Impaired immune responses in diabetes mellitus: Analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabete Metab. 1992, 18, 187–201. [PubMed] World Health Organization. Collaborative Framework for Care and Control of Tuberculosis and Diabetes. 2011. Available online: https://apps.who.int/iris/bitstream/handle/10665/44698/9789241502252_eng.pdf (accessed on 25 February 2022). Ekrikpo, U.E.; Kengne, A.P.; Bello, A.K.; Effa, E.; Noubiap, J.J.; Salako, B.L.; Rayner, B.L.; Remuzzi, G.; Okpechi, I.G. Chronic kidney disease in the global adult HIV-infected population: A systematic review and meta-analysis. PLoS ONE 2018, 13, e0195443. [CrossRef] [PubMed] Kopp, J.B.; Nelson, G.W.; Sampath, K.; Johnson, R.C.; Genovese, G.; An, P.; Friedman, D.; Briggs, W.; Dart, R.; Korbet, S.; et al. APOL1 Genetic Variants in Focal Segmental Glomerulosclerosis and HIV-Associated Nephropathy. J. Am. Soc. Nephrol. 2011, 22, 2129–2137. [CrossRef] Kasembeli, A.N.; Duarte, R.; Ramsay, M.; Mosiane, P.; Dickens, C.; Dix-Peek, T.; Limou, S.; Sezgin, E.; Nelson, G.W.; Fogo, A.B.; et al. APOL1 Risk Variants Are Strongly Associated with HIV-Associated Nephropathy in Black South Africans. J. Am. Soc. Nephrol. 2015, 26, 2882–2890. [CrossRef] Ulasi, I.I.; Tzur, S.; Wasser, W.G.; Shemer, R.; Kruzel, E.; Feigin, E.; Ijoma, C.K.; Onodugo, O.D.; Okoye, J.U.; Arodiwe, E.B.; et al. High population frequencies of APOL1 risk variants are associated with increased prevalence of non-diabetic chronic kidney disease in the Igbo people from south-eastern Nigeria. Nephron Clin. Pract. 2013, 123, 123–128. [CrossRef] Epstein, F.H.; Pantaleo, G.; Graziosi, C.; Fauci, A.S. The Immunopathogenesis of Human Immunodeficiency Virus Infection. N. Engl. J. Med. 1993, 328, 327–335. [CrossRef]
https://openalex.org/W4220889378	https://link.springer.com/content/pdf/10.1007/s11154-022-09718-9.pdf	English	null	Fluorescence-guided detection of pituitary neuroendocrine tumor (PitNET) tissue during endoscopic transsphenoidal surgery available agents, their potential, and technical aspects	Reviews in endocrine and metabolic disorders	2,022	cc-by	8,587	Abstractf Differentiation of pituitary neuroendocrine tumor (PitNET) tissue from surrounding normal tissue during surgery is chal- lenging. A number of fluorescent agents is available for visualization of tissue discrepancy, with the potential of improving total tumor resection. This review evaluates the availability, clinical and technical applicability of the various fluorescent agents within the field of pituitary surgery. According to PRISMA guidelines, a systematic review was performed to identify reports describing results of in vivo application of fluorescent agents. In this review, 15 publications were included. Sodium Fluorescein (FNa) was considered in two studies. The first study reported noticeable fluorescence in adenoma tissue, the second demonstrated the strongest fluorescence in non-functioning pituitary adenomas. 5-Aminolevulinic acid (5-ALA) was investigated in three studies. One study compared laser-based optical biopsy system (OBS) with photo-diagnostic filter (PD) and found that the OBS was able to detect all microadenomas, even when MRI was negative. The second study retrospectively analyzed twelve pituitary adenomas and found only one positive for fluorescence. The third investigated fifteen pituitary adenomas of which one displayed vague fluorescence. Indocyanine green (ICG) was researched in four studies with variable results. Second-Window ICG yielded no significant difference between functioning and non-functioning adenomas in one study, while a second study displayed 4 times higher fluorescence in tumor tissue than in normal tissue. In three studies, OTL38 showed potential in non-functioning pituitary adenomas. At present, evidence for fluorescent agents to benefit total resection of PitNETs is lacking. OTL38 can potentially serve as a selective fluorescent agent in non-functioning pituitary adenomas in the near future. Keyword Transsphenoidal surgery—pituitary adenoma—intraoperative fluorescence—fluorescence-guided surgery Keyword Transsphenoidal surgery—pituitary adenoma—intraoperative fluorescence—fluorescence-guided surgery Fluorescence‑guided detection of pituitary neuroendocrine tumor (PitNET) tissue during endoscopic transsphenoidal surgery available agents, their potential, and technical aspects Rob A Vergeer1 · Robin E P Theunissen1 · Theodora van Elk1 · Iris Schmidt3 · Mark R Postma2 · Katalin Tamasi1,4 · J Marc C van Dijk1 · Jos M A Kuijlen1 Accepted: 3 March 2022 © The Author(s) 2022 / Published online: 28 March 2022 1 Department of Neurosurgery, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 1 Introduction Pituitary neuroendocrine tumors (PitNETs) are benign tumors arising from adenohypophyseal cells. They account for approximately 10–15% of all intracranial tumors and represent a heterogeneous group of tumors in their origin, growth patterns, and biological behavior [1–3]. PitNETs are divided into functioning and non-functioning. The latter are typically macroadenomas (≥ 10 mm) that generally present due to local space-occupying effects (optico-chiasmatic compression or cavernous sinus invasion), with or without hormone hyposecretion. Functioning PitNETs are mostly microadenomas (< 10 mm) and generally present with hypersecretory syndromes such as Cushing syndrome and acromegaly. Transsphenoidal surgery is the primary treat- ment of all PitNETs except for lactotroph tumors, which are https://doi.org/10.1007/s11154-022-09718-9 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 https://doi.org/10.1007/s11154-022-09718-9 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 * Rob A Vergeer ra.vergeer@umcg.nl * Rob A Vergeer ra.vergeer@umcg.nl ra.vergeer@umcg.nl 1 Department of Neurosurgery, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 2 Department of Endocrinology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 3 Department of Gastroenterology and Hepatology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 4 Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 1 Department of Neurosurgery, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 1 Department of Neurosurgery, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 2 Department of Endocrinology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 3 Department of Gastroenterology and Hepatology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 4 Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands (0121 3456789) 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 648 PubMed, 25 publications from Embase, 17 publications from Google Scholar and the Cochrane database showed no publications. Duplicated, non-English, and abstract-only articles were excluded. Publications were reviewed for meet- ing inclusion criteria: primarily treated with dopamine agonists. Surgery removes the tumor mass and thereby decompresses vital structures. Total resection (TR) is particularly important in functioning pituitary adenomas to achieve endocrine remission (ER). Traditionally, differentiation between normal tissue and pituitary adenoma is made by the neurosurgeon based on macroscopic characteristics, e.g. tissue color and consist- ency. As such, ER for ACTH-producing adenomas is around 70%, and for GH-producing tumors 50–60% [4–6]. There- fore, more objective intraoperative tools can be valuable in the treatment of patients with functioning PitNETs. 1. Regarding neurosurgery 2. Regarding adults 3. Regarding PitNET surgery 4. And/or regarding the technical aspects of fluorescent agent usel Strategies to augment resection rates vary from measures that enhance preoperative diagnosis to improve visualization during surgery. Intraoperative fluorescence has been proven to enhance intraoperative differentiation and is currently applied in multiple surgical-oncology specialties such as gastrointestinal surgery, for applications as sentinel lymph node mapping, tumor detection, and tissue perfusion [7]. In 1948, a fluorophore named Sodium Fluorescein (FNa) was for the first time used in the field of neurosurgery for localization of brain tumors [8, 9]. Since then, fluorescence was further explored and is nowadays routinely used in other surgical-oncology specialties. In PitNET surgery, this imag- ing tool has not proven its value yet and therefore it is not part of the standard/routine surgical armentarium for remov- ing PiTNET’s. * Rob A Vergeer ra.vergeer@umcg.nl The question remains if fluorescence guide surgery will be part of the routine surgical procedure for Pit- NET removal in future and thereby increasing TR and ER. 5. Use of fluorescent agents Two reviewers (T.v.E. and R.E.P.T.) independently screened the titles and abstracts of all records identi- fied. The full text of potentially eligible publications was retrieved and read for final selection. In case of disagree- ment, a third reviewer (R.A.V. or J.M.A.K.) was consulted to reach consensus. This process resulted in 15 publica- tions on fluorescent agent use in PitNETs and on technical aspects considered relevant for the topic. See Fig. 1 for a flow diagram. The results are described in this systematic literature review. 3.1 Sodium fluorescein 3. Which technical aspects are to be considered in the resection of PitNETs?li Sodium fluorescein (FNa) is an organic compound (C20H12O5) that functions as a dye. The light emitted from this yellow-green fluorophore is visible to the naked eye. FNa is used in the medical setting in ophthalmology for diagnostic use and in neurosurgery during the resection of high grade gliomas, metastatic lesions, treatment of cerebral aneurysms, and vascular malformations [11, 12]. FNa accu- mulation is the result of disturbed vascular permeability due to a disrupted blood–brain barrier [1]. 4. Does fluorescence-guided surgery provide a benefit to the resection of PitNETs in adults? 3 Results It is important to distinguish the various fluorophores in the way in which they act and how these fluorophores can be visualized. Fluorophores can be divided into 2 groups: (1) visible light fluorophores, e.g. Sodium Fluorescein and 5-ALA, that can be seen with the naked eye, thus preclud- ing additional imaging instruments, and (2) Near Infrared (NIR) fluorophores, such as ICG and OTL38. To visualize NIR fluorophores additional imaging instruments, such as an infrared camera/filters, are needed (or a dedicated NIR imag- ing system). In the past years there has been an exponential growth in the publication of articles pertaining to fluorescence-guided surgery [10]. The aim of this review is therefore to identify reports describing the availability of fluorescent agents as well as the technical considerations. 1. Which fluorophores are currently being studied for use in the resection of PitNETs?l 2. What is the mechanism of action of these fluorescent agents? 2 Methods Two separate systematic literature searches were conducted using PubMed, EmBase, and Cochrane (Table 1). Articles were reviewed in accordance with PRISMA guidelines. The search terms can be reviewed in table 1 in the appen- dix. All relevant articles until March 2021 were included in this review. The search resulted in 282 publications from Two separate systematic literature searches were conducted using PubMed, EmBase, and Cochrane (Table 1). Articles were reviewed in accordance with PRISMA guidelines. The search terms can be reviewed in table 1 in the appen- dix. All relevant articles until March 2021 were included in this review. The search resulted in 282 publications from In 2010, da Silva et al. performed a small observational FNa-study in skull base tumors(n = 6), which included only one PitNET. The authors reported a significant 1 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 649 Table 1 Search method Table 1 Search method Table 1 Search method Table 1 Search method SEARCH TERMS ("Adenoma"[Mesh] OR "Pituitary Neoplasms"[Mesh] OR pituitar[ab]) AND ("Fluorescent Dyes"[Mesh] OR "Fluorescent Dyes" [Pharmacological Acon] OR "Infrared Rays"[Mesh] OR infrared[ab] OR ﬂuorescen[ab]) AND ("Surgery, Computer-Assisted"[Mesh] OR "surgery" [Subheading] OR surg[ab] OR neurosurg[ab] OR "Neurosurgery"[Mesh]) SEARCH TERMS ("Adenoma"[Mesh] OR "Pituitary Neoplasms"[Mesh] OR pituitar[ab]) AND ("Fluorescent Dyes"[Mesh] OR "Fluorescent Dyes" [Pharmacological Acon] OR "Infrared Rays"[Mesh] OR infrared[ab] OR ﬂuorescen[ab]) AND ("Surgery, Computer-Assisted"[Mesh] OR "surgery" [Subheading] OR surg[ab] OR neurosurg[ab] OR "Neurosurgery"[Mesh]) NAL DATABASES SUCH AS COCHRANE AND EMBASE WERE ALSO USED. THE SEARCH TERMS WERE: AND FOR TECHNICAL ASPECTS: AND FOR TECHNICAL ASPECTS: ('ﬂuorescence'/exp OR ﬂuorescence) AND guided AND ('surgery'/exp OR surgery) AND ('pituitary'/exp OR pituitary) AND ('adenoma'/exp OR adenoma) AND tech* AND FOR TECHNICAL ASPECTS: ('ﬂuorescence'/exp OR ﬂuorescence) AND guided AND ('surgery'/exp OR surgery) AND ('pituitary'/exp OR pituitary) AND ('adenoma'/exp OR adenoma) AND tech* wavelength difference before and after FNa injection in all tumors [13]. resected totally, three sub-totally and six partially. Func- tioning adenomas were found to have the highest inten- sity of fluorescence (p = 0.003) [12]. In 2019, Romano-Feinholz et al. performed a pilot study to assess safety and feasibility of a hybrid tech- nique using both endoscopy and microscopy in pitui- tary adenomas with intraoperative FNa, including 15 patients. Seven adenomas were non-functioning, six GH-secreting, one prolactin-secreting, and one ACTH- secreting. The authors compared pre- and postopera- tive MRI and categorized the extent of resection into total (100%), subtotal (80%-99%), and partial (< 80%). Intraoperative fluorescence was quantified with Adobe Photoshop using intraoperative images. Six tumors were Key points: Sodium Fluorescein (FNa) Key points: Sodium Fluorescein (FNa) - Peak emission at 521 nm Key points: Sodium Fluorescein (FNa) 1 3 - The light emitted from this organic compound is visible to the naked eye - Peak emission at 521 nm - Used in ophthalmology and neurosurgery - Functioning adenoma tumor tissue has the highest intensity of fluorescence - Literature on FNa fluorescence in pituitary adenoma surgery is sparse and quality is low - The light emitted from this organic compound is visible to the naked eye - Used in ophthalmology and neurosurgery - Functioning adenoma tumor tissue has the highest intensity of fluorescence - Literature on FNa fluorescence in pituitary adenoma surgery is sparse and quality is low 1 3 650 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 Titles, abstracts and full reports were analysed relevance to pituitary adenomas, neurosurgical specialty, study type, transsphenoidal surgical approaches and fluorescence. Records identified from: PubMed n = 282 Embase n = 25 Google Scholar n = 17 Cochrane n=0 Records removed before screening: Duplicate records removed n = 16 Titles and abstracts screened n = 308 Records excluded for irrelevance n = 232 Reports assessed for eligibility n = 76 Reports excluded: Not meeting inclusion criteria n = 61 Studies included in review n = 15 Identification of studies via databases and registers n o it a c ifit n e d I Screening Included Eligibility Fig. 1 Study collection Identification of studies via databases and registers Identification of studies via databases and registers Records excluded for irrelevance* n = 232 Reports excluded: Not meeting inclusion criteria n = 61 Studies included in review n = 15 *Titles, abstracts and full reports were analysed relevance to pituitary adenomas, neurosurgical specialty, study type, transsphenoidal surgical approaches and fluorescence. 3.2 5‑ALA Eljamel et al. studied 30 patients with pituitary adeno- mas; non-functioning macroadenomas (n = 14), gonado- trophin macroadenomas (n = 5), GH-secreting macroad- enoma (n = 1), prolactinomas (n = 2), ACTH secreting microadenomas (n = 3), GH-secreting microadenoma (n = 1), and pituitary cysts (n = 4). Patients received 20 mg/kg body weight of 5-ALA, 3 h before surgery. Intra- operative they used a Photo-Diagnostic filter (PD) and a laser-based Optical Biopsy System (OBS). The OBS con- sisted of a Gallium nitride-laser unit emitting blue light at 440 nm and a compact spectrometer. Once the adenoma was localized, it was exposed and illuminated by violet- blue light (375–440 nm) using an endoscopic PD-system (Olympus® QTV55 and CLV-520, Germany) attached to an endoscope, and any fluorescent adenoma was removed until no further fluorescence was detectable. The sensitiv- ity and specificity of the PD was found to be 80,8% and 5-ALA is a natural precursor for heme-synthesis. Cells metabolize ALA to heme in mitochondria. This pathway produces protoporphyrin-IX (PpIX). The conversion of PpIX to heme is rate-limiting in tumor cells, leading to accumulation of PpIX. PpIX is a photoactive molecule that absorbs violet-blue light and transmits in the red spectrum (600-650 nm) with peak emission at 635 nm [14, 15]. There are limitations to using 5-ALA. The auto- fluorescent characteristics of endogenous fluorophores such as lipofuscin and flavin lowers the signal-to-back- ground ratio (SBR). Another disadvantage of 5-ALA is the low tissue penetration (micrometers) [16]. Also, a relative discomforting situation is that after administra- tion patients have to be protected from sunlight and other ultraviolet radiation sources for 24 h. 1 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 651 term coined by Cho et al. ICG is administered as an infusion 24 h pre-operatively [16]. 75% respectively. The OBS sensitivity and specificity was found to be 95,5% and 100% respectively. PD showed a false negative rate of 19.2% and OBS 4.5%. The false posi- tive rate for OBS was 0% versus 25% for PD. The OBS system was able to detect 100% of microadenomas, even when the MRI was negative [14]. Litvack et al. performed an observational study in ten pituitary adenomas (five null-cell adenomas, one ACTH secreting adenoma, two GH secreting adenomas, one prolac- tin adenoma, and one mammosomatotroph adenoma). 3.2 5‑ALA Key points 5-AminoLevulinic Acid (5-ALA) - 5-ALA is metabolized to PpIX and accumulates in tumor cells - PpIX absorbs violet-blue light and transmits in the red spectrum of 600 nm to 650 nm with peak emission at 635 nm - 5-ALA has been shown to improve glioma resection rates and progression-free survival rates - Literature is sparse concerning 5-ALA usage during PitNET surgery. No strong added benefit for pituitary adenoma surgery has been proven yet Key points 5-AminoLevulinic Acid (5-ALA) - PpIX absorbs violet-blue light and transmits in the red spectrum of 600 nm to 650 nm with peak emission at 635 nm Verstegen et al. performed a clinical trial, including 10 patients with pituitary adenomas (four non-functioning, four ACTH-secreting, one GH-secreting, one prolactin-secreting tumor). Endoscopic transsphenoidal surgery was performed. Peroperatively 5 mg ICG was administered. All 10 patients displayed a minimum Fluorescence Contrast Ratio (FCR) of 1.3 (mean (SD) 1.5 ± (0.2)). FCR is calculated by dividing two regions of interest on a still image showing fluorescence of the adenoma and surrounding tissue. The authors noted that optimal FCR is generally considered to be > 2. They also noted that the timing of ICG is crucial for achieving optimal fluorescence [25]. - 5-ALA has been shown to improve glioma resection rates and progression-free survival rates - Literature is sparse concerning 5-ALA usage during PitNET surgery. No strong added benefit for pituitary adenoma surgery has been proven yet - Literature is sparse concerning 5-ALA usage during PitNET surgery. No strong added benefit for pituitary adenoma surgery has been proven yet Key points 5-AminoLevulinic Acid (5-ALA) - 5-ALA is metabolized to PpIX and accumulates in tumor cells 3.2 5‑ALA Fluo- rescence was analyzed with a photo diagnostic endoscope 45–90 s after ICG injection, before and after opening of the dura. A strong fluorescent nodule in the dura was noted, most likely due to hypervascularity by dural invasion of the adenoma. These fluorescent dural nodules were found in all patients with acromegaly, prolactinoma, and mammosoma- totroph adenoma, but was not seen in null-cell adenomas (n = 5) that displayed lower fluorescence in tumor tissue ver- sus higher fluorescence in pituitary gland tissue [24]. Marbacher et al. retrospectively analyzed 458 cases with a variety of intracranial tumors. Positivity for 5-ALA was evaluated intermittently using the 440-nm blue-violet light source of a microscope (OPMI Pentero, Carl Zeiss). Twelve pituitary adenomas were assessed, but only 8% (1/12) dis- played a positive fluorescence with 5-ALA. The fluorescence was evaluated by the surgeon. No further information on type of pituitary adenoma was available [17]. Micko et al. performed a multicenter retrospective study to investigate 5-ALA in endoscopic endonasal approaches. Twenty-eight patients with skull base pathologies were included and evaluated on their fluorescence status, qual- ity, and homogeneity. The evaluation was done by both the surgical team and an external reviewer. Teams were blinded for each other’s results. Of the 28 pathologies, 15 were pitui- tary adenomas (seven functioning and eight non-functioning adenomas). No fluorescence was identified in all pituitary adenomas. Vague florescence was identified from the normal pituitary gland in a case of a gonadotroph macroadenoma, displayed vague fluorescence [18]. Sandow et al. performed an observational study in 22 patients. Thirteen patients were diagnosed with acromegaly, six with Cushing disease, and three with vision disorders or dizziness. In 11 patients, direct visualization of fluores- cence in the lesion was noted, meaning that the adenoma displayed higher fluorescence than healthy tissue. The other 11 showed indirect visualization, in which the tissue sur- rounding the adenoma displayed higher fluorescence. All patients with Cushing disease displayed a higher ICG uptake than healthy tissue. It is interesting that all ACTH producing tumors (n = 6) were visualized directly, in comparison to the findings of Litvack et al. in which two hormone-producing adenomas displayed lower fluorescence than other adeno- mas. Apparently, timing of ICG injection affects fluores- cence intensity, with a measured optimal fluorescence at 2.4 min after injection [23]. 3.4 OLT38 OTL38 (On Target Laboratories, West Lafayette, Indiana) is a bioconjugate of a folate analog and a cyanine dye target- ing folate receptor alpha (FRα) [27], which may be overex- pressed in nonfunctioning pituitary adenomas [2]. A draw- back of OTL38 is that FRα overexpression is not known before surgery, thus potentially resulting in significant false- negative results [27]. OTL38 has an excitation wavelength of 785 nm, and an emission wavelength spectrum of 800- 835 nm, classifying OTL38 as NIR-fluorophore. The mechanism of SWIG is based on the Enhanced Perme- ability Retention effect (EPR). EPR describes the enhanced permeability of tumor vasculature through which macromol- ecules can enter the tumor. It is hypothesized that ICG enters the cell through the EPR effect and is being retained in the cell. After a certain time period, healthy tissue has cleared ICG but tumor cells have become fluorescent by accumu- lation of ICG [15]. It is to be noted that ICG is not tumor specific, since disturbed permeability is seen in necrosis and inflammation as well. In a study performed by Lee et al. 15 patients with func- tioning or nonfunctioning adenomas were given OTL38. Six patients had non-functioning adenomas (four null-cell, one clinically silent gonadotroph, one fully silent somato- troph) and nine functioning adenomas (five corticotrope, three somatotroph, one somatocorticotrope). Three non- functioning PitNETs (50%) were found to overexpress FRα, while the functioning adenomas were found to have little to no FRα overexpression. Tumors with high FRα expression (n = 3) had an SBR 3.0 ± 0.29, while low FRα expression (n = 12) resulted in an SBR 1.6 ± 0.43 [28]. Jeon et al. studied the use of SWIG on 8 patients (5 non- functioning adenomas, 1 somatotroph, and 3 corticotrope). ICG was administered 16–30 h before surgery. Pituitary adenomas demonstrated a signal-to-background (SBR) of 3.9 ± 0.8, which was 4 times higher than the fluorescence of surrounding tissue. While all 15 samples were positive for NIR signal, only 10 specimens were histopathologi- cally proven adenoma tissue. The detection of false posi- tive NIR signal, resulted in a decrease in specificity (20%). The sensitivity, PPV, and NPV of NIR-fluorescence were 100%, 71,43%, and 100%, respectively. The neurosurgeon was asked to evaluate intraoperatively whether the tissue was tumor or healthy tissue and was compared to histologi- cal findings. Key points Second Window ICG (SWIG) Key points Second Window ICG (SWIG) - ICG is a near-infrared dye - SWIG mechanism is based on Enhanced Permeability Retention (EPR) - SWIG mechanism is based on Enhanced Permeability Retention (EPR) - It has an excitation wavelength of 805 nm and emission in the 820- 860 nm range - PPV 71,43%—82%, NPV 100% - Intra-operative dosages of ICG are timing sensitive for optimal results - Hormonal status of the adenoma has no effect on intraoperative fluorescence - Administering multiple dosages is possible - Administering multiple dosages is possible 3.3.1 Intraoperative bolus Indocyanine green (ICG) is a near-infrared (NIR) fluorescent dye that has been approved by the Food and Drug Adminis- tration in 1959. Since then, it has been widely used in oph- thalmology, reconstructive surgery, and abdominal surgery [19, 20]. In neurosurgery, ICG is utilized for intraoperative visualization of real-time cerebral perfusion. It is used for monitoring cerebral perfusion during surgery for neurovas- cular disorders [21]. Recently, the ICG-indication range of neurological pathology is broadened to gliomas, meningi- omas, and pituitary adenomas [22, 23]. ICG is used as an IV bolus injection intraoperatively. In Second Window ICG, a Amano et al. studied 33 ICG bolus injections in 20 patients with a dosage of 6.25 or 12.5 mg/injection. Fifteen patients were diagnosed with a pituitary adenoma (nine GH- producing and six non-functioning adenomas). Two minutes post-injection, the posterior lobe was differentiated from the anterior lobe. After three minutes, the anterior lobe was fluo- rescent. The adenomas in the non-functioning group (n = 6) showed highest fluorescence after 1–7 min and gradually decreased, while the healthy tissue started to fluoresce 10–20 min post-injection. GH-producing adenomas showed 1 3 1 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 652 opening of the dura and direct exposure of the tumor, the SBR increased to 4.1 ± 0.72. In addition, the pituitary adenomas were grouped by preoperative hormonal status. The hormonally active adenomas expressed a mean SBR of 4.1 ± 0.63, while in the non- functioning group an SBR of 4.0 ± 0.82 was detected. SWIG was 100% sensitive, but only 29% specific for neoplastic tissue, with a PPV of 82% and an NPV of 100% [16]. fluorescence six minutes after injection. After debulking, the normal gland gradually became more fluorescent eight min- utes post-injection. No growth hormone deficiencies were found post-operative. This is the first study that explored the extent of fluorescence versus time. These findings might explain the variable results in earlier studies using ICG [21]. Key points ICG - ICG is a near-infrared dye - It has an excitation wavelength of 805 nm and emission in the 820- 860 nm range - Intra-operative dosages of ICG are timing sensitive for optimal results - Administering multiple dosages is possible Key points ICG 3.4 OLT38 The sensitivity, specificity, PPV, and NPV of the neurosurgeon were 90%, 100%, 100%, and 83%, respec- tively. Jeon and colleagues suggested that a trained neuro- surgeon in combination with NIR-fluorescence could lead to better resection results [26]. Remarkably, adenomas without overexpressed FRα should not have demonstrated fluorescence and yet did exhibit fluorescence. Thus, adenomas with little or no FRα expression showed an increased SBR as time passed because normal tissue cleared OTL38 faster than the tumor. The authors believe that these fluorescent adenomas showed an accumulation of OTL38, likely due to the tumors vasculature being leakier than healthy tissue, leading to enhanced per- meability and retention effect in the tumor tissue, allowing passive accumulation. Cho et al. set up a study with two groups simultaneously. In the first, 16 patients received ICG 24 h pre-operatively (SWIG), while in the latter 23 patients received a folate analog, OTL-38. The SWIG-group consisted of seven non-functioning adenomas, four GH-secreting, two ACTH-secreting, one TSH-secreting, and 2 PRL-secreting adenomas. All 16 patients demonstrated NIR-fluorescence through the dura (SBR 2.7 ± 0.63). Upon The OTL38 patient subgroup of the mentioned study by Cho et al. [16] consisted of 23 patients (14 non-functioning adenomas, three GH-secreting, and six ACTH-secreting ade- nomas) and received OTL38 2–4 h preoperatively. All pitui- tary adenomas fluoresced with a mean SBR 1.6 ± 0.43 through 3 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 653 the dura, and the mean NIR-SBR increased to 2.2 ± 0.88 upon tumor exposure. The 14 non-functioning adenomas showed a mean SBR 2.6 ± 0.91, the nine functioning adenomas a mean SBR 1.7 ± 0.47. When divided based on postoperatively deter- mined FRα expression, the nine FRα-overexpressing adeno- mas demonstrated a mean SBR 3.2 ± 0.53. The 14 adenomas with low FRα expression had a mean SBR 1.6 ± 0.42. OTL38 had higher sensitivity and specificity when limited to the FRα-overexpressing adenomas, compared to assessment of all adenomas. Thus, OTL38 is highly sensitive and specific for pituitary adenomas that overexpress FRα but has limited utility in functioning adenomas. are the most difficult to achieve total resection, since these tumors invade into the cavernous sinus with or without entanglement of the carotid artery. Especially for these com- plex tumors a supplemental diagnostic tool could be help- ful in the intraoperative detection of residual tumor tissue. 4.1 Sodium fluorescein and 5‑ALA Sodium Fluorescein (FNa) and 5–ALA both belong to the group of ‘visible light’ fluorophores. 5-ALA has proven its feasibility in glioma surgery and FNa has an application in the detection of CSF leakage [31]. The application of FNa to aid in the resection of PitNETs is lacking evidence. The sparse results in the literature imply that FNa could be of use in the detection of functioning PitNETs but due to the limited evidence, the added value of FNa in PitNET surgery remains unclear. Eljamel et al. [14] concluded that 5-ALA might be of supportive value in the resection of the PitNETs. Specifically, in this study specialized hardware, such as a photo diagnostic filter and an Optical Biopsy System, was used to detect the 5-ALA fluorescence during surgery. In glioma surgery the violet/blue filter is integrated in the microscope, but a specific laser controlled Optical biopsy system is not present. Inter- esting to notice is that the OBS system was able to detect all of the microadenomas that were not seen on the MRI (6/6). Since OTL38 is folate receptor-alpha targeted (which is mainly expressed in non-functioning adenomas), the sensi- tivity and specificity was calculated only in non-functioning adenomas with (n = 28) and without (n = 18) overexpression of FRα. In the first, the sensitivity and specificity are 75% (51–90%) and 100% (60–100%) respectively, with PPV 100% (75–100%) and NPV 62% (43–77%). In the FRα overexpress- ing adenoma group (n = 18), the sensitivity and specificity are 100% (75–100%) and 100% (31–100%) respectively, with PPV 100% (75–100%) and NPV 100% (31–100%). Another study by Cho et al. included 14 patients with non-functioning adenomas. Prior to opening of the dura, the group had a mean SBR 1.8 ± 0.40 that increased to 2.6 ± 0.92 after opening of the dura. The group was then divided into an overexpressing FRα and a non-FRα-overexpressing group. Nine adenomas overexpressed FRα with NIR-SBR 3.2 ± 0.52, whereas 5 non-FRα-overexpressing adenomas fluoresced with NIR-SBR 1.5 ± 0.21. Margin sample analysis demonstrated that the surgeon’s impression of the tissue had 83% sensitivity, 100% specificity, 100% PPV, and 89% NPV, while NIR-fluorescence demonstrated 100% for all values [3]. The evidence of 5-ALA as an additive diagnostic instru- ment for detection of residual pituitary tumor is not fully explored. Data is sparse and the described methods deter- mining the fluorescence are limited and therefore difficult to reproduce. 3.4 OLT38 Therefore, we reviewed the literature to identify and value the various fluorescent agents specifically available in the context of endoscopic transsphenoidal surgery. 4.2 Indocyanine green (ICG) - OTL38 has an excitation wavelength of 785 nm, and an emission wavelength spectrum of 800-835 nm, classifying OTL38 as a NIR- fluorophore ICG demonstrated interesting results. A few studies demon- strated that non-functioning adenomas displayed a lower sig- nal than the normal pituitary gland. Thus, indirect imaging of the tumor could be used to differentiate between normal tissue and tumor tissue [23, 24]. Furthermore, two small studies [23, 24] suggested that hormone producing tumors were visualized directly. However, this conclusion was not absolute as two hormone-producing adenomas displayed lower fluorescence [24]. Also in ICG, the described meth- ods determining the fluorescence are limited and therefore difficult to reproduce. An explanation could be that timing of ICG injection and local tumor permeability are of key importance in relation to the optimal time frame of visuali- zation and whether visualization is indirect or direct. - Cho et al. demonstrated proof of concept and usability in non-functioning PitNETs pituitary adenoma 4.1 Sodium fluorescein and 5‑ALA For certainly, more research will be needed to support 5-ALA as a suitable agent for detection of pituitary tumor tissue. - OTL38 is a bioconjugate of a folate analog and a cyanine dye that targets folate receptor alpha (FRα), which is mainly overexpressed in nonfunctioning pituitary adenomas 4 Discussion As gross total or total resection in PitNET surgery is around 70–80% [29, 30], there is a window of opportunity to raise the overall resection percentage. Particularly in functioning PitNETs, total resection is mandatory to cure the patient. Surgical aspects are of paramount importance in the gen- eral treatment of patients with PitNET. High-grade PitNETs 1 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 654 A technique that eliminates necessity for timing (time sensitiveness) is second window ICG (SWIG). Neverthe- less, studies demonstrated that specificity of SWIG is low due to disturbed permeability. Since lower specificity leads to increased false-positive outcomes, this might be harmful as during surgery normal tissue could be falsely interpreted as tumor tissue. The SWIG study by Zhang et al. [27] dem- onstrated SBR to be 4 times higher in tumor than in sur- rounding tissue, but in histopathological examination not all fluorescent tissue could be classified as tumor tissue (false positive). In the study by Cho et al. [3], all PitNETs dem- onstrated NIR fluorescence, lacking significant difference in SWIG fluorescence signal between functioning and non- functioning PitNETs. As such, studies [16, 21, 23–26] seem to show contradictory results on ICG and SWIG application. However, SWIG has better results than intraoperative ICG bolus injection, considering the high NPV [16]. study by Cho et al. [3], the dedicated NIR-system showed superior SBR compared to the surgical microscope with NIR module. At the moment, dedicated NIR endoscopes do not have enough distinctiveness with respect to 2D HD 4 K endoscopes. Therefore, both NIR and 2D HD systems must be compared during surgery in order to define tumor tissue from non-tumorous tissue.l When using fluorescence during surgery, ideally only tumor tissue emits fluorescent signals, while other tissues such as healthy pituitary tissue should not. Thus, the detec- tion of true positives is important. However, there are fac- tors that can inhibit the detection of true positives. First, if there is a significant amount of fluorescence in the endonasal mucosa or dura surrounding the surgical field, the NIR cam- era may not properly detect smaller areas of fluorescence that could be tumor tissue. This can be combated by cover- ing the surrounding area with surgical towels or gauze to isolate the surgical area, allowing smaller areas of residual fluorescence to be detected more easily [33]. However, this can be challenging in PitNET surgery. 4.3 OTL38 In studies on the use of OTL38, this fluorescent agent dem- onstrates its additive diagnostic use only in the population of non-functioning PitNETs. Histopathological analysis detected that mainly non-functioning PitNETs express the folate receptor alpha (FRα). Functioning PitNETs have none or less expression of FRα. Thus, OTL38 can only be applied in the removal of non-functioning PitNETS. Also, the factor time sensitivity played a role in the evaluation of the peri- operative results. PitNETs that did not overexpress FRα on histopathological analysis did exhibit fluorescence during surgery, because normal tissue cleared OTL38 faster than the tumor [28]. 4 Discussion Second, care must be taken not to miss fluorescence hidden by overhanging tissue, necrosis, blood, or hemostatic adjuncts such as Surgicel® [33, 34]. The fluorophore must be directly excited by the laser. Thus, fluorophore concentrations hidden from direct light by overhanging structures, such as tumors that invade the cavernous sinus, can be easily missed. In one patient in a study by Verstegen et al. [25] intracavernous sinus bleed- ing prohibited assessment of the fluorescent signal of the adenoma tissue. Pooled blood can also obscure areas of NIR fluorescence by physically blocking excitation and emission light. Therefore, it is critical that hemostasis is achieved so that the tissue of interest can be placed in the direct line of the excitation source [33]. 3 4.5 Avoidance of false positivity False positive signal is where healthy tissue produces a positive fluorescence signal. In order to detect tumor tissue alone, false positive signals need to be minimized. First, an important factor in the creation of false positives is the dis- tance between the sensor and the tissue. Bringing the scope too proximal to the area of interest may result in artificially increased NIR fluorescence by flooding the area with excita- tion photons [32, 33]. Light decays exponentially as a factor of distance squared. Hence, when the scope is twice as close to the tissue, this results in a fluorescence signal that is 16 times greater, because the excitation light is 4 times stronger and 4 times as much emission is reaching the sensor. Mul- tiple articles describe the use of the medial optico-carotid recess (MOCR) [3, 28, 32] as a useful landmark, maintain- ing the distance between the bilateral MOCR to be less than 50% of the overall field of view [32]. This offers a better dis- crimination between false-positive and true positive areas. In Understanding technical aspects of fluorescent imaging is crucial, especially within the sphenoidal cavity as its bony walls reflect the endoscopic light and blood remnants may disturb visibility, it is paramount to minimize all confound- ers that influence the detection quality of the fluorescent signal. NIR fluorophores can emit their light through a thicker layer of tissue in comparison to visible light fluorophores, which is mainly absorbed by hemoglobin/blood. A second advantage is that NIR fluorophores are less influenced by the autofluorescence of nearby situated normal structures [3, 24, 28, 32]. On the other hand, the technique to visu- alize NIR fluorophores requires additional detection filters and hardware in comparison to visible light fluorophores. This can either be a microscope or endoscope with add- on NIR modules or a dedicated NIR imaging system. In a 3 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 655 OT38 article [32], adenomas that did not overexpress FRα only fluoresced when the scope was brought into close prox- imity to the tumor, whereas adenomas that overexpressed FRα fluoresced strongly regardless of the scope distance. could be especially problematic if the status of the samples originating from the same patient were highly correlated (e.g., sample A and sample B are more likely to be true positives if they were contributed by the same patient than by two different patients). 4.7 Future perspectives Currently, the use of fluorescent agents as a diagnostic dur- ing pituitary surgery is not standard care and most studies are proof of principle studies of non-targeted fluorophores. These fluorophores have their principal action through vas- cular permeability of the tumor, are non-targeted, and have more or less a narrow time window of imaging opportunity. In that perspective, there are many disadvantages. Neverthe- less, the field of molecular biology is continuously develop- ing and there is a high intensity quest for patient specific or tumor associated targets. Software, hardware, and high throughput analytical methods are available to unleash tumor associated markers in a short period of time. 4.6 The benefit of fluorescence‑guided surgery in the resection of PitNETs In general, fluorescence guided resection of PitNETs is not common practice. Most publications are proof of principle studies with small numbers of included patients and out- come is not related to clinical outcome, but have technical aspects of measuring and analyzing the fluorescent signals as the primary outcome parameters. Although test accuracy data (sensitivity, specificity NPV and PPV) presented in some studies show positive results we argue that interpreta- tion of these figures must be done with great caution. It is questionable if these figures represent actual accuracy. We tried to verify, from the presented data, if these accuracy calculations were correct. In order to do so we contacted the respective authors by email to provide additional informa- tion, Unfortunately we could get no contact to date 1st of October 2021. Monoclonal antibodies such as bevacizumab (anti-VEGF) [35–37], girentuximab (anti carbonic anhydrase IX (CAIX) [38], cetuximab (anti EGFR) [39], and panitumumab (anti EGFR) [40, 41] are clinically approved oncolytic drugs that target specific cell membranous/cytoplasmatic bound anti- bodies and have been evaluated in clinical trials. Because the pharmacokinetic and pharmacodynamic properties of the monoclonal antibody and the NIR dye are already known from previous performed studies [42, 43], the combination of both can be constructed and evaluated [44]. 4.5 Avoidance of false positivity Hence, the reported test character- istics (sensitivity, specificity, PPV, NPV) from studies that assume samples to be independent may not be representa- tive of the patient-level test characteristics. Furthermore, in order to be able to generalize the results, it is important to consider the uncertainty of the estimates. Studies typically did only report the point estimate of test characteristics but no confidence intervals, which cannot be calculated without patient-level information. With the few data presented in several small cohorts it is impossible to draw any conclu- sion whether fluorescent supported surgery of PitNETs can lead to a higher track record of total resection and will be beneficial for patients with these tumors. Second, many NIR systems have variable gains that reflect the sensitivity of the camera sensor to emission pho- tons. To be able to detect weaker signals a more sensitive camera is needed, and thus a higher gain can be applied. However, both signal and noise are enhanced, increasing the percentage of false positive signals. Many papers have described methods to minimize false positives caused by high gain settings. Cho et al. [33] describe a technique where the gain setting was preoperatively adjusted when the tumor was first exposed and maintained at that level throughout the rest of the surgery. Another technique is by Jeon et al. [26], where a positive control is used to set the gain. They dipped the control paper in an ICG solution and then waited till dry. An image was made from 12 mm away with an endoscope using both white light and NIR. The NIR detector was set to 50% gain, which was previously observed to provide a strong NIR signal without significant background noise. They noted that when using the automatic sensor gain adjustment feature inherent in the VisionSense camera software, true positive samples had a lower mean gain percent (76%) compared to that of false positive samples (97.5%). This suggests that the gain setting and thus the NIR sensor’s sensitivity was automatically increased to the point of creating false-positive fluorescence signals [26]. 5 Conclusion Fluorescent molecular endoscopy (FME) distinguishes from MRI, PET, and other radiological diagnostics in that it is real-time guidance in discriminating between healthy and diseased tissues. Properties of the imaging hardware and software can influence the images and can either limit or improve the ability to differentiate the targeted tissue from its surroundings. Interpretation of the endoscopic images The pitfall of the reviewed studies which presented accu- racy data was that they included multiple samples per patient and provided no information as to which patient contributed to which sample, precluding the possibility for a patient- level analysis and an accurate estimation of confidence intervals. Disregarding the clustered nature of the data it 1 656 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 and Specific Intraoperative Visualization of Nonfunctional Pitui- tary Adenomas. Oper Neurosurg (Hagerstown). 2019;16:59–70. requires comprehensive understanding of this technique, and the surgical team should carefully consider the limitations during the process of image-guided pituitary surgery. Atten- tion must also be given to the accuracy and standardization of imaging systems and probes, in terms of clinical grade GMP production, quality control, and standard operating procedures related to stability [45]. y p g g 4. Shin SS, Tormenti MJ, Paluzzi A, Rothfus WE, Chang Y-F, Zainah H, et al. Endoscopic endonasal approach for growth hor- mone secreting pituitary adenomas: outcomes in 53 patients using 2010 consensus criteria for remission. Pituitary. 2013;16:435–44.l 5. Schöfl C, Franz H, Grussendorf M, Honegger J, Jaursch-Hancke C, Mayr B, et al. Long-term outcome in patients with acromegaly: analysis of 1344 patients from the German Acromegaly Register. Eur J Endocrinol. 2013;168:39–47. Furthermore, special software and hardware must be facilitated in order to use FME during pituitary surgery. Therefore, it is to be expected that FME for pituitary surgery will initially be applied in high volume centers. For certain, further innovation on the field of FME in pituitary surgery will progress, but at this moment conventional endoscopic surgery still is the gold standard for maximum safe resection of pituitary tumors. 6. Jane JA Jr, Starke RM, Elzoghby MA, Reames DL, Payne SC, Thorner MO, et al. Endoscopic transsphenoidal surgery for acro- megaly: remission using modern criteria, complications, and pre- dictors of outcome. J Clin Endocrinol Metab. 2011;96:2732–40. 7. van Manen L, Handgraaf HJM, Diana M, Dijkstra J, Ishizawa T, Vahrmeijer AL, et al. Supplementary information The online version contains supplemen- tary material available at https://doi.org/10.1007/s11154-022-09718-9. 9. Landau MJ, Gould DJ, Patel KM. Advances in fluorescent-image guided surgery. Annals of Translational Medicine. 2016;4:3–3. Author contributions Design of the study was performed by Vergeer RA and Kuijlen JMA. literature search and data analysis was performed by Vergeer RA, Theunissen REP, van ELK T and Kuijlen JMA. The first draft of the manuscript was written by Vergeer RA, Theunissen REP and van ELK T. The review was critically revised by Schmidt I, Postma MR, Tamasi K, Dijk JMC and Kuijlen JMA. All authors read and approved the final manuscript. 10. DSouza AV, Lin H, Henderson ER, Samkoe KS, Pogue BW. Review of fluorescence guided surgery systems: identification of key performance capabilities beyond indocyanine green imaging. J Biomed Opt 2016;21:80901. 11. Shinoda J, Yano H, Yoshimura S-I, Okumura A, Kaku Y, Iwama T, et al. Fluorescence-guided resection of glioblastoma multi- forme by using high-dose fluorescein sodium: Technical note. J Neurosurg. 2003;99:597–603. 12. Romano-Feinholz S, Alcocer-Barradas V, Benítez-Gasca A, la Maza EM, Valencia-Ramos C, Gómez-Amador JL. Hybrid fluorescein- guided surgery for pituitary adenoma resection: a pilot study. J Neu- rosurg. 2019;132:1490–8.l Funding The authors did not receive support from any organization for the submitted work. Declarations 13. da Silva CE, da Silva JLB, da Silva VD. Use of sodium fluorescein in skull base tumors. Surg Neurol Int. 2010;1:70.i Conflict of interest The authors declare that they have no conflicts of interest. 14. Eljamel MS, Leese G, Moseley H. Intraoperative optical identifi- cation of pituitary adenomas. J Neurooncol. 2009;92:417–21. 15. Nemes A, Fortmann T, Poeschke S, Greve B, Prevedello D, Santacroce A, et al. 5-ALA Fluorescence in Native Pituitary Adenoma Cell Lines: Resection Control and Basis for Photodynamic Therapy (PDT)? PLoS One 2016;11:e0161364. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 16. Cho SS, Jeon J, Buch L, Nag S, Nasrallah M, Low PS, et al. Intraoperative near-infrared imaging with receptor-specific ver- sus passive delivery of fluorescent agents in pituitary adenomas. J Neurosurg. 2018;131:1974–84. 17. Marbacher S, Klinger E, Schwzer L, Fischer I, Nevzati E, Diepers M, et al. Use of Fluorescence to Guide Resection or Biopsy of Pri- mary Brain Tumors and Brain Metastases. Journal of Neurological Surgery Part A: Central European Neurosurgery 2014;75. https:// doi.org/10.1055/s-0034-1383791. 18. Micko A, Rapoport BI, Youngerman BE, Fong RP, Kosty J, Brunswick A, et al. Limited utility of 5-ALA optical fluorescence in endoscopic endonasal skull base surgery: a multicenter retrospective study. J Neu- rosurg 2020;-1:1–7. 5 Conclusion A practical guide for the use of indocyanine green and methylene blue in fluorescence-guided abdominal sur- gery. J Surg Oncol. 2018;118:283. g y g 8. Moore GE, Peyton WT, French LA, Walker WW. The Clinical Use of Fluorescein in Neurosurgery: The localization of brain tumors. J Neurosurg. 1948;5:392–8.l References 19. Kimura T, Muguruma N, Ito S, Okamura S, Imoto Y, Miyamoto H, et al. Infrared fluorescence endoscopy for the diagnosis of superficial gastric tumors. Gastrointest Endosc. 2007;66:37–43. https://doi.org/10.1016/j.gie.2007.01.009.f 1. Chang SW, Donoho DA, Zada G. Use of optical fluorescence agents during surgery for pituitary adenomas: current state of the field. J Neurooncol. 2019;141:585–93. 20. Holm C, Tegeler J, Mayr M, Becker A, Pfeiffer UJ, Mühlbauer W. Monitoring free flaps using laser-induced fluorescence of indocya- nine green: a preliminary experience. Microsurgery. 2002;22:278–87. 2. Lakomkin N, Van Gompel JJ, Post KD, Cho SS, Lee JYK, Hadjipanayis CG. Fluorescence guided surgery for pituitary adenomas. J Neuroon- col. 2021;151(3):403–13. 21. Amano K, Aihara Y, Tsuzuki S, Okada Y, Kawamata T. Appli- cation of indocyanine green fluorescence endoscopic system in 3. Cho SS, Zeh R, Pierce JT, Jeon J, Nasrallah M, Adappa ND, et al. Folate Receptor Near-Infrared Optical Imaging Provides Sensitive 3. Cho SS, Zeh R, Pierce JT, Jeon J, Nasrallah M, Adappa ND, et al. Folate Receptor Near-Infrared Optical Imaging Provides Sensitive 1 3 Reviews in Endocrine and Metabolic Disorders (2022) 23:647–657 657 transsphenoidal surgery for pituitary tumors. Acta Neurochir. 2019;161:695–706. 34. Stummer W, Molina ES. Fluorescence Imaging/Agents in Tumor Resection. Neurosurg Clin N Am. 2017;28:569–83. https://doi. org/10.1016/j.nec.2017.05.009. 22. Kim EH, Cho JM, Chang JH, Kim SH, Lee KS. Application of intraoperative indocyanine green videoangiography to brain tumor surgery. Acta Neurochir 2011;153:1487–95; discussion 1494–5. 35. Harlaar NJ, Koller M, de Jongh SJ, van Leeuwen BL, Hemmer PH, Kruijff S, et al. Molecular fluorescence-guided surgery of peritoneal carcinomatosis of colorectal origin: a single-centre feasibility study. Lancet Gastroenterol Hepatol. 2016;1:283–90. 23. Sandow N, Klene W, Elbelt U, Strasburger CJ, Vajkoczy P. Intra- operative indocyanine green videoangiography for identifica- tion of pituitary adenomas using a microscopic transsphenoidal approach. Pituitary. 2015;18:613–20.l 36. Lamberts LE, Koch M, de Jong JS, Adams ALL, Glatz J, Kranendonk MEG, et al. Tumor-Specific Uptake of Fluorescent Bevacizumab- IRDye800CW Microdosing in Patients with Primary Breast Cancer: A Phase I Feasibility Study. Clin Cancer Res. 2017;23:2730–41. 24. Litvack ZN, Zada G, Laws ER. Indocyanine green fluorescence endoscopy for visual differentiation of pituitary tumor from sur- rounding structures. J Neurosurg. 2012;116:935–41. https://doi. org/10.3171/2012.1.jns11601. 37. Koch M, de Jong JS, Glatz J, Symvoulidis P, Lamberts LE, Adams ALL, et al. Threshold Analysis and Biodistribution of Fluores- cently Labeled Bevacizumab in Human Breast Cancer. Cancer Res. 2017;77:623–31. g j 25. References Verstegen MJT, Tummers QR, Schutte PJ, Pereira AM, van Furth WR, van de Velde CJH, et al. Intraoperative Identification of a Normal Pituitary Gland and an Adenoma Using Near-Infrared Fluorescence Imaging and Low-Dose Indocyanine Green. Opera- tive Neurosurgery. 2016;12:260–8. https://doi.org/10.1227/neu. 0000000000001328. 38. Hekman MCH, Boerman OC, de Weijert M, Bos DL, Oosterwijk E, Langenhuijsen HF, et al. Targeted Dual-Modality Imaging in Renal Cell Carcinoma: An Ex Vivo Kidney Perfusion Study. Clin Cancer Res. 2016;22:4634–42. https://doi.org/10.1158/1078- 0432.ccr-15-2937. 26. Jeon JW, Cho SS, Nag S, Buch L, Pierce J, Su YS, et al. Near- Infrared Optical Contrast of Skull Base Tumors During Endo- scopic Endonasal Surgery. Oper Neurosurg (Hagerstown). 2019;17:32–42. 39. Rosenthal EL, Warram JM, de Boer E, Chung TK, Korb ML, Brandwein-Gensler M, et al. Safety and Tumor Specificity of Cetuximab-IRDye800 for Surgical Navigation in Head and Neck Cancer. Clin Cancer Res. 2015;21:3658–66. 27. Zhang DY, Singhal S, Lee JYK. Optical Principles of Fluorescence- Guided Brain Tumor Surgery: A Practical Primer for the Neurosur- geon. Neurosurgery. 2019;85:312–24. 40. Tummers WS, Warram JM, Tipirneni KE, Fengler J, Jacobs P, Shankar L, et al. Regulatory Aspects of Optical Methods and Exogenous Targets for Cancer Detection. Cancer Res. 2017;77:2197–206. 28. Lee JYK, Cho SS, Zeh R, Pierce JT, Martinez-Lage M, Adappa ND, et al. Folate receptor overexpression can be visualized in real time during pituitary adenoma endoscopic transsphenoidal surgery with near-infrared imaging. J Neurosurg. 2018;129:390–4030. 41. Rosenthal EL, Warram JM, de Boer E, Basilion JP, Biel MA, Bogyo M, et al. Successful Translation of Fluorescence Naviga- tion During Oncologic Surgery: A Consensus Report. J Nucl Med. 2016;57:144–50. 29. Almutairi RD, Muskens IS, Cote DJ, Dijkman MD, Kavouridis VK, Crocker E, Ghazawi K, Broekman MLD, Smith TR, Mekary RA, Zaidi HA. Gross total resection of pituitary adenomas after endo- scopic vs. microscopic transsphenoidal surgery: a meta-analysis. Acta Neurochir (Wien). 2018;160(5):1005–1021.f 42. Scheuer W, van Dam GM, Dobosz M, Schwaiger M, Ntziachristos V. Drug-based optical agents: infiltrating clinics at lower risk. Sci Transl Med 2012;4:134ps11. 43. Day KE, Sweeny L, Kulbersh B, Zinn KR, Rosenthal EL. Pre- clinical comparison of near-infrared-labeled cetuximab and pani- tumumab for optical imaging of head and neck squamous cell carcinoma. Mol Imaging Biol. 2013;15:722–9. 30. Soneru CP, Riley CA, Hoffman K, Tabaee A, Schwartz TH. Intra- operative MRI vs endoscopy in achieving gross total resection of pituitary adenomas: a systematic review. Acta Neurochir (Wien). 2019;161(8):1683–98. https://doi.org/10.1007/s00701-019- 03955-9. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Epub 28 May 2019 PMID: 31139934. 44. Linssen MD, Ter Weele EJ, Allersma DP, Lub-de Hooge MN, van Dam GM, Jorritsma-Smit A, et al. Roadmap for the Development and Clinical Translation of Optical Tracers Cetuximab-800CW and Trastuzumab-800CW. J Nucl Med. 2019;60:418–23. 31. Jakimovski D, Bonci G, Attia M, Shao H, Hofstetter C, Tsiouris AJ, et al. Incidence and significance of intraoperative cerebrospi- nal fluid leak in endoscopic pituitary surgery using intrathecal fluorescein. World Neurosurg. 2014;82:e513–23. 45. de Boer E, Harlaar NJ, Taruttis A, Nagengast WB, Rosenthal EL, Ntziachristos V, et al. Optical innovations in surgery. Br J Surg. 2015;102:e56-72. 32. Cho SS, Lee JYK. Intraoperative Fluorescent Visualization of Pituitary Adenomas. Neurosurg Clin N Am. 2019;30:401–12. 33. Cho SS, Salinas R, Lee JYK. Indocyanine-Green for Fluorescence- Guided Surgery of Brain Tumors: Evidence, Techniques, and Prac- tical Experience. Front Surg 2019;6. https://doi.org/10.3389/fsurg. 2019.00011. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 1 3 1 3 1 3 3
https://openalex.org/W2605263792	https://hal.science/hal-02084940/document	English	null	Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station	Sensors	2,017	cc-by	13,544	Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station Elise Lachat, Tania Landes, Pierre Grussenmeyer To cite this version: Elise Lachat, Tania Landes, Pierre Grussenmeyer. Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station. Sensors, 2017, 17 (4), pp.730. 10.3390/s17040730. hal-02084940 To cite this version: Elise Lachat, Tania Landes, Pierre Grussenmeyer. Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station. Sensors, 2017, 17 (4), pp.730. 10.3390/s17040730. hal-02084940 Elise Lachat, Tania Landes, Pierre Grussenmeyer. Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station. Sensors, 2017, 17 (4), pp.730. 10.3390/s17040730. hal-02084940 Elise Lachat , Tania Landes and Pierre Grussenmeyer Elise Lachat , Tania Landes and Pierre Grussenmeyer ICube Laboratory, Photogrammetry and Geomatics Group, National Institute of Applied Sciences (INSA), 24 Boulevard de la Victoire, 67084 Strasbourg CEDEX, France; tania.landes@insa-strasbourg.fr (T.L.); pierre.grussenmeyer@insa-strasbourg.fr (P.G.) Correspondence: elise.lachat@insa-strasbourg.fr; Tel.: +33-388-14-47-33 ICube Laboratory, Photogrammetry and Geomatics Group, National Institute of Applied Sciences (INSA), 24 Boulevard de la Victoire, 67084 Strasbourg CEDEX, France; tania.landes@insa-strasbourg.fr (T.L.); pierre.grussenmeyer@insa-strasbourg.fr (P.G.) ube Laboratory, Photogrammetry and Geomatics Group, National Institute of Applied Sciences (INSA), Boulevard de la Victoire, 67084 Strasbourg CEDEX, France; tania.landes@insa-strasbourg.fr (T.L.); rre.grussenmeyer@insa-strasbourg.fr (P.G.) p g y g * Correspondence: elise.lachat@insa-strasbourg.fr; Tel.: +33-388-14-47-33 * Correspondence: elise.lachat@insa-strasbourg.fr; Tel.: +33-388-14-47-33 Academic Editor: Assefa M. Melesse Received: 23 January 2017; Accepted: 28 March 2017; Published: 31 March 2017 Received: 23 January 2017; Accepted: 28 March 2017; Published: 31 March 2017 Abstract: Surveying ﬁelds from geosciences to infrastructure monitoring make use of a wide range of instruments for accurate 3D geometry acquisition. In many cases, the Terrestrial Laser Scanner (TLS) tends to become an optimal alternative to total station measurements thanks to the high point acquisition rate it offers, but also to ever deeper data processing software functionalities. Nevertheless, traditional surveying techniques are valuable in some kinds of projects. Nowadays, a few modern total stations combine their conventional capabilities with those of a laser scanner in a unique device. The recent Trimble SX10 scanning total station is a survey instrument merging high-speed 3D scanning and the capabilities of an image-assisted total station. In this paper this new instrument is introduced and ﬁrst compared to state-of-the-art image-assisted total stations. The paper also addresses the topic of various laser scanning projects and the delivered point clouds are compared with those of other TLS. Directly and indirectly georeferenced projects have been carried out and are investigated in this paper, and a polygonal traverse is performed through a building. Comparisons with the results delivered by well-established survey instruments show the reliability of the Trimble SX10 for geodetic work as well as for scanning projects. Keywords: Terrestrial Laser Scanning; robotic image-assisted total station; trimble SX10 scanning total station; surveying and scanning projects; geometric investigations; comparisons HAL Id: hal-02084940 https://hal.science/hal-02084940v1 Submitted on 29 Mar 2019 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. sensors sensors Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station Elise Lachat , Tania Landes and Pierre Grussenmeyer Article Article Investigation of a Combined Surveying and Scanning Device: The Trimble SX10 Scanning Total Station Elise Lachat , Tania Landes and Pierre Grussenmeyer ICube Laboratory, Photogrammetry and Geomatics Group, National Institute of Applied Sciences (INSA), 24 Boulevard de la Victoire, 67084 Strasbourg CEDEX, France; tania.landes@insa-strasbourg.fr (T.L.); pierre.grussenmeyer@insa-strasbourg.fr (P.G.) * Correspondence: elise lachat@insa strasbourg fr; Tel : +33 388 14 47 33 1. Introduction As with almost every research area, the ﬁeld of surveying and its equipment evolve with the technological innovations leading to ever faster, more accurate and more versatile devices. Not only land surveyors but also specialists of various research communities make use of surveying techniques. These can be applied for tasks varying from detection of changes to structural monitoring, whether in geosciences or for industrial applications. Depending on the scale of the surveyed area and on the kind of data required for the analysis, various surveying technologies can be employed. Next to large-scale remote sensing methods that are more suited for covering wide study areas, geodetic measurements through Global Navigation Satellite Systems (GNSS), total stations or Terrestrial Laser Scanners (TLS) are commonly applied. Whereas individual points of interest are selected by the user who guides their direct measurements while using GNSS or total stations, TLS provide dense point clouds measured along a predeﬁned frame and representing the 3D geometry of a scene surface. The 3D coordinates of points of interest are then indirectly obtained through point cloud processing. Due to the development of ever more powerful stereo-matching algorithms, photogrammetry methods also lead to the creation of dense point clouds. p As regards total stations, over the last ﬁfty years this speciﬁc equipment has known drastic changes. From manual tacheometers to motorized robotic total stations, the electronics have brought Sensors 2017, 17, 730; doi:10.3390/s17040730 www.mdpi.com/journal/sensors www.mdpi.com/journal/sensors 2 of 23 Sensors 2017, 17, 730 numerous new facilities to surveyor instrumentation. Nowadays, almost every total station is capable both of performing fast reﬂectorless measurements and of prism autotracking. Besides, steering the total station through one or more cameras integrated into the hardware is possible and prototypes were studied in the 1990s as in [1]. The adjunction of a GNSS receiver is also possible on some current instruments. In fact, total stations now tend to become multi-task devices composed of several built-in sensors. A comprehensive review of the total station evolution is proposed in [2]. Parallel to these developments, in recent decades TLS have gained wide acceptance and their use is more and more generalized for surveying and monitoring tasks. In some cases, TLS even tend to become optimal alternatives to total stations [3,4], even if the latter remain valued devices. 2. Modern Total Stations—State of the Art and Case Studies Before using the Trimble SX10 in various projects to investigate the geometric quality of the data it delivers, in this section some of its technical speciﬁcations are confronted to those of other instruments offering similar functionalities. 1. Introduction Thanks to ever more accurate and faster sensor technology, as well as always deeper data processing software capabilities, terrestrial laser scanning has now been proved as a reliable technique for high accuracy, 3D mapping and documentation of physical scenes and structures [5,6]. Besides, the research topics around scanning technology are still growing, regarding for example the geometric accuracy of the produced data [7] or the inﬂuence of different parameters [8]. However using most of the TLS, the dense point clouds captured are known in a local reference system related to the scanner. Thus conventional surveying techniques are still required to achieve what is known as the georeferencing of the project. To take advantage of both geodetic measurement principles as well as scanning facilities, the trend is now to combine a laser scanning module within the total station hardware to build a unique device. The recent Trimble SX10 scanning total station [9] released in October 2016 belongs to this category of innovative and versatile multi-sensor systems. This article aims at observing and assessing the Trimble SX10 capabilities by means of experiments carried out based on its different measurement modules. To put the recent device in the context of modern total stations development, it is ﬁrst compared to recent total stations. The proposed review is illustrated by examples of case studies applying these techniques. A second part is dedicated to the investigation respectively of the Electronic Distance Measurement (EDM) unit (Section 4), and of the scanning module (Section 5). The idea is to focus on real projects which have been formerly handled using other instruments, but for which the use of the Trimble SX10 would represent an added value during the survey. The collected datasets have thus been compared to the results obtained with conventional total station and Terrestrial Laser Scanner. 2.1. Overview of Image-Assisted and Scanning Total Stations Despite the diversity of surveying instruments available to land surveyors, the total station is still a staple instrument in recording punctual and precise measurements. From the initial theodolites allowing the user to measure horizontal and vertical angles, to the instrument known as tacheometer or total station including an Electronic Distance Measurement (EDM) unit to record distances, the electronics have contributed to the improvement of hardware. Today total stations are moving towards robotic multi-sensor systems which combine ever more sophisticated functionalities. Thus the broad range of instruments available on the market may appear staggering, as described in the recent report in [10]. Scherer and Lerma [2] provide a comprehensive and thorough review about the historic and constant evolution of this technology, as also do [11]. It shows that not only the EDM has evolved, allowing for example reﬂectorless distance measurements, but also the adjunction of built-in sensors has contributed to changing the hardware. In fact, most of the current total stations are robotic and motorized, achieving among other features the automatic recognition and tracking of reﬂector prisms or the remote steering of the device by the user. All these new functionalities offered by total stations have triggered further improvements and innovations in these surveying instruments. 3 of 23 Sensors 2017, 17, 730 Total stations which are equipped with one or more built-in digital camera(s) for targeting and steering operations are reported in the literature as Image-Assisted Total Stations (IATS), see for example [2,12,13]. Even if the idea of combining theodolite and camera does not have a recent origin as explained in [2], the ﬁrst prototypes of a modern device combination were constructed in the 2000s. Prototypes applying this major improvement are described notably in projects from universities [1,12,14–16]. While dealing with such an instrument, its calibration is highly important. Calibration models for the on-board cameras are described in [12,17,18]. The calibration ensures the perfect coaxiality between the built-in camera and the telescope line of sight, so that the observed image corresponds to the targeted area containing the point to be measured. Ehrhart and Lienhart [18] speciﬁcally analyze different error sources related to the use of an IATS as well as their impact on experimental geodetic network measurements. p g More recently a few manufacturers of surveying instruments have introduced scanning functions into their total station hardware. 2.1. Overview of Image-Assisted and Scanning Total Stations This innovation was driven by the generalized use of TLS in surveying activities, and made possible by the combination of instrument motorization and reﬂectorless distance measurement. First releases of such modern total stations offer some pre-built scanning modules such as line or grid scanning. This is the case of the Topcon Imaging Station series launched on the market in 2008 and illustrated by the Topcon IS-3 [19] in Table 1. In addition to imaging facilities and the possible creation of 360° panoramas, this instrument also offers simple scanning functions with a measuring rate lower than 20 points per second. This means that points are captured in a deﬁned window area along a grid interval speciﬁed by the user, but without user intervention during measurements and in particular no need to aim at each individual point. Table 1. Overview of scanning total stations with some of their technical speciﬁcations according to manufacturers. Topcon IS-3 Leica MS60 Trimble SX10 Date of release 2011 2015 2016 Angular accuracy 1”, 3” or 5” 1” 1” EDM maximum range 5000 m (prism) 10,000 m (prism) 5500 m (prism) 2000 m (non prism) 2000 m (non prism) 800 m (non prism) EDM accuracy (prism) 2 mm + 2 ppm 1 mm + 1.5 ppm 1 mm + 1.5 ppm EDM accuracy (non-prism) 10 mm + 10 ppm 2 mm + 2 ppm 2 mm + 1.5 ppm Imaging Number and kind of cameras wide-angle + coaxial overview + telescope overview + primary + coaxial Resolution 1.3 megapixel 5 megapixel 5 megapixel Frame rate up to 10 Hz up to 20 Hz up to 15 Hz Scanning Maximum rate 20 pts/s 1000 pts/s @ 300 m 26,600 pts/s Maximum range 2000 m 1000 m (limited to 1 Hz) 600 m Scanning range noise - 1 mm @ 50 m 1.5 mm @ 50 m Table 1. Overview of scanning total stations with some of their technical speciﬁcations according to manufacturers. Table 1. Overview of scanning total stations with some of their technical speciﬁcations according to manufacturers. 2.1. Overview of Image-Assisted and Scanning Total Stations There is quite a signiﬁcant gap between scanning rates of total stations such as the previous Trimble VX Spatial Station or more recently Trimble S9 (Trimble Inc., Sunnyvale, CA, USA) (15 pts/s) but also Topcon IS-3 (Topcon Positioning Systems Inc., Livermore, CA, USA) (20 pts/s, see Table 1), and scanning rates of current TLS which are able to capture millions of points a second. This gap was slightly reduced as Leica introduced its Nova series total stations. The Leica Nova MS50 4 of 23 Sensors 2017, 17, 730 (Leica Geosystems, Heerbrugg, Switzerland) was launched in 2013 and a new release called Leica MS60 has been marketed since 2015 [20]. They present the same scanning speed of about 1000 points per second as indicated in Table 1. The very latest release of Trimble SX10 in October 2016 shows a promising improvement since it offers an increased scanning rate of 26,600 points per second [9]. Further speciﬁcations of Topcon IS-3 (Topcon Positioning Systems Inc., Livermore, CA, USA), Leica MS60 (Leica Geosystems, Heerbrugg, Switzerland) and Trimble SX10 (Trimble Inc., Sunnyvale, CA, USA) are listed in Table 1. Besides, because the capability of total stations to feature scanning facilities is a quite recent improvement, the names given by manufacturers or stated in the literature for this technology can vary. Leica often calls its instruments of Nova series multi-stations, whereas Trimble refers to SX10 device as a scanning total station. Since the second acronym is more general, it will be used in the remaining parts of this article to refer to this technology. g g p gy Wagner [13] presents the speciﬁcations of two other modern total stations in a table similar to Table 1. When having a look at these two tables, it appears that all major surveying instrument manufacturers (Leica, Trimble and Topcon among others) currently develop IATS which integrate some scanning functions. Nevertheless regarding scanning speciﬁcations and more particularly the scanning rate, one can say that actually only Leica MS50 or MS60 and Trimble SX10 total stations are capable of efﬁcient scanning. A big advantage of these devices is the possibility to measure all point coordinates in a same unique reference system. It should be noticed that, nowadays, the high difference of scanning rate between scanning total stations and TLS is mainly due to the mass of the total station telescope that has to be moved during the scanning process [2,13]. 2.1. Overview of Image-Assisted and Scanning Total Stations The considerably higher measuring speed of Trimble SX10 compared to Leica MS60 is due to the absence of telescope on the SX10 device, which is one of the most original features of this instrument. Since most of the total station users are used to manually aiming by looking through a telescope, this is a daring decision of the manufacturer which clearly beneﬁts the scanning rate. As explained later in Section 3.1, the telescope functionality is replaced by the presence of cameras and the use of a remote controller shown in the corresponding picture in Table 1. 2.2. Related Work A literature review shows that IATS and scanning total stations have been adopted in several ﬁelds which usually apply conventional tacheometry. The advantages related to the use of modern devices is often underlined in these papers. Unfortunately most of the time contributions do not focus on the assessment of the devices but rather on their use in speciﬁc case studies, except when considering contributions related to prototypes of IATS. This is even more true for scanning total stations which were released later on. Besides, it appears that scanning total stations are sometimes only applied for the cameras they contain. In the following subsection an overview of contributions from different application ﬁelds using IATS and scanning total stations is proposed. It mainly encompasses contributions where the devices of Table 1 were used. A ﬁrst application ﬁeld where the results are promising deals with the monitoring of civil engineering structures. Examples of bridge monitoring [21], dam monitoring [22] as well as chimney deformation monitoring [23] are reported. The main advantage mentioned of IATS is the possibility of automatically detecting targets thanks to the camera-based views without the necessary use of reﬂective prisms, as do [21] to investigate vibrations and displacements on a footbridge. While this previous work only uses the camera capabilities of the Leica MS50 to prevent structural failure, in [22] the grid scanning module of this same device was used to model a dam surface. The dam surface was scanned as well as speciﬁc cross-sections using the line scanning module to carry out subsidence analysis during loading experiments on the dam. It appears that the data captured by the scanning total station was valuable to investigate the overall deformation behavior since the acquired point clouds enabled the identiﬁcation of different deformation patterns. For industrial chimney deformation monitoring, Zheng et al. [23] use a Topcon IS to capture oriented images from several stations distributed in a network around the chimney. Outlines of the chimney are extracted from 5 of 23 Sensors 2017, 17, 730 these pictures after different image processing steps, and a model of the surveyed object is ﬁnally reconstructed. A last example related to civil engineering reports on the survey of reservoirs with UAV photogrammetry [24]. To validate the photogrammetry-based model, a reference model of the reservoirs is created thanks to a Leica MS60. 2.2. Related Work This kind of instrument appears as a suitable solution to the authors since a station network is necessary to obtain a complete model thanks to multiple scan stations. In the ﬁeld of geosciences, deformations or changes of natural areas are often monitored using GNSS and conventional tacheometry. An example of a project whose aim is the development of an early warning system for alpine instable slopes is presented in Thuro et al. [25]. In this context, monitoring of landslides is performed thanks to an IATS prototype based on a conventional Leica total station. The authors use this prototype since the control of the mounted camera with external devices is required. Indeed the video frames are used for the automatic detection of natural targets such as surface rocks or debris in this case, rather than artiﬁcial marks or reﬂectors used in conventional methods. A ﬂowchart that recaps the steps of natural or artiﬁcial target detection with an IATS is also proposed in [25]. Looking at a similar topic, Gomez-Vasconcelos et al. [26] do not only use pictures captured by a Leica Nova MS50 but also the georeferenced point clouds it delivers to investigate tectonic areas. The authors explain that after meshing the point clouds they are able to create a digital outcrop model, and fault displacements are measured within this model. This data is combined with information coming from other sensors in order to predict potential interactions between volcanism and seismic hazards in these areas. Then regarding change detection, erosion of a beach is surveyed in [27]. The changed areas within a determined time interval are studied thanks to the grid scanning module of a Topcon IS. Wagner [13] describes an application example of geo-monitoring that is also related to the study of changes. In this application, displacement vectors between two periods in a mountainous terrain have been computed thanks to a Leica MS50. In such a challenging terrain, the authors highlight the advantage of carrying only one surveying instrument up to the observation point. Some contributions report the use of modern total stations for building survey and cultural heritage applications. Wozniak et al. [28] present results obtained for the creation of a building facade 3D model using data from a Topcon IS-3 as well as images acquired with a conventional digital camera. 3.1. Overview of Built-In Digital Cameras Among the speciﬁcities of the SX10 device, the camera technology is particularly developed since three different calibrated cameras are built into the rotating part of the hardware. Unlike in previous IATS from Trimble (e.g., Trimble S9), not only a wide-angle camera is available but also two other cameras with smaller ﬁelds of view. According to manufacturer documentation, these cameras are called overview, primary and telescope cameras. The overview or wide-angle camera as well as the primary camera are mounted parallel with the EDM axis but with an offset. The telescope camera located in the path of measurement axis is introduced for magniﬁcation. Each camera chip has a 5 megapixel resolution (i.e., 2592 × 1944 pixels), but given their different ﬁelds of view the pixel size differs for each camera as described in Table 2. Table 2. Some speciﬁcations of Trimble SX10 on-board cameras according to manufacturer. Speciﬁcations Overview Camera Primary Camera Telescope Camera Position relative to EDM axis parallel parallel coaxial Field of view 54° 12° 2° Pixel size @ 50 m 20 mm 4.4 mm 0.88 mm Table 2. Some speciﬁcations of Trimble SX10 on-board cameras according to manufacturer. Since there is no true telescope in the hardware, optical sighting is no longer foreseen and is replaced by the use of digital cameras. The device can thus only be controlled remotely through the video stream displayed on an external tablet or a remote controller, given that the hardware does not contain any screen as visible in the ﬁgure of Table 1 in the previous section. The ﬁeld software Trimble Access is used for controlling the device in each kind of operation. After coarsely aiming the device at a survey point, navigation within the video is ensured thanks to the overview and primary cameras which both offer two zoom levels. The user just has to tap on the touch-sensitive screen to more ﬁnely aim at the survey point by overlaying digital cross-hairs with the target. Zoom in for ﬁnal accurate targeting is then enabled by the telescope camera. The different zoom levels offered through the three cameras are also useful while framing an area to be scanned. Not only steering of the device and point measurement are possible with the innovative on-board camera technology used. In many cases, it can be interesting to document the project by taking pictures of the measured points or targets as a reminder. 2.2. Related Work In this study aiming at documenting geometrical features of buildings, the Topcon Imaging Station is used in its standard scanning mode for reﬂectorless point acquisition. Despite a stated advantage of requiring one unique device capable of multiple survey tasks, some main limitations of the device are underlined, such as the low measurement speed or the low image resolution. In the context of Building Information Modeling (BIM), Sepasgozar et al. [29] relate on mobile and static acquisition methods for as-built modeling, given that point clouds are increasingly used as data in construction. Static acquisitions performed with a Leica MS50 are compared with data obtained based on mobile technologies, providing a better spatial accuracy than mobile data, although requiring more time. The same Leica MS50 is used to provide reference data in a similar approach described in [30]. In this second work the authors aim at evaluating a mobile mapping system on a larger scale test ﬁeld, in which control points on building facades and street proﬁles were previously captured with the multi-station. Some more examples rather related to cultural heritage are reported in [31,32] which both use the Topcon IS. Evgenikou and Georgopoulos [31] investigate the 3D modeling of small artifacts thanks to different sensors, and a brief overview of results obtained for architectural survey of a bridge are shown in [32]. Even if point density seems to be the major issue in both cases, the device could be adapted for documenting objects of similar size as a bridge whereas it is visibly not adapted for small artifacts reconstruction. Metrology applications can ﬁnally be reported, considering leveling capabilities with a Leica MS60 [33], or collimation measurements with an experimental camera module on a total station [14]. 6 of 23 Sensors 2017, 17, 730 3. Trimble SX10 Technical Properties Because of the recent release of the Trimble SX10 scanning total station, no scientiﬁc review relates to its use yet. The purpose of this section is to describe and to verify some of the announced device speciﬁcities by focusing mainly on its cameras and scanning modes. 3.2. Speciﬁcations of SX10 Scanning Modes The Trimble SX10 has been devised as an innovative instrument combining the capabilities of a total station with those of a powerful laser scanner. Due to its particular design and to its capacities, it can even be seen as a TLS with total station measurement facilities rather than the contrary. In fact, the Trimble SX10 scanning rate tends to approach the scanning rate of a conventional TLS as exposed in Table 1. This subsection seeks to describe the scanning module contained in the device. The Trimble SX10 makes it possible to acquire dense point clouds of a scene in a very timely manner. The area to be scanned can be directly selected by the user without the necessity to perform a 360° preview point cloud for the selection beforehand. This area is deﬁned through tapping on the touchscreen between four possible speciﬁcities. Full dome acquisition of dense point clouds can be performed to obtain a global overview of the surrounding environment with a maximal ﬁeld of view of 360° × 300°. But the user can also choose to select a viewing window according to three possible kinds of frames: a band deﬁned by two points placed on the upper and lower limits, a rectangle deﬁned by two points or a polygon for more complex shapes. After framing the area of interest to be scanned, other settings can be modiﬁed as for example the point density. The four point densities correspond to 1, 2, 4 or 8 slightly shifted passes of the scanning line from coarse to superﬁne density, and the corresponding point spacings are listed in Table 3. A last setting to parametrize is the choice of the camera that is used to colorize the point cloud, as well as the overlap between these pictures. Table 3. Point spacings achievable with the Trimble SX10 depending on scanning density. Table 3. Point spacings achievable with the Trimble SX10 depending on scanning density. Table 3. Point spacings achievable with the Trimble SX10 depending on scanning density. Coarse Standard Fine Superﬁne Point spacing @ 10 m 10 mm 5 mm 2 mm 1 mm Coarse Standard Fine Superﬁne Point spacing @ 10 m 10 mm 5 mm 2 mm 1 mm A valuable asset while using the SX10 for point cloud acquisition is the possibility of visualizing the acquired data in the ﬁeld. 3.1. Overview of Built-In Digital Cameras Some notes and comments can then be added on the picture through the touchscreen. When opening an exported project in one of the manufacturer provided ofﬁce software, the pictures captured during ﬁeld operations are automatically referenced to the station from which they were taken, as well as to the measured point they correspond to. Pictures can be exported for further ofﬁce applications such as for example post-acquisition documenting. Because of the pixel size offered by the primary camera, it would be interesting in the future to investigate its use in photogrammetry projects. However, the limited resolution of the acquired pictures might be restrictive for such approaches. Another interesting feature is the creation of panoramic images. By selecting an area on the screen, imaging total stations can capture several overlapping pictures in the horizontal and vertical directions. The ﬁeld software Trimble Access offers the possibility to choose the camera that should be used to create the panorama, as well as the overlapping rate between pictures. When the pictures are captured within a geodetic network of known points, the imaging data is directly geolocalized. An example of a panorama performed on a bridge is shown in Figure 1. In addition, the user can choose between 7 of 23 Sensors 2017, 17, 730 the overview and primary camera for point cloud colorization in scanning projects, depending on the appropriate acquisition time and texture quality for the project considered. the overview and primary camera for point cloud colorization in scanning projects, depending on the appropriate acquisition time and texture quality for the project considered. It should be noted that the station setup uses a fourth camera for centering the device above the reference point. A plummet camera with a ﬁxed focus and a 6° ﬁeld of view replaces the optical or laser plummet. This camera-based solution may be confusing for uninitiated users at the beginning. Nevertheless this solution offers a ground pixel size of about 0.2 mm at a 1.55 m instrument height with a quoted accuracy of 0.5 mm at this same height according to the manufacturer. Figure 1. Example of a panoramic image captured with Trimble SX10 on a bridge using the primary camera and an overlap of 20% between individual pictures. Figure 1. Example of a panoramic image captured with Trimble SX10 on a bridge using the primary camera and an overlap of 20% between individual pictures. 3.2. Speciﬁcations of SX10 Scanning Modes With the Trimble Access software a review of the captured point clouds is possible and they can be displayed either in true color, with intensity values or with one color per scan according to the scan station it is related to. Zoom in on the captured point cloud through the tablet is also foreseen, which is valuable in the ﬁeld to ensure there are no missing parts in the surveyed area. The scanning function presented in this subsection is assessed later through a set of comparisons made between some SX10 produced point clouds and datasets captured with other TLS. In this context, 8 of 23 Sensors 2017, 17, 730 Table 4 gives an overview of some major speciﬁcities of Trimble SX10 as well as those from laser scanners used later in this article. It appears that the Trimble SX10 scanning module offers a wider measurement range compared to FARO Focus3D (FARO Technologies, Lake Mary, FL, USA) [34] and Leica ScanStation C10 (Leica Geosystems, Heerbrugg, Switzerland) [35]. The FARO Focus3D is based on phase-shift measurement principle unlike the two other time-of-ﬂight laser scanners, which explains the high scanning rate difference. More recent ScanStations from Leica exist with increased performances in terms of precision and scanning rate, however the C10 was the most recent laser scanner from Leica available in our equipment. Table 4. Comparison of some technical properties of the three scanning devices used in the article, according to manufacturer speciﬁcations. according to manufacturer speciﬁcations. Leica ScanStation C10 FARO Focus3D X330 Trimble SX10 Scanning Total Station Measurement principle time-of-ﬂight phase-shift time-of-ﬂight Scanning rate up to 50 kHz up to 976 kHz up to 26.6 kHz Max. measurement range 300 m 330 m 600 m Accuracy criterion * Accuracy of single dist. measurement: 4 mm @ 1–50 m (one sigma) Ranging error: 2 mm @ 25 m (one sigma) Distance measurement accuracy (DR mode): 2 mm + 1.5 ppm Range noise * 2 mm max. 0.5 mm @ 25 m 1.5 mm @ 50 m Point spacing <1 mm minimum through full range * from 1.5 mm to 49 mm @ 10 m from 1 mm to 10 mm @ 10 m Laser wavelength 532 nm 1550 nm 1550 nm Date of release 2009 2013 2016 * terminology and values directly derived from technical datasheets of respective manufacturers. according to manufacturer speciﬁcations. 3.2. Speciﬁcations of SX10 Scanning Modes Leica ScanStation C10 FARO Focus3D X330 Trimble SX10 Scanning Total Station Measurement principle time-of-ﬂight phase-shift time-of-ﬂight Scanning rate up to 50 kHz up to 976 kHz up to 26.6 kHz Max. measurement range 300 m 330 m 600 m Accuracy criterion * Accuracy of single dist. measurement: 4 mm @ 1–50 m (one sigma) Ranging error: 2 mm @ 25 m (one sigma) Distance measurement accuracy (DR mode): 2 mm + 1.5 ppm Range noise * 2 mm max. 0.5 mm @ 25 m 1.5 mm @ 50 m Point spacing <1 mm minimum through full range * from 1.5 mm to 49 mm @ 10 m from 1 mm to 10 mm @ 10 m Laser wavelength 532 nm 1550 nm 1550 nm Date of release 2009 2013 2016 * terminology and values directly derived from technical datasheets of respective manufacturers. 3.3. Practical Investigation of Maximal Measurement Range In order to assess the maximal measurement range announced by the manufacturer, an outdoor experiment was carried out. To ensure a large visibility, the tests were performed from a building roof surrounded by other buildings in the city center of Strasbourg, France. A 360° band-shaped point cloud was acquired from the roof with a low point density to keep a moderate scanning duration. Figure 2 shows a top view of the acquired point cloud. Centered on the scanner location which is known, circles have been drawn with rays representing different acquisition ranges. The color codiﬁcation represents the points divided per 100 m range. In Table 5, the number of points belonging to each 100 m portion is reported, as well as the percentage it represents with respect to the number of points of the whole point cloud. 9 of 23 Sensors 2017, 17, 730 Figure 2. Top view of a 360° point cloud acquired in band scanning mode from a building roof. The point cloud is colorized according to range and signiﬁcant ranges are represented. Figure 2. Top view of a 360° point cloud acquired in band scanning mode from a building roof. The point cloud is colorized according to range and signiﬁcant ranges are represented. Table 5. Percentage of points contained in each 100 m portion around the scanner station. Range Interval Number of Points Percentage According to Whole Point Cloud <100 m 412,861 32.95% 100 m–200 m 446,500 35.64% 200 m–300 m 235,924 18.83% 300 m–400 m 71,201 5.68% 400 m–500 m 38,714 3.09% 500 m–600 m 33,311 2.66% 600 m–700 m 11,558 0.92% 700 m–800 m 1313 0.10% 800 m–900 m 1425 0.11% >900 m 125 0.01% Range Interval Number of Points Percentage According to Whole Point Cloud It appears that the furthest point detected is about 900 m from the scanner location, which means that the announced maximal measurement range of 600 m is well respected. However, the farther the points are located, the lower the point density. Indeed, points acquired beyond the announced maximal range are more scattered. This is conﬁrmed while cumulating the proportion of points contained within the ﬁrst intervals and listed in Table 5. About 99% of the total amount of points are concentrated before 600 m range. 4.1. Total Station Working Principle Standard surveying solutions are included in the SX10 device. Regarding station setup, similar options to those present in conventional tacheometers are available. Depending on the existing references on the ﬁeld, the user can choose between resection or stationing on a known point with orientation on one or several reference(s). Measurements can be automatically repeated in both telescope positions through direct and reverse angle measurements. The user can deﬁne beforehand which points should be observed twice, such as references or control points for example. Besides, numerous coordinate geometry (known as COGO) computations are handled by the device. These functionalities include point coordinates calculations, but also intersection computations or point and line stakeout among other features. Station setup as well as COGO computations are controlled through the ﬁeld software Trimble Access during surveying and scanning projects. All the measured data are integrated and stored into a same job created in the Trimble Access software. A very useful functionality offered by the software interface is the real-time viewing of the acquired data directly on the tablet or controller screen. The whole project and thus topographic, optical and scanning data it contains can be displayed all together. This helps in the ﬁeld by making it possible to control the potential missing areas that still have to be captured. Another particularity while using the Trimble SX10 as a total station is the presence of a camera system. As a matter of fact, manual sighting is no longer foreseen because of the absence of an optical telescope on the device. The user thus steers the instrument through real-time video streaming. Movements of the sighting axis to aim the instrument towards the targeted point are directed by tapping directly on the screen. As during optical sighting, coarse and ﬁne targeting are necessary to accurately reach out to the target. Coarse targeting is ﬁrst achieved using the overview and primary cameras. Accurate targeting is performed in a second step thanks to the telescope camera which offers a better image magniﬁcation. This process is detailed in Section 3.1 related to on-board cameras. 4. Investigation of the EDM Unit and General Surveying Functionalities The Trimble SX10 scanning total station can be seen as a hybrid solution. Since the device offers more than the powerful laser scanning functionalities described in Section 3.2, it also belongs to the category of total stations. One can even speak of an Image-Assisted Total Station because of the cameras it contains as described in Section 3.1. In this section the use of total station functionalities is introduced and illustrated by a network survey carried out in a building. 3.3. Practical Investigation of Maximal Measurement Range The last portions of points between 600 m and 900 m count about 14,000 points, which represents only 1% of the total amount of acquired points. Besides, the geometry of the buildings located at this range is hardly distinguishable, since most of the time only a few points were acquired on these faraway buildings. 10 of 23 Sensors 2017, 17, 730 10 of 23 4.2. Geodetic Traverse Survey The surveying solutions proposed by Trimble SX10 device have been investigated through the achievement of a geodetic closed traverse. The measurements were performed through a building and some point clouds have been acquired on rooms along the traverse. Topographic data are displayed together with captured point clouds in Figure 3a. This traverse begins outdoors at the building ground ﬂoor level on an unknown point, using references to determine its coordinates through resection. It then passes through the ﬁrst ﬂoor before closing on the ﬁrst survey point located on the ground ﬂoor. This is the reason why the measurements appear to be superimposed on the top view as proposed in Figure 3a. A prism ﬁxed on a tripod was used to signal each network point. Laser scanning survey as well as traverse measurement were carried out within the same day, and the ﬁnal loop closure of the traverse is useful to control the coordinates of the two last measured points in this particular case. The fact that the traverse is closed also enables the computation of the adjusted coordinates of the measured points. Both ﬁeld and ofﬁce applications have been experimented during this project, using the manufacturer ofﬁce software Trimble Business Center (TBC) for data post processing. This software offers several facilities including visualization of the acquired data but also adjustment computations. Figure 3a is a screenshot obtained from TBC and shows the considered network points which are 11 of 23 Sensors 2017, 17, 730 linked. Since measurements on references have been carried out from some traverse points, the network adjustment tool was chosen to compute the adjusted values of the point coordinates. In this particular project, the standard deviations do not exceed 8 mm for all adjusted network points with a mean value of about 4 mm. Error ellipsoids are also computed for each network point during adjustment computation in order to provide an idea of point precision after adjustment. They can be plotted together with the points they correspond to, as can be seen in Figure 3a. This ﬁgure illustrates that error ellipses in the planar view are larger particularly for points obtained through polar measurements (points 101, 1000 and 1001). The knowledge of these statistic values provides valuable information for further inspection and conﬁdence analysis of the project. (a) (b) (c) Figure 3. 4.2. Geodetic Traverse Survey (a) Top view of the network performed within a building with error ellipsoids computed after adjustment for each network point. The point clouds acquired during this project appear in grayscale. (b) and (c): Visualization under two different views of cloud-to-cloud distances computed between point clouds of the room acquired from station 1001, before and after network adjustment. Deviations unit: meter. (a) (a) (b) (c) (b) (c) Figure 3. (a) Top view of the network performed within a building with error ellipsoids computed after adjustment for each network point. The point clouds acquired during this project appear in grayscale. (b) and (c): Visualization under two different views of cloud-to-cloud distances computed between point clouds of the room acquired from station 1001, before and after network adjustment. Deviations unit: meter. Network adjustment has been recomputed using another software dedicated to the adjustment of geodetic networks (Covadis, Geomedia). Since the deviations between the ﬁnal coordinates adjusted by this means and through TBC never exceed 1 cm, the results obtained thanks to the manufacturer software can be validated. Besides, network adjustment has a direct inﬂuence on the coordinates of 12 of 23 Sensors 2017, 17, 730 point clouds contained within the project. To illustrate this inﬂuence, the whole point cloud acquired during the project has been exported twice, ﬁrst as raw data and then after network adjustment. The distances computed between both point clouds focusing on one room are shown from two opposite points of view in Figure 3b,c. The cloud-to-cloud comparison has been performed using the dedicated free software CloudCompare [36]. Deviations no larger than 2.5 cm and with a mean value of about 5 mm are observed. It also appears that the larger deviations are predominantly visible on two opposite faces of the room, which means that the offset between both datasets mainly occurs along one direction. Their order of magnitude near to 1 cm almost corresponds to the deviation between station coordinates before and after adjustment. These deviations caused by the polar measurement of successive points within a traverse attest to the adjustment inﬂuence on scanning data. 5. Assessment of Laser Scanning Projects Some laser scanning projects have been conducted with the Trimble SX10, whereby different objects and georeferencing methods were considered. When the coordinates of reference points are known in the project area, a direct georeferencing of the acquired point clouds can be performed. This consists in the capture of point clouds which are directly known in a required reference system. Due to surveying facilities included in the Trimble SX10 device, the scanning projects it provides may be directly georeferenced without supplementary material. On the other hand, georeferencing can be performed as a post-processing step, consisting of the transformation of all points of the data into the required geodetic coordinate system. This process, known as indirect georeferencing is often made based on the measurements of targets with known coordinates in the destination system [37]. Two ﬁeld cases are presented to illustrate this section. The ﬁrst one is a church facade scanned from three different setup locations without consideration of georeferencing, whereas the second ﬁeld case is a bridge scanned from six different setup locations and directly georeferenced. 5.1. Investigation of a Building Facade Survey A ﬁrst scanning project has been performed on the facade of a church which is located in Strasbourg, France. The facade has been captured using the Trimble SX10 from three different stations as shown in Figure 4a. The aim of this ﬁrst project was to investigate the quality of the delivered point cloud on a detailed facade, and not to study the georeferencing capability. A network of known points available in the near environment of the church has still been used to roughly align the three point clouds in the ﬁeld, as illustrated in Figure 4b which presents the three segmented point clouds. To correct the inﬂuence of potential uncertainties due to the survey and to really concentrate on the overall geometry, the point cloud alignment was reﬁned by applying an algorithm based on the Iterative Closest Point (ICP) principle. This was made thanks to the manufacturer software Trimble RealWorks which delivered the lower remaining error after performing ICP in this project. Figure 4c depicts the result obtained after the three point clouds have been ﬁnely registered. With remaining alignment errors no larger than 1 cm between these individual point clouds, it clearly appears in this Figure that the data better overlap. The ﬁnal segmented point cloud counts about 13 millions of points. Depending on the range between instrument and surveyed object, the ﬁne resolution (stations 11 and 12) or the standard resolution (station 13) were used, leading to a global acquisition time of about 30 min. In order to assess the quality of the global point cloud issued by the recent device, several comparisons are carried out with a reference dataset of the same church facade. The reference dataset has been acquired on the same day with a commonly used laser scanner Focus3D from FARO. The reference point cloud counts about 50 millions of points captured from three scanning stations. Since the FARO device cannot be centered on a known point, these stations slightly differ from the ones used during the SX10 acquisitions. To perform comparisons between the Trimble SX10 and FARO Focus3D point clouds, the free software CloudCompare is used. According to the computed cloud-to-cloud deviations presented in Figure 4d, a mean deviation smaller than 1 cm can be observed. 13 of 23 Sensors 2017, 17, 730 Cloud-to-mesh distance computations are performed on these limited areas, which has the advantage of delivering signed values of the deviations unlike previous cloud-to-cloud comparisons. The Poisson surface reconstruction algorithm available in CloudCompare was applied on the dense point cloud acquired by FARO Focus3D in order to create a reference mesh. This mesh and the areas segmented from the SX10 point cloud were ﬁnely registered using an ICP-like algorithm, delivering a ﬁnal alignment error of about 9 mm and 1 cm respectively for the rose window and tympanum, and about 2 mm for planar surfaces. (a) (b) (c) Figure 5. (a) Overview of the facade containing control points and locations of the areas selected for ﬁner analyses; (b,c) Cloud-to-mesh comparisons performed on detailed areas between the SX10 point cloud and meshed parts of FARO Focus3D point cloud: the rose window (b) and the tympanum (c). Deviations are in meters and projected on SX10 point cloud. (b) (b) (c) (a) (c) Figure 5. (a) Overview of the facade containing control points and locations of the areas selected for ﬁner analyses; (b,c) Cloud-to-mesh comparisons performed on detailed areas between the SX10 point cloud and meshed parts of FARO Focus3D point cloud: the rose window (b) and the tympanum (c). Deviations are in meters and projected on SX10 point cloud. Figures 5b,c show the deviations that were computed between the mesh and segmented point clouds for rose window and tympanum. These ﬁgures ﬁrstly illustrate the ﬁne level of detail that is reached thanks to the SX10 point clouds. The resulting point clouds appear to be geometrically accurate since a standard deviation of about 2 mm is observed in both cases. As visible on both ﬁgures, the highest deviations are present in areas which are hardly accessible for the laser beam from the ground, producing grazing angles. It should also be noticed that the mesh creation process may have a small inﬂuence on the initial point cloud geometry, but this does not have much impact on the ﬁnal analysis. Then regarding planar areas on the left and right parts of the facade (blue rectangular areas in Figure 5a), a standard deviation which varies between 3 and 4 mm with respect to adjusted planes is computed. These results are reliable considering the fact that the surveyed planar surfaces are made of stones which are thus not perfectly planar and aligned. Sensors 2017, 17, 730 Sensors 2017, 17, 730 Sensors 2017, 17, 730 However some high deviations appear in particular on the windows located in the lower part of the facade, but this can be explained by the noise caused by this reﬂective material. It should also be noticed that the high values visible above tympanum are caused by a banner which was present during the acquisitions. Finally, the presence of non-overlapping parts between both datasets due to slightly different viewing angles from the scan stations produce small deviations, which may be visible in particular on edges next to cornices, windows and pillars, and more generally on complex parts of a facade. Given that the deviations provided by a cloud-to-cloud comparison are non-signed, further comparisons have been performed in order to derive ﬁner statistical values. (a) (b) (c) (d) Figure 4. (a) Overview of the area around church facade; (b) Three point clouds acquired on the surveyed church facade without alignment reﬁnement; (c) Point clouds ﬁnely registered after Iterative Closest Point (ICP); (d) Cloud-to-cloud comparison between FARO Focus3D and Trimble SX10 point clouds of the facade. Deviations (unit: m) are projected on SX10 point cloud. To gain a more reliable idea of the Trimble SX10 point cloud quality, a statistical analysis of (a) (a) (a) (d) (b) (c) (d) Figure 4. (a) Overview of the area around church facade; (b) Three point clouds acquired on the surveyed church facade without alignment reﬁnement; (c) Point clouds ﬁnely registered after Iterative Closest Point (ICP); (d) Cloud-to-cloud comparison between FARO Focus3D and Trimble SX10 point (c) (b) (d) (b) (c) Figure 4. (a) Overview of the area around church facade; (b) Three point clouds acquired on the surveyed church facade without alignment reﬁnement; (c) Point clouds ﬁnely registered after Iterative Closest Point (ICP); (d) Cloud-to-cloud comparison between FARO Focus3D and Trimble SX10 point clouds of the facade. Deviations (unit: m) are projected on SX10 point cloud. To gain a more reliable idea of the Trimble SX10 point cloud quality, a statistical analysis of the computed deviations has been performed based on some detailed and representative parts of the facade. The selected areas highlighted in Figure 5a are a rose window and a tympanum with 14 of 23 14 of 23 Sensors 2017, 17, 730 moldings (green areas in the Figure), as well as planar surfaces (blue rectangular areas). 5.2. Assessment of a Directly Georeferenced Project The second scanning project which is presented in this article involves a road bridge located near the Franco-German border in the Alsace region, France. A network of known points has been previously built in this area by determining the coordinates of some points with a GNSS receiver. Figure 6a presents an overview of the area containing reference points used during the project. This ﬁgure also shows that the four reference points below the bridge on the riverbank have been used as scanning stations. For each of these stations, the Trimble SX10 was set up on one considered reference point and oriented on the three others. To complete the point cloud of the bridge, two further scanning stations have been set up, under the bridge deck on one side and on the deck on the other. The new station coordinates have been determined through resection based each time on three visible references. Thus each of the six acquired point clouds is known in the coordinate system applied in this study, so that the whole scanning project is directly georeferenced. A station view from TBC software is presented in Figure 6b. On such a view, the user can visualize the georeferenced point clouds and pictures acquired from the selected station, as well as references and surveyed points. The same scene was then acquired using a Leica ScanStation C10 which also makes it possible to directly georeference scanning projects. Compared to the Trimble SX10, targeting of individual points is not foreseen in the hardware surveying facilities. This is the reason why georeferencing is performed by stationing on a known point and scanning a small area around a speciﬁc target placed on another known point. Six scanning stations were also useful with the Leica C10 to capture not only the structure geometry but also some parts under and on the bridge deck. The high resolution parameter was chosen during Leica C10 acquisitions, corresponding to a point spacing of 5 cm at 100 m. In comparison, the Trimble SX10 point clouds have been acquired using the standard resolution parameter (5 mm @ 10 m) except for the station under the bridge where the coarse parameter (10 mm @ 10 m) has been used. After segmenting the datasets to keep only the considered structure and to remove the dense surrounding vegetation, Trimble SX10 whole point cloud counts about 9 millions of points. Sensors 2017, 17, 730 The same statement is made while adjusting planes on the same segmented areas from the FARO Focus3D point cloud, where standard deviations between 5 and 6 mm are observed. This difference in terms of deviation is mainly caused by the higher amount of points in the FARO Focus3D segmented point clouds which are about twice as dense as those from the Trimble SX10. To complete these analyses, further comparisons performed on the 15 of 23 15 of 23 Sensors 2017, 17, 730 same Trimble SX10 dataset but using other devices than the FARO laser scanner are presented in [38]. The computed deviations are very similar to those obtained here with the FARO Focus3D. Finally, some control points were chosen on the facade as depicted in Figure 5a. Using a Leica TS02 total station, their 3D coordinates were determined in the local network partially illustrated in Figure 4a. These points were then measured thanks to the Trimble SX10 surveying facilities from station 1 in front of the facade (see Figure 5a). It should be noted that due to the height of the facade, it was impossible to physically mark these points on the ediﬁce. The 3D deviations computed between the two surveys vary from 5 mm up to 2.5 cm. These results are acceptable given that the control points are natural features of the facade. 5.2. Assessment of a Directly Georeferenced Project p The aim of this second scanning project is mainly to assess the performed georeferencing. In the following comparisons, the segmented raw data are considered without reﬁnement of the individual point clouds alignment. This is meant to avoid skewing the georeferencing if the initial registration is modiﬁed. Cloud-to-cloud comparisons using CloudCompare software have been carried out between Leica C10 and Trimble SX10 point clouds. Because the free stations for which the coordinates are computed by resection (see stations St.1 and St.2 in Figure 6a) were not located exactly on the same points during both scanning campaigns, hidden parts differ in both point clouds, in particular under the bridge deck. Such areas with little overlap between the two datasets may cause the computation of high cloud-to-cloud deviations which are not relevant. Besides, the amount and scattering of points acquired on the deck signiﬁcantly vary between the two point clouds. The two main reasons which explain these differences are unavoidable and cannot be managed by the user: ﬁrst the high road trafﬁc on such a structure creates different masks between two acquisition campaigns; then the way grazing angles are managed and ﬁltered by the instruments is very likely different for both devices. In order to better visualize the small deviations that are meaningful between the datasets, points belonging to 16 of 23 16 of 23 Sensors 2017, 17, 730 previously described areas and presenting obviously too high deviations have been ﬁltered. The result of cloud-to-cloud distance computation on the remaining bridge structure is shown in Figure 7a. (a) (b) Figure 6. (a) Overview of the area around surveyed bridge with four of the Global Navigation Satellite Systems (GNSS) reference points; (b) Station view captured from Trimble Business Center ofﬁce software. Georeferenced point cloud (in grey) and images are visible as seen from the considered station (station G1 in this case). (a) (a) (a) (b) (b) Figure 6. (a) Overview of the area around surveyed bridge with four of the Global Navigation Satellite Systems (GNSS) reference points; (b) Station view captured from Trimble Business Center ofﬁce software. Georeferenced point cloud (in grey) and images are visible as seen from the considered station (station G1 in this case). (b) Figure 6. (a) Overview of the area around surveyed bridge with four of the Global Navigation Satellite Systems (GNSS) reference points; (b) Station view captured from Trimble Business Center ofﬁce software. 5.2. Assessment of a Directly Georeferenced Project Georeferenced point cloud (in grey) and images are visible as seen from the considered station (station G1 in this case). A ﬁrst look at the comparison shows that no signiﬁcant geometrical deviations of the structure are observed between both datasets. A mean deviation smaller than 2 cm can be derived from this analysis for the bridge main structure, where points belonging to upper and lower parts of the deck have been ﬁltered. A closer look at the comparison result provides relevant information about the georeferencing. As a matter of fact, deviations of the same order of magnitude are observed on the opposite sides of each bridge abutment. Since this appears on opposite sides with respect to the road axis, it illustrates that C10 and SX10 point clouds have a slightly different orientation due to their respective georeferencing. While superimposing both raw segmented point clouds, this trend becomes clearer. Figure 7 shows that on the presented side, the blue dataset (Trimble SX10) stands out on the right abutment whereas the red one stands out on the left part. This is also visible from the other side of the bridge point cloud. On these particular endpoints of the bridge, computed deviations do not exceed 6 cm and are almost the same for each opposite abutment. However the overall structure of the point clouds do not suffer from any signiﬁcant distortion. Regarding the bridge 17 of 23 17 of 23 Sensors 2017, 17, 730 length, the orientation deviation between both bridge axes is acceptable. The presence of this deviation highlights how important the reference targeting performed to compute georeferencing is. The result obtained underlines that georeferencing mainly depends on the network quality and can vary while using different instruments. (a) (b) Figure 7. (a) Cloud-to-cloud comparison between Leica C10 and Trimble SX10 directly georeferenced point clouds of the bridge. Deviations (in m) are projected on SX10 point cloud; (b) Overview of both Leica C10 (in red) and Trimble SX10 (in blue) point clouds displayed together. (a) (b) (b) Figure 7. (a) Cloud-to-cloud comparison between Leica C10 and Trimble SX10 directly georeferenced point clouds of the bridge. Deviations (in m) are projected on SX10 point cloud; (b) Overview of both Leica C10 (in red) and Trimble SX10 (in blue) point clouds displayed together. 5.2. Assessment of a Directly Georeferenced Project To complete the study in particular on the bridge deck, the Trimble SX10 point cloud has been confronted to a previous total station survey. Altitudes of some signiﬁcant points of the deck have been compared to existing georeferenced proﬁles. The considered points which have been manually selected in the point cloud are the road axis, the limits of the pavement and the sidewalks. Based on comparisons performed on 11 successive proﬁles 2 m apart (see Figure 8a), the mean altitude deviation computed was about 7 mm. A detailed analysis of each proﬁle shows that the highest deviations are mainly observed on the last sections near the middle of the bridge deck because of the lower point density. Points acquired on this part of the deck are more scattered due to grazing angles and thus more noise is present in the point cloud. For the remaining sections, the deviations are randomly distributed between negative and positive values and no systematic error is observed. Since GNSS has been used to observe the initial network (Figure 6a), the achieved georeferencing could not offer better results. To complete the analysis, Figure 8b presents a section carried out in the point cloud 18 of 23 Sensors 2017, 17, 730 displayed together with its corresponding tacheometric proﬁle, in red. It can be seen that the point cloud geometry accurately follows the previous surveying measurements. (a) (a) (b) Figure 8. (a) Top view of the bridge point cloud with locations of the investigated sections (red lines); (b) Section A performed in Trimble SX10 point cloud and displayed together with the tacheometric proﬁle data (in red). (b) (b) (a) Figure 8. (a) Top view of the bridge point cloud with locations of the investigated sections (red lines); (b) Section A performed in Trimble SX10 point cloud and displayed together with the tacheometric proﬁle data (in red). 5.3. Uncertainty Analysis Related to Object Color and Materials 5.3. Uncertainty Analysis Related to Object Color and Materials It has been shown many times in the literature that the properties of the scanned object such as color, roughness or reﬂectivity [39,40] and even the moisture [41] of the scanned material can have an inﬂuence on the distance measured by laser scanners. The incidence angle of the laser beam on the reﬂected surface is also of relevance [8]; nevertheless this speciﬁc point was not examined during our tests. Since the previous scanning projects were carried out in dry weather, the performed laboratory analyses mainly focus on the type of material, as well as on its color and reﬂectivity. A board composed of various samples has been scanned with the Trimble SX10 at a distance of about 9 m. Using the superﬁne scanning resolution, the point cloud representing the sample board counts slightly more than 1 million points. On this colorized point cloud shown in Figure 9a, it appears that the board is composed of color samples on the right part, and different kinds of wood samples in the middle. On the left part, samples with different textures and roughnesses can be seen. Flat black and white targets have also been placed in the scene, as well as a mirror in the lower right corner or bricks and cement blocks on each side of the board. Regarding the intensity values provided by the laser scanner and depicted in Figure 9b, it is clearly visible that black and reﬂective materials present low intensity values. As a matter of fact, reﬂective materials such as the mirror surface or the compact disk cause a specular reﬂection of the laser beam whose few signal is returned back to the TLS. Then, since black materials are very absorbing, the amount of signal they reﬂect back to the scanner is also small. Considering the remaining samples, it appears that the color variations do not really inﬂuence the amount of returned intensity. To better investigate the geometry of the resulting point cloud by comparing it to data coming from another laser scanner, the board was also scanned with a FARO Focus3D placed at the same location. Both point clouds were ﬁrst aligned using an ICP-like algorithm (alignment error: 4 mm), and were then compared using the CloudCompare software. Figure 9c shows a cloud-to-cloud comparison between both Trimble and FARO point clouds, whereby the computed distances are projected on the Trimble SX10 point cloud. 6. Discussion It is obvious that total stations are nowadays becoming smarter thanks to the integration of new capabilities and sensors into the hardware. This trend corresponds to a new way of thinking and performing surveying, aided by constant technological evolutions. Meanwhile laser scanning is a more and more appreciated solution for the fast acquisition of three dimensional geometry. In this context, the concept of scanning total station has been recently introduced to qualify instruments devised not only for scanning purposes, but also to perform conventional tacheometric tasks. The Trimble SX10 scanning total station belongs to this category and is intended for use in traditional surveying as well as in larger laser scanning projects. The absence of an optical telescope replaced by a powerful camera technology contributes to its originality. This daring decision to remove the telescope is made possible thanks to the use of a remote controller as well as of a set of cameras, and it clearly beneﬁts the measuring rate of the device. More generally, the versatility of this new generation of instruments probably represents its most valuable aspect since a wide variety of measurement tasks can be carried out with only one unique instrument. This has already been suggested in the past years such as in the conclusion of [7], where the authors were regretting the absence of facilities other than resection for the device location during georeferencing with scanning devices. Beyond competitiveness and veriﬁed high geometrical accuracy, the Trimble SX10 singular all-in-one design and its on-board camera technology among other features are signiﬁcant added values. However depending on the surveying project, a common total station may sometimes be sufﬁcient. Even if direct georeferencing is rather easy to perform with such an instrument, it may not be the most adapted solution in some particular scanning projects such as complex buildings for example. This is also true for common laser scanning projects, to which a total station survey may be preferred depending on the required deliverable. The high number of provided facilities should obviously not bypass a careful preparation and analysis of the necessary data by the user before acquisition. The absence of a telescope which may be confusing for uninitiated users also corresponds to a digital evolution that will probably impact most of the future device developments. 6. Discussion Regarding all the previously mentioned attributes, the Trimble SX10 deﬁnitely belongs to the next generation of equipment for land surveyors and people related to ﬁeld surveying in their speciﬁc research topics. 5.3. Uncertainty Analysis Related to Object Color and Materials Sensors 2017, 17, 730 20 of 23 20 of 23 5.3. Uncertainty Analysis Related to Object Color and Materials 19 of 23 Sensors 2017, 17, 730 (a) (b) (c) Figure 9. (a) Trimble SX10 colorized point cloud of a board composed of various color and material samples; (b) Visualization of intensity values returned by the laser scanner; (c) Cloud-to-cloud comparison between FARO Focus3D and Trimble SX10 point clouds of the same board. Deviations (unit: m) are projected on SX10 point cloud. (a) (a) (a) (c) (b) (b) (c) Figure 9. (a) Trimble SX10 colorized point cloud of a board composed of various color and material samples; (b) Visualization of intensity values returned by the laser scanner; (c) Cloud-to-cloud comparison between FARO Focus3D and Trimble SX10 point clouds of the same board. Deviations (unit: m) are projected on SX10 point cloud. It can be seen on this ﬁgure that the geometries of the point clouds are very similar since most of the deviations are smaller than 5 mm, which is within the order of magnitude of the alignment error. The main difference between both datasets is visible on the mirror, since no points were measured by the FARO Focus3D on such a highly reﬂective surface. Deviations can also be observed on the edges of the board which are covered with large black tape. This means that the ﬁltering of highly reﬂective (mirror) and very absorbing (black patterns) surfaces is probably managed differently in both devices. Deviations observed on the edges of wood samples for example, are mainly due to the alignment between both datasets and to the grazing angles of the laser beam on these parts. Regarding color variations, this has obviously no signiﬁcant inﬂuence on the range measurement. A small amount of measurement noise is only visible on black patterns, and this amount is not exactly the same between both datasets. Besides, intensity values provided by each laser scanner are very similar for the data acquired on the considered samples. One can assume that ﬁlters are applied on the observations to reduce the inﬂuence of object color on the measured data. All the previously assessed projects have shown a high similarity between Trimble SX10 data and datasets issued from other laser scanners. The results of these comparisons have proved that the device is efﬁcient and as reliable as other commonly used instruments, and this validates the announced performance. Some of the major characteristics of the Trimble SX10 scanning total station are summarized and discussed in the next section. 7. Conclusions The aim of this article was to give an overview of the major capabilities of the recent Trimble SX10 scanning total station. Based on a review of existing instruments providing similar functionalities, the innovative device was ﬁrst put in the current context of surveying instrument development. Some major technical properties of the Trimble SX10 were confronted to those of currently commercialized IATS as well as TLS. Based on this purely technical comparison, the device is competitive considering the multiple facilities it delivers in one unique piece of hardware. Advantages such as the possibility to document a project with pictures or to capture georeferenced point clouds and panoramas have been highlighted, as well as its originality which is based on the absence of a telescope. The quality of the produced data was then investigated through a set of acquisitions carried out under various conditions. Most of the Trimble SX10 capacities were tested, not only scanning performances but also surveying facilities. Their use has been reported through different studies presented in this article, such as building facade or bridge survey. It is important to note that the presented case studies are real projects for which the use of such an “all-in-one” instrument could bring an added value. To assess the quality of the produced data, they were confronted to results issued by similar instruments and methods. Scanning projects in particular were compared to datasets produced with other TLS. Direct georeferencing also belongs to the investigated capabilities through a civil engineering structural survey. A ﬁner assessment of the real precision achievable using the Trimble SX10 would require the performance of experiments on a speciﬁcally designed calibration benchmark. Nevertheless, the few case studies presented throughout the article correspond to typical applications which could beneﬁt from the facilities offered by this particular instrument. Based on these experiments, the results 21 of 23 Sensors 2017, 17, 730 obtained are valid compared to those delivered by the instruments used in our surveys. For these reasons one can conclude that the Trimble SX10 scanning total station meets the needs of users in terms of convenience and precision for common surveying tasks, depending on the expectations required by the project. Due to the advantages related to a powerful and competitive multi-sensor system, it also answers to further needs and enlarges the choice of users in the range of smart surveying instruments. References 1. Scherer, M. Advantages of the Integration of Image Processing and Direct Coordinate Measurement for Architectural Surveying—Development of the System TOTAL. In Proceedings of the XXII FIG International Congress, Washington, DC, USA, 19–26 April 2002. 1. Scherer, M. Advantages of the Integration of Image Processing and Direct Coordinate Measurement for Architectural Surveying—Development of the System TOTAL. In Proceedings of the XXII FIG International Congress, Washington, DC, USA, 19–26 April 2002. 2. Scherer, M.; Lerma, J.L. From the Conventional Total Station to the Prospective Image Assisted Photogrammetric Scanning Total Station: Comprehensive Review. J. Surv. Eng. 2009, 135, 173–178. 2. Scherer, M.; Lerma, J.L. From the Conventional Total Station to the Prospective Image Assisted Photogrammetric Scanning Total Station: Comprehensive Review. J. Surv. Eng. 2009, 135, 173–178. 3. Heritage, G.; Hetherington, D. Towards a protocol for laser scanning in ﬂuvial geomorphology. Earth Surface Process. Landf. 2007, 32, 66–74. 3. Heritage, G.; Hetherington, D. Towards a protocol for laser scanning in ﬂuvial geomorphology. Earth Surface Process. Landf. 2007, 32, 66–74. 4. Soni, A.; Robson, S.; Gleeson, B. Structural monitoring for the rail industry using conventional survey, laser scanning and photogrammetry. Appl. Geomat. 2015, 7, 123–138. 4. Soni, A.; Robson, S.; Gleeson, B. Structural monitoring for the rail industry using conventional survey, laser scanning and photogrammetry. Appl. Geomat. 2015, 7, 123–138. 5. Böhler, W. Comparison of 3D laser scanning and other 3D measurement techniques. In Recording, Modeling and Visualization of Cultural Heritage; Baltsavias, E., Grün, A., Van Gool, L., Pateraki, M., Eds.; Taylor & Francis Group: London, UK, 2006; pp. 89–99. 5. Böhler, W. Comparison of 3D laser scanning and other 3D measurement techniques. In Recording, Modeling and Visualization of Cultural Heritage; Baltsavias, E., Grün, A., Van Gool, L., Pateraki, M., Eds.; Taylor & Francis Group: London, UK, 2006; pp. 89–99. 6. Slob, S.; Hack, R. 3D terrestrial laser scanning as a new ﬁeld measurement and monitoring technique. In Engineering Geology for Infrastructure Planning in Europe; Hack, R., Azzam, R., Charlier, R., Eds.; Springer: Berlin/Heidelberg, Germany, 2004; Volume 104, pp. 179–189. 6. Slob, S.; Hack, R. 3D terrestrial laser scanning as a new ﬁeld measurement and monitoring technique. In Engineering Geology for Infrastructure Planning in Europe; Hack, R., Azzam, R., Charlier, R., Eds.; Springer: Berlin/Heidelberg, Germany, 2004; Volume 104, pp. 179–189. 7. Gordon, S.J.; Lichti, D.D. Terrestrial Laser Scanners with a Narrow Field of View: The Effect on 3D Resection Solutions. Surv. Rev. 7. Conclusions Acknowledgments: The authors would like to thank the Trimble company for lending the SX10 device, as well as Volker Köhler from the company for its support during the test phase. Author Contributions: Experiments were designed and performed by Elise Lachat with the help of Tania Landes and Pierre Grussenmeyer. All the authors contributed to the writing of the paper. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations The following abbreviations are used in this manuscript: TLS Terrestrial Laser Scanner GNSS Global Navigation Satellite System EDM Electronic Distance Measurement IATS Image-Assisted Total Station COGO Coordinate Geometry TBC Trimble Business Center ICP Iterative Closest Point TLS Terrestrial Laser Scanner GNSS Global Navigation Satellite System EDM Electronic Distance Measurement IATS Image-Assisted Total Station COGO Coordinate Geometry TBC Trimble Business Center ICP Iterative Closest Point References 2004, 37, 448–468. 8. Soudarissanane, S.; Lindenbergh, R.; Menenti, M.; Teunissen, P. Scanning geometry: Inﬂuencing factor on the quality of terrestrial laser scanning points. ISPRS J. Photogramm. Remote Sens. 2011, 66, 389–399. 9. Trimble SX10 Scanning Total Station Information and Datasheet. Available online: http://infogeospatial. trimble.com/SX10.html (accessed on 10 March 2017). 10. Lemmens, M. Total Stations: The Surveyor’s Workhorse. Available online: https://www.gim-international. com/content/article/total-stations-the-surveyor-s-workhorse (accessed on 10 March 2017). 11. Ferreira, R.; Santos, V. State-of-the-Art Total Stations. Available online: https://www.gim-international. com/content/article/state-of-the-art-total-stations (accessed on 10 March 2017). 12. Walser, B.H. Development and Calibration of an Image Assisted Total Station. Ph.D. Thesis, Swiss Federal Institute of Technology (ETH Zurich), Zurich, Switzerland, 2004. Institute of Technology (ETH Zurich), Zurich, Switzerland, 2004. 13. Wagner, A. A new approach for geo-monitoring using modern total stations and RGB + D images. Meas. J. Institute of Technology (ETH Zurich), Zurich, Switzerland, 2004. 13. Wagner, A. A new approach for geo-monitoring using modern total stations and RGB + D images. Meas. J. Int Meas Confed 2016 82 64 74 13. Wagner, A. A new approach for geo-monitoring using modern total stations and RGB + D images. Meas. J. Int. Meas. Confed. 2016, 82, 64–74. 22 of 23 Sensors 2017, 17, 730 14. Hauth, S.; Schlüter, M.; Thiery, F. Modular Imaging Total Stations—Sensor Fusion for high precision alignment. In Proceedings of the 3rd International Conference on Machine Control & Guidance, Stuttgart, Germany, 27–29 March 2012; pp. 202–210. 5. Juretzko, M. Reﬂektorlose Video-Tachymetrie—Ein Integrales Verfahren Zur Erfassung Geometrischer U Visueller Informationen. Ph.D. Thesis, Ruhr-Universität Bochum, Bochum, Germany, 2004. 16. Wasmeier, P. Grundlagen der Deformationsbestimmung Mit Messdaten Bildgebender Tachymeter. Ph.D. Thesis, Technische Universität München, Munich, Germany, 2009. 17. Zhou, Y.; Wagner, A.; Wunderlich, T.; Wasmeier, P. Calibration Method for IATS and Application in Multi-Target Monitoring using Coded Targets. In Proceedings of the 3rd Joint International Symposium on Deformation Monitoring, Vienna, Austria, 30 March–1 April 2016. 18. Ehrhart, M.; Lienhart, W. Accurate Measurements with Image-Assisted Total Stations and Their Prerequisites. J. Surv. Eng. 2016, doi:10.1061/(ASCE)SU.1943-5428.0000208. 19. Topcon IS-3 Series Information and Datasheet. Available online: https://www.topconpositioning.com/ total-stations/robotic-total-stations/3-series (accessed on 10 March 2017). 0. Leica Nova MS60 Information and Datasheet. Available online: http://www.leica-geosystems.us/ Leica-Nova-MS60_106652.html (accessed on 10 March 2017). 1. Ehrhart, M.; Lienhart, W. Monitoring of Civil Engineering Structures using a State-of-the-art Image Assi Total Station. J. Appl. Geod. 2015, 9, 174–182. 22. Lackner, S.; Lienhart, W.; Supp, G.; Marte, R. References Geodetic and ﬁbre optic measurements of a full-scale bi-axial compressional test. Surv. Rev. 2016, 48, 86–93. 23. Zheng, S.; Ma, D.; Zhang, Z.; Hu, H.; Gui, L. A novel measurement method based on silhouette for chimney quasi-static deformation monitoring. Measurement 2012, 45, 226–234. 24. Pagliari, D.; Rossi, L.; Passoni, D.; Pinto, L.; De Michele, C.; Avanzi, F. Measuring the volume of ﬂushed sediments in a reservoir using multi-temporal images acquired with UAS. Geomat. Nat. Hazards Risk 2016, doi:10.1080/19475705.2016.1188423. 25. Thuro, K.; Singer, J.; Festl, J.; Wunderlich, T.; Wasmeier, P.; Reith, C.; Heunecke, O.; Glabsch, J.; Schuhbäck, S. New landslide monitoring techniques—Developments and experiences of the alpEWAS project. J. Appl. Geod. 2010, 4, 69–90. 26. Gómez-Vasconcelos, M.; Villamor, P.; Cronin, S.; Procter, J.; Kereszturi, G.; Palmer, A.; Townsend, D.; Leonard, G.; Berryman, K.R.; Ashraf, S. Earthquake history at the eastern boundary of the South Taupo Volcanic Zone, New Zealand. N. Z. J. Geol. Geophys. 2016, 59, 522–543. 27. Lee, H.S.; Lim, S.S.; Park, D.W. Application of Terrestrial Laser Scanner and Raster operations to Change Detection of Beach. J. Coast. Res. 2011, 64, 1692–1696. 28. Wo´zniak, M.; ´Swierczy´nska, E.; Jastrze¸bski, S. The Use of Video-Tacheometric Technology for Documenting and Analysing Geometric Features of Objects. Rep. Geod. Geoinform. 2015, 99, 28–43. 29. Sepasgozar, S.M.E.; Lim, S.; Shirowzhan, S.; Kim, Y.M. Implementation of As-Built Information Modelling Using Mobile and Terrestrial Lidar Systems. In Proceedings of the 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC), Sydney, Australia, 9–11 July 2014. 30. Hofmann, S.; Brenner, C. Accuracy Assessment of Mobile Mapping Point Clouds Using the Existing Environment As Terrestrial Reference. In Proceedings of the XXIII ISPRS Congress, Prague, Czech Republic, 12–19 July 2016; Volume XLI-B1, pp. 601–608. 31. Evgenikou, V.; Georgopoulos, A. Investigating 3D reconstruction methods for small artifacts. In Proceedings of the ISPRS International Workshop 3D-Arch 2015, Avila, Spain, 25–27 February 2015; Volume XL-5/W4, pp. 101–108. 32. Caprioli, M.; Scognamiglio, A. Low Cost Methodology for 3D Modeling and Metric Description in Architectural Heritage. In Proceedings of the ISPRS International Workshop 3D-ARCH 2009, Trento, Italy, 25–28 February 2009; Volume XXXVIII-5. 33. Wagner, A.; Wiedemann, W.; Wunderlich, T. Using IATS to Read and Analyze Digital Leveling Staffs. In Proceedings of the SIG 2016 “International Symposium on Engineering Geodesy”, Varazdin, Croatia, 20–22 May 2016; pp. 515–526. 34. FARO Focus3D Information and Datasheet. Available online: http://www.faro.com/products/ 3d-surveying/laser-scanner-faro-focus-3d/overview (accessed on 10 March 2017). © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References 23 of 23 Sensors 2017, 17, 730 35. Leica ScanStation C10 Information and Datasheet. Available online: http://leica-geosystems.com/products/ laser-scanners/scanners/leica-scanstation-c10 (accessed on 10 March 2017). 35. Leica ScanStation C10 Information and Datasheet. Available online: http://leica-geosystems.com/products/ laser-scanners/scanners/leica-scanstation-c10 (accessed on 10 March 2017). 36. CloudCompare Free Software. Available online: http://www.danielgm.net/cc/ (accessed on 10 M 36. CloudCompare Free Software. Available online: http://www.danielgm.net/cc/ (accessed on 10 March 2017). 37. Grussenmeyer, P.; Landes, T.; Doneus, M.; Lerma, J.L. Basics of range-based modelling techniques in 36. CloudCompare Free Software. Available online: http://www.danielgm.net/cc/ (accessed on 10 March 2017). 37. Grussenmeyer, P.; Landes, T.; Doneus, M.; Lerma, J.L. Basics of range-based modelling techniques in Cultural Heritage. In 3D Recording, Documentation and Management of Cultural Heritage; Stylianidis, E., Remondino F Eds ; Whittles Publishing: Dunbeath UK 2016; pp 305 368 37. Grussenmeyer, P.; Landes, T.; Doneus, M.; Lerma, J.L. Basics of range-based modelling techniques in Cultural Heritage. In 3D Recording, Documentation and Management of Cultural Heritage; Stylianidis, E., Remondino, F., Eds.; Whittles Publishing: Dunbeath, UK, 2016; pp. 305–368. 38. Lachat, E.; Landes, T.; Grussenmeyer, P. First experiences with the Trimble SX10 Scanning Total Station for building facade survey. In Proceedings of the ISPRS International Workshop 3D-Arch 2017, Nafplio, Greece, 1–3 March 2017; Volume XLII-2/W3, pp. 405–412. 39. Kaasalainen, S.; Krooks, A.; Kukko, A.; Kaartinen, H. Radiometric Calibration of Terrestrial Laser Scanners with External Reference Targets. Remote Sens. 2009, 1, 144–158. 40. Lichti, D.D.; Harvey, B. The Effects of Reﬂecting Surface Material Properties on Time-of-Flight Laser Scanner Measurements. In Proceedings of the ISPRS Commission IV Symposium “Geospatial Theory, Processing and Applications”, Ottawa, ON, Canada, 9–12 July 2002. 41. Voegtle, T.; Schwab, I.; Landes, T. Inﬂuences of different materials on the measurements of a terrestrial laser scanner (TLS). In Proceedings of the XXI ISPRS Congress, Beijing, China, 3–11 July 2008; Volume XXXVII-B5, pp. 1061–1066. © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4244553621	https://www.qeios.com/read/PBR0CY/pdf	English	null	PNGVL3-hICD Vaccine	Definitions	2,020	cc-by	126	Qeios · Definition, February 7, 2020 Open Peer Review on Qeios Open Peer Review on Qeios PNGVL3-hICD Vaccine National Cancer Institute National Cancer Institute Source National Cancer Institute. pNGVL3-hICD Vaccine. NCI Thesaurus. Code C61146. Qeios ID: PBR0CY · https://doi.org/10.32388/PBR0CY National Cancer Institute. pNGVL3-hICD Vaccine. NCI Thesaurus. Code C61146. A plasmid DNA cancer vaccine encoding the intracellular domain (ICD) of the HER-2/neu proto-oncogene. Upon administration and after cellular uptake by skin or muscle cells, the pNGVL3-hICD vaccine plasmid expresses the HER-2/neu protein, which, after intracellular processing, may elicit both antigen-specific cytotoxic T-lymphocyte (CTL) and humoral immune responses against tumor cells expressing HER-2. The HER-2/neu ICD protein is highly immunogenic and, as a subdominant epitope, may be associated with decreased immune tolerance. Qeios ID: PBR0CY · https://doi.org/10.32388/PBR0CY 1/1
https://openalex.org/W2129004487	https://europepmc.org/articles/pmc2831024?pdf=render	English	null	Standardized observation of neighbourhood disorder: does it work in Canada?	International journal of health geographics	2,010	cc-by	13,529	© 2010 Parsons et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Background: There is a growing body of evidence that where you live is important to your health. Despite numerous previous studies investigating the relationship between neighbourhood deprivation (and structure) and residents' health, the precise nature of this relationship remains unclear. Relatively few investigations have relied on direct observation of neighbourhoods, while those that have were developed primarily in US settings. Evaluation of the transferability of such tools to other contexts is an important first step before applying such instruments to the investigation of health and well-being. This study evaluated the performance of a systematic social observational (SSO) tool (adapted from previous studies of American and British neighbourhoods) in a Canadian urban context. Methods: This was a mixed-methods study. Quantitative SSO ratings and qualitative descriptions of 176 block faces were obtained in six Toronto neighbourhoods (4 low-income, and 2 middle/high-income) by trained raters. Exploratory factor analysis was conducted with the quantitative SSO ratings. Content analysis consisted of independent coding of qualitative data by three members of the research team to yield common themes and categories. Results: Factor analysis identified three factors (physical decay/disorder, social accessibility, recreational opportunities), but only 'physical decay/disorder' reflected previous findings in the literature. Qualitative results (based on raters' fieldwork experiences) revealed the tool's shortcomings in capturing important features of the neighbourhoods under study, and informed interpretation of the quantitative findings. Conclusions: This study tested the performance of an SSO tool in a Canadian context, which is an important initial step before applying it to the study of health and disease. The tool demonstrated important shortcomings when applied to six diverse Toronto neighbourhoods. The study's analyses challenge previously held assumptions (e.g. social 'disorder') regarding neighbourhood social and built environments. For example, neighbourhood 'order' has traditionally been assumed to be synonymous with a certain degree of homogeneity, however the neighbourhoods under study were characterized by high degrees of heterogeneity and low levels of disorder. Heterogeneity was seen as an appealing feature of a block face. Employing qualitative techniques with SSO represents a unique contribution, enhancing both our understanding of the quantitative ratings obtained and of neighbourhood characteristics that are not currently captured by such instruments. hood conditions and residents' health (and the mediator and moderator factors at play) remains unclear. Open Access Open Access Research Standardized observation of neighbourhood disorder: does it work in Canada? Research Standardized observation of neighbourhood disorder: does it work in Canada? Janet A Parsons1,2, Gita Singh3, Allison N Scott4, Rosane Nisenbaum4,5, Priya Balasubramaniam8, Amina Jabbar4, Qamar Zaidi4, Amanda Sheppard6,7, Jason Ramsay9, Patricia O'Campo4,5 and James Dunn4,5 Correspondence: ParsonsJ@smh.toronto.on.ca 1 Applied Health Research Centre, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 INTERNATIONAL JOURNAL OF HEALTH GEOGRAPHICS Background There is a growing body of evidence that where you live is important to your health [1-4]. Environmental factors (e.g. air quality, proximity to industrial pollutants), neigh- bourhood income, and neighbourhood structure have all been linked to a variety of health outcomes [5-7]. How- ever the nature of the relationship between neighbour- Characterizing neighbourhoods and the specific fea- tures of neighbourhoods that may contribute to residents' health and well being is a complex and difficult undertak- ing. Considerable attention has been devoted to the role of neighbourhood poverty (and its potential health effects), in part secondary to trends in recent decades where poverty has become more spatially concentrated in inner-city neighbourhoods, at least in the United States * Correspondence: ParsonsJ@smh.toronto.on.ca 1 Applied Health Research Centre, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 2 of 19 [6,8]. Investigators have linked neighbourhood income to individual-level health outcomes, using either adminis- trative datasets or individual-level survey responses [1,7]. Most studies using objective data sources have relied on census data, which provide information about socioeco- nomic position of a given census tract relative to others (e.g. median household income), population stability, and ethnic composition of a neighourhood [9]. There are a number of problems with using such census data, one of which is that they provide neither information regarding the social life of the neighbourhood nor the physical characteristics of the built environment [3,9]. these measures have focused on observational data at the level of randomly selected block faces within specific cen- sus tracts [7,14]. Such observational tools are essentially checklists used to inventory and rate aspects of the social and physical environments of each block face. Data are typically aggregated up to whatever definition of neigh- bourhood is being used, and associations with various health outcomes investigated. To date, these scales have never been applied in a Cana- dian context to a range of urban block faces within a sin- gle city. While Kohen and colleagues (2002) applied a scale ostensibly capturing both physical and social disor- der in a large Canadian sample (n = 3,350), the authors only looked at a very limited range of physical (traffic vol- ume, presence of garbage, building conditions) and social attributes (loitering, hostile behaviour, drunkenness/ intoxication) [19]. Background The focus was at the level of individual families (not at the city or neighbourhood level) and only a very limited portion of the block face surrounding the participants' home (same side of street) was examined. It did not investigate neighbourhoods or their characteris- tics in depth. A study by Coen and Ross (2006) has applied such scales specifically to the study of Montreal neighbourhood parks, but did not look beyond these par- ticular features [20]. Relatively few studies have employed direct observation of neighbourhood environments. Of those that have, most have concentrated on both physical (geographical) features of urban environments and/or on social features to which residents are exposed [10]. While social envi- ronmental influences on health have been documented [1,3,4,11], the relationship between neighbourhood dis- advantage, physical disorder, and social disorder/social cohesion remains unclear. Much of this research is grounded in the field of criminology [12-14] and has then been adopted as a starting point for health research [15,16]. One influential model (Skogan, 1990)[17] sug- gests that the more prevalent physical incivilities/disor- der become, residents' perceptions of their neighbourhood shift, leading to decreased social cohe- sion (and increasing crime) ([17] cited in [12]; [14]). Markowitz and colleagues (2001) posit a feedback loop whereby the effect of disorder on neighbourhood cohe- sion is mediated by fear [12]. The concept of 'social capi- tal' is another theoretical construct underlying much of the neighbourhoods-and-crime literature, and is defined by Sampson and colleagues (1997) as 'collective efficacy' (premised on "'mutual trust and a shared willingness to intervene for the common good' of the community") (Sampson et al., 1997 in Franzini et al., 2005)[18](p. 1136). Such collective efficacy is posited to be comprised of two processes: "social cohesion (the sense of connect- edness) and informal social control (the willingness to intervene in community problems)" [18]. However the link between social disorder and social cohesion (and how this relates to health) has yet to be definitively dem- onstrated. Nevertheless it is important to understand the theoretical assumptions underlying much of the literature on systematic social observations (SSO). There are several reasons for questioning the applica- bility and appropriateness of such instruments for study- ing public health in urban contexts outside the United States. There has been relatively little critical scrutiny of the theoretical assumptions linking physical and social disorder with crime, and their subsequent extension to the arena of public health. Background Few have questioned the appropriateness of this extrapolation. For example, some investigators drawing on these assumptions have hypoth- esized that evidence of territoriality and defensible space could represent physical manifestations of social cohe- sion, suggesting that if a criminal offender were to cross territorial boundaries that residents would take defensive action [14,21]. This appears to us a potentially problem- atic assumption, and that gardens, shrubs and low rail- ings may not be evidence of such defensive thinking, but may have quite different meanings for residents. Caughy et al. (2001) problematize the assumption that the pres- ence of physical incivilities (presence of trash or graffiti) reflects an 'uncaring' attitude by residents [22], pointing out these may be the combined result of municipal prior- ity setting and resident will [21]. Furthermore, research by Oreopoulos (2005) highlights some important differ- ences between the five largest Canadian cities and their five largest counterparts in the US [23]. Oreopoulos notes that low-income census tracts in Canada are not charac- terized by segregation of visible minorities (as they are in the US), and Canadian residents of low-income census tracts are exposed to much lower rates of crime than their What then does 'neighbourhood disorder' look like? Prior research in the US and the UK has resulted in the development of standardized observational tools in an attempt to quantify features indicative of disorder in both the immediate social and physical environments. These standardized instruments have further informed the development of measurement scales with which to com- pare neighbourhoods. To date, most studies employing Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 3 of 19 Page 3 of 19 nized the subjective nature of observation (even when undertaken systematically) and we drew upon this rich tradition in the social sciences to inform our study design [26]. It is increasingly recognized that innovative mixed- method approaches are important, because our under- standing of quantitative evidence (the measured relation- ship between variables) can be enhanced by qualitative evidence to understand the 'how' and 'why' of (and pro- cesses underlying) those relationships [27]. Cuthchin (2007) and others emphasize the importance of recogniz- ing culture and its implications for geographic spaces/ built environments, and their meanings to observers [5,28,29]. Data collection procedures This mixed-methods study entailed compiling an obser- vational tool comprised of the union set of items from four widely used instruments (described below) and test- ing its performance in a variety of Toronto neighbour- hoods. A consensus process with community partners defined the following Toronto neighbourhoods to be sur- veyed: Eglinton East, South Parkdale, St. Jamestown, and Weston (all considered low-income) and Banbury-Don Mills and North Riverdale (both middle/high income). All four low-income neighbourhoods are characterized by higher incidence of low-income households (e.g. St. Jamestown 47.9%, South Parkdale 46.4%), compared with the mid/high-income neighbourhoods (Banbury-Don Mills 14.3%, North Riverdale 15.6%) (City of Toronto sta- tistics) [30]. The neighbourhoods were defined by exist- ing City of Toronto criteria (which are numerous) and are further defined by Statistics Canada census tracts (for purposes of statistical reporting) [30]. No Toronto neigh- bourhood is comprised of a single census tract, and each has a minimum population of between 7,000 and 10,000 people. Of the six study neighbourhoods, five had popu- lations > 16,000 (two with populations over 22,000), and one (North Riverdale) had approximately 11,000 per- sons[30]. Toronto is one of the most ethnically diverse cities in the world and the most diverse nationally; most of its neighbourhoods have significant proportions of new immigrant and visible minority populations. That being said, the four low-income neighbourhoods are con- sidered some of the most ethnically diverse with some of the largest populations of new immigrants in the city [30]. We did not attempt to link the observational data to health outcomes per se in our study, but rather were con- cerned with testing the elements of these various instru- ments to determine if they were sensitive to detecting physical and social disorder in a Canadian context. Such validation studies are a necessary first step in this area of research and are of interest to health researchers, aiding them in the interpretation of findings acquired using these instruments. While we originally intended to apply quantitative methods in isolation, it became apparent at the outset of data collection that qualitative techniques would provide important insights regarding the tool's performance. As the research team comprised members with expertise in both methodological approaches, a qualitative component of the study was incorporated into the study design. While this has not been done in prior SSO studies, the rationale for including qualitative approaches is sound. Background To our knowledge the inclusion of qualitative methods represents a unique contribution to the field of SSO. US counterparts [23]. He indicates that the poverty lines in the two nations are not directly comparable, and that it should not be assumed that the experiences of low- income households in the lowest income census tracts are similar in both countries [23]. Thus it may be that apply- ing such observational tools in urban settings outside the US is of questionable utility, given the stark differences in the social geography of US cities when compared to those of other nations. In this paper we applied observational tools measuring physical and social disorder (developed for use in US and UK cities) to a specific Canadian urban context. This study is one in a series of investigations regarding neigh- bourhood-level influences on health being conducted in Toronto, Canada's largest city. Prior to examining link- ages with residents' health, it was first necessary to deter- mine if previously-developed observational tools were transferable to the Toronto context. Drawing on prior work by Raudenbush & Sampson (1999), Caughy and col- leagues (2001) [18], Weich et al. (2001) and others, [7,9,18,21,24,25] we began from an assumption that "neighbourhood impoverishment" is a "source of neigh- bourhood social and physical characteristics"[18]. Raudenbush & Sampson (1999) developed a reliable cod- ing method for neighbourhood block faces (using video- taped data), premised on the notion that structural characteristics can influence neighbourhood social orga- nization and collective efficacy [8]. Weich et al. (2001) validated a survey checklist of built environment features for use in UK cities (specifically, structural and building features; use of green, public and vacant spaces; security/ safety; and accessibility to amenities). The items chosen for inclusion in our study incorporated those which had previously demonstrated reliability and validity, and whether they made conceptual sense (based on knowl- edge of the existing literature). Data collection procedures The qualitative find- ings inform the results of the quantitative analysis. SSO tool We reviewed the literature to identify constructs, scales, or items related to SSO of neighbourhoods previously employed in cities in the US and UK [7,21,25]. We cre- ated a pilot observational tool that included 98 items from four sources: (1) Raudenbush & Sampson (1999) who reported on items related to the constructs Social Disorder, Physical Disorder and Physical Decay [7]; (2) Caughy et al. (2001) who examined items related to the constructs Physical Incivilities, Territoriality and Avail- able Play Resources [21]; (3) Weich et al. (2001) who investigated the association of individual 'Built Environ- ment' items with depression [25]; and (4) Sastry & Pebley (2003) who employed a public-access neighbourhood survey instrument [24]. The domains of the resulting composite tool included: (1) Characteristics of streets and Block face diagram Figure 1 Block face diagram. Block face diagram Figure 1 Block face diagram. g Figure 1 Block face diagram. grid. The total number of concordant answers was calcu- lated for each BF, and the proportion of correct answers (or percent agreement) was calculated by dividing it by 56. We further documented the data collection processes using qualitative methods, encouraging raters to reflect on their fieldwork experiences and the perceived utility of the tool in situ. In this report, we examine the properties of the data collected using this mixed-method design; qualitative techniques, descriptive statistics and explor- atory factor analysis are employed. The qualitative find- ings inform the results of the quantitative analysis. tures, varied residential layouts (grid versus curvilinear streets), mixture of high-rise and low-rise buildings, var- ied population densities, road zoning, etc.). A block face (BF) was defined as "any street between two intersections" and included both sides of the street (Fig. 1); an intersection was characterized by the presence of street signs, breaks in sidewalks, or a dead end. Due to budget and time constraints, the study was designed to collect data on a maximum of 180 BFs. It was decided that to better allocate resources, the 180 BFs would be distributed in a ratio of 2:1 for low to middle/high income neighbourhoods (120:60). Within each income level, we selected BFs with probability proportional to total num- ber of BFs (497 and 455 respectively for low to middle/ high income neighbourhoods). Data collection procedures Within each neighbour- hood, BFs were listed in a quasi-geographic order, by looking at the neighbourhood street map and recording the streets between two intersections. Systematic random sampling with probability proportional to neighbourhood size, with the appropriate sampling rates, was applied. This meant that for the low-income neighbourhoods, the sampling rate was 120/497 (or 24.15%), and for middle/ high income neighbourhoods was 60/455 (or 13.19%). The final sample included 176 BFs. Data collection procedures Anthropologists have long recog- Data collection took place in the six neighbourhoods, in order to determine whether the composite SSO instru- ment could detect differences between neighbourhoods of disparate median incomes. These neighbourhoods are also characterized by diverse built forms (i.e. character- ized by a mixture of residential and commercial struc- Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 4 of 19 Block face diagram Figure 1 Block face diagram. tures, varied residential layouts (grid versus curvilinear streets), mixture of high-rise and low-rise buildings, var- ied population densities, road zoning, etc.). A block face (BF) was defined as "any street between two intersections" and included both sides of the street (Fig. 1); an intersection was characterized by the presence of street signs, breaks in sidewalks, or a dead end. Due to budget and time constraints, the study was designed to collect data on a maximum of 180 BFs. It was decided that to better allocate resources, the 180 BFs would be distributed in a ratio of 2:1 for low to middle/high income neighbourhoods (120:60). Within each income level, we selected BFs with probability proportional to total num- ber of BFs (497 and 455 respectively for low to middle/ high income neighbourhoods). Within each neighbour- hood, BFs were listed in a quasi-geographic order, by looking at the neighbourhood street map and recording the streets between two intersections. Systematic random sampling with probability proportional to neighbourhood size, with the appropriate sampling rates, was applied. This meant that for the low-income neighbourhoods, the sampling rate was 120/497 (or 24.15%), and for middle/ high income neighbourhoods was 60/455 (or 13.19%). The final sample included 176 BFs. Data collection occurred at two different times of day (daylight and evening hours) to account for temporal dif- ferences in certain variable items (described below). We also conducted repeated observations on a sub sample of grid. The total number of concordant answers was calcu- lated for each BF, and the proportion of correct answers (or percent agreement) was calculated by dividing it by 56. We further documented the data collection processes using qualitative methods, encouraging raters to reflect on their fieldwork experiences and the perceived utility of the tool in situ. In this report, we examine the properties of the data collected using this mixed-method design; qualitative techniques, descriptive statistics and explor- atory factor analysis are employed. SSO tool We reviewed the literature to identify constructs, scales, or items related to SSO of neighbourhoods previously employed in cities in the US and UK [7,21,25]. We cre- ated a pilot observational tool that included 98 items from four sources: (1) Raudenbush & Sampson (1999) who reported on items related to the constructs Social Disorder, Physical Disorder and Physical Decay [7]; (2) Caughy et al. (2001) who examined items related to the constructs Physical Incivilities, Territoriality and Avail- able Play Resources [21]; (3) Weich et al. (2001) who investigated the association of individual 'Built Environ- ment' items with depression [25]; and (4) Sastry & Pebley (2003) who employed a public-access neighbourhood survey instrument [24]. The domains of the resulting composite tool included: (1) Characteristics of streets and sidewalks, (2) Residential land use, (3) Non-residential land use (commercial, industrial or vacant), (4) Parks and playgrounds, (5) Commercial establishments, services, and institutions, (6) People, and (7) Incivilities (copies of Data collection occurred at two different times of day (daylight and evening hours) to account for temporal dif- ferences in certain variable items (described below). We also conducted repeated observations on a sub-sample of BFs (n = 16 or 9.1% of total BFs) to ensure inter-rater reli- ability. The answer to each of 56 SSO fixed items was compared between two 'gold standard' trainers and rat- ers, and concordance was indicated by a check in a paper Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 5 of 19 the pilot SSO instrument are available upon request from the authors). cussion. Based on the emergent themes and categories, sub-categories were created. In addition to the written comments analysis, a focus group discussion was conducted with all raters once data collection had been completed, in order to allow for fur- ther elaboration on their field-work experiences. In addi- tion, one of the expert trainers and one of the senior research team members attended the focus group. The focus group (facilitated by a member of the research team with expertise in qualitative research methods - JAP) was audio-taped and transcribed verbatim, providing an addi- tional source of qualitative data for analysis [31]. SSO tool Topics for discussion included: perceived positive features of the instrument, perceived shortcomings, suggestions for improving the instrument, reflections on fieldwork expe- riences (including surprises, expectations, the role of social interactions with residents), and perceptions regarding what constitutes an 'appealing' BF. These data served to clarify and expand on some of the themes that emerged during analysis of the written comments. All observations were completed between August and October 2006. The 98-item checklist was completed by raters between the hours of 09:00 and 15:30. Because some of the SSO items could potentially vary with time of day (e.g. presence of police officer on the BF, number of visible children) a subset of 41 'variable' items was identi- fied. Raters returned to each BF to evaluate these variable items between 17:00 and 20:00 hrs. We consolidated the repeated information of variable items by indicating pres- ence (or attribute) observed either during the first or sec- ond observation. In the case of ordinal items, we selected the highest category observed (e.g., some garbage was selected over little garbage). All raters underwent a week-long intensive training program conducted by two project coordinators experi- enced with SSO. Training consisted of in-depth item review via a standardized protocol, training slideshow presentation, and informal classroom exercises. In addi- tion, formal field exercises were conducted in two neigh- bourhoods which simulated the experience of performing SSOs (similar to the methodology described by Caughy) [21]. Competency of raters was determined by the achievement of a minimum score on the 98-item compre- hensive checklist. Per cent agreement was evaluated for 16 BFs (or 9.1% of the total sample of BFs) and confirmed a high level of agreement between raters and the most experienced trainer (median agreement = 89.3%, SD ± 15.6%). Exploratory Factor Analyses Our analyses started by identifying items linked to a pub- lished construct or potentially linked to the construct based on its face validity. The process of organizing the items led to the creation of 3 major meta-categories of neighbourhood items: 1. Physicality; 2. Social; and 3. Resources. Built environment items were used only for descriptive purposes. Qualitative Methods International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 6 of 19 don't see play equipment or any sign denoting that it's a park." (Eglinton East). iate normal distribution for each pair of binary items. For example, presence of 'some/a lot of garbage' (observed) is indicated only if the amount of garbage (latent) is greater than a certain threshold (assumed to be consistently eval- uated by all raters because of standardized training). Dur- ing the exploratory factor analysis, items were further excluded because of: (1) high correlation with another item; (2) failure to load in any factor; (3) factor loading <0.40; and (4) loading in multiple factors. The adequacy of the number of factors was evaluated using the chi- square test for overall model fit (p-value > 0.05) and the root mean square residual (RMSR < 0.05). Binary items were compared across neighbourhoods using Fisher's exact tests; analysis of variance and the Wilcoxon-rank sum test assessed differences with respect to factor scores. don't see play equipment or any sign denoting that it's a park." (Eglinton East). iate normal distribution for each pair of binary items. For example, presence of 'some/a lot of garbage' (observed) is indicated only if the amount of garbage (latent) is greater than a certain threshold (assumed to be consistently eval- uated by all raters because of standardized training). Dur- ing the exploratory factor analysis, items were further excluded because of: (1) high correlation with another item; (2) failure to load in any factor; (3) factor loading <0.40; and (4) loading in multiple factors. The adequacy of the number of factors was evaluated using the chi- square test for overall model fit (p-value > 0.05) and the root mean square residual (RMSR < 0.05). Binary items were compared across neighbourhoods using Fisher's exact tests; analysis of variance and the Wilcoxon-rank sum test assessed differences with respect to factor scores. don't see play equipment or any sign denoting that it's a park." (Eglinton East). It would have been inappropriate to indicate that the area was a park, yet the response category of 'vacant/ undeveloped lots' did not seem to capture the land use adequately in this instance. Features of the BF relative to the surrounding neighbourhood Of the 176 BFs evaluated, raters documented comments corresponding to 112 (64%) of the systematic social observations. The qualitative analyses revealed that rat- ers' comments related to three broad levels of interpreta- tion. These comments described: (1) features of the BF being observed, (2) features of the BF relative to the sur- rounding neighbourhood, and (3) features of the rater's experience while observing the BF. The qualitative sub- committee revisited the 21 emergent codes and the raw data to determine if it was more useful to fit the themes to the meta-categories derived from the quantitative data. However, all analysts agreed that the qualitative data were much better suited to the three specific levels of interpre- tation outlined above. As discussed previously, a random sample of BFs was examined within each neighbourhood, and raters often expressed in their comments how the BF under observa- tion did or did not appear to correspond with the sur- rounding area. A couple of examples of such data follow: "Very messy block face compared to surrounding neighbourhood" (South Parkdale) As discussed previously, a random sample of BFs was examined within each neighbourhood, and raters often expressed in their comments how the BF under observa- tion did or did not appear to correspond with the sur- rounding area. A couple of examples of such data follow: "Very messy block face compared to surrounding neighbourhood" (South Parkdale) "Very messy block face compared to surrounding neighbourhood" (South Parkdale) "Sense of neighbourhood pride, heritage sign outside on 2 houses and child made artwork outside another with label 'beautiful Weston' " (Weston) These comments provide important contextual infor- mation, situating the BF observed within the surrounding neighbourhood. Raters' comments indicated that BFs within all neighbourhoods evaluated were characterized by considerable heterogeneity (with the sole exception of Banbury-Don Mills). Furthermore, the comments suggest that this variability or heterogeneity could be seen as pos- itive or negative depending on the particular BF charac- teristic or feature under observation. This finding was reinforced during the group discussion as raters unani- mously agreed that diversity of land use within BFs was viewed as an attribute of an appealing BF. Moreover, rat- ers viewed heterogeneity (in the context of diversity) as a positive attribute of both BFs and neighbourhoods. Qualitative Methods In previous studies using sim- ilar SSO instruments, vacant lots were often presumed to be indicators of physical disorder [7]. In this case, how- ever -- given the evidence of a community garden -- the comment suggests that this land use cannot be easily cat- egorized as either vacant or undeveloped as suggested by the instrument. The community garden not only implies a certain level of physical order in this undeveloped lot, but also hints at a level of social cohesion by the residents' use of common lands. Moreover, this example represents numerous instances within the qualitative data that speak to the challenges and difficulties of capturing neighbour- hood characteristics according to the mutually exclusive categories dictated by the SSO instrument. At the same time, the additional contextual information provided by the qualitative data demonstrates the additional strength and depth provided by the use of mixed-methods. Factor analyses were performed using Mplus version 3.1 (Muthen & Muthen, Los Angeles, CA) software; all other analyses were performed using SAS version 9.1 (SAS Institute Inc., Cary, NC). Features of the BF relative to the surrounding neighbourhood This was especially the case with respect to aesthetic and natu- ral aspects of the physical environment, diversity of archi- tectural styles, and multicultural resident profiles. These findings were somewhat surprising and stand in contrast with previous literature in the area which has emphasized Qualitative Methods A large proportion of items in the observational tool were on a binary scale (Yes/No) and many ordinal-scaled items presented skewed distributions suggesting two or at most three possible values. Therefore, for the purposes of factor analyses, these items were dichotomized. We created composite indicators to include items with very low prevalence (<5% or 9 of 176 block faces). Items that were too rare or could not be included in composite indi- cators were excluded from the analyses. Items that could vary over the day were observed twice. These were called 'variable items'. We consolidated information of variable items observed on two occasions by indicating presence (or the attribute) observed either in the first or second survey. In each case of an ordinal item, we selected the highest category observed (e.g. 'some garbage ' was selected over 'little garbage '). For the qualitative component of the study, an emergent design was adopted based on the initial implementation of the SSO tool in the neighbourhoods. Raters com- mented at team meetings (early on in the data collection process) that the tool failed to capture certain aspects of the BFs being evaluated - for example, interactions with residents, raters' sense of personal safety, and aesthetic appeal of the block face. They also voiced concerns that a failure to capture these aspects (owing to the fixed response categories of the observational tool) might result in misrepresentation of the BFs under study. Fol- lowing group discussions a decision was made early on to create a separate "Comments" section within the SSO tool. This encouraged the observers to document their experiences and additional observations not otherwise effectively captured by the instrument. We conducted exploratory factor analysis to elicit underlying dimensions of the 3 meta-categories in the studied neighbourhoods. Because items were binary, we estimated the tetrachoric correlations and performed dichotomous factor analysis using the weighted least- squares with mean and variance adjustment estimator [32,33]. In brief, tetrachoric correlations assume a thresh- old model for the observed binary items and latent bivar- All hand-written field comments were transcribed into an electronic text document and labelled with BF and neighbourhood identifiers. The comments were then independently coded by three members of the research team using a content analysis approach. Emergent cate- gories and themes were identified and presented to the qualitative sub-committee for further analysis and dis- Parsons et al. Descriptive Statistics The prevalence of the remaining 58 items was calculated and compared 1) between low and middle/high income neighborhoods; and 2) between individual neighbor- hoods (Table 1). Statistically significant differences (based on income and neighbourhood) were identified for: items regarding the physical condition of the neigh- bourhood (garbage; cigarette butts; poor condition of buildings, grounds, and public spaces; flow of traffic; number of trees), social aspects of the neighbourhood (the presence of drinking establishments; the presence of drunken, disorderly adults or gangs; residents socializing in mixed racial groups; the presence of languages other than English on the BF), and other factors such as the presence of usable public phones, signs denoting a neigh- bourhood name, and the presence of toys in private resi- dential grounds. Several items did not demonstrate significant differences between low and high income neighbourhoods, but achieved statistical significance upon stratification by individual neighbourhood. These variables included resident reaction to raters, presence of public courtesies, graffiti, presence of children/teenagers/ adults, and others. This example speaks to instances of commonality as well as differences in perception. Such commentary from the qualitative component of the SSO and the retrospec- tive group discussion with raters raised issues relating to their preconceived notions of the neighbourhoods under study - perceptions which were often based on neigh- bourhood reputation, media accounts and previous per- sonal experiences. Our findings revealed that a rater's preconceived notions about a given neighbourhood could be - and often were - challenged by their fieldwork expe- riences. In addition, these experiences provided richer contextual information not immediately accessible from straightforward application of the SSO tool, especially in relation to heterogeneity and covert disorder. For exam- ple, in the group discussion, one rater commented, Features of the rater's experience while observing the BF Raters spent hours at a time performing SSOs, and this seemed to provoke reflections about their experiences during data collection. Not surprisingly, the findings stemming from these personal experiences emerged as an independent theme. A characteristic example follows: These qualitative findings are crucial to understanding and interpreting the quantitative data, its limitations and strengths, and the performance of the tool in the field. It is to this analysis which we now turn. "I talked to two people on the block. A lady saw me looking at the neighbour's driveway and wanted to know what I was doing. She told me that there were lots of historic homes in the area, that this area was really nice. Her neighbour (a soldier in the army) ... started talking about how there are too many taxes, the street's aren't kept up, that Weston Road's a bad neighbourhood - apparently there's a really nice house near it that's not getting sold because there was a shooting 30 ft. away. He then went around and cleaned up not only his litter but the litter at his neighbour's house." (Weston) Quantitative Findings The quantitative analysis assesses the 'performance' of the SSO tool in the field. Individual items were evaluated with respect to their construction and sensitivity, and to determine whether it could be used to distinguish between neighbourhoods, and specifically between neighbourhoods of differing median income. Finally, fac- tor analysis was conducted to determine whether the instrument items could be grouped into meaningful higher-level constructs related to physical and social dis- order. Though seemingly simple in its narrative quality, this example represents a highly complex account of a neigh- bourhood interaction. Firstly, the neighbours feel com- fortable approaching a stranger and the woman appears to exercise vigilance with respect to her neighbour's property. Secondly, the female resident provides a posi- tive account of her neighbourhood, which is then con- trasted by the account of the second neighbour, who readily joins the conversation. In addition, there is the stark contrast between the pleasantness of the houses in the area coupled with the story of violence at one nearby residence, which again reinforces the notion of consider- able heterogeneity over small spaces and the concept of covert disorder. Finally, the male neighbour takes care of his own yard as well as his neighbour's suggesting social cohesion along the BF. the significant correlation between uniformity and lack of disorder [15]. The group discussion revealed that raters felt diversity, positive social interactions, familiarity, aesthetic qualities, structural attributes (e.g. the presence of sidewalks) and a sense of personal safety - all were attributes that contrib- uted to the impression of an appealing BF. Features of the BF being observed Raters expressed concern about misrepresenting a BF by capturing only limited inventory-type data contained within the SSO instrument. As a result, many raters attempted to provide additional contextual information during their observations. An example of a comment which attempted to justify marking a certain response is presented below. In this instance, the rater is uncomfort- able with completing Question 47 of the instrument, which documents the presence of 'vacant/undeveloped lots'. The rater wrote, "One very big undeveloped lot that looks like it may be used as a park or reclaimed as a park. There's a com- munity vegetable garden within it and a flock of Cana- dian geese are resting in it and the grass is cut, but I Page 7 of 19 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 the significant correlation between uniformity and lack of disorder [15]. y Physicality Meta Category We obtained 4 competing, interpretable, one-factor mod- els (Table 2). Each model was evaluated on the basis of theoretical justification and relative fit indices (RMSR) (3 yielded RMSR slightly above 0.05 and one had an RMSR < 0.05). The models had comparable RMSR values, but Models 1, 2, and 3 omitted items that were important in discriminating between neighborhoods on the basis of "diversity played definitely, I think we'd all agree, on providing a more positive experience......" Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 8 of 19 http://www.ij healthgeographics.com/content/9/1/6 Table 1: Prevalence of items in 176 block faces by neighbourhood Neighbourhood Total Low Income Middle/High Income Items (divided by meta- category) % No. EE, N = 39 % SP, N = 24 % SJT, N = 9 % W, N = 43 % NR, N = 11 % BDM, N = 50 % Physicality Items Any Abandoned/Beat-Up Vehicles 8.0 14 0 4.2 0 16.3 9.1 10.0 Some or a lot Garbage Or Odors,¶ 43.2 76 43.6 79.2 44.4 48.8 63.6 14.0 Any beer containers or liquor bottles on the street or sidewalks, in yards, or vacant lots 8.5 15 12.8 8.3 11.1 11.6 9.1 2.0 Some/A Lot cigarettes/cigar butts or discarded cigarette packages on any of the streets or sidewalks, in yards/lots or gutters,¶ 59.7 105 66.7 91.7 100.0 51.2 63.6 38.0 Any graffiti on any of the buildings, sidewalks, walls, or signs,¶ 41.5 73 33.3 66.7 100.0 32.6 45.5 32.0 Any Vacant/Undeveloped Land/Lots 11.4 20 12.8 12.5 22.2 4.7 9.1 14.0 Poor/fair/deteriorated condition of the public spaces,¶ 63.6 112 69.2 83.3 88.9 79.1 54.6 34.0 Poor/Fair/Abandoned condition of residential buildings,¶ 19.3 34 20.5 45.8 55.6 16.3 18.2 2.0 Poor/Deteriorated/Fair overall condition of residential grounds,¶ 22.2 39 23.1 45.8 33.3 27.9 18.2 4.0 Poor/Fair overall condition of the vacant/undeveloped property 8.5 15 12.8 12.5 22.2 4.7 0.0 6.0 Poor/Fair condition of commercial buildings 16.5 29 18.0 25.0 33.3 20.9 9.1 6.0 Table 1: Prevalence of items in 176 block faces by neighbourhood Parsons et al. y Physicality Meta Category International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 9 of 19 Any residences with "for sale" or "for rent" signs* 27.8 49 33.3 37.5 22.2 34.9 36.4 12.0 Any commercial/industrial buildings with "for sale" or "for rent" signs* 10.2 18 5.1 25.0 11.1 11.6 9.1 6.0 Social Items Presence of a police officer or private security guard 9.7 17 12.8 16.7 33.3 4.7 9.1 4.0 Presence of adults loitering or selling drugs 8.0 14 10.3 8.3 33.3 7.0 9.1 2.0 Presence of gangs or drinking, drunken or disorderly groups of adults*,¶ 6.3 11 7.7 8.3 55.6 2.3 0.0 0.0 Presence of indications of neighbourhood uniformity 10.2 18 2.6 12.5 22.2 7.0 0.0 0.0 Presence of signs which denote a neighbourhood name,¶ 10.2 18 5.1 37.5 33.3 4.7 9. 2.0 Presence of neighbourhood crime watch signs 19.3 34 20.5 8.3 33.3 16.3 27.3 22.0 Presence of no trespassing or security warnings signs 38.1 67 30.8 45.8 44.4 39.5 45.5 36.0 Presence of any residences with signs indicating they are protected by dogs* 14.2 25 23.1 16.7 11.1 18.6 0.0 0.0 Presence of any residences with some form of decoration (e.g. decorated gardens/ balconies, nameplates, window boxes)? 76.7 135 82.1 66.7 66.7 76.7 81.8 78.0 Presence of any residential building with a fence or border 74.4 131 84.6 62.5 66.7 76.7 81.8 70.0 Presence of any residential buildings with window bars or door grates¶ 17.6 31 18.0 33.3 33.3 9.3 63.6 4.0 Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Parsons et al. y Physicality Meta Category International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 10 of 19 http://www.ij healthgeographics.com/content/9/1/6 Presence of any residential buildings with signs indicating they are protected by private security services 51.1 90 56.4 50.0 44.4 51.2 18.2 56.0 Presence of any commercial/ industrial buildings with barred windows 10.2 18 12.8 20.8 22.2 7.0 9.1 4.0 Resident react to the presence of raters\|\| 39.2 69 56.4 58.3 33.3 25.6 36.4 30.0 Three or more people present on the block face 35.8 63 20.5 37.5 100.0 37.2 63.6 28.0 Heard or saw a language other than English on the block face,¶ 30.1 53 35.9 41.7 77.8 25.6 36.4 14.0 Presence of any people socializing in mixed racial groups,\|\| 29.6 52 25.6 45.8 66.7 34.9 18.2 16.0 Presence of adults walking, socializing, doing home repair, sitting on the porch, supervising children, exercising, or patronizing businesses on the block face¶ 83.5 147 76.9 95.8 100.0 81.4 100.0 78.0 Presence of signs advertising cultural, political or social events or neighbourhood helper signs¶ 15.9 28 5.1 41.7 66.7 9.3 45.5 2.0 Presence of children¶ 49.4 87 51.3 37.5 88.9 53.5 90.9 34.0 Presence of teenagers¶ 37.5 66 30.8 25.0 88.9 53.5 63.6 20.0 Presence of public seating 10.8 19 7.7 16.7 22.2 7.0 18.2 10.0 Presence of children or teenagers in the parks or playgrounds\|\| 9.7 17 7.7 0.0 44.4 14.0 9.1 6.0 Presence of children or teenagers in playing in public spaces other than parks or playgrounds¶ 13.1 23 7.7 4.2 33.3 9.3 63.6 10.0 Presence of any residential grounds with toys or play equipment,¶ 25.0 44 33.3 0.0 0.0 20.9 36.4 36.0 Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Parsons et al. y Physicality Meta Category International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 11 of 19 Resources Access to public transportation on block face* 26.1 46 23.1 37.5 44.4 32.6 9.1 18.0 Presence of public courtesies, such as public seating, trash cans, newspaper dispensers and public pay phones¶ 33.5 59 23.1 58.3 77.8 30.2 36.4 24.0 Presence of usable public phones,¶ 10.8 19 12.8 16.7 44.4 11.6 0.0 2.0 Presences of handicap accessibility facilities such as sidewalk ramps, wheel-trans stops, or reserved parking 18.2 32 10.3 12.5 55.6 20.9 18.2 18.0 Presence of any bicycle facilities such as bike lanes and parking stands¶ 12.5 22 2.6 33.3 44.4 9.3 27.3 4.0 Presence of a park or playground. 14.2 25 12.8 4.2 44.4 16.3 9.1 14.0 Presence of other play spaces (sports fields or children playing in other public spaces)¶ 30.1 53 18.0 4.2 55.6 41.9 72.7 28.0 Good or well-kept public spaces,¶ 31.8 56 20.5 12.5 11.1 20.9 45.5 60.0 Playgrounds/parks in good or excellent condition 8.0 14 7.7 0.0 0.0 9.3 9.1 12.0 Other play spaces in good or excellent condition,¶ 19.9 35 15.4 0.0 11.1 23.3 54.6 24.0 Presence of commercial establishments\|\| 26.7 47 23.1 50.0 44.4 25.6 27.3 16.0 Presence of institutions (eg. Schools, churches, medical clinics, libraries) 19.9 35 18.0 29.2 33.3 23.3 27.3 10.0 Presence of drinking establishments,¶ 10.2 18 12.8 33.3 11.1 4.7 18.2 0.0 Light or very light traffic*,\|\| 63.1 111 59.0 41.7 44.4 62.8 72.7 78.0 Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Parsons et al. y Physicality Meta Category Jamestown, W = Weston, NR = North Riverdale, BDM = Banbury Don Mills Significant difference p < .05, two-sided Fisher's exact test comparing low income and high income neighbourhoods ** Significant difference p < .01, two-sided Fisher's exact test comparing low income and high income neighbourhoods \|\| Significant difference p < 0.05, Fisher's exact test comparing all six neighbourhoods ¶ Significant difference p < 0.01, Fisher's exact test comparing all six neighbourhoods establishments where alcohol was served) constituted the "Neighbourhood Social Accessibility" factor. physical features, namely the amount of cigarette butts and the condition of abandoned buildings (see prevalence Table 1). Model 4 included these items and had similar factor loadings for the common items. Therefore we con- sidered Model 4 with 7 items as the most interpretable model, capturing the greatest number of physical BF characteristics, with no loss of relative fit to the data. The factor outlined in Model 4 was named "Physical Decay and Disorder" and retained for use in neighborhood com- parisons. The factor labeled "Recreational opportunities" was composed of Public spaces condition, Good parks condi- tion, Good other play space condition and Light/Very Light Flow of Traffic items. This factor reflected the opportunities for recreation and safe play that were afforded by the block face. Comparing Neighbourhood Physical Decay and Disorder A scale was created from the Physical Decay and Disorder factor identified in the Physicality exploratory factor analysis. A summary score for each BF was created by adding one point for each of the seven items endorsed. The potential minimum score was 0 and the potential maximum score was 7. A box and whisker plot was gener- ated comparing Physical Decay and Disorder scores for BFs stratified by neighbourhood (Fig. 2). y Physicality Meta Category International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 12 of 19 Page 12 of 19 Built Environment Presence of single family homes\|\| 66.5 117 76.9 50.0 22.2 74.4 81.8 64.0 Presence of buildings greater than two stories tall¶ 30.7 54 7.7 70.8 100.0 18.6 90.9 14.0 Greater or equal to 20 residences on the block face¶ 54.0 95 61.5 16.7 33.3 37.2 27.3 62.0 Greater or equal to 20 trees on the block face,¶ 31.3 55 38.5 45.8 33.3 41.9 18.2 12.0 All residences have access to a private yard or garden¶ 54.0 95 79.5 8.3 0.0 53.5 27.3 72.0 Greater or equal to 50% of residences have a private balcony, porch or terrace¶ 47.2 83 25.6 58.3 77.8 41.9 36.4 60.0 EE = Eglington East, SP = South Parkdale, SJT = St. Jamestown, W = Weston, NR = North Riverdale, BDM = Banbury Don Mills Significant difference p < .05, two-sided Fisher's exact test comparing low income and high income neighbourhoods ** Significant difference p < .01, two-sided Fisher's exact test comparing low income and high income neighbourhoods \|\| Significant difference p < 0.05, Fisher's exact test comparing all six neighbourhoods ¶ Significant difference p < 0.01, Fisher's exact test comparing all six neighbourhoods Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) Table 1: Prevalence of items in 176 block faces by neighbourhood (Continued) EE = Eglington East, SP = South Parkdale, SJT = St. Jamestown, W = Weston, NR = North Riverdale, BDM = Banbury Don Mills * Significant difference p < .05, two-sided Fisher's exact test comparing low income and high income neighbourhoods ** Significant difference p < .01, two-sided Fisher's exact test comparing low income and high income neighbourhoods \|\| Significant difference p < 0.05, Fisher's exact test comparing all six neighbourhoods ¶ Significant difference p < 0.01, Fisher's exact test comparing all six neighbourhoods EE = Eglington East, SP = South Parkdale, SJT = St. Social Meta Category It was not possible to extract a model with interpretable factor structure, overall or stratified by income level. Fac- tors reported in published studies [7,21] were not vali- dated in this sample. Resource Meta Category Fifteen Resource items could not be factor analyzed in the same model. Two subsets of items yielded 2 separate one-factor models (Table 3). Items that reflected accessi- bility within the neighbourhood (bike facilities, and pub- lic transport) and social activities (cultural events and Mean Physicality scores across neighbourhoods were statistically different (ANOVA, p < 0.001). As variances among neighbourhoods were equal (Levene test p-value = 0.193), we compared pairs of neighbourhoods using Tukey's Honestly Significant Test at α = 0.05. Banbury- Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 13 of 19 Table 2: Exploratory factor analysis of Physicality items Item Model 1 Model 2 Model 3 Model 4 Poor/Fair condition of commercial buildings 0.725 0.711 0.766 0.749 Poor/fair/deteriorated condition of public spaces 0.562 0.604 0.597 0.618 Poor/fair/abandoned condition of residential buildings 0.640 0.630 Some or a lot of Garbage or Odors 0.923 0.907 0.814 0.826 Any beer containers or liquor bottles 0.660 0.644 0.696 0.694 Any graffiti 0.496 0.505 0.547 0.547 Some/A Lot cigarettes/ cigar butts 0.937 0.930 χ2 p-value 0.949 0.938 0.870 0.913 RMSR 0.043 0.054 0.053 0.060 Table 2: Exploratory factor analysis of Physicality items demonstrates the depth and richness of analysis that can be obtained by using a mixed-methods approach for SSOs. To our knowledge this is one of the first studies to incorporate qualitative observational findings and researcher reflections into an investigation employing quantitative systematic observational tools. As such it represents an important contribution to our understand- ing of neighbourhood evaluation. One British study (Morrow, 2000) [34] has examined youth residents' self- reported perceptions of neighbourhood physical context and its impact on youth well-being using qualitative methods, however we use these techniques to tap raters' perceptions, not those of residents. Furthermore, Mor- row does not attempt to tie her findings to observational data collected in the neighbourhoods under study[34]. Don Mills had significantly lower average scores than Eglinton East, South Parkdale, St Jamestown and Weston; Eglinton East had lower average scores than South Parkdale; and South Parkdale had lower average scores than Weston. The Wilcoxon rank-sum test using a Bon- ferroni correction for multiple comparisons yielded the same results. Mean Social and Resource scores could not be compared across neighbourhoods because it was not possible to extract models with interpretable factor struc- tures for either of these meta-categories. Discussion This study sought to evaluate the performance of an SSO instrument in the context of six Toronto neighbour- hoods. This instrument was based on scales and con- structs from prior studies largely generated in US settings. The factor analysis based on the quantitative data, taken together with the qualitative findings, raises important questions and concerns with respect to the 'transferability' of such constructs to the Toronto context. Interpreting the Data g y Analysis of the qualitative data revealed three broad themes related to use of the SSO instrument in the field: (1) features of the BF being observed, (2) features of the BF relative to the neighbourhood, and (3) features of the rater's experience while observing the BF. Each of these themes poses further questions with regard to the utility of the SSO instrument in the Toronto context and chal- lenges some of the assumptions upon which SSO research has been based to date. In particular, our quali- tative findings urge us to question the meaning of the quantitative results with respect to the underlying social processes relating to neighbourhood disorder. transferability of such Interpreting the Data The qualitative component of this study contributes to the literature on SSO by acknowledging the inherently subjective nature of neighbourhood observations (and recognizing the positive contributions of such subjective data sources) - a consideration absent from the literature to date. The inclusion of qualitative methods to our study Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 14 of 19 Low Income Neighbourhoods High Income N’hoods Box and whisker plot comparing Physical Decay and Disorder Scale by neighbourhood Figure 2 Box and whisker plot comparing Physical Decay and Disorder Scale by neighbourhood. Median, IQR and range of Physical Decay and Disorder Scores stratified by neighbourhood. Physical Decay and Disorder score is the sum of 7 items: Poor/Fair condition of commercial buildings; Poor/fair/deteriorated condition of public spaces; Poor/fair/abandoned condition of residential buildings; Some or a lot of Garbage or Odors; Any Graf- fiti; Any beer containers or liquor bottles; Some or a lot of cigarette butts. Low Income Neighbourhoods High Income N’hoods Low Income Neighbourhoods High Income N’hoods Box and whisker plot comparing Physical Decay and Disorder Scale by neighbourhood Figure 2 Box and whisker plot comparing Physical Decay and Disorder Scale by neighbourhood. Median, IQR and range of Physical Decay and Disorder Scores stratified by neighbourhood. Physical Decay and Disorder score is the sum of 7 items: Poor/Fair condition of commercial buildings; Poor/fair/deteriorated condition of public spaces; Poor/fair/abandoned condition of residential buildings; Some or a lot of Garbage or Odors; Any Graf- fiti; Any beer containers or liquor bottles; Some or a lot of cigarette butts. The first theme (features of the BF being observed) problematizes the 'objective' nature of this form of data collection. There were numerous instances where raters found it challenging to appropriately characterize BFs, and they feared misrepresenting BFs - and the implica- tions of such misrepresentations when drawing compari- sons between neighbourhoods. This may be a concern when examining urban settings which have low levels of severe disorder and substantial intra-neighbourhood variation as was noted in our study. The quantitative find- ings are more readily interpretable in light of the complex nature of the data collection process. For example, the factor 'physical decay and disorder' was the only one which resembled the constructs generated in the US studies. The high prevalence of cigarette butts (noted by raters in all neighbourhoods) accounted for a significant proportion of this factor. transferability of such Interpreting the Data The physical decay/disorder construct was further influenced by the high prevalence of ratings of 'poor/fair/deteriorated condition of public spaces' (with this rating assigned to 63.6% of BFs). The standardized instructions may have skewed these results in favor of a preponderance of 'fair' ratings. The standard- ized instructions indicated that any street, sidewalk, pub- lic transit stop, public parks or grounds, public schools or any non-private land should be marked in 'fair' condition if it showed irregular maintenance (including those with even small amounts of cracked concrete or paint or mod- erately overgrown vegetation) and overall the space was "in decent condition, but (rater) would recommend addi- tional upkeep." Such instructions logically resulted in Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 15 of 19 Table 3: Two one-factor models of separate subsets of Resource items One-Factor Model Factor loading χ2 p-value RMSR "Neighbourhood social accessibility" Access to public transportation on the block face 0.731 0.523 0.0547 Presence of any bicycle facilities such as bike lanes and parking stands 0.893 Presence of signs advertising cultural, political or social events or neighbourhood helper signs 0.531 Presence of drinking establishments 0.818 "Recreational opportunities" Good or well-kept public spaces 0.795 0.5644 0.0516 Parks/playgrounds in good or excellent condition 0.609 Other play spaces in good or excellent condition 0.589 Light or very light traffic 0.648 Table 3: Two one-factor models of separate subsets of Resource items 0.0516 order' scale revealed that low-income neighbourhoods were more likely to be characterized by greater levels of disorder/decay than middle/high-income neighbour- hoods. However, the box-and-whisker plots indicate that considerable heterogeneity exists within each neighbour- hood, regardless of income. Taken together, these mixed- method findings suggest that the impact of concentrated disorder (evident in smaller pockets within neighbour- hoods) may be diluted when describing neighbourhoods more broadly, implying a notion of "covert disorder" in the Toronto setting. most raters ranking public spaces as being in fair condi- tion. However it is questionable whether this degree of disorder on its own would result in a negative experience for persons using the BF. It was discomfort with this type of rating that raters' comments reflected. The second theme addressing features of the BF relative to the surrounding neighbourhood challenges commonly held assumptions regarding neighbourhood homogeneity as these relate to Toronto neighbourhoods. transferability of such Interpreting the Data This is because, as observers on foot, raters could observe in 360° over a longer period of time, since most observations required at least 30 min- utes for completion. Moreover, the narrative accounts of raters' experiences often revealed unanticipated information concerning the neighbourhoods. For example, raters suggested in group discussion that evidence of extreme social disorder was often fleeting to the outsider--erupting to the surface at intervals, but not always obvious at first glance. As well, information obtained through interaction and observa- tion frequently challenged raters' preconceived notions of the BF or neighbourhood. The qualitative findings rein- force the importance of heterogeneity and covert disor- der in explaining features of the neighbourhoods under study. By capturing the raters' experiences in a systematic way, the qualitative portion of our study was able to access yet another level of rich contextual information that supported and helped in interpreting our quantita- tive results. It was not possible to extract a model with interpretable factor structure for the Social meta-category. Possible explanations include lack of power due to rare items or small sample size, or an artifact due to how the items were dichotomized. However this may also reflect funda- mental differences between the neighbourhoods included in the present investigation and the neighbourhoods that were used in the Chicago and Baltimore studies - mean- ing that the constructs of Territoriality and Social Disor- der may not be applicable to Toronto. The quantitative findings also provided new insights regarding SSO. The generation of the two distinct con- structs from the resources meta-category is an original contribution to our understanding of neighbourhoods. The 'neighbourhood social accessibility' factor speaks to neighbourhoods as dynamic entities rather than static ones. It reflects the ease with which one can enter and leave a neighbourhood and - coupled with signs advertis- ing social and cultural events as well as the presence of drinking establishments - suggests features of neighbour- hoods that make them desirable places for both residents and non-residents alike. The 'recreational opportunities' factor represents a related construct in that places to play or meet in public spaces without being overwhelmed by traffic (and its attendant congestion, parking difficulties, noise and pollution) might also prove appealing to resi- dents. In a forthcoming study employing concept map- ping (Sheppard et al: "Are Canadians influenced by their urban neighbourhoods? transferability of such Interpreting the Data The qualita- tive findings underline the importance of heterogeneity within the neighbourhood and how this sense of hetero- geneity may impact the overall impression of a specific area within a neighbourhood or the entire neighbour- hood itself. The impact of such heterogeneity challenges the classic understanding of a neighbourhood which pre- supposes certain levels of homogeneity within a specified bounded area [35]. It may also provoke questions con- cerning the significance of the BF or smaller bounded communities within neighbourhoods in the presence of considerable neighbourhood variability. Certainly the quantitative findings reflected neighbourhood heteroge- neity as well. The creation of the 'physical decay and dis- The third theme examining features of the rater's expe- rience while observing the BF speaks to the access to infor- mation that would not have been otherwise obtained by using a purely quantitative approach or even the use of other data collection methods such as observations per- formed by driving through neighbourhoods [7,21]. The fact that raters performed observations on foot, walking up and down a BF numerous times, offered residents the opportunity to interact with them, in turn yielding detailed narrative accounts. When raters did not engage with residents, their very presence on the BF provided Page 16 of 19 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 der' and on some individual items. Several items did not demonstrate significant differences between low and high income neighborhoods, but achieved statistical signifi- cance upon stratification by individual neighbourhood. These variables (including resident reaction to raters, presence of public courtesies, graffiti, signs advertising cultural or social events, the presence of children, teenag- ers and adults, features of the built environment) may be features of neighbourhoods that are not necessarily linked to income, but are rather descriptors of the unique character of neighbourhoods. Conversely, several items were only statistically significant upon stratification by income (poor/fair condition of commercial buildings, buildings for sale or rent, residences protected by dogs). These items may be more strongly linked to income than neighbourhood, or there may be insufficient power to resolve them. them access to observations that might have escaped notice using other methods of data collection such as drive-by observation. transferability of such Interpreting the Data Neighbourhood characteristics and their perceived impact on self-rated mental well- being," submitted)- which asked residents for their per- spectives on neighbourhoods and mental health - resi- dents indicated that pedestrian-friendly neighbourhoods, accessible by public transit or other means, with plenty of public services, places to meet and occasions to celebrate - all were reported as contributing to residents' mental well-being. The quantitative findings in the present study suggest that such neighbourhood factors are 'observable' (physically quantifiable) and are important to residents. By combining the quantitative and qualitative analyses, a number of interesting points for discussion are posed. With respect to the transportability of previously employed SSO tools into a Canadian context, it is fair to ask whether the notion of 'social disorder' is the most appropriate to this setting [36]. In particular, the choice of variables included in the tool and how these were opera- tionalized were of concern to raters. For example, raters were concerned by the limited ability of the SSO instru- ment to capture certain characteristics of the BF that they felt were important for the Toronto setting, such as the aesthetic appeal of a BF (e.g. the degree of order and diversity of land use), the rater's sense of personal safety, and their experiences during data collection. In contrast, as elicited during the group discussion, many raters felt that items within the SSO tool relating to extreme physi- cal or social disorder were not as relevant to the study of Toronto neighbourhoods, but that these items accounted for a considerable proportion of the overall observation. Oreopoulos (2005) also posited differences in neighbour- hood disorder as experienced by Canadian and US resi- dents [23]. Instead of focusing on disorder, the mixed method findings in our study suggest that perhaps the notion of 'order' may be more pertinent in the context of Toronto neighbourhoods (but not 'order' conceived of as a mere corollary to that of 'disorder' premised in most of the SSO literature). We caution that the raters' reflections Within the quantitative analysis, we were able to iden- tify some differences between low and high income neighbourhoods on the factor 'physical decay and disor- Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 17 of 19 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 17 of 19 bourhoods? transferability of such Interpreting the Data Neighbourhood characteristics and their perceived impact on self-rated mental well-being," sub- mitted) suggests that residents of apartment buildings (particularly high-rise buildings) include the internal spaces between apartments (lobbies, elevators, common areas) as important features of 'neighbourhood' for them. Perhaps we need to adapt a tool that will capture both 'internal' and 'external' neighbourhood characteristics (and the relative order or disorder therein). What is the 'appropriate' level of observation when evaluating neigh- bourhoods? are those of a very select group of participants - academic researchers, not laymen. Nevertheless, the positive emphasis that all raters gave to diversity (as an appealing attribute at both the BF and neighbourhood level) chal- lenges prevailing planning notions that stress uniformity. The qualitative data therefore help to highlight and rein- force specific challenges regarding the transfer and subse- quent application of an SSO tool from one urban context to another. In the case of Toronto, additional variables corresponding to more specific concepts of safety, aes- thetic appeal and order, and heterogeneity might be con- sidered in any future revision of the SSO instrument. As noted already, this study has a number of important limitations. We were unable to generate factors for the social meta-category, with potential explanations includ- ing: a lack of power (secondary to either low prevalence or small sample size), artifact from dichotomizing the variables, or a fundamental problem with the construct of social disorder underlying the adapted scales (adapted from US and UK contexts). Anecdotally, it is not unusual for visitors to Toronto (or even agency representatives providing funding to low- income inner-city neighbourhoods) to ask "when are we going to get to the 'bad' neighbourhood?" As such, rela- tive disorder (both physical and social) and decay are not always obvious to the casual observer. This is not to sug- gest that Toronto does not have its share of both physical and social disorder. For example, while few homeless peo- ple were observed during the study, we know that Toronto has many homeless residents [37]. Rather the tool failed to capture this facet of city life. If we are to use the concept of disorder (rather than order), perhaps a rec- ognition of the concealed nature of disorder (particularly social disorder) is more reflective of the Toronto experi- ence. Another important question that we cannot answer by this investigation relates to linkages between the con- structs generated by SSO and health. transferability of such Interpreting the Data This study rather represents an initial step in understanding the utility and applicability of these tools in different contexts. As such our findings will assist health researchers in interpreting the findings they acquire when using these measures. It is important to note that the nature of the relationship between neighbourhoods and health has been difficult to delineate and such linkages are no doubt complex. Given the increasing interest in this area of research and the use of SSO tools, improved understanding of the instruments themselves (evaluating both their strengths and limita- tions) is a valuable contribution. For example, if we take the prevalence of homelessness and relate it to health, for example, what are the attributes at the individual level that contribute to vulnerability, and what aspects of place make individuals vulnerable? These are questions which will only be answered by studies employing a variety of methodologies (observation, concept mapping, surveys, interviews, document analysis, policy analysis, etc.), such as Klinenberg's approach of 'social autopsy' [38]. Taken together, our qualitative and quantitative find- ings compel us to interrogate the theoretical assumptions underlying social and physical disorder in a Toronto con- text. As such, we propose that alternative theoretical con- cepts might be more relevant to this setting given the complexity of the phenomena under investigation. For example, the finding of considerable heterogeneity and and the discussions amongst raters concerning 'covert disorder' might suggest that residents perceive their local BF or immediate surroundings as more representative of their functional neighbourhood. From this perspective, the concept of smaller functional geospatial communities bounded within traditionally defined neighbourhoods may very well have a substantial impact on any concep- tual framework attempting to describe how neighbour- hoods affect health - particularly as they relate to health in Toronto or similar urban settings. Our employment of mixed quantitative and qualitative methods represents a unique contribution to the field of neighbourhoods-and-health research. The qualitative findings enhance our understanding of the quantitative data and analyses, but also add new and important infor- mation regarding raters' experiences in conducting such research. Both forms of data (and their interpretation) contribute to our understanding of neighbourhood-level characteristics and both pose important questions regarding the best ways to characterize neighbourhoods. This observational study (like others before it) chose to focus on the facades of BFs as units for observation. Conclusions 2Department of Physical Therapy, University of Toronto, Toronto, Canada, 3Department of Obstetrics & Gynecology, Markham-Stouffville Hospital, Toronto, Canada, This study tested the performance of an SSO tool (adapted from previous studies in the US and UK) in a Canadian context. Based on our analysis, the SSO tool as implemented here demonstrated considerable shortcom- ings when applied to these six Toronto neighbourhoods. While we were able to generate three 'factors' using the quantitative data (physical decay and disorder, neigh- bourhood social accessibility, and recreational opportuni- ties), only the first reflected the findings of other investigators in that it was able to differentiate between low and higher income neighbourhoods. Unlike investi- gators in other jurisdictions, we found relatively few instances of items linked to 'disorder' in implementing the SSO tool. Coupled with the qualitative findings, which related to raters' experiences of conducting neigh- bourhood observations, our study raises important ques- tions regarding the theoretical premises (such as social disorder) underlying much neighbourhoods-and-health research. To our knowledge, this study is the first to employ a qualitative component to SSO. Our findings demonstrate that US and UK-generated constructs are not readily adaptable to a Canadian context, and suggests other constructs that may be more appropriate to apply in this setting. It poses some interesting directions for future research, namely the development of new instru- ments, different levels of observation, new methodologi- cal approaches (particularly greater use of mixed methods), and new theories with which to understand neighbourhood-level effects on health. 4Centre for Research on Inner City Health, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada, 9Neurorehabilitation Program, Toronto Rehabilitation Institute, Toronto, Canada 9Neurorehabilitation Program, Toronto Rehabilitation Institute, Toronto, Canada Acknowledgements This study was funded through a Canadian Institutes of Health Research Strate- gic Training Program in the Transdisciplinary Approach to the Health of Mar- ginalized Populations, hosted by the Centre for Research on Inner City Health, Li Ka Shing Knowledge Institute of St. Michael's Hospital. 17. Skogan W: Disorder and decline: Crime and the spiral of decay in American neighboorhoods. New York: Free Press; 1990. 18. Franzini L, Caughy M, Spears W, Fernandez Esquer ME: Neighborhood economic conditions, social processes, and self-rated health in low- income neighborhoods in Texas: A multilevel latent variables model. Social Science & Medicine 2005, 61:1135-1150. Authors' contributions JAP d h d JAP participated in the design of the study (including conceptualization of the qualitative components), data collection and analysis, and drafted the manu- script. GS participated in the data collection and analysis and assisted with pre- paring the manuscript. ANS carried out data collection and analysis (quantitative and qualitative), and assisted in manuscript preparation. RN per- formed the statistical analyses and contributed to the study design and manu- script preparation. PB assisted with data collection and analysis and assisted with manuscript preparation. AJ assisted with the qualitative data analysis. QZ contributed to data collection and analyses. AS contributed to data collection and analysis (qualitative and quantitative). JR assisted with data collection and quantitative analyses, as well as manuscript preparation. PO conceived the study, directed its design and coordination, contributed to the analyses and assisted with manuscript preparation. JD (with PO) conceived the study, directed its design and coordination, contributed to the analyses and assisted with manuscript preparation. All authors read and approved the final manu- script. 11. Pickett KE, Pearl M: Multilevel analyses of neighbourhood socioeconomic context and health outcomes: a critical review. Journal of Epidemiology and Community Health 2001, 55(2):12. 12. Markowitz FE, Bellair PE, Liska AE, Liu J: Extending social disorganization theory: Modeling the realtionships between cohesion, disorder and fear. Criminology 2001, 39(2):28. y 13. Sampson RJ, Raudenbush SW, Earls : Neighborhoods and violent crime: A multilevel study of collective efficacy. Science 1997, 277:7. 14. Perkins DD, Meeks JW, Taylor RB: The physical environment of street blocks and resident perceptions of crime and disorder: Implications for theory and measurement. Journal of Environmental Psychology 1992, 12:21-34. 15. Raudenbush SW: The quantitative assessment of neighborhood social environments. In Neighbourhoods and Health Edited by: Kawachi I, Berknman LF. New York: Oxford University Press; 2003:112-131. 16. Sampson RJ: Neighborhood-level context and health: Lessons from sociology. In Neighbourhoods and Health Edited by: Kawachi I, Berknman LF. New York: Oxford University Press; 2003:132-146. transferability of such Interpreting the Data Given the questions raised here regarding the nature of disorder, it is fair to ask whether we are looking for disor- der in the 'right' places. Perhaps we should be sampling the considerable network of 'back alleys' that are a staple of Toronto's inner-city neighbourhoods (a suggestion offered by raters during the group discussion). The data from our forthcoming concept mapping study (Sheppard et al.: "Are Canadians influenced by their urban neigh- Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Page 18 of 19 2Department of Physical Therapy, University of Toronto, Toronto, Canada, 3Department of Obstetrics & Gynecology, Markham-Stouffville Hospital, Toronto, Canada, 4Centre for Research on Inner City Health, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada, 5Dalla Lana School of Public Health, University of Toronto, Toronto, Canada, 6Institute of Medical Science, University of Toronto, Toronto, Canada, 7Cancer Care Ontario, Toronto, Canada, 8FutureHealth Inc., Toronto, Canada and 9Neurorehabilitation Program, Toronto Rehabilitation Institute, Toronto, Canada Competing interests Th h d l h Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. 10. MacIntyre S, Ellaway A: Neighborhoods and health: An overview. In Neighborhoods and Health Edited by: Kawachi I, Berkman LF. New York: Oxford University Press; 2003:20-42. References Raudenbush SW, Sampson RJ: Ecometrics: Toward a science of assessing ecological settings, with application to the systematic social observation of neighborhoods. Sociological Methodology 1999, 29:41. 8. Sampson RJ, Morenoff JD: Ecological perspectives on the neighborhood context of urban poverty: Past and present. In Neighborhood Poverty Volume II. Edited by: Brooks-Gunn J, Duncan GJ, Aber JL. New York: Russell Sage Foundation; 1997:1-22. 8. Sampson RJ, Morenoff JD: Ecological perspectives on the neighborhood context of urban poverty: Past and present. In Neighborhood Poverty Volume II. Edited by: Brooks-Gunn J, Duncan GJ, Aber JL. New York: Russell Sage Foundation; 1997:1-22. 9. Schaefer-McDaniel N, Caughy M, O'Campo P, Gearey W: Examining methodological details of neighborhood observations and the relationship to health: A literature review. Social Science & Medicine 2009, 70(2):277-92. References 1. Fone D, Dunstan F, Lloyd K, Williams G, Watkins J, Palmer S: Does social cohesion modify the association between area income deprivation and mental health? A multilevel analysis. International Journal of Epidemiology 2007, 36:338-345. 1. Fone D, Dunstan F, Lloyd K, Williams G, Watkins J, Palmer S: Does social cohesion modify the association between area income deprivation and mental health? A multilevel analysis. International Journal of Epidemiology 2007, 36:338-345. 2. Kawachi I, Berkman LF: Introduction. In Neighborhoods and Health Edited by: Kawachi I, Berkman LF. New York: Oxford University Press; 2003:1-19. 2. Kawachi I, Berkman LF: Introduction. In Neighborhoods and Health Edited by: Kawachi I, Berkman LF. New York: Oxford University Press; 2003:1-19. 2. Kawachi I, Berkman LF: Introduction. In Neighborhoods and Health Edited by: Kawachi I, Berkman LF. New York: Oxford University Press; 2003:1-19. 3. Macintyre S, Ellaway A, Cummins S: Place effects on health: how can we conceptualise, operationalise and measure them? Social Science and Medicine 2002, 55(1):15. 4. Marmot MG: Improvement of social environment to improve health. Lancet 1998, 351:57-60. 4. Marmot MG: Improvement of social environment to improve health. Lancet 1998, 351:57-60. 5. Cutchin MP: The need for the 'new health geography' in epidemiologic studies of environment and health. Health & Place 2007, 13:725-742. 6. Gephart MA, Brooks-Gunn J: Introduction. In Neighborhood Poverty Volume II. Edited by: Brooks-Gunn J, Duncan GJ, Aber JL. New York: Russell Sage Foundation; 1997:xiii-xxii. 5. Cutchin MP: The need for the 'new health geography' in epidemiologic studies of environment and health. Health & Place 2007, 13:725-742. 6 G h MA B k G J I d i I N i hb h d P 5. Cutchin MP: The need for the 'new health geography' in epidemiologic studies of environment and health. Health & Place 2007, 13:725-742. 5. Cutchin MP: The need for the new health geography in epidemiologic studies of environment and health. Health & Place 2007, 13:725-742. 6. Gephart MA, Brooks-Gunn J: Introduction. In Neighborhood Poverty Volume II. Edited by: Brooks-Gunn J, Duncan GJ, Aber JL. New York: Russell Sage Foundation; 1997:xiii-xxii. , 6. Gephart MA, Brooks-Gunn J: Introduction. In Neighborhood Poverty Volume II. Edited by: Brooks-Gunn J, Duncan GJ, Aber JL. New York: Russell Sage Foundation; 1997:xiii-xxii. 7. Raudenbush SW, Sampson RJ: Ecometrics: Toward a science of assessing ecological settings, with application to the systematic social observation of neighborhoods. Sociological Methodology 1999, 29:41. 7. 1Applied Health Research Centre, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada, Author Details 1Applied Health Research Centre, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada, Page 19 of 19 Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 19. Kohen DE, Brooks-Gunn J, Leventhal T, Hertzman C: Neighborhood income and physical and social disorder in Canada: Associations with young children's competencies. Child Development 2002, 73(6):1844-1860. 20. Coen SE, Ross NA: Exploring the material basis for health: characteristics of parks in Montreal neighborhoods with contrasting health outcomes. Health and Place 2006, 12(4):11. 21. Caughy MO, O'Campo PJ, Patterson J: A brief observational measure for urban neighborhoods. Health and Place 2001, 7(3):12. 22. Cohen D, Spear S, Scribner R, Kissinger P, Mason K, Wildgen J: "Broken windows" and the risk of gonorrhea. American Journal of Public Health 2000, 90(2):7. 23. Oreopoulos P: A Critique of Neighbourhood Effects in Canada. In New Approaches for Addressing Poverty and Exclusion Toronto: Policy Research Initiative, University of Toronto; 2005:52. 24. Sastry , Narayan , Pebley AR: The Los Angeles Family and Neighbourhood Survey: Neighbourhood Observation Forms and Interviewer Manual DRU-2400/6-LAFANS 2003. 2003. 25. Weich S, Burton E, Blanchard M, Prince M, Sproston K, Erens B: Measuring the built environment: validity of a site survey instrument for use in urban settings. Health and Place 2001, 7(4):10. g 26. Coffey A: The Ethnographic Self: Fieldwork and the Representation of Identity. London: Sage; 1999. 27. Maxwell JA: Evidence: A critical realist perspective for qualitative research. In Qualitative Inquiry and Social Justice Edited by: Denzin NK, Giardina MD. Walnut Creek, Calif.: Left Coast Press, Inc; 2009:108-122. 28. Glass T, McAtee MJ: Behavioral science at the crossroads in public health: extending horizons, envisioning the future. Social Science and Medicine 2006, 62(7):22. 29. Snyder EE: Teaching qualitative research: The meanings associated with farm buildings. Teaching Sociology 1995, 23(1):5. farm buildings. Teaching Sociology 1995, 23(1):5. 30. Toronto Neighbourhood Profiles [http://www.toronto.ca/ demographics/neighbourhoods.htm] 30. Toronto Neighbourhood Profiles [http demographics/neighbourhoods.htm] 31. Madriz E: Focus groups in feminist research. In Handbook of Qualitative Research 2nd edition. Edited by: Denzin NK, Lincoln YS. Thousand Oaks: Sage Publications Inc; 2000:835-850. 31. Madriz E: Focus groups in feminist research. In Handbook of Qualitative Research 2nd edition. Edited by: Denzin NK, Lincoln YS. Thousand Oaks: Sage Publications Inc; 2000:835-850. 32. McLeod LD, Swygert KA, Thissen D: Factor analysis for items scored in two categories. In Test scoring Edited by: Wainer DTH. Mahwah, NJ: Lawrence Erlbaum; 2001:189-216. 33. Muthén B, du Toit SHC, Spisic D: Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continous outcomes. Psychometrika 1997. 34. Parsons et al. International Journal of Health Geographics 2010, 9:6 http://www.ij-healthgeographics.com/content/9/1/6 Morrow V: 'Dirty looks' and 'trampy places' in young people's accounts of community and neighbourhood: Implications for health inequalities. Critical Public Health 2000, 10(2):141-152. 35. O'Campo P, Salmon C, Burke J: Neighbourhoods and mental well-being: What are the pathways? Health and Place 2009, 15:56-68. 36. Sampson RJ, Raudenbush SW: Seeing disorder: Neighbourhood stigma and the social construction of "broken windows". Social Psychology Quarterly 2004, 67(4):24. 37. Cowan L, Khandor E, Mason K, Hwang S: The Street Health Report. Toronto: Street Health; 2007. 37. Cowan L, Khandor E, Mason K, Hwang S: The Street Health Report. Toronto: Street Health; 2007. 38. Klinenberg E: Heat Wave: A Social Autopsy of Disaster in Chicago. Chicago: University of Chicago Press; 2002. doi: 10.1186/1476-072X-9-6 Cite this article as: Parsons et al., Standardized observation of neighbour- hood disorder: does it work in Canada? International Journal of Health Geo- graphics 2010, 9:6
https://openalex.org/W4255268491	https://www.qeios.com/read/6EGTUM/pdf	English	null	Thymic Mucosa-Associated Lymphoid Tissue Lymphoma	Definitions	2,020	cc-by	142	Open Peer Review on Qeios Open Peer Review on Qeios Thymic Mucosa-Associated Lymphoid Tissue Lymphoma National Cancer Institute Qeios ID: 6EGTUM · https://doi.org/10.32388/6EGTUM Qeios · Definition, February 7, 2020 Open Peer Review on Qeios Source National Cancer Institute. Thymic Mucosa-Associated Lymphoid Tissue Lymphoma. NCI Thesaurus. Code C45380. A low-grade extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue that affects the thymus. It is characterized by the presence of centrocyte-like or monocytoid lymphocytes that surround reactive lymphoid follicles. There is infiltration of the thymic epithelium by lymphocytes and formation of lymphoepithelial lesions. It is associated with the presence of autoimmune disease, most frequently Sjogren syndrome. Patients are usually asymptomatic and the tumor is detected incidentally as a mediastinal mass on chest x-ray. A minority of patients present with dyspnea, hemoptysis, and chest pain. The prognosis is excellent. Qeios ID: 6EGTUM · https://doi.org/10.32388/6EGTUM 1/1
https://openalex.org/W3214523043	http://repository.essex.ac.uk/34103/1/Using%20breeding%20and%20quantitative%20genetics%20to%20understand%20the%20C4%20pathway.pdf	English	null	Using breeding and quantitative genetics to understand the C4 pathway	Journal of experimental botany	2,021	cc-by	12,033	Using breeding and quantitative genetics to understand the C4 pathway Conor J.C. Simpson†, Gregory Reeves†, Anoop Tripathi, Pallavi Singh and Julian M. Hibberd, Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EA, UK † These authors contributed equally to this work. Correspondence: jmh65@cam.ac.uk † These authors contributed equally to this work. * Correspondence: jmh65@cam.ac.uk * Correspondence: jmh65@cam.ac.uk Received 6 September 2021; Editorial decision 21 October 2021; Accepted 3 November 2021 Editor: Alistair McCormick, University of Edinburgh, UK Abstract Reducing photorespiration in C3 crops could significantly increase rates of photosynthesis and yield. One method to achieve this would be to integrate C4 photosynthesis into C3 species. This objective is challenging as it involves en- gineering incompletely understood traits into C3 leaves, including complex changes to their biochemistry, cell biology, and anatomy. Quantitative genetics and selective breeding offer underexplored routes to identify regulators of these processes. We first review examples of natural intraspecific variation in C4 photosynthesis as well as the potential for hybridization between C3 and C4 species. We then discuss how quantitative genetic approaches including artificial selection and genome-wide association could be used to better understand the C4 syndrome and in so doing guide the engineering of the C4 pathway into C3 crops. Keywords: C4 photosynthesis, natural variation, hybridization, mapping population designs. © The Author(s) 2021. Published by Oxford University Press on behalf of the Society for Experimental Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. s) p y, , pp , https://doi.org/10.1093/jxb/erab486 Advance Access Publication 8 November 2021 Introduction After diffusing to an adjacent cell layer such as the bundle or mestome sheath, malate or aspartate are decarboxylated such that high concentrations of CO2 accumulate around Rubisco and so allow high rates of carboxylation (Fig. 1A). Finally, in species that use NAD-dependent malic enzyme (NAD-ME) or NADP-ME to release CO2 around Rubisco, the 3-carbon molecule produced from decarboxylation is regenerated to PEP in mesophyll cells by pyruvate orthophosphate dikinase (PPDK) to continue the cycle (Fig. 1A). types, plasmodesmatal frequency is increased compared with the C3 state (Botha, 1992; Danila et al., 2016). Some lineages contain suberin in the bundle sheath cell wall whilst others do not (Mertz and Brutnell, 2014), and whilst some C4 lineages arrange chloroplasts in bundle sheath cells centripetally, others do this centrifugally with respect to the veins (Edwards and Voznesenskaya, 2011). Lastly, soon after the discovery of C4 photosynthesis, dif- ferences in the biochemistry of the pathway were discovered among C4 species (Hatch et al. 1975). These different pathways were termed C4 ‘subtypes’ due to the fact that decarboxyl- ation is associated with three separate C4 acid decarboxylases, NADP-ME, NAD-ME, and phosphoenolpyruvate carboxykinase (PEPCK). Although there is growing support for the notion that species can modify the extent to which each C4 acid decarboxylases is engaged (Omoto et al., 2012; Sharwood et al., 2014; Sales et al., 2018), the differences in bio- chemistry associated with the subtypes exemplify the fact that the C4 pathway is a convergent phenomenon, and that its op- eration varies between species. The differences in leaf anatomy, cell biology, and biochem- istry between independent C4 lineages have frequently been summarized (Edwards and Voznesenskaya, 2011; Sage, 2016). In contrast, there have been fewer recent attempts to synthe- size the literature relating to forced hybridizations between C3 and C4 species. Studies have included somatic hybridizations of phylogenetically distant C3 and C4 plants, as well as sexual hybridizations of congeneric species. Whilst these wide hy- bridizations have provided insight into the extent to which C4 traits can be maintained and inherited in C3 species, a growing body of evidence documents variation in C4 traits within a species. We summarize examples of this work and suggest that there are opportunities to use quantitative trait mapping to better understand the C4 pathway. Introduction Photosynthetic plants provide humanity’s food, many textiles, and building materials, and represent the source of numerous medicines and fuels. Understanding how improvements in photosynthesis could be achieved therefore has the potential to impact many aspects of human life. Photosynthesis requires the enzyme Rubisco to fix atmospheric carbon dioxide (CO2) into 3-phosphoglycerate (Calvin and Benson, 1948). Species that only use Rubisco for carbon fixation are known as ‘C3’ plants, as 3-phosphoglycerate contains three carbon atoms. Rubisco, however, is also able to react with oxygen in add- ition to CO2. This oxygenation reaction produces the toxic molecule 2-phosphoglycolate, which must be metabolized and recycled via the photorespiratory cycle. Photorespiration leads to loss of carbon fixed by Rubisco and release of ammonia from amino acids at the expense of both ATP and reducing power (Bowes et al., 1971). Rates of photorespiration typically increase at higher temperatures because, under these condi- tions, the oxygenation reaction of Rubisco is favoured (Portis and Parry, 2007), but photorespiration can also increase during periods of drought when stomatal closure limits CO2 supply to the Rubisco active site. In extreme conditions, photorespiratory rates can use ~25% of photosynthetic outputs (Sharkey, 1988). Land plants have evolved two carbon-concentrating mechan- isms to reduce photorespiration. These are termed Crassulacean acid metabolism (CAM) and C4 photosynthesis. Whilst in both cases rates of photorespiration are reduced because compared C4 photosynthesis and quantitative genetics 3073 with the C3 state, ~10-fold higher concentrations of CO2 are supplied to Rubisco, CAM and C4 species use temporal and spatial systems, respectively. It is estimated that the C4 pathway has evolved independently from C3 ancestors at least 60 times to yield numerous phenotypes that concentrate CO2 around Rubisco (Sage et al., 2011). In all cases, in the C4 leaf Rubisco- dependent fixation of CO2 takes place in a specific compart- ment supplied with high concentrations of CO2 such that the oxygenase activity of Rubisco is almost completely abolished (Fig. 1A). In most C4 species, photosynthesis is compartmented between two cell types so that they are unified by a general pathway in which CO2 is converted to bicarbonate (HCO3 –) by carbonic anhydrase (CA) in mesophyll cells, and then com- bined with the 3-carbon molecule phosphoenolpyruvate (PEP) by the enzyme phosphoenolpyruvate carboxylase (PEPC) into the 4-carbon molecule oxaloacetate (Fig. 1A). Oxaloacetate is then either reduced to malate or transaminated to aspartate. Introduction Not only could these clas- sical approaches provide insight into the evolution and genetic basis of C4 photosynthesis, they may also inform efforts to en- gineer more efficient C3 crops. Traits underpinning C4 photosynthesis vary widely between species (Edwards and Voznesenskaya, 2011; Furbank, 2011; Sage and Stata, 2015; Sedelnikova et al., 2018). This interspe- cific variation in C4 traits includes differences in leaf anatomy, cell biology, and biochemistry, as well as the patterns of gene expression that determine these characteristics. For example, the cell types and arrangement of veins used by C4 species vary between lineages that have independently evolved the pathway (Fig. 1B). At least nine anatomical types have been described in the grasses (Poaceae) (Edwards and Voznesenskaya, 2011). Examples of this variation include in the number of layers of mestome and/or bundle sheath cells, and whether Rubisco is compartmented into the bundle or the mestome sheath. Although much of this variation associated with C4 photosynthesis is found in lineages that are separated by deep evolutionary time, Kranz anatomy also differs in species within families including the Amaranthaceae (Kadereit et al., 2003; Muhaidat et al., 2007; Sage, 2016), Asteraceae (Peter and Katinas, 2003), Cleomaceae (Koteyeva et al., 2011), Portulaceae (Voznesenskaya et al., 2017), and Poaceae (Ohsugi and Murata, 1985; Edwards and Voznesenskaya, 2011). Of the ~8100 C4 species defined to date, six operate the C4 pathway in a single cell (Fig. 1B). In these single-celled C4 species, the pathway is distributed between separate populations of chloroplasts such that the cell biology of these species has been modified com- pared with the C3 state. However, modifications to the cell biology of C4 leaves is not restricted to these single-cell species. In C4 species that separate photosynthesis between two cell Somatic hybridization of C3 and C4 species Approaches such as protoplast fusion allow somatic or asexual hybridization. Protoplasts from somatic cells from separate spe- cies are fused and regenerated into hybrid plants (Carlson et al., 1972; Evans, 1983). In many cases, asexual hybridization can lead to fertile hybrids between species that are considered sexually incompatible. Attempts to form hybrids via somatic hybridization of C3 rice (Oryza sativa) and other C4 grasses have been moderately successful. Terada et al. (1987) produced somatic hybrids between rice and C4 Echinochloa oryzicola that were morphologically different from either parent. Some con- tained 60 chromosomes which corresponded to the full hybrid complement, but plants developed necrosis and died before forming roots. Moreover, rice and C4 Panicum maximum (now 74 \| Simpson et al. . 1. Natural variation in C4 biochemistry and anatomy. (A) An overview of C4 biochemical subtypes. Although all forms of two-celled C4 photo olve initial CO2 fixation to generate four-carbon intermediates in mesophyll cells and diffusion to bundle sheath cells, the method of decarboxy reate a high-CO2 environment around Rubisco varies between C4 species. Solid and dashed lines show enzymatic and diffusion steps of the hway, respectively. (B) Examples of leaf anatomies seen in C4 species. Exemplar species that use each anatomical variant are shown below e e. Many more anatomical types have been described, which suggests that multiple leaf morphologies can facilitate the C4 pathway. Abbrevia mesophyll; B, bundle sheath; VB, vascular bundle; CCC, central cytoplasmic compartment; PC, peripheral chloroplast; WS, water storage ce oroplast. 3074 3074 \| Simpson et al. Simpson et al. Fig. 1. Natural variation in C4 biochemistry and anatomy. (A) An overview of C4 biochemical subtypes. Although all forms of two-celled C4 photosynthesis involve initial CO2 fixation to generate four-carbon intermediates in mesophyll cells and diffusion to bundle sheath cells, the method of decarboxylation to create a high-CO2 environment around Rubisco varies between C4 species. Solid and dashed lines show enzymatic and diffusion steps of the C4 pathway, respectively. (B) Examples of leaf anatomies seen in C4 species. Exemplar species that use each anatomical variant are shown below each type. Many more anatomical types have been described, which suggests that multiple leaf morphologies can facilitate the C4 pathway. Abbreviations: M, mesophyll; B, bundle sheath; VB, vascular bundle; CCC, central cytoplasmic compartment; PC, peripheral chloroplast; WS, water storage cell; ch, chloroplast. 1. Natural variation in C4 biochemistry and anatomy. Somatic hybridization of C3 and C4 species To our knowledge, whilst no hybrids have been re- ported in Blepharis, natural hybrids between Cynodon dactylon (NAD-ME) and Chloris sp. (PEPCK) display intermediate ac- tivities of NAD-ME and PEPCK (Prendergast, 1987). C3–C4 hybrids have been generated in the grasses by two broad approaches. First, as with dicotyledons, congeners using Megathyrsus maximus) were successfully fused to form hybrids with abnormal floral structures with lowered fertility (Xin et al., 1997). In all, 28 hybrids flowered but only five set fertile seed. To our knowledge, this work has never been repeated. g , p There have also been attempts to form hybrids between wheat and C4 grasses. A cell suspension of Trititrigia (a peren- nial hybrid of Triticum durum and Thinopyrum intermedium) was hybridized with maize (Wang et al., 1993; Wang and Niizeki, 1994). Plants that regenerated were aneuploids carrying in- complete sets of chromosomes from both species. Although the progeny were not full hybrids, this study demonstrated that after asexual hybridization, maize and Triticum chromosomes were not eliminated during successive cell divisions despite the uniparental genome elimination that occurs when both spe- cies are hybridized sexually (Laurie and Bennett, 1986, 1989; Laurie et al., 1990). Szarka et al. (2002) fused a cell suspen- sion of an albino maize mutant with wheat protoplasts. Plants that regenerated resembled maize but were green, indicating that photosynthesis from wheat rescued the albino phenotype in maize. Cytological observations showed the plants had all parental chromosomes, but no morphological traits associated with C4 photosynthesis were detected and, although the plants produced male and female flowers, all were sterile (Szarka et al. 2002). Independently, Xu et al. (2003) reported wheat–maize hybrids that contained nuclear and mitochondrial genomes of both species but plastid DNA only from wheat. These som- atic hybrids resembled wheat and, although many flowered, they were all sterile. This may have been due, at least in part, to the fact that the wheat and maize cell suspension cultures had chromosomal aberrations prior to fusion. Thus, taken as a whole, work on asexual hybridization of C3 and C4 cereals indicates that chromosomes of both photosynthetic types are stable in fused cells. However, in reports such as those from Xu et al. (2003) and Szarka et al. (2002), plants were not viable after transfer from tissue culture. Somatic hybridization of C3 and C4 species In these reports above, no F1 individual, nor any within segregating F2 and F3 populations, showed a full transfer of C4 photosynthesis. More recently, F2 individuals de- rived from a resynthesized C4 A. rosea×C3 A. prostrata cross showed large variation in leaf anatomy and nearly intermediate CO2 compensation points, but individuals in the F3 gener- ation seemed to revert to C3-like values (Oakley et al., 2014). Hybrids have also been made between C3 and C4-like species of Flaveria (Apel et al., 1988; Cameron et al., 1989) and C3–C4 intermediate and C4 Flaveria species (Brown et al., 1986, 1992). Significant F1 sterility was encountered (Brown and Bouton, 1993) but F2 were obtained and, although they possessed con- tinuous variation with regard to C4 leaf anatomy and carbon isotope discrimination characteristics, it was skewed away from the mid-parental mean towards a C3 or C3–C4 phenotype. This would indicate dominance deviation towards a C3 phenotype despite the presence of genes that allow C4 photosynthesis. In F1 hybrids derived from a C3×C4-like Flaveria cross, enzyme ac- tivities of PEPC, PPDK, and NADP-ME were skewed towards those associated with C3 photosynthesis, but C4-like activities were reported for NADP-malate dehydrogenase (Holaday et al., 1988), indicating that incomplete dominance for certain genes may exist while others show dominant activity patterns. In summary, although many C3×C4 hybrids in the dicotyle- dons showed reduced fertility and limited penetrance of C4 traits, these studies also indicate that aspects of C4 photosyn- thesis are heritable in a C3 background. As many other closely related C3 and C4 species exist (Fig. 2C), it is possible that add- itional stable hybrids could be generated that exhibit increased genomic stability and/or better trait segregation between the C3, C3–C4, and C4 types. Hybrids between different C4 decarb- oxylation subtypes may also be possible. Closely related species such as Blepharis cilaris and Blepharis attenuata that use NAD-ME and NADP-ME, respectively, have been described (Akhani et al., 2008). To our knowledge, whilst no hybrids have been re- ported in Blepharis, natural hybrids between Cynodon dactylon (NAD-ME) and Chloris sp. (PEPCK) display intermediate ac- tivities of NAD-ME and PEPCK (Prendergast, 1987). C C h b id h b t d i th b t compensation point, 178 individuals showed values similar to the C3 parent, 19 showed intermediate phenotypes, and three were similar to the C4 parent (Björkman et al., 1969). Somatic hybridization of C3 and C4 species Thus, in a small number of individuals, it appears that crossing was able to integrate loci associated with the compensation point. When F1 derived from a C4 A. rosea×C3 A. patula hy- bridization were backcrossed to C4 A. rosea, these BC1 off- spring segregated for either C4 or C3 photosynthesis, with only two individuals showing C4 photosynthesis (Rikiishi et al., 1988), suggesting dominance towards a C3 state in this hy- brid combination. In these reports above, no F1 individual, nor any within segregating F2 and F3 populations, showed a full transfer of C4 photosynthesis. More recently, F2 individuals de- rived from a resynthesized C4 A. rosea×C3 A. prostrata cross showed large variation in leaf anatomy and nearly intermediate CO2 compensation points, but individuals in the F3 gener- ation seemed to revert to C3-like values (Oakley et al., 2014). Hybrids have also been made between C3 and C4-like species of Flaveria (Apel et al., 1988; Cameron et al., 1989) and C3–C4 intermediate and C4 Flaveria species (Brown et al., 1986, 1992). Significant F1 sterility was encountered (Brown and Bouton, 1993) but F2 were obtained and, although they possessed con- tinuous variation with regard to C4 leaf anatomy and carbon isotope discrimination characteristics, it was skewed away from the mid-parental mean towards a C3 or C3–C4 phenotype. This would indicate dominance deviation towards a C3 phenotype despite the presence of genes that allow C4 photosynthesis. In F1 hybrids derived from a C3×C4-like Flaveria cross, enzyme ac- tivities of PEPC, PPDK, and NADP-ME were skewed towards those associated with C3 photosynthesis, but C4-like activities were reported for NADP-malate dehydrogenase (Holaday et al., 1988), indicating that incomplete dominance for certain genes may exist while others show dominant activity patterns. In summary, although many C3×C4 hybrids in the dicotyle- dons showed reduced fertility and limited penetrance of C4 traits, these studies also indicate that aspects of C4 photosyn- thesis are heritable in a C3 background. As many other closely related C3 and C4 species exist (Fig. 2C), it is possible that add- itional stable hybrids could be generated that exhibit increased genomic stability and/or better trait segregation between the C3, C3–C4, and C4 types. Hybrids between different C4 decarb- oxylation subtypes may also be possible. Closely related species such as Blepharis cilaris and Blepharis attenuata that use NAD-ME and NADP-ME, respectively, have been described (Akhani et al., 2008). Somatic hybridization of C3 and C4 species In contrast, sexual hybrid- ization of closely related C3 and C4 species has in some cases allowed production of fertile plants and their progeny assessed over multiple generations. We address this next. Somatic hybridization of C3 and C4 species (A) An overview of C4 biochemical subtypes. Although all forms of two-celled C4 photosynthe Fig. 1. Natural variation in C4 biochemistry and anatomy. (A) An overview of C4 biochemical subtypes. Although all forms of two-celled C4 photosynthesis involve initial CO2 fixation to generate four-carbon intermediates in mesophyll cells and diffusion to bundle sheath cells, the method of decarboxylation to create a high-CO2 environment around Rubisco varies between C4 species. Solid and dashed lines show enzymatic and diffusion steps of the C4 pathway, respectively. (B) Examples of leaf anatomies seen in C4 species. Exemplar species that use each anatomical variant are shown below each type. Many more anatomical types have been described, which suggests that multiple leaf morphologies can facilitate the C4 pathway. Abbreviations: M, mesophyll; B, bundle sheath; VB, vascular bundle; CCC, central cytoplasmic compartment; PC, peripheral chloroplast; WS, water storage cell; ch, chloroplast. Fig. 1. Natural variation in C4 biochemistry and anatomy. (A) An overview of C4 biochemical subtypes. Although all forms of two-celled C4 photosynthesis involve initial CO2 fixation to generate four-carbon intermediates in mesophyll cells and diffusion to bundle sheath cells, the method of decarboxylation to create a high-CO2 environment around Rubisco varies between C4 species. Solid and dashed lines show enzymatic and diffusion steps of the C4 pathway, respectively. (B) Examples of leaf anatomies seen in C4 species. Exemplar species that use each anatomical variant are shown below each type. Many more anatomical types have been described, which suggests that multiple leaf morphologies can facilitate the C4 pathway. Abbreviations: M, mesophyll; B, bundle sheath; VB, vascular bundle; CCC, central cytoplasmic compartment; PC, peripheral chloroplast; WS, water storage cell; ch, chloroplast C4 photosynthesis and quantitative genetics 3075 4 p y q g \| compensation point, 178 individuals showed values similar to the C3 parent, 19 showed intermediate phenotypes, and three were similar to the C4 parent (Björkman et al., 1969). Thus, in a small number of individuals, it appears that crossing was able to integrate loci associated with the compensation point. When F1 derived from a C4 A. rosea×C3 A. patula hy- bridization were backcrossed to C4 A. rosea, these BC1 off- spring segregated for either C4 or C3 photosynthesis, with only two individuals showing C4 photosynthesis (Rikiishi et al., 1988), suggesting dominance towards a C3 state in this hy- brid combination. Sexual hybridization of C3 and C4 species e Contains: C4 species C3-C4 species C3 species Cleomaceae family Contains: C4 species C3-C4 species C3 species Potential breeding systems S. spathellosum P. milioids F. ﬂoridana F. chloraefolia F. ramosissima F. linearis F. anomala ailo fitis o p p o .F g. 2. Examples of successful as well as potential hybridizations between C3 and C4 species (A) Phylogenetic reconstruction of the orders constitu B B C A Panicum/ m Steinchisma F1 hybrids C S. sp P. Potential breeding systems Lamiids C4 Fig. 2. Examples of successful as well as potential hybridizations between C3 and C4 species. (A) Phylogenetic reconstruction of the orders constituting flowering plants according to The Angiosperm Phylogeny Group (2016). Orders containing C4 lineages are shown in bold. (B) Exemplar hybridization webs that have resulted in successful F1 hybrids between C3, C4, and C3–C4 intermediate photosynthetic types. (C) Taxa that contain closely related C3, C4, or C3–C4 intermediate species or accessions for which hybridization has not been reported, but may be possible. These groups are potential systems where C4 genes could be mapped. Arrows from the phylogenetic tree indicate from which order the plant species originate (B, C). Fig. 2. Examples of successful as well as potential hybridizations between C3 and C4 species. (A) Phylogenetic reconstruction of the orders constituting flowering plants according to The Angiosperm Phylogeny Group (2016). Orders containing C4 lineages are shown in bold. (B) Exemplar hybridization webs that have resulted in successful F1 hybrids between C3, C4, and C3–C4 intermediate photosynthetic types. (C) Taxa that contain closely related C3, C4, or C3–C4 intermediate species or accessions for which hybridization has not been reported, but may be possible. These groups are potential systems where C4 genes could be mapped. Arrows from the phylogenetic tree indicate from which order the plant species originate (B, C). attempts at wide hybridization have also been reported. For example, although maize pollen germinates and fertilizes the ovule of wheat to form zygotes containing a full haploid set of each parental genome (Laurie and Bennett, 1986), these hybrids were unstable and after three rounds of mitotic cell divisions during embryogenesis all maize chromosomes were lost (Laurie and Bennett, 1986, 1989). Sexual hybridization of C3 and C4 species A number of taxa containing either congeneric C3 and C4 species or C3, C3–C4 intermediates, and C4 species have been successfully hybridized (Fig. 2A, B). Although the outcome of these analyses varied, whilst wholesale transfer of C4 traits have not been reported in some instances, specific traits were introgressed into a C3 background. For example, crosses be- tween C4 Atriplex rosea and C3 Atriplex prostrata (formerly A. patula ssp. hastata and A. triangularis, respectively), C3 A. rosea and C3 A. glabriuscula have been made (Björkman et al., 1969; Nobs et al., 1970). Populations derived from such crosses were progressed and C4-like characteristics assessed (Björkmann et al., 1971). Among 200 F3 individuals screened for the CO2 C3–C4 hybrids have been generated in the grasses by two broad approaches. First, as with dicotyledons, congeners using either C3 or C3–C4 photosynthesis have been crossed. Second, much wider crosses of distantly related species have been per- formed. Examples of crosses within a genus include C3 and C3–C4 intermediate Steinchisma (formally Panicum) species 3076 \| Simpson et al. 3076 F. australasica F. bidentis F. campestris F. palmeri F. trinervia F. pringlei A. rosea A. lindleyi subsp. infata A. patula A. glabriuscula A. prostrata S. laxa P. boliviense Oryza sativa P. maximum S. hians C3 C3-C4 C4 C4-like F. brownii Basal angiosperms Magnoliids g Arecales g Poales* Commelinales oa es Zingiberales Asparagales g Liliales p Dioscoreales Pandanales Petrosaviales Alismatales* Acorales Ceratophyllales Ranunculales p y p y Proteales Trochodendrales Buxales Gunnerales Fabales Rosales Fagales Cucurbitales g Oxalidales Malpighiales* Celastrales p g p g Zygophyllales* Geraniales yg p y Myrtales Crossosomatales y Picramniales Malvales Brassicales* Huerteales Sapindales ue ea es Vitales Saxifragales Dilleniales g Berberidopsidales Santalales p Caryophyllales* Cornales y p y Ericales Aquifoliales Asterale q s* Escalloniales Bruniales Apiales Dipsacales p Paracryphiales p Solanale yp yp s* Lamiales* Vahliales Gentianales Boraginales Garryales g Metteniusales y Icaninales MONOCOTS EUDICOTS Fabids Malvids Alismatids Liloids Commelinids Campanulids Lamiids Rosids Superrosids Superasterids Asterids Flaveria r F a 1 hybrids C3 C4 Atriplex F x 1 hybrids C3 C4 Panicum/ m Steinchisma F1 hybrids C3-C4 Pr ot o pl a st f u si o n Botanical Order or Group A B *Order contains C4 species C Blepharis sp. Contains: C4 species C4-like species C3 species Alloteropsis se r mialata subs Contains: C4 accessions C3-C4 accessions C3 accessions Heliotrt opiu r m sp. Contains: C4 species C3-C4 species Neura r chne sp. Sexual hybridization of C3 and C4 species Thus, wheat–pearl millet hybrids may be more stable than wheat–maize hybrids, but problems maintaining chromosomes from both parents still appear to exist. Unlike wheat–maize hybrids, maize chromosomes have successfully been integrated into oat. This allowed the synthesis of so-called oat–maize chromosome addition lines that stably inherit single chromosome pairs from maize (Kynast et al., 2001, 2004). As with the pearl millet–oat crosses (Ishii et al., 2015), stability of the oat–maize addition lines appears to be mediated by incorporation of centromeric oat histones into the maize chromosomes such that proper chromosomal seg- regation can take place during mitosis (Jin et al., 2004; Wang et al., 2014). In some maize–oat lines, C4 characteristics such as abundant transcripts of PEPC or C4-like bundle sheath cell size and vein spacing were detected (Tolley et al., 2012). (Ishii et al., 2013). It appears that the pearl millet chromosomes had incorporated centromeric oat histones (Ishii et al., 2015), but these haploid oat–millet F1 hybrids developed necrosis and died. This may have been caused by incompatibility between the species or non-ideal tissue culture conditions. Crosses be- tween wheat and grain pearl millet (Pennisetum americanum) or oat and maize both allowed individual chromosomes from one species to be incorporated into the other. In the case of wheat and grain pearl millet from 958 hybridizations, one wheat plant carrying an additional pearl millet chromosome was identified (Ahmad and Comeau, 1990). Although this chromosome was maintained until flowering, it was not de- tected in the next generation. Thus, wheat–pearl millet hybrids may be more stable than wheat–maize hybrids, but problems maintaining chromosomes from both parents still appear to exist. Unlike wheat–maize hybrids, maize chromosomes have successfully been integrated into oat. This allowed the synthesis of so-called oat–maize chromosome addition lines that stably inherit single chromosome pairs from maize (Kynast et al., 2001, 2004). As with the pearl millet–oat crosses (Ishii et al., 2015), stability of the oat–maize addition lines appears to be mediated by incorporation of centromeric oat histones into the maize chromosomes such that proper chromosomal seg- regation can take place during mitosis (Jin et al., 2004; Wang et al., 2014). In some maize–oat lines, C4 characteristics such as abundant transcripts of PEPC or C4-like bundle sheath cell size and vein spacing were detected (Tolley et al., 2012). Sexual hybridization of C3 and C4 species pi yi While bundle sheath cells are always greener in C4 com- pared with C3 species, the proportion of leaf tissue allocated to bundle sheath cells compared with the mesophyll cells can be caused by either increased bundle sheath cell size or vein density (Sedelnikova et al., 2018). Interestingly, within G. gynandra, these characteristics co-vary and correlate negatively with one another (Reeves et al., 2018). In addition to variation in Kranz anatomy in a species, there is also evidence that the cell biology of C4 leaves can differ. For example, some acces- sions of Panicum coloratum possess a suberized bundle sheath whilst others do not (Ohsugi and Murata, 1985). There is also variation in chloroplast organization, with some accessions arranging chloroplasts centrifugally and others centripetally compared with veins (Ohsugi and Murata, 1985). Interestingly, Cynodon dactylon, an NAD-ME subtype with centripetal chloroplasts and a suberized bundle sheath, hybridizes nat- urally with Chloris that uses PEPCK as the primary C4 acid decarboxylase, has centrifugally arranged chloroplasts, and no suberization of the bundle sheath (Prendergast, 1987). F1s demonstrated intermediacy for these traits (Prendergast, 1987). Thus, these species offer an interesting system to study regu- lators of bundle sheath cell biology. In summary, the findings based on wide hybridization of maize and oat indicate that breeding offers a possible route to incorporate some C4 traits into C3 crops without prior know- ledge of the underlying genetics. Although additional parental combinations may exist that allow greater trait stability in pro- geny, this approach has not yet allowed loci controlling C4 traits to be identified. In contrast, quantitative variation in C4 char- acteristics within a C4 species would allow trait mapping, and there is increasing evidence that this could be informative. To our knowledge, there are no clear examples of quantita- tive variation in the extent to which accessions of an individual C4 species use the various C4 acid decarboxylases. However, there are two reasons to consider this likely. First, in 26 founder lines of a maize multiparent population, variation in the activ- ities of C4 enzymes has been reported (McMullen et al., 2009; Kolbe et al., 2018). As the founders show differences in enzyme activity, it is likely that lines of the mapping population pos- sess similar variation. Accessions of A. semialata (Dunning et al., 2017) and G. Sexual hybridization of C3 and C4 species gynandra (Reeves et al., 2018) demonstrate differ- ences in transcript abundance and so it appears likely that these species will also demonstrate variation in activity of C4 acid decarboxylases. Second, the extent to which the different C4 acid decarboxylases are engaged can vary with the environment. For example, in G. gynandra and maize, increased abundance of transcripts encoding C4 enzymes did not correlate with photosynthetic efficiency (Kolbe and Cousins, 2018; Reeves et al., 2018) but in G. gynandra they were associated with in- creased water use efficiency. Additionally, the PEPCK subtype Sexual hybridization of C3 and C4 species In contrast, after hybrid- ization of oat and pearl millet (Pennisetum glaucum) (Gernand et al., 2005; Ishii et al., 2010), some oat embryos contained all pearl millet chromosomes, and embryo rescue allowed hybrids possessing the haploid genomes of both species to be obtained from the Poaceae (Bouton et al., 1986; Brown et al., 1986; Sternberg et al., 1986). F2 and F5 individuals derived from hy- bridization of Steinchisma milioides (C3–C4) and Steinchisma laxum (C3), or S. spathellosum (C3–C4) and S. boliviense (C3) ex- hibited intermediate leaf morphologies, CO2 compensation points, and δ13C values. Also within the Poaceae, C3 and C4 accessions of Alloteropsis semialata have been hybridized, pro- ducing plants with intermediate anatomical traits as well as C4 gene expression (Bianconi et al., 2021, Preprint). Thus, in these hybridizations, some traits important for C4 photosynthesis could be introduced into an otherwise C3 leaf. A variety of C4 photosynthesis and quantitative genetics 3077 Thus, natural variation in Kranz anatomy is found within spe- cies of C4 monocotyledons and dicotyledons. Statistical model- ling suggests that evolution of enlarged bundle sheath cells and vein density were among the first changes to occur during the transition from C3 to C4 photosynthesis (Williams et al., 2013), and phylogenetic reconstructions reveal that these changes probably happened in response to reduced water availability (Edwards and Smith, 2010). As bundle sheath cell size and vein density were found to be correlated with water use efficiency in maize (Yabiku and Ueno, 2017) and G. gynandra (Reeves et al., 2018), it is possible that analysis of C4 accessions adapted to different water availabilities will allow additional examples of intraspecific variation in Kranz anatomy to be identified. (Ishii et al., 2013). It appears that the pearl millet chromosomes had incorporated centromeric oat histones (Ishii et al., 2015), but these haploid oat–millet F1 hybrids developed necrosis and died. This may have been caused by incompatibility between the species or non-ideal tissue culture conditions. Crosses be- tween wheat and grain pearl millet (Pennisetum americanum) or oat and maize both allowed individual chromosomes from one species to be incorporated into the other. In the case of wheat and grain pearl millet from 958 hybridizations, one wheat plant carrying an additional pearl millet chromosome was identified (Ahmad and Comeau, 1990). Although this chromosome was maintained until flowering, it was not de- tected in the next generation. Intraspecific variation in C4 photosynthesis As PEPC discriminates less than Rubisco against the 13C isotope, a stronger C4 cycle leads to lower incorporation of 13C into tissue and so less negative δ13C values (Leary, 1988). Intraspecific variation in δ13C has been reported in maize and Gynandropsis gynandra (Voznesenskaya et al., 2007; Kolbe and Cousins, 2018; Kolbe et al., 2018; Reeves et al., 2018; Twohey et al., 2019). To our knowledge, the extent to which this variation in C4 efficiency is caused by differences in Kranz anatomy, cell biology, or C4 biochemistry has not been determined but, as summarized next, variation in some of these traits within a species has been reported. This includes variation in vein density in maize (Yabiku and Ueno, 2017; Kolbe and Cousins, 2018) as well as bundle sheath cell size in Alloteropsis semialata (Lundgren et al., 2016) and G. gynandra (Reeves et al., 2018). Simpson et al. 3078 is considered more efficient under lower levels of light since it theoretically requires fewer quanta of light per CO2 mol- ecule fixed (Furbank, 2011; Yin and Struik, 2020). Consistent with this, sugarcane (Saccharum offiniarum) and maize which predominantly use NADP-ME showed lower and higher ac- tivities of NADP-ME and PEPCK, respectively, after either shade or salt stress (Omoto et al., 2012; Sharwood et al., 2014; Sales et al., 2018). Increased CO2 leakage from bundle sheath cells has also been reported, and it has been proposed that this is caused by increased use of cytosolic PEPCK compared with the chloroplastic NADP-ME (Sales et al., 2018). If populations of these species have become reproductively isolated in habitats with distinct light supplies, differences in subtype preference may have evolved. Thus, C4 traits ranging from discrimination against δ13C, C4 leaf anatomy, bundle sheath cell biology, and C4 transcript abundance have been documented within a spe- cies. In each case, breeding and quantitative genetics offer an opportunity to identify loci controlling these traits. Within this context, we next assess opportunities associated with quantita- tive genetics to better understand C4 photosynthesis. specifically for the investigation of C4 traits. High-throughput phenotyping combined with the convoluted neural net- work Mask R-CNN (He et al., 2017) has been used for QTL mapping of C4-relevant traits in biparental populations. This allowed rapid assessment of thousands of images and identifi- cation of QTL for stomatal traits such as size and density (Xie et al., 2021). Intraspecific variation in C4 photosynthesis Although QTL mapping is used extensively, its power is limited if the trait is responsive to the environment and so has low heritability. The heritability of many C4 traits re- mains poorly understood, but there is growing evidence that variations in CO2 fixation processes and leaf anatomy exist (Table 1) and so estimates of heritability of such C4 traits should be possible. Given the complexity of photosynthesis, its ability to respond to the environment, and temporal vari- ation in its efficiency, it is highly likely that low-heritability traits will be encountered (Flood et al., 2016). Although traits with low heritability can be investigated using highly con- trolled environments, highly inbred populations in com- bination with high-density marker systems are necessary to capture the multiple small-effect QTL contributing to the low-heritability trait of interest. An alternative approach in- volves genome-wide association studies (GWAS) or linkage disequilibrium (LD) mapping, which identifies markers such as single nucleotide polymorphisms (SNPs) that are in LD with the phenotype of interest (Tam et al., 2019). GWAS does not require a segregating population but rather uses many diverse accessions that represent thousands of years of recom- bination to capture multiple alleles, allowing marker groups (haplotypes) to be identified in close association with causal loci. Additionally, it has the advantage of being feasible for obligate outcrossers. In order to work successfully, GWAS re- quires many markers since it relies on LD decay (Mackay and Powell, 2007) and, as pedigrees are unknown, physical maps are also needed. Although population structure increases the number of false positives derived from GWAS (Korte and Farlow, 2013), this is increasingly being overcome by stat- istical modelling (Cortes et al., 2021). GWAS has identified QTL associated with photosynthetic performance during chilling in maize (Strigens et al., 2013) and sorghum (Ortiz et al., 2017). More recently, a sorghum diversity panel of 756 African accessions was described (Faye et al., 2021) and a di- verse 869 line panel (Valluru et al., 2019) was subjected to GWAS to identify genes controlling stomatal conductance and water use efficiency (Ferguson et al., 2021; Pignon et al., 2021). The latter two studies used transcriptome data to allow transcriptome-wide association as well as GWAS (re- viewed by Wainberg et al., 2019) to increase the likelihood of identifying candidate genes. Intraspecific variation in C4 photosynthesis Association mapping has also been used to study the light-dependent reactions of photo- synthesis (van Bezouw et al., 2019) but, to our knowledge, QTL determining differences in C4 carbon fixation or Kranz anatomy have not yet been identified. The sorghum and maize mapping panels present an avenue through which Quantitative genetics and C4 photosynthesis Quantitative genetics allow traits exhibiting continuous vari- ation to be linked to genomic regions termed quantitative trait loci (QTL). Advances in high-throughput phenotyping rele- vant to photosynthetic performance (reviewed by Choudhury et al., 2019; van Bezouw et al., 2019) mean that quantitative genetics now offers a path to dissect the genetics underlying photosynthesis. Traditional QTL mapping requires a linkage map (or gen- etic map) to order loci. Using a population derived from two parents that differ in a trait of interest, associations between the trait and molecular markers can identify genes in close prox- imity to the trait (Mauricio, 2001). Advantages of QTL map- ping are that limited knowledge of the genome is necessary and producing bi-parental populations is relatively rapid (Fig. 3A). Recombinant inbred lines (RILs) can be produced, for example, from a segregating F2 generation through rounds of self-fertilization and so generate an immortalized population that can be genotyped once but phenotyped repeatedly. This is especially useful for heritability estimates and mapping QTL in different environments or years (Broman, 2005). Due to con- siderable differences in the biochemistry and physiology of C3 and C4 plants, if mapping populations derived from C3 and C4 parents of Atriplex, Alloteropsis, or Flaveria were generated, QTL mapping could probably associate genes with a wide variety of C4 phenotypes. Alloteropsis semialata could be of particular interest here because of the presence of both C3 and C4 sub- species that hybridize to produce offspring with intermediate characteristics (Bianconi et al., 2021, Preprint). As self-fertiliza- tion is also possible, a population of RILs could be designed C4 photosynthesis and quantitative genetics \| 3079 C4 photosynthesis and quantitative genetics \| 3079 3079 Fig. 3. Quantitative genetics in the context of C4 photosynthesis. (A) A schematic for QTL mapping of leaf anatomical traits. Two homozygous parents, genotyped for four markers, A, B, C, and D, and differing in vein density are hybridized and advanced to form a bi-parental population that can be used to identify QTL associated with vein density (here located near markers C and D). Numbers show recombination fractions, which are used to position the QTL relative to flanking markers. (B) Population structure of a MAGIC pedigree followed by four generations of intercrossing and self-fertilization. Progeny contain more genetic variation than that derived from a bi-parental design. Quantitative genetics and C4 photosynthesis Hypothetical plot showing how QTL associated with individually mapped C4 phenotypes such as gene expression, bundle sheath cell size, or gas exchange parameters (e.g. stomatal conductance, CO2 assimilation, etc,) can be mapped with one population. matic for QTL mapping of leaf anatomical traits. Two homozygous parents, ybridized and advanced to form a bi-parental population that can be used d D). Numbers show recombination fractions, which are used to position the e followed by four generations of intercrossing and self-fertilization. Progeny ypothetical plot showing how QTL associated with individually mapped C4 ( l d CO i il i ) b Fig. 3. Quantitative genetics in the context of C4 photosynthesis. (A) A schematic for QTL mapping of leaf anatomical traits. Two homozygous parents, genotyped for four markers, A, B, C, and D, and differing in vein density are hybridized and advanced to form a bi-parental population that can be used to identify QTL associated with vein density (here located near markers C and D). Numbers show recombination fractions, which are used to position the QTL relative to flanking markers. (B) Population structure of a MAGIC pedigree followed by four generations of intercrossing and self-fertilization. Progeny contain more genetic variation than that derived from a bi-parental design. Hypothetical plot showing how QTL associated with individually mapped C4 phenotypes such as gene expression, bundle sheath cell size, or gas exchange parameters (e.g. stomatal conductance, CO2 assimilation, etc,) can be mapped with one population. Fig. 3. Quantitative genetics in the context of C4 photosynthesis. (A) A schematic for QTL mapping of leaf anatomical traits. Two homozygous parents, genotyped for four markers, A, B, C, and D, and differing in vein density are hybridized and advanced to form a bi-parental population that can be used to identify QTL associated with vein density (here located near markers C and D). Numbers show recombination fractions, which are used to position the QTL relative to flanking markers. (B) Population structure of a MAGIC pedigree followed by four generations of intercrossing and self-fertilization. Progeny contain more genetic variation than that derived from a bi-parental design. Hypothetical plot showing how QTL associated with individually mapped C4 phenotypes such as gene expression, bundle sheath cell size, or gas exchange parameters (e.g. stomatal conductance, CO2 assimilation, etc,) can be mapped with one population. resolution mapping. These include nested-association mapping (NAM) and multiparent advanced generation inter-crossing (MAGIC) population designs. Quantitative genetics and C4 photosynthesis Both address issues with GWAS and capture more allelic variation than bi-parental popula- tions. Whilst allelic diversity is reduced in these multiparent designs compared with GWAS, linkage mapping as well as as- sociation mapping are possible, and this is particularly useful when a physical map is not available (Broman et al., 2018). Thus, NAM and MAGIC are currently particularly relevant targeted phenotyping of C4-specific traits could be used to identify genes responsible for the C4 syndrome. For example, if a gene controlling bundle sheath cell size was identified through mapping in maize or sorghum, this could then be introduced in a C3 crop such as rice to determine whether this allowed engineering of this trait.i Association mapping can be combined with specific breeding pedigrees to capture multiple recombination events, account for population structure, and so allow higher 3080 \| Simpson et al. 3080 3080 Table 1. Summary of publications documenting intraspecific variation in traits relevant to C4 photosynthesis-associated traits Species Varying trait Reference Alloteropsis semialata (C4 accessions) Abundance of PEPC and PEPCK transcripts Dunning et al. (2017) PEPC content Carbon isotope discrimination Mesophyll cell size Bundle sheath cell size Leaf physiology Lundgren et al. (2016) Gynandropsis gynandra C4 transcript abundance, physiology, and leaf morphology Reeves et al. (2018) Panicum coloratum Chloroplast location Bundle sheath suberization Ohsugi and Murata (1985) Setaria italica Carbon isotope Differing intensities of green’ Lightfoot et al. (2016) Sorghum bicolor Net assimilation rate Kataria and Guruprasad (2012) Zea mays CA transcript abundance Zhang et al. (2015) Zea mays CA, PEPC, and Rubisco activity Net assimilation rate Interveinal distance Mesophyll thickness Maximum assimilation rate Kolbe and Cousins (2018) CA, PEPC, and Rubisco activity C4 transcript abundance Carbon isotope Kolbe et al. (2018) Vein density Gas exchange traits PEPC, NADP-ME, PEPCK, and Rubisco activity Yabiku and Ueno (2017) CA, carbonic anhydrase, NADP-ME; NADP-dependent malic enzyme; PEPC, phosphoenolpyruvate carboxylase; PEPCK, phosphoenolpyruvate carboxykinase. Reeves et al. (2018) Ohsugi and Murata (1985) Lightfoot et al. (2016) Kataria and Guruprasad (2012) Zhang et al. (2015) Kolbe and Cousins (2018) Gynandropsis gynandra Panicum coloratum Setaria italica Sorghum bicolor Zea mays Zea mays Kolbe et al. (2018) Yabiku and Ueno (2017) Kolbe et al. (2018) Kolbe et al. (2018) Yabiku and Ueno (2017) The MAGIC design also relies on homozygous founder lines that differ in traits of interest. Quantitative genetics and C4 photosynthesis Intercrossing for multiple generations allows segregating populations to be formed con- sisting of lines that capture the founder genomes in unique recombinants (Fig. 3B). Such segregating lines then undergo self-fertilization for several generations to generate RILs that capture multiple allele combinations from the various parents (Cavanagh et al., 2008). With MAGIC, haplotype diversity is not limited by the use of a single recurrent parent (Ladejobi et al., 2016) and, although the MAGIC design requires large amounts of hybridization and significant time to produce the final population (Huang et al., 2015; Pascual et al., 2015; Ongom and Ejeta, 2017; Mahan et al., 2018), simplified strat- egies can be implemented (Stadlmeier et al., 2018). In the con- text of C4 photosynthesis, MAGIC RILs are available for maize and sorghum (Dell’Acqua et al., 2015; Ongom and Ejeta, 2017; Mahan et al., 2018; Butrón et al., 2019). Additionally, transcrip- tome data exist for the founders of one maize MAGIC popu- lation (Dell’Acqua et al., 2015) and 94 of the MAGIC RILs (Baute et al., 2016). Should these RILs possess variation in ac- tivity of C4 enzymes or components of Kranz anatomy, QTL could be identified. To our knowledge, there is currently no MAGIC population available for a C4 dicotyledon, nor a map- ping panel designed explicitly to map C4 photosynthetic traits. As variation in C4 traits has been reported in A. semialata and for C4 photosynthesis because although annotated genome sequences are being developed for, for example, Alloteropsis sp., Flaveria sp., and G. gynandra, complete and well-annotated gen- omes for many C4 model species have not yet been developed. The NAM design involves crossing one recurrent parent with many other accessions. Progeny from each cross are initially bulked and then self-fertilized for multiple generations, leading to multiple RIL families (one family per unique founder) that then constitute the final NAM population (Yu et al., 2008; McMullen et al., 2009). At least two NAM populations exist for maize (Yu et al., 2008; Chen et al., 2019) and, as mentioned above, significant variation for δ13C as well as CA, PEPC, and Rubisco activities has been reported in the founder lines (Zhang et al., 2015; Kolbe et al., 2018; Twohey et al., 2019). Despite this, QTL for these traits have to our knowledge not yet been determined. Quantitative genetics and C4 photosynthesis A sorghum NAM population has been used in conjunction with an association panel to identify QTL for grain filling (Tao et al., 2020). NAM populations offer the chance to study an extremely divergent line, such as a pre- domesticated species in the background of a stable population. This has been done with teosinte and maize as the recurrent parent (Chen et al., 2019). Given the noted differences in maize and teosinte photosynthetic capacity (Yabiku and Ueno, 2017), this offers an interesting resource to map traits that differ be- tween these species. C4 photosynthesis and quantitative genetics 3081 G. gynandra (Lundgren et al., 2016; Reeves et al., 2018) and they can be crossed (Sogbohossou et al., 2018; Bianconi et al., 2020), mapping resources in these species would be useful. G. gynandra (Lundgren et al., 2016; Reeves et al., 2018) and they can be crossed (Sogbohossou et al., 2018; Bianconi et al., 2020), mapping resources in these species would be useful. a combination of both mechanisms, and whether these mech- anisms are functioning in an activating or repressive manner (Fig. 4). This technique has been deployed in F1 hybrids de- rived from a cross between the C3–C4 intermediate Moricandia arvensis and the C3 M. moricandiodes to show that cis-regulation dominates control of photosynthetic and anatomical pheno- types (Lin et al., 2021, Preprint). Information from such studies could inform mapping strategies and marker placement for associations. Once a QTL is identified using any of the above popula- tion types, fine mapping enables causative genes to be iden- tified (Hormozdiari et al., 2014; Tam et al., 2019). Parsing C4 photosynthesis into individual components, such genes con- trolling C4 enzyme activity or bundle sheath cell size (Dunning et al., 2017) are identified by different phenotyping techniques which, combined with fine mapping, could identify additional genes required for C4 photosynthesis. Exploiting the high de- gree of natural variation among C3 and C4 species will enable genome-wide associations to help map critical photosynthesis regulators. Furthermore, inferences into the inheritance of C4 components such as cell-specific gene expression can be parsed even without proper segregation or recombination in C3 and C4 hybrids (Fig. 4). While such methods cannot iden- tify QTL, they can at least establish broad modes of inherit- ance (Charlesworth and Willis, 2009). Quantitative genetics and C4 photosynthesis For example, sterile F1 populations derived from C3 and C4 parents that show altered transcript abundance or cellular localization of C4 enzymes can provide insight into whether genes are controlled in cis, trans, or In summary, in order to modify C3 leaves to perform C4 photosynthesis, an improved understanding of C4 anatomy, cell biology, and biochemistry is needed. Wide hybridization by either sexual or asexual means to recombine interspecific variation found in C3 and C4 species or intraspecific photo- synthetic variation in C4 species, combined with mapping populations and high-throughput phenotyping, should facili- tate a better understanding of C4 photosynthesis. Quantitative genetics then offer robust methods to better understand the regulatory mechanisms behind these traits. Applying these techniques therefore promises to enhance photosynthetic effi- ciency of C3 and C4 crops and so contribute to a more robust world agriculture in the future. Fig. 4. Using breeding to understand the molecular basis of C4 gene regulation. Parental populations that differ in transcript abundance can be due to multiple genetic effects that can be parsed by quantitative genetics. A simplified two loci model where one locus is a cis-element and the other an activating trans-factor is presented to illustrate how the molecular basis underpinning variations in gene expression can be determined by inheritance gene expression in F1 hybrids. If expression of a gene is controlled by changes in cis-regulation between parents, offspring exhibit additive expression patterns. If variation in expression is due to changes in trans between parents, then offspring exhibit dominance deviation towards one parent. Lastly, differences in gene expression between parents is due to both cis and trans factors, offspring demonstrate heterosis or overdominance. Fig. 4. Using breeding to understand the molecular basis of C4 gene regulation. Parental populations that differ in transcript abundance can be due to multiple genetic effects that can be parsed by quantitative genetics. A simplified two loci model where one locus is a cis-element and the other an activating trans-factor is presented to illustrate how the molecular basis underpinning variations in gene expression can be determined by inheritance of gene expression in F1 hybrids. If expression of a gene is controlled by changes in cis-regulation between parents, offspring exhibit additive expression patterns. If variation in expression is due to changes in trans between parents, then offspring exhibit dominance deviation towards one parent. References Ahmad F, Comeau A. 1990. Wheat × pearl millet hybridization: conse- quence and potential. Euphytica 50, 181–190. Cavanagh C, Morell M, Mackay I, Powell W. 2008. From mutations to MAGIC: resources for gene discovery, validation and delivery in crop plants. Current Opinion in Plant Biology 11, 215–221. Akhani H, Ghasemkhani M, Chuong SDX, Edwards GE. 2008. Occurrence and forms of Kranz anatomy in photosynthetic organs and characterization of NAD-ME subtype C4 photosynthesis in Blepharis ciliaris (L.) B. L. Burtt (Acanthaceae). Journal of Experimental Botany 59, 1755–1765. Charlesworth D, Willis JH. 2009. The genetics of inbreeding depression. Nature Reviews. Genetics 10, 783–796. Chen Q, Yang CJ, York AM, et al. 2019. TeoNAM: a nested associ- ation mapping population for domestication and agronomic trait analysis in maize. Genetics 213, 1065–1078. Apel P, Bauwe H, Bassüner B, Maass I. 1988. Photosynthetic properties of Flaveria cronquistii, F. palmeri, and hybrids between them. Biochemie und Physiologie der Pflanzen 183, 291–299. Choudhury SD, Samal A, Awada T. 2019. Leveraging image analysis for high-throughput. Plant Phenotyping 10, 1–8. Baute J, Herman D, Coppens F, Block J De, Slabbinck B, Acqua MD, Pè ME, Maere S, Nelissen H, Inzé D. 2016. Combined large-scale phenotyping and transcriptomics in maize reveals a robust growth regula- tory network. Plant Physiology 170, 1848–1867. Cortes LT, Zhang Z, Yu J. 2021. Status and prospects of genome-wide association studies in plants. Plant Genome 14, 1–17. Danila FR, Quick WP, White RG, Furbank RT. 2016. The metab- olite pathway between bundle sheath and mesophyll: quantification of plasmodesmata in leaves of C3 and C4 monocots. The Plant Cell 28, 1461–1471. Bianconi ME, Dunning LT, Curran EV, et al. 2020. Contrasted histories of organelle and nuclear genomes underlying physiological diversification in a grass species. Proceedings of the Royal Society B: Biological Sciences 287, 20201960. Dell’Acqua M, Gatti DM, Pea G, et al. 2015. Genetic properties of the MAGIC maize population: a new platform for high definition QTL mapping in Zea mays. Genome Biology 16, 1–23. Bianconi ME, Sotelo G, Curran EV, Milenkovic V, Samaritani E, Dunning LT, Osborne CP, Christin P. 2021. Upregulation of C4 char- acteristics does not consistently improve photosynthetic performance in intraspecific hybrids of a grass. bioRxiv. doi:10.1101/2021.08.10.455822. [Preprint]. Dunning LT, Lundgren MR, Moreno-Villena JJ, Namaganda M, Edwards EJ, Nosil P, Osborne CP, Christin PA. 2017. Introgression and repeated co-option facilitated the recurrent emergence of C4 photosyn- thesis among close relatives. Evolution 71, 1541–1555. Quantitative genetics and C4 photosynthesis Lastly, if differences in gene expression between parents is due to both cis and trans factors, offspring demonstrate heterosis or overdominance. Fig. 4. Using breeding to understand the molecular basis of C4 gene regulation. Parental populations that differ in transcript abundance can be due to multiple genetic effects that can be parsed by quantitative genetics. A simplified two loci model where one locus is a cis-element and the other an activating trans-factor is presented to illustrate how the molecular basis underpinning variations in gene expression can be determined by inheritance of gene expression in F1 hybrids. If expression of a gene is controlled by changes in cis-regulation between parents, offspring exhibit additive expression patterns. If variation in expression is due to changes in trans between parents, then offspring exhibit dominance deviation towards one parent. Lastly, if differences in gene expression between parents is due to both cis and trans factors, offspring demonstrate heterosis or overdominance. 3082 \| Simpson et al. with high-dimensional data and multiparent populations high-dimensional data and multiparent populations. Genetics 211, 495–502. with high-dimensional data and multiparent populations high-dimensional data and multiparent populations. Genetics 211, 495–502. Author contributions All authors contributed to the analysis of literature and writing of this review. Brown HR, Bouton JH. 1993. Interspecific hybrids between photosyn- thetic types. Annual Review of Plant Physiology 44, 435–436. Brown RH, Bassett CL, Cameron RG, Evans PT, Bouton JH, Black CC, Sternberg LO, Deniro MJ. 1986. Photosynthesis of F1 hybrids be- tween C4 and C3–C4 species of Flaveria. Plant Physiology 82, 211–217. Funding SCJS was supported by a BBSRC PhD studentship; GR by a Gates Cambridge Trust PhD Student Fellowship; GR, PS and AT by ERC Grant 694733 Revolution and BB/P003117/1 awarded to JMH; and GR and AT by a Ceres Agri-Tech Fund award. Calvin M, Benson AA. 1948. The path of carbon in photosynthesis. Encyclopedia of the Environment 107, 476–480. Cameron RG, Bassett CL, Bouton JH, Brown RH. 1989. Transfer of C4 photosynthetic characters through hybridization of Flaveria species. Plant Physiology 90, 1538–1545. Carlson PS, Smith HH, Dearing R. 1972. Parasexual interspecific plant hybridization. Proceedings of the National Academy of Sciences, USA 69, 2292–2294. The authors declare no conflicts of interest. The authors declare no conflicts of interest. Butrón A, Santiago R, Cao A, Samayoa LF, Malvar RA. 2019. QTLs for resistance to Fusarium ear rot in a Multiparent Advanced Generation Intercross (MAGIC) maize population. Plant Disease 109, 897–904. Conflict of interest Brown RH, Byrd GT, Black CC. 1992. Degree of C4 photosynthesis in C4 and C3–C4 Flaveria species and their hybrids: II. Inhibition of apparent photosynthesis by a phosphoenolpyruvate carboxylase inhibitor. Plant Physiology 100, 947–950. C4 photosynthesis and quantitative genetics Trends in Plant Science 12, 57–63. Mahan AL, Murray SC, Klein PE. 2018. Four-Parent Maize (FPM) popu- lation: development and phenotypic characterization. Crop Science 58, 1106–1117. Ishii T, Ueda T, Tanaka H, Tsujimoto H. 2010. Chromosome elimin- ation by wide hybridization between Triticeae or oat plant and pearl millet: pearl millet chromosome dynamics in hybrid embryo cells. Chromosome Research 18, 821–831. Mauricio R. 2001. Mapping quantitative trait loci in plants: uses and cav- eats for evolutionary biology. Nature Reviews. Genetics 2, 370–381. Ishii T, Tanaka H, Eltayeb AE, Tsujimoto H. 2013. Wide hybridization between oat and pearl millet belonging to different subfamilies of Poaceae. Plant Reproduction 26, 25–32. McMullen MD, Kresovich S, Villeda HS, et al. 2009. Genetic proper- ties of the maize nested association mapping population. Science 325, 737–740. Ishii T, Sunamura N, Matsumoto A, Eltayeb AE, Tsujimoto H. 2015. Preferential recruitment of the maternal centromere-specific histone H3 (CENH3) in oat (Avena sativa L.) × pearl millet (Pennisetum glaucum L.) hy- brid embryos. Chromosome Research 23, 709–718. Mertz RA, Brutnell TP. 2014. Bundle sheath suberization in grass leaves: multiple barriers to characterization. Journal of Experimental Botany 65, 3371–3380. Muhaidat R, Sage RF, Dengler NG. 2007. Diversity of Kranz anatomy and biochemistry in C4 eudicots. American Journal of Botany 94, 362–381. Jin W, Melo JR, Nagaki K, Talbert PB, Henikoff S, Dawe RK, Jiang J. 2004. Maize centromeres: organization and functional adaptation in the genetic background of oat. The Plant Cell 16, 571–581. Nobs MA, Björkmann O, Pearcy RW, Boynton JE. 1970. Hybrids be- tween Atriplex species with and without beta-carboxylation photosynthesis. Carnegie Institution of Washington Yearbook 69, 617–662. Kadereit G, Borsch T, Weising K, Freitag H. 2003. Phylogeny of Amaranthaceae and Chenopodiaceae and the evolution of C4 photosyn- thesis. International Journal of Plant Sciences 164, 959–986. Oakley JC, Sultmanis S, Stinson CR, Sage TL, Sage RF. 2014. Comparative studies of C3 and C4 Atriplex hybrids in the genomics era: physiological assessments. Journal of Experimental Botany 65, 3637–3647. Kataria S, Guruprasad KN. 2012. Intraspecific variations in growth, yield and photosynthesis of sorghum varieties to ambient UV (280–400nm) radi- ation. Plant Science 196, 85–92. Ohsugi R, Murata T. 1985. C4 photosynthetic characteristics of Panicum species in the Dichotomiflora group. Japan Agricultural Research Quarterly 19, 125–131. Kolbe AR, Cousins AB. 2018. Mesophyll conductance in Zea mays re- sponds transiently to CO2 availability: implications for transpiration efficiency in C4 crops. C4 photosynthesis and quantitative genetics 3083 Furbank RT. 2011. Evolution of the C4 photosynthetic mechanism: are there really three C4 acid decarboxylation types? Journal of Experimental Botany 62, 3103–3108. Laurie DA, Bennett MD. 1986. Wheat × maize hybridization. Canadian Journal of Genetics and Cytology 28, 313–316. Laurie DA, Bennett MD. 1989. The timing of chromosome elimination in hexaploid wheat × maize crosses. Genome 32, 953–961. Gernand D, Rutten T, Varshney A, Rubtsova M, Prodanovic S, Brüß C, Kumlehn J, Matzk F, Houben A. 2005. Uniparental chromosome elim- ination at mitosis and interphase in wheat and pearl millet crosses involves micronucleus formation, progressive heterochromatinization, and DNA frag- mentation. The Plant Cell 17, 2431–2438. Laurie DA, O’Donoughue LS, Bennett MD. 1990. Wheat × maize and other wide sexual hybrids: their potential for genetic manipulation and crop improvement. In: Gustafson JP, ed. Gene manipulation in plant improve- ment II. Boston, MA: Springer, 95–126. Hatch M, Kagawa T, Craig S. 1975. Subdivision of C4-pathway species based on differing C4 acid decarboxylating systems and ultrastructural fea- tures. Functional Plant Biology 2, 111. Leary MHO. 1988. Carbon isotopes in photosynthesis. BioScience 38, 328–336. Lightfoot E, Przelomska N, Craven M, O Connell TC, He L, Hunt HV, Jones MK. 2016. Intraspecific carbon and nitrogen isotopic variability in foxtail millet (Setaria italica). Rapid Communications in Mass Spectrometry 30, 1475–1487. He K, Gkioxari G, Dollar P, Girshick R. 2017. Mask R-CNN. In: IEEE International Conference on Computer Vision, 2980–2988. Holaday AS, Brown RH, Bartlett JM, Sandlin EA, Jackson RC. 1988. Enzymic and photosynthetic characteristics of reciprocal F1 hybrids of Flaveria pringlei (C3) and Flaveria brownii (C4-like species). Plant Physiology 87, 484–490. Lin M, Schlüter U, Stich B, Weber APM. 2021. Cis -regulatory diver- gence underpins the evolution of C3–C4 intermediate photosynthesis in Moricandia. bioRxiv. doi:10.1101/2021.05.10.443365. [Preprint]. Hormozdiari F, Kostem E, Kang EY, Pasaniuc B, Eskin E. 2014. Identifying causal variants at loci with multiple signals of association. Genetics 198, 497–508. Lundgren MR, Christin PA, Escobar EG, Ripley BS, Besnard G, Long CM, Hattersley PW, Ellis RP, Leegood RC, Osborne CP. 2016. Evolutionary implications of C3–C4 intermediates in the grass Alloteropsis semialata. Plant, Cell & Environment 39, 1874–1885. Huang BE, Verbyla KL, Verbyla AP, Raghavan C, Singh VK, Gaur P, Leung H, Varshney RK, Cavanagh CR. 2015. MAGIC populations in crops: current status and future prospects. Theoretical and Applied Genetics 128, 999–1017. Mackay I, Powell W. 2007. Methods for linkage disequilibrium mapping in crops. References Björkman O, Gauhl E, Nobs M. 1969. Comparative studies of Atriplex species with and without β-carboxylation photosynthesis. Carnegie Institution of Washington Yearbook 68, 620–633. Edwards EJ, Smith SA. 2010. Phylogenetic analyses reveal the shady history of C4 grasses. Proceedings of the National Academy of Sciences, USA 107, 2532–2537. Björkmann O, Nobs MA, Berry J. 1971. Further studies on hybrids beteen C3 and C4 species of Atriplex. Carnegie Institiute of Washington Annual Report 70, 507–511. Edwards GE, Voznesenskaya EV. 2011. C4 phtosynthesis: Kranz forms and single-cell C4 in terrestrial plants. In: Raghavendra AS, Sage RF, eds. C4 photosynthesis and related CO2 concentrating mechanisms. Dordrecht: Springer, 29–61. Botha CEJ. 1992. Plasmodesmatal distribution, structure and frequency in relation to assimilation in C3 and C4 grasses in southern Africa. Planta 187, 348–358. Evans DA. 1983. Agricultural applications of plant protoplast fusion. Nature Biotechnology 1, 253–261. Bouton JH, Brown RH, Evans PT, Jernstedt JA. 1986. Photosynthesis, leaf anatomy, and morphology of progeny from hybrids between C3 and C3/ C4 Panicum species. Plant Physiology 80, 487–492. Faye JM, Maina F, Akata EA, et al. 2021. A genomics resource for gen- etics, physiology, and breeding of West African sorghum. Plant Genome 14, 1–18. Bowes G, Ogren WL, Hageman RH. 1971. Phosphoglycolate produc- tion catalyzed by ribulose diphosphate carboxylase. Biochemical and Biophysical Research Communications 45, 716–722. Ferguson JN, Fernandes SB, Monier B, et al. 2021. Machine leaning enabled phenotyping for GWAS and TWAS of WUE traits in 869 field-grown sorghum accessions. Plant Physiology 187, 1481–1500. Broman KW. 2005. The genomes of recombinant inbred lines. Genetics 169, 1133–1146. Flood PJ, Kruijer W, Schnabel SK, Schoor R, Jalink H, Snel JFH, Harbinson J, Aarts MGM. 2016. Phenomics for photosynthesis, growth and reflectance in Arabidopsis thaliana reveals circadian and long-term fluc- tuations in heritability. Plant Methods 12, 1–14. Broman KW, Gatti DM, Simecek P, Furlotte NA, Prins P, Sen S, Yandell BS, Churchill GA. 2018. R/qtl2: software for mapping quantitative trait loci with high-dimensional data and for mapping quantitative trait loci C4 photosynthesis and quantitative genetics New Phytologist 217, 1463–1474. Omoto E, Taniguchi M, Miyake H. 2012. Adaptation responses in C4 photosynthesis of maize under salinity. Journal of Plant Physiology 169, 469–477. Kolbe AR, Studer AJ, Cousins AB. 2018. Biochemical and transcriptomic analysis of maize diversity to elucidate drivers of leaf carbon isotope com- position. Functional Plant Biology 45, 489–500. Ongom PO, Ejeta G. 2017. Mating design and genetic structure of a multi- parent advanced generation intercross (MAGIC) population of sorghum (Sorghum bicolor L. Moench). G3 8, 331–341. Korte A, Farlow A. 2013. The advantages and limitations of trait analysis with GWAS: a review. Plant Methods 9, 291. Ortiz D, Hu J, Salas Fernandez MG. 2017. Genetic architecture of photo- synthesis in Sorghum bicolor under non-stress and cold stress conditions. Journal of Experimental Botany 68, 4545–4557. Koteyeva NK, Voznesenskaya EV, Roalson EH, Edwards GE. 2011. Diversity in forms of C4 in the genus Cleome (Cleomaceae). Annals of Botany 107, 269–283. Pascual L, Desplat N, Huang BE, Desgroux A, Bruguier L, Bouchet JP, Le QH, Chauchard B, Verschave P, Causse M. 2015. Potential of a to- mato MAGIC population to decipher the genetic control of quantitative traits and detect causal variants in the resequencing era. Plant Biotechnology Journal 13, 565–577. Kynast RG, Riera-Lizarazu O, Vales MI, et al. 2001. A complete set of maize individual chromosome additions to the oat genome. Plant Physiology 125, 1216–1227. Kynast RG, Okagaki RJ, Galatowitsch MW, Granath SR, Jacobs MS, Stec AO, Rines HW, Phillips RL. 2004. Dissecting the maize genome by using chromosome addition and radiation hybrid lines. Proceedings of the National Academy of Sciences, USA 101, 9921–9926. Peter G, Katinas L. 2003. A new type of Kranz anatomy in Asteraceae. Australian Journal of Botany 51, 217–226. Ladejobi O, Elderfield J, Gardner KA, Gaynor RC, Hickey J, Hibberd JM, Mackay IJ, Bentley AR. 2016. Maximizing the potential of multi- parental crop populations. Applied and Translational Genomics 11, 9–17. Pignon CP, Fernandes SB, Valluru R, Bandillo N, Lozano R, Buckler E, Gore MA, Long SP, Brown PJ, Leakey ADB. 2021. Phenotyping stomatal closure by thermal imaging for GWAS and TWAS of water use 3084 3084 \| Simpson et al. 3084 Simpson et al. efficiency-related genes. Plant Physiology doi: https://doi.org/10.1093/ plphys/kiab395. The Angiosperm Phylogeny Group. 2016. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Botanical Journal of the Linnean Society 181, 1–20. C4 photosynthesis and quantitative genetics Portis AR, Parry MAJ. 2007. Discoveries in Rubisco (Ribulose 1,5-bisphosphate carboxylase/oxygenase): a historical perspective. Photosynthesis Research 94, 121–143. Tolley BJ, Sage TL, Langdale JA, Hibberd JM. 2012. Individual maize chromosomes in the C3 plant oat can increase bundle sheath cell size and vein density. Plant Physiology 159, 1418–1427. Prendergast HDV. 1987. Structural, biochemical and geographical re- lationships in Australian C4 grasses. PhD thesis, Australian National University. Twohey RJ III, Roberts LM, Studer AJ. 2019. Leaf stable carbon isotope composition reflects transpiration efficiency in Zea mays. The Plant Journal 97, 475–484. Reeves G, Singh P, Rossberg TA, Sogbohossou EOD, Schranz ME, Hibberd JM. 2018. Natural variation within a species for traits underpinning C4 photosynthesis. Plant Physiology 177, 504–512. Valluru R, Gazave EE, Fernandes SB, Ferguson JN, Lozano R, Hirannaiah P, Zuo T, Brown PJ, Leakey ADB, Gore MA. 2019. Deleterious mutation burden and its association with complex traits in sor- ghum (Sorghum bicolor). Genetics 211, 1075–1087. p p g C4 photosynthesis. Plant Physiology 177, 504–512. Rikiishi K, Oguro H, Samejima M, Sugiyama T, Hinata K. 1988. C4-like plants derived from a cross (Atriplex rosea (C4) × A. patula (C3)) × A. rosea. Japanese Journal of Breeding 38, 397–408. van Bezouw RFHM, Keurentjes JJB, Harbinson J, Aarts MGM. 2019. Converging phenomics and genomics to study natural variation in plant photosynthetic efficiency. The Plant Journal 97, 112–133. Sage RF. 2016. A portrait of the C4 photosynthetic family on the 50th anni- versary of its discovery: species number, evolutionary lineages, and hall of fame. Journal of Experimental Botany 67, 4039–4056. Voznesenskaya EV, Koteyeva NK, Chuong SDX, Ivanova AN, Barroca J, Craven LA, Edwards GE. 2007. Physiological, anatomical and biochemical characterisation of photosynthetic types in genus Cleome (Cleomaceae). Functional Plant Biology 34, 247–267. Sage RF, Stata M. 2015. Photosynthetic diversity meets biodiversity: the C4 plant example. Journal of Plant Physiology 172, 104–119. Sage RF, Christin P-A, Edwards EJ. 2011. The C4 plant lineages of planet Earth. Journal of Experimental Botany 62, 3155–3169. Voznesenskaya EV, Koteyeva NK, Edwards GE, Ocampo G. 2017. Unique photosynthetic phenotypes in Portulaca (Portulacaceae): C3–C4 intermediates and NAD-ME C4 species with Pilosoid-type Kranz anatomy. Journal of Experimental Botany 68, 225–239. Sales CRG, Ribeiro RV, Hayashi AH, Marchiori PER, Silva KI, Martins MO, Silveira JAG, Silveira NM, Machado EC. 2018. Flexibility of C4 decarboxylation and photosynthetic plasticity in sugarcane plants under shading. Environmental and Experimental Botany 149, 34–42. C4 photosynthesis and quantitative genetics Wainberg M, Sinnott-Armstrong N, Mancuso N, et al. 2019. Opportunities and challenges for transcriptome-wide association studies. Nature Genetics 51, 592–599. Sedelnikova OV, Hughes TE, Langdale JA. 2018. Understanding the genetic basis of C4 Kranz anatomy with a view to engineering C3 crops. Annual Review of Genetics 52, 249–270. Wang K, Wu Y, Zhang W, Dawe RK, Jiang J. 2014. Maize centromeres expand and adopt a uniform size in the genetic background of oat. Genome Research 24, 107–116. Sharkey TD. 1988. Estimating the rate of photorespiration in leaves. Physiologia Plantarum 73, 147–152. Wang TB, Niizeki M. 1994. Somatic hybridization between Zea mays and Triticum sect. tritirigia. In: Bajaj YPS, ed. Biotechnology in agriculture and forestry. Berlin, Heidelberg: Springer Berlin Heidelberg, 99–111. Sharwood RE, Sonawane BV, Ghannoum O. 2014. Photosynthetic flexi- bility in maize exposed to salinity and shade. Journal of Experimental Botany 65, 3715–3724. Wang TB, Niizeki M, Harada T, Ishikawa R, Qian YQ, Saito K. 1993. Establishment of somatic hybrid cell lines between Zea mays L. (maize) and Triticum sect, trititrigia MacKey (trititrigia). Theoretical and Applied Genetics 86, 371–376. Sogbohossou EOD, Achigan-Dako EG, Maundu P, Solberg S, Deguenon EMS, Mumm RH, Hale I, Van Deynze A, Schranz ME. 2018. A roadmap for breeding orphan leafy vegetable species: a case study of Gynandropsis gynandra (Cleomaceae). Horticulture Research 5, 1–15. Williams BP, Johnston IG, Covshoff S, Hibberd JM. 2013. Phenotypic landscape inference reveals multiple evolutionary paths to C4 photosyn- thesis. eLife 2, 1–19. Stadlmeier M, Hartl L, Mohler V. 2018. Usefulness of a multiparent ad- vanced generation intercross population with a greatly reduced mating de- sign for genetic studies in winter wheat. Frontiers in Plant Science 871, 1–12. Xie J, Fernandes SB, Mayfield-Jones D, Erice G, Choi M, E Lipka A, Leakey ADB. 2021. Optical topometry and machine learning to rap- idly phenotype stomatal patterning traits for maize QTL mapping. Plant Physiology 187, 1462–1480. Sternberg LDSL, Deniro MJ, Sloan ME, Black CC. 1986. Compensation point and isotopic characteristics of C3/C4 intermediates and hybrids in Panicum. Plant Physiology 80, 242–245. Xin HW, Sun JS, Yan QS, Zhang XQ. 1997. Plant regeneration from asymmetric somatic hybrids of Oryza sativa and Panicum maximum. Acta Botanica Sinica 39, 717–724. Strigens A, Freitag NM, Gilbert X, Grieder C, Riedelsheimer C, Schrag TA, Messmer R, Melchinger AE. 2013. Association mapping for chilling tolerance in elite flint and dent maize inbred lines evaluated in growth chamber and field experiments. C4 photosynthesis and quantitative genetics Plant, Cell & Environment 36, 1871–1887. Xu C, Xia G, Zhi D, Xiang F, Chen H. 2003. Integration of maize nuclear and mitochondrial DNA into the wheat genome through somatic hybridiza- tion. Plant Science 165, 1001–1008. Szarka B, Göntér I, Molnár-Láng M, Mórocz S, Dudits D. 2002. Mixing of maize and wheat genomic DNA by somatic hybridization in regenerated sterile maize plants. Theoretical and Applied Genetics 105, 1–7. Yabiku T, Ueno O. 2017. Variations in physiological, biochemical, and structural traits of photosynthesis and resource use efficiency in maize and teosintes (NADP-ME-type C4). Plant Production Science 20, 448–458. Tam V, Patel N, Turcotte M, Bossé Y, Paré G, Meyre D. 2019. Benefits and limitations of genome-wide association studies. Nature Reviews. Genetics 20, 467–484. Yin X, Struik PC. 2020. Viewpoints: Exploiting differences in the energy budget among C4 subtypes to improve crop productivity. New Phytologist 229, 2400–2409. Tao Y, Zhao X, Wang X, Hathorn A, Hunt C, Cruickshank AW, Erik J, Godwin ID, Mace ES, Jordan DR. 2020. Large-scale GWAS in sor- ghum reveals common genetic control of grain size among cereals. Plant Biotechnology Journal 18, 1093–1105. Yu J, Holland JB, McMullen MD, Buckler ES. 2008. Genetic design and statistical power of nested association mapping in maize. Genetics 178, 539–551. Zhang N, Gibon Y, Wallace JG, et al. 2015. Genome-wide association of carbon and nitrogen metabolism in the maize nested association mapping population. Plant Physiology 168, 575–583. Terada R, Kyozuka J, Nishibayashi S, Shimamoto K. 1987. Plantlet re- generation from somatic hybrids of rice (Oryza sativa L.) and barnyard grass (Echinochloa oryzicola Vasing). Molecular & General Genetics 210, 39–43.
https://openalex.org/W2169749928	https://europepmc.org/articles/pmc2699347?pdf=render	English	null	An early burst of IFN-γ induced by the pre-erythrocytic stage favours Plasmodium yoelii parasitaemia in B6 mice	Malaria journal	2,009	cc-by	9,589	Published: 9 June 2009 Malaria Journal 2009, 8:128 doi:10.1186/1475-2875-8-128 This article is available from: http://www.malariajournal.com/content/8/1/128 © 2009 Soulard et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BioMed Central BioMed Central BioMed Central Ope Research An early burst of IFN-γ induced by the pre-erythrocytic stage favours Plasmodium yoelii parasitaemia in B6 mice Valérie Soulard†1,2, Jacques Roland†1,3, Olivier Gorgette1,4, Eliane Barbie Pierre-André Cazenave1,5 and Sylviane Pied1,3 Address: 1Unité d'Immunophysiopathologie Infectieuse, Centre National de la Recherche Scientifique (CNRS) URA 1961, Université Paris VI, Institut Pasteur, 25-28 rue du Dr Roux, 75015 Paris, France, 2UR010, Santé de la mère et de l'enfant en milieu tropical, Institut de Recherche pour le Développement (IRD), Université Paris Descartes, Faculté de Pharmacie, 4 avenue de l'Observatoire, 75270 PARIS Cedex 06, France, 3U547, Institut National de la Santé et de la Recherche Médicale (INSERM), Institut Pasteur de Lille, 1 rue du Pr. Calmette, BP 245, 59019 LILLE Cedex, France, 4Unité d'Immunologie Moléculaire des Parasites, Institut Pasteur, 25-28 rue du Dr Roux, 75015 Paris, France and 5UPMC-CNRS 7087, Université Pierre et Marie Curie, GH La Pitié Salpétrière, 83 Boulevard de l'Hôpital, 75651 Paris Cedex 13, France mail: Valérie Soulard - valerie_soulard@yahoo.fr; Jacques Roland - jacques.roland@pasteur-lille.fr; Olivier Gorgett liane Barbier - elianecolettebarbier@orange.fr; Pierre-André Cazenave - cazenave@pasteur.fr; Sylviane Pied - sylvian Email: Valérie Soulard - valerie_soulard@yahoo.fr; Jacques Roland* - jacques.roland@pasteur-lille.fr; Olivier Gorgette - ogorgett@pasteur.fr; Eliane Barbier - elianecolettebarbier@orange.fr; Pierre-André Cazenave - cazenave@pasteur.fr; Sylviane Pied - sylviane.pied@pasteur-lille.fr * Corresponding author †Equal contributors * Corresponding author †Equal contributors Received: 16 January 2009 Accepted: 9 June 2009 Malaria Journal Open Access Background this early IFN-γ production on the control of infection in vivo. Surprisingly, the parasitaemia of infected B6.IFN-γ-/- was significantly lower than the one of B6 mice. These results were confirmed by in vivo neutralization of the early production of IFN-γ in infected B6 mice using anti- IFN-γ Abs. Finally, the stage-specificity of this early IFN-γ burst was addressed and, following infection of B6 mice with P. yoelii-infected RBCs, no peak of IFN-γ was detecta- ble in their serum during the first week of infection. g Malaria is initiated when sporozoites are injected into the mammalian host during the blood meal of an infected mosquito. The sporozoites reach the liver where they mature and divide within hepatocytes, thus completing the first phase of the parasite life cycle, the pre-erythro- cytic stage. When the hepatic forms are mature, thousands of merozoites are released, reach the blood stream and invade red blood cells, initiating the erythrocytic stage. Parasitized RBCs (pRBCs) subsequently release new merozoites, which perpetuate the erythrocytic cycle by invading new RBCs. In conclusion, these results show that, during a primary infection initiated with P. yoelii sporozoites, IFN-γ plays two opposite roles in the control of parasitaemia and that, unexpectedly, B6 mice control their parasitaemia better in the absence of an early IFN-γ production. Together with a recent report from Couper et al [13], the results presented here challenge the current view regarding the role of IFN- γ on the control of the parasite growth and suggest that an early peak of circulating IFN-γ promotes Plasmodium's growth during a primary infection. In addition to that, these data show that the immune response induced by the pre-erythrocytic stage impacts on the control of the subse- quent erythrocytic stage. These results highlight the importance of studying the immune response to Plasmo- dium in models as close as possible to the physiology of the natural infection. In mice, as well as in humans, early immune events play a determinant role in the outcome of malaria, which depends, in part, on a subtle balance between pro- and anti-inflammatory responses. An early proinflammatory immune response has not only been associated with pro- tection, through its contribution to parasite elimination, but also with the severe complications of the disease [1-3]. In murine models of malaria, the early production of IL- 12, IL-18, TNF and IFN-γ has been associated with the res- olution of blood-stage infection [1,4-6]. Background Particularly, an early IFN-γ production was shown to be essential to con- trol parasitaemia and to be associated with a better sur- vival prognostic [7-9]. IFN-γ indeed promotes protective Th1 T cell responses in vivo, and favours the Th1-associ- ated IgG2a response involved in the clearance of blood- stage parasites [9]. Abstract When the hepatic forms are of merozoites are released, reach the invade red blood cells, initiating the Parasitized RBCs (pRBCs) subsequ merozoites, which perpetuate the ery invading new RBCs. In mice, as well as in humans, early im determinant role in the outcome depends, in part, on a subtle balance anti-inflammatory responses. An earl immune response has not only been a tection, through its contribution to pa but also with the severe complications In murine models of malaria, the earl 12, IL-18, TNF and IFN-γ has been asso olution of blood-stage infection [1,4- early IFN-γ production was shown to trol parasitaemia and to be associated vival prognostic [7-9]. IFN-γ indeed p Th1 T cell responses in vivo, and favo ated IgG2a response involved in the c stage parasites [9]. Recently, NK, NKT, and γδT cells were IFN-γ during the blood-stage of murin in response to P. falciparum-infected R 12]. To better define the early immune me mote the control of Plasmodium grow tion during primary infection, the model of B6 mice infected with P. y zoites, the natural invasive form of the Such an experimental condition allow in vivo of the full Plasmodium life cycle natural infection. In this model, the production of Th1 c first analysed at early post-infection This showed that P. yoelii 265BY infe characterized by a peak of IFN-γ in the while concentrations of circulating TN at the levels found in non-infected anim ric analyses revealed that splenic NKT, cells simultaneously produced this cyt the latter ones being the major produc - mice were used to evaluate the physio Methods Mice C57Bl/6N@Ico (referred to as B6) mice were purchased from Charles River-Iffa Credo (St-Aubin les Elbeufs, France). CD1d.1-/-, IFN-γ-/- and RAG2-/- mice on a C57BL/ 6 genetic background were provided by A. Bendelac [14], J.F. Bureau (mice bred at the Institut Pasteur, originally from the Jackson Laboratory, Bar Harbor, Maine, USA), and J.P. Di Santo [15] respectively (referred to as B6.CD1d-/-, B6. IFN-γ-/- and B6.RAG2-/-). All animals were housed and bred in the animal facilities of the Institut Pas- teur (Paris, France) under standard conditions. Only 8- to 12-week-old females were used and experiments were conducted in accordance with institutional guidelines for animal care and use. Recently, NK, NKT, and γδT cells were shown to produce IFN-γ during the blood-stage of murine malaria and also in response to P. falciparum-infected RBCs in vitro [3,10- 12]. To better define the early immune mechanisms that pro- mote the control of Plasmodium growth and its elimina- tion during primary infection, the non-lethal malaria model of B6 mice infected with P. yoelii 265BY sporo- zoites, the natural invasive form of the parasite, was used. Such an experimental condition allows the development in vivo of the full Plasmodium life cycle that is closer to the natural infection. Abstract Background: In murine models of malaria, an early proinflammatory response has been associated with the resolution of blood-stage infection. To dissect the protective immune mechanims that allow the control of parasitaemia, the early immune response of C57BL/6 mice induced during a non-lethal plasmodial infection was analysed. Methods: Mice were infected with Plasmodium yoelii 265BY sporozoites, the natural invasive form of the parasite, in order to complete its full life cycle. The concentrations of three proinflammatory cytokines in the sera of mice were determined by ELISA at different time points of infection. The contribution of the liver and the spleen to this cytokinic response was evaluated and the cytokine-producing lymphocytes were identified by flow cytometry. The physiological relevance of these results was tested by monitoring parasitaemia in genetically deficient C57BL/6 mice or wild-type mice treated with anti-cytokine neutralizing antibody. Finally, the cytokinic response in sera of mice infected with parasitized-RBCs was analysed. Results: The early immune response of C57BL/6 mice to sporozoite-induced malaria is characterized by a peak of IFN-γ in the serum at day 5 of infection and splenic CD4 T lymphocytes are the major producer of this cytokine at this time point. Somewhat unexpected, the parasitaemia is significantly lower in P. yoelii- infected mice in the absence of IFN-γ. More precisely, at early time points of infection, IFN-γ favours parasitaemia, whereas helping to clear efficiently the blood-stage parasites at later time points. Interestingly, the early IFN-γ burst is induced by the pre-erythrocytic stage. Conclusion: These results challenge the current view regarding the role of IFN-γ on the control of parasite growth since they show that IFN-γ is not an essential mediator of protection in P. yoelii-infected C57BL/6 mice. Moreover, the mice parasitaemia is more efficiently controlled in the absence of an early IFN-γ production, suggesting that this cytokine promotes parasite's growth. Finally, this early burst of IFN- γ is induced by the pre-erythrocytic stage, showing the impact of this stage on the immune response taking place during the subsequent erythrocytic stage. Page 1 of 11 (page number not for citation purposes) http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 http://www.malariajournal.com/content/8/1/128 http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 Background Malaria is initiated when sporozoites a mammalian host during the blood m mosquito. The sporozoites reach th mature and divide within hepatocyte the first phase of the parasite life cyc cytic stage. Isolation of hepatic and splenic mononuclear cells f p p Hepatic mononuclear cells were prepared as previously described [20]. Briefly, livers from control and infected mice were perfused in situ with sterile DMEM, removed and homogenized using a Potter-Elvehjem homogenizer. Cells were washed, resuspended in a 35% Percoll solution (Pharmacia Biotech, Uppsala, Sweden) and centrifuged at 1400 g for 25 min at room temperature (RT). The pellet (containing mononuclear cells) was washed with DMEM. Spleens from control and infected mice were gently smashed between two glass slides in sterile D-MEM (Gibco Invitrogen, Cergy Pontoise, France). The cell sus- pensions were then washed with D-MEM before erythro- cyte lysis with an ACK lysing buffer (0.15 M NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA). Finally, cells were resus- pended in sterile RPMI 1640 medium + GlutaMAX I (Gibco Invitrogen) containing 3% FCS before counting of living cells in eosin. For quantification of IFN-γ in culture supernatants, total splenocytes and iHLs were isolated, resuspended at a con- centration of 1 × 106 cells/ml in complete RPMI 1640 GlutaMAX I (10% FCS, 100 IU/ml penicillin-streptomy- cin) and distributed at 200 μl/well in 96-well tissue cul- ture plates. Duplicate cultures were done for each sample. After 72 hours of culture, plates were centrifuged, 150 μl of supernatants were collected and stored frozen at -20°C until use. DO means of duplicate wells were used to deter- mine the concentration of IFN-γ in the samples according to the standard. In vivo treatment of C57Bl/6 mice with anti-IFN-γ mAb Mice were infected with 4,000 sporozoites and, on the indicated day, were injected intraperitoneally with 2 mg of anti-IFN-γ mAb (XMG1.2, purified from ascites fluid by DEAE cellulose) or with 2 mg of irrelevant control rat IgG (ref. I4131, Sigma). Intracellular cytokine staining for flow cytometric analysis Splenic cells from control and infected mice were isolated and incubated at a concentration of 1 × 106 cells per ml, for 1 hour at 37°C in a 5% CO2 atmosphere, in RPMI 1640 medium + GlutaMAX I containing 10% FCS, penicil- lin-streptomycin (100 IU/ml, Gibco Invitrogen), and brefeldin A (10 μg/ml, Sigma Aldrich). Cells were then washed once in PBS-3% FCS and surface antigens were stained at 4°C, in the dark, for 20 min, using the follow- ing monoclonal antibodies purchased from BD Bio- sciences and conjugated to biotine, FITC, PE, APC or PE- cyanine7: anti-NK1.1 (PK136), anti-TCRγδ (GL3), anti- CD3ε (145-2C11), anti-CD4 (L3T4), anti-CD8α (Ly-2). Parasites, in vivo infection, and parasitaemia Sporozoites of the uncloned line of the 265BY strain of Plasmodium yoelii yoelii were obtained by dissecting the sal- ivary glands of infected Anopheles stephensi mosquitoes as previously described [16]. The mosquitoes were bred, maintained, and infected at the CEPIA (Centre de Produc- tion et Infection des Anophèles, Institut Pasteur, Paris, France). Mice were infected intravenously with 4,000 spo- rozoites diluted in sterile PBS or intraperitoneally with 106 parasitized-RBCs (pRBCs). Parasitaemia was meas- ured by flow cytometry following a protocol adapted from Jouin et al [17,18], and Lee et al [19]. Briefly, 3 μl of blood In this model, the production of Th1 cytokines in vivo was first analysed at early post-infection (p.i.) time points. This showed that P. yoelii 265BY infection in B6 mice is characterized by a peak of IFN-γ in the serum at day 5 p.i., while concentrations of circulating TNF and IL-12 stayed at the levels found in non-infected animals. Flow cytomet- ric analyses revealed that splenic NKT, NK, γδ and CD4 T cells simultaneously produced this cytokine at day 5 p.i., the latter ones being the major producer. Then, B6.IFN-γ-/ - mice were used to evaluate the physiological relevance of Page 2 of 11 (page number not for citation purposes) Malaria Journal 2009, 8:128 http://www.malariajournal.com/content/8/1/128 Statistical analyses Statistical analyses were performed using the non-para- metric Mann-Whitney test and Statview 5.0 software (SAS Institute Inc., Cary, NC). A p-value < 0.05 was considered significant. Isolation of hepatic and splenic mononuclear cells Biotinylated mAbs were revealed with streptavidine-PE- cyanine7 (BD Biosciences). Then, cells were fixed for one hour at RT, in the dark, with 2% PFA, and subsequently treated with Perm/Wash solution (BD Biosciences) before incubation with APC-conjugated anti-IFN-γ (XMG1.2, BD Biosciences) or isotype-matched control mAb (rat IgG1, BD Biosciences) in Perm/Wash solution at RT, in the dark, for 30 min. Finally, cells were washed in Perm/Wash solu- tion and then in PBS-3% FCS. Stained cells were analysed on a six-color LSR flow cytometer with the CellQuest Pro software (BD Biosciences, San Diego, CA). Quantification of cytokines in the sera and in culture supernatants by ELISA taken from the tail vein were fixed in 500 μl of PBS-glutar- aldehyde 0.25% (grade I, ref. G5882, Sigma Aldrich, Lyon, France) and stored at 4°C until use. Once all the samples were collected, 50 μl of each were incubated in 400 μl of PBS containing 0.5 mM Hoechst (Bisbenzimide H33258, ref. B 2883, Sigma Aldrich) and 0.1 mg/ml Thi- azole Orange (ref. 39 006.2, Sigma Aldrich) for 1 hour, at RT, in the dark. Analysis of the staining was performed on a LSR cytometer using the CellQuest Pro software (BD Biosciences, San Diego, California, USA). Results are expressed as the percentage of pRBCs. taken from the tail vein were fixed in 500 μl of PBS-glutar- aldehyde 0.25% (grade I, ref. G5882, Sigma Aldrich, Lyon, France) and stored at 4°C until use. Once all the samples were collected, 50 μl of each were incubated in 400 μl of PBS containing 0.5 mM Hoechst (Bisbenzimide H33258, ref. B 2883, Sigma Aldrich) and 0.1 mg/ml Thi- azole Orange (ref. 39 006.2, Sigma Aldrich) for 1 hour, at RT, in the dark. Analysis of the staining was performed on a LSR cytometer using the CellQuest Pro software (BD Biosciences, San Diego, California, USA). Results are expressed as the percentage of pRBCs. Mice were bled at the indicated days and sera were aliq- uoted and stored at -20°C until use. The mouse IFN-γ ELISA Set (Cat. No. 555138), mouse TNF (Mono/Mono) ELISA Set (Cat. No. 555268), and mouse IL-12p40 ELISA Set (Cat. No. 555165) from BD Biosciences were used to quantify each cytokine following the manufacturer's instructions. Duplicate serial dilutions were performed for each serum, and DO means were used to determine the concentrations of cytokines in the samples according to the standard. Plasmodium yoelii sporozoite infection induces an early peak of IFN-γ in the serum of B6 mice p γ The in vivo production of three Th1-type cytokines, namely IFN-γ, TNF and IL-12, during P. yoelii primary infection in B6 mice was first adressed. For that purpose, the concentrations of these cytokines in the sera of non- infected and infected mice were determined by ELISA, between day 3 and day 10 post-injection of sporozoites. TNF and IL-12 concentrations did not change during the course of infection and remained at the level found in non-infected control mice. Conversely, the concentration of IFN-γ peaked at day 5 p.i. (Figure 1; median value at day 5 p.i. = 11.51 ng/ml, n = 21; median value in non-infected control mice = 1.66 ng/ml, n = 22; p < 0.0001). A large range of IFN-γ concentrations Page 3 of 11 (page number not for citation purposes) Page 3 of 11 (page number not for citation purposes) http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 Primary infection of B6 mice with P. yoelii sporozoites is characterized by a peak of IFN-γ in the sera at day 5 post-infection Figure 1 Primary infection of B6 mice with P. yoelii sporozoites is characterized by a peak of IFN-γ in the sera at day 5 post-infection. B6 mice were infected with 4,000 sporozoites of P. yoelii 265BY and the serum level of IFN-γ at days 0, 3, 4, 5, 6, 7, and 10 p.i. was determined by ELISA. Results from a pool of 6 independent experiments (3 to 21 mice per time point) are shown. Individual values (circles) and mean values (bars) are shown. NiSG: day-5 serum from B6 mice which received an injec- tion of non-infected salivary gland extract. The statistical difference between day 0 (n = 17) and day 5 p.i. (n = 21) was calcu- lated with the Mann-Whitney test. Primary infection of B6 mice with P. yoelii sporozoites is characterized by a peak of IFN-γ in the sera at day 5 post-infection Figure 1 Primary infection of B6 mice with P. yoelii sporozoites is characterized by a peak of IFN γ in the sera at day 5 post infection Figure 1 Primary infection of B6 mice with P. yoelii sporozoites is characterized by a peak of IFN-γ in the sera at day 5 post-infection. B6 mice were infected with 4,000 sporozoites of P. yoelii 265BY and the serum level of IFN-γ at days 0, 3, 4, 5, 6, 7, and 10 p.i. was determined by ELISA. Plasmodium yoelii sporozoite infection induces an early peak of IFN-γ in the serum of B6 mice Results from a pool of 6 independent experiments (3 to 21 mice per time point) are shown. Individual values (circles) and mean values (bars) are shown. NiSG: day-5 serum from B6 mice which received an injec- tion of non-infected salivary gland extract. The statistical difference between day 0 (n = 17) and day 5 p.i. (n = 21) was calcu- lated with the Mann-Whitney test. The early peak of IFN-γ in the serum is dependent on T and/or B lymphocytes, but independent of CD1d-restricted NKT cells at day 5 p.i. was observed among mice, which could be explained by the transient nature of IFN-γ production. No increase in the IFN-γ concentration was detected in the sera of mice 5 days after injection of salivary gland extracts from non-infected mosquitoes (NiSG, Figure 1). This result shows that the peak of IFN-γ observed at day 5 p.i. is not the result of a non-specific immune response induced by salivary gland extracts, but is specifically induced by P. yoelii infection. This peak of IFN-γ was absent from the sera of B6.IFN-γ-/- mice. Thus, P. yoelii pri- mary infection in B6 mice, induced by the injection of sporozoites, is characterized by a peak of IFN-γ in the blood at day 5 of infection. at day 5 p.i. was observed among mice, which could be explained by the transient nature of IFN-γ production. No increase in the IFN-γ concentration was detected in the sera of mice 5 days after injection of salivary gland extracts from non-infected mosquitoes (NiSG, Figure 1). This result shows that the peak of IFN-γ observed at day 5 p.i. is not the result of a non-specific immune response induced by salivary gland extracts, but is specifically induced by P. yoelii infection. This peak of IFN-γ was absent from the sera of B6.IFN-γ-/- mice. Thus, P. yoelii pri- mary infection in B6 mice, induced by the injection of sporozoites, is characterized by a peak of IFN-γ in the blood at day 5 of infection. To identify the lymphoid cell populations involved in this early IFN-γ production, sera of B6.RAG2-/- mice (which lack T, B, and NKT cells, but possess NK cells) were tested by ELISA. No peak of IFN-γ was detected in the sera of these mice between day 0 and day 10 p.i. (Figure 2), dem- onstrating that this early production of IFN-γ is dependent on T, and/or B, and/or NKT lymphocytes. Moreover, these results suggest that NK cells alone cannot initiate the pro- duction of this cytokine at a level detectable in the serum. Since this burst of IFN-γ occurs early after parasite injec- tion, it was postulated that other innate lymphocytes could be the source of this cytokine. In this context, involvement of CD1d-restricted NKT cells, a population Page 4 of 11 (page number not for citation purposes) Malaria Journal 2009, 8:128 http://www.malariajournal.com/content/8/1/128 The early peak of IFN-γ in the serum of P. The early peak of IFN-γ in the serum is dependent on T and/or B lymphocytes, but independent of CD1d-restricted NKT cells yoelii-infected mice is dependent on T and/or B lymphocytes, but independent of CD1d-dependent NKT cells Figure 2 The early peak of IFN-γ in the serum of P. yoelii- infected mice is dependent on T and/or B lym- phocytes, but independent of CD1d-dependent NKT cells. B6.RAG2-/- mice and B6.CD1d-/- mice were infected with 4,000 sporozoites of P. yoelii 265BY and the serum level of IFN-γ at days 0 and 5 p.i. was determined by ELISA. Results from a pool of 2 to 3 independent experiments per mouse strain (4 to 17 mice per time point) are shown. Indi- vidual values (circles) and mean values (bars) are shown. The statistical difference between day 0 (n = 8 B6.CD1d-/-) and day 5 p.i. (n = 10 B6.CD1d-/-) was calculated with the Mann- Whitney test. The early p is depende CD1d-dep Figure 2 The early p is depende CD1d-dep Figure 2 In order to precise the cellular source of IFN-γ among splenocytes, the intracellular production of this cytokine by splenic NK cells, NKT cells, γδ CD4 and CD8 T cells during infection was analysed by flow cytometry. Exam- ples of FACS analysis for minor lymphocyte populations are shown in Figure 4A. As shown in Figure 4B, the pro- portion of IFN-γ-positive cells among NKT, NK, and γδT cells peaked at day 5 p.i., matching with the peak of this cytokine detected in the sera. It can also be noted that the frequency of IFN-γ-positive cells decreases very rapidly after day 5 of infection, except for NKT cells. In parallel, the percentage of IFN-γ-positive conventional CD4T cells started to increase from day 5 p.i. and was maintained sta- ble, at least, until day 7 p.i. (Figure 4B). y p γ y p y p y p p g The early peak of IFN-γ in the serum of P. yoelii- infected mice is dependent on T and/or B lym- phocytes, but independent of CD1d-dependent NKT cells. B6.RAG2-/- mice and B6.CD1d-/- mice were infected with 4,000 sporozoites of P. yoelii 265BY and the serum level of IFN-γ at days 0 and 5 p.i. was determined by ELISA. Results from a pool of 2 to 3 independent experiments per mouse strain (4 to 17 mice per time point) are shown. Indi- vidual values (circles) and mean values (bars) are shown. The statistical difference between day 0 (n = 8 B6.CD1d-/-) and day 5 p.i. (n = 10 B6.CD1d-/-) was calculated with the Mann- Whitney test. of lymphocytes involved in early immune responses to several pathogens, including Plasmodium [21-23], was tested. B6.CD1d-/- mice, which lack CD1d-restricted NKT cells, were infected and IFN-γ in serum was quantified by ELISA at different time points p.i. As shown in Figure 2, the IFN-γ concentration in the sera of B6.CD1d-/- mice increased at day 5 p.i. and reached a similar amount to that detected in day 5-infected B6 mice (B6.CD1d-/- median value at day 5 p.i. = 6.760 ng/ml, n = 10; B6 median value at day 5 p.i. = 11.51 ng/ml, n = 21; p = 0.3932). At days 3, 4, 6, and 7 p.i., IFN-γ concentrations determined in the sera of B6.CD1d-/- mice were similar to those found in non-infected mice. The early p is depende CD1d-dep Figure 2 Thus, these results show that CD1d-restricted NKT cells are not necessary for the initiation of the IFN-γ burst detected in the serum at day 5 p.i. of lymphocytes involved in early immune responses to several pathogens, including Plasmodium [21-23], was tested. B6.CD1d-/- mice, which lack CD1d-restricted NKT cells, were infected and IFN-γ in serum was quantified by ELISA at different time points p.i. As shown in Figure 2, the IFN-γ concentration in the sera of B6.CD1d-/- mice increased at day 5 p.i. and reached a similar amount to that detected in day 5-infected B6 mice (B6.CD1d-/- median value at day 5 p.i. = 6.760 ng/ml, n = 10; B6 median value at day 5 p.i. = 11.51 ng/ml, n = 21; p = 0.3932). At days 3, 4, 6, and 7 p.i., IFN-γ concentrations determined in the sera of B6.CD1d-/- mice were similar to those found in non-infected mice. Thus, these results show that CD1d-restricted NKT cells are not necessary for the initiation of the IFN-γ burst detected in the serum at day 5 p.i. However, the number of splenic CD4 T cells secreting IFN- γ at day 5 p.i. was 2.4 and 2.5 times superior to the num- bers of IFN-γ-positive innate lymphocytes and CD8 T cells, respectively (Figure 4C). Thus, the production of IFN-γ by splenic innate lymphocytes is simultaneous and peaks at day 5 p.i., but conventional CD4 T cells represent the major source of this cytokine at this time point. Splenocytes are the major source of IFN-γ at early time points of P. yoelii infection Since it has previously been observed that the innate immune response to P. yoelii infection was compartimen- talised and differed between spleen and liver [11,12], the contribution of splenocytes and iHLs to this production of IFN-γ in B6 mice was analysed at days 0 and 5 pi. Splenic and hepatic cells were isolated and IFN-γ was quantified in supernatants after 72 h of culture without any stimulation. As shown in Figure 3A, the concentration of IFN-γ was significantly increased in supernatants of both splenocytes and iHLs isolated from day 5-infected B6 mice. In addition, splenocytes and iHLs produced compa- rable amounts of IFN-γ for a similar number of cells per culture well (Figure 3A). However, the number of total splenocytes at day 5 of infection was about 60 times higher than the one of total iHLs. So, considering the IFN- γ production that can be attributed to each organ, the spleen appeared to be the main source of IFN-γ (Figure 3B). Consequently, further analyses were focussed on this organ. Splenic conventional T CD4 lymphocytes are the major producers of IFN-γ at day 5 p.i Splenic conventional T CD4 lymphocytes are the major producers of IFN-γ at day 5 p.i Page 5 of 11 (page number not for citation purposes) Absence of the early production of IFN-γ associates with a better control of parasitaemia Then, the physiological relevance of this early burst of IFN-γ in the control of infection was addressed. B6 and B6.IFN-γ-/- mice were infected with sporozoites and their survival and parasitaemia were monitored. As shown in Figure 5A, B6.IFN-γ-/- mice survived the infection as well as B6 control mice. The parasitaemia of B6.IFN-γ-/- mice was Page 5 of 11 (page number not for citation purposes) Page 5 of 11 (page number not for citation purposes) Malaria Journal 2009, 8:128 http://www.malariajournal.com/content/8/1 Splenocytes are the major source of IFN-γ ex vivo Figure 3 Splenocytes are the major source of IFN-γ ex vivo. B6 mice were either infected with 4,000 sporozoites of P. yoelii ( 10 mice) or received an injection of salivary gland extracts from non-infected mosquitoes (n = 10 NiSG control mice). Five days later, splenocytes and iHLs were isolated and cultured for 3 days without any stimulation. Then, supernatants were c lected and IFN-γ concentrations were determined by ELISA. A. Results are expressed as mean values ± SD of IFN-γ in pg 2.5 × 105 cells. B. Results are expressed as mean values ± SD of IFN-γ in ng per organ (meaning per total number of spleno cytes or iHLs at the indicated days p.i.). Data from a representative experiment out of two are shown. http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 Splenocytes are the major source of IFN-γ ex vivo Figure 3 Splenocytes are the major source of IFN-γ ex vivo. B6 mice were either infected with 4,000 sporozoites of P. yoelii (n = 10 mice) or received an injection of salivary gland extracts from non-infected mosquitoes (n = 10 NiSG control mice). Five days later, splenocytes and iHLs were isolated and cultured for 3 days without any stimulation. Then, supernatants were col- lected and IFN-γ concentrations were determined by ELISA. A. Results are expressed as mean values ± SD of IFN-γ in pg per 2.5 × 105 cells. B. Results are expressed as mean values ± SD of IFN-γ in ng per organ (meaning per total number of spleno- cytes or iHLs at the indicated days p.i.). Data from a representative experiment out of two are shown. Splenocytes are the major source of IFN-γ ex vivo Figure 3 Splenocytes are the major source of IFN-γ ex vivo. B6 mice were either infected with 4,000 sporozoites of P. Absence of the early production of IFN-γ associates with a better control of parasitaemia yoelii (n = 10 mice) or received an injection of salivary gland extracts from non-infected mosquitoes (n = 10 NiSG control mice). Five days later, splenocytes and iHLs were isolated and cultured for 3 days without any stimulation. Then, supernatants were col- lected and IFN-γ concentrations were determined by ELISA. A. Results are expressed as mean values ± SD of IFN-γ in pg pe 2.5 × 105 cells. B. Results are expressed as mean values ± SD of IFN-γ in ng per organ (meaning per total number of spleno- cytes or iHLs at the indicated days p.i.). Data from a representative experiment out of two are shown. Splenocyte Figure 3 Splenocytes are the major source of IFN-γ ex vivo Figure 3 Splenocytes are the major source of IFN-γ ex vivo. B6 mice were either infected with 4,000 sporozoites of P. yoelii (n = 10 mice) or received an injection of salivary gland extracts from non-infected mosquitoes (n = 10 NiSG control mice). Five days later, splenocytes and iHLs were isolated and cultured for 3 days without any stimulation. Then, supernatants were col- lected and IFN-γ concentrations were determined by ELISA. A. Results are expressed as mean values ± SD of IFN-γ in pg per 2.5 × 105 cells. B. Results are expressed as mean values ± SD of IFN-γ in ng per organ (meaning per total number of spleno- cytes or iHLs at the indicated days p.i.). Data from a representative experiment out of two are shown. Page 6 of 11 (page number not for citation purposes) http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 KT, γδT and CD4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i NKT, γδT and CD4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i. Sple n-infected and infected B6 mice at days 4, 5, 6, and 7 after injection of 4,000 P. yoelii 265BY sp n of IFN-γ among NK cells (gated on NK1.1+CD3-), NKT cells (gated on NK1.1+TCRβ+), γδT , CD4 T cells (gated on CD4+) and CD8 T cells (gated on CD8+) cells was analysed by flow cy wing IFN-γ expression among splenic NK, NKT and γδT cells at days 0 and 5 p.i. IC: isotype-ma ages of positive cells are indicated. B and C. Kinetics of intracellular IFN-γ expression among th s. Analysis was performed as shown in A. Results are expressed as mean values ± SD of the p C) of IFN-γ-positive cells among each subset. Data are representative of two independent ex ce per time point. D4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i CD4 T lymphocytes produce IFN-γ simultaneously at day 5 fected B6 mice at days 4, 5, 6, and 7 after injection of 4,000 P. yoelii 2 ng NK cells (gated on NK1.1+CD3-), NKT cells (gated on NK1.1+TC ated on CD4+) and CD8 T cells (gated on CD8+) cells was analysed ssion among splenic NK, NKT and γδT cells at days 0 and 5 p.i. IC: is cells are indicated. B and C. Splenocyte Figure 3 Page 7 of 11 (page number not for citation purposes) http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 ower in the absence of IFN-γ during primary P. yoelii infection e is lower in the absence of IFN-γ during primary P. yoelii infectio - mice following injection of 4,000 sporozoites of P. yoelii 265BY. B. Parasit N-γ mAb or 2 mg of rat control Ab, at day 3 p.i. with 4,000 sporozoites o with 2 mg of anti- IFN-γ mAb at day 4 p.i. with 4,000 sporozoites of P. yoe oups of mice were identified: "high" and "low" parasitaemia. D. Parasitaem 06 P. yoelii 265BY pRBCs. Each graph is representative of two independen s are expressed as mean value ± SD. The statistical differences between g t (* = p < 0.05). Parasitaemia of B6 mice is lower in the absence of IFN-γ during primary P. yoelii infection Figure 5 Parasitaemia of B6 mice is lower in the absence of IFN-γ during primary P. yoelii infection. A. Parasitaemia of B6 control mice and B6.IFN-γ-/- mice following injection of 4,000 sporozoites of P. yoelii 265BY. B. Parasitaemia of B6 mice treated with either 2 mg of anti- IFN-γ mAb or 2 mg of rat control Ab, at day 3 p.i. with 4,000 sporozoites of P. yoelii 265BY. C. Parasi- taemia of B6 mice treated with 2 mg of anti- IFN-γ mAb at day 4 p.i. with 4,000 sporozoites of P. yoelii 265BY. Regarding their parasitaemia curves, two groups of mice were identified: "high" and "low" parasitaemia. D. Parasitaemia of B6 and B6.IFN-γ-/- mice following injection of 106 P. yoelii 265BY pRBCs. Each graph is representative of two independent experiments with 3 to 9 mice per time point. Results are expressed as mean value ± SD. The statistical differences between groups were determined with the Mann-Whitney test (* = p < 0.05). Parasitaemia of B6 mice is lower in the absence of IFN γ during primary P. yoelii infection Figure 5 Parasitaemia of B6 mice is lower in the absence of IFN-γ during primary P. yoelii infection. A. Parasitaemia of B6 control mice and B6.IFN-γ-/- mice following injection of 4,000 sporozoites of P. yoelii 265BY. B. Parasitaemia of B6 mice treated with either 2 mg of anti- IFN-γ mAb or 2 mg of rat control Ab, at day 3 p.i. with 4,000 sporozoites of P. yoelii 265BY. C. Splenocyte Figure 3 Kinetics of intracellular IFN-γ expression Splenic NK, NKT, γδT and CD4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i Figure 4 Splenic NK, NKT, γδT and CD4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i. Splenocytes were iso- lated from non-infected and infected B6 mice at days 4, 5, 6, and 7 after injection of 4,000 P. yoelii 265BY sporozoites. Intracel- lular expression of IFN-γ among NK cells (gated on NK1.1+CD3-), NKT cells (gated on NK1.1+TCRβ+), γδT cells (gated on CD3+TCRγδ+), CD4 T cells (gated on CD4+) and CD8 T cells (gated on CD8+) cells was analysed by flow cytometry. A. FACS dot plots showing IFN-γ expression among splenic NK, NKT and γδT cells at days 0 and 5 p.i. IC: isotype-matched control anti- body. Percentages of positive cells are indicated. B and C. Kinetics of intracellular IFN-γ expression among the different subsets of lymphocytes. Analysis was performed as shown in A. Results are expressed as mean values ± SD of the percentages (B) or the numbers (C) of IFN-γ-positive cells among each subset. Data are representative of two independent experiments with at least three mice per time point. Splenic NK Figure 4 Splenic NK, NKT, γδT and CD4 T lymphocytes produce IFN γ simultaneously at day 5 p.i Figure 4 Splenic NK, NKT, γδT and CD4 T lymphocytes produce IFN-γ simultaneously at day 5 p.i. Splenocytes were iso- lated from non-infected and infected B6 mice at days 4, 5, 6, and 7 after injection of 4,000 P. yoelii 265BY sporozoites. Intracel- lular expression of IFN-γ among NK cells (gated on NK1.1+CD3-), NKT cells (gated on NK1.1+TCRβ+), γδT cells (gated on CD3+TCRγδ+), CD4 T cells (gated on CD4+) and CD8 T cells (gated on CD8+) cells was analysed by flow cytometry. A. FACS dot plots showing IFN-γ expression among splenic NK, NKT and γδT cells at days 0 and 5 p.i. IC: isotype-matched control anti- body. Percentages of positive cells are indicated. B and C. Kinetics of intracellular IFN-γ expression among the different subsets of lymphocytes. Analysis was performed as shown in A. Results are expressed as mean values ± SD of the percentages (B) or the numbers (C) of IFN-γ-positive cells among each subset. Data are representative of two independent experiments with at least three mice per time point. Discussion The data presented here show that the immune response of B6 mice to a P. yoelii primary infection initiated with sporozoites is characterized by an early burst of IFN-γ, dependent on the pre-erythrocytic stage. It is also interesting to note that B6 mice, which received anti-IFN-γ mAb on day 4 of infection behaved differently. Indeed, from three independent experiments, 36.8% (7 out of 19) of the mice behave like B6 mice treated at day 3 p.i. ("low parasitaemia" group; Figure 5C), and the other mice displayed parasitaemia similar to that of B6 control mice ("high parasitaemia" group; Figure 5C). Ab treatment at days 5 and 7 p.i. had no effect on the parasi- taemia of B6 mice. This shows that the decrease in parasi- taemia, observed between day 6 and day 20 p.i., is a consequence of the neutralization of IFN-γ at days 3–4 p.i., and is not due to persistent anti-IFN-γ mAb, which would have neutralized subsequent production of IFN-γ. Somewhat unexpectedly, IFN-γ was not essential to elimi- nate blood-stage parasites, and moreover, mice controlled their parasitaemia in a more effective manner in the absence of this early burst of IFN-γ. These results are in contradiction with other studies showing that IFN-γ con- tributes to blood-stage clearance. Indeed, in the well- described P. chabaudi model, IFN-γ-deficient mice infected with pRBCs control their parasitaemia less efficiently and die from infection [6,8,9]. This discrepancy cannot be attributed to the parasite stage used to initiate the infec- tion. In addition, it cannot be a consequence of differ- ences in the host genetic background since B6.IFN-γ-/- mice show an increased parasitaemia and exhibit higher mortality following P. chabaudi infection [9]. This suggests that the parasite species likely accounts for these differ- ences. In accordance with this latter hypothesis and also with the results presented here, a recent report showed that the control of the primary wave of P. yoelii 17× para- sitaemia is independent of IFN-γ [13]. Altogether, these results clearly show that, in B6 mice, IFN-γ is not essential to cure of P. yoelii primary infection and to eliminate blood-stage parasites, but in its absence, parasitaemia is delayed and lowered. In other models of non-cerebral malaria, such as P. chabaudi infection, mice are infected with pRBCs. As the results presented here were unexpected compared to the ones observed in the well-described P. The early burst of IFN-γ is induced by the pre-erythrocytic stage of P. yoelii infection The early burst of IFN-γ is induced by the pre-erythrocytic stage of P. yoelii infection The early burst of IFN-γ is induced by the pre-erythrocytic stage of P. yoelii infection significantly lower than that of B6 mice after day 8 of infection. In addition, the duration of parasitaemia was longer in B6. IFN-γ-/- mice than in control animals, as par- asites were cleared from the blood only at day 32 p.i., compared to day 25 in B6 mice (Figure 5A). Finally, in order to determine the parasite developmental stage that induces the early peak of IFN-γ, B6 mice were infected with P. yoelii-infected RBCs and the concentration of IFN-γ was determined in their sera. No significant increase of IFN-γ concentrations was observed in the sera of mice at days 3, 4, 5, 6 and 7 post-injection of pRBCs (two independent experiments). Moreover, the level of IFN-γ stayed at the level found in non-infected animals, excepted for two mice out of 36 infected, which showed a small increase of IFN-γ in their serum (one out of five day- 3 mice had 4.98 ng/ml of IFN-γ, and one out of nine day 6-mice had 6.96 ng/ml of IFN-γ). These results clearly indicate that the early IFN-γ burst is induced by the pre- erythrocytic stage, and not by the blood-stage. To analyse more precisely the consequence of the early peak of IFN-γ on the control of infection, a single injection of anti-IFN-γ mAb was performed in B6 mice, at day 3 or day 4 p.i., and their survival and parasitaemia were mon- itored. As shown in Figure 5B, B6 mice treated at day 3 p.i. with anti-IFN-γ mAb survived the same as control mice which received irrelevant rat IgG, but their parasitaemia rose significantly slower (since day 6 of infection), and their peak of parasitaemia was significantly lower than that of B6 control mice. However, in contrast to B6.IFN-γ- /- mice, B6 mice treated with anti-IFN-γ mAb at day 3 p.i. eliminated blood stage parasites at the same time as the B6 mice which received irrelevant rat IgG (Figure 5B). Splenocyte Figure 3 Parasi- taemia of B6 mice treated with 2 mg of anti- IFN-γ mAb at day 4 p.i. with 4,000 sporozoites of P. yoelii 265BY. Regarding their parasitaemia curves, two groups of mice were identified: "high" and "low" parasitaemia. D. Parasitaemia of B6 and B6.IFN-γ-/- mice following injection of 106 P. yoelii 265BY pRBCs. Each graph is representative of two independent experiments with 3 to 9 mice per time point. Results are expressed as mean value ± SD. The statistical differences between groups were determined with the Mann-Whitney test (* = p < 0.05). Page 8 of 11 (page number not for citation purposes) http://www.malariajournal.com/content/8/1/128 http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 Discussion had no effect on the parasitaemia, and injection at day 4 p.i. lowered the parasitaemia of approximatively half of the treated mice. These results show that early immune events, occurring in a very narrow window of time are crucial for the control of the subsequent growth of P. yoelii in B6 mice. Moreo- ver, these data reflect the importance of the timing and quantity of the release of an inflammatory mediator, i.e. in appropriate amounts at appropriate time during the inflammatory response [26]. anti-IFN-γ mAb delay the course of parasitaemia and decrease the level of the parasite load. However, only the constitutive deficiency of IFN-γ makes the parasitaemia last longer. This latter observation is reminiscent of the one made in B6 mice infected with P. chabaudi pRBCs, showing that IFN-γ favours the appropriate antibody response required to eliminate efficiently the blood-stage parasites after day 20 of infection [24,25]. The results pre- sented here thus argue in favour of two opposite roles played by IFN-γ during P. yoelii primary infection in B6 mice: in the first days following sporozoites injection, IFN-γ would be deleterious for the host as it favours the parasitaemia, while, after the first week of infection, it would be beneficial for the host, helping to clear effi- ciently the blood-stage parasites. In addition, regarding the role of IFN-γ early in infection, only the injection of anti-IFN-γ Abs in B6 mice at day 3 p.i., two days before the cytokine increases in the serum, lowered the parasitaemia in all the treated mice. Injection of the same concentration of antibody performed at days 5 or 7 p.i. had no effect on the parasitaemia, and injection at day 4 p.i. lowered the parasitaemia of approximatively half of the treated mice. These results show that early immune events, occurring in a very narrow window of time are crucial for the control of the subsequent growth of P. yoelii in B6 mice. Moreo- ver, these data reflect the importance of the timing and quantity of the release of an inflammatory mediator, i.e. in appropriate amounts at appropriate time during the inflammatory response [26]. The concentration of IFN-γ in the blood rapidly decreased. This could be due to its capture by target receptors and/or to a quick shut down of its production by regulatory mechanisms, such as TGF-β and/or IL-10 production [13,27]. Authors' contributions VS, JR and SP designed the study. VS and JR performed the experiments and wrote the manuscript. OG performed the statistical analyses. EB performed some of the ELISA experiments. PAC and SP helped to write the manuscript. All authors read and approved the final manuscript. Discussion Finally, this early burst of IFN-γ was induced by the pre- erythrocytic stage, showing the impact of this latter one on the immune response taking place during the subsequent blood stage. Conclusion Overall, the results presented here argue in favour of two opposite roles for IFN-γ during P. yoelii primary infection in B6 mice and show that an early IFN-γ response can be deleterious for the host regarding the control of parasite growth. Taken together with the recent report published by Couper et al [13], these data challenge the prevailing idea that IFN-γ is an essential mediator of protection in malaria. These results also support the idea that the com- mitment towards protection or pathology takes place very early after infection since we show that precocious immune events impact on the control of the parasite's growth much later in infection. Finally, these data highlight the complex regulation of the primary immune response to P. yoelii sporozoite-induced malaria, since it is shown that the pre-erythrocytic stage induces an early burst of IFN-γ that, directly or indirectly, favours parasitaemia. These results also strenghten the need for studying the immune response to Plasmodium in models closest as possible to the physiology of the natural infection. Regarding the cell types involved in this early burst of IFN- γ it appears that conventional splenic CD4 T cells are likely to be the main source of this cytokine at day 5 p.i.. These results are coherent with observations made in infected B6.RAG2-/- mice (lacking T and B cells), which showed no early peak of IFN-γ and exhibited a lower par- asitaemia than B6 mice following injection of P. yoelii spo- rozoites [11]. Nevertheless, the first cells to produce IFN-γ are innate NK, NKT and γδT cells, which responded simul- taneously, suggesting their concomitant activation. Fur- ther investigations are required to determine whether this IFN-γ response by innate and CD4 lymphocytes depends on the pre-erythrocytic stage only, or both the pre-eryth- rocytic and the erythrocytic stages. Competing interests The authors declare that they have no competing interests. Discussion chabaudi model [8,9], the hypothesis that the parasite stage used for infection could explain these discrepancies was tested. To clarify this point, B6 and B6.IFN-γ-/- mice were infected with P. yoelii pRBCs and their survival and parasitaemia were monitored. As shown in Figure 5D, B6.IFN-γ-/- mice infected with pRBCs behave in the same way as when infected with sporozoites: they survived the infection, their parasitaemia was significantly lower after day 9 of infection, and it lasted longer than that of B6 mice. This shows that the results obtained are not related to the parasite stage used to initiate the infec- tion. As the parasitaemia was lower in the absence of IFN-γ, this suggests that IFN-γ would either facilitate P. yoelii's repli- cation, and/or disadvantage its control by the immune system. One possibility could be that IFN-γ favours the parasitaemia through promoting the production of P. yoe- lii's blood stage target cell, namely the reticulocyte. Fur- ther investigations are required to address precisely the mode of action of IFN-γ on the erythrocytic stage in this model. It is interesting to note that both the constitutive genetic deficiency in IFN-γ and the in vivo neutralization with Page 9 of 11 (page number not for citation purposes) Malaria Journal 2009, 8:128 http://www.malariajournal.com/content/8/1/128 http://www.malariajournal.com/content/8/1/128 anti-IFN-γ mAb delay the course of parasitaemia and decrease the level of the parasite load. However, only the constitutive deficiency of IFN-γ makes the parasitaemia last longer. This latter observation is reminiscent of the one made in B6 mice infected with P. chabaudi pRBCs, showing that IFN-γ favours the appropriate antibody response required to eliminate efficiently the blood-stage parasites after day 20 of infection [24,25]. The results pre- sented here thus argue in favour of two opposite roles played by IFN-γ during P. yoelii primary infection in B6 mice: in the first days following sporozoites injection, IFN-γ would be deleterious for the host as it favours the parasitaemia, while, after the first week of infection, it would be beneficial for the host, helping to clear effi- ciently the blood-stage parasites. In addition, regarding the role of IFN-γ early in infection, only the injection of anti-IFN-γ Abs in B6 mice at day 3 p.i., two days before the cytokine increases in the serum, lowered the parasitaemia in all the treated mice. Injection of the same concentration of antibody performed at days 5 or 7 p.i. References 1. Stevenson MM, Riley EM: Innate immunity to malaria. Nat Rev Immunol 2004, 4:169-180. 22. Hansen DS, Evans KJ, D'Ombrain MC, Bernard NJ, Sexton AC, Buck- ingham L, Scalzo AA, Schofield L: The natural killer complex reg- ulates severe malarial pathogenesis and influences acquired immune responses to Plasmodium berghei ANKA. Infect Immun 2005, 73:2288-2297. 2. Schofield L, Grau GE: Immunological processes in malaria pathogenesis. Nat Rev Immunol 2005, 5:722-35. p g 3. Walther M, Woodruff J, Edele F, Jeffries D, Tongren JE, King E, Andrews L, Bejon P, Gilbert SC, De Souza JB, Sinden R, Hill AV, Riley EM: Innate immune responses to human malaria: heteroge- neous cytokine responses to blood-stage Plasmodium falci- parum correlate with parasitological and clinical outcomes. J Immunol 2006, 177:5736-45. 23. 23. Pied S, Roland J, Louise A, Voegtle D, Soulard V, Mazier D, Cazenave PA: Liver CD4-CD8- NK1.1+ TCR alpha beta intermediate cells increase during experimental malaria infection and are able to exhibit inhibitory activity against the parasite liver stage in vitro. J Immunol 2000, 164:1463-1469. 4. Amani V, Vigario AM, Belnoue E, Marussig M, Fonseca L, Mazier D, Rénia L: Involvement of IFN-γ receptor-mediated signaling in pathology and anti-malarial immunity induced by Plasmo- dium berghei infection. Eur J Immunol 2000, 30:1646-1655. g J 24. Su Z, Stevenson MM: IL-12 is required for antibody-mediated protective immunity against blood-stage Plasmodium chabaudi AS malaria infection in mice. J Immunol 2002, 168:1348-1355. g J 5. Singh RP, Kashiwamura S, Rao P, Okamura H, Mukherjee A, Chauhan VS: The role of IL-18 in blood-stage immunity against murine malaria Plasmodium yoelii 265 and Plasmodium berghei ANKA. J Immunol 2002, 168:4674-4681. 25. Seixas E, Ostler D: Plasmodium chabaudi chabaudi (AS): differ- ential cellular responses to infection in resistant and suscep- tible mice. Exp Parasitol 2005, 110:394-405. p 26. Riley EM, Wahl S, Perkins DJ, Schofield L: Regulating immunity to malaria. Parasite Immunol 2006, 28:35-49. J 6. Langhorne J, Albano FR, Hensmann M, Sanni L, Cadman E, Voisine C, Sponaas AS: Dendritic cells, pro-inflammatory responses, and antigen presentation in a rodent malaria infection. Immunol Rev 2004, 201:35-47. 27. de Souza JB, Riley EM, Omer FM: Differential induction of TGF-β regulates proinflammatory cytokine production and deter- mines the outcome of lethal and nonlethal Plasmodium yoelii infections. J Immunol 2003, 171:5430-5436. , 7. http://www.malariajournal.com/content/8/1/128 http://www.malariajournal.com/content/8/1/128 Malaria Journal 2009, 8:128 platform, Institut Pasteur, Paris) for their help with the setting of parasitae- mia analysis by flow cytometry and Lobna Gaayeb for critical reading of the manuscript. We are very grateful to Prof. Monique Capron for her constant help. 19. Lee LG, Chen CH, Chiu LA: Thiazole orange: a new dye for reticulocyte analysis. Cytometry 1986, 7:508-517. 20. 20. Watanabe H, Ohtsuka K, Kimura M, Ikarashi Y, Ohmori K, Kusumi A, Ohteki T, Seki S, Abo T: Details of an isolation method for hepatic lymphocytes in mice. J Immunol Methods 1992, 146:145-154. 21. Brigl M, Brenner MB: CD1: antigen presentation and T cell function. Annu Rev Immunol 2004, 22:817-890. Acknowledgements g This work was partly supported by the "Programme de Recherche Fonda- mentale en Microbiologie, Maladies Infectieuses et Parasitaires (PRFMMIP) AO 2000" of the French Ministry of Research. V.S. was a recipient of fel- lowships from the French Ministry of Education and Research, the "Fonda- tion pour la Recherche Médicale" and the "Fonds Inkermann" from the "Fondation de France". It is also interesting to note that CD1d-restricted NKT cells are not necessary for the initiation of this early production of IFN-γ. Together with previously published results showing that CD1d-independent NKT cells are induced in the spleen of P. yoelii-infected B6 mice and are biased towards the production of Th1-type cytokines [12], these data strengthen the idea that CD1d-independent NKT cells are involved in P. yoelii-induced early Th1 immune response. We thank Isabelle Thiery and Catherine Bourgouin for providing us with Anopheles stephensi mosquitoes (CEPIA, Institut Pasteur, Paris). We are grateful to Christèle Sellier for her help with mosquito infection and dissec- tion and to Jean-François Bureau for the gift of B6.IFN-γ-/- mice. We thank Sébastien Bagot, Anne Louise, and Marie-Christine Wagner (Cytometry Page 10 of 11 (page number not for citation purposes) Page 10 of 11 (page number not for citation purposes) References Shear HL, Srinivasan R, Nolan T, Ng C: Role of IFN-γ in lethal and nonlethal malaria in susceptible and resistant murine hosts. J Immunol 1989, 143:2038-2044. J 8. Meding SJ, Cheng SC, Simon-Haarhaus B, Langhorne J: Role of gamma interferon during infection with Plasmodium chabaudi chabaudi. Infect Immun 1990, 58:3671-3678. f 9. Su Z, Stevenson MM: Central role of endogenous gamma inter- feron in protective immunity against blood-stage Plasmo- dium chabaudi AS infection. Infect Immun 2000, 68:4399-4406. f , 10. Urban BC, Ing R, Stevenson MM: Early interactions between blood-stage plasmodium parasites and the immune system. Curr Top Microbiol Immunol 2005, 297:25-70. p 11. Roland J, Soulard V, Sellier C, Drapier AM, Di Santo JP, Cazenave PA, Pied S: NK cell responses to Plasmodium infection and con- trol of intrahepatic parasite development. J Immunol 2006, 177:1229-1239. 12. Soulard V, Roland J, Sellier C, Gruner AC, Leite-de-Moraes M, Fran- etich JF, Renia L, Cazenave PA, Pied S: Primary infection of C57BL/6 mice with Plasmodium yoelii induces a heterogene- ous response of NKT cells. Infect Immun 2007, 75:2511-2522. p 13. Couper KN, Blount DG, Hafalla JC, van Rooijen N, de Souza JB, Riley EM: Macrophage-mediated but gamma-interferon-independ- ent innate immune responses control the primary wave of Plasmodium yoelii parasitaemia. Infect Immun 2007, 75:5806-5818. 14. Carnaud C, Lee D, Donnars O, Park SH, Beavis A, Koezuka Y, Bend- elac A: Cutting edge: Cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J Immu- nol 1999, 163:4647-50. 15. Di Santo JP, Kuhn R, Muller W: Common cytokine receptor gamma chain (gamma-c)-dependent cytokines: understand- ing in vivo functions by gene targeting. Immunol Rev 1995, 148:19-34. References Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 11 of 11 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge 16. Mazier D, Beaudoin RL, Mellouk S, Druilhe P, Texier B, Trosper J, Miltgen F, Landau I, Paul C, Brandicourt O: Complete develop- ment of hepatic stages of Plasmodium falciparum in vitro. Sci- ence 1985, 227:440-442. 17. Jouin H, Goguet de la Salmoniere YO, Behr C, Huyin Qan Dat M, Michel JC, Sarthou JL, Pereira da Silva L, Dubois P: Flow cytometry detection of surface antigens on fresh, unfixed red blood cells infected by Plasmodium falciparum. J Immunol Methods 1995, 179:1-12. 18. Jouin H, Daher W, Khalife J, Ricard I, Puijalon OM, Capron M, Dive D: Double staining of Plasmodium falciparum nucleic acids with hydroethidine and thiazole orange for cell cycle stage analysis by flow cytometry. Cytometry A 2004, 57:34-38.
https://openalex.org/W2335331919	https://figshare.com/ndownloader/files/4794175	Aragonese	null	New Rhinocerotidae from the Kisingiri localities (lower Miocene of western Kenya)	Journal of vertebrate paleontology	2,016	cc-by	1,947	New Rhinocerotidae from the Kisingiri localities (Lower Miocene of Western Kenya) Denis Geraads, Thomas Lehmann, Daniel J. Peppe, and Kieran McNulty Journal of Vertebrate Paleontology Supplementary data Consensus tree of the two most parsimonious trees (L = 1546; ci = 18; ri = 58) obtained by TNT from matrix Geraads&al_TNTmatrix.txt New Rhinocerotidae from the Kisingiri localities (Lower Miocene of Western Kenya) Denis Geraads, Thomas Lehmann, Daniel J. Peppe, and Kieran McNulty Supplementary data Consensus tree of the two most parsimonious trees (L = 1546; ci = 18; ri = 58) obtained by TNT from matrix Geraads&al_TNTmatrix.txt S t r i c t c o n s e n s u s o f 2 t r e e s ( 0 t a x a e x c l u d e d ) S t r i c t c o n s e n s u s o f 2 t r e e s ( 0 t a x a e x c l u d e d ) S t r i c t c o n s e n s u s o f 2 t r e e s ( 0 t a x a e x c l u d e d ) M o l a s s i t h e r i u m _ a l b i g e n s e T u r k a n a t h e r i u m _ a c u t i r o s t r a t u m T r i g o n i a s _ o s b o r n i T e l e o c e r a s _ f o s s i g e r S u b h y r a c o d o n _ o c c i d e n t a l i s S i n o t h e r i u m _ l a g r e l i i S h e n n o n g t h e r i u m _ h y p s o d o n t u s S h a n s i r h i n u s _ r i n g s t r o m i R u s i n g a c e r o s _ l e a k e y i R o n z o t h e r i u m _ f i l h o l i R h i n o c e r o s _ u n i c o r n i s R h i n o c e r o s _ s o n d a i c u s P r o t a c e r a t h e r i u m _ m i n u t u m P r o s a n t o r h i n u s _ d o u v i l l e i P l e u r o c e r o s _ p l e u r o c e r o s P l e u r o c e r o s _ b l a n f o r d i P l e s i a c e r a t h e r i u m _ m i r a l l e s i P l e s i a c e r a t h e r i u m _ g r a c i l e P a r e l a s m o t h e r i u m _ s c h a n s i e n s e P a r e l a s m o t h e r i u m _ l i n x i a e n s e O u g a n d a t h e r i u m _ n a p a k e n s e N i n x i a t h e r i u m _ l o n g i r h i n u s N i n x i a t h e r i u m _ e u r y r h i n u s M e s a c e r a t h e r i u m _ w e l c o m m i M e s a c e r a t h e r i u m _ p a u l h i a c e n s e M e s a c e r a t h e r i u m _ g a i m e r s h e i m e n s e M e n o c e r a s _ a r i k a r e n s e L a r t e t o t h e r i u m _ s a n s a n i e n s e K e n y a t h e r i u m _ b i s h o p i V i c t o r i a c e r o s _ k e n y e n s i s K a r u n g u K a c h u k u I r a n o t h e r i u m _ m o r g a n i H y r a c h y u s _ e x i m i u s H o p l o a c e r a t h e r i u m _ t e t r a d a c t y l u m H i s p a n o t h e r i u m _ b e o n e n s e H i s p a n o t h e r i u m _ t u n g u r e n s e H i s p a n o t h e r i u m _ m a t r i t e n s e H i s p a n o t h e r i u m _ c o r c o l e n s e G o b i t h e r i u m _ m o n g o l i e n s e G a i n d a t h e r i u m _ b r o w n i E l a s m o t h e r i u m _ s i b i r i c u m E l a s m o t h e r i u m _ c a u c a s i c u m D i c e r o r h i n u s _ s u m a t r e n s i s D i c e r o s _ b i c o r n i s D i c e r a t h e r i u m _ a r m a t u m D i a c e r a t h e r i u m _ a g i n e n s e C o e l o d o n t a _ a n t i q u i t a t i s S u b c h i l o t h e r i u m _ i n t e r m e d i u m C h i l o t h e r i d i u m _ p a t t e r s o n i C h i l o t h e r i u m _ w i m a n i C h i l o t h e r i u m _ a n d e r s s o n i C e r a t o t h e r i u m _ s i m u m C a e m e n t o d o n _ o e t t i n g e n a e C a e m e n t o d o n _ c a u c a s i c u m B r a c h y p o t h e r i u m _ m i n o r B u g t i r h i n u s _ p r a e c u r s o r B r a c h y p o t h e r i u m _ b r a c h y p u s B e l i a j e v i n a _ t e k k a y a i B e g e r t h e r i u m _ g r i m m i B e g e r t h e r i u m _ b o r i s s i a k i A p r o t o d o n _ f a t e h j a n g e n s e A l i c o r n o p s _ s i m o r r e n s e A l i c o r n o p s _ c o m p l a n a t u m A l i c o r n o p s _ a l f a m b r e n s e A c e r o r h i n u s _ p a l a e o s i n e n s i s A c e r o r h i n u s _ y u a n m o u e n s i s A c e r o r h i n u s _ t s a i d a m e n s i s A c e r o r h i n u s _ z e r n o w i A c e r o r h i n u s _ f u g u e n s i s A c e r o r h i n u s _ l u f e n g e n s i s A c e r o r h i n u s _ h e z h e n g e n s i s A c e r a t h e r i u m _ i n c i s i v u m A c e r a t h e r i u m _ h u a d e e n s i s T a p i r u s _ t e r r e s t r i s A l i c o r n o p s _ s i m o r r e n s e A l i c o r n o p s _ c o m p l a n a t u m A p r o t o d o n
https://openalex.org/W2574986000	https://iris.unimore.it/bitstream/11380/1209646/2/PhysRevLett.118.111803.pdf	Latin	null	Observation of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:msubsup><mml:mi>B</mml:mi><mml:mi>c</mml:mi><mml:mo>+</mml:mo></mml:msubsup><mml:mo stretchy="false">→</mml:mo><mml:msup><mml:mi>D</mml:mi><mml:mn>0</mml:mn></mml:msup><mml:msup><mml:mi>K</mml:mi><mml:mo>+</mml:mo></mml:msup></mml:math> Decays	Physical review letters	2,017	cc-by	8,548	DOI: 10.1103/PhysRevLett.118.111803 The decay Bþ →¯D0πþ is used for normalization. Since the ratio of production rates for Bþc and Bþ mesons within the LHCb acceptance, fc=fu, is unknown, the measured observables are The Bþc meson is the only ground-state meson consisting of two heavy quarks of different flavor, namely a ¯b and a c quark. As such, its formation in pp collisions is suppressed relative to the lighter B mesons. Unlike B0, Bþ and B0s mesons, the b-quark decay accounts for only ∼20% of the Bþc width [1]. Around 70% of its width is due to c-quark decays, where the c-quark transition has been observed with Bþc →B0sπþ decays [2]. This leaves ∼10% for ¯bc →Wþ →¯qq annihilation amplitudes, which can be unambiguously probed in charmless final states. No charm- less Bþc decays have been reported to date, although searches show an indication at the level of 2.4 standard deviations (σ) [3]. RDðÞ0h ¼ fc fu × BðBþc →DðÞ0hþÞ; ð1Þ ð1Þ where h is π or K and BðBþc →DðÞ0hþÞ represents the corresponding branching fraction. The four observables are measured with a simultaneous fit to the D0πþ and D0Kþ invariant mass distributions. Theoretical estimates for BðBþc →J=ψπþÞ range from 6.0 × 10−4 [9] to 1.8 × 10−3 [10], which implies fc=fu values in the range 0.004–0.012 using the production ratio measured in Ref. [5] and the branching fraction BðBþ →J=ψKþÞ [11]. Estimates for BðBþc →D0KþÞ vary from 1.3 × 10−7 [6] to 6.6 × 10−5 [8], while estimates for BðBþc →D0πþÞ vary from 2.3 × 10−7 [6] to 2.3 × 10−6 [7]. Using Eq. (1), the expectation for RD0π is seen to cover the range 9 × 10−10 −3 × 10−8, while RD0K covers the range 5 × 10−10 −8 × 10−7. To test QCD factorization and explore the new physics potential of Bþc decays, rarer decays such as suppressed tree-level b →u transitions and b →s loop-mediated (penguin) decays can be studied, where the charm quantum number remains unchanged. The simplest decay is the color-allowed Bþc →DðÞ0πþ decay, illustrated in Fig. 1(a). The expected branching fraction for this decay is a factor jVub=Vcbj2 ≈0.007 lower than the favored b →c and color-allowed Bþc →J=ψπþ decay [4,5], placing this mode at the limit of sensitivity with current LHCb data. Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. week ending 17 MARCH 2017 week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) Full author list given at the end of the article. Observation of Bþc →D0Kþ Decays R. Aaij et al. (LHCb Collaboration) j (LHCb Collaboration) ( ) (Received 7 January 2017; revised manuscript received 17 February 2017; published 15 March 2017) (Received 7 January 2017; revised manuscript received 17 February 2017; published 15 March 2017) Using proton-proton collision data corresponding to an integrated luminosity of 3.0 fb−1, recorded by the LHCb detector at center-of-mass energies of 7 and 8 TeV, the Bþc →D0Kþ decay is observed with a statistical significance of 5.1 standard deviations. By normalizing to Bþ →¯D0πþ decays, a measurement of the branching fraction multiplied by the production rates for Bþc relative to Bþ mesons in the LHCb acceptance is obtained, RD0K ¼ ðfc=fuÞ × BðBþc →D0KþÞ ¼ ð9.3þ2.8 −2.5 0.6Þ × 10−7, where the first uncertainty is statistical and the second is systematic. This decay is expected to proceed predominantly through weak annihilation and penguin amplitudes, and is the first Bþc decay of this nature to be observed. DOI: 10.1103/PhysRevLett.118.111803 DOI: 10.1103/PhysRevLett.118.111803 A two-dimensional optimization is performed to determine the second stage BDT requirements for the two-body and four-body modes, where the signal S is compared to the number of background events B in data using a figure of merit S=ð ﬃﬃﬃﬃ B p þ 3=2Þ [25]. The value of B is determined within 50 MeV=c2 of the known Bþc mass. No PID information is used in the BDT training, so that the efficiency for B →D0Kþ and B →D0πþ decays is similar. The use of BDTs to select signal decays was validated by comparing the efficiency of the BDT requirements for Bþ →¯D0πþ decays in data and simulation, where close agreement was found across a wide range of BDT cuts. The purity of the selection is further improved by requiring all kaons and pions in the D0 decay to be identified with a PID selection that has an efficiency of about 85% per particle. RDðÞ0h ¼ N ðBþc →DðÞ0hþÞ N ðBþ →¯D0πþÞ × BðBþ →¯D0πþÞ × ξ; ð2Þ ð2Þ where N ðBþc →DðÞ0hþÞ represents the Bþc →DðÞ0hþ yield, N ðBþ →¯D0πþÞ represents the yield of Bþ →¯D0πþ normalization decays, BðBþ →¯D0πþÞ is the normalization mode branching fraction [11], and ξ is the ratio of efficiencies for reconstructing and selecting Bþ and Bþc mesons decaying to these final states. The LHCb detector is a single-arm forward spectrometer covering the pseudorapidity range 2 < η < 5, described in detail in Refs. [12,13]. The detector allows the reconstruction of both charged and neutral particles. For this analysis, the ring-imaging Cherenkov (RICH) detectors [14], distinguishing pions, kaons, and protons, are particu- larly important. Simulated events are produced using the software described in Refs. [15–22]. After reconstruction of the D0 meson candidate, the same selection is applied to the Bþc and Bþ candidates. The invariant mass of the D0 candidate must be within 25 MeV=c2 of its known value [11]. The other hadron originating from the B decay must have transverse momen- tum (pT) in the range 0.5–10.0 GeV=c and momentum (p) in the range 5–100 GeV=c, ensuring that the track is within the kinematic coverage of the RICH detectors that provide particle identification (PID) information. A kinematic fit is performed to each decay chain [23], with vertex constraints applied to both the B and D vertices, and the D0 candidate mass constrained to its known value. DOI: 10.1103/PhysRevLett.118.111803 However, this expectation may be enhanced by penguin and weak annihilation amplitudes, which will be more pronounced in the Bþc →DðÞ0Kþ mode [see Fig. 1(b) and 1(c)]. This motivates a search for the Bþc →DðÞ0Kþ and Bþc → DðÞ0πþ decays, particularly as the branching fraction estimates in the literature vary considerably [6–8]. This Letter reports a search for Bþc →D0πþ and Bþc → D0Kþ decays in pp collision data corresponding to FIG. 1. Tree (a), penguin (b), and weak annihilation (c) diagrams for the decays studied. In each case, the meson appearing before the comma denotes the favored decay. 111803-1 © 2017 CERN, for the LHCb Collaboration 111803-1 0031-9007=17=118(11)=111803(9) week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) combinations from data with invariant mass in the range 5900–7200 MeV=c2. For the first BDT, background can- didates with a D0 invariant mass more than 30 MeV=c2 away from the known D0 mass are used. In the second BDT, background candidates with a D0 invariant mass within 25 MeV=c2 of the known D0 mass are used. A loose cut on the classifier response of the first BDT is applied before training the second one. This focuses the second BDT training on backgrounds enriched with fully reconstructed D0 mesons. integrated luminosities of 1.0 and 2.0 fb−1 taken by the LHCb experiment at center-of-mass energies of 7 and 8 TeV, respectively, where the D0 meson is reconstructed in the Cabibbo-favored final states D0 →K−πþ or D0 →K−πþπ−πþ (inclusion of charge-conjugate proc- esses is implied throughout). Partially reconstructed Bþc →ðD0 →D0fπ0; γgÞhþ decays, where the neutral particle indicated in braces is not considered in the invariant mass calculation, are treated as additional signal channels. The number of Bþc decays is normalized by comparison to the number of Bþ →½ ¯D0 →Kþπ−ðπþπ−Þπþ decays. A fit to the invariant mass distribution of DðÞ0hþ candidates in the range 5800–6900 MeV=c2 enables a measurement of The inputs to all BDTs include properties of each particle (p, pT, and the IP significance) and additional properties of the B and D0 candidates (decay time, flight distance, decay vertex quality, radial distance between the decay vertex and the PV, and the angle between the reconstructed momentum vector and the line connecting the production and decay vertices). DOI: 10.1103/PhysRevLett.118.111803 The red solid curve illustrates Bþc →D0Kþ decays, the red dashed curve illustrates Bþc →D0Kþ decays, the green dashed curve represents Bþc →D0πþ decays, the gray shaded region represents partially reconstructed background decays, the cyan dashed line represents the combinatorial background, and the total PDF is displayed as a blue solid line. The small drop visible in the total Bþc →DðÞ0πþ PDF around the Bþc mass arises from the fact that the fit finds a small negative value for the Bþc →D0πþ yield. Partially reconstructed decays form a background at invariant masses lower than that of the signal peak. This background is described by a combination of parametric PDFs, with yield and shape parameters that are allowed to vary. A linear function describes the combinatorial back- ground. The yield of Bþ →¯D0Kþ decays, where the kaon is misidentified as a pion, is fixed using a simultaneous fit to correctly identified Bþ →¯D0Kþ events. Using a data- driven analysis of approximately 20 million Dþ decays reconstructed as Dþ →D0πþ, D0 →K−πþ, the proba- bility of kaon misidentification is determined to be 32%. The invariant mass fits to Bþ →ð ¯D0 →Kþπ−Þπþ and Bþ →ð ¯D0 →Kþπ−πþπ−Þπþ decays determine a total observed yield N ðBþ →¯D0πþÞ ¼ 309462 550. two particles are missed, with shape parameters taken from simulated Bþ →D0πþπ0 decays and scaled to account for the different momenta of the decay products in Bþc and Bþ decays. y ð Þ To measure N ðBþc →DðÞ0hþÞ, a simultaneous invariant mass fit to the Bþc →D0πþ and Bþc →D0Kþ samples is performed in the region 5800–6900 MeV=c2. Two-body and four-body D -decay candidates are included, where a Gaussian PDF describes the fully reconstructed Bþc signals. The mean of this Gaussian is fixed to the known Bþc mass [11]. The width of the Bþc →D0πþ PDF is taken from a fit to suppressed Bþ →ð ¯D0 →πþK−Þπþ decays, scaled up by a factor 1.3 to account for the difference in momenta of the decay products in Bþc →D0πþ and Bþ →¯D0πþ decays. The width of the Bþc →D0Kþ peak is related to that of Bþc →D0πþ decays by the ratio of the widths of the Bþ →¯D0Kþ and Bþ →¯D0πþ peaks found in the normali- zation mode fits. DOI: 10.1103/PhysRevLett.118.111803 The Bþ (Bþc ) meson candidates with an invariant mass in the interval 5080–5900 MeV=c2 (5800–6900 MeV=c2) and with a proper decay time above 0.2 ps are retained. Each B candidate is associated with the primary vertex (PV) to which it has the smallest impact parameter (IP), defined as the distance of closest approach of the candidate’s trajec- tory to a given PV. p Simulated signal samples are used to evaluate the relative efficiency for selecting Bþc and Bþ decays. The efficiency ratio is ξ ¼ ϵðBþÞ=ϵðBþc Þ, where ϵðBþÞ and ϵðBþc Þ re- present the combined efficiencies of detector acceptance, trigger, reconstruction, and offline selection. As both Bþc and Bþ mesons are required to decay to the same final-state particles, differences between ϵðBþÞ and ϵðBþc Þ arise due to differences in their masses and lifetimes. The Bþc meson lifetime is ð0.507 0.009Þ ps, which is 3.2 times shorter than that of the Bþ meson [11]. This results in a lower Bþc efficiency relative to Bþ by a factor 2.4, due to the proper decay time cut. The Bþc meson is heavier than the Bþ, which reduces by a factor 1.3 the fraction of Bþc decays in which all final-state particles are within the detector acceptance. However, as the BDTs are trained specifically on Bþc simulated decays, the offline selection efficiency is lower for Bþ decays, contributing a relative efficiency of 0.94. Overall, the efficiency ratio is ξ ¼ 3.04 0.16 ð2.88 0.15Þ for the two-body (four-body) D0 decay. The uncertainties are systematic, arising from the use of Two boosted decision tree (BDT) discriminators [24] are used for further background suppression. They are trained using simulated Bþc →½D0 →K−πþðπþπ−Þhþ signal decays and a sample of wrong-sign Kþπ−ðπþπ−Þhþ 111803-2 111803-2 ] 2 c ) [MeV/ ± h 0 D ( m 5800 6000 6200 6400 6600 6800 5 10 15 20 25 ± π 0 D → ± c B LHCb ) 2 c Candidates / (40 MeV/ 5 10 15 20 25 ± K 0 D → ± c B LHCb FIG. 2. Results of the simultaneous fit to the D0Kþ (top plot) and D0πþ (bottom plot) invariant mass distributions in the Bþc mass region, including the D0 →K−πþ and D0 →K−πþπ−πþ final states. Inclusion of the charge conjugate decays is implied. DOI: 10.1103/PhysRevLett.118.111803 The red solid curve illustrates Bþc →D0Kþ decays, the red dashed curve illustrates Bþc →D0Kþ decays, the green dashed curve represents Bþc →D0πþ decays, the gray shaded region represents partially reconstructed background decays, the cyan dashed line represents the combinatorial background, and the total PDF is displayed as a blue solid line. The small drop visible in the total Bþc →DðÞ0πþ PDF around the Bþc mass arises from the fact that the fit finds a small negative value for the Bþc →D0πþ yield. V I E W L E T T E R S week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) ] 2 c ) [MeV/ ± h 0 D ( m 5800 6000 6200 6400 6600 6800 5 10 15 20 25 ± π 0 D → ± c B LHCb ) 2 c Candidates / (40 MeV/ 5 10 15 20 25 ± K 0 D → ± c B LHCb 17 MARCH 2017 finite simulated samples and possible mismodeling of the simulated Bþc lifetime and production kinematics. To measure N ðBþ →¯D0πþÞ, binned maximum like- lihood fits to the invariant mass distributions of selected Bþ candidates are performed, where separate fits are employed for the two-body and four-body ¯D0 modes. The total probability density function (PDF) is built from four contributions. The Bþ →¯D0πþ decays are modeled by the sum of two modified Gaussian functions with asym- metric power-law tails and an additional Gaussian function as used in Ref. [26], all of which share a common peak position. Misidentified Bþ →¯D0Kþ candidates have an incorrect mass assignment and form a distribution dis- placed downward in mass, with a tail extending to lower invariant masses. They are modeled by the sum of two modified Gaussian PDFs with low-mass power-law tails. All PDF parameters are allowed to vary, with the exception of the tail parameters which are fixed to the values found in simulation. FIG. 2. Results of the simultaneous fit to the D0Kþ (top plot) and D0πþ (bottom plot) invariant mass distributions in the Bþc mass region, including the D0 →K−πþ and D0 →K−πþπ−πþ final states. Inclusion of the charge conjugate decays is implied. DOI: 10.1103/PhysRevLett.118.111803 π 0 D R 0 0.5 1 1.5 6 − 10 × -value p 0 0.5 1 LHCb K 0 D R 0 0.5 1 1.5 6 − 10 × -value p 0 0.5 1 LHCb decay. The value of RD0K is at the high end of theoretical predictions [6–8] and an expectation based on the observed Bþc →J=ψπþ yield at LHCb [28]. From Refs. [5] and [11], RJ=ψπ ¼ ð7.0 0.3Þ × 10−6 is obtained. As fc=fu is common to both RJ=ψπ and RD0K, the ratio of branching fractions is measured to be BðBþc →D0KþÞ= BðBþc →J=ψπþÞ ¼ 0.13 0.04 0.01 0.01, where the first uncertainty is statistical, the second is systematic, and the third comes from RJ=ψπ. FIG. 3. CLs p-value distributions for the RD0h observables. The dashed line represents the expected CLs values, where the 1σ and 2σ contours are indicated by the green (dark) and yellow (light) shaded regions, respectively. Upper limits are determined by the points at which the observed CLsþb p-values (black points connected by straight lines) fall below 5% (red solid line). Also displayed are the corresponding CLs ¼ CLsþb=CLb values (blue points connected by straight dotted lines). The absence of the Bþc →D0πþ mode shows that the Bþc →D0Kþ amplitude is not dominated by the tree-level b →u transition shown in Fig. 1(a), but rather by the penguin 1(b) and/or weak annihilation 1(c) diagrams. This result constitutes the first observation of such amplitudes in the decay of a Bþc meson. We express our gratitude to our colleagues in the CERN accelerator departments for the excellent performance of the LHC. We thank the technical and administrative staff at the LHCb institutes. We acknowledge support from CERN and from the national agencies: CAPES, CNPq, FAPERJ and FINEP (Brazil); NSFC (China); CNRS/IN2P3 (France); BMBF, DFG and MPG (Germany); INFN (Italy); FOM and NWO (Netherlands); MNiSW and NCN (Poland); MEN/IFA (Romania); MinES and FASO (Russia); MinECo (Spain); SNSF and SER (Switzerland); NASU (Ukraine); STFC (United Kingdom); NSF (USA). We acknowledge the computing resources that are provided by CERN, IN2P3 (France), KIT and DESY (Germany), INFN (Italy), SURF (Netherlands), PIC (Spain), GridPP (United Kingdom), RRCKI and Yandex LLC (Russia), CSCS (Switzerland), IFIN-HH (Romania), CBPF (Brazil), PL-GRID (Poland) and OSC (USA). We are indebted to the communities behind the multiple open source software packages on which we depend. DOI: 10.1103/PhysRevLett.118.111803 Partially reconstructed Bþc →D0hþ signal decays are modeled using a combination of para- metric PDFs, with yield and shape parameters that are allowed to vary. These decays contribute at lower invariant masses than the fully reconstructed signal decays, as a result of not considering the natural particle in the invariant mass calculation. An additional background component at low invariant mass is included to describe Bþc decays where Misidentified Bþc →D0πþðKþÞ decays in the Bþc →D0KþðπþÞ sample are modeled using the same PDFs as the normalization fits, with widths and peak positions scaled for the decay momentum difference. These shapes are fixed in the fit. Signal decays are split into separate samples with correct and incorrect kaon identification, with a kaon misidentification rate of 7% and a corresponding pion identification efficiency of 91% fixed using the data-driven Dþ analysis described above. An exponential function describes the combinatorial back- ground, which is fitted independently in the Bþc →D0πþ and Bþc →D0Kþ samples. The combinatorial yields, signal yields, and partially reconstructed Bþc →D0hþfπ0g and Bþc →D0hþfπ0g background yields are all free to vary. The fit to data is shown in Fig. 2, where a Bþc →D0Kþ yield of 20 5 events is found. All other signal yields are consistent with zero. To test the significance of each signal yield, CLs hypothesis tests [27] are performed. Upper limits at 95% confidence level (C.L.) are determined by the point 111803-3 week ending 17 MARCH 2017 week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) π 0 D R 0 0.5 1 1.5 6 − 10 × -value p 0 0.5 1 LHCb K 0 D R 0 0.5 1 1.5 6 − 10 × -value p 0 0.5 1 LHCb FIG. 3. CLs p-value distributions for the RD0h observables. The dashed line represents the expected CLs values, where the 1σ and 2σ contours are indicated by the green (dark) and yellow (light) shaded regions, respectively. Upper limits are determined by the points at which the observed CLsþb p-values (black points connected by straight lines) fall below 5% (red solid line). Also displayed are the corresponding CLs ¼ CLsþb=CLb values (blue points connected by straight dotted lines). [1] I. P. Gouz, V. V. Kiselev, A. K. Likhoded, V. I. Romanovsky, and O. P. Yushchenko, Prospects for the Bc studies at LHCb, Phys. At. Nucl. 67, 1559 (2004). y [2] R. Aaij et al. (LHCb Collaboration), Observation of the Decay Bþc →B0sπþ, Phys. Rev. Lett. 111, 181801 (2013). [3] R. Aaij et al. (LHCb Collaboration), Study of Bþc decays to the KþK−πþ final state and evidence for the decay Bþc →χc0πþ, Phys. Rev. D 94, 091102 (2016). [1] I. P. Gouz, V. V. Kiselev, A. K. Likhoded, V. I. Romanovsky, and O. P. Yushchenko, Prospects for the Bc studies at LHCb, Phys. At. Nucl. 67, 1559 (2004). [2] R. Aaij et al. (LHCb Collaboration), Observation of the Decay Bþc →B0sπþ, Phys. Rev. Lett. 111, 181801 (2013). [3] R. Aaij et al. (LHCb Collaboration), Study of Bþc decays to the KþK−πþ final state and evidence for the decay Bþc →χc0πþ, Phys. Rev. D 94, 091102 (2016). DOI: 10.1103/PhysRevLett.118.111803 This is the first observation of the Bþc →D0Kþ 111803-4 week ending 17 MARCH 2017 week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) [18] D. J. Lange, The EvtGen particle decay simulation package, Nucl. Instrum. Methods Phys. Res., Sect. A 462, 152 (2001). [4] V. M. Abazov et al. (D0 Collaboration), Observation of the Bc Meson in the Exclusive Decay Bc →J=ψπ, Phys. Rev. Lett. 101, 012001 (2008). [5] R. Aaij et al. (LHCb Collaboration), Measurement of Bþc Production at ﬃﬃﬃs p ¼ 8 TeV, Phys. Rev. Lett. 114, 132001 (2015). [19] J. Allison et al. (Geant4 Collaboration), Geant4 develop- ments and applications, IEEE Trans. Nucl. Sci. 53, 270 (2006). ( ) [6] H. F. Fu, Y. Jiang, C. S. Kim, and G.-L. Wang, Probing non- leptonic two-body decays of Bc meson, J. High Energy Phys. 06 (2011) 15. [20] S. Agostinelli et al. (Geant4 Collaboration), Geant4: A simulation toolkit, Nucl. Instrum. Methods Phys. Res., Sect. A 506, 250 (2003). y ( ) [7] D.-S. Du and Z.-T. Wei, Space-like Penguin effects in Bc decays, Eur. Phys. J. C 5, 705 (1998). [21] M. Clemencic, G. Corti, S. Easo, C. R Jones, S. Miglioranzi, M. Pappagallo, and P. Robbe, The LHCb simulation application, Gauss: Design, evolution and experience, J. Phys. Conf. Ser. 331, 032023 (2011). [8] J. Zhang and X.-Q. Yu, Branching ratio and CP violation of Bc →DK decays in the perturbative QCD approach, Eur. Phys. J. C 63, 435 (2009). [22] C.-H. Chang, C. Driouichi, P. Eerola, and X.-G. Wu, BCVEGPY: An event generator for hadronic production of the Bc meson, Comput. Phys. Commun. 159, 192 (2004). [9] D. Ebert, R. N. Faustov, and V. O. Galkin, Weak decays of the Bc meson to charmonium and D mesons in the relativistic quark model, Phys. Rev. D 68, 094020 (2003). [23] W. D. Hulsbergen, Decay chain fitting with a Kalman filter, Nucl. Instrum. Methods Phys. Res., Sect. A 552, 566 (2005). [10] C.-H. Chang and Y.-Q. Chen, Decays of the Bc meson, Phys. Rev. D 49, 3399 (1994). [11] C. Patrignani et al. (Particle Data Group Collaboration), Review of particle physics, Chin. Phys. C 40, 100001 (2016). [24] B. P. Roe, H.-J. Yang, J. Zhu, Y. Liu, I. Stancu, and G. DOI: 10.1103/PhysRevLett.118.111803 Individual groups or members have received support from AvH Foundation (Germany), EPLANET, Marie Skłodowska-Curie Actions and ERC (European Union), Conseil Général de Haute- Savoie, Labex ENIGMASS and OCEVU, Région Auvergne (France), RFBR and Yandex LLC (Russia), GVA, XuntaGal and GENCAT (Spain), Herchel Smith Fund, The Royal Society, Royal Commission for the Exhibition of 1851 and the Leverhulme Trust (United Kingdom). at which the p-value falls below 5%. All free variables in the fit are considered as nuisance parameters in this procedure. The p-value distributions for each RD0h meas- urement are shown in Fig. 3. The Bþc →D0hþ modes demonstrate no excess, and the RD0h CLs confidence intervals are determined similarly to that of RD0π. The upper limits at 95% confidence level found for RD0π, RD0π, and RD0K are RD0π < 3.9 × 10−7; RD0π < 1.1 × 10−6; RD0K < 1.1 × 10−6: RD0π < 3.9 × 10−7; RD0π < 1.1 × 10−6; RD0K < 1.1 × 10−6: The systematic uncertainties affecting the measurements are found to be much smaller than the statistical uncer- tainty, and do not alter the above upper limits. In the case of RD0K, the observed signal is of much higher significance. To determine the full uncertainty for RD0K, the systematic uncertainties affecting the measure- ment are accounted for. A systematic uncertainty of 1.1 × 10−8 is incurred from the use of fixed terms in the invariant mass fit. According to Eq. (2), several terms with associated relative uncertainties scale the measured signal yield: ξ with 5.3% uncertainty, BðBþ →¯D0πþÞ with 3.1% uncertainty [11], and N ðBþ →¯D0πþÞ with 0.14% uncer- tainty. The total systematic uncertainty, given by the sum in quadrature, is 6.2%. To determine the significance of the Bþc →D0Kþ peak, a likelihood scan is performed. The resulting −Δ logðLÞ value for the RD0K ¼ 0 hypothesis corresponds to a statistical significance of ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ −2Δ logðLÞ p ¼ 5.1σ for the signal. The final result is RD0K ¼ ð9.3þ2.8 −2.5 0.6Þ × 10−7; [3] R. Aaij et al. (LHCb Collaboration), Study of Bþc decays to the KþK−πþ final state and evidence for the decay Bþc →χc0πþ, Phys. Rev. D 94, 091102 (2016). where the first uncertainty is statistical and the second is systematic. R. Aaij,40 B. Adeva,39 M. Adinolfi,48 Z. Ajaltouni,5 S. Akar,59 J. Albrecht,10 F. Alessio,40 M. Alexander,53 S. Ali,43 G. Alkhazov,31 P. Alvarez Cartelle,55 A. A. Alves Jr.,59 S. Amato,2 S. Amerio,23 Y. Amhis,7 L. An,3 L. Anderlini,18 G. Andreassi,41 M. Andreotti,17,g J. E. Andrews,60 R. B. Appleby,56 F. Archilli,43 P. d’Argent,12 J. Arnau Romeu,6 A. Artamonov,37 M. Artuso,61 E. Aslanides,6 G. Auriemma,26 M. Baalouch,5 I. Babuschkin,56 S. Bachmann,12 J. J. Back,50 A. Badalov,38 C. Baesso,62 S. Baker,55 V. Balagura,7,c W. Baldini,17 R. J. Barlow,56 C. Barschel,40 S. Barsuk,7 W. Barter,56 F. Baryshnikov,32 M. Baszczyk,27 V. Batozskaya,29 B. Batsukh,61 V. Battista,41 A. Bay,41 L. Beaucourt,4 J. Beddow,53 F. Bedeschi,24 I. Bediaga,1 L. J. Bel,43 V. Bellee,41 N. Belloli,21,i K. Belous,37 I. Belyaev,32 E. Ben-Haim,8 G. Bencivenni,19 S. Benson,43 A. Berezhnoy,33 R. Bernet,42 A. Bertolin,23 C. Betancourt,42 F. Betti,15 M.-O. Bettler,40 M. van Beuzekom,43 Ia. Bezshyiko,42 S. Bifani,47 P. Billoir,8 T. Bird,56 A. Birnkraut,10 A. Bitadze,56 A. Bizzeti,18,u T. Blake,50 F. Blanc,41 J. Blouw,11,† S. Blusk,61 V. Bocci,26 T. Boettcher,58 A. Bondar,36,w N. Bondar,31,40 W. Bonivento,16 I. Bordyuzhin,32 A. Borgheresi,21,i S. Borghi,56 M. Borisyak,35 M. Borsato,39 F. Bossu,7 M. Boubdir,9 T. J. V. Bowcock,54 E. Bowen,42 C. Bozzi,17,40 S. Braun,12 M. Britsch,12 T. Britton,61 J. Brodzicka,56 E. Buchanan,48 C. Burr,56 A. Bursche,2 J. Buytaert,40 S. Cadeddu,16 R. Calabrese,17,g M. Calvi,21,i M. Calvo Gomez,38,m A. Camboni,38 P. Campana,19 D. H. Campora Perez,40 L. Capriotti,56 A. Carbone,15,e G. Carboni,25,j R. Cardinale,20,h A. Cardini,16 P. Carniti,21,i L. Carson,52 K. Carvalho Akiba,2 G. Casse,54 L. Cassina,21,i L. Castillo Garcia,41 M. Cattaneo,40 G. Cavallero,20 R. Cenci,24,t D. Chamont,7 M. Charles,8 Ph. Charpentier,40 G. Chatzikonstantinidis,47 M. Chefdeville,4 S. Chen,56 S. -F. Cheung,57 V. Chobanova,39 DOI: 10.1103/PhysRevLett.118.111803 McGregor, Boosted decision trees as an alternative to artificial neural networks for particle identification, Nucl. Instrum. Methods Phys. Res., Sect. A 543, 577 (2005). [12] A. A. Alves Jr. et al. (LHCb Collaboration), The LHCb detector at the LHC, J. Instrum. 3, S08005 (2008). [25] G. Punzi, in Statistical Problems in Particle Physics, Astrophysics, and Cosmology, edited by L. Lyons, R. Mount, and R. Reitmeyer (Stanford Linear Accelerator Center, Stanford, California, 2003), p. 79. [13] R. Aaij et al. (LHCb Collaboration), LHCb detector performance, Int. J. Mod. Phys. A 30, 1530022 (2015). [14] M. Adinolfi et al., Performance of the LHCb RICH detector at the LHC, Eur. Phys. J. C 73, 2431 (2013). [26] R. Aaij et al. (LHCb Collaboration), Measurement of CP observables in B →DK and B →Dπ with two- and four-body D meson decays, Phys. Lett. B 760, 117 (2016). [15] T. Sjöstrand, S. Mrenna, and P. Skands, PYTHIA 6.4 physics and manual, J. High Energy Phys. 05 (2006) 026. [16] T. Sjöstrand, S. Mrenna, and P. Skands, A brief introduction to PYTHIA 8.1, Comput. Phys. Commun. 178, 852 (2008). [27] L. Moneta et al., The RooStats Project, Proc. Sci., ACAT20102010 (2010) 057. [17] I. Belyaev et al., Handling of the generation of primary events in Gauss, the LHCb simulation framework, J. Phys. Conf. Ser. 331, 032047 (2011). [28] R. Aaij et al. (LHCb Collaboration), Measurement of the branching fraction ratio BðBþc →ψð2SÞπþÞ= BðBþc →J=ψπþÞ, Phys. Rev. D 92, 072007 (2015). R. Aaij,40 B. Adeva,39 M. Adinolfi,48 Z. Ajaltouni,5 S. Akar,59 J. Albrecht,10 F. Alessio,40 M. Alexander,53 S. Ali,43 G. Alkhazov,31 P. Alvarez Cartelle,55 A. A. Alves Jr.,59 S. Amato,2 S. Amerio,23 Y. Amhis,7 L. An,3 L. Anderlini,18 G. Andreassi,41 M. Andreotti,17,g J. E. Andrews,60 R. B. Appleby,56 F. Archilli,43 P. d’Argent,12 J. Arnau Romeu,6 A. Artamonov,37 M. Artuso,61 E. Aslanides,6 G. Auriemma,26 M. Baalouch,5 I. Babuschkin,56 S. Bachmann,12 J. J. Back,50 A. Badalov,38 C. Baesso,62 S. Baker,55 V. Balagura,7,c W. Baldini,17 R. J. Barlow,56 C. Barschel,40 S. Barsuk,7 W. Barter,56 F. Baryshnikov,32 M. Baszczyk,27 V. Batozskaya,29 B. Batsukh,61 V. Battista,41 A. Bay,41 L. Beaucourt,4 J. Beddow,53 F. Bedeschi,24 I. Bediaga,1 L. J. Bel,43 V. Bellee,41 N. Belloli,21,i K. Belous,37 I. Belyaev,32 E. Ben-Haim,8 G. Bencivenni,19 S. Benson,43 A. Berezhnoy,33 R. Bernet,42 A. Bertolin,23 C. Betancourt,42 F. Betti,15 M.-O. Bettler,40 M. van Beuzekom,43 Ia. Bezshyiko,42 S. Bifani,47 P. Billoir,8 T. Bird,56 A. Birnkraut,10 A. Bitadze,56 A. M. Chrzaszcz,42,27 X. Cid Vidal,39 G. Ciezarek,43 P. E. L. Clarke,52 M. Clemencic,40 H. V. Cliff,49 J. Closier,40 V. Coco,59 J. Cogan,6 E. Cogneras,5 V. Cogoni,16,40,f L. Cojocariu,30 G. Collazuol,23,o P. Collins,40 A. Comerma-Montells,12 A. Contu,40 A. Cook,48 G. Coombs,40 S. Coquereau,38 G. Corti,40 M. Corvo,17,g C. M. Costa Sobral,50 B. Couturier,40 G. A. Cowan,52 D. C. Craik,52 A. Crocombe,50 M. Cruz Torres,62 S. Cunliffe,55 R. Currie,55 C. D’Ambrosio,40 F. Da Cunha Marinho,2 E. Dall’Occo,43 J. Dalseno,48 P. N. Y. David,43 A. Davis,3 K. De Bruyn,6 S. De Capua,56 M. De Cian,12 J. M. De Miranda,1 L. De Paula,2 M. De Serio,14,d P. De Simone,19 C. T. Dean,53 D. Decamp,4 M. Deckenhoff,10 L. Del Buono,8 M. Demmer,10 A. Dendek,28 D. Derkach,35 O. Deschamps,5 F. Dettori,40 B. Dey,22 A. Di Canto,40 H. Dijkstra,40 F. Dordei,40 M. Dorigo,41 A. Dosil Suárez,39 A. Dovbnya,45 K. Dreimanis,54 L. Dufour,43 G. Dujany,56 K. Dungs,40 P. Durante,40 R. Dzhelyadin,37 A. Dziurda,40 A. Dzyuba,31 N. Déléage,4 S. Easo,51 M. Ebert,52 U. Egede,55 V. Egorychev,32 S. Eidelman,36,w S. Eisenhardt,52 U. Eitschberger,10 R. Ekelhof,10 L. Eklund,53 S. Ely,61 S. Esen,12 H. M. Evans,49 T. Evans,57 A. Falabella,15 N. Farley,47 S. Farry,54 R. Fay,54 D. Fazzini,21,i D. Ferguson,52 A. Fernandez Prieto,39 F. Ferrari,15,40 F. Ferreira Rodrigues,2 M. Ferro-Luzzi,40 S. Filippov,34 R. A. Fini,14 M. Fiore,17,g M. Fiorini,17,g M. Firlej,28 C. Fitzpatrick,41 T. Fiutowski,28 F. Fleuret,7,b K. Fohl,40 M. Fontana,16,40 F. Fontanelli,20,h D. C. Forshaw,61 R. Forty,40 V. Franco Lima,54 M. Frank,40 C. Frei,40 J. Fu,22,q W. Funk,40 E. Furfaro,25,j C. Färber,40 A. Gallas Torreira,39 D. Galli,15,e S. Gallorini,23 S. Gambetta,52 M. Gandelman,2 P. Gandini,57 Y. Gao,3 L. M. Garcia Martin,69 J. García Pardiñas,39 J. Garra Tico,49 L. Garrido,38 P. J. Garsed,49 D. Gascon,38 C. Gaspar,40 L. Gavardi,10 G. Gazzoni,5 D. Gerick,12 E. Gersabeck,12 M. Gersabeck,56 T. Gershon,50 Ph. Ghez,4 S. Gianì,41 V. Gibson,49 O. G. Girard,41 L. Giubega,30 K. Gizdov,52 V. V. Gligorov,8 D. Golubkov,32 A. Golutvin,55,40 A. Gomes,1,a I. V. Gorelov,33 C. Gotti,21,i R. Graciani Diaz,38 L. A. Granado Cardoso,40 E. Graugés,38 E. Graverini,42 G. Graziani,18 A. Grecu,30 P. Griffith,47 L. Grillo,21,40,i B. R. Gruberg Cazon,57 O. Grünberg,67 E. Gushchin,34 Yu. Guz,37 T. Gys,40 C. Göbel,62 T. Hadavizadeh,57 C. Hadjivasiliou,5 G. Haefeli,41 C. Haen,40 S. C. Haines,49 B. Hamilton,60 X. Han,12 S. Hansmann-Menzemer,12 N. Harnew,57 S. T. Harnew,48 J. Harrison,56 M. Hatch,40 J. He,63 T. Head,41 A. Heister,9 K. Hennessy,54 P. Henrard,5 L. Henry,8 E. van Herwijnen,40 M. Heß,67 A. Hicheur,2 D. Hill,57 C. Hombach,56 H. Hopchev,41 W. Hulsbergen,43 T. Humair,55 M. Hushchyn,35 D. Hutchcroft,54 M. Idzik,28 P. Ilten,58 R. Jacobsson,40 A. Jaeger,12 J. Jalocha,57 E. Jans,43 A. Jawahery,60 F. Jiang,3 M. John,57 D. Johnson,40 C. R. Jones,49 C. Joram,40 B. Jost,40 N. Jurik,57 S. Kandybei,45 M. Karacson,40 J. M. Kariuki,48 S. Karodia,53 M. Kecke,12 M. Kelsey,61 M. Kenzie,49 T. Ketel,44 E. Khairullin,35 B. Khanji,12 C. Khurewathanakul,41 T. Kirn,9 S. Klaver,56 K. Klimaszewski,29 S. Koliiev,46 M. Kolpin,12 I. Komarov,41 R. F. Koopman,44 P. Koppenburg,43 A. Kosmyntseva,32 A. Kozachuk,33 M. Kozeiha,5 L. Kravchuk,34 K. Kreplin,12 M. Kreps,50 P. Krokovny,36,w F. Kruse,10 W. Krzemien,29 W. Kucewicz,27,l M. Kucharczyk,27 V. Kudryavtsev,36,w A. K. Kuonen,41 K. Kurek,29 T. Kvaratskheliya,32,40 D. Lacarrere,40 G. Lafferty,56 A. Lai,16 G. Lanfranchi,19 C. Langenbruch,9 T. Latham,50 C. Lazzeroni,47 R. Le Gac,6 J. van Leerdam,43 A. Leflat,33,40 J. Lefrançois,7 R. Lefèvre,5 F. Lemaitre,40 E. Lemos Cid,39 O. Leroy,6 T. Lesiak,27 B. Leverington,12 T. Li,3 Y. Li,7 T. Likhomanenko,35,68 R. Lindner,40 C. Linn,40 F. Lionetto,42 X. Liu,3 D. Loh,50 I. Longstaff,53 J. H. Lopes,2 D. Lucchesi,23,o M. Lucio Martinez,39 H. Luo,52 A. Lupato,23 E. Luppi,17,g O. Lupton,40 A. Lusiani,24 X. Lyu,63 F. Machefert,7 F. Maciuc,30 O. Maev,31 K. Maguire,56 S. Malde,57 A. Malinin,68 T. Maltsev,36 G. Manca,16,f G. Mancinelli,6 P. Manning,61 J. Maratas,5,v J. F. Marchand,4 U. Marconi,15 C. Marin Benito,38 M. Marinangeli,41 P. Marino,24,t J. Marks,12 G. Martellotti,26 M. Martin,6 M. Martinelli,41 D. Martinez Santos,39 F. Martinez Vidal,69 D. Martins Tostes,2 L. M. Massacrier,7 A. Massafferri,1 R. Matev,40 A. Mathad,50 Z. Mathe,40 C. Matteuzzi,21 A. Mauri,42 E. Maurice,7,b B. Maurin,41 A. Mazurov,47 M. McCann,55,40 A. McNab,56 R. McNulty,13 B. Meadows,59 F. Meier,10 M. Meissner,12 D. Melnychuk,29 M. Merk,43 A. Merli,22,q E. Michielin,23 D. A. Milanes,66 M. -N. Minard,4 D. S. Mitzel,12 A. Mogini,8 J. Molina Rodriguez,1 I. A. Monroy,66 S. Monteil,5 M. Morandin,23 P. Morawski,28 A. Mordà,6 M. J. Morello,24,t O. Morgunova,68 J. Moron,28 A B M i 52 R M t i 61 F M h i 52 M M ld 43 M M i i 15 D Müll 56 J Müll 10 K Müll 42 V Müll 10 DOI: 10.1103/PhysRevLett.118.111803 Bizzeti,18,u T. Blake,50 F. Blanc,41 J. Blouw,11,† S. Blusk,61 V. Bocci,26 T. Boettcher,58 A. Bondar,36,w N. Bondar,31,40 W. Bonivento,16 I. Bordyuzhin,32 A. Borgheresi,21,i S. Borghi,56 M. Borisyak,35 M. Borsato,39 F. Bossu,7 M. Boubdir,9 T. J. V. Bowcock,54 E. Bowen,42 C. Bozzi,17,40 S. Braun,12 M. Britsch,12 T. Britton,61 J. Brodzicka,56 E. Buchanan,48 C. Burr,56 A. Bursche,2 J. Buytaert,40 S. Cadeddu,16 R. Calabrese,17,g M. Calvi,21,i M. Calvo Gomez,38,m A. Camboni,38 P. Campana,19 D. H. Campora Perez,40 L. Capriotti,56 A. Carbone,15,e G. Carboni,25,j R. Cardinale,20,h A. Cardini,16 P. Carniti,21,i L. Carson,52 K. Carvalho Akiba,2 G. Casse,54 L. Cassina,21,i L. Castillo Garcia,41 M. Cattaneo,40 G. Cavallero,20 R. Cenci,24,t D. Chamont,7 M. Charles,8 Ph. Charpentier,40 G. Chatzikonstantinidis,47 M. Chefdeville,4 S. Chen,56 S. -F. Cheung,57 V. Chobanova,39 111803-5 week ending 17 MARCH 2017 week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) J. Cogan,6 E. Cogneras,5 V. Cogoni,16,40,f L. Cojocariu,30 G. Collazuol,23,o P. Collins,40 A. Comer L. De Paula,2 M. De Serio,14,d P. De Simone,19 C. T. Dean,53 D. Decamp,4 M. Deckenhoff,10 L. A. Dendek,28 D. Derkach,35 O. Deschamps,5 F. Dettori,40 B. Dey,22 A. Di Canto,40 H. Dijkstra, g y N. Farley,47 S. Farry,54 R. Fay,54 D. Fazzini,21,i D. Ferguson,52 A. Fernandez Prieto,39 F. Ferrari,15,40 F. Ferreira Rodrigues,2 M. Ferro-Luzzi,40 S. Filippov,34 R. A. Fini,14 M. Fiore,17,g M. Fiorini,17,g M. Firlej,28 C. Fit pp j p F. Fleuret,7,b K. Fohl,40 M. Fontana,16,40 F. Fontanelli,20,h D. C. Forshaw,61 R. Forty,40 V. Franco Lima,54 M. Frank,40 C. Frei,40 J. Fu,22,q W. Funk,40 E. Furfaro,25,j C. Färber,40 A. Gallas Torreira,39 D. Galli,15,e S. Gallorini,23 S. Gambetta,52 M. Lucio Martinez,39 H. Luo,52 A. Lupato,23 E. Luppi,17,g O. Lupton,40 A. Lusiani,24 X. Lyu,63 F. Machefert,7 F. Maciuc,30 J. F. Marchand,4 U. Marconi,15 C. Marin Benito,38 M. Marinangeli,41 P. Marino,24,t J. Marks,12 G. Martellotti,26 M. Martin,6 M. Martinelli,41 D. Martinez Santos,39 F. Martinez Vidal,69 D. Martins Tostes,2 L. M. Massacrier,7 A. Massafferri,1 R. Matev,40 A. Mathad,50 Z. Mathe,40 C. Matteuzzi,21 A. Mauri,42 E. Maurice,7,b B. Maurin,41 A. Mazurov,47 M. McCann,55,40 A. McNab,56 R. McNulty,13 B. Meadows,59 F. Meier,10 M. Meissner,12 D. Melnychuk,29 M. Merk,43 A M li 22 q E Mi hi li 23 D A Mil 66 M N Mi d 4 D S Mi l 12 A M i i 8 J M li R d i 1 M. DOI: 10.1103/PhysRevLett.118.111803 Viaud,7 D. Vieira,63 M. Vieites Diaz,39 H. Viemann,67 X. Vilasis-Cardona,38,m M. Vitti,49 V. Volkov,33 A. Vollhardt,42 B. Voneki,40 A. Vorobyev,31 V. Vorobyev,36,w C. Voß,9 J. A. de Vries,43 C. Vázquez Sierra,39 R. Waldi,67 C. Wallace,50 R. Wallace,13 J. Walsh,24 J. Wang,61 D. R. Ward,49 H. M. Wark,54 N. K. Watson,47 D. Websdale,55 A. Weiden,42 M. Whitehead,40 J. Wicht,50 G. Wilkinson,57,40 M. Wilkinson,61 M. Williams,40 M. P. Williams,47 M. Williams,58 T. Williams,47 F. F. Wilson,51 J. Wimberley,60 J. Wishahi,10 W. Wislicki,29 M. Witek,27 G. Wormser,7 S. A. Wotton,49 K. Wraight,53 K. Wyllie,40 Y. Xie,65 Z. Xing,61 Z. Xu,4 Z. Yang,3 Y. Yao,61 H. Yin,65 J. Yu,65 X. Yuan,36,w O. Yushchenko,37 K. A. Zarebski,47 M. Zavertyaev,11,c L. Zhang,3 Y. Zhang,7 Y. Zhang,63 A. Zhelezov,12 Y. Zheng,63 X. Zhu,3 V. Zhukov,33 and S. Zucchelli15 (LHCb Collaboration) 1Centro Brasileiro de Pesquisas Físicas (CBPF), Rio de Janeiro, Brazil 2Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, Brazil 3Center for High Energy Physics, Tsinghua University, Beijing, China 4LAPP, Université Savoie Mont-Blanc, CNRS/IN2P3, Annecy-Le-Vieux, France 5Clermont Université, Université Blaise Pascal, CNRS/IN2P3, LPC, Clermont-Ferrand, France 6CPPM, Aix-Marseille Université, CNRS/IN2P3, Marseille, France 7LAL, Université Paris-Sud, CNRS/IN2P3, Orsay, France 8LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, France 9I. Physikalisches Institut, RWTH Aachen University, Aachen, Germany 10Fakultät Physik, Technische Universität Dortmund, Dortmund, Germany 11Max-Planck-Institut für Kernphysik (MPIK), Heidelberg, Germany 12Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 13School of Physics, University College Dublin, Dublin, Ireland 14Sezione INFN di Bari, Bari, Italy 15Sezione INFN di Bologna, Bologna, Italy 16Sezione INFN di Cagliari, Cagliari, Italy DOI: 10.1103/PhysRevLett.118.111803 McCann,55,40 A. McNab,56 R. McNulty,13 B. Meadows,59 F. Meier,10 M. Meissner,12 D. Melnychuk,29 M. Merk,43 A. Merli,22,q E. Michielin,23 D. A. Milanes,66 M. -N. Minard,4 D. S. Mitzel,12 A. Mogini,8 J. Molina Rodriguez,1 I. A. Monroy,66 S. Monteil,5 M. Morandin,23 P. Morawski,28 A. Mordà,6 M. J. Morello,24,t O. Morgunova,68 J. Moron,28 , , , , , , , , , P. Naik,48 T. Nakada,41 R. Nandakumar,51 A. Nandi,57 I. Nasteva,2 M. Needham,52 N. Neri,22 S. Neubert,12 N. Neufeld,40 M. Neuner,12 T. D. Nguyen,41 C. Nguyen-Mau,41,n S. Nieswand,9 R. Niet,10 N. Nikitin,33 T. Nikodem,12 A. Nogay,68 A. Novoselov,37 D. P. O’Hanlon,50 A. Oblakowska-Mucha,28 V. Obraztsov,37 S. Ogilvy,19 R. Oldeman,16,f C. J. G. Onderwater,70 J. M. Otalora Goicochea,2 A. Otto,40 P. Owen,42 A. Oyanguren,69 P. R. Pais,41 A. Palano,14,d A. Novoselov, D. P. O Hanlon, A. Oblakowska Mucha, V. Obraztsov, S. Ogilvy, R. Oldeman, C. J. G. Onderwater,70 J. M. Otalora Goicochea,2 A. Otto,40 P. Owen,42 A. Oyanguren,69 P. R. Pais,41 A. Palano,14,d M. Palutan,19 A. Papanestis,51 M. Pappagallo,14,d L. L. Pappalardo,17,g W. Parker,60 C. Parkes,56 G. Passaleva,18 g g p S. Perazzini,40 P. Perret,5 L. Pescatore,47 K. Petridis,48 A. Petrolini,20,h A. Petrov,68 M. Petruzzo,22,q E. Picatoste Olloqui,38 111803-6 week ending 17 MARCH 2017 week ending 17 MARCH 2017 P H Y S I C A L R E V I E W L E T T E R S PRL 118, 111803 (2017) S. Poslavskii,37 C. Potterat,2 E. Price,48 J. D. Price,54 J. Prisciandaro,39,40 A. Pritchard,54 C. Prouve,48 V. Pugatch,46 A. Puig Navarro,42 G. Punzi,24,p W. Qian,50 R. Quagliani,7,48 B. Rachwal,27 J. H. Rademacker,48 M. Rama,24 C. Sanchez Mayordomo,69 B. Sanmartin Sedes,39 R. Santacesaria,26 C. Santamarina Rios,39 M. Santimaria,19 E. Santovetti,25,j A. Sarti,19,k C. Satriano,26,s A. Satta,25 D. M. Saunders,48 D. Savrina,32,33 S. Schael,9 M. Schellenberg,10 C. Sanchez Mayordomo,69 B. Sanmartin Sedes,39 R. Santacesaria,26 C. Santamarina Rios,39 M. Santimaria,19 E. Santovetti,25,j A. Sarti,19,k C. Satriano,26,s A. Satta,25 D. M. Saunders,48 D. Savrina,32,33 S. Schael,9 M. Schellenberg,10 M. Schiller,53 H. Schindler,40 M. Schlupp,10 M. Schmelling,11 T. Schmelzer,10 B. Schmidt,40 O. Schneider,41 A. Schopper,40 K. Schubert,10 M. Schubiger,41 M. -H. Schune,7 R. Schwemmer,40 B. Sciascia,19 A. Sciubba,26,k A. Semennikov,32 6 5 5 U. Uwer,12 C. Vacca,16,f V. Vagnoni,15,40 A. Valassi,40 S. Valat,40 G. Valenti,15 R. Vazquez Gomez,19 P. Vazquez Regueiro,39 S. Vecchi,17 M. van Veghel,43 J. J. Velthuis,48 M. Veltri,18,r G. Veneziano,57 A. Venkateswaran,61 M. Vernet,5 M. Vesterinen,12 J. V. Viana Barbosa,40 B. 1Centro Brasileiro de Pesquisas Físicas (CBPF), Rio de Janeiro, Brazil 2Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, Brazil 3Center for High Energy Physics, Tsinghua University, Beijing, China 4LAPP, Université Savoie Mont-Blanc, CNRS/IN2P3, Annecy-Le-Vieux, France 5Clermont Université, Université Blaise Pascal, CNRS/IN2P3, LPC, Clermont-Ferrand, France 6CPPM, Aix-Marseille Université, CNRS/IN2P3, Marseille, France 7LAL, Université Paris-Sud, CNRS/IN2P3, Orsay, France 8LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, France 9I. Physikalisches Institut, RWTH Aachen University, Aachen, Germany 10Fakultät Physik, Technische Universität Dortmund, Dortmund, Germany 11Max-Planck-Institut für Kernphysik (MPIK), Heidelberg, Germany 12Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 13School of Physics, University College Dublin, Dublin, Ireland 14Sezione INFN di Bari, Bari, Italy 15Sezione INFN di Bologna, Bologna, Italy 16Sezione INFN di Cagliari, Cagliari, Italy F. Polci,8 A. Poluektov,50,36 I. Polyakov,61 E. Polycarpo,2 G. J. Pomery,48 A. Popov,37 D. Popov,11,40 B. Popovici,30 S. Poslavskii,37 C. Potterat,2 E. Price,48 J. D. Price,54 J. Prisciandaro,39,40 A. Pritchard,54 C. Prouve,48 V. Pugatch,46 A. Puig Navarro,42 G. Punzi,24,p W. Qian,50 R. Quagliani,7,48 B. Rachwal,27 J. H. Rademacker,48 M. Rama,24 M. Ramos Pernas,39 M. S. Rangel,2 I. Raniuk,45 F. Ratnikov,35 G. Raven,44 F. Redi,55 S. Reichert,10 A. C. dos Reis,1 C. Remon Alepuz,69 V. Renaudin,7 S. Ricciardi,51 S. Richards,48 M. Rihl,40 K. Rinnert,54 V. Rives Molina,38 P. Robbe,7,40 A. B. Rodrigues,1 E. Rodrigues,59 J. A. Rodriguez Lopez,66 P. Rodriguez Perez,56,† A. Rogozhnikov,35 S. Roiser,40 A. Rollings,57 V. Romanovskiy,37 A. Romero Vidal,39 J. W. Ronayne,13 M. Rotondo,19 M. S. Rudolph,61 T. Ruf,40 P. Ruiz Valls,69 J. J. Saborido Silva,39 E. Sadykhov,32 N. Sagidova,31 B. Saitta,16,f V. Salustino Guimaraes,1 C. Sanchez Mayordomo,69 B. Sanmartin Sedes,39 R. Santacesaria,26 C. Santamarina Rios,39 M. Santimaria,19 E. Santovetti,25,j A. Sarti,19,k C. Satriano,26,s A. Satta,25 D. M. Saunders,48 D. Savrina,32,33 S. Schael,9 M. Schellenberg,10 M. Schiller,53 H. Schindler,40 M. Schlupp,10 M. Schmelling,11 T. Schmelzer,10 B. Schmidt,40 O. Schneider,41 A. Schopper,40 K. Schubert,10 M. Schubiger,41 M. -H. Schune,7 R. Schwemmer,40 B. Sciascia,19 A. Sciubba,26,k A. Semennikov,32 A. Sergi,47 N. Serra,42 J. Serrano,6 L. Sestini,23 P. Seyfert,21 M. Shapkin,37 I. Shapoval,45 Y. Shcheglov,31 T. Shears,54 L. Shekhtman,36,w V. Shevchenko,68 B. G. Siddi,17,40 R. Silva Coutinho,42 L. Silva de Oliveira,2 G. Simi,23,o S. Simone,14,d M. Sirendi,49 N. Skidmore,48 T. Skwarnicki,61 E. Smith,55 I. T. Smith,52 J. Smith,49 M. Smith,55 H. Snoek,43 l. Soares Lavra,1 M. D. Sokoloff,59 F. J. P. Soler,53 B. Souza De Paula,2 B. Spaan,10 P. Spradlin,53 S. Sridharan,40 F. Stagni,40 M. Stahl,12 S. Stahl,40 P. Stefko,41 S. Stefkova,55 O. Steinkamp,42 S. Stemmle,12 O. Stenyakin,37 H. Stevens,10 S. Stevenson,57 S. Stoica,30 S. Stone,61 B. Storaci,42 S. Stracka,24,p M. Straticiuc,30 U. Straumann,42 L. Sun,64 W. Sutcliffe,55 K. Swientek,28 V. Syropoulos,44 M. Szczekowski,29 T. Szumlak,28 S. T’Jampens,4 A. Tayduganov,6 T. Tekampe,10 G. Tellarini,17,g F. Teubert,40 E. Thomas,40 J. van Tilburg,43 M. J. Tilley,55 V. Tisserand,4 M. Tobin,41 S. Tolk,49 L. Tomassetti,17,g D. Tonelli,40 S. Topp-Joergensen,57 F. Toriello,61 E. Tournefier,4 S. Tourneur,41 K. Trabelsi,41 M. Traill,53 M. T. Tran,41 M. Tresch,42 A. Trisovic,40 A. Tsaregorodtsev,6 P. Tsopelas,43 A. Tully,49 N. Tuning,43 A. Ukleja,29 A. Ustyuzhanin,35 U. Uwer,12 C. Vacca,16,f V. Vagnoni,15,40 A. Valassi,40 S. Valat,40 G. Valenti,15 R. Vazquez Gomez,19 P. Vazquez Regueiro,39 S. Vecchi,17 M. van Veghel,43 J. J. Velthuis,48 M. Veltri,18,r G. Veneziano,57 A. Venkateswaran,61 M. Vernet,5 M. Vesterinen,12 J. V. Viana Barbosa,40 B. Viaud,7 D. Vieira,63 M. Vieites Diaz,39 H. Viemann,67 X. Vilasis-Cardona,38,m M. Vitti,49 V. Volkov,33 A. Vollhardt,42 B. Voneki,40 A. Vorobyev,31 V. Vorobyev,36,w C. Voß,9 J. A. de Vries,43 C. Vázquez Sierra,39 R. Waldi,67 C. Wallace,50 R. Wallace,13 J. Walsh,24 J. Wang,61 D. R. Ward,49 H. M. Wark,54 N. K. Watson,47 D. Websdale,55 A. Weiden,42 M. Whitehead,40 J. Wicht,50 G. Wilkinson,57,40 M. Wilkinson,61 M. Williams,40 M. P. Williams,47 M. Williams,58 T. Williams,47 F. F. Wilson,51 J. Wimberley,60 J. Wishahi,10 W. Wislicki,29 M. Witek,27 G. Wormser,7 S. A. Wotton,49 K. Wraight,53 K. Wyllie,40 Y. Xie,65 Z. Xing,61 Z. Xu,4 Z. Yang,3 Y. Yao,61 H. Yin,65 J. Yu,65 X. Yuan,36,w O. Yushchenko,37 K. A. Zarebski,47 M. Zavertyaev,11,c L. Zhang,3 Y. Zhang,7 Y. Zhang,63 A. Zhelezov,12 Y. Zheng,63 X. Zhu,3 V. Zhukov,33 and S. Zucchelli15 P H Y S I C A L R E V I E W L E T T E R S Wills Physics Laboratory, University of Bristol, Bristol, United Kingdom 49Cavendish Laboratory, University of Cambridge, Cambridge, United Kingdom 50Department of Physics, University of Warwick, Coventry, United Kingdom 51STFC Rutherford Appleton Laboratory, Didcot, United Kingdom 52School of Physics and Astronomy, University of Edinburgh, Edinburgh, United Kingdom 53School of Physics and Astronomy, University of Glasgow, Glasgow, United Kingdom 54Oliver Lodge Laboratory, University of Liverpool, Liverpool, United Kingdom 55Imperial College London, London, United Kingdom 56School of Physics and Astronomy, University of Manchester, Manchester, United Kingdom 57Department of Physics, University of Oxford, Oxford, United Kingdom 58Massachusetts Institute of Technology, Cambridge, Massachusetts, USA 59University of Cincinnati, Cincinnati, Ohio, USA 60University of Maryland, College Park, Maryland, USA 61Syracuse University, Syracuse, New York, USA 62Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil, associated with Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, Brazil 63University of Chinese Academy of Sciences, Beijing, China, associated with Center for High Energy Physics, Tsinghua University, Beijing, China 64School of Physics and Technology, Wuhan University, Wuhan, China, associated with Center for High Energy Physics, Tsinghua University, Beijing, China 65Institute of Particle Physics, Central China Normal University, Wuhan, Hubei, China, associated with Center for High Energy Physics, Tsinghua University, Beijing, China 66Departamento de Fisica, Universidad Nacional de Colombia, Bogota, Colombia, associated with LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, France 67Institut für Physik, Universität Rostock, Rostock, Germany, associated with Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 68National Research Centre Kurchatov Institute, Moscow, Russia, associated with Institute of Theoretical and Experimental Physics (ITEP), Moscow, Russia Imperial College London, London, United Kingdom 58Massachusetts Institute of Technology, Cambridge, Massachusetts, USA 59 59University of Cincinnati, Cincinnati, Ohio, USA 60 61Syracuse University, Syracuse, New York, USA 62Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil, 63University of Chinese Academy of Sciences, Beijing, China, f g gy y g y j g 64School of Physics and Technology, Wuhan University, Wuhan, China, 66Departamento de Fisica, Universidad Nacional de Colombia, Bogota, Colombia, associated with LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, Fran 67 associated with LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, Fra 67Institut für Physik Universität Rostock Rostock Germany with Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 68 68National Research Centre Kurchatov Institute, Moscow, Russia, associated with Institute of Theoretical and Experimental Physics (ITEP), Moscow, Russia 111803-8 111803-7 111803-7 week ending 17 MARCH 2017 PRL 118, 111803 (2017) P H Y S I C A L R E V I E W L E T T E R S 17Sezione INFN di Ferrara, Ferrara, Italy 18Sezione INFN di Firenze, Firenze, Italy 19Laboratori Nazionali dell’INFN di Frascati, Frascati, Italy 20Sezione INFN di Genova, Genova, Italy 21Sezione INFN di Milano Bicocca, Milano, Italy 22Sezione INFN di Milano, Milano, Italy 23Sezione INFN di Padova, Padova, Italy 24Sezione INFN di Pisa, Pisa, Italy 25Sezione INFN di Roma Tor Vergata, Roma, Italy 26Sezione INFN di Roma La Sapienza, Roma, Italy 27Henryk Niewodniczanski Institute of Nuclear Physics Polish Academy of Sciences, Kraków, Poland 28AGH - University of Science and Technology, Faculty of Physics and Applied Computer Science, Kraków, Poland 29National Center for Nuclear Research (NCBJ), Warsaw, Poland 30Horia Hulubei National Institute of Physics and Nuclear Engineering, Bucharest-Magurele, Romania 31Petersburg Nuclear Physics Institute (PNPI), Gatchina, Russia 32Institute of Theoretical and Experimental Physics (ITEP), Moscow, Russia 33Institute of Nuclear Physics, Moscow State University (SINP MSU), Moscow, Russia 34Institute for Nuclear Research of the Russian Academy of Sciences (INR RAN), Moscow, Russia 35Yandex School of Data Analysis, Moscow, Russia 36Budker Institute of Nuclear Physics (SB RAS), Novosibirsk, Russia 37Institute for High Energy Physics (IHEP), Protvino, Russia 38ICCUB, Universitat de Barcelona, Barcelona, Spain 39Universidad de Santiago de Compostela, Santiago de Compostela, Spain 40European Organization for Nuclear Research (CERN), Geneva, Switzerland 41Institute of Physics, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland 42Physik-Institut, Universität Zürich, Zürich, Switzerland 43Nikhef National Institute for Subatomic Physics, Amsterdam, Netherlands 44Nikhef National Institute for Subatomic Physics and VU University Amsterdam, Amsterdam, Netherlands 45NSC Kharkiv Institute of Physics and Technology (NSC KIPT), Kharkiv, Ukraine 46Institute for Nuclear Research of the National Academy of Sciences (KINR), Kyiv, Ukraine 47University of Birmingham, Birmingham, United Kingdom 48H.H. 111803-8 week ending 17 MARCH 2017 PRL 118, 111803 (2017) PRL 118, 111803 (2017) 69Instituto de Fisica Corpuscular, Centro Mixto Universidad de Valencia - CSIC, Valencia, Spain, associated with ICCUB, Universitat de Barcelona, Barcelona, Spain 70Van Swinderen Institute, University of Groningen, Groningen, Netherlands, associated with Nikhef National Institute for Subatomic Physics, Amsterdam, Netherlands †Deceased. aUniversidade Federal do Triângulo Mineiro (UFTM), Uberaba-MG, Brazil. b bLaboratoire Leprince-Ringuet, Palaiseau, France. cP.N. Lebedev Physical Institute, Russian Academy of Science (LPI RAS), Moscow, Russia d cP.N. Lebedev Physical Institute, Russian Academy of Science (LPI RAS) d dUniversità di Bari, Bari, Italy. eUniversità di Bologna, Bologna, Italy. f fUniversità di Cagliari, Cagliari, Italy. fUniversità di Cagliari, Cagliari, Italy. gUniversità di Ferrara, Ferrara, Italy. h gUniversità di Ferrara, Ferrara, Italy. h hUniversità di Genova, Genova, Italy. I IUniversità di Milano Bicocca, Milano, Italy. j IUniversità di Milano Bicocca, Milano, Italy. j jUniversità di Roma Tor Vergata, Roma, Italy. k jUniversità di Roma Tor Vergata, Roma, Italy. k kUniversità di Roma La Sapienza, Roma, Italy. l Università di Roma La Sapienza, Roma, Italy. lAGH - University of Science and Technology, Faculty of Computer Science, Electronics and Telecommunications, Kraków, Poland. mLIFAELS, La Salle, Universitat Ramon Llull, Barcelona, Spain. p , , y lAGH - University of Science and Technology, Faculty of Computer Science, Electronics and Telecommunications, Kraków, Poland. mLIFAELS L S ll U i it t R Ll ll B l S i lAGH - University of Science and Technology, Faculty of Computer Science, Electronics and Telecommunications, Kraków, Poland. mLIFAELS L S ll U i i R Ll ll B l S i lAGH - University of Science and Technology, Faculty of Computer Science, Electronics a mLIFAELS, La Salle, Universitat Ramon Llull, Barcelona, Spain. LIFAELS, La Salle, Universitat Ramon Llull, Barc nHanoi University of Science, Hanoi, Viet Nam. nHanoi University of Science, Hanoi, Viet Nam. oUniversità di Padova, Padova, Italy. pUniversità di Pisa, Pisa, Italy. pUniversità di Pisa, Pisa, Italy. qUniversità degli Studi di Milano, Milano, Italy. r qUniversità degli Studi di Milano, Milano, Italy. rUniversità di Urbino, Urbino, Italy. rUniversità di Urbino, Urbino, Italy. sUniversità della Basilicata, Potenza, Italy. sUniversità della Basilicata, Potenza, Italy. tScuola Normale Superiore, Pisa, Italy. tScuola Normale Superiore, Pisa, Italy. uUniversità di Modena e Reggio Emilia, Modena, Italy. uUniversità di Modena e Reggio Emilia, Modena, Italy. vIligan Institute of Technology (IIT), Iligan, Philippines. vIligan Institute of Technology (IIT), Iligan, Philippines. wNovosibirsk State University, Novosibirsk, Russia. wNovosibirsk State University, Novosibirsk, Russia. 111803-9
https://openalex.org/W2757453573	https://europepmc.org/articles/pmc5853288?pdf=render	English	null	The Crux of Ebola Diagnostics	The journal of infectious diseases (Online. University of Chicago Press)/The Journal of infectious diseases	2,017	public-domain	2,275	Keywords. Ebola; diagnostics; infectivity assay; outbreak. (immunoglobulin M, immunoglobulin G, and rarely neutralizing antibodies); and (4) molecular detection for viral genome sequences [2, 4, 6] (Table 1). Virus iso- lation and subsequent characterization (ie, by immunofluorescence or electron microscopy) has historically been the gold standard of testing, but is rapidly being challenged by PCR assays, other nucleic amplification strategies (eg, loop-mediated isothermal amplification), and next-gen- eration sequencing. Today these molecu- lar detection methods may be considered the new gold standard, especially after the West African Ebola outbreak, because of their wider application, higher sensitivity and specificity, rapid performance, and ability for field deployment [6, 7]. Although there are no approved thera- pies for Ebola virus disease (EVD), early and reliable diagnosis is paramount to managing cases so that isolation, contact tracing, and infection control procedures can be implemented in a timely fashion. Because of the nonpathognomonic clin- ical presentation of EVD, any diagnostic uncertainty or delay may expose individ- uals who are not infected, with devastat- ing consequences [1]. In addition, delays in initial recognition of the outbreak may delay appropriate local, national, and international responses. patients should be closely monitored, and isolated and tested if they were to become symptomatic. These risk stratification measures allow for more rapid case iden- tification, help prioritize diagnostic test- ing, and lessen further transmission. We recognize that risk stratifications of EVD patient management may also be applied differently in distinct countries and under certain circumstances according to spe- cific national contingency plans. Although the techniques utilized for laboratory testing are not vastly different for other pathogens, diagnostic testing for ebolaviruses and other such high-con- sequence pathogens has been limited to laboratories with highly trained staff and enhanced biosafety and biosecur- ity requirements [1, 2, 4]. These issues are still pertinent during outbreaks, as typically initial diagnosis is performed in international high-containment refer- ence laboratories, as exemplified by the recent West African outbreak [5]. Once the outbreak has been identified and established, diagnostics can be handled by mobile laboratories deployed into remote regions, such as in sub-Saharan Africa, with limited but rapid and tar- geted diagnostic approaches such as real- time reverse-transcription polymerase chain reaction (PCR) or other, largely molecular detection methods [6, 7]. Obviously, nonlaboratory parameters such as epidemiological risk as well as the presence or absence of EVD symp- toms are key factors in the evaluation and management of suspect cases [1–3]. The Crux of Ebola Diagnostics James E. Strong1,2,3 and Heinz Feldmann2,4 1Special Pathogens, National Microbiology Laboratory, Public Health Agency of Canada, and Departments of 2Medical Microbiology and Infectious Diseases and 3Paediatrics and Child Health, College of Medicine, Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada; and 4Laboratory of Virology, Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana The Journal of Infectious Diseases The Journal of Infectious Diseases The Journal of Infectious Diseases® 2017;216:1340–2 Published by Oxford University Press for the Infectious Diseases Society of America 2017. This work is written by (a) US Government employee(s) and is in the public domain in the US. DOI: 10.1093/infdis/jix490 Received 11 September 2017; editorial decision 11 September 2017; accepted 20 September 2017; published online September 23, 2017. Correspondence: H. Feldmann, MD, National Institute of Allergy and Infectious Diseases, 903 S Fourth St, Hamilton, MT 59840 (feldmannh@niaid.nih.gov). (See the brief report by Kainulainen et al, on pages 1380–5.) Keywords. Ebola; diagnostics; infectivity assay; outbreak. Received 11 September 2017; editorial decision 11 September 2017; accepted 20 September 2017; published online September 23, 2017. Correspondence: H. Feldmann, MD, National Institute of Allergy and Infectious Diseases, 903 S Fourth St, Hamilton, MT 59840 (feldmannh@niaid.nih.gov). Keywords. Ebola; diagnostics; infectivity assay; outbreak. This is par- ticularly pertinent for patient discharge decisions, in convalescing patients or with nontraditional specimens (including semen, urine, breast milk), as virus com- ponents, such as viral proteins or nucleic acid, may be detectable by sensitive meth- ods and yet the patients may not be infec- tious. For example, immune-complexed virus from convalescing patients may not be viable or infectious. Semen, urine, and breast milk may also contain immuno- logic or other, yet unidentified factors that lower or abrogate infectivity. Currently, only culture-based techniques will defin- itively demonstrate virus viability and help to define the transmission risk, an important public health control parame- ter. This is particularly important for con- valescent patients who have shown to be persistently infected with ebolavirus for several months with potential for sexual transmission [13, 14]. assay developed by Kainulainen et al [8] addresses some of those drawbacks, but for biosafety reasons cannot complement the molecular assays in field conditions. One could perhaps foresee that in-coun- try or regional reference laboratories may one day allow for limited virus replication under safe conditions and, thus, those laboratories may be able to safely accom- modate this or other new assays in the future to address issues of infectivity for certain outbreak response activities. and titer rises. Reliable diagnosis is likely more important than rapid diagnosis because of the tremendous implications on public health measures. A good com- promise is often necessary. A major drawback to the current main- stream approaches to ebolavirus diagnos- tics (PCR, serology, antigen detection) is that they fail to determine virus infectiv- ity. In this issue of The Journal of Infectious Diseases, Kainulainen and colleagues [8] have partially addressed this dilemma with a novel reporter cell line capable of detecting live ebolavirus replication. A minigenome expression cassette was intro- duced into Vero E6 cells, the standard cell line for ebolavirus isolation. Upon infec- tion, the ebolavirus replication complex recognizes the promoter on the minige- nome leading to expression of a reporter gene, here zsGreen (ZSG, Clontech), that can be easily detected by its fluorescence. Interestingly, despite being based on Ebola virus Makona, the causative agent of the West African Ebola outbreak, the minige- nome promoter is detected by members of all other known ebolavirus species (Zaire, Sudan, Bundibugyo, Reston, and Tai Forest ebolaviruses), but not by related members of the genus Marburgvirus. Keywords. Ebola; diagnostics; infectivity assay; outbreak. To this end, individuals who have had possi- ble exposure to ebolaviruses but manifest no symptoms may not warrant testing, as none of the currently available diagnostic tests have demonstrated the sensitivity to reliably detect ebolaviruses prior to the onset of symptoms. Furthermore, such asymptomatic patients may not need to be strictly isolated as they pose a minimal risk of transmission. However, such exposed i Equally important is a standardized laboratory algorithm that should be applied [4, 7]. The algorithm likely dif- fers in a high-containment reference laboratory from that employed in a field laboratory or any interim diagnostic arrangement [4, 6, 7]. Testing individual specimens with a single approach and the right algorithm is probably accepta- ble in an outbreak setting with an estab- lished causative agent, even though not optimal, but certainly not adequate for initial diagnosis, which should include confirmatory testing by an independent approach. General diagnostic princi- ples should be considered such as mul- tiple target molecular detection to avoid false-negatives due to genetic changes or follow-up sampling for serological assays to demonstrate antibody class switch Currently, ebolavirus-specific testing can be achieved in several ways: (1) virus isolation and characterization; (2) anti- gen detection for viral proteins; (3) sero- logic testing for virus-specific antibodies 1340 • JID 2017:216 (1 December) • EDITORIAL COMMENTARY Table 1. Established Diagnostic Approaches pp Approach Platform Advantages Disadvantages Comments Antigen detection Antigen capture ELISA Rapid; fieldable; potential for bedside testing Lower sensitivity Confirmatory test Serology Direct (IgG) or antibody-cap- ture (IgM) ELISA Convalescence; helpful in retrospec- tive and ecologic analyses Lower sensitivity; limitations in specificity; delayed response Confirmatory test; serosurveys Nucleic acid detection Real time RT-PCR, loop-mediated isothermal amplification Rapid; sensitive; fieldable Potential for contamination; poten- tial for false-negatives due to genome mutations New gold standard Isolation Cell culture, animal Virus isolate for characterization Slow; containment required Old gold standard Particle structure Electron microscopy Structure Slow; expensive; potential for false-positives Rarely used Abbreviations: ELISA, enzyme-linked immunosorbent assay; IgG, immunoglobulin G; IgM, immunoglobulin M. key component of outbreak control [6, 7, 9, 10]. Bedside testing based on antigen detection is being developed for faster diagnosis and higher safety, as these tests would eliminate or at least limit specimen transport [11, 12]. These efforts address the need for immediate outbreak response and patient triage, but do not answer the question of virus infectivity. Keywords. Ebola; diagnostics; infectivity assay; outbreak. Even though this assay has not eliminated the necessity of high biocontainment, it is simpler, faster, and more easily scalable than usual isola- tion in tissue culture. Therefore, this is a very welcome improvement for diagnostics, with important public health implications.h In general, initial diagnosis of ebola- virus infections is still made in reference laboratories largely located in developed countries [4]. This was not any different with the recent West African epidemic [5]. Normally, the algorithm includes confirmation by independent assays based on distinct principles, one of which being virus infectivity testing [4, 7]. Therefore, the new assay by Kainulainen et al [8] would be helpful in replacing or complementing virus isolation. It likely could also be adapted to serve as a new and faster platform to test for ebolavirus neutralization and serve for better testing of non-blood samples. A key problem remains early detection, which starts with clinical awareness, rec- ognition of suspected cases, and the sub- sequent confirmation through laboratory diagnosis—a sequence of events that has caused the largest delays in the past. Once an outbreak is established, early mobilization of on-site laboratory response is paramount but has been slow, even though effective. Therefore, more effective strategies are Unfortunately, current assays to determine infectivity prevent rapid diagnosis and screening of larger sam- ple numbers, as they require high con- tainment conditions and will not lend themselves to outbreak situations such as those in sub-Saharan Africa. The novel The recent West African Ebola outbreak has established molecular diagnostics as a EDITORIAL COMMENTARY • JID 2017:216 (1 December) • 1341 Available at: https://www.cdc.gov/ vhf/ebola/index.html. Accessed 8 September 2017. needed for clinical education, assay devel- opment, and effective deployment [15]. Despite having powerful diagnostics and algorithms in place, the high-containment diagnostic laboratory world is still in search for the ideal ebolavirus test: one that in the field can safely, rapidly, and inexpensively detect minute quantities of only viable virus at the bedside. 9. Cherpillod P, Schibler M, Vieille G, et al. Ebola virus disease diagnosis by real-time RT-PCR: a comparative study of 11 different procedures. J Clin Virol 2016; 77:9–14. 3. Baseler L, Chertow DS, Johnson KM, Feldmann H, Morens DM. The pathogenesis of Ebola virus disease. Annu Rev Pathol 2017; 12:387–418. 10. Gire SK, Goba A, Andersen KG, et al. Genomic surveillance elucidates Ebola virus origin and transmission during the 2014 outbreak. Science 2014; 345:1369–72. 4. Keywords. Ebola; diagnostics; infectivity assay; outbreak. Strong JE, Grolla A, Jahrling PB, Feldmann H. Filoviruses and arena- viruses. In: Detrick B, Hamilton RG, Folds JD, ed. Manual of molecular and clinical laboratory immunology, 7th ed. Washington, DC: ASM Press, 2006:774–90. Potential conflicts of interest. All authors: No reported conflicts of interest. Potential conflicts of interest. All authors: No reported conflicts of interest. Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors con- sider relevant to the content of the manu- script have been disclosed. 13. Jacobs M, Rodger A, Bell DJ, et al. Late Ebola virus relapse causing meningoencephalitis: a case report. Lancet 2016; 388:498–503. 7. de Wit E, Rosenke K, Fischer RJ, et al. Ebola laboratory response at the eternal love winning Africa campus, Monrovia, Liberia, 2014– 2015. J Infect Dis 2016; 214(Suppl 3):S169–76. 14. Mate SE, Kugelman JR, Nyenswah TG, et al. Molecular evidence of sex- ual transmission of Ebola virus. N Engl J Med 2015; 373:2448–54. 1342 • JID 2017:216 (1 December) • EDITORIAL COMMENTARY Notes 11. Walker NF, Brown CS, Youkee D, et al. Evaluation of a point-of-care blood test for identification of Ebola virus disease at Ebola holding units, Western Area, Sierra Leone, January to February 2015. Euro Surveill 2015; 20:21073. Financial support. Work on filo- viruses for J. E. S is supported by the Canadian Institutes of Health Research (CIHR) and the Public Health Agency of Canada. Work on filoviruses for H. F. is supported by the Intramural Research Program, National Institute of Allergy and Infectious Diseases, National Institutes of Health. 5. Baize S, Pannetier D, Oestereich L, et al. Emergence of Zaire Ebola virus disease in Guinea. N Engl J Med 2014; 371:1418–25. 12. Coarsey CT, Esiobu N, Narayanan R, Pavlovic M, Shafiee H, Asghar W. Strategies in Ebola virus dis- ease (EVD) diagnostics at the point of care. Crit Rev Microbiol 2017; 43:779–98. 6. Shorten RJ, Brown CS, Jacobs M, Rattenbury S, Simpson AJ, Mepham S. Diagnostics in Ebola virus disease in resource-rich and resource-limited settings. PLoS Negl Trop Dis 2016; 10:e0004948. References 1. World Health Organization. Manual for the care and management of patients in Ebola care units/community care centers; interim emergency guidance. Geneva, Switzerland: WHO, 2015. 8. Kainulainen MH, Nichol ST, Albariño CG, Spiropoulou CF. Rapid determination of ebolavirus infectiv- ity in clinical samples using a novel reporter cell line. J Infect Dis 2017; 216:1380–5. 8. Kainulainen MH, Nichol ST, Albariño CG, Spiropoulou CF. Rapid determination of ebolavirus infectiv- ity in clinical samples using a novel reporter cell line. J Infect Dis 2017; 216:1380–5. 15. Perkins MD, Dye C, Balasegaram M, et al. Diagnostic prepared- ness for infectious disease out- breaks. Lancet 2017. doi:10.1016/ S0140-6736(17)31224-2. 2. Centers for Disease Control and Prevention. Ebola management.
https://openalex.org/W4295925413	https://aladabj.uobaghdad.edu.iq/index.php/aladabjournal/article/download/1837/3255	Arabic	null	حب التقاط صور السيلفي وعلاقته ببعض سمات الشخصية لدى طلبة الجامعة	Al-ādāb	2,022	cc-by	14,765	DOI: https://doi.org/10.31973/aj.v1i142.1837 DOI: https://doi.org/10.31973/aj.v1i142.1837 Selfie- liking and its relationship to some personality Traits among a sample of university students g p p y among a sample of university students Bushra Othman Ahmad, (PHD) Ministry of Higher Education and Scientific Research - Psychological Research Center Bushra.19742@yahoo.com Bushra Othman Ahmad, (PHD) Ministry of Higher Education and Scientific Research - Psychological Research Center Bushra.19742@yahoo.com ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) Abstract: The aim of the present study is to identify the level of selfie- liking among students of Baghdad University and recognize the differences between males and females, and the correlation between selfie- liking and the personality traits. A random sample of (135) University students has been selected (64males and 61 females), selfie- liking (6 items) and narcissism (8items) scales responding according to 5-Likert scale ranging from (strongly agree- strongly disagree). And Loneliness (6 items), and attention- seeking(4items), and self-centered behavior(5items) scales responding according to 5- Likert scale ranging from (always- never). The results showed that the level of selfie- liking is high among the research sample, and that there are significant differences in selfie- liking, due to the gender, with high level for females., and that there is high correlation between the two variables. After conducting a regression analysis, it was found that the narcissistic personality trait and self-centered behavior made the greatest contribution to the selfie- liking among Baghdad University students. Keywords: Selfie liking, personality traits, Narcissism, attention- seeking, Self –centered behavior, Loneliness, University instructors. 372 2222 / م1111 هـ م. د بذرى عثمان احمد وزارة التعليم العالي والبحث العلمي- مركز البحهث ا لنفدية م. د بذرى عثمان احمد وزارة التعليم العالي والبحث العلمي- مركز البحهث ا لنفدية )(مُلخَصُ البَحث هددت ا درتسد ددى إرددة مست ددى مدددقاط صدد درقلدد ل ىدداس دردددة بغ رددتط ددى مسددى ددتد وقي ه ع ة و ق مق ةتدرجشس. وإيج درسبلقى بةن ص درقل ل ىاس درددة بغ وعسدس دس درذخردديى لدرشت ددديى ودراصددتس درشبددديى و دد ا در حدد عددن دال نق دد و دد ا درقست دد صددا درددتد. Abstract: إم درلددددا ددددمم درا ددددة ى هددددغ س دددد رى يرددددتي McLuhanل2:75 إرة أ متط ؤثت دأل ود درقغ ندقسس ه ع ة س و ش و مثةت س د و ش ع ة درحي س دال قس ليى. و رلا أم لشيى د ر تيتس خ ق مد صى صي س تيتس وأم ت ق دألشي يق ةت غ هدت درسدد صى. سسشدة ،خدت أىد حا لشيد دال رد ال ذدن أ ً رجسي شتدوح درحي س غ درستدسص ودربرا درتسد يى وأم كن درسس وصقة درسش ز . وهتد يسشدغ أم درسادمدد درقكشارا يددى سد وسًد صةاقًدد ددغ كداقن درش دد غ درثلدد غ و زد ع ةشدد ددغ ع رسش درخ ص(Seyfi & Arpaci,2016, p.144). أ ط دنقذددد س و ددد و درقادىددد دال قسددد عغ ذدددن وةدددت إردددة دددا ةت قدددتس ال يردددت مدددن دراقددا درددت يخرردده درشدد ص رهددتد درشددا. مددن درا دد و . وأىدد حا مشقجدد درا دد و درجتيددتس مجددت عش ىددت ال عشددة عشهدد ددغ صي شدد . وأصددت هددت درسش ىددت هددا دردددة بغ ودرددت أىدد ح ددد د أ دددي مدددن صي شددد . رلدددت حدددا إردددة دألكثدددت د دددقسس ال دددغ رويلددد و ددد و درقادىددد دال قس عغ و غ دإلعبلن درقج سقدى ودرتع يدى دردي ديى ودرسبلقد دأل دتقى و دغ إ سد د خدت و س نبدك مبهام رآلختقن . مدد درقردداسد ودرق ةددتد دردددتقسى ددغ لشيدد دال ردد ال ةددت أنسدد ل درحيدد س وعس يددى درقشذددددالى دال قس ليددددى رؤل ددددتد دخدددد درسجقسدددد . إم درلددددا ددددمم درا ددددة ى هددددغ س دددد رى يرددددتي McLuhanل2:75 إرة أ متط ؤثت دأل ود درقغ ندقسس ه ع ة س و ش و مثةت س د و ش ع ة درحي س دال قس ليى. و رلا أم لشيى د ر تيتس خ ق مد صى صي س تيتس وأم ت ق دألشي يق ةت غ هدت درسدد صى. سسشدة ،خدت أىد حا لشيد دال رد ال ذدن أ ً رجسي شتدوح درحي س غ درستدسص ودربرا درتسد يى وأم كن درسس وصقة درسش ز . وهتد يسشدغ أم درسادمدد درقكشارا يددى سد وسًد صةاقًدد ددغ كداقن درش دد غ درثلدد غ و زد ع ةشدد ددغ ع رسش درخ ص(Seyfi & Arpaci,2016, p.144). أ ط دنقذددد س و ددد و درقادىددد دال قسددد عغ ذدددن وةدددت إردددة دددا ةت قدددتس ال يردددت مدددن دراقددا درددت يخرردده درشدد ص رهددتد درشددا. Abstract: ودخقةددت عةشددى عذددادويى منانددى مددنل246 ردد و ر ددى بادقدد ل75 ذ دداسول82 إندد و ددن وشددغ و ت سددى ود ددقسس متيدد ص صدد درقلدد ل ىدداس دردددة بغSelfie- liking scale ومل ييس س درذخريى درسقسث ى رشت ديى و د ا در حد عدن دالنق د ومتيد ص دردددددد ا درسقست ددددد صدددددا دردددددتد ومتيددددد ص دراصدددددتس درشبدددددديى دردددددت دددددن دعدددددتد ه مدددددن قوددددد لCharoensukmongkol 3127 يقددددمري متيدددد ص صدددد درقلدددد ل ىدددداس دردددددة بغ مددددنل7 لتد ومتي ص درشت ديى منل9 لتد ودال دقج ى ع دة و دق دتسا خس دغ يقدتدوي مد بدةن ل د بددق س مدد ل6 - ال د بددق س مدد ل2 أمدد متيدد ص دراصددتس درشبددديى ةقكددام مددنل7 لددتد ومتيددد ص در حددد عدددن دالنق ددد ل5 لدددتد ومتيددد ص ددد ا درقست ددد صدددا دردددتد ل6 لدددتد ودال قج ى ع ة و ق تسا خس غ يقتدوي م بةنل دوس ل6 - دبتدل2 و اى ا درشقد و دردة ص درقل ل ىاس دردة بغ ها ع رتط عةشدى در حد ودنده هشد دتو ذد الردى دغ صد در قل ل ىاس دردة بغ يس ط درة مق ةت درجشس ورر رح دإلن ودرقاى دردة و دا عبلقدى دس يى بةن ص درقل ل ىاس دردة بغ و س درذخرديى درسحدت س دغ درتسد دى وعسدت د دتد ح ة دالنحتدس بةن مق ةدتد درتسد دى ودةن أم درددسى درذخرديى درشت دديى ودردد ا درسقست د صا درتد ودرسد. هسى غ ص درقل ل ىاس دردة بغ رتط ى مسى تد الكلمااات المفتاحيااة : صدد درقلدد ل ىدداس دردددة بغ ددس درذخردديى درشت ددديى در حدد عددن .دالنق درد ا درسقست صا درتد دراصتس درشبديى ى درج مسى و ا ط ة غ اس غ الكلمااات المفتاحيااة :صدد درقلدد ل ىدداس دردددة بغ ددس درذخردديى درشت ددديى در حدد عددن .دالنق درد ا درسقست صا درتد دراصتس درشبديى ى درج مسى 373 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( مةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ مذكلة البحث مذكلة البحث مذكلة البحث مدد درقردداسد ودرق ةددتد دردددتقسى ددغ لشيدد دال ردد ال ةددت أنسدد ل درحيدد س وعس يددى درقشذددددالى دال قس ليددددى رؤل ددددتد دخدددد درسجقسدددد . Abstract: درسادقدد ع ددة درددتعن مددن أم درقلدد ل دألشددخ ص رردداسهن درخ ىددى أو س ددن ىدداسًد ذد يددى يت دد سقخه إرة درسراس درلتيسى ضسن درسس يى درق سقخيى إال أم هشد درستيدت مدن د سد درسخق بدى صا دال قسس دألو رقلشي د ه س دره ي درسحسدا ومد ذردك قود أم يدقن درقسدتا ع يده رسدد درسقق دد ر ددادهت درست رددى نقذدد س درقكشارا يدد ال يسنددن ردده أم ي بدد عددن ددد هتس دنقذددت ذدددن وةددت وأىددد حا عشادندد ألصدددتد درةدداغ درقدددغ ياد ههدد مددددقسس ا درقكشارا يددد ورسدد أبددتز هددت در ادهتل دد هتس دردددة بغ ودرقددغ سشددغ أم ي ددقل درذددخف رشبددده ىدداسس ددغ مند م مدد أو وضدد يى مسةشددى وو لد رساقدد وقنةوةددتي ددتم دررداسس درس قلرددى ذد يدد هددغ ىدداسس شخريى ي قلره ى صوه رشبده قسس ،رى راقت أو قسس ه ي ذ دغ مجهد ةردى سقسيدددى ثدددن يشذدددته ع دددة درذددد ن دال قس ليىلهيددددوا اققت دنددددق تدغ وعةدددت ... أل ددد ب مقشاعددى وال لقرددت هددت در دد هتس ع ددة الددى عستقددى محددت س بدد دنقذددت رقزددن أع دد دربالدد .درسستقى ع ة دخقبلا مادقسهن ومتدك هن ذس ا در در ومذ هةت دربن ودأل ب ودردي ى و لًدد رلدد ماصOxford Dictionary ذددةت ىدداسس دردددة بغ إرددة درقلدد ل دربددت رردداسس ا اعتدهيى ذد يد رشبدده لأو مد أشدخ ص ،خدتقن و دقل ند مةتد خ ىدى أو ند مةتد درهد ي درسحسددا مسدددا ى سردد دردددة بغ أو ما هددى نحددا درسددت،س وقددقن مذدد س قه عدد سً عوددت و دد و درقادى دال قس عغ درسخق بى(Sorokowski et al ,2015) غا( ) أى حا سدى دة بغselfie سدى ش ندى و ا ده شد و و وةدت بدةن درستيدت مدن درذد ب ودرستدهلةن غ دردش اد دألخةتس. و دغ أي مشد هدت ع ر د مد ندتط دألشدخ ص وخ ىدى درذد ب و درستدهلةن وهن يحس ام درك مةتد غ أذسعهن و يا هانه نحا أنبدهن الرقلد ل دررداس دغ كدد مندد م لتق دد . Abstract: ورس ىاس دردة بغ ،خت هت دررتع درقكشارا يى درقغ عةت درسب هين دال قس ليى ودرسبلق دربت يى س أنق هتد درشس درجتيت من درقكشارا ي درحتيثى ها عةت مدوا مدن ددتا دأل ددتد دده اندده أىدد ح ماضددى درسرددت درجتيددت و دد هتس دنقذددت ذددن ود دد ددتد خ ىى بةن الى درذ ب غ د ونى د خةتس و م درقرداس درقكشاردا غ و خدا درهاد دي درت يدى ع رسشددد و رددداس و ددد و درقادىددد دال قسددد عغ درقدددمثةت دركوةدددت دددغ هددداس مددد يسدددتا ددد هتس درددة بغ أو عرددت دردددة بغ دردت أىدد ح ماضددى بددةن درذددساب ود مد م بددةن شددساب أخددتط مىددد ح ي ذدددن هددد س درجسيددد مشدددت هددداس دددن سدددت درددددة بغ مجدددت ىددداسس عباقدددى ي قلرهددد 374 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ (جمل د بة مج )(ي ول)م درذددددخف رشبددددده بدددد هشدددد لشيدددد وىددددتع تيددددتس ردددده ورسدددد الددددى درذدددد ب هددددغ دألكثددددت د ددقسس ال رهددت درقلشيددى درحتيثددى إذ القددا سود دد ود ددس ددغ أو دد هن مسدد س هدد ددؤثت ع ددةهن ربى ع مى وع ة درسبلق دال قس لي ى رتيهن وع ة خر وف درذخرديى و كاقش هد ردبى : خ ىى رتد البت من تي درقد ؤ د غ هد هشد عبلقدى مد بةن د ا درقلد ل ىداس درددة بغ وعسدس دس درذخرديى درسقسث ددى . رشت ديى ودرد ا درسقست صا درتد ودر ح عن دالنق ودراصتس درشبديى؟ أوال: أىمية البحث والحاجة ال : يو درذددددخف رشبددددده بدددد هشدددد لشيدددد وىددددتع تيددددتس ردددده ورسدددد الددددى درذدددد ب هددددغ دألكثددددت د ددقسس ال رهددت درقلشيددى درحتيثددى إذ القددا سود دد ود ددس ددغ أو دد هن مسدد س هدد ددؤثت ع ددةهن ربى ع مى وع ة درسبلق دال قس لي ى رتيهن وع ة خر وف درذخرديى و كاقش هد ردبى : خ ىى رتد البت من تي درقد ؤ د غ هد هشد عبلقدى مد بةن د ا درقلد ل ىداس درددة بغ وعسدس دس درذخرديى درسقسث ددى . Abstract: رشت ديى ودرد ا درسقست صا درتد ودر ح عن دالنق ودراصتس درشبديى؟ أوال: أىمية البحث والحاجة ال : يو ج و ب يا و ي رسدد درسقق دد ر ددادهت درست رددى نقذدد س درقكشارا يدد ال يسنددن ردده أم ي بدد عددن ددد هتس دنقذددت ذدددن وةددت وأىددد حا عشادندد ألصدددتد درةدداغ درقدددغ ياد ههدد مددددقسس ا درقكشارا يددد ورسدد أبددتز هددت در ادهتل دد هتس دردددة بغ ودرقددغ سشددغ أم ي ددقل درذددخف رشبددده ىدداسس ددغ مند م مدد أو وضدد يى مسةشددى وو لد رساقدد وقنةوةددتي ددتم دررداسس درس قلرددى ذد يدد هددغ ىدداسس شخريى ي قلره ى صوه رشبده قسس ،رى راقت أو قسس ه ي ذ دغ مجهد ةردى سقسيدددى ثدددن يشذدددته ع دددة درذددد ن دال قس ليىلهيددددوا اققت دنددددق تدغ وعةدددت ... أل ددد ب مقشاعددى وال لقرددت هددت در دد هتس ع ددة الددى عستقددى محددت س بدد دنقذددت رقزددن أع دد دربالدد .درسستقى ع ة دخقبلا مادقسهن ومتدك هن ذس ا در در ومذ هةت دربن ودأل ب ودردي ى و لًدد رلدد ماصOxford Dictionary ذددةت ىدداسس دردددة بغ إرددة درقلدد ل دربددت رردداسس ا اعتدهيى ذد يد رشبدده لأو مد أشدخ ص ،خدتقن و دقل ند مةتد خ ىدى أو ند مةتد درهد ي درسحسددا مسدددا ى سردد دردددة بغ أو ما هددى نحددا درسددت،س وقددقن مذدد س قه عدد سً عوددت و دد و درقادى دال قس عغ درسخق بى(Sorokowski et al ,2015) أى حا سدى دة بغselfie سدى ش ندى و ا ده شد و و وةدت بدةن درستيدت مدن درذد ب ودرستدهلةن غ دردش اد دألخةتس. و دغ أي مشد هدت ع ر د مد ندتط دألشدخ ص وخ ىدى درذد ب و درستدهلةن وهن يحس ام درك مةتد غ أذسعهن و يا هانه نحا أنبدهن الرقلد ل دررداس دغ كدد مندد م لتق دد . وعذددن أكثددت حتيددتد أىدد ح درقلدد ل درردداس درذخردديىل دردددة بغ نذدد شدد وس يلددتغ درشدد ص مددن خبلردده أنبددده ن ر جسهدداس دو ع مددى درش ص.وعدد س مدد يلامددام بقحسةدد ىدداس دردددة بغ درقددغ ي قلرانهدد ع ددة مادقدد درذدد نى دال قس ليددىSocial networking sites(SNS مثددددددد در يددددددددواFacebook ودالنددددددددق تدغInstagram و دددددددش ب شددددددد Snap chat وعةتهدد مددن مادقدد درقادىدد دال قسدد عغ ومذدد س قه بددةن دالىددتق ددغ هددت . Abstract: مددن درا دد و . وأىدد حا مشقجدد درا دد و درجتيددتس مجددت عش ىددت ال عشددة عشهدد ددغ صي شدد . وأصددت هددت درسش ىددت هددا دردددة بغ ودرددت أىدد ح ددد د أ دددي مدددن صي شددد . رلدددت حدددا إردددة دألكثدددت د دددقسس ال دددغ رويلددد و ددد و درقادىددد دال قس عغ و غ دإلعبلن درقج سقدى ودرتع يدى دردي ديى ودرسبلقد دأل دتقى و دغ إ سد د خدت و س نبدك مبهام رآلختقن . يسنششدد درلددا إندده يذددن دروشيددى دأل دديى رثل ددى درلددتم درحدد ودرسذددتقن. وأىدد ح درلددا درسذدددهاس ألندددت ودسهددداAndy Warhol صتيتيددد دددغ أنددده دددينام درجسيددد دددغ رويلددد دردددة بغ ددغ درسدددقلو مذددهاسًد رسددتس26 قيلددى ذددن س ددغ. ألم درقس يلدد درقددغ يددقن دإل ال به ع ة ىاس دردة بغ ومذ س ى هدت دررداس جدمس زد دأل دتد دغ ماقدي يقحدت ده مدن درجسي . رهتد دردو يتس دأل تد درددة بغ ع دة أنده أ سدت مدن مجدت انهد ىداسس دنده مند م مهن إذ يناندام مقاد دتين هيده منبددهن دغ هدت درحيد س. ردتد دتم سد دى دمثةتد هدتد درقروةدق دردددت يودددتو أنددده مقاد دددت دددغ صي شددد إردددة دألبدددت ع دددة شخرددديى دربدددت و دددس ه درسخق بدددى أمدددت ضت وس(Seyfi & Arpaci, 2016, p.144). Abstract: وعذددن أكثددت حتيددتد أىدد ح درقلدد ل درردداس درذخردديىل دردددة بغ نذدد شدد وس يلددتغ درشدد ص مددن خبلردده أنبددده ن ر جسهدداس دو ع مددى درش ص.وعدد س مدد يلامددام بقحسةدد ىدداس دردددة بغ درقددغ ي قلرانهدد ع ددة مادقدد درذدد نى دال قس ليددىSocial networking sites(SNS مثددددددد در يددددددددواFacebook ودالنددددددددق تدغInstagram و دددددددش ب شددددددد Snap chat وعةتهدد مددن مادقدد درقادىدد دال قسدد عغ ومذدد س قه بددةن دالىددتق ددغ هددت . درسادقدد ع ددة درددتعن مددن أم درقلدد ل دألشددخ ص رردداسهن درخ ىددى أو س ددن ىدداسًد ذد يددى يت دد سقخه إرة درسراس درلتيسى ضسن درسس يى درق سقخيى إال أم هشد درستيدت مدن د سد درسخق بدى صا دال قسس دألو رقلشي د ه س دره ي درسحسدا ومد ذردك قود أم يدقن درقسدتا ع يده 375 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) دمجةج ةدددتًد دددغ درااليددد درسقحدددتس دددن حتيدددت سدددى دددة كSelca دددغ اسقددد درجشاعيدددى دددغ عددد غ 3122 . Selca هدغ سدى مت دى جسد بدةن "self وcamera" مسد يدت ع دة دررداسس درتد يى ر ذخف(Kwon & Kwon,2011, p.660). و غ عد غ3125 نذدت درجس يدى دألمتقنيدى ر رد درشبددغ(APA) ًملد ال ردشي هيده هاص دردة بغ(selfitis) ك ضرتدب عل غ تيت لVincent 2014) د عة درسل أم هاص دردددة بغ هددا درتا دددى درلهتقددى ردددتط دربددت الرقلددد ل ىدداس رشبدددده ونذددته ع دددة و دد و درقادىددد دال قس عغ ا ة ى ر قساقس عدن دتنغ لدتيت دردتد ومد دربجداس دغ درسبلقدى درحسيسدى رتيده (Balakrishnan & Griffiths,2018,p.722). وأشدد س درددة أم هشدد ثبلثددى مدددقاق رهددتد دالضددرتدب وهددا درسدددقاط درخ يدد لدرقلدد ل دربدددددت ر رددددداس بشبدددددده ثدددددبل مدددددتد ع دددددة دألقددددد ياميًددددد وم نذدددددته ع دددددة مادقددددد درقادىددددد دال قسد عغ درح لدرقلدد ل ىدداس رتد دده ثددبل مددتد ع دة دألقدد ياميًدد و نذددت دد ىدداسس مددن دررددداس ع دددة و ددد و درقادىددد دال قسددد عغ ودرس منلص ردددى اليسندددن دردددقحنن بهددد دددغ درقلددد ل درراس ذد ي ع ة متدس درد عى ونذته ع ة مادق درقادى دال قس عغ أكثت من دا مدتد غ درةاغ .وأعةت نذت درلرى ع دة درستيدت مدن درسادقد دإلخ سقدى دغ سيد أنحد درسد رن وركدن ددتع م مدد وددةن أ م درلرددى نددا مجددت ختعددى. Abstract: ومدد ذرددك مث سدد بددتأ در حدد درقجتقوددغ ددغ إ م م دالنقتنا ست نذت مس يةت خ عى رئل م م ع ة دإلنقتنا من قو إيبد م اردتبةتا دغ عدد غ2::6 (Widyanto and Griffith,2010) . دد م نبددس درذددغ صددت مدد ماضددا دضرتدب دردة بغ(Balakrishnan & Griffiths,2018, pp. 722-723). وىدي د مدن(Ma et al.2017) درقلد ل ىداس درددة بغ مدن صةد ن تقدى لدتين درتدself- presentation ودرقغ يقن رويله ر قمثةت ع ة د ختقن. إذ سا دإلشد سس دردة أم درقلددد ل ىددداس درددددة بغ رددديس سدددبل دو ددد ا ً مددددقلبلً ألنددده يمخدددت أ سددد ًد أخدددتط عشدددتم يدددقن مذددد س قه عودددت و دددد و درقادىددد دال قسدددد عغ. مثددد هددددت دردددد ا ي سنددددن م قلردددغ ىدددداس ً دردة بغ من لتين أنبدهن رتقلى مديرت ع ةه . ودرقغ أى حا غ دردشاد دألخةدتس نذد شددد وسً ذدددن ال يردددت ع دددة دإلنقتندددا دددغ ثةدددت مدددن دألصيددد م عشدددت مذددد س قه عودددت مادقددد درقادى دال قس عغ ودرقغ أى حا و ة ى تيتس ر قسوةت عن درتد ودرقسثة دردتد غ( Ma et al,2017, p.139). وع ر م يقن نذت ىداس درددة بغ ع دة مشرد درقادىد دال قسد عغ درسددقسس ى وذردك ك شدد سس رقلددتين درددتد لMehdizadeh,2010; Papachaissi 3122 . وقددت ددن درقاىدد درة أم درتد ر قسوةت عن درتد وصتقى درقحنن ست مؤشتد قاقى عن درذخريى(Gosling et al., 2002) لدت دا ت ىداس درددة بغ ىداسس أ زد عدن درددس درذخرديى ألىدح به ةدددتًد دددغ درااليددد درسقحدددتس دددن حتيدددت سدددى دددة كSelca دددغ اسقددد درجشاعيدددى دددغ عددد غ 3122 . Selca هدغ سدى مت دى جسد بدةن "self وcamera" مسد يدت ع دة دررداسس درتد يى ر ذخف(Kwon & Kwon,2011, p.660). ةدددتًد دددغ درااليددد درسقحدددتس دددن حتيدددت سدددى دددة كSelca دددغ اسقددد درجشاعيدددى دددغ عددد غ 3122 . Selca هدغ سدى مت دى جسد بدةن "self وcamera" مسد يدت ع دة دررداسس درتد يى ر ذخف(Kwon & Kwon,2011, p.660). و غ عد غ3125 نذدت درجس يدى دألمتقنيدى ر رد درشبددغ(APA) ًملد ال ردشي هيده هاص دردة بغ(selfitis) ك ضرتدب عل غ تيت لVincent 2014) د عة درسل أم هاص دردددة بغ هددا درتا دددى درلهتقددى ردددتط دربددت الرقلددد ل ىدداس رشبدددده ونذددته ع دددة و دد و درقادىددد دال قس عغ ا ة ى ر قساقس عدن دتنغ لدتيت دردتد ومد دربجداس دغ درسبلقدى درحسيسدى رتيده (Balakrishnan & Griffiths,2018,p.722). Abstract: وأشدد س درددة أم هشدد ثبلثددى مدددقاق رهددتد دالضددرتدب وهددا درسدددقاط درخ يدد لدرقلدد ل دربدددددت ر رددددداس بشبدددددده ثدددددبل مدددددتد ع دددددة دألقددددد ياميًددددد وم نذدددددته ع دددددة مادقددددد درقادىددددد دال قسد عغ درح لدرقلدد ل ىدداس رتد دده ثددبل مددتد ع دة دألقدد ياميًدد و نذددت دد ىدداسس مددن دررددداس ع دددة و ددد و درقادىددد دال قسددد عغ ودرس منلص ردددى اليسندددن دردددقحنن بهددد دددغ درقلددد ل درراس ذد ي ع ة متدس درد عى ونذته ع ة مادق درقادى دال قس عغ أكثت من دا مدتد غ درةاغ .وأعةت نذت درلرى ع دة درستيدت مدن درسادقد دإلخ سقدى دغ سيد أنحد درسد رن وركدن ددتع م مدد وددةن أ م درلرددى نددا مجددت ختعددى. ومدد ذرددك مث سدد بددتأ در حدد درقجتقوددغ ددغ إ م م دالنقتنا ست نذت مس يةت خ عى رئل م م ع ة دإلنقتنا من قو إيبد م اردتبةتا دغ عدد غ2::6 (Widyanto and Griffith,2010) . دد م نبددس درذددغ صددت مدد ماضددا وع ر م يقن نذت ىداس درددة بغ ع دة مشرد درقادىد دال قسد عغ درسددقسس ى وذردك ك شدد سس رقلددتين درددتد لMehdizadeh,2010; Papachaissi 3122 . وقددت ددن درقاىدد درة أم درتد ر قسوةت عن درتد وصتقى درقحنن ست مؤشتد قاقى عن درذخريى(Gosling et al., 2002) لدت دا ت ىداس درددة بغ ىداسس أ زد عدن درددس درذخرديى ألىدح به 376 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( ل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ مل سنى رراس دألختط. وم ذرك لت أ هت درتسد أم دأل تد يناناد ق لةن ذمم ىاسس درتد درخ ىى بهن ع ة دإلنقتنا و يسستوم درة مس رجى لتيسهن ر تد إلنذد ىداسس ر دتد متعاعى د قس ليً . قت ن درقاى إرة أنهن يس زوم ذود هن ودرحرا ع دة س و سد إيج بيدى عن تقق ش ن درقادى دال قس عغ عوت ىاس درس ي درقستقبغ درذخرغ(Qiu et al., Social comparison theory Social rank theory وهدددتد مددد أشددد س إريددده عدددت مددددن در صثةن درة أنه غ دراقا درت يدسح هيه درقل ل مث هت درراس رؤل تد رقتوق ألنبدهن رس مددى درشدد ص ودال مدد م ع ددة درقلدد ل ىدداس درددد بغ قددت يقدددو ددغ ردداقت دال ددتد ر ددد ا ي عةددددت دررددددحيى مثدددد درشت دددددديى ودالن نيددددى وقددددت يددددؤثت ددددد ع ددددة عبلقدددد هن مدددد دالخدددددتقن لSorokowski et al., 2015, p . 235 . أش سHess(2015) إرة أنه يقن د قسس ىاس ة بغ غ د مدن دألمد كن درخ ىدى ودرس مددى زددبل عددن ذرددك يسددت دد ا دردددة بغ ردديس مجددت درقلدد ل ىدداسس وإنسدد يذددس حتقددت در ددام ودرق دد ين و ةةددت درخ يدد وإضدد ى ددمثةتد أخددتط ع ددة درردداسس قودد حسة هدد ع ددة مشرددى درقادىدد دال قسدد عغ. أ هددت درخيدد سد درسزدد ى ود ددقسس درقحتقددت درسقك مدد إرددة زقدددددد س شددددددسويى دددددد ا درقلدددددد ل درردددددداس دردددددددة بغ إذ يذددددددةت(Liuubiniene & Keturakisل3125 درددة أندده يسنددن ر سدددقسس ةن متدق ددى إبددتدع ىدداسهن درذخردديى ع ددة أنهدد ىدداسس مسنا ددى سة ددى رهددن. ودز د وعذددن م حددا ددغ دردددشاد دألخةددتس شددتد در زدد و درست رى مخت ىاس ة بغ لمث عر ىاس دردة بغ رقحدةن نرد دررداس(Flaherty and Choi, 2016, p.1). وقددت شددس در حدد ددغ درتسد دد درقددغ ش ورددا دد ا دردددة بغ ددغ درستيددت مددن درسجدد ال درسخق بددى. سص در دد صثام دد ا دردددة بغ ددغ عبلققدده بدد سس درسق ةددتد مثدد درجددشس ودرسددت (Allburry.2015) ود دقسس ره دغ درددي دردي دغ(Baishya.2015; Deller and Tilton.2015) وىددددداس درددددددة بغ درسددددددنتقى(Dishy . 3128 وىددددداس در ددددددة بغ را ددددد و درقتا لMarwick 3126 وعبلقى ا دردة بغ ب سس س درذخريى (Choi et al. 2017؛ Qiu et al . 3126 . و دغ إ د س سد دقه ع دة و دق مق ةدت درجدشس و دت د مدن Fox and Rooneyل3126 دم درددت اس يسة ددام درددة لتقددت نذددت ىدداس دردددة بغ ذددن مقكتس ودع ة من دإلن . وم ذرك دكقذي من اس ا دنغ وزمبل ؤSorokowski et al.ل3126 دم دالن دكثت نذتد رراس دردة بغ من درت اس. Social comparison theory Social rank theory ومدن ن صيدى أرخدتط دتم ن تقدى درسل سندى دال قس ليدىSocial comparison theory (Festinger 2:65 ون تقددددى درقرددددشي دال قسدددد عغSocial rank theory (Gilbert 3111 درقدددغ ؤ دددت ع دددة وس دروةالدددى دال قس ليدددى دددغ بددددةت ددد ا درشذدددت عودددت مادق درذ نى دال قس ليى لSNS ودرقغ سشغ ضدسش دم دأل دتد يشقسدام دردة مجساعد مدن دألقددتدم يقرددبام بتس ددى ع ريددى مددن درس ش دددى ودرسل سنددى دال قس ليددى بددةن دالعزدد هددن دألكثددت دصقسدد ال ددغ درقلدد ل ىدداس دردددة بغ ونذددته ددغ ددك درسادقدد ريجس دداد أنبدددهن أكثددت بتدعددى مددن دقتدنهنلTandoc, Ferrucci & Duffy,2015, p . 253 . وعسدد دد مددنGoffman and Mezirow دد رقت ة ع ددة دألىدد درددد ا غ رهددت در د هتس. سدت دم ثتهن بش تقدى دأل دس دألخبلقيدى Moral foundations theory إذ يت د Goffman(1959) ع ددة دررتقلددى درقددغ نلددتغ بهدد أنبدددش دو ذود شدد ددغ درحيدد س درةاميددى بةشسدد يؤ ددتMezirow(1991) ع ددة دركي يددى درقددغ نتددين بهدد ذد يدد ألنبدددش عددن تقددق س و سدد دألقدددتدم وذردددك ددددو سعوقشددد دددغ أم نكدددا م ددد ًد مدددن مجقسددد أقتدنشددد . وقذدددةتDuver Miclot(2015) إرددة أم هددت دأل كدد س مش دد ى ر ددتد درحددتي أل ددتد درجةدد درحدد رغ درددتين يشخت ام غ درقادى دال قس عغ من خبلselfie . وقتقتوم من د خدتقن سؤقدقهن س يًد و ددغ دراقددا درحدد رغ مددن دررددس ددتًد مدد مبهدداغ دردددة بغ ددغ ست قدد ودصددت ألم أعتدضدده وأهتد دده ق ةددت ددقستدس. ومدد ذرددك يسنددن ستقبدده سرددر ح عدد غ ع ددة أندده ردداقت دأل ددتد ألنبدهن منبدهن لSeyfi & Arpaci,2016, p. 145). 377 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) و ددددغ هددددتد درسجدددد دنقلددددت عددددت مددددن دررددددحبةةن أم درقلدددد ل درردددداس درذخردددديى درتد يددددىل دردددة بغ ع ددة أندده دد ا أندد نغ الندده يسةدد درشدد ص ثةددتد درددة داله قسدد غ س هددتهن ددغ درردداس وهدددتد يجس هدددن أصي نددد يقجددد ه ام دألشدددخ ص درسحيردددةن بهددددن. Social comparison theory Social rank theory أم مدن صةد درسسدت و دت Weiser ل3126 دم س دردن يسة ام إرة لتقت نذت ىاس دردة بغ ذن مقكتس ودع ة مددن درذدد ب إال أندده و ددغ د ونددى دألخةددتس ش مددا مخدد وا درسجقسدد الس دد ل سددس خردد وف درذخردديى درددد ويى ردددة بغ وهددتد مدد أشدد س إريدده عددت مددن درتسد دد مشهدد درسسدد درتدوددت ددغ هدتد درسجد ندا ر تسد دى درقدغ أ تدهد د مدنFox and Rooney ل3126 إذ س د ع دددددددددة دس ددددددددد ل صددددددددد درقلددددددددد ل ىددددددددداس درددددددددددة بغ رشت دددددددددديى Naracissim ودرسين ة يددددددددد 378 ( مةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) Mechiavellianism ودردددددددددديناع ثيى Psychopathy . وقدددددددددت أ لددددددددداد ع ةهددددددددد ددددددددددس درذخريى درثبلثيى درس سى ودرست رى رقل ل ىاس دردة بغ. إذ قسة درسين ة يى رتا ى دغ ويدددى دصقي ددد دربدددت مددد متدعددد س درل ةددد رؤلخدددبل و قسةددد درشت دددديى إلصدددد ص رس سدددى ودرهيسشدددى ودال ددددقحل دردددت يجسدددد د أل دددتد يش ددددتوم إرددددة أنبددددهن ع ددددة أنهدددن مقباقددددام ع ددددة د خدددتقن ودرذخرددديى درددددديناع ثيى ال قلددد س إردددة درقسدددد ي ودردددد ا ي دالنت ليدددى و دددد ا در ح عن دالنق س درش ت عن درقك بى درقغ يقحس ه د ختوم و هتد در ح دألو حا أرخدتط مدن قود د مدنSorokowski et al(2015) and Weiser ل3126 دغ د قكذ ا درسبلقى بةن درقل ل ىاس دردة بغ ودرشت ديى. د قسس اد عةشد مدن بارشدتد ودرااليد درسقحدتس. و دغ در حد دردت أ دت مدن قود د مدنQiu, et al . ل3126 دن د قكذد ا درسبلقى بدةن درددس درخسدس دركودتط ر ذخرديى و د ا درقلد ل ىداس درددة بغ ردتط عةشدى مد ن دررةشةةن مشه أم رشذت ىاس دردة بغ دس ل دس درذخريى درلواagreeableness ودالن د يىextraversion ودرقدغ سندس صد دخدقبلل دأل دتد د ختقن ردترك دتم دأل دتد درتين يحوام راقت درسبلق دال قس ليى م د ختقن يسة ام درة أم ينام رتيهن د د وةدت الرقلد ل ىداس درددة بغ (Fox & Rooney ,2015,p.161; Sorokowski et al, 2015, p.123; Qiu et al, 2015, p.443). ردترك ومدن درسدت ح أم يندام سوةدت ًدرذخريى غ ىاس درددة بغ خق دي عدن دك درسا دا س دغ أنداد. أخدتط مدن دررداس. Social comparison theory Social rank theory غ درراس درتي يى أو دالعقي يى وع رش ت درة درستيت من در حا درد لى درقغ سعرا بةن س درذخريى ودرقل ل ىاس دردة بغ ودرقغ حثا ع ة نرد ود د دتم درهدتا مدن هدت درتسد دى هدا لدتين درس قدت صدا هتد درساضا. قسس عةشى من درذد بل دى درج مسدى وع دة و ده درخرد اص يقشد و در ح درح رغ درخر وف درذخريى رؤل تد درقغ ره دس ل ح درقل ل دردة بغ يت در ص غ درتسد ى درح ريى ع ة أسع خر وف ر ذخريى قت دت حد درقلد ل ىدداس دردددة بغ قزددسن : ل2 درشت ددديى ل3 ودرذددساس راصددتس درشبددديى ل4 و دد ا در حدد عن دالنق ل5 و ا درقست صا درتد. 2 . . مدقاط ص درقل ل ىاس دردة بغ رتط ى مسى تد 3 . .مدقاط ص درقل ل ىاس دردة بغ رتط ى مسى تد ع ة و ق مق ةت درجشس 4 . درسبلقدددى بدددةن صددد درقلددد ل ىددداس درددددة بغ و عسدددس دددس درذخرددديىل درشت دددديى ودراصدددتس . درشبديى و ا در ح عن دالنق و ا درقست صا درتد حددددت در حدددد درحدددد رغ ر ددددى مسددددى ددددتد ر سدددد غ درتسد ددددغ ل3129 - 312: درتسد ددددى درر صيى ر قخررةن درس سغ ودإلند نغ ومن بل درجشدةنل ذ اس- . إن تحديد المرظلح ات Social comparison theory Social rank theory زدبل عددن أم ىدداس دردددة بغ حقددا ع ددة أع ددة إشدد سد تقددتس ال قددا ت ددغ أندداد. أخددتط مددن درردداس. ع ددة ددوة درسثدد و دده در رددىduck face وهددا سوةددت درددت يددقن إ ددتدؤ عددن تقق درذبقةن رؤلم غ ورؤلع ة إلعر م هت شب وةتس وع ر ً م يقن سؤققه غ ىاس دردة بغ وركدن رديس دغ أنداد. أخدتط مدن دررداس. قدت كذدي مثد هدت دالشد سد سوةدتًد تيد تًد ر ذخريى غ درراس. رترك(Qiu et al, 2015, p443). أ هددددددت درتسد دددددد درددددددد لى أم درردددددداس حقددددددا ع ددددددة إشدددددد سد صتيتيددددددى ذد عبلقددددددى رذخريى. س منNeslteret et al.( 2012) ع ة مجساعى من درراس درسحت س درخو س ي ع دالنبق ه ذبيى در ت د أم دالن ل ق درذب ر ذخريى غ درراس. رترك(Qiu et al, 2015, p443). أ هددددددت درتسد دددددد درددددددد لى أم درردددددداس حقددددددا ع ددددددة إشدددددد سد صتيتيددددددى ذد عبلقددددددى رذخريى. س منNeslteret et al.( 2012) ع ة مجساعى من درراس درسحت س وو تود أم دالن د ل مت ج ذبيى درا ه ودالنبق ي ع ة درخوتس ي ت حجدن درذدب وقدت صةاقددى درزددسةت مناثددى درا دده. هددت دإلشدد سد قس ددق ذددن أ ددغ سةدد د درا دده وال يسنددن ةةتهد عشدت درقلد ل دررداس(Neslter et al., 2012, p.695 . بةشسد د دقسس ا سد د أخددتط ىدداسًد عباقددى درقلرهدد درسذددقت ةن رقجتعددى وو ددت أم دالن ددد ل دد م مت رًدد روهجددى ودالبقددد غ بةشسدد دس رددا درشت ددديى رج ذبيددى ودرسبل ددس دروتدقددى ودرسنيدد ا زددبل عددن ذرددك كددام وقبددى درسش دددرام رتقلددى أكثددت نذدد ً بةشسدد وقبددى دالنرددادوةةن رتقلددى أكثددت ددا تد ددغ دررداس درس قلردى رك مد درجددن(Borkenau et al. 2009, p.703; Naumann et 379 2222 / م1111 هـ al., 2009, p.1666; Vazire et al., 2008, p. 1439). .وع رسل سندى مد دألنداد دألختط من درراس سشح ىاس ة بغ دأل تد م قتًد من درحتقى غ دردقحنن دغ سؤقدى و ههدن ودرقسوةددت درسدد بغ وماضدد دركدد مةتد. رددترك قددت حقددا ع ددة إشدد سد تيددتس مثدد و دده در رددى ودس ب . درك مةتد ودرقغ ال قا ت . غ درراس درتي يى أو دالعقي يى ودس ب . درك مةتد ودرقغ ال قا ت . أوال: الديلفيSelfie . McCullough et al(2002) محتس ق ماص أو دباس دالنج ة ل3124 ك سى دردة بغ ها ا ده دادهت رد عت ه ود- ر رداسس درقدغ ي قلر هد درسدت بشبدده وعدددد س مدددد يددددقن درقل هدددد بهدددد ي ذ ددددغ أو دددد مةتد وقجدددد مذدددد س قه عوددددت و دددد و درقادىدددد . دال قسدد عغ وقسددتا صددد درقلدد ل درددددة بغ ددغ هدددت درتسد ددى ع دددة أندده شدددساس دأل ددتد دددمنهن مت رام ع ي رقل ل ىاس دردة بغ و مجه دغ أنذدرقهن درةاميدى. وإم أخدت هدت دررداس .ها نذ ل مهن غ صي هن عةشى مسث دى رسحقداط درشرد دردد ا غ رسبهداغ صد درقلد ل ىداس درددة بغ درسقزدسشى دغ دأل. دس وقسوت عشه بتس ى ألعتد هتد در ح 37: ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ ثانيا : سمات الذخرية: ثانيا : سمات الذخرية: دربتو بةن دأل تد وهغ عنس درح رى أرواس لAllport 2:74 : منه ن غ نبدغ عروغ يقسة رقسسين ودرقست وقخدقف دد ربت ورتيدده درلددتسس ع ددة نلدد درستيددت مددن درسشوهدد درسقس رددى و ي يدد ً وع ددة درخ ددق ودرقا يدده درسدقستقن ألشن مقس رى من درد ا درقسوةت ودرقاد لغ ودردسى عشت ره و ا صتيلغ غ . دربت سسشة أم دردس هغ درحل وق درشه ويى ر قش ين درشبدغ غ درذخريى ك لCattle 2:61 : منه مجساعى س و دأل س ودال قج درقغ يتعره ندا. مدن دراصددتس درقددغ دددسح رهددت دال ددقج درقشوددا سدد دديبس ه دربددت وأم اضدد حددا د ددن ودصددت ومس رجقه ررتقلى ذد ه غ مس ن دألصاد . ودردسى عشت ترك هغ ن ث بدا نددوي مدن خردد وف درذخردديى وهددغ سددت عدد م غ يدددقختا باد ددرى درقح ةدد درسدد م غ ربلخق دد سد أ . دربتو بةن دأل تد وهغ عنس درح رى دي ندددك لEysenck . 2:64 : مجساعدددى مدددن دأل سددد دردددد ا يى درقدددغ ق ةدددت مسددد و سدددت . دردس عشت مب هين ن تقى أكثت مشه وصتد صديى ة باس لGuilford 2:6: : ن يسندن سةةد وذو ودغ نددوغ ع دة أ ده يخق دي دربت عن عةت .لعوت درخ رق2:98 ص76 - 77 ة باس لGuilford 2:6: : ن يسندن سةةد وذو ودغ نددوغ ع دة أ ده يخق دي دربت عن عةت .لعوت درخ رق2:98 ص76 - 77 و ست هدددد در صثددددى إ تدويدددد : عةشددددى مسث ددددى رسحقدددداط درشردددد درددددد ا غ رسبهددددا غل درشت ددددديى ودراصتس درشبديى و ا در ح عن دالنق ودرد ا درسقست د صدا دردتد درسقزدسشى دغ . أ ود در ح وقسوت عشه بتس ى ألعتد هتد در ح إ و ست هدددد در صثددددى إ تدويدددد : عةشددددى مسث ددددى رسحقدددداط درشردددد درددددد ا غ رسبهددددا غل درشت ددددديى ودراصتس درشبديى و ا در ح عن دالنق ودرد ا درسقست د صدا دردتد درسقزدسشى دغ . أ ود در ح وقسوت عشه بتس ى ألعتد هتد در ح إ مدد درقردداسد ودرق ةددتد دردددتقسى ددغ لشيدد دال ردد ال ةددت أنسدد ل درحيدد س وعس يددى درقش ذدددالى دال قس ليدددى رؤل دددتد دخددد درسجقسددد . ددد م ك اهددد مMcluhan(1969) أم ددد . ثانيا : سمات الذخرية: أرواس لAllport 2:74 : منه ن غ نبدغ عروغ يقسة رقسسين ودرقست وقخدقف دد ربت ورتيدده درلددتسس ع ددة نلدد درستيددت مددن درسشوهدد درسقس رددى و ي يدد ً وع ددة درخ ددق ودرقا يدده درسدقستقن ألشن مقس رى من درد ا درقسوةت ودرقاد لغ ودردسى عشت ره و ا صتيلغ غ . دربت سسشة أم دردس هغ درحل وق درشه ويى ر قش ين درشبدغ غ درذخريى ك لCattle 2:61 : منه مجساعى س و دأل س ودال قج درقغ يتعره ندا. مدن دراصددتس درقددغ دددسح رهددت دال ددقج درقشوددا سدد دديبس ه دربددت وأم اضدد حددا د ددن ودصددت ومس رجقه ررتقلى ذد ه غ مس ن دألصاد . ودردسى عشت ترك هغ ن ث بدا نددوي مدن خردد وف درذخردديى وهددغ سددت عدد م غ يدددقختا باد ددرى درقح ةدد درسدد م غ ربلخق دد سد أ . دربتو بةن دأل تد وهغ عنس درح رى دي ندددك لEysenck . 2:64 : مجساعدددى مدددن دأل سددد دردددد ا يى درقدددغ ق ةدددت مسددد و سدددت . دردس عشت مب هين ن تقى أكثت مشه وصتد صديى ة باس لGuilford 2:6: : ن يسندن سةةد وذو ودغ نددوغ ع دة أ ده يخق دي دربت عن عةت .لعوت درخ رق2:98 ص76 - 77 و ست هدددد در صثددددى إ تدويدددد : عةشددددى مسث ددددى رسحقدددداط درشردددد درددددد ا غ رسبهددددا غل درشت ددددديى ودراصتس درشبديى و ا در ح عن دالنق ودرد ا درسقست د صدا دردتد درسقزدسشى دغ . أ ود در ح وقسوت عشه بتس ى ألعتد هتد در ح ثانيا : اإلطار النظري المقدمة أرواس لAllport 2:74 : منه ن غ نبدغ عروغ يقسة رقسسين ودرقست وقخدقف دد ربت ورتيدده درلددتسس ع ددة نلدد درستيددت مددن درسشوهدد درسقس رددى و ي يدد ً وع ددة درخ ددق ودرقا يدده درسدقستقن ألشن مقس رى من درد ا درقسوةت ودرقاد لغ ودردسى عشت ره و ا صتيلغ غ . دربت سسشة أم دردس هغ درحل وق درشه ويى ر قش ين درشبدغ غ درذخريى ك لCattle 2:61 : منه مجساعى س و دأل س ودال قج درقغ يتعره ندا. مدن دراصددتس درقددغ دددسح رهددت دال ددقج درقشوددا سدد دديبس ه دربددت وأم اضدد حددا د ددن ودصددت ومس رجقه ررتقلى ذد ه غ مس ن دألصاد . ودردسى عشت ترك هغ ن ث بدا نددوي مدن خردد وف درذخردديى وهددغ سددت عدد م غ يدددقختا باد ددرى درقح ةدد درسدد م غ ربلخق دد سد أ . ثانيا : سمات الذخرية: رلت حا إرة دألكثت د قسس ال دغ رويلد 381 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ و دد و درقادىدد دال قسدد عغ و ددغ دإلعبلندد درقج سقددى ودرتع يددى دردي دديى ودرسبلقدد دأل ددتقى و غ إ سد د خت و س نبدك مبهام رآلختقن (Seyfi & Arpaci, 2016, p.144). يسنششددد درلدددا إنددده يذدددن دروشيدددى دأل ددديى رثل دددى درلدددتم درحددد ودرسذدددتقن ومدددن خدددبل درقرددداسد درقدددغ صدددتثا دددغ لشيددد دال رددد ال ةدددت عس يددد درقلةدددين دردددتد غ رؤل شدددخ ص و راس . بتأ درش ص د م غ م ذود هن سذ عت دإلعج ب درسلتمى مدن درسجقسد مدن خدبل مذددد س قهن ىددداسهن درددددة بغ ع دددة و ددد و درقادىددد دال قسددد عغ وع رقددد رغ دددمثةت ذردددك ع دددة . س هن درذخريى ثانيا : سمات الذخرية: لشيددى د ردد تيددتس خ ددق مددد صى صيدد س تيددتس وأم ددت ق دألشددي يق ةددت سدد رهددت درسددد صى سسشددة أندده أىدد حا لشيدد دال ردد ال ذددن أ ً دد رجسيدد شددتدوح درحيدد س ددغ درسددتدسص ودربرددا درتسد دديى وأمدد كن درس سدد وصقددة درسشدد ز . وهددتد يسشددغ أم درسادمدد درقكشارا يددى سدد وسًد صةاقًدد ددغ كدداقن درش دد غ درثلدد غ و زدد ع ةشدد ددغ عدد رسهن درخدد ص ددغ(Seyfi & Arpaci, 2016, p.144). أ ط سسين ودنقذ س و و درقادىد دال قسد عغ دغ أثشد درددشاد درسذدت درس ضديى إردة ددا ةت قددتس ال يرددت مددن دراقدد ا درددت يخرردده درشدد ص رهددتد درشددا. مددن درا دد و . أىدد حا مشقجدد درا دد و درجتيددتس مجددت عش ىددت ال عشددة عشهدد ددغ صي شدد . ودصددت هددت درسش ىددت هددا .دردة بغ ودرت أى ح د أ ي من صي ش . رلت حا إرة دألكثت د قسس ال دغ رويلد مدد درقردداسد ودرق ةددتد دردددتقسى ددغ لشيدد دال ردد ال ةددت أنسدد ل درحيدد س وعس يددى درقش ذدددالى دال قس ليدددى رؤل دددتد دخددد درسجقسددد . ددد م ك اهددد مMcluhan(1969) أم ددد . لشيددى د ردد تيددتس خ ددق مددد صى صيدد س تيددتس وأم ددت ق دألشددي يق ةددت سدد رهددت درسددد صى سسشددة أندده أىدد حا لشيدد دال ردد ال ذددن أ ً دد رجسيدد شددتدوح درحيدد س ددغ درسددتدسص ودربرددا درتسد دديى وأمدد كن درس سدد وصقددة درسشدد ز . وهددتد يسشددغ أم درسادمدد درقكشارا يددى سدد وسًد صةاقًدد ددغ كدداقن درش دد غ درثلدد غ و زدد ع ةشدد ددغ عدد رسهن درخدد ص ددغ(Seyfi & Arpaci, 2016, p.144). أ ط سسين ودنقذ س و و درقادىد دال قسد عغ دغ أثشد درددشاد درسذدت درس ضديى إردة ددا ةت قددتس ال يرددت مددن دراقدد ا درددت يخرردده درشدد ص رهددتد درشددا. مددن درا دد و . أىدد حا مشقجدد درا دد و درجتيددتس مجددت عش ىددت ال عشددة عشهدد ددغ صي شدد . ودصددت هددت درسش ىددت هددا .دردة بغ ودرت أى ح د أ ي من صي ش . الديلفي وإدراك الذاتSelfie and self- perception . رلت م مبهاغ درتد ماضاعً ر ستيت من درس اغ مث ع ن درشبس وع ن دال قسد . ودرس داغ درقتعاقدددى. خ ىدددى مددد درقرددداسد دددغ لشيددد دال رددد ال دددغ درددددشاد دألخةدددتس لدددت دكقددددوا من نددى مهسددى ددغ ع دداغ دال ردد ديزدد . و لًدد ردددOyserman(2004) ي حدد مبهدداغ درددتد دال ى عن دأل ال ى مث من أن ؟ و أين أنقسدغ؟ ر دت د وس مهدن ر يدى دغ ىدش درلدتدس و ود د و د ا ي دأل دتد(Hagger & Chatzisarantis,2005,p.71). أمد رشدد ى ردد Purkey(1970) هيسددتا مبهدداغ درددتد مندده ن دد غ مسلددت و يشدد مينغ رسسقلددتد دربددت عددن نبدده وقدتطYavuzer(2001) منده دررتقلدى درقدغ يش دت بهد دربدت إردة نبدده و ك ت ده عدن مددن ومدد ها وهاققدده. وال قددمري درددتد لدد مددن ردداسد دربددت و اقس دده وإنسدد ،سد د خددتقن .درسهسددةن ددغ صي دده مىددتق وه وع و قدده ودرستس ددةن .إر صاردده وصددا دد ا ي هن ج هدده وعسسشة ،خت سد لقدتب ىداسس دردتدSelf- image دردتد درقدغ يدقن إ سدكهد لكيد يدتط نبدده مدن دردتد درسث ريدىIdeal self لكيد يتقدت أم يندام يد د لدتيت دربدت رتد ده أمد Baymur(1994) دردددتد هدددغ قددداس دخ يدددى لددداغ رسبل دددى ودرحندددن ودرقلةدددين ودإل دسس مدددن خبل ش ين ا ي ش. وع رشد ى رراس دردة بغ تم درلداس درتدخ يدى درقدغ تدقوشد كدام مشبقحدى ر قمثةتد درخ س يدى وأىد حا أ ده دغ درقبلعد بهد . هشد يدقن ذدنة درباضدة بدةن رداس دربت درحتيلغ ر تد وإ سدكه درسث رغ ر تد ألم مذ س ى ىاس دردة بغ عوت و و درقادىد دال قسدد عغ وإعجدد بهن بهدد ودرقس ةددق ع ةهدد يددؤثت ذددن م شددت ع ددة درقلةددين درددتد غ ر ذددخف لSeyfi & Arpaci,2016, pp, 145-146). ومن درادضح أم أ ود دال ر ق ةت و ا ه ن ت درش ص إرة درتد . و لً ر تسد ى درقدغ لSeyfi & Arpaci,2016, pp, 145 146). ومن درادضح أم أ ود دال ر ق ةت و ا ه ن ت درش ص إرة درتد . الديلفي وإدراك الذاتSelfie and self- perception . يددقن د ددقسس ره ذددن عدد غ عشددتم ينددام درسسثدد ودليدد درددة دندده مبلصدد مددن درجسهدداس ومددن ثددن يددؤ نذدد ه درقسثة ددغ الورالددك درددتين يذدد هتونه عددن تقددق إبدددتدز سدددس دردددد ا ي وإخبددد ددد ا ي دخدددتط رةقش ددد مددد قادعدددت مسةشدددى و، دب دردددد ا دال قس عغ(Swihart et al,2017, p.43). يوشدددغ درقح ةددد درسددددتصغ إ ددد سًد ربهدددن درقب عددد دإلندددد ن غ إذ يقردددتا دأل دددتد دددد مسث دددةن مدتصةةن ع ة خذ ى درسدتي لGoffman 2:67 . أثش درقب ع يرش ت إرة دأل تد ع ة أنهن أشخ ص مدؤ ين مدن أ د ت ةد دالنق د ع دة دالنر د . دردت يتقدتوم إيرد ره منبددهن رآلخدددتقن. وىدددي ا سددد م عس يدددى درق ةةدددت ودرقودددتي دددغ دردددد ا ودرسدددت دالن قلددد وغ رددد سس درخردد وف ددت دسس دالنر دد . . يددقن د ددقسس ره ذددن عدد غ عشددتم ينددام درسسثدد ودليدد درددة دندده مبلصدد مددن درجسهدداس ومددن ثددن يددؤ نذدد ه درقسثة ددغ الورالددك درددتين يذدد هتونه عددن تقددق إبدددتدز سدددس دردددد ا ي وإخبددد ددد ا ي دخدددتط رةقش ددد مددد قادعدددت مسةشدددى و، دب دردددد ا دال قس عغ(Swihart et al,2017, p.43). ا دددتط زدددسةن درسادمددد درسدددد نتس رسس رجدددى دنر ددد . درجسهددداس ع دددة خذددد ى درسددددتي و لًددد ر اضدد دال قسدد عغ ودرس هددت درخدد س غ و تقلددى درقب عدد . ع ددة ددوة درسثدد يحدد ضدد درذت ى ع ة م هت دصقتد غ مهشغ زبل عن دس تد در درت دسغ؛ أمد غ درجسهداس إذ يشلد هد تد درسدت ر زد ع دة أنده دت دد ومحقدتا. يدتعغ ا سد م ل2:67 أنده مدن أ دد ارةت راس ث با ر تد يج ع ة درسدت درحبد ع دة سو دةن ث بدا مدن دأل د ألم دربذد غ درتي غ بترك يت ع ة خد سس هتد درس هت درخ س غ(Nesvadba,2017,p.5). قت يقدو عتغ دال د غ راس وس مقش قس ر تد م دكقذد ا درجسهداس مس امد قحددتط مردددتدقيى درسددؤ(Aerni,2014, p.10) . ودرشقيجدددى هدددغ ردداقت شددد ر رددداسس درسوشيددى درستعاعدددى. وعدد س مددد يخق ددي ددد ا درسسثدد دددغ خذدد ى درسددددتي درخ يددى إذ ال يندددام هش ص ى إرة دأل د . وإنده مند م مدت دم د مسدةن. الديلفي وإدراك الذاتSelfie and self- perception . و لً ر تسد ى درقدغ أ تدهDeniz(2012) رقحتيت درسبلقى بةن و و درقادىد دال قسد عغ وإ سد دردتد ردتط ددددى درسددددتدسص درث ناقددددى ؛ يذدددد س دررددددبلب رتقلددددى أو ددددمختط سددددس درسس امدددد درسقس لددددى ألشدددي در قدددغ يلدددتسونه وقارانهددد أهسيدددى رشدددد ى رهدددن دددغ صدددد هن ع دددة و ددد و درقادىددد .دال قس عغ هت درتسد ى أم راس درسذ س ةن ر تد غ درذ ن دال قس ليى ها إيج بغ 382 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ وقست درربلب م ب هن درذخريى درادق يى غ درذ ن دال قس ليى وم أ ت . وقسندشهن دددهارى مددن مذدد س ى درسس امدد درحتيتيددى ودرردداس درخ ىددى بهددن. وعشدد د ع يدده يددارغ مس ددن دررددددبلب دالهقسدددد غ ر ردددداسس رقكدددددام ةددددتس رهددددن وألنبدددددهن ؛ وقناندددددام ددددستد عشددددتم يسجددددد دىتق وهن راسهن(Seyfi & Arpaci,2016, p.146). وقست درربلب م ب هن درذخريى درادق يى غ درذ ن دال قس ليى وم أ ت . وقسندشهن دددهارى مددن مذدد س ى درسس امدد درحتيتيددى ودرردداس درخ ىددى بهددن. وعشدد د ع يدده يددارغ مس ددن دررددددبلب دالهقسدددد غ ر ردددداسس رقكدددددام ةددددتس رهددددن وألنبدددددهن ؛ وقناندددددام ددددستد عشددددتم يسجددددد دىتق وهن راسهن(Seyfi & Arpaci,2016, p.146). الديلفي وإستراتيجيات تقديم الذاتself presentation )(الذات على المدرح يت درقل ةت درش دت ر دتد سد رن دال قسد . درلدتم درسذدتقن إيدت ةش ا سد م دغ ق ده درسؤثت لتين درتد غ درحي س درةاميى ل2:67 يدقسس دربكتس درسج زقى ر تسدم رقلتين مبهاغ . تقت رقذنة درهاقى غ سد ى درقب ع دالند نغ يوشدددغ درقح ةددد درسددددتصغ إ ددد سًد ربهدددن درقب عددد دإلندددد ن غ إذ يقردددتا دأل دددتد دددد مسث دددةن مدتصةةن ع ة خذ ى درسدتي لGoffman 2:67 . أثش درقب ع يرش ت إرة دأل تد ع ة أنهن أشخ ص مدؤ ين مدن أ د ت ةد دالنق د ع دة دالنر د . دردت يتقدتوم إيرد ره منبددهن رآلخدددتقن. وىدددي ا سددد م عس يدددى درق ةةدددت ودرقودددتي دددغ دردددد ا ودرسدددت دالن قلددد وغ رددد سس درخردد وف ددت دسس دالنر دد . . الديلفي وإدراك الذاتSelfie and self- perception . دال دقب س مدن درسثد درسدت اس أعدبل رزدد درذددت ى هددا / هددغ قدد سس ع ددة دال ددقتخ ددغ بةالددى وسد دركددادريس وعددتغ دالضددرتدس إرة دس دتد درد درت دسغ أو درقردتا رتقلدى مقاقسدى د قس ليد ألنده رديس هشد صزداس أو مذ هتين(Nesvadba,2017,p.5-6). 383 383 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ و دغ نبددس دال جدد دداس سو وميددقتRoy Baumeister(1982,1987) ن تقددى ددددغ ع ددددن درددددشبس دال قسدددد عغ رقح ةدددد مددددتط ود دددد مددددن درددددد ا در ذددددت درسذددددن ر دددد هتس دال قس ليى وقذةت ومدقت أم لتين درتد ا يدقهتا مشه اىة سس درسس ام صدا دردتد أو سدس مدن ىدداسس دردتد ع دة د خدتقن . وقشدق عددن هدتد دردد ا نداعةن مددن درتود س أش:س إرةه وميدقةت 1 ارضاء الجميهرAudience amuse: وهت دده هددا مر لددى مدد يستضدده دربددت مددن درددتد مدد اقسدد د خددتقن و بزدديبل هن ت درحرا ع ة دعج بهن وم ي س حتيت م يقن عتضه مدن دردتد هدا درجسهداس ومدن ثددن ددينام مذددتو دد خقبلا درادقددي ودرجسهدداس. وقسددت هددتد درددتد شددن مددن أشددن د دسس دنر دد . د خددتقن عددن دربددت عددن تقددق مدد ي هددت رهددن مددن نبددده ردديجس هن ينانددام دنر عدد .محت د(Baumeister& Debra,1987, p.77). 2 بناء الذاتSelf- construction : هت ه مر لى م يلتمه دربت عن نبده مد درقرداس درشسداذ غ عدن ذد ده. أ أم قرد بق . ذد ه درست اس م ذد ه دربس يى. وقنام هتد درتد مذتو بقراس درست رتد ه درشساذ يى يسسددد دربدددت ع دددة لدددتين دردددتد ددددغ شددد ن درقادىددد دال قسددد عغ مدددن خدددبل درس بدددد درذخردديى ودرسددتون وىدداس درددد ة بغ الىددتق ودرسقدد سام ع ددة هددت درذدد ن هددن سث ددى درجسهاس درت يست دربت من ذد ه أم مهن أم ر دت د دسس دنر د عهن عشده أو ر دت إقشدد عهن وإقشددد . نبدددده منددده ربسددد ع ددة دررددداسس درقدددغ يدددا أم يندددام ع ةه .لعثسددد م3128 ص7 - 8 وقذةت ك منDiefenbach & Christoforakos(2017) أم ا دردة بغ هددا أصددت أشددن درددد ا درقددغ يلددتغ ةهدد دربددت ذد دده عددن تقددق شدد ن درقادىدد دال قسدد عغ وذردددك ألنهددد مقست ددد س صدددا دردددتد وأكثدددت أنددداد. دردددتود درست ردددى دددد ا درددددة بغ هدددا د ددد .. إسض درتد دو إ دسس دالنر أشد س د مدن دان وعةقسد مJones & Pittman(1982) دردة خسدس د دقتد يجي شددد وسى رقلدددتين دردددتد وهدددغل درقددداIngratiation و دددتوق دردددتدSelf- promotion ودرقتهةددددددد دو درقخاقدددددددIntimidation ودألمث دددددددىExemplifiction ودال دددددددقجتد Supplication ودضدددد ا مةتزعدددد خت(Merzbacher ندددداعةن ،خددددتقن مددددن دال ددددقتد يجي وهدددددددددددددغل عسدددددددددددددد دردددددددددددددش بسUnderstatement و ذددددددددددددددي دردددددددددددددتدSelf- disclosure لعثسددددددد م3128 ص8 - 9 . وقجددددددد دددددددDiefenbach & Christoforakos(2017) : م دردة بغ ير ح رقشبةت د قتد يجةقةن ل مشهس 384 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) توق درتد : درسقسث دسغ درسدت ألم يجسد د خدتقن يتونده منده شدخف ذو دتدسسل ذ د ..ماه ى ب س وذرك عن تقق إ ه س درست رلتسد ه ودنج زد ه و هةيه ود قخ س به ذن . من أشن درتع يى أو درقتوق ر تد كذدددي دردددتد : درشتددديس مدددن د دددقتد يجيى درقدددتوق ر دددتد دددبل يسسددد دربدددت ع دددة لدددتين ذد دددهل درس سسى وإنس نذي سس درسس ام عن نبده ومد يسندن أم يندام دتيتد رقسد ي مشهد كذن من أشن درواي ودالنبق يلعثس م3128 ص8 - 9 . 2 بناء الذاتSelf- construction : خرائص الذخرية المرتبظة بذدة نذر صهر الديلفي النرجدية سدددتا درر سدددى درخ مددددى مدددن دردددترة درقذخيردددغ ودإلصرددد وغ ربلضدددرتد درسل يدددى لDSM - 6 درددددت نذددددت ه درجس يددددى دألمتقنيددددى ر ردددد درشبدددددغ ل3124 درشت ددددديى منهدددد دإلعجدددددد ب رددددددتد درددددددت يقسةدددددد دددددد رسةا نحددددددا دأل كدددددد س درس يسددددددى ودرسادهدددددد درسقخة ددددددى ودال قستدضيى ودال اب دردت عغ س دً ع دة درشلدت ودرسبلقد درذخرديى درقدغ قسةد سذد عت دال ددقحل ودال ددق بل وق ددى درقسدد ي. يسةدد دأل ددتد درشت دددةام إرددة ردداقت و هدد ن ددت إيج بيى نحا درتد و دررب دراسدثيى درت يقزسن دردت ودرج ذبيدى درجددسيى و دردد رى أو درلداس(Emmons,1987; Mehdizadeh,2010) . يسندن رخرد وف درشت دديى أم بددت ددو هدداص سددس دأل ددتد رقلدد ل ىدداس دردددة بغ وقسةدد درشت دددةام ددغ دأل دد ص إرددة درل ددق ذدمم درس هدت درجددسغ(Bleske- Recheck, Remiker, & Baker, 2008) يحودام در س و قةن أنبدهن رتدودق د دقب دزقى و تد دى ربلنق د وإ دتدل دغ لدتيت د ذبةقهن دغ عةدام د خددتقن وعدددو هددتد درددد ا يسةدد درشت دددةام إرددة دال ددقسق . رقلدد ل ىدداس ددة بغ. وأم درتي غ بترك يدسح رهدن د رقحنن درك مد دغ ي يدى هداسهن دغ دررداس. عد سً مد كدام ىداس درددددة بغ ىددداس مجس دددى ذدددن ةدددت ددددسة إردددة درقدددمثةت ع دددة د خدددتقن . وهدددتد مددد أشددد س إريددده Weiser(2015 قزسن ىاس ة بغ ع سً درسش ىت دألكثدت وضداصً ر قبد خت ودالنقذد س درددتد غ. يسنددن أم ينددام درقلدد ل ىدداس دردددة بغ سث ددى مشدد وسس نبددديى تيددتس يسنددن مددن خبلرهدد رؤل ددتد درش ت دددةةن حلةددق ود سهددن دددهارى. عددبلوس ع ددة ذرددك يسةدد درشت دددةام إرددة درقس ددق ذددددن وةددددت رقلدددد ل ىدددداس دردددددة بغ ألندددده يسنددددشهن مددددن حسة هدددد ع ددددة مادقدددد درقادىدددد دال قس عغ درسخق بىSNSs مس يدسح رهدن تسضد سعودقهن دغ دالنخدتدل دغ د ا س قد درتد(Sorokowski et al., 2015, p.125). ذدددةت درتسد ددد دردددد لى دددمم هشددد رةددد ودضدددح ع دددة دررددد ى بدددةن دررددداسSelfies ودرشت ددديى. ع ددة ددوة درسثدد و ددت سد ددىFox and Rooney(2015 ددغ دروتديددى عبلقى قاقى بةن درشت ديى وعدت ىداس درددة بغ درقدغ دن نذدته . 2 بناء الذاتSelf- construction : وس د درتسد دى درقدغ قد غ بهد 385 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( مةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ Weiser(2015) ع دددة ثبلثدددى أ سددد تليدددى مدددن درشت دددديى سددد دددغ ذردددك درتيددد س/ دردددد رى ودال قستدضدددد دركوددددتط دو درسبددددتل ودال ددددقحل / دال ددددق بل ؛ و ددددت أم دددداد ت نذددددت ىدددداس .دردددددددددددة بغ ددددددددددد م مت رًددددددددددد ذددددددددددتس م سددددددددددد درتيددددددددددد س/ درددددددددددد رى ودال قستدضددددددددددديى درسبت دددددددددددى د ددقسسSorokowski et al (2015) أسعسددى ملدد ييس تليددى مددن در شت ددديى زددسشا دالكقبدد درددتد غSelf- sufficiency ودر ددتوسVanity ودرتيدد سLeadership و دد دإلعجدد بadmiration demand ؛ وو ددتود أم درردد ى بددةن درددتس دإل س ريددى ر شت ددديى ونذت ىاس دردة بغ درخ ىى بهدن ندا أقداط عشدت دردت اس مل سندى إلند لSorokowski et al., 2015, p.125). سلهك البحث عن االنتباهAttention-seeking behavior. وقسة ددام إرددة درقبكةددت ددغ دصقي دد هن 386 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ وسا دد هن درخ ىددى وعددتالً مددن مح ورددى هددن د خددتقن أو درقسدد ي مسهددن يقاقسددام أم يبهسهددن د ختوم. ذن خ ص ن تًد ألم دأل تد يسة ام إرة درقت ة ذن سويدغ ع ة أنبددهن عشدت .درقل ل ىاس دردة بغ لت يت هدتد دال جد ردد ا درسقست د صدا دردتد دردت ي هتونده بتأ درسجقسد مدؤختًد دغ درقسوةدت عدن درل دق مدن أم درقلد ل ىداس درددة بغ قدت يسقودت سدبلً أن نيًد ألنه يجسد درشد ص يهقسدام ثةدتًد س هدتهن دغ دررداس وقبذد ام دغ درقبكةدت دغ د خدتقن مدن صددارهن. وهددت دد مقدددلى مدد سد ددىFox and Rooney(2015) ذددمم هددت درحجددى صة أ هت أم دألشخ ص درتين يشذتوم ىاس دردة بغ نق غ ع ة درذد ن دال قس ليدى سة إرة إ ه س عتغ درقس ي(Charoensukmongkol,2016, p.3). عردتِّفا درذدساس راصدتس درشبدديى منده مددقاط مدن درسبلقد دال قس ليدى دألوريدى أقد مددن درسدددددقاط درستعدددداب دو درسقحلددددق و قزددددسن درسذدددد عت درسزددددرتعى ودرزددددةق ود سد ألو دددده درلردداس ددغ ددك درسبلقدد دال قس ليددى . يسددت درذددساس راصددتس خ ىدديى شخردديى مت رددى مدد درقلددد ل ىددداس د رددددة بغ ألم دأل حددد دردددد لى أ هدددت أنهددد سةددد إردددة دالس ددد ل لددداس نث دددى دالنذرى ع ة مادق درقادى دال قس عغSNS ال يس نذت درسس ام ومذ س قه لLou et al,2012, p . 217 . و ن دتًد ألم درقلد ل ىداس درددة بغ هدا نذد ل يددقهتا إردة س قد دركذدددي عدددن دردددتد ودرقادىددد دال قسددد عغ ع دددة مادقددد درقادىددد دال قسددد عغ ردددترك يسندددن دعق دد س مهسًددد ألورالددك دردددتين يسددد نام مددن درذدددساس راصددتس درشبدددديى. وعذدددن أكثددت حتيدددتدً قدددت يدددقسق دألشددخ ص مخددت ىدداس ددة بغ رقل ةدد درذددساس راصددتس ألم نذددت ىدداس ددة بغ ع ددة هددت درسادقددSNSs ودرحرددا ع ددة س و أ سدد مددن أىددتق وهن يددد عت ع ددة س قدد دركذددي عددن درددتد ودرقادىدد دال قسدد عغ. زددبل عددن ذرددك دد دد مددنSteinfield et al . ل3119 دددمم صددد درقلدددد ل ىددداس دردددددة بغ يسندددن أم يقدددددو دددغ ردددداقت دأل دددتد عبلقدددد رحيى م د ختقن ودرقغ يسنن أم خ ق درس قت من درذدساس راصدتس. سلهك البحث عن االنتباهAttention-seeking behavior. سلهك البحث عن االنتباهAttention seeking behavior. زددبل عددن دالس دد ل درددت قددت ينددام رددتط درشت ددديى مدد صدد درقلدد ل ىدداس دردددة بغ ً ج درتسد ى مم دس ه يسنن أم يلتغ ذن عةت م شت عن تقق ا يت دس وثيلً رشت ديى س غ ذرك ا در ح عن ًدالنق د ودردد ا درسقست د صدا دردتد . أوال يحت درد ا در ص عن دالنق ع س عشتم يقردتا دألشدخ ص رتقلدى هدتا إردة سد د ختقن يشقوهام إرةهن. و غ درادق يت ا در ح عن دالنق ردس درشت دديى ن دتًد ألم انده مت د دالهقسد غ يحلدق هدتا انده مت د درلداس و درقدمثةت لDeWall, Buffardi, Bonser, & Campbell 3122 ودققدددتيWeiserل3126 أم يندددام درشت ددددةام .مقحسدددةن رجددتب دنق دد د خددتقن وإعجدد بهن ر حبدد ع ددة و هدد ن ددتهن ر ددتد درسقزددخسى .ع ة درتعن من أم هتد درد ا يحت ع س عشت دأل ب إال أنه ويسغ أيزً عشت در د ر ةن و غ ثةت من درح ال يسنن أم يؤ دردسغ وسد دالنق إرة د ا مشدق لوذردك رسسد درجدد ر حرددا ع ددة دعقددتدا أو د ددتد مددن د خددتقن . ومدد ذرددك ددتم در حدد عددن دإلهقسدد غ درسبتل يسنن أم يؤثت ع ة دررحى درشبديى ر سس دأل تد وقت يقدو ذرك دغ دالنخدتدل دغ سس درد ا ي عةت درسش ى لAngstman & Rasmussen, 2011, p 2366 . ذددن عدد غ ن ددتًد ألم دألشددخ ص درددتين رددتيهن دد ا در حدد عددن دالنق دد يتعوددام دددغ دالنخددتدل ددغ دد ا يهددتا إرددة ددتب دنق دد د خددتقن ددتنهن يسة ددام إرددة صدد درقلدد ل ىدداس دردة بغ وقتس ام أم درتي غ بترك ع ة أنه عس مهن ألنه يسندشهن مدن نذدت هدت دررداس ع دة مادق درقادى دال قس عغSNS .رجتب ودرحرا ع ة س يل إيج بيى من أقتدنهن الدلهك المتمركز حهل الذاتSelf-centered behavior. هشدد دد ا ،خددت مددت رشت ددديى وقددت يددت أيزً دد حدد درقلدد ل ىدداس دردددة بغ هددا درددد ا درسقست دد صددا درددتد . عسامًدد يس ةدد دأل ددتد ذوو درددد ا درسقست دد صددا درددتد إرددة دالهقسدد غ أكثددت منبدددهن أكثددت مددن دهقسدد مهن دد ختقن. سلهك البحث عن االنتباهAttention-seeking behavior. يدقن عدن هدت درحجدى مددن خددبل حدد يذددة ت إرددة أم درسذدد س ةن درددتين ذ ددتود أنهددن نذددتود درس قددت مددن ىددا دردددة بغ سةد إردة درحردا ع دة س دى أقد دغ متيد ص دألربدىIntimacy (Steinfield et al, 2008, p . 549 . ثالثا: إجراءات البحث اوال: مجتمع البحث رلددت حلددق مجقسدد در حدد درحدد رغ ر ددى مسددى ددتد ر سدد غ درتسد ددغ ل3129 - 312: درتسد ى درر صيى ر قخررةن درس سغ ودإلندد نغ ومدن دبل درجشددةنل ذ داس- دند ودر د ر عت هنل2645: . ر و ر ى ا رلددت حلددق مجقسدد در حدد درحدد رغ ر ددى مسددى ددتد ر سدد غ درتسد ددغ ل3129 - 312: درتسد ى درر صيى ر قخررةن درس سغ ودإلندد نغ ومدن دبل درجشددةنل ذ داس- دند ودر د ر عت هنل2645: . ر و ر ى 387 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ ثانيا: عينة البحث عت هنل246 بادق ل75 ذ ت ول82 دنثة ن دخقي سهن رتقلى.عذادويى ثالثا: أداة البحث تط وشغ متي ص ص درقل ل ىاس درددة بغSelfie- liking scale وملد ييس دس درذخرديى درسقسث دى رشت دديى و دد ا در حد عدن دالنق د ومتيدد ص دردد ا درسقست د صددا درددددددددددددددددددتد ومتيدددددددددددددددددد ص دراصددددددددددددددددددتس درشبددددددددددددددددددديى درددددددددددددددددددت ددددددددددددددددددن دعددددددددددددددددددتد ه مددددددددددددددددددن قودددددددددددددددددد (Charoensukmongkol,2016, p.7) سدا ت سدى لدتد متيد ص صد درقلد ل ىداس دردة بغ ومتي ص س درذخريى من در ى دالنك ة قدى إردة در دى درستعيدى ودرقمكدت مدن دبلمى ىدي عى لدتد درستيد ص ر اقد ً وذردك عدن تقدق عتضده ع دة أصدت دالخقر ىدةةن دغ در ددى درستعيى. ست ذرك ن ت سى درستيد ص مدتس أخدتط مدن در دى درس تعيدى دردة در دى دالنك ة قدى وذردك ر قمكدددت مدددن أم درقت سدددى درستعيدددى ر ستيددد ص سندددس درسسشدددة دربس دددغ ودرحتيلدددغ ربلدددتد درستيددد ص دألىد غ ومدن ثدن دن درحردا ع دة ندددخى أوريدى مقت سدى. سلهك البحث عن االنتباهAttention-seeking behavior. يقدمري متيد ص صد درقلد ل ىدداس دردددة بغ مددنل7 لددتد ومتيدد ص درشت ددديى مددنل9 لددتد ودال ددقج ى و ددق ددتسا خس ددغ ل د بق س م د بق د بق درة صت مد ال أ بدق ال أ بدق س مد و سردة دردتس د غ ل6 - 5 - 4 - 3 - 2 أم متي ص دراصتس درشبديى ةقكام مدنل7 لدتد ومتيد ص د ا در حد عدن دالنق دد ل5 لدددتد ومتيددد ص ددد ا درقست ددد صددا دردددتد ل6 لدددتد ودال دددقج ى و دددق دددتسا خس غل دو س ع ر دصي ن ن سد دبتد و سرة درتس غ ل6 - 5 - 4 - 3 - 2 . صدق المقياس : : و ن درقحلق من خبل صدق المحكمين : عت درستي ص ع ة محنسةن غ مجد درس داغ درشبدديى و درقتعاقدى يدى درقحلق من وضاي دربلتد و ىي عقه و مبلوسقهد ر دت دردت وضدسا أل ده و ندا ند ى د ب درسحنسةن ل211 . و رجسي دربلتد% صااادق الءنااااء : و قدددت حلدددق مدددن خدددبل إيجددد مسددد مبل دس ددد ل س دددى دربلدددتس رتس دددى درك يدددى ر ستي ص و درجتو ل2 ول3 ول4 ول5 ول6 . اضح درشق و تو ل2 عبلقى دربلتس رتس ى درك يى رستي ص ص درقل ل ىاس دردة بغ د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.874 3 0.923 5 0.906 2 0.931 4 0.907 6 0.900 ثالثا: أداة البحث تط وشغ متي ص ص درقل ل ىاس درددة بغSelfie- liking scale وملد ييس دس درذخرديى درسقسث دى رشت دديى و دد ا در حد عدن دالنق د ومتيدد ص دردد ا درسقست د صددا درددددددددددددددددددتد ومتيدددددددددددددددددد ص دراصددددددددددددددددددتس درشبددددددددددددددددددديى درددددددددددددددددددت ددددددددددددددددددن دعددددددددددددددددددتد ه مددددددددددددددددددن قودددددددددددددددددد (Charoensukmongkol,2016, p.7) سدا ت سدى لدتد متيد ص صد درقلد ل ىداس دردة بغ ومتي ص س درذخريى من در ى دالنك ة قدى إردة در دى درستعيدى ودرقمكدت مدن دبلمى ىدي عى لدتد درستيد ص ر اقد ً وذردك عدن تقدق عتضده ع دة أصدت دالخقر ىدةةن دغ در ددى درستعيى. سلهك البحث عن االنتباهAttention-seeking behavior. ست ذرك ن ت سى درستيد ص مدتس أخدتط مدن در دى درس تعيدى دردة در دى دالنك ة قدى وذردك ر قمكدددت مدددن أم درقت سدددى درستعيدددى ر ستيددد ص سندددس درسسشدددة دربس دددغ ودرحتيلدددغ ربلدددتد درستيددد ص دألىد غ ومدن ثدن دن درحردا ع دة ندددخى أوريدى مقت سدى. يقدمري متيد ص صد درقلد ل ىدداس دردددة بغ مددنل7 لددتد ومتيدد ص درشت ددديى مددنل9 لددتد ودال ددقج ى و ددق ددتسا خس ددغ ل د بق س م د بق د بق درة صت مد ال أ بدق ال أ بدق س مد و سردة دردتس د غ ل6 - 5 - 4 - 3 - 2 أم متي ص دراصتس درشبديى ةقكام مدنل7 لدتد ومتيد ص د ا در حد عدن دالنق دد ل5 لدددتد ومتيددد ص ددد ا درقست ددد صددا دردددتد ل6 لدددتد ودال دددقج ى و دددق دددتسا خس غل دو س ع ر دصي ن ن سد دبتد و سرة درتس غ ل6 - 5 - 4 - 3 - 2 . صدق المقياس : : و ن درقحلق من خبل صدق المحكمين : عت درستي ص ع ة محنسةن غ مجد درس داغ درشبدديى و درقتعاقدى يدى درقحلق من وضاي دربلتد و ىي عقه و مبلوسقهد ر دت دردت وضدسا أل ده و ندا ند ى د ب درسحنسةن ل211 . و رجسي دربلتد% صااادق الءنااااء : و قدددت حلدددق مدددن خدددبل إيجددد مسددد مبل دس ددد ل س دددى دربلدددتس رتس دددى درك يدددى ر ستي ص و درجتو ل2 ول3 ول4 ول5 ول6 . سلهك البحث عن االنتباهAttention-seeking behavior. اضح درشق و تو ل2 عبلقى دربلتس رتس ى درك يى رستي ص ص درقل ل ىاس دردة بغ د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.874 3 0.923 5 0.906 2 0.931 4 0.907 6 0.900 388 388 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ 2222 / م1111 هـ (جمةل ل ب / جملدل1د )1 2 (ي ول) 2 22 /م1 تو ل3 عبلقى دربلتس رتس ى درك يى رستي ص درشت ديى د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.883 5 0.856 2 0.843 6 0.859 3 0.832 7 0.786 4 0.797 8 0.854 تو ل4 عبلقى دربلتس رتس ى درك يى رستي ص دراصتس درشبديى د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.829 4 0.835 2 0.815 5 0.751 3 0.772 6 0.778 تو ل5 عبلقى دربلتس رتس ى درك يى رستي ص ا در ح عن دالنق د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.865 3 0.822 2 0.839 4 0.872 تو ل6 عبلقى دربلتس رتس ى درك يى رستي ص ا درقست صا درتد د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.888 4 0.887 2 0.770 5 0.754 3 0.850 وقددت أ هددت نقدد و درقح ةدد دإلصردد وغ مددن إم مسدد مبل دالس دد ل س ددى دربلددتس رتس ددى درك يى ر ستي ص سيسه درى و عشت مدقاط الرى ل1.12 و درت يست مؤشتد ررت دروش. سلهك البحث عن االنتباهAttention-seeking behavior. درث : و ن درقحلق مشه رتقلى دال د دردتدخ غ لدربد تون ل ر سلد ييس إذ ودق درستيد ص ع ددة عةشددى قادمهدد ل41 ردد /س دد مسغ و أ ددبت درشقدد و عددن مسدد مبل درث دد د يددى و كس ماضح غ درجتو ل7 تو ل7 مس م درث درستي ص مس م درب تون ل ص درقل ل ىاس دردة بغ 0.816 درشت ديى 0.794 دراصتس درشبديى 0.798 ا در ح عن دالنق 0.829 ا درقست صا درتد 0.813 تو ل3 عبلقى دربلتس رتس ى درك يى رستي ص درشت ديى د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.883 5 0.856 2 0.843 6 0.859 3 0.832 7 0.786 4 0.797 8 0.854 تو ل4 عبلقى دربلتس رتس ى درك يى رستي ص دراصتس درشبديى د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.829 4 0.835 2 0.815 5 0.751 3 0.772 6 0.778 تو ل5 عبلقى دربلتس رتس ى درك يى رستي ص ا در ح عن دالنق د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.865 3 0.822 2 0.839 4 0.872 تو ل6 عبلقى دربلتس رتس ى درك يى رستي ص ا درقست صا درتد د د دربلتس مس م دالس ل د د دربلتس مس م دالس ل 1 0.888 4 0.887 2 0.770 5 0.754 3 0.850 وقددت أ هددت نقدد و درقح ةدد دإلصردد وغ مددن إم مسدد مبل دالس دد ل س ددى دربلددتس رتس ددى وقددت أ هددت نقدد و درقح ةدد دإلصردد وغ مددن إم مسدد مبل دالس دد ل س ددى دربلددتس رتس ددى درك يى ر ستي ص سيسه درى و عشت مدقاط الرى ل1.12 و درت يست مؤشتد ررت دروش. سلهك البحث عن االنتباهAttention-seeking behavior. وال: التعرف على مدتهى حب التقاط صهر الديلفي لدى طلبة جامعة بغداد إذ دددن صدددد ب درسقا دددر درحددددد بيى ودالنحتد ددد درس ي سقدددى أل ددددتد عةشدددى در حددد ع ددددة متيددد ص صددد درقلددد ل ىددداس درددددة بغ ور قسدددتا ع دددة الردددى دربدددت دال صرددد وغ بدددةن درسقا ددد درحد بغ ودربتضغ ن د قسس دالخق س درق وغt-test ً رسةشى ودصدتس ود زدح دم هشد توقد درى دصر وي وذرك عن تقق مل سندى درتدين درق ويدى درسحدداعى رجتوريدى در ر دىل2.:7 عشدت مدقاطل1.16 و س ى صتقىل245 ودرجتو ل8 يوةن ذرك تو ل8 دالخق س درق وغ رسةشى ودصتس ر قستا ع ة دربت بةن درا رةن درحد بغ ودربتضغ درستي ص درسقا دالنحتدا درس ي س س ى درحتقى درا دربتضغ درق ويى درسحداعى درق ويى درجتوريى درتالرى ص درقل ل ىاس دردة بغ 23.192 6.924 134 18 8.713 1.96 د و ذةت درشقيجى أعبل دم رتط عةشى در ح مدقاط ع من ص درقل ل ىاس دردة بغ درهتا درث نغ: درقستا ع ة مدقاط ص درقل ل ىاس دردة بغ ردتط دى مسدى دتد ع دة و دددق مق ةدددت درجدددشس . إذ دددن صدددد ب درسقا دددر درحدددد بيى ودالنحتد ددد درس ي سقدددى ردددتس درسةشى ود قختدا درتيسى درق ويى رسةشقةن مدقل قةن ودرجتو ل9 ي وةن ذرك تو ل9 دالخق س درق وغ رسةشقةن مدقل قةن ر قستا ع ة دربتو غ ص درقل ل ىاس دردة بغ و ق مق ةت درجشس درستي ص درجشس درست درسقا دالنحتدا درس ي س س ى درحتقى درق ويى درسحداعى درق ويى درجتوريى درتالرى ص درقل ل ىاس دردة بغ دن 71 26.0141 4.274 133 5.505 1.96 د ذ اس 64 20.0625 7.921 و ذةت درشقيجى درة أم هش توق ً ذد الرى دغ صد درقلد ل ىداس درددة بغ ردتط دى . مسى تد يس ط درة مق ةت درجشسلذ اس ودن ورر رح دإلن درهتا درث ر : درقستا ع ة درسبلقى بةن ص درقل ل ىاس دردة بغ وعسس س درذخ رديى . سلهك البحث عن االنتباهAttention-seeking behavior. لدرشت دددديى ودراصدددتس درشبدددديى و ددد ا در حددد عدددن دالنق ددد و ددد ا درقست ددد صدددا دردددتد ورقحلةددق هددتد درهددتا ددن د ددقختدا مس مدد دس دد ل بةت ددام بددةن درددتس درك يددى درقددغ صردد ع ةهدد أ ددتد عةشددى در حددد ع ددة متيدد ص صددد درقلدد ل ىدداس درددددة بغ و س دد هن ع ددة ملددد ييس دمجةج م عرض النتائج ومناقذتيا أ. وال: التعرف على مدتهى حب التقاط صهر الديلفي لدى طلبة جامعة بغداد إذ دددن صدددد ب درسقا دددر درحددددد بيى ودالنحتد ددد درس ي سقدددى أل ددددتد عةشدددى در حددد ع ددددة متيددد ص صددد درقلددد ل ىددداس درددددة بغ ور قسدددتا ع دددة الردددى دربدددت دال صرددد وغ بدددةن درسقا ددد درحد بغ ودربتضغ ن د قسس دالخق س درق وغt-test ً رسةشى ودصدتس ود زدح دم هشد توقد درى دصر وي وذرك عن تقق مل سندى درتدين درق ويدى درسحدداعى رجتوريدى در ر دىل2.:7 عشدت مدقاطل1.16 و س ى صتقىل245 ودرجتو ل8 يوةن ذرك تو ل8 دالخق س درق وغ رسةشى ودصتس ر قستا ع ة دربت بةن درا رةن درحد بغ ودربتضغ درستي ص درسقا دالنحتدا درس ي س س ى درحتقى درا دربتضغ درق ويى درسحداعى درق ويى درجتوريى درتالرى ص درقل ل ىاس دردة بغ 23.192 6.924 134 18 8.713 1.96 د و ذةت درشقيجى أعبل دم رتط عةشى در ح مدقاط ع من ص درقل ل ىاس دردة بغ درهتا درث نغ: درقستا ع ة مدقاط ص درقل ل ىاس دردة بغ ردتط دى مسدى دتد ع دة و دددق مق ةدددت درجدددشس . إذ دددن صدددد ب درسقا دددر درحدددد بيى ودالنحتد ددد درس ي سقدددى ردددتس درسةشى ود قختدا درتيسى درق ويى رسةشقةن مدقل قةن ودرجتو ل9 ي وةن ذرك تو ل9 دالخق س درق وغ رسةشقةن مدقل قةن ر قستا ع ة دربتو غ ص درقل ل ىاس دردة بغ و ق مق ةت درجشس درستي ص درجشس درست درسقا دالنحتدا درس ي س س ى درحتقى درق ويى درسحداعى درق ويى درجتوريى درتالرى ص درقل ل ىاس دردة بغ دن 71 26.0141 4.274 133 5.505 1.96 د ذ اس 64 20.0625 7.921 و ذةت درشقيجى درة أم هش توق ً ذد الرى دغ صد درقلد ل ىداس درددة بغ ردتط دى . سلهك البحث عن االنتباهAttention-seeking behavior. درث : و ن درقحلق مشه رتقلى دال د دردتدخ غ لدربد تون ل ر سلد ييس إذ ودق درستيد ص ع ددة عةشددى قادمهدد ل41 ردد /س دد مسغ و أ ددبت درشقدد و عددن مسدد مبل درث دد د يددى و كس ماضح غ درجتو ل7 تو ل7 مس م درث درستي ص مس م درب تون ل ص درقل ل ىاس دردة بغ 0.816 درشت ديى 0.794 دراصتس درشبديى 0.798 ا در ح عن دالنق 0.829 ا درقست صا درتد 0.813 وقددت أ هددت نقدد و درقح ةدد دإلصردد وغ مددن إم مسدد مبل دالس دد ل س ددى دربلددتس رتس ددى درك يى ر ستي ص سيسه درى و عشت مدقاط الرى ل1.12 و درت يست مؤشتد ررت دروش. درث : و ن درقحلق مشه رتقلى دال د دردتدخ غ لدربد تون ل ر سلد ييس إذ ودق درستيد ص ع ددة عةشددى قادمهدد ل41 ردد /س دد مسغ و أ ددبت درشقدد و عددن مسدد مبل درث دد د يددى و كس ماضح غ درجتو ل7 تو ل7 مس م درث درستي ص مس م درب تون ل ص درقل ل ىاس دردة بغ 0.816 درشت ديى 0.794 دراصتس درشبديى 0.798 ا در ح عن دالنق 0.829 ا درقست صا درتد 0.813 2222 / م1111 هـ ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) ( جمةل الداب / اجملدل1 ) العدد112 (أيلول) 2 22 / م1111 هـ عرض النتائج ومناقذتيا أ. سلهك البحث عن االنتباهAttention-seeking behavior. لدرشت دددديى ودراصدددتس درشبدددديى و ددد ا در حددد عدددن دالنق ددد و ددد ا درقست ددد صدددا دردددتد ورقحلةددق هددتد درهددتا ددن د ددقختدا مس مدد دس دد ل بةت ددام بددةن درددتس درك يددى درقددغ صردد ع ةهدد أ ددتد عةشددى در حددد ع ددة متيدد ص صددد درقلدد ل ىدداس درددددة بغ و س دد هن ع ددة ملددد ييس 38: ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ دس درذخردديىل درشت دديى ودراصددتس درشبددديى و د ا در حدد عددن دالنق د و دد ا درقست دد صا درتد ودرجتو ل21 . ياضح ذرك تو ل: دالس ل در دي بةن ص درقل ل ىاس دردة بغ وعسس س درذخريى ل درشت ديى ودراصتس درشبديى و ا در ح عن دال نق و ا درقست صا درتد س درذخري ى ص درقل ل ىاس دردة بغ درشت ديى دراصتس درشبديى ا در ح عن دالنق ا درقست صا درتد مس م دالس ل 0.824 0.370 0.317 0.266 ب أب أب ومن خبل درجتو أعبل يقوةن أم هش عبلقى دس يى دردى دصرد وي بدةن صد درق لد ل ىداس دردددة بغ و ددس درذخردديى دل رشت ددديى ودر اصددتس درشبددديى و دد ا در حدد عددن دالنق دد و دد ا درقست دد صددا درددتد ورسست ددى ع يددى دد ددسى مددن ددس درذخردديى درسددت اسس ددغ ص درقل ل ىاس دردة بغ وإ تد ح ة دالنحتدس درسقست ودرجتو ل21 . ياضح ذرك تو ل21 ح ة دالنحتدس ر قستا ع ة درتالرى دإلصر ويى وإ ه غ س درذخريى غ ص درقل ل ىاس دردة بغ مرتس درق ين s.of.v .مجسا درستعس s.of.s س ى درحتقى D.F مقا درستعس M.S درتيسى درب ويى F درتالرى sig دالنحتدس 4443.66 4 1110.91 72.890 0.05 درسق لغ 1981.31 130 15.241 درك غ 6424.99 134 ببا من خبل درجتو أعبل يقو ةن أم س درذخريى سجساعه دهن بتالرى دصر ويى غ ص درقل ل ىاس دردة بغ رتط ى مسدى دتد دذ ب دا درتيسدى درب ويدى درسحدداعى رقح ةد ين دالنحتدسل83.9: وهغ أع ة من درب ويى درجتوريى در ر دىل3.48 عشدت مددقاطل1.16 و س ى صتقىل24135 . ور قستا ع دة دإل ده غ درشددوغ ركد دسى مدن دس درذخرديى دغ ص درقل ل ىاس دردة بغ لت ن د قختدا مس م لبةقBeta ودرجتو ل22 . سلهك البحث عن االنتباهAttention-seeking behavior. مسى تد يس ط درة مق ةت درجشسلذ اس ودن ورر رح دإلن درهتا درث ر : درقستا ع ة درسبلقى بةن ص درقل ل ىاس دردة بغ وعسس س درذخ رديى . لدرشت دددديى ودراصدددتس درشبدددديى و ددد ا در حددد عدددن دالنق ددد و ددد ا درقست ددد صدددا دردددتد ورقحلةددق هددتد درهددتا ددن د ددقختدا مس مدد دس دد ل بةت ددام بددةن درددتس درك يددى درقددغ صردد ع ةهدد أ ددتد عةشددى در حددد ع ددة متيدد ص صددد درقلدد ل ىدداس درددددة بغ و س دد هن ع ددة ملددد ييس عرض النتائج ومناقذتيا أ. وال: التعرف على مدتهى حب التقاط صهر الديلفي لدى طلبة جامعة بغداد إذ دددن صدددد ب درسقا دددر درحددددد بيى ودالنحتد ددد درس ي سقدددى أل ددددتد عةشدددى در حددد ع ددددة متيددد ص صددد درقلددد ل ىددداس درددددة بغ ور قسدددتا ع دددة الردددى دربدددت دال صرددد وغ بدددةن درسقا ددد درحد بغ ودربتضغ ن د قسس دالخق س درق وغt-test ً رسةشى ودصدتس ود زدح دم هشد توقد درى دصر وي وذرك عن تقق مل سندى درتدين درق ويدى درسحدداعى رجتوريدى در ر دىل2.:7 عشدت مدقاطل1.16 و س ى صتقىل245 ودرجتو ل8 يوةن ذرك تو ل8 دالخق س درق وغ رسةشى ودصتس ر قستا ع ة دربت بةن درا رةن درحد بغ ودربتضغ درستي ص درسقا دالنحتدا درس ي س س ى درحتقى درا دربتضغ درق ويى درسحداعى درق ويى درجتوريى درتالرى ص درقل ل ىاس دردة بغ 23.192 6.924 134 18 8.713 1.96 د ب و ذةت درشقيجى درة أم هش توق ً ذد الرى دغ صد درقلد ل ىداس درددة بغ ردتط دى . مسى تد يس ط درة مق ةت درجشسلذ اس ودن ورر رح دإلن درهتا درث ر : درقستا ع ة درسبلقى بةن ص درقل ل ىاس دردة بغ وعسس س درذخ رديى . سلهك البحث عن االنتباهAttention-seeking behavior. ياضح ذرك و من خبل درجتو ل22 ي قوةن أم ر شت ديى و ا درقست صا درتد ره د ه غ د دصردد وي ددغ صدد درقلدد ل ىدداس دردددة بغ ودم نددد ى إ دده غ درشت ددديى دد م أع ددة مددن درقست دد صددا درددتد إذ ب ددا مس مدد بةقدد ر شت ددديىل1.94 ورددد ا درقست دد صددا درددتد ل1.28 وعشددت تعيدد هددت درتددين نجددت أم درشت دددديى و دد ل1.7: ودرقست دد صددا درددتد ل1.14 . مسددد 391 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ يذددةت إرددة أمل1.83 درق ةددت ددغ صدد درقلدد ل ىدداس دردددة بغ يت دد إرددة درشت ددديى و دد ا . درقست صا درتد تو ل22 مس م بةق رئل ه غ درشدوغ و الرقه دالصر ويى دردسى دالس ل درسقست مس م درقحتيت قيسى بةق T درسحداعى درتالرى درشت ديى 0.832 0.962 0.829 15.240 د دراصتس درشبديى 0.115 1.512 عةت د ا در ح عن دالنق 0.035 0.362 عةت د ا درقست صا درتد 0.168 1.917 د :مناقذة النتائج رتط رشد ى رشقيجى درهتا دألو درسقسث رتس ى درس ريى رح درقل ل ىاس دردة ب ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ يذددةت إرددة أمل1.83 درق ةددت ددغ صدد درقلدد ل ىدداس دردددة بغ يت دد إرددة درشت ددديى و دد ا . درقست صا درتد ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) . سلهك البحث عن االنتباهAttention-seeking behavior. درقست صا درتد تو ل22 مس م بةق رئل ه غ درشدوغ و الرقه دالصر ويى دردسى دالس ل درسقست مس م درقحتيت قيسى بةق T درسحداعى درتالرى درشت ديى 0.832 0.962 0.829 15.240 د دراصتس درشبديى 0.115 1.512 عةت د ا در ح عن دالنق 0.035 0.362 عةت د ا درقست صا درتد 0.168 1.917 د :مناقذة النتائج رشددد ى رشقيجددى درهددتا دألو درسقسثدد رتس ددى درس ريددى رحدد درقلدد ل ىدداس دردددة بغ رددتط عةشى در ح يسنن بدةت درة دنه دىد حا لشيد دال رد ال درسخق بدى ذدن أ د رجسيد شددتدوح درسجقسدد سدد ةهدد الددى درذدد ب دداد ددغ درسددتدسص ودربرددا درتسد دديى أو درج م يددى وأمدد كن درسسدد وصقددة ددغ درسشدد ز ..در ودرقددغ سدد وسد صةاقدد ددغ كدداقن درش دد غ درثلدد غ وع يدددده ةددددت عس يدددد درقلةددددين درددددتد غ رؤل ددددتد أنبدددددهن و ردددداس سدددد ر ق ةددددتد ددددغ لشيدددد دال رددد ال درسخق بدددى صةددد بدددتأ دأل دددتد بدددتم ذود هدددن سذددد عت دإلعجددد ب درسلتمدددى مدددن قوددد درجسي ودردة درتا دى دقث سس دهقسد غ دالخدتقن وربدا دالنق د ودرتا دى دغ س د درسن ندى درشبدديى ودرذساس النقس درة س عى م ودرقس سا دال قس عغ من خدبل مذد س قهن ررداس درددة بغ درقددغ ي قلرانهدد ألنبدددهن عوددت و دد و درقادىدد دال قسدد عغ ودرقددغ دىدد ح ا د دس مهسددى ر يددى رقسنةددددشهن مددددن درقسوةددددت عددددن أنبددددددهن وهددددتد مدددد أكددددت ع يددددده درستيددددت مددددن درتسد دددد مشهددددد ل Hess,2015, p.1626; Roberts & koliska,2017,p.10 ; Moon et al,2016, p.23; Gabriel, 2014, p . 215 . أم نقيجى درهتا درث نغ هتد م يسنن بدةت و ل رس أش س إريه Allbery(2015) إرة أم سيدد دإلندد يدددسةفنا هددتد ر حرددا ع ددة ىدداسس مشدددلى كددام ةهدد درددتد درجسة ددى ودرجتد ى ودرسقادزنى م سس درقمكةتد ع ة دالىد رى وهدتد يسندن حتيلده عدن تقدق درقلد ل ىدداس دردددة بغ مددن خددبل إبددتدز ادندد مسةشددى مددن درس هددت درخدد س غ ودخبدد ادندد أخددتط 392 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) إ يسندن رخرد وف درشت دديى أم بددت دو هداص سدس دأل دتد رقلد ل ىداس درددة بغ وقسة درشت دةام غ دأل د ص إردة درل دق ذدمم درس هدت درجددسغ وعددو هدتد دردد ا يسةد درش ت دددةام إرددة دال ددقسق . سلهك البحث عن االنتباهAttention-seeking behavior. رقلدد ل ىدداس ددة بغ يحوددام در دد س و دد قةن أنبدددهن رددت د ددددقب دزقى و تد ددددى ربلنق دددد وإ ددددتدل ددددغ لددددتيت دددد ذبةقهن ددددغ عةددددام د خددددتقن و دددددو هددددتد درد ا يسة درشت دةام إردة دال دقسق . رقلد ل ىداس دة بغ( Halpern et al.2016, p. 99; kaur, maheshwari and Sharma, 2018, p.56; Arpaci, 2018. P.71). أم رشد ى ر د ا درقست صا درتد ذن خ ص ن تًد ألم دأل دتد يسة دام إردة درقت ةدد ذددن سويدددغ ع ددة أنبدددهن عشدددت درقلدد ل ىدداس دردددة بغ يسةدد دأل ددتد ذو دردددد ا درسقست ددد صدددا دردددتد إردددة دالهقسددد غ منبددددهن أكثدددت مدددن دهقسددد مه ن ددد ختقن. وقسة دددام إردددة درقبكةت غ دصقي هن وسا هن درخ ىى وعتالً من مح ورى هن د ختقن أو درقس ي مسهن يقاقسام أم يبهسهن د ختوم. وهتد م أكت ه درتسد ى درقغ أ تقا من قو من لSeyfi & Arpaci 2016 p 143) ● ع ة مخقرغ دررحى درشبديى ودإلسش د رشبدغ غ درج مسى درسس ع ة د قكذ ا ي يى د قختدغ درذ ب رراس دردة بغ وو و درقادى دال قس عغ ودرقادى درتقسغ ج مهن من لةدين درردحى درشبدديى رهدن و دقح در د ب ر سش قذد صدا لدتيت دردتد ىداسس درجددن .ودرسبلق دررحيى ودردبلمى درتقسيى ● يجد حتيددت دألمد كن ودألو قدد درقدغ خ ددا مددن دال دقختدغ رددؤل ود درقكش ا يدى ددغ درسشدد ز . ومتدك درقس ن ودألم كن درس مى وم إرة ذرك رقل ة مثةته ودالعقس ع يه اأ ● يج بت درسحد وال ع دة درسددقاقةن درذخردغ ودرسجقسسدغ رقل ةد دال دقختدغ دإلركقتوندغ . درسبتل ال قس س ادزم درحي س ودرسس درسلقتص ● يج بت درسحد وال ع دة درسددقاقةن درذخردغ ودرسجقسسدغ رقل ةد دال دقختدغ دإلركقتوندغ . درسبتل ال قس س ادزم درحي س ودرسس درسلقتص ● إ تد ح ة دالنحتدس درسقست رسست ى مثةت مق ةتد أختط ع ة د ا صد درقلد ل ىداس . دردة بغ ● درسبلقدددى بدددةن صددد درقلددد ل ىددداس درددددة بغ ومق ةدددتد أخدددتط لدرح ددد درشبدددديى وأهدددتدا . درحي س ودردس درثل هيى ودرسس يةت دال قس ليى ● درسبلقدددى بدددةن صددد درقلددد ل ىددداس درددددة بغ ومق ةدددتد أخدددتط لدرح ددد درشبدددديى وأهدددتدا . سلهك البحث عن االنتباهAttention-seeking behavior. درحي س ودردس درثل هيى ودرسس يةت دال قس ليى أا ● إ دددتد سد دددى مس ث دددى قشددد و عةشددد أخدددتط مخق بدددى مددد دألخدددت بش دددت دالعق ددد س رسق ةدددتد . يساعتدهيىل درسست ودرجشس ودرسدقاط درقس يسغ ودرسهشى..در ● إ دددتد سد دددى مس ث دددى قشددد و عةشددد أخدددتط مخق بدددى مددد دألخدددت بش دددت دالعق ددد س رسق ةدددتد . يساعتدهيىل درسست ودرجشس ودرسدقاط درقس يسغ ودرسهشى..در ● إ دددتد سد دددى مس ث دددى قشددد و عةشددد أخدددتط مخق بدددى مددد دألخدددت بش دددت دالعق ددد س رسق ةدددتد . يساعتدهيىل درسست ودرجشس ودرسدقاط درقس يسغ ودرسهشى..در 393 393 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ :المرادر 1. (.الشربيني، لطفي2001 ). موسوعة شرح المصطلحات النفسية انجليزي- عربزي، اا النضةزة البربيزة .للطباعة والنشر، بيروت وسوعة شرح المصطلحات النفسية انجليزي- عربزي، اا النضةزة البربيزة (حمدأأ )إ 1. Aerni, M. (2014). The passionate „sharing‟ of creative women: A Study of self- portrayal on Facebook and Instagram. (Masters Thesis, Stockholm University).Retrieved from http://www.divaportal.org/smash/get/diva2:724725/FULLTEXT01.pdf http://www.divaportal.org/smash/get/diva2:724725/FULLTEXT01.pdf 2. Angstman, K. B., & Rasmussen, N. H. (2011). Personality disorders: Review and clinical application in daily practice. American Family Physician, 8, 1253– 1260. 3. Arpaci, Ibrahim.(2018). The moderating effect of gender in the relationship between narcissism and selfie-posting behavior, Personality and Individual Differences. 134 ,71–74 4. Balakrishnan, J.,& Griffiths,MD(2018). An Exploratory Study of “Selfitis” and the Development of the Selfitis Behavior Scale. International Journal of Mental Health and Addiction, 16,722-736. 5. Baumeister Roy F., Hutten, Debra G. (1987). Self-presentation theory: self- construction and audience pleasing. In Goethals, George and Mullen, Brian (ed.) theories of group behavior. New york: springer –Verlag Inc. pp. 71-87. g p y p g g pp 6. Borkenau, P., Brecke, S., Möttig, C., & Paelecke, M. (2009). Extraversion is accurately perceived after a 50-ms exposure to a face. Journal of Research in Personality, 43(4), 703-706. y 7. Charoensukmongkol, Peerayuth. (2016). Exploring personal characteristics associated with selfie-liking. Cyberpsychology. 10(2).1-13. 8. El-Sherbiny, Lotfi (2001), Encyclopedia of Explaining Arabic Psychological Terms, Dar Al-Nahda Al-Arabiya for Printing and Publishing, Beirut. 9. Flaherty, G. T., & Choi, J. (2016). The „selfie‟phenomenon: reducing the risk of harm while using smartphones during international travel. Journal of Travel Medicine, 23(2), 1–3. 10. Fox, J., & Rooney, M. C. (2015). The Dark Triad and trait self objectification as predictors of men‟s use and self- presentation behaviors on social networking sites. سلهك البحث عن االنتباهAttention-seeking behavior. Personality and Individual Differences, 76, 161-165. y 11. Gabriel, Fleur. (2014). Sexting, Selfies and Self-Harm: Young People, Social Media and the Performance of Self-Development. Media International Australia incorporating Culture and Policy. 151. 104-112. 12. Halpern, D., Valenzuela, S., & Katz, J. E. (2016). “Selfie-ists” or “Narci- selfiers”?: A cross-lagged panel analysis of selfie taking and narcissism. Personality and Individual Differences, 97, 98–101. y 13. Hess, A. (2015). Selfies\| the selfie assemblage. International Journal of Communication, 9, 1629–1646. 14. Kaur, Sukhdeep & Maheshwari, Dr S K & Preksha, Dr. (2018). Narcissistic Personality and Selfie Taking Behavior among College Students. International Journal of medical and health research. 3. 56-60. 15. Kwon, Y. J. & Kwon, K. N. (2011). Self in Selca, Self-Portrait Photography, as a Model, Photographer, And Consumer. Building Connections, 39, 660-661. 394 ( جمةل الآداب / اجملدل1 ) العدد112 (أ يلول) 2222 / م1111 هـ 16. Ma, J. W., Yang, Y., & Wilson, J. A. (2017). A window to the ideal self: a study of UK Twitter and Chinese Sina Weibo selfie-takers and the implications for marketers. Journal of Business Research, 74, 139–142. 17. Moon, J. H., Lee, E., Lee, J. A., Choi, T. R., & Sung, Y. (2016). The role of narcissism in self-promotion on Instagram. Personality and Individual Differences, 101, 22–25. 18. Naumann, L. P., Vazire, S., Rentfrow, P. J., & Gosling, S. D. (2009). Personality judgments based on physical appearance. Personality and Social Psychology Bulletin, 35(12), 1661-1671 19. Nestler, S., Egloff, B., Küfner, A. C., & Back, M. D. (2012). An integrative lens model approach to bias and accuracy in human inferences: Hindsight effects and knowledge updating in personality judgments. Journal of personality and social psychology, 103(4), 689-717. 20. Nesvadba, Noé(2017) The presentation of self in the digital world: Goffman and the study of women‟s self-portrayal on istagram through selfies, A Master‟s Dissertation Submitted in partial fulfillment of the requirements for the award of M.A in Digital Media and Society, Loughborough University, Retrieved from https://www.academia.edu/36582334/ 21. Qiu, L., Lu, J., Yang, S., Qu, W., & Zhu, T. (2015). What does your selfie say about you? Computers in Human Behavior, 52, 443–449. 22. Roberts, J., & Koliska, M. (2017). Comparing the use of space in selfies on Chinese Weibo and Twitter. Global Media and China, 2(2), 1–16. 23. Sorokowski, P., Sorokowska, A., Oleszkiewicz, A., Frackowiak, T., Huk, A., & Pisanski, K. (2015). سلهك البحث عن االنتباهAttention-seeking behavior. Selfie posting behaviors are associated with narcissism among men. Personality and Individual Differences, 85, 123–127. 24. Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self-esteem, and use of online social network sites: A longitudinal analysis. Journal of Applied Developmental Psychology, 29, 434–445. p y gy 25. Swihart,JD., Polack,C., Gilbert, E., Essa,I. (2017). Selfie-Presentation in Everyday Life: A Large-Scale Characterization of Selfie Contexts on Instagram, Proceedings of the Eleventh International AAAI Conference on Web and Social Media,42-51 26. Tandoc, E. C., Jr., Ferrucci, P., & Duffy, M. (2015). Facebook use, envy, and depression among college students: Is facebooking depressing? Computers in Human Behavior, 43, 139–146. 27. Vazire, S., Naumann, L. P., Rentfrow, P. J., & Gosling, S. D. (2008). Portrait of a narcissist: Manifestations of narcissism in physical appearance. Journal of Research in Personality, 42(6), 1439-1447. 28. Weiser, E. B. (2015). Me: Narcissism and its facets as predictors of selfie- posting frequency. Personality and Individual Differences, 86, 477–481. 29. Weiten & Margaret (1994) Psychology Applied Modern Life, Cole PUBLISHING Company Pacific Grove, California. 395
https://openalex.org/W3198777459	https://europepmc.org/articles/pmc8464812?pdf=render	English	null	Regression Modeling of the Antioxidant-to-Nephroprotective Relation Shows the Pivotal Role of Oxidative Stress in Cisplatin Nephrotoxicity	Antioxidants	2,021	cc-by	13,546	7 Group of Biomedical Research on Critical Care (BioCritic), Valladolid University Hospital, 47003 Valladolid, Spain Citation: Casanova, A.G.; Harvat, M.; Vicente-Vicente, L.; Pellicer-Valero, Ó.J.; Morales, A.I.; López-Hernández, F.J.; Martín-Guerrero, J.D. Regression Modeling of the Antioxidant-to-Nephroprotective Relation Shows the Pivotal Role of Oxidative Stress in Cisplatin Nephrotoxicity. Antioxidants 2021, 10, 1355. https://doi.org/10.3390/ antiox10091355 p 8 Disease and Theranostic Modelling (DisMOD) Working Group, IBSAL, 37007 Salamanca, Spain p 8 Disease and Theranostic Modelling (DisMOD) Working Group, IBSAL, 37007 Salamanca, Spain * Correspondence: jose d martin@uves; Tel : +34 963 544 022 8 Disease and Theranostic Modelling (DisMOD) Working Group, IBSAL, 37007 Salamanca, Spain * Correspondence: jose.d.martin@uv.es; Tel.: +34-963-544-022 † These authors share ﬁrst authorship. ‡ These authors share senior authorship. Abstract: The clinical utility of the chemotherapeutic drug cisplatin is signiﬁcantly limited by its nephrotoxicity, which is characterized by electrolytic disorders, glomerular ﬁltration rate decline, and azotemia. These alterations are consequences of a primary tubulopathy causing injury to proximal and distal epithelial cells, and thus tubular dysfunction. Oxidative stress plays a role in cisplatin nephrotoxicity and cytotoxicity, but its relative contribution to overall toxicity remains unknown. We studied the relation between the degree of oxidative reduction (provided by antioxidant treatment) and the extent of nephrotoxicity amelioration (i.e., nephroprotection) by means of a regression analysis of studies in animal models. Our results indicate that a linear relation exists between these two parameters, and that this relation very nearly crosses the value of maximal nephroprotection at maximal antioxidant effect, suggesting that oxidative stress seems to be a pivotal and mandatory mechanism of cisplatin nephrotoxicity, and, hence, an interesting, rationale-based target for clinical use. Our model also serves to identify antioxidants with enhanced effectiveness by comparing their actual nephroprotective power with that predicted by their antioxidant effect. Among those, this study identiﬁed nanoceria, erythropoietin, and maltol as highly effective candidates affording more nephroprotection than expected from their antioxidant effect for prospective clinical development. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Keywords: cisplatin; nephrotoxicity; prevention; antioxidants; preclinical; linear ﬁt Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Regression Modeling of the Antioxidant-to-Nephroprotective Relation Shows the Pivotal Role of Oxidative Stress in Cisplatin Nephrotoxicity Alfredo G. Casanova 1,2,3,4,5,† , Mykola Harvat 6,†, Laura Vicente-Vicente 1,4, Óscar J. Pellicer-Valero 6 , Ana I. Morales 1,2,3,5,7,8 , Francisco J. López-Hernández 1,2,4,5,7,8,‡ and José D. Martín-Guerrero 6,*,‡ 1 Institute of Biomedical Research of Salamanca (IBSAL), 37007 Salamanca, Spain; alfredogcp@usal.es (A.G.C.); lauravicente@usal.es (L.V.-V.); amorales@usal.es (A.I.M.); ﬂopezher@usal.es (F.J.L.-H.) 2 Department of Physiology and Pharmacology, University of Salamanca, 37007 Salamanca, Spain 3 T i l A U i it f S l 37007 S l S i 2 Department of Physiology and Pharmacology, University of Salamanca, 37007 Salamanca, Spain 3 Toxicology Area, University of Salamanca, 37007 Salamanca, Spain 4 Fundación Instituto de Estudios de Ciencias de la Salud de Castilla y León, 42002 Soria, Spain 5 Group of Translational Research on Renal and Cardiovascular Diseases (TRECARD), National Network for Kidney Research REDINREN, RD016/0009/0025, Instituto de Salud Carlos III, 37007 Salamanca, Spain 6 Intelligent Data Analysis Laboratory (IDAL), Dpt. Enginyeria Electrònica, ETSE-UV, Universitat de València, 46100 Valencia, Spain; Mykola.Harvat@uv.es (M.H.); oscar.pellicer@uv.es (Ó.J.P.-V.) Citation: Casanova, A.G.; Harvat, M.; Vicente-Vicente, L.; Pellicer-Valero, Ó.J.; Morales, A.I.; López-Hernández, F.J.; Martín-Guerrero, J.D. Regression Modeling of the Antioxidant-to-Nephroprotective Relation Shows the Pivotal Role of Oxidative Stress in Cisplatin Nephrotoxicity. Antioxidants 2021, 10, 1355. https://doi.org/10.3390/ antiox10091355 Academic Editor: Nicole Schupp Received: 16 July 2021 Accepted: 23 August 2021 Published: 26 August 2021 antioxidants antioxidants 1. Introduction Cisplatin is one of the most potent and widely used chemotherapeutic drugs for the treatment of a variety of solid cancers [1], but its dosage and clinical utility are limited by nephrotoxicity [2]. Nephrotoxicity occurs in 25–35% of adult [3] and 70% of pedi- atric [4] therapeutic courses. Direct effects on the renal vasculature are involved [2], but cisplatin nephrotoxicity mostly shows a tubular damage pattern of dysfunction and de- rangement, producing electrolytic disturbances (i.e., most typically hypomagnesemia and https://www.mdpi.com/journal/antioxidants Antioxidants 2021, 10, 1355. https://doi.org/10.3390/antiox10091355 2 of 21 Antioxidants 2021, 10, 1355 hypokalemia), acute tubular injury (ATI), and acute kidney injury (AKI), with elevated plasma creatinine (pCr) and urea (pUrea) levels [2,5–7], which may occasionally progress to chronic ﬁbrotic nephropathy [8,9]. As shown in Figure 1, tubular damage causes a reduc- tion in glomerular ﬁltration rate (GFR) by a number of mechanisms, including activation of the tubuloglomerular feedback (TGF) mechanism and renal vasoconstriction induced by inﬂammation and factors released by activated renal cells [2,10]. Figure 1. Mechanisms of cisplatin nephrotoxicity, including oxidative stress as a contributing factor. A, autoregulation of renal blood ﬂow and intraglomerular blood pressure. GFR, glomerular ﬁltration rate. Kf, ultraﬁltration coefﬁcient. pCr, plasma creatinine concentration. pUrea, plasma urea concentration. RBF, renal blood ﬂow. Figure 1. Mechanisms of cisplatin nephrotoxicity, including oxidative stress as a contributing factor. A, autoregulation of renal blood ﬂow and intraglomerular blood pressure. GFR, glomerular ﬁltration rate. Kf, ultraﬁltration coefﬁcient. pCr, plasma creatinine concentration. pUrea, plasma urea concentration. RBF, renal blood ﬂow. This pathophysiological pattern results from cisplatin accumulation in proximal (mainly the S3 segment) [11,12] and distal tubule cells [2,13], which causes diverse cellular alterations, chieﬂy including inhibition of membrane transporters [2,14], interference with metabolic pathways [15], and cell death [16,17]. Tubular cell death shows apoptotic and nonapoptotic phenotypes, depending on the level of exposure to cisplatin [16]. While lower concentrations induce apoptosis, higher concentrations cause a necrotic-like pheno- Antioxidants 2021, 10, 1355 3 of 21 3 of 21 type [18,19]. Inside the cells, cisplatin becomes aquated and turns into a potent nucleophilic that binds to numerous targets, most prominently nucleic acids and many proteins [18,20]. Cisplatin cytotoxicity has been traditionally explained by formation of inter- and intra- strand adducts with nuclear DNA, which activates DNA repair mechanisms that, when overwhelmed, in turn, activate apoptosis. This cytotoxic mechanism is very effective in rapidly dividing cells, because nonrepaired DNA activates the p53–p21 cyclin-dependent kinase 2 (cdk2) pathway to make death/life decisions at cell division checkpoints [21]. type [18,19]. Inside the cells, cisplatin becomes aquated and turns into a potent nucleophilic that binds to numerous targets, most prominently nucleic acids and many proteins [18,20]. Cisplatin cytotoxicity has been traditionally explained by formation of inter- and intra- strand adducts with nuclear DNA, which activates DNA repair mechanisms that, when overwhelmed, in turn, activate apoptosis. This cytotoxic mechanism is very effective in rapidly dividing cells, because nonrepaired DNA activates the p53–p21 cyclin-dependent kinase 2 (cdk2) pathway to make death/life decisions at cell division checkpoints [21]. p y p Despite bearing a high and ready division capacity (for regeneration purposes), the proliferation rate of tubular epithelial cells is, however, very low under normal condi- tions [22]. Cell-cycle-independent mechanisms have been described, which might explain cisplatin cytotoxicity in target, nonproliferating epithelial cells, in which the drug accu- mulates [2,13]. Apoptotic and necrotic signaling is induced from damaged structures and organelles, such as mitochondria, endoplasmic reticulum, lysosomes, and others [16,21]. Cisplatin also induces oxidative stress in tubule epithelial cells in culture and in animal models [23–25] by accumulating in mitochondria and interfering with mitochondrial home- ostasis and respiration [16,23]. Oxidative stress causes, or contributes to causing cell death, in general [26,27], and speciﬁcally after exposure to cisplatin [28–30]. In addition, oxidative stress also participates in other mechanisms of nephrotoxicity, such as renal vascular [31–34] and mesangial [35] contraction, endothelial dysfunction [36,37], inﬂamma- tion [38–40], and TGF enhancement [32,33], leading to renal blood ﬂow and GFR reduction and damage ampliﬁcation [2,21] (Figure 1). Oxidative stress has been proposed as a prominent event and mediator of cisplatin cytotoxicity and nephrotoxicity [2,13,21], but its relative weight among other pathophys- iological mechanisms, and its hierarchical and causality relation with them, are mostly unknown. In this article, we studied and modeled the relation between the degree of reduction in oxidative stress and the degree of protection of cisplatin nephrotoxicity be- stowed by exogenous antioxidants in a number of studies with animal models. A key role of oxidative stress in cisplatin nephrotoxicity was inferred from the linear relation between the antioxidant and nephroprotective effects, with almost complete prevention of nephrotoxicity at maximal antioxidant effect. 2. Materials and Methods 2.1. Data Mining The data used for this study were obtained from the literature search carried out in our previous meta-analysis [41], in which preclinical studies reporting molecules or products preventing cisplatin nephrotoxicity were identiﬁed. Among them, only those articles meeting the following criteria were used: (1) evaluating antioxidant nephropro- tectants, (2) conducted on experimental animals, (3) providing number of individuals per experimental group, (4) using cisplatin as the nephrotoxic agent, (5) written in English, (6) fully accessible for authors (through journal subscriptions, request to authors, or open access), (7) using pUrea or blood urea nitrogen (BUN) level as the parameter to estimate nephroprotection, and (8) using malonyldialdehyde (MDA) to evaluate oxidative stress, as previously described [42]. The subsequent mathematical analysis was performed only with those studies reporting statistically signiﬁcant nephroprotective and antioxidant effects (with respect to the group that received cisplatin and no nephroprotectant). Publication bias was evaluated with the asymmetry tests of Begg and Mazumdar [43], and Egger et al. [44]. • Nephroprotection index (Enep): Enep = 1 −MaxNP −BasNP MaxNA −BasNA where MaxNP is the value of the nephrotoxicity biomarker (i.e., pUrea or pBUN) at the maximum toxicity time in the nephroprotectant+cisplatin group; BasNP is the value of the nephrotoxicity biomarker at basal time point in the nephroprotectant+cisplatin group; MaxNA is the value of the nephrotoxicity biomarker at the maximum toxicity time point in the cisplatin group; and BasNA is the value of the nephrotoxicity biomarker at the basal time point in the cisplatin group. Thereof, MaxNP-BasNP corresponds to the increment in the level of the nephrotoxicity biomarker in the nephroprotectant+cisplatin group; and MaxNA- BasNA corresponds to the increment in the level of the same nephrotoxicity biomarker in the aminoglycoside group. Enep > 0 denotes nephroprotection (i.e., reduced cisplatin nephrotoxicity due to the action of the nephroprotectant), with the higher the value of Enep, the higher the nephroprotective effect. Enep = 1 represents total nephroprotection. Enep = 0 means there is no effect exerted by the nephroprotectant. • Antioxidant index (Eoxi): • Antioxidant index (Eoxi): Eoxi = 1 −MaxOP −BasOP MaxOA −BasOA where MaxOP is the value of the oxidative stress biomarker (i.e., MDA) at the maximum toxicity time point in the nephroprotectant+cisplatin group; BasOP is the value of the oxidative stress biomarker at the basal time point in the nephroprotectant+cisplatin group; MaxOA is the value of the oxidative stress biomarker at the maximum toxicity time point in the cisplatin group; and BasOA is the value of the oxidative stress biomarker at the basal time point in the cisplatin group. Thereof, MaxOP-BasOP corresponds to the increment in the level of the oxidative stress biomarker in the nephroprotectant+cisplatin-treated group; and MaxOA-BasOA corresponds to the increment in the level of the same oxidative stress biomarker in the cisplatin-treated group. Eoxi > 0 denotes antioxidant activity due to the protector, with the higher the value of Eoxi, the higher the antioxidant effect. Eoxi = 1 represents a complete antioxidant effect. Eoxi = 0 means no antioxidant effect is exerted by the nephroprotectant. The Eoxi versus Enep relation was represented. We used an ordinary least squares (OLS) approach for building a linear regression model between Eoxi as an independent variable, and Enep as a dependent variable. Nonlinear models were also taken into account, but they did not improve the performance achieved by their linear counterparts. 2.2. Mathematical Modeling With the objective of evaluating a potential relation between the antioxidant and the nephroprotective activity of the nephroprotectants included in the study, the following parameters were deﬁned: Antioxidants 2021, 10, 1355 4 of 21 • Nephroprotection index (Enep): • Nephroprotection index (Enep): • Antioxidant index (Eoxi): • Nephroprotection index (Enep): As on the basal state of Eoxi = 0, we should expect no nephroprotection effect (Enep = 0); we used this fact in the model assessment and supposed a zero-centered model that was tested with a proper model assessment using the Akaike information criterion (AIC). In particular, the ﬁnal linear regression model was given by a weighted linear combination: Enep = w ∗Eoxi The model was assessed by measuring the statistical signiﬁcance of the coefﬁcients; the variability of the relationship between the predictors and the target value was determined by the corresponding R2 coefﬁcient [45]. As the presence of outliers was considerably high, two additional robust techniques were considered, namely the Huber regression and the random sample consensus (RANSAC) algorithm [46,47]. The Huber regression is a robust technique that uses a Huber loss func- tion instead of the standard least squares in order to penalize the error depending on their magnitude [47]. RANSAC is an iterative estimation algorithm which ﬁts several iterative models on subsets of data, and then selects the subset with the least average error that, by assumption, is the subset with no outlier points [46]. The value of the slope coefﬁcient corresponding to each of the three models was eventually compared as an evaluation Antioxidants 2021, 10, 1355 5 of 21 metric about the inﬂuence of outliers in the coefﬁcient estimation carried out by OLS. All three models were ﬁtted using Python module Scikit-learn [48]; the rest of the processing was performed in Python programming language [49]. 3. Results The characteristics of the studies included in this work are provided in Table 1. Table 1. Descriptive data of the studies that met the inclusion criteria. CP, cisplatin; i.p., intraperitoneal; i.v., intravenous; NPT, nephroprotectant; p.o., per os (i.e., oral administration). Reference Animal Species Study Length Cisplatin NPT Daily Dose, Administration Route, Dose CP + NPT (n) Dose and Route n Abdel-Aziz et al., 2018 [50] Rat 15 days 5 mg/kg, i.p. 8 Diacerein 50 mg/kg, p.o. 14 doses 100 mg/kg, p.o. 14 doses 8 8 Abdel Moneim et al., 2014 [51] Rat 10 days 5 mg/kg, i.p. 7 Azadirachta indica leaf methanolic extract 500 mg/kg, p.o. 5 doses before CP 500 mg/kg, p.o. 5 doses after CP 7 7 Abdel-Wahab et al., 2017 [52] Rat 4 weeks 6 mg/kg, i.p. 6 N-acetylcisteine 50 mg/kg, i.p. 12 doses Taurine 50 mg/kg, i.p. • Nephroprotection index (Enep): 12 doses N-acetylcisteine + Taurine 50 mg/kg, i.p. + 50 mg/kg, i.p. 12 doses 6 6 6 Alibakhshi et al., 2018 [53] Rat 7 days 7.5 mg/kg, i.p. 5 Zingerone 50 mg/kg, p.o. 7 doses 5 Al-Husseiny et al. 2016 [54] Rat 30 days 5 mg/kg, i.p. 20 Human amniotic ﬂuid stem cells (5 × 106) 20 Amirshahrokhi and Khalili, 2015 [55] Mouse 4 days 15 mg/kg, i.p. 8 Thalidomide 100 mg/kg, p.o. 4 doses. 8 An et al., 2011 [56] Mouse 5 days 5 mg/kg, i.p. once daily for 5 days 9 Pravastatin 80 mg/kg, p.o. 5 doses 9 Badawy et al., 2019 [57] Rat 12 days 7 mg/kg, i.p. 10 Wogonin 40 mg/kg, i.p. 12 days 10 Bami et al., 2017 [58] Rat 5 days 10 mg/kg, i.p. 6 Ferulic acid 50 mg/kg, p.o. 5 doses 6 Bayomi et al., 2013 [59] Rat 7 days 10 mg/kg, i.p. 10 SB-4315421 mg/kg, i.p. 3 doses 10 Bazmandegan et al., 2019 [60] Mouse 4 days 20 mg/kg, i.p 7 Sumatriptan 0.3 mg/kg, i.p. 3 doses 7 Chen et al., 2019 [61] Mouse 3 days 20 mg/kg, i.p 5 Hesperetin 50 mg/kg, i.p. 3 doses 5 Chirino et al., 2008 [62] Rat 10 days 7.5 mg/kg, i.p. 10 Apocynin 2 g/L in drinking water, p.o. for 10 days 10 Darwish et al., 2017 [63] Rat 14 days 6 mg/kg, i.p. 6 Vitamin E 75 mg/kg, i.p. 14 doses 6 Dehnamaki et al., 2019 [64] Mouse 5 days 20 mg/kg, i.p 7 Troxerutin 150 mg/kg 7 metric about the inﬂuence of outliers in the coefﬁcient estimation carried out by OLS. All three models were ﬁtted using Python module Scikit-learn [48]; the rest of the processing was performed in Python programming language [49]. metric about the inﬂuence of outliers in the coefﬁcient estimation carried out by OLS. All three models were ﬁtted using Python module Scikit-learn [48]; the rest of the processing was performed in Python programming language [49]. Antioxidants 2021, 10, 1355 6 of 21 Table 1. Cont. Reference Animal Species Study Length Cisplatin NPT Daily Dose, Administration Route, Dose CP + NPT (n) Dose and Route n Divya et al., 2016 [65] Rat 6 days 16 mg/kg, i.p. 6 Silymarin 100 mg/kg, p.o. 5 doses Apodytes dimidiata leaf methanolic extract 250 mg/kg, p.o. 5 doses before CP 250 mg/kg, p.o. • Nephroprotection index (Enep): 5 doses after CP 6 6 6 Elhusseini et al., 2016 [66] Rat 30 days 10 mg/kg, i.p. 20 Human adipose-derived mesenchymal stem cells (5 × 106), i.v. 20 El-Naga, 2014 [67] Rat 2 weeks 7 mg/kg, i.p. 10 Cardamonin 10 mg/kg, p.o. 14 doses 30 mg/kg, p.o. 14 doses 10 10 El-Naga and Mahran, 2016 [7] Rat 2 weeks 7 mg/kg, i.p. 10 Indole-3-carbinol 20 mg/kg, p.o. 14 doses 10 Elsherbiny et al., 2016 [68] Rat 10 days 10 mg/kg, i.p. 10 Arjunolic acid 100 mg/kg, p.o. 3 doses 250 mg/kg, p.o. 3 doses 10 10 Fatima et al., 2016 [69] Rat 6 days 7 mg/kg, i.p. 8 Epigallocatechin gallate + coenzyme Q10 15 mg/kg, i.p. + 5 mg/kg, i.p. 6 doses 8 Fernández- Rojas et al., 2014 [70] Mouse 4 days 18 mg/kg, i.p. 4 C-phycocyanin 10 mg/kg, i.p. 1 dose 30 mg/kg, i.p. 1 dose 4 4 Hassan et al., 2014 [71] Rat 6 weeks 7.5 mg/kg, i.p. 8 Grape seed proanthocyanidin extract 100 mg/kg, p.o. 42 doses 8 Helmy et al., 2014 [72] Rat 4 days 6 mg/kg, i.p. 7– 8 BQ-123 1 mg/kg, i.p. 2 doses 7–8 Hosseini et al., 2018 [73] Rat 3 days 8 mg/kg, i.p. 6 Rheum turkestanicum root extract 100 mg/kg, i.p. 1 dose 200 mg/kg, i.p. 1 dose 6 6 Y.C. Huang et al., 2017 [74] Mouse 3 days 20 mg/kg, i.p. 5 Galangin 75 mg/kg, p.o. 3 doses 5 H. Huang et al., 2017 [75] Rat 12 days 8 mg/kg, i.p. 6 Schisandra chinensis bee pollen extract 400 mg/kg, p.o. 12 doses starting 7 days prior to CP 800 mg/kg, p.o. 12 doses starting 7 days prior to CP 1200 mg/kg, p.o. 12 doses starting 7 days prior to CP after 6 6 6 Huang et al., 2019 [76] Mouse 4 days 22 mg/kg, i.p. 8 N-Acetylcysteine 50 mg/kg, i.p. 3 doses 8 Kandemir et al., 2019 [77] Rat 8 days 7 mg/kg, i.p. 8 Zingerone 25 mg/kg, p.o. 7 doses 50 mg/kg, p.o. 7 doses 8 8 Kang et al., 2016 [78] Mouse 7 days 20 mg/kg, i.p. 12 Sappanone A 10 mg/kg, i.p. 3 doses 20 mg/kg, i.p. 3 doses 40 mg/kg, i.p. 3 doses 12 12 12 Table 1. Cont. 7 of 21 Antioxidants 2021, 10, 1355 Table 1. Cont. • Nephroprotection index (Enep): Reference Animal Species Study Length Cisplatin NPT Daily Dose, Administration Route, Dose CP + NPT (n) Dose and Route n Kenza et al., 2017 [79] Mouse 11 days 8 mg/kg, i.p. 6 Vitamin E 100 mg/kg, p.o. 10 doses Centaurea choulettiana Pomel leaf butanolic extract 150 mg/kg, p.o. 10 doses 6 6 Khairnar et al., 2020 [80] Rat 6 days 5 mg/kg, i.p. 6 Disulﬁram 50 mM/kg, p.o. 5 doses Disulﬁram + CuCl2 50 mM/kg + 50 mM/kg, p.o. 5 doses Disulﬁram cooper chelate (Cu-DEDC) 50 mM/kg, p.o. 5 doses Amifostine 100 mg/kg i.v. 1 dose 6 6 6 6 Kim et al., 2018 [81] Mouse 6 days 15 mg/kg, i.p. 8 Ac-YVAD-cmk 10 mg/kg, i.p. 3 doses 8 F. Li et al., 2018 [82] Mouse 4 days 20 mg/kg, i.p. 12 Xanthohumol 12.5 mg/kg, i.p. 3 doses 25 mg/kg, i.p. 3 doses 50 mg/kg, i.p. 3 doses 12 12 12 Y.Z. Li et al., 2018 [83] Mouse 10 days 20 mg/kg, i.p. 8 Schisandra chinensis extract 300 mg/kg, p.o. 10 doses 600 mg/kg, p.o. 10 doses 8 8 Li et al., 2019 [84] Mouse 11 days 20 mg/kg, i.p. 10 Arginyl-fructosyl-glucose 40 mg/kg, p.o. 10 doses starting 3 days prior to CP 80 mg/kg, p.o. 10 doses starting 3 days prior to CP 10 10 Ma et al., 2015 [85] Mouse 7 days 15 mg/kg, i.p. 10 Icariin 30 mg/kg, p.o. 6 doses 60 mg/kg, p.o. 6 doses 10 10 Ma et al., 2017 [86] Rat 5 days 7 mg/kg, i.p. 7 Puerarin 3 days before CP and 5 days after CP 30 mg/kg i.v. 50 mg/kg i.v. 7 7 Malik et al., 2015 [87] Rat 10 days 8 mg/kg, i.p. 6 Nobiletin 5 mg/kg, i.p. 10 doses 6 Mi et al., 2018 [88] Mouse 10 25 mg/kg i.p. 8 Maltol 100 mg/kg p.o. 10 doses starting 7 days prior to CP 8 Mohamed et al., 2013 [89] Rat 2 weeks 9 mg/kg, i.p. divided in two doses once a week for two weeks. 20 Recombinant human erythropoietin 100 IU/kg, i.p. 14 doses 20 Morsy and Heeba, 2016 [90] Rat 7 days 6 mg/kg, i.p. 6– 8 Nebivolol 10 mg/kg, p.o. 7 doses 6–8 Mundhe et al., 2015 [91] Rat 5 days 7.5 mg/kg, i.p. 6 Nordihydroguaiaretic acid 10 mg/kg, i.p. 5 doses 6 Table 1. Cont. Antioxidants 2021, 10, 1355 8 of 21 Table 1. Cont. • Nephroprotection index (Enep): Reference Animal Species Study Length Cisplatin NPT Daily Dose, Administration Route, Dose CP + NPT (n) Dose and Route n Mundhe et al., 2019 [92] Rat 10 days 7.5 mg/kg i.p. 8 Nordihydroguaiarectic acid 10 mg/kg i.p. 5 doses before CP 10 mg/kg i.p. 5 doses after CP 8 8 Nazari Soltan Ahmad et al., 2018 [93] Rat 10 days 8 mg/kg i.p. 6 Tangeretin 2.5 mg/kg i.p. 7 doses before CP and 3 after 5 mg/kg i.p. 7 doses before CP and 3 after 6 6 Nazari Soltan Ahmad et al., 2018 [94] Rat 4 days 20 mg/kg i.p. 5 Dunnione 10 mg/kg p.o. 4 doses starting 12 h prior to CP 20 mg/kg p.o. 4 doses starting 12 h prior to CP 5 5 Neamatallah et al., 2018 [95] Rat 11 days 7.5 mg/kg, i.p. 6 Talh honey 2.5 g/kg, p.o. 10 doses 6 Purena et al., 2018 [96] Rat 14 days 12 mg/kg, i.p. 5 Emblica ofﬁcinalis leaf ethanolic extract 100 mg/kg, p.o. 14 doses 200 mg/kg, p.o. 14 doses 5 5 Qi et al., 2018 [97] Mouse 10 days 20 mg/kg i.p. 8 Pseudoginsengenin DQ 30 mg/kg p.o. 10 doses starting 7 days prior to CP 60 mg/kg p.o. 10 doses starting 7 days prior to CP 8 8 Radwan et al., 2017 [98] Rat 10 days 7.5 mg/kg, i.p. 6 Rutin 200 mg/kg p.o. Low dose gamma radiation (LDR) Rutin 200 mg/kg p.o. + LDR 6 6 6 Rana et al., 2016 [99] Rat 10 days 6 mg/kg, i.p. 6 Bauhinia purpurea bark ethanolic extract 400 mg/kg, p.o. 9 doses Bauhinia purpurea unripe pod extract 400 mg/kg, p.o. 9 doses 6 6 Sahin et al., 2014 [100] Rat 12 days 7 mg/kg, i.p. 7 Curcumin diﬂuorinated 50 mg, p.o. 12 doses Curcumin 50 mg, p.o. 12 doses 7 7 Saiﬁet al., 2019 [101] Mouse 14 days 28 days 10 mg/kg i.p. 5 mg/kg i.p. CP every week for 3 consecutive weeks 6 6 Nanoceria 2 mg/kg i.p. 14 doses starting 5 days prior to CP Nanoceria 0.2 mg/kg i.p. 28 doses starting 5 days prior to 1st dose of CP 2 mg/kg i.p. 28 doses after 1st dose of CP 6 6 6 Sen et al., 2018 [102] Rat 25 days 5 mg/kg, i.p. every ﬁve days (four injections) for 25 days 6 Dillenia indica fruit methanolic extract 300 mg/kg, p.o. • Nephroprotection index (Enep): 25 doses Dillenia indica fruit ethanolic extract 300 mg/kg, p.o. 25 doses 6 6 Sener et al., 2012 [103] Rat 14 days 10 mg/kg, i.p. 14 days 7 Mirtazapine 15 mg/kg, p.o. 14 doses 30 mg/kg, p.o. 14 doses 7 7 Table 1. Cont. Antioxidants 2021, 10, 1355 9 of 21 Table 1. Cont. Reference Animal Species Study Length Cisplatin NPT Daily Dose, Administration Route, Dose CP + NPT (n) Dose and Route n Sharma and Goyal, 2012 [104] Mouse 7 days 16 mg/kg, i.p. 6 Heliotropium eichwaldii root methanolic extract 400 mg/kg, p.o. 7 doses 6 Sherif, 2015 [105] Rat 10 days 7 mg/kg, i.p. 10 Arjunolic acid 20 mg/kg, p.o. 10 doses 10 Shi et al., 2019 [106] Mouse 7 days 20 mg/kg i.p. 8 Docosahexaenoic acid (DHA) 12.5 mg per day p.o. for 4 days prior to CP Docosahexaenoic acid-phosphatidylcholine (DHA-PC) 12.5 mg per day p.o. for 4 days prior to CP 8 8 Topcu- Tarladacalisir et al., 2016 [107] Rat 10 days 7.5 mg/kg, i.p. 6 Curcumin 200 mg/kg, p.o. 4 doses 6 Z. Wang et al., 2018 [108] Mouse 10 days 25 mg/kg i.p. 8 Ginsenoside-Re (G-Re) 25 mg/kg p.o. 10 doses starting 7 days prior to CP 8 Yang et al., 2017 [109] Mouse 10 days 25 mg/kg, i.p. 8 Sika deer antler protein powder 10 mg/kg, p.o. 10 doses 20 mg/kg, p.o. 10 doses 8 8 Yu et al., 2018 [110] Mouse 96 h 20 mg/kg i.p. 8 Celastrol 1 mg/kg i.p. 24 h before CP 2 mg/kg i.p. 24 h before CP 8 8 L. Zhang et al., 2018 [111] Mouse 7 days 20 mg/kg i.p. 3 Daphnetin 2.5 mg/kg i.p. 3 doses before CP 5 mg/kg i.p. 3 doses before CP 10 mg/kg i.p. 3 doses before CP Pyrrolidine dithiocarbamate (PDTC) 3 3 3 3 Table 1. Cont. The Begg–Mazumdar test applied to assess potential publication bias yielded a Kendall’s tau of 0.74 (p < 0.001). Similarly, the Egger test provided a bias of 10.49 (95% CI = 9.56, 11.43; p < 0.001). Both tests showed the presence of asymmetry. How- ever, in our study, this result was expected and is not necessarily reﬂective of publication bias. In fact, pursuant of our objective, only studies reporting a statistically signiﬁcant nephroprotective effect were included, as stated in the Methods. • Nephroprotection index (Enep): The OLS model was evaluated by checking the statistical signiﬁcance of the coefﬁcients with an alpha error threshold of 0.01. We obtained the following results: w = 0.938 (95% CI = (0.89, 0.987), p < 0.0001). We used the Akaike information criterion (AIC) [112] to assess the choice of including or excluding the bias term from the ﬁnal model. We obtained an AICintercept = −66.32 for the linear model with bias term, and AICbase = −61.90 for the model without bias term. Therefore, based on this result, we only kept the slope term in the resulting model, R2 = 0.932, meaning that 93.2% of the variability of Enep could be explained by Eoxi. The potential inﬂuence of outliers in the ﬁnal model was also assessed. In particular, Huber and RANSAC regressions were obtained, assuming the same model as for the OLS case (i.e., without intercept term). Both algorithms yielded similar slope values to that obtained by the OLS regression model: wHuber = 0.957 and wRANSAC = 0.965. We concluded that, under our assumptions, the outliers had no signiﬁcant inﬂuence on the ﬁnal model. The three models are depicted in Figure 2. Studies in which Eoxi > 1 were removed from the models. In these studies, the antioxidant reduced oxidative stress beyond the basal Antioxidants 2021, 10, 1355 10 of 21 10 of 21 level (i.e., the level of oxidative stress in the control group), which had a negative impact on nephroprotection. Speciﬁcally, Enep showed a negative slope beyond Eoxi = 1 (data not shown). This is because normal (i.e., basal) production of reactive oxygen species (ROS) has been shown to have homeostatic signaling roles [113–115]. As a corollary, inhibition of basal ROS production may reasonably result in deleterious effects for cell and organ function [116]. level (i.e., the level of oxidative stress in the control group), which had a negative impact on nephroprotection. Speciﬁcally, Enep showed a negative slope beyond Eoxi = 1 (data not shown). This is because normal (i.e., basal) production of reactive oxygen species (ROS) has been shown to have homeostatic signaling roles [113–115]. As a corollary, inhibition of basal ROS production may reasonably result in deleterious effects for cell and organ function [116]. Figure 2. Regression results for the three tested models. • Nephroprotection index (Enep): The nephroprotection index (Enep) is represented versus the antioxidant index (Eoxi) for the OLS, Huber regression, and RANSAC regression models, all of them yielding similar results. The RANSAC model provided the most robust ﬁt, as it empirically ignored some outlier points, giving them a zero weight in the ﬁnal adjustment. MSE, mean squared error. Figure 2. Regression results for the three tested models. The nephroprotection index (Enep) is represented versu antioxidant index (Eoxi) for the OLS, Huber regression, and RANSAC regression models, all of them yielding similar re Figure 2. Regression results for the three tested models. The nephroprotection index (Enep) is represented versus the antioxidant index (Eoxi) for the OLS, Huber regression, and RANSAC regression models, all of them yielding similar results. The RANSAC model provided the most robust ﬁt, as it empirically ignored some outlier points, giving them a zero weight in the ﬁnal adjustment. MSE, mean squared error. Products located over the model provided more nephroprotection than expected from their antioxidant effect. Based on the RANSAC regression (the model with the most robust ﬁt), these products were subclassiﬁed as showing 25, 50, 75, or 100% of additional nephroprotection (Figure 3); they are identiﬁed and listed in Table 2. 11 of 21 Antioxidants 2021, 10, 1355 Figure 3. RANSAC linear regression and 25, 50, 75, and 100% relative error areas. Products over the model afford a greater nephroprotection than expected from their antioxidant effect. Products within these areas are identiﬁed and shown in Table 2. Letters (a through v) identify individual products, as listed in Table 2. Figure 3. RANSAC linear regression and 25, 50, 75, and 100% relative error areas. Products over the model afford a greater nephroprotection than expected from their antioxidant effect. Products within these areas are identiﬁed and shown in Figure 3. RANSAC linear regression and 25, 50, 75, and 100% relative error areas. Products over the model afford a greater nephroprotection than expected from their antioxidant effect. Products within these areas are identiﬁed and shown in Table 2. Letters (a through v) identify individual products, as listed in Table 2. Table 2. Products providing higher nephroprotection than expected from their antioxidant effect, according to their relative position with respect to the RANSAC linear regression model. Table 2. Products providing higher nephroprotection than expected from their antioxidant effect, according to their relative position with respect to the RANSAC linear regression model. Table 2. 4. Discussion The regression model best adjusting our experimental data shows a linear relation- ship between inhibition of oxidative stress and amelioration of cisplatin nephrotoxicity (Figure 2). This relation intercepts the nephroprotection axis (i.e., the y-axis) very near the Enep = 1 value at the maximal antioxidant point (i.e., Eoxi = 1 in the x-axis). This indicates that a complete abrogation of oxidative stress apparently leads to a complete prevention of nephrotoxicity. Thus, oxidative stress might not only be a contributing, but a pivotal mechanism of cisplatin nephrotoxicity. Cisplatin nephrotoxicity is a tubulopathy, in which all pathophysiological and clinical manifestations derive from cytotoxic tubular injury as the primary event (Figure 1) [2,13]. Consequently, oxidative stress must also be in the core of cisplatin cytotoxicity. p y y Mitochondria are the main intracellular site of cell life/death decision [117–119]. Mi- tochondria funnel and integrate stress signals arising from damaged subcellular structures and organelles, including themselves, and activate apoptotic and necrotic death programs that mostly pose no-return points for cell demise [120]. One of these signals is oxidative stress. Extramitochondrial sources of ROS exist (e.g., the cytosol and the endoplasmic reticulum) [121], but mitochondria are the main source of ROS production and overproduc- tion [115]. Mitochondrial outer membrane permeabilization (MOMP) is a mandatory event for the release of proapoptotic factors (e.g., cytochrome c and AIF), apoptosome formation in the cytosol, and initiation of intrinsic apoptosis [119]. Intracellular death signals regulate MOMP by targeting the outer membrane through pro- and anti-apoptotic Bcl-2 family members, which directly modulate its permeability [118,122]. Inner membrane perme- abilization (i.e., mitochondrial permeability transition, MPT) is also intimately related to cell death. MPT is bidirectionally linked to transmembrane mitochondrial potential (∆Ψ) dissipation, and causes intermembrane swelling, outer membrane disruption, and MOMP. MPT is mediated by a multiprotein complex, the permeability transition pore (PT pore or PTP). PTP is located at sites of inner–outer membrane connections (where Bcl-2 family members accumulate), is inhibited by anti-apoptotic Bcl-2 members, is critical for apoptosis, and participates in MOMP [123–125]. In isolated mitochondria [126], cisplatin interferes with the respiratory chain, produces oxidative stress [127] and rapid cytochrome c release [128], and causes calcium-dependent mitochondrial swelling and mitochondrial depolarization, as a consequence of PT pore opening [129]. In this scenario, oxidative stress may be the cause or the consequence of the other events. • Nephroprotection index (Enep): Products providing higher nephroprotection than expected from their antioxidant effect, according to their relative position with respect to the RANSAC linear regression model. ID Product Relative Error (%) a Nanoceria 0.2 mg/kg (before CP) (28 doses) >100% b Nanoceria 2 mg/kg (after CP) (28 doses) >75% c RH Erythropoietin 100 IU/kg >75% d Maltol 100 mg/kg >50% e Centaurea choulettiana Pomel leaf butanolic extract 150 mg/kg >50% f Wogonin 40 mg/kg >25% g Rutin 200 mg/kg >25% h Celastrol 1 mg/kg >25% i SB-431542 1 mg/kg >25% j Arjunolic acid 100 mg/kg >25% k Troxerutin 150 mg/kg >25% 12 of 21 Antioxidants 2021, 10, 1355 Table 2. Cont. Table 2. Cont. ID Product Relative Error (%) l Sika deer antler protein powder 10 mg/kg >25% m Puerarin 50 mg/kg >25% n N-acetylcisteine 50 mg/kg >25% o Mesenchymal stem cells 5 × 106 cells >25% p Human amniotic ﬂuid stem cells 5 × 106 cells >25% q Vitamin E 100 mg/kg >25% r Sumatriptan 0.3 mg/kg >25% s Celastrol 2 mg/kg >25% t Taurine 50 mg/kg >25% u Daphnetin 2.5 mg/kg >25% v Sappanone A 10 mg/kg >25% 4. Discussion In fact, decoupling or inhibition of mitochondrial respiration induces both PT pore opening and oxidative stress [117,130,131]. PT pore opening (and, thus, MPT) is triggered by mitochondrial Ca2+ and potentiated by oxidative stress [125,132,133], suggest- ing that alterations in respiration induce oxidative stress, and this, in turn, contributes to 13 of 21 Antioxidants 2021, 10, 1355 13 of 21 the opening of the PT pore. In agreement, antioxidants inhibit MPT [134]. However, vice versa is also possible: PT pore opening produces ∆Ψ dissipation, respiratory uncoupling, and oxidative stress [124,132,135]. As such, oxidative stress and mitochondrial dysfunction induce one another [136,137], and so a causality dilemma existed for cisplatin cytotoxic- ity [16]. Cisplatin also causes oxidative stress by directly damaging mitochondrial DNA (mtDNA) [128,138–140], which impairs appropriate expression of mitochondrial enzymes forming the respiratory chain, and thus induces oxidative stress. Finally, cisplatin abates the antioxidant barrier by inhibiting superoxide dismutase (SOD), catalase, glutathione peroxidase, glutathione S-transferase [141–143], and glutathione reductase [144] in kidney tissues. Figure 4 summarizes the participation of oxidative stress in the tubular pathophysi- ological scenario. The results of the present study are more congruent with oxidative stress being mainly upstream of MPT and MOMP, because, after these mitochondrial events have occurred, the cell is irreversibly committed to dying [120]. Figure 4. Schematic depiction of the pivotal role of oxidative stress in the tubular dysfunction induced by cisplatin. ATP, adenosine triphosphate. MOMP, mitochondrial outer membrane permeability. MPT, mitochondrial permeability transition. mtDNA, mitochondrial DNA. ROS, reactive oxygen species. Figure 4. Schematic depiction of the pivotal role of oxidative stress in the tubular dysfunction induced by cisplatin. ATP, adenosine triphosphate. MOMP, mitochondrial outer membrane permeability. MPT, mitochondrial permeability transition. mtDNA, mitochondrial DNA. ROS, reactive oxygen species. Our study closely relates oxidative stress to the reduction in glomerular ﬁltration (using pUrea as a proxy). GFR reduction is a pivotal alteration in cisplatin nephrotoxicity, derived mostly from tubular cytotoxicity (as shown in Figure 1) [2], and an internationally recognized hallmark of AKI, regardless of etiology [145]. However, tubular damage and GFR decline are not directly proportional. In fact, an undetermined degree of tubular damage may occur without affecting GFR [146,147], as undamaged nephrons may, to a certain extent, sustain (total) GFR by increasing their single-nephron GFR (SNGFR) [148]. 4. Discussion This implies that additional injury mechanisms (unrelated to oxidative stress) might re- Antioxidants 2021, 10, 1355 14 of 21 14 of 21 main under maximal antioxidant circumstances. Potential oxidative stress-independent mechanisms are known (see Figures 1 and 4, and [2,16,21]), but their weight in cisplatin toxicity is unknown. They would pose potential targets for pharmacological intervention in combination with antioxidants to optimize cisplatin nephrotoxicity prophylaxis. As previously reported [42], nephroprotectants whose effect lies above the model line are products showing greater protective effect than expected from their antioxidant effect. This suggests that additional protection mechanisms are involved, which makes them especially interesting candidates for clinical application. The most effective candidates include nanoceria, recombinant human erythropoietin, maltol, and the butanolic extract of Centaurea choulettiana Pomel (Table 2). On the contrary, those compounds lying below the model line are less effective than expected, implying that they also activate counteracting mechanisms, and are thus less interesting. Along with its antioxidant effect, nanoceria (cerium oxide nanoparticles) also shows anti-inﬂammatory [101] and antiapoptotic properties [149]. Its anti-inﬂammatory effect has been shown to derive from the inhibition of inducible nitric oxide synthase (iNOS) expression [150] and of the NF-κB signaling pathway [151]. With regard to erythropoi- etin, multiple additional mechanisms have been invoked, including (i) the promotion of tubular cell regeneration, (ii) the reduction in vascular endothelial growth factor (VEGF), hemeoxygenase-1 (HO-1) and iNOS expression [89], (iii) the inactivation of macrophages via NF-κB [152], (iv) the inhibition of TGF-β1 expression [153], and (v) the reduction in polymorphonuclear cell inﬁltration [154]. Anti-inﬂammatory and antiapoptotic properties with involvement of the AMPK/PI3K/Akt pathway have also been attributed to maltol, an ingredient in the food industry [88]. Finally, traditional medicine has attributed anti- inﬂammatory properties to Centaurea choulettiana [155]. However, oxidative stress is known to be involved in the development and perpetuation of inﬂammation [156,157], and in the activation of apoptosis [158]. Accordingly, their anti-inﬂammatory and antiapoptotic properties might be the consequence of their antioxidant capacity, and would thus not explain their additional properties, which need to be further explored. Because drug discovery from plant extracts is a complicated and long process, nanoceria, erythropoietin, and maltol hoard readier potential to become clinical applications, and should thus be further explored. References 1. Florea, A.M.; Büsselberg, D. Cisplatin as an Anti-Tumor Drug: Cellular Mechanisms of Activity, Drug Resistance and Induced Side Effects. Cancers 2011, 3, 1351–1371. [CrossRef] 2. Sánchez-González, P.D.; López-Hernández, F.J.; López-Novoa, J.M.; Morales, A.I. An Integrative View of the Pathophysiological Events Leading to Cisplatin Nephrotoxicity. Crit. Rev. Toxicol. 2011, 41, 803–821. [CrossRef] [PubMed] S.; Ohtsubo, A.; Shoji, S.; Ishikawa, D.; Tanaka, T.; Nozaki, K.; Kondo, R.; Okajima, M.; Miura, S.; et al 3. Sato, K.; Watanabe, S.; Ohtsubo, A.; Shoji, S.; Ishikawa, D.; Tanaka, T.; Nozaki, K.; Kondo, R.; Okajima, M.; Miura, S.; et al. Nephrotoxicity of Cisplatin Combination Chemotherapy in Thoracic Malignancy Patients with CKD Risk Factors. BMC Cancer 2016, 16, 222. [CrossRef] [PubMed] 4. Finkel, M.; Goldstein, A.; Steinberg, Y.; Granowetter, L.; Trachtman, H. Cisplatinum Nephrotoxicity in Oncology Therapeutics: Retrospective Review of Patients Treated between 2005 and 2012. Pediatr. Nephrol. 2014, 29, 2421–2424. [CrossRef] 5. Sharbaf, F.; Farhangi, H.; Assadi, F. Prevention of Chemotherapy-Induced Nephrotoxicity in Children with Cancer. Int. J. Prev. Med. 2017, 8, 76. [CrossRef] [PubMed] 6. Hoek, J.; Bloemendal, K.M.; van der Velden, L.A.A.; van Diessen, J.N.A.; van Werkhoven, E.; Klop, W.M.C.; Tesselaar, M.E.T. Nephrotoxicity as a Dose-Limiting Factor in a High-Dose Cisplatin-Based Chemoradiotherapy Regimen for Head and Neck Carcinomas. Cancers 2016, 8, 21. [CrossRef] 7. El-Naga, R.N.; Mahran, Y.F. Indole-3-Carbinol Protects against Cisplatin-Induced Acute Nephrotoxicity: Role of Calcitonin Gene-Related Peptide and Insulin-like Growth Factor-1. Sci. Rep. 2016, 6. [CrossRef] [PubMed] 8. Pabla, N.; Dong, Z. Cisplatin Nephrotoxicity: Mechanisms and Renoprotective Strategies. Kidney Int. 2008, 73, 994–1007. [CrossRef] 9. Landau, S.I.; Guo, X.; Velazquez, H.; Torres, R.; Olson, E.; Garcia-Milian, R.; Moeckel, G.W.; Desir, G.V.; Saﬁrstein, R. Regulated Necrosis and Failed Repair in Cisplatin-Induced Chronic Kidney Disease. Kidney Int. 2019, 95, 797–814. [CrossRef] , ; , ; q , ; , ; , ; , ; , ; , ; , g Necrosis and Failed Repair in Cisplatin-Induced Chronic Kidney Disease. Kidney Int. 2019, 95, 797–814. [CrossRef] p p y y 10. Lopez-Novoa, J.M.; Quiros, Y.; Vicente, L.; Morales, A.I.; Lopez-Hernandez, F.J. New Insights into the M side Nephrotoxicity: An Integrative Point of View. Kidney Int. 2011, 79, 33–45. [CrossRef] ovoa, J.M.; Quiros, Y.; Vicente, L.; Morales, A.I.; Lopez-Hernandez, F.J. New Insights into the Mechanism of A hrotoxicity: An Integrative Point of View. Kidney Int. 2011, 79, 33–45. [CrossRef] 11. Kuhlmann, M.K.; Burkhardt, G.; Köhler, H. 5. Conclusions Our results have revealed that oxidative stress is not a contributing, but a central medi- ator of cisplatin nephrotoxicity in preclinical models. In agreement, a recent meta-analysis identiﬁed antioxidants as the most effective protectants of cisplatin nephrotoxicity in clinical studies [159]. Interestingly, several antioxidants have shown, in animal models, nephro- protective properties without interfering with the antitumor effect of cisplatin [160–164], a critical issue for clinical application. This might be attributed to cisplatin genotoxicity mostly impacting on proliferating cells, such as tumor cells. In perspective, this study provides a rationale for further clinical development of preventive strategies based on single or combined therapies containing antioxidants. Author Contributions: Conceptualization, L.V.-V., A.G.C., A.I.M. and F.J.L.-H.; methodology, M.H., Ó.J.P.-V. and J.D.M.-G.; validation, M.H. and A.G.C.; formal analysis, all authors.; data curation, A.G.C., A.I.M. and L.V.-V.; writing—original draft preparation, A.G.C., L.V.-V., F.J.L.-H. and J.D.M.-G.; writing—review and editing, all authors; supervision, F.J.L.-H. and J.D.M.-G.; funding acquisi- tion, A.I.M., F.J.L.-H. and J.D.M.-G. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Instituto de Salud Carlos III (Madrid, Spain), grant number PI20/01351 and PI18/00996 and FEDER funds. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. 15 of 21 Antioxidants 2021, 10, 1355 Data Availability Statement: The data presented in this study are available in the article. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Insights into Potential Cellular Mechanisms of Cisplatin Nephrotoxicity and Their Clinical Application. Nephrol. Dial. Transpl. 1997, 12, 2478–2480. [CrossRef] pp p p 12. Price, P.M.; Saﬁrstein, R.L.; Megyesi, J. Protection of Renal Cells from Cisplatin Toxicity by Cell Cycle Inhibitors. Am. J. Physiol. Ren. Physiol. 2004, 286. [CrossRef] [PubMed] 13. Karasawa, T.; Steyger, P.S. An Integrated View of Cisplatin-Induced Nephrotoxicity and Ototoxicity. Toxicol. Lett. 2015, 237, 219–227. [CrossRef] 14. Casanova, A.G.; Fuentes-Calvo, I.; Hernández-Sánchez, M.T.; Quintero, M.; Toral, P.; Caballero, M.T.; Martínez-Salgado, C.; Morales, A.I.; Layton, A.T.; Eleno, N.; et al. The Furosemide Stress Test and Computational Modeling Identify Renal Damage Sites Associated with Predisposition to Acute Kidney Injury in Rats. Transl. Res. 2021, 231, 76–91. [CrossRef] p y j y 15. Macciò, A.; Madeddu, C. Cisplatin: An Old Drug with a Newfound Efﬁcacy-from Mechanisms of Action to Cytotoxicity. Expert Opin. Pharm. 2013, 14, 1839–1857. [CrossRef] 16. Sancho-Martínez, S.M.; Prieto-García, L.; Prieto, M.; López-Novoa, J.M.; López-Hernández, F.J. Subcellular Targets of Cisplatin Cytotoxicity: An Integrated View. Pharmacol. Ther. 2012, 136, 35–55. [CrossRef] 17. Volarevic, V.; Djokovic, B.; Jankovic, M.G.; Harrell, C.R.; Fellabaum, C.; Djonov, V.; Arsenijevic, N. Molecular Mechanisms of Cisplatin-Induced Nephrotoxicity: A Balance on the Knife Edge between Renoprotection and Tumor Toxicity. J. Biomed. Sci. 2019, 26, 1–14. [CrossRef] 18. Sancho-Martínez, S.M.; Prieto-García, L.; Prieto, M.; Fuentes-Calvo, I.; López-Novoa, J.M.; Morales, A.I.; Martínez-Salgado, C.; López-Hernández, F.J. N-Acetylcysteine Transforms Necrosis into Apoptosis and Affords Tailored Protection from Cisplatin Cytotoxicity. Toxicol. Appl. Pharmacol. 2018, 349, 83–93. [CrossRef] 19. Sancho-Martínez, S.M.; Piedraﬁta, F.J.; Cannata-Andía, J.B.; López-Novoa, J.M.; López-Hernández, F.J. Necrotic Concentrations of Cisplatin Activate the Apoptotic Machinery but Inhibit Effector Caspases and Interfere with the Execution of Apoptosis. Toxicol. Sci. 2011, 122, 73–85. [CrossRef] 20. Kartalou, M.; Essigmann, J.M. Mechanisms of Resistance to Cisplatin. Mutat. Res. Fundam. Mol. Mech. Mutagenesis 2001, 478, 23–43. [CrossRef] 21. Chirino, Y.I.; Pedraza-Chaverri, J. Role of Oxidative and Nitrosative Stress in Cisplatin-Induced Nephrotoxicity. Exp. Toxicol. Pathol. 2009, 61, 223–242. [CrossRef] 22. Vogetseder, A.; Picard, N.; Gaspert, A.; Walch, M.; Kaissling, B.; Le Hir, M. Proliferation Capacity of the Renal Proximal Tubule Involves the Bulk of Differentiated Epithelial Cells. Am. J. Physiol. Cell Physiol. 2008, 294, C22–C28. [CrossRef] [PubMed] p y y 23. Yuan, Y.; Wang, H.; Wu, Y.; Zhang, B.; Wang, N.; Mao, H.; Xing, C. P53 Contributes to Cisplatin Induced Renal Oxidative Damage via Regulating P66shc and MnSOD. Cell. Physiol. Biochem. 2015, 37, 1240–1256. [CrossRef] [PubMed] 23. References [CrossRef] [PubMed] y g g p y 34. Wang, D.; Chabrashvili, T.; Wilcox, C.S. Enhanced Contractility of Renal Afferent Arterioles from Angiotensin-Infused Rabbits: Roles of Oxidative Stress, Thromboxane Prostanoid Receptors, and Endothelium. Circ. Res. 2004, 94, 1436–1442. [CrossRef] 35. Dunlop, M.E.; Muggli, E.E. Small Heat Shock Protein Alteration Provides a Mechanism to Reduce Mesa Diabetes and Oxidative Stress. Kidney Int. 2000, 57, 464–475. [CrossRef] [PubMed] 36. Stenvinkel, P. Interactions between Inﬂammation, Oxidative Stress, and Endothelial Dysfunction in End-Stage Renal Disease. J. Ren. Nutr. 2003, 13, 144–148. [CrossRef] 37. Matsuoka, H. Endothelial Dysfunction Associated with Oxidative Stress in Human. Diabetes Res. Clin. Pract. 2001, 54, S65–S72. [CrossRef] 38. Jha, J.C.; Ho, F.; Dan, C.; Jandeleit-Dahm, K. A Causal Link between Oxidative Stress and Inﬂammation in Cardiovascular and Renal Complications of Diabetes. Clin. Sci. 2018, 132, 1811–1836. [CrossRef] 39. Tucker, P.S.; Scanlan, A.T.; Dalbo, V.J. Chronic Kidney Disease Inﬂuences Multiple Systems: Describing the Relationship between Oxidative Stress, Inﬂammation, Kidney Damage, and Concomitant Disease. Oxidative Med. Cell. Longev. 2015, 2015. [CrossRef] 40 Himmelfarb J Linking Oxidative Stress and Inﬂammation in Kidney Disease: Which Is the Chicken and Which Is the Egg? Semin 39. Tucker, P.S.; Scanlan, A.T.; Dalbo, V.J. Chronic Kidney Disease Inﬂuences Multiple Systems: Describing the Relationship between Oxidative Stress, Inﬂammation, Kidney Damage, and Concomitant Disease. Oxidative Med. Cell. Longev. 2015, 2015. [CrossRef] 40. Himmelfarb, J. Linking Oxidative Stress and Inﬂammation in Kidney Disease: Which Is the Chicken and Which Is the Egg? Semin. Dial. 2004, 17, 449–454. [CrossRef] 40. Himmelfarb, J. Linking Oxidative Stress and Inﬂammation in Kidney Disease: Which Is the Chicken and Which Is the Egg? Semin. Dial. 2004, 17, 449–454. [CrossRef] 41. Casanova, A.G.; Hernández-Sánchez, M.T.; Martínez-Salgado, C.; Morales, A.I.; Vicente-Vicente, L.; López-Hernández, F.J. A Meta-Analysis of Preclinical Studies Using Antioxidants for the Prevention of Cisplatin Nephrotoxicity: Implications for Clinical Application. Crit. Rev. Toxicol. 2020, 1–21. [CrossRef] [PubMed] pp a, A.G.; Vicente-Vicente, L.; Hernández-Sánchez, M.T.; Pescador, M.; Prieto, M.; Martínez-Salgado, C.; Mo 42. Casanova, A.G.; Vicente-Vicente, L.; Hernández-Sánchez, M.T.; Pescador, M.; Prieto, M.; Martínez-Salgado, C.; Morales, A.I.; López-Hernández, F.J. Key Role of Oxidative Stress in Animal Models of Aminoglycoside Nephrotoxicity Revealed by a Systematic Analysis of the Antioxidant-to-Nephroprotective Correlation. Toxicology 2017, 385, 10–17. [CrossRef] [PubMed] López-Hernández, F.J. Key Role of Oxidative Stress in Animal Models of Aminoglycoside Nephrotoxicity Revealed by a Systematic Analysis of the Antioxidant-to-Nephroprotective Correlation. Toxicology 2017, 385, 10–17. References Yuan, Y.; Wang, H.; Wu, Y.; Zhang, B.; Wang, N.; Mao, H.; Xing, C. P53 Contributes to Cisplatin Induced Renal Oxidative Damage via Regulating P66shc and MnSOD. Cell. Physiol. Biochem. 2015, 37, 1240–1256. [CrossRef] [PubMed] 24. Nishikawa, M.; Nagatomi, H.; Nishijima, M.; Ohira, G.; Chang, B.J.; Sato, E.; Inoue, M. Targeting Superoxide Dismutase to Renal Proximal Tubule Cells Inhibits Nephrotoxicity of Cisplatin and Increases the Survival of Cancer-Bearing Mice. Cancer Lett. 2001, 171, 133–138. [CrossRef] 16 of 21 16 of 21 Antioxidants 2021, 10, 1355 25. Kong, M.J.; Han, S.J.; Kim, J.I.; Park, J.W.; Park, K.M. Mitochondrial NADP+-Dependent Isocitrate Dehydrogenase Deﬁciency Increases Cisplatin-Induced Oxidative Damage in the Kidney Tubule Cells Article. Cell Death Dis. 2018, 9, 488. [CrossRef] 26. Ott, M.; Gogvadze, V.; Orrenius, S.; Zhivotovsky, B. Mitochondria, Oxidative Stress and Cell Death. Apoptosis 2007, 12, 913–922. [CrossRef] 25. Kong, M.J.; Han, S.J.; Kim, J.I.; Park, J.W.; Park, K.M. Mitochondrial NADP+-Dependent Isocitrate Dehydrogenase Deﬁciency Increases Cisplatin-Induced Oxidative Damage in the Kidney Tubule Cells Article. Cell Death Dis. 2018, 9, 488. [CrossRef] p g y 26. Ott, M.; Gogvadze, V.; Orrenius, S.; Zhivotovsky, B. Mitochondria, Oxidative Stress and Cell Death. Apoptosis 2007, 12, 913–922. [CrossRef] [ ] 27. Ryter, S.W.; Hong, P.K.; Hoetzel, A.; Park, J.W.; Nakahira, K.; Wang, X.; Choi, A.M.K. Mechanisms of Cell Death in Oxidative Stress. Antioxid. Redox Signal. 2007, 9, 49–89. [CrossRef] [PubMed] g , S.; Leung, N. Cisplatin Nephrotoxicity: A Review of the Literature. J. Nephrol. 2018, 31, 15–25. [CrossRef] g 28. Manohar, S.; Leung, N. Cisplatin Nephrotoxicity: A Review of the Literature. J. Nephrol. 2018, 31, 1 29. Xiao, T.; Choudhary, S.; Zhang, W.; Ansari, N.H.; Salahudeen, A. Possible Involvement of Oxidative Stress in Cisplatin-Induced Apoptosis in LLC-PK1 Cells. J. Toxicol. Environ. Health Part A 2003, 66, 469–479. [CrossRef] 30. Jiang, M.; Wei, Q.; Pabla, N.; Dong, G.; Wang, C.Y.; Yang, T.; Smith, S.B.; Dong, Z. Effects of Hydroxyl Radical Scavenging on Cisplatin-Induced P53 Activation, Tubular Cell Apoptosis and Nephrotoxicity. Biochem. Pharmacol. 2007, 73, 1499–1510. [CrossRef] [PubMed] 31. Araujo, M.; Wilcox, C.S. Oxidative Stress in Hypertension: Role of the Kidney. Antioxid. Redox Signal. 2014, 20, 74–101. [CrossRef] [PubMed] 32. Wilcox, C.S. Redox Regulation of the Afferent Arteriole and Tubuloglomerular Feedback. Acta Physiol. Scand. 2003, 179, 217–223. [CrossRef] 33. Schnackenberg, C.G. Physiological and Pathophysiological Roles of Oxygen Radicals in the Renal Microvasculature. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2002, 282, R335–R342. p y g 53. Alibakhshi, T.; Khodayar, M.J.; Khorsandi, L.; Rashno, M.; Zeidooni, L. Protective Effects of Zingerone on Oxidative Stress and Inﬂammation in Cisplatin-Induced Rat Nephrotoxicity. Biomed. Pharm. 2018, 105, 225–232. [CrossRef] References [CrossRef] [PubMed] y y p p gy 43. Begg, C.B.; Mazumdar, M. Operating Characteristics of a Rank Correlation Test for Publication Bias. Biometrics 1994, 50, 1088–1101. [CrossRef] [PubMed] 44. Egger, M.; Smith, G.D.; Schneider, M.; Minder, C. Bias in Meta-Analysis Detected by a Simple, Graphical Test. BMJ 1997, 315, 629–634. [CrossRef] 45. Kutner, M.H.; Nachtsheim, C.J.; Neter, J.; Li, W. Applied Linear Statistical Models; McGraw-Hill Irwin Volume 5. 46. Fischler, M.A.; Bolles, R.C. Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Commun. ACM 1981, 24, 381–395. [CrossRef] g y 47. Huber, P.J. Robust estimation of a location parameter. In Breakthroughs in Statistics; Springer: New York, NY, USA, 1992; pp. 492–518. pp 48. Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V. Scikit-Learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. 49. Python Language Reference; Python Software Foundation: Wilmington, DE, USA, 2013. 50. Abdel-Aziz, A.M.; Ibrahim, M.A.; El-Shiekh, A.A.; Osman, N.A.T.; Geddawy, A.; Abdelrahman, A. Prophylactic Effect of Diacerein against Cisplatin-Induced Nephrotoxicity in Rats. Int. J. Pharmacol. 2018, 14, 384–390. [CrossRef] g p p y J , , [ ] 51. Abdel Moneim, A.E.; Othman, M.S.; Aref, A.M. Azadirachta Indica Attenuates Cisplatin-Induced Nephrotoxicity and Oxidative Stress. Biomed. Res. Int. 2014, 2014, 647131. [CrossRef] 51. Abdel Moneim, A.E.; Othman, M.S.; Aref, A.M. Azadirachta Indica Attenuates Cisplatin-Induced Neph Stress. Biomed. Res. Int. 2014, 2014, 647131. [CrossRef] .; Moussa, F.I.; Saad, N.A. Synergistic Protective Effect of N-Acetylcysteine and Taurine against Cisplatin icity in Rats Drug Des Dev 2017 11 901–908 [CrossRef] 52. Abdel-Wahab, W.M.; Moussa, F.I.; Saad, N.A. Synergistic Protective Effect of N-Acetylcysteine and Taurine against Cisplatin- Induced Nephrotoxicity in Rats. Drug Des. Dev. 2017, 11, 901–908. [CrossRef] 52. Abdel-Wahab, W.M.; Moussa, F.I.; Saad, N.A. Synergistic Protective Effect of N-A Induced Nephrotoxicity in Rats. Drug Des. Dev. 2017, 11, 901–908. [CrossRef] 53. Alibakhshi, T.; Khodayar, M.J.; Khorsandi, L.; Rashno, M.; Zeidooni, L. Protective Effects of Zingerone on Oxidative Stress and Inﬂammation in Cisplatin-Induced Rat Nephrotoxicity. Biomed. Pharm. 2018, 105, 225–232. [CrossRef] 17 of 21 17 of 21 Antioxidants 2021, 10, 1355 54. Al-Husseiny, F.; Sobh, M.A.; Ashour, R.H.; Foud, S.; Medhat, T.; El-Gilany, A.-H.; Elghannam, D.; Abdel-Ghaffar, H.; Saad, M.-A.; Sobh, M. Amniotic Fluid-Derived Mesenchymal Stem Cells Cut Short the Acuteness of Cisplatin-Induced Nephrotoxicity in Sprague-Dawley Rats. Int. J. References Stem Cells 2016, 9, 70–78. [CrossRef] [PubMed] p g y 55. Amirshahrokhi, K.; Khalili, A.-R. Thalidomide Ameliorates Cisplatin-Induced Nephrotoxicity by Inhib in an Experimental Model. Inﬂammation 2015, 38, 476–484. [CrossRef] in an Experimental Model. Inﬂammation 2015, 38, 476–484. [CrossRef] 56. An, Y.; Xin, H.; Yan, W.; Zhou, X. Amelioration of Cisplatin-Induced Nephrotoxicity by Pravastatin in Mice. Exp. Toxicol. Pathol. 2011, 63, 215–219. [CrossRef] [PubMed] 56. An, Y.; Xin, H.; Yan, W.; Zhou, X. Amelioration of Cisplatin-Induced Nephrotoxicity by Pravastatin in 2011, 63, 215–219. [CrossRef] [PubMed] 57. Badawy, A.M.; El-Naga, R.N.; Gad, A.M.; Tadros, M.G.; Fawzy, H.M. Wogonin Pre-Treatment Attenuates Cisplatin-Induced Nephrotoxicity in Rats: Impact on PPAR-γ, Inﬂammation, Apoptosis and Wnt/β-Catenin Pathway. Chem. Biol. Interact. 2019, 308, 137–146. [CrossRef] 58. Bami, E.; Ozakpınar, O.B.; Ozdemir-Kumral, Z.N.; Köroglu, K.; Ercan, F.; Cirakli, Z.; Sekerler, T.; Izzettin, F.V.; Sancar, M.; Okuyan, B. Protective Effect of Ferulic Acid on Cisplatin Induced Nephrotoxicity in Rats. Environ. Toxicol Pharm. 2017, 54, 105–111. [CrossRef] [PubMed] [CrossRef] [PubMed] 59. Bayomi, H.S.; Elsherbiny, N.M.; El-Gayar, A.M.; Al-Gayyar, M.M.H. Evaluation of Renal Protective Effects of Inhibiting TGF-β Type I Receptor in a Cisplatin-Induced Nephrotoxicity Model. Eur. Cytokine Netw. 2013, 24, 139–147. [CrossRef] 59. Bayomi, H.S.; Elsherbiny, N.M.; El Gayar, A.M.; Al Gayyar, M.M.H. Evaluation of Renal Protective Effects of Inhibiting TGF β Type I Receptor in a Cisplatin-Induced Nephrotoxicity Model. Eur. Cytokine Netw. 2013, 24, 139–147. [CrossRef] Type I Receptor in a Cisplatin-Induced Nephrotoxicity Model. Eur. Cytokine Netw. 2013, 24, 139–147. [CrossRef] 60 B d G A i t i M K idi A Sh i d h A Kh d lh i i M N t ll hi M H H i M I Receptor in a Cisplatin-Induced Nephrotoxicity Model. Eur. Cytokine Netw. 2013, 24, 139–147. [CrossRef] Type I Receptor in a Cisplatin-Induced Nephrotoxicity Model. Eur. Cytokine Netw. 2013, 24, 139–147. [Cro Bazmandegan, G.; Amirteimoury, M.; Kaeidi, A.; Shamsizadeh, A.; Khademalhosseini, M.; Nematollahi, M p p p y y , , [ ] andegan, G.; Amirteimoury, M.; Kaeidi, A.; Shamsizadeh, A.; Khademalhosseini, M.; Nematollahi, M.H.; Ha 60. Bazmandegan, G.; Amirteimoury, M.; Kaeidi, A.; Shamsizadeh, A.; Khademalhosseini, M.; Nematollahi, M.H.; Hassanipour, M.; Fatemi, I. Sumatriptan Ameliorates Renal Injury Induced by Cisplatin in Mice. Iran. J. Basic Med. Sci. 2019, 22, 563–567. [CrossRef] 61. Chen, X.; Wei, W.; Li, Y.; Huang, J.; Ci, X. Hesperetin Relieves Cisplatin-Induced Acute Kidney Injury by Mitigating Oxidative Stress, Inﬂammation and Apoptosis. Chem. Biol. Interact. References 2019, 308, 269–278. [CrossRef] Fatemi, I. Sumatriptan Ameliorates Renal Injury Induced by Cisplatin in Mice. Iran. J. Basic Med. Sci. 2019, 22, 563–567. [CrossRef] 61. Chen, X.; Wei, W.; Li, Y.; Huang, J.; Ci, X. Hesperetin Relieves Cisplatin-Induced Acute Kidney Injury by Mitigating Oxidative Stress, Inﬂammation and Apoptosis. Chem. Biol. Interact. 2019, 308, 269–278. [CrossRef] p p 62. Chirino, Y.I.; Sánchez-González, D.J.; Martínez-Martínez, C.M.; Cruz, C.; Pedraza-Chaverri, J. Protective Effects of Apocynin against Cisplatin-Induced Oxidative Stress and Nephrotoxicity. Toxicology 2008, 245, 18–23. [CrossRef] st Cisplatin-Induced Oxidative Stress and Nephrotoxicity. Toxicology 2008, 245, 18–23. [CrossRef] 63. Darwish, M.A.; Abo-Youssef, A.M.; Khalaf, M.M.; Abo-Saif, A.A.; Saleh, I.G.; Abdelghany, T.M. Vitamin E Mitigates Cisplatin- Induced Nephrotoxicity Due to Reversal of Oxidative/Nitrosative Stress, Suppression of Inﬂammation and Reduction of Total Renal Platinum Accumulation. J. Biochem. Mol. Toxicol. 2017, 31, 1–9. [CrossRef] [PubMed] 64. Dehnamaki, F.; Karimi, A.; Pilevarian, A.A.; Fatemi, I.; Hakimizadeh, E.; Kaeidi, A.; Allahtavakoli, M.; Rahmani, M.R.; Khademalhosseini, M.; Bazmandegan, G. Treatment with Troxerutin Protects against Cisplatin-Induced Kidney Injury in Mice. Acta Chir. Belg. 2019, 119, 31–37. [CrossRef] [PubMed] 65. Divya, M.K.; Lincy, L.; Raghavamenon, A.C.; Babu, T.D. Ameliorative Effect of Apodytes Dim Nephrotoxicity in Wistar Rats. Pharm. Biol. 2016, 54, 2149–2157. [CrossRef] 66. Elhusseini, F.M.; Saad, M.-A.A.A.; Anber, N.; Elghannam, D.; Sobh, M.-A.; Alsayed, A.; El-Dusoky, S.; Sheashaa, H.; Abdel- Ghaffar, H.; Sobh, M. Long Term Study of Protective Mechanisms of Human Adipose Derived Mesenchymal Stem Cells on Cisplatin Induced Kidney Injury in Sprague-Dawley Rats. J. Stem Cells Regen. Med. 2016, 12, 36–48. [PubMed] p y j y p g y g 67. El-Naga, R.N. Pre-Treatment with Cardamonin Protects against Cisplatin-Induced Nephrotoxicity in Inﬂammation and Apoptosis. Toxicol. Appl. Pharm. 2014, 274, 87–95. [CrossRef] 68. Elsherbiny, N.M.; Eladl, M.A.; Al-Gayyar, M.M.H. Renal Protective Effects of Arjunolic Acid in a Cisplatin-Induced Nephrotoxicity Model. Cytokine 2016, 77, 26–34. [CrossRef] [PubMed] y 69. Fatima, S.; Al-Mohaimeed, N.; Al-Shaikh, Y.; Tyagi, P.; Banu, N.; Hasan, S.; Arjumand, S. Combined Treatment of Epigallocatechin Gallate and Coenzyme Q10 Attenuates Cisplatin-Induced Nephrotoxicity via Suppression of Oxidative/Nitrosative Stress, Inﬂammation and Cellular Damage. Food Chem. Toxicol. 2016, 94, 213–220. [CrossRef] [PubMed] 70. Fernández-Rojas, B.; Medina-Campos, O.N.; Hernández-Pando, R.; Negrette-Guzmán, M.; Huerta-Yepez, S.; Pedraza-Chaverri, J. C-Phycocyanin Prevents Cisplatin-Induced Nephrotoxicity through Inhibition of Oxidative Stress. Food Funct 2014, 5, 480–490. [CrossRef] 71. Hassan, H.A.; Edrees, G.M.; El-Gamel, E.M.; El-Sayed, E.A. y 78. Kang, L.; Zhao, H.; Chen, C.; Zhang, X.; Xu, M.; Duan, H. Sappanone A Protects Mice against Cisplatin-Induced Kidney Injury. Int. Immunopharmacol. 2016, 38, 246–251. [CrossRef] References Amelioration of Cisplatin-Induced Neph Extract and Fish Oil Is Mediated by Lowering Oxidative Stress and DNA Damage. Cytotechnology 2014 71. Hassan, H.A.; Edrees, G.M.; El-Gamel, E.M.; El-Sayed, E.A. Amelioration of Cisplatin-Induced Nephrotoxicity by Grape Seed Extract and Fish Oil Is Mediated by Lowering Oxidative Stress and DNA Damage. Cytotechnology 2014, 66, 419–429. [CrossRef] 72. Helmy, M.M.; Helmy, M.W.; Abd Allah, D.M.; Abo Zaid, A.M.; Mohy El-Din, M.M. Selective ETA Receptor Blockade Protects 72. Helmy, M.M.; Helmy, M.W.; Abd Allah, D.M.; Abo Zaid, A.M.; Mohy El-Din, M.M. Selective ETA Re against Cisplatin-Induced Acute Renal Failure in Male Rats. Eur. J. Pharmacol. 2014, 730, 133–139. [Cro g p 73. Hosseini, A.; Fanoudi, S.; Mollazadeh, H.; Aghaei, A.; Boroushaki, M.T. Protective Effect of Rheum Turkestanicum against Cisplatin by Reducing Oxidative Stress in Kidney Tissue. J. Pharm. Bioallied Sci. 2018, 10, 66–71. [CrossRef] 74. Huang, Y.-C.; Tsai, M.-S.; Hsieh, P.-C.; Shih, J.-H.; Wang, T.-S.; Wang, Y.-C.; Lin, T.-H.; Wang, S.-H. Galangin Ameliorates Cisplatin-Induced Nephrotoxicity by Attenuating Oxidative Stress, Inﬂammation and Cell Death in Mice through Inhibition of ERK and NF-KappaB Signaling. Toxicol. Appl. Pharm. 2017, 329, 128–139. [CrossRef] pp g g pp 75. Huang, H.; Shen, Z.; Geng, Q.; Wu, Z.; Shi, P.; Miao, X. Protective Effect of Schisandra Chinensis Bee Pollen Extract on Liver and Kidney Injury Induced by Cisplatin in Rats. Biomed. Pharm. 2017, 95, 1765–1776. [CrossRef] 76. Huang, S.; You, J.; Wang, K.; Li, Y.; Zhang, Y.; Wei, H.; Liang, X.; Liu, Y. N-Acetylcysteine Attenuates Cisplatin-Induced Acute Kid I j b I hibi i h C5 R Bi d R I t 2019 2019 4805853 [C R f] [P bM d] pp g g pp 75. Huang, H.; Shen, Z.; Geng, Q.; Wu, Z.; Shi, P.; Miao, X. Protective Effect of Schisandra Chinensis Bee Pollen Extract on Liver and Kidney Injury Induced by Cisplatin in Rats. Biomed. Pharm. 2017, 95, 1765–1776. [CrossRef] y j y y p [ ] 76. Huang, S.; You, J.; Wang, K.; Li, Y.; Zhang, Y.; Wei, H.; Liang, X.; Liu, Y. N-Acetylcysteine Attenuates Cisplatin-Induced Acute Kidney Injury by Inhibiting the C5a Receptor. Biomed. Res. Int. 2019, 2019, 4805853. [CrossRef] [PubMed] 77. Kandemir, F.M.; Yildirim, S.; Caglayan, C.; Kucukler, S.; Eser, G. Protective Effects of Zingerone on Cisplatin-Induced Nephrotoxi- city in Female Rats. Environ. Sci. Pollut. Res. Int. 2019, 26, 22562–22574. [CrossRef] [PubMed] 77. Kandemir, F.M.; Yildirim, S.; Caglayan, C.; Kucukler, S.; Eser, G. References [CrossRef] [PubMed] g g y 87. Malik, S.; Bhatia, J.; Suchal, K.; Gamad, N.; Dinda, A.K.; Gupta, Y.K.; Arya, D.S. Nobiletin Ameliorates Cisplatin-Induced Acute Kidney Injury Due to Its Anti-Oxidant, Anti-Inﬂammatory and Anti-Apoptotic Effects. Exp. Toxicol. Pathol. 2015, 67, 427–433. [CrossRef] Mi, X.; Hou, J.; Wang, Z.; Han, Y.; Ren, S.; Hu, J.; Chen, C.; Li, W. The Protective Effects of Maltol o Nephrotoxicity through the AMPK-Mediated PI3K/Akt and P53 Signaling Pathways. Sci. Rep. 2018, 8, 15 88. Mi, X.; Hou, J.; Wang, Z.; Han, Y.; Ren, S.; Hu, J.; Chen, C.; Li, W. The Protective Effects of Maltol on Cisplatin-Induced Nephrotoxicity through the AMPK-Mediated PI3K/Akt and P53 Signaling Pathways. Sci. Rep. 2018, 8, 15922. [CrossRef] p y g g g y p 89. Mohamed, H.E.; El-Swefy, S.E.; Mohamed, R.H.; Ghanim, A.M.H. Effect of Erythropoietin Therapy on Induced Renal Injury in Rats. Exp. Toxicol. Pathol. 2013, 65, 197–203. [CrossRef] 90. Morsy, M.A.; Heeba, G.H. Nebivolol Ameliorates Cisplatin-Induced Nephrotoxicity in Rats. Basic Clin. Pharm. Toxicol. 2016, 118, 449–455. [CrossRef] 91. Mundhe, N.A.; Kumar, P.; Ahmed, S.; Jamdade, V.; Mundhe, S.; Lahkar, M. Nordihydroguaiaretic Acid Ameliorates Cisplatin Induced Nephrotoxicity and Potentiates Its Anti-Tumor Activity in DMBA Induced Breast Cancer in Female Sprague-Dawley Rats. Int. Immunopharmacol. 2015, 28, 634–642. [CrossRef] p [ ] 92. Mundhe, N.; Kumar, P.; Arora, I.; Ahmed, S.; Lahkar, M. Differential Effect of NDGA on Cisplatin-Induced Nephrotoxicity in Spargue-Dawley Rats. Immunopharmacol. Immunotoxicol. 2019, 41, 68–75. [CrossRef] 93. Nazari Soltan Ahmad, S.; Rashtchizadeh, N.; Argani, H.; Roshangar, L.; Ghorbanihaghjo, A.; Sanajou, D.; Panah, F.; Ashraﬁ Jigheh, Z.; Dastmalchi, S.; Kalantary-Charvadeh, A. Tangeretin Protects Renal Tubular Epithelial Cells against Experimental Cisplatin Toxicity. Iran. J. Basic Med. Sci. 2019, 22, 179–186. [CrossRef] 94. Nazari Soltan Ahmad, S.; Rashtchizadeh, N.; Argani, H.; Roshangar, L.; Ghorbani Haghjo, A.; Sanajou, D.; Panah, F.; Ashraﬁ Jigheh, Z.; Dastmalchi, S.; Mesgari-Abbasi, M. Dunnione Protects against Experimental Cisplatin-Induced Nephrotoxicity by Modulating NQO1 and NAD+ Levels. Free Radic. Res. 2018, 52, 808–817. [CrossRef] 95. Neamatallah, T.; El-Shitany, N.A.; Abbas, A.T.; Ali, S.S.; Eid, B.G. Honey Protects against Cisplatin-Induced Hepatic and Renal Toxicity through Inhibition of NF-KB-Mediated COX-2 Expression and the Oxidative Stress Dependent BAX/Bcl-2/Caspase-3 Apoptotic Pathway. Food Funct. 2018, 9, 3743–3754. [CrossRef] p p y 96. Purena, R.; Seth, R.; Bhatt, R. Protective Role of Emblica Ofﬁcinalis Hydro-Ethanolic Leaf Extract in Cisplatin Induced Nephrotox- icity in Rats. Toxicol. Rep. 2018, 5, 270–277. [CrossRef] 97. References Protective Effects of Zingerone on Cisplatin Induced Nephrotoxi city in Female Rats. Environ. Sci. Pollut. Res. Int. 2019, 26, 22562–22574. [CrossRef] [PubMed] y 78. Kang, L.; Zhao, H.; Chen, C.; Zhang, X.; Xu, M.; Duan, H. Sappanone A Protects Mice against Cisplatin-Induced Kidney Injury. Int. Immunopharmacol. 2016, 38, 246–251. [CrossRef] 18 of 21 18 of 21 Antioxidants 2021, 10, 1355 79. Kenza, B.; Djihane, A.; Mouad, B.; Ratiba, M.; Samir, B.; Fadila, B.; Souad, A. Renoprotective Effect of Centaurea Choulettiana Pomel (Asteraceae) Leaves on Cisplatin-Induced Oxidative Stress and Renal Dysfunction in Mice. J. Appl. Pharm. Sci. 2017, 7, 147–154. 80. Khairnar, S.I.; Mahajan, U.B.; Patil, K.R.; Patel, H.M.; Shinde, S.D.; Goyal, S.N.; Belemkar, S.; Ojha, S.; Patil, C.R. Disulﬁram and Its Copper Chelate Attenuate Cisplatin-Induced Acute Nephrotoxicity in Rats Via Reduction of Oxidative Stress and Inﬂammation. Biol. Trace Elem. Res. 2020, 193, 174–184. [CrossRef] , , [ ] 81. Kim, J.-Y.; Park, J.-H.; Kim, K.; Jo, J.; Leem, J.; Park, K.-K. Pharmacological Inhibition of Caspase-1 Ameliorates Cisplatin-Induced Nephrotoxicity through Suppression of Apoptosis, Oxidative Stress, and Inﬂammation in Mice. Mediat. Inﬂamm. 2018, 2018, e6571676. [CrossRef] , F.; Yao, Y.; Huang, H.; Hao, H.; Ying, M. Xanthohumol Attenuates Cisplatin-Induced Nephrotoxicity th d f l h h l [ f] 82. Li, F.; Yao, Y.; Huang, H.; Hao, H.; Ying, M. Xanthohumol Attenuates Cisplatin-Induced Nephrotoxicity through Inhibiting NF-KB and Activating Nrf2 Signaling Pathways. Int. Immunopharmacol. 2018, 61, 277–282. [CrossRef] F-KB and Activating Nrf2 Signaling Pathways. Int. Immunopharmacol. 2018, 61, 277–282. [CrossRef] 83. Li, Y.-Z.; Ren, S.; Yan, X.-T.; Li, H.-P.; Li, W.; Zheng, B.; Wang, Z.; Liu, Y.-Y. Improvement of Cisplatin-Induced Renal Dysfunction by Schisandra Chinensis Stems via Anti-Inﬂammation and Anti-Apoptosis Effects. J. Ethnopharmacol. 2018, 217, 228–237. [CrossRef] 84. Li, R.-Y.; Zhang, W.-Z.; Yan, X.-T.; Hou, J.-G.; Wang, Z.; Ding, C.-B.; Liu, W.-C.; Zheng, Y.-N.; Chen, C.; Li, Y.-R.; et al. Arginyl- Fructosyl-Glucose, a Major Maillard Reaction Product of Red Ginseng, Attenuates Cisplatin-Induced Acute Kidney Injury by Regulating Nuclear Factor KB and Phosphatidylinositol 3-Kinase/Protein Kinase B Signaling Pathways. J. Agric. Food Chem. 2019, 67, 5754–5763. [CrossRef] [PubMed] 85. Ma, P.; Zhang, S.; Su, X.; Qiu, G.; Wu, Z. Protective Effects of Icariin on Cisplatin-Induced Acute Renal Injury in Mice. Am. J. Transl. Res. 2015, 7, 2105–2114. [PubMed] 86. Ma, X.; Yan, L.; Zhu, Q.; Shao, F. Puerarin Attenuates Cisplatin-Induced Rat Nephrotoxicity: The Invo Signaling Pathway. PLoS ONE 2017, 12, e0171612. References 2010, 11, 621–632. [CrossRef] Mitochondrial Membrane Permeabilization: The Sine qua Non for Cell Death. BioEssays 2006, 28, 253– 119. Armstrong, J.S. Mitochondrial Membrane Permeabilization: The Sine qua Non for Cell Death. Bio [CrossRef] 120. Green, D.R.; Amarante-Mendes, G.P. The Point of No Return: Mitochondria, Caspases, and the Commitment to Cell Death. Results Probl. Cell Differ. 1998, 24, 45–61. 121. Cao, S.S.; Kaufman, R.J. Endoplasmic Reticulum Stress and Oxidative Stress in Cell Fate Decision and Human Disease. Antioxid. Redox Signal. 2014, 21, 396–413. [CrossRef] emer, G. The Pathophysiology of Mitochondrial Cell Death. Science 2004, 305, 626–629. [CrossRef] g 122. Green, D.R.; Kroemer, G. The Pathophysiology of Mitochondrial Cell Death. Science 2004, 305, 626–629 122. Green, D.R.; Kroemer, G. The Pathophysiol 122. Green, D.R.; Kroemer, G. The Pathophysiology of Mitochondrial Cell Death. Science 2004, 305, 626–629. [CrossRef] 123. Marzo, I.; Brenner, C.; Kroemer, G. The Central Role of the Mitochondrial Megachannel in Apoptosis: Evidence Obtained with Intact Cells Isolated Mitochondria and Puriﬁed Protein Complexes Biomed Pharmacother 1998 52 248 251 [CrossRef] 123. Marzo, I.; Brenner, C.; Kroemer, G. The Central Role of the Mitochondrial Megachannel in Apoptosis: Intact Cells, Isolated Mitochondria, and Puriﬁed Protein Complexes. Biomed. Pharmacother. 1998, 52, 24 p 124. Bernardi, P.; Rasola, A.; Forte, M.; Lippe, G. The Mitochondrial Permeability Transition Pore: Channel Formation by F-ATP Synthase, Integration in Signal Transduction, and Role in Pathophysiology. Physiol. Rev. 2015, 95, 1111–1155. [CrossRef] [PubMed] 124. Bernardi, P.; Rasola, A.; Forte, M.; Lippe, G. The Mitochondrial Permeability Transition Pore: Channel Formation by F-ATP Synthase, Integration in Signal Transduction, and Role in Pathophysiology. Physiol. Rev. 2015, 95, 1111–1155. [CrossRef] [PubMed] 125. Crompton, M. The Mitochondrial Permeabitity Transition Pore and Its Role in Cell Death. Biochem. J. 1999, 341, 233–249. [CrossRef] [PubMed] y g g p y gy y 125. Crompton, M. The Mitochondrial Permeabitity Transition Pore and Its Role in Cell Death. Biochem. J. 1999, 341, 233–249. [CrossRef] [PubMed] 126. Cullen, K.J.; Yang, Z.; Schumaker, L.; Guo, Z. Mitochondria as a Critical Target of the Chemotheraputic Agent Cisplatin in Head and Neck Cancer. J. Bioenerg. Biomembr. 2007, 39, 43–50. [CrossRef] [PubMed] 127. Santandreu, F.M.; Roca, P.; Oliver, J. Uncoupling Protein-2 Knockdown Mediates the Cytotoxic Effects of Cisplatin. Free Radic. Biol. Med. 2010, 49, 658–666. [CrossRef] 128. Yang, Z.; Schumaker, L.M.; Egorin, M.J.; Zuhowski, E.G.; Quo, Z.; Cullen, K.J. References Wang, Z.; Li, Y.-F.; Han, X.-Y.; Sun, Y.-S.; Zhang, L.-X.; Liu, W.; Liu, X.-X.; Li, W.; Liu, Y.-Y. Kidney Protec Re and Its Underlying Mechanisms on Cisplatin Induced Kidney Injury Cell Physiol Biochem 2018 4 108. Wang, Z.; Li, Y.-F.; Han, X.-Y.; Sun, Y.-S.; Zhang, L.-X.; Liu, W.; Liu, X.-X.; Li, W.; Liu, Y.-Y. Kidney Protec Re and Its Underlying Mechanisms on Cisplatin-Induced Kidney Injury. Cell. Physiol. Biochem. 2018, 4 Re and Its Underlying Mechanisms on Cisplatin-Induced Kidney Injury. Cell. Physiol. Biochem. 2018, 48, 2 ng Mechanisms on Cisplatin-Induced Kidney Injury. Cell. Physiol. Biochem. 2018, 48, 2219–2229. [CrossRef] 109. Yang, H.; Li, W.; Wang, L.; Li, W.; Sun, H.; He, X.; Zhang, J. The Protective Effects of Sika Deer Antler Protein on Cisplatin-Induced Nephrotoxicity. Cell. Physiol. Biochem. 2017, 43, 395–404. [CrossRef] [PubMed] 110. Yu, X.; Meng, X.; Xu, M.; Zhang, X.; Zhang, Y.; Ding, G.; Huang, S.; Zhang, A.; Jia, Z. Celastrol Ameliorates Cisplatin Nephrotoxicity by Inhibiting NF-KB and Improving Mitochondrial Function. EBioMedicine 2018, 36, 266–280. [CrossRef] [PubMed] 111. Zhang, L.; Gu, Y.; Li, H.; Cao, H.; Liu, B.; Zhang, H.; Shao, F. Daphnetin Protects against Cisplatin-Induced Nephrotoxicity by Inhibiting Inﬂammatory and Oxidative Response. Int. Immunopharmacol. 2018, 65, 402–407. [CrossRef] [PubMed] 112. Burnham, K.; Anderson, D. Model Selection and Multi-Model Inference, 2nd ed.; Springer: New York, NY, USA, 2002; Volume 63, ISBN 978-0-387-95364-9. 113. Nolﬁ-Donegan, D.; Braganza, A.; Shiva, S. Mitochondrial Electron Transport Chain: Oxidative Phosphorylation, Oxidant Production, and Methods of Measurement. Redox Biol. 2020, 37, 101674. [CrossRef] [PubMed] 114. Murphy, M.P. How Mitochondria Produce Reactive Oxygen Species. Biochem. J. 2009, 417, 1–13. [Cross 114. Murphy, M.P. How Mitochondria Produce Reactive Oxygen Species. Biochem. J. 2009, 417, 1–13. [CrossRef] [PubMed] 115. Dan Dunn, J.; Alvarez, L.A.J.; Zhang, X.; Soldati, T. Reactive Oxygen Species and Mitochondria: A Nexus of Cellular Homeostasis. Redox Biol. 2015, 6, 472–485. [CrossRef] 115. Dan Dunn, J.; Alvarez, L.A.J.; Zhang, X.; Soldati, T. Reactive Oxygen Species and Mitochondria: A Nexus of Cellular Homeostasis. Redox Biol. 2015, 6, 472–485. [CrossRef] 116. Forrester, S.J.; Kikuchi, D.S.; Hernandes, M.S.; Xu, Q.; Griendling, K.K. Reactive Oxygen Species in Metabolic and Inﬂammatory Signaling. Circ. Res. 2018, 122, 877–902. [CrossRef] [PubMed] 117. Galluzzi, L.; Kepp, O.; Trojel-Hansen, C.; Kroemer, G. Mitochondrial Control of Cellular Life, Stress, and Death. Circ. Res. 2012, 111, 1198–1207. [CrossRef] 118. Tait, S.W.G.; Green, D.R. Mitochondria and Cell Death: Outer Membrane Permeabilization and Beyond. Nat. Rev. Mol. Cell Biol. 128. Yang, Z.; Schumaker, L.M.; Egorin, M.J.; Zuhowski, E.G.; Quo, Z.; Cullen, K.J. Cisplatin Preferentially Binds Mitochondrial DNA and Voltage-Dependent Anion Channel Protein in the Mitochondrial Membrane of Head and Neck Squamous Cell Carcinoma: Possible Role in Apoptosis. Clin. Cancer Res. 2006, 12, 5817–5825. [CrossRef] References Qi, Z.; Li, Z.; Li, W.; Liu, Y.; Wang, C.; Lin, H.; Liu, J.; Li, P. Pseudoginsengenin DQ Exhibits Therapeutic Effects in Cisplatin- Induced Acute Kidney Injury via Sirt1/NF-KB and Caspase Signaling Pathway without Compromising Its Antitumor Activity in Mice. Molecules 2018, 23, 3038. [CrossRef] 98. Radwan, R.R.; Abdel Fattah, S.M. Mechanisms Involved in the Possible Nephroprotective Effect of Rutin and Low Dose γ Irradiation against Cisplatin-Induced Nephropathy in Rats. J. Photochem. Photobiol. B 2017, 169, 56–62. [CrossRef] 99. Rana, M.A.; Khan, R.A.; Nasiruddin, M.; Khan, A.A. Amelioration of Cisplatin-Induced Nephrotoxicity by Ethanolic Extract of Bauhinia Purpurea: An in Vivo Study in Rats. Saudi J. Kidney Dis. Transpl. 2016, 27, 41–48. [CrossRef] [PubMed] 100. Sahin, K.; Orhan, C.; Tuzcu, M.; Muqbil, I.; Sahin, N.; Gencoglu, H.; Guler, O.; Padhye, S.B.; Sarkar, F.H.; Mohammad, R.M. Comparative in Vivo Evaluations of Curcumin and Its Analog Diﬂuorinated Curcumin against Cisplatin-Induced Nephrotoxicity. Biol. Trace Elem. Res. 2014, 157, 156–163. [CrossRef] [PubMed] 19 of 21 19 of 21 Antioxidants 2021, 10, 1355 101. Saiﬁ, M.A.; Sangomla, S.; Khurana, A.; Godugu, C. Protective Effect of Nanoceria on Cisplatin-Induced Nephrotoxicity by Amelioration of Oxidative Stress and Pro-Inﬂammatory Mechanisms. Biol. Trace Elem. Res. 2019, 189, 145–156. [CrossRef] [PubMed] 102. Sen, S.; Chakraborty, R.; Kalita, P. Dillenia Indica Fruit Prevents Cisplatin-Induced Kidney Injury in Modulation of Oxidative Stress, Marker Enzyme, and Biochemical Changes. Nutrire 2018, 43, 15. [C 103. Sener, M.T.; Sener, E.; Tok, A.; Polat, B.; Cinar, I.; Polat, H.; Akcay, F.; Suleyman, H. Biochemical and H Cisplatin Nephrotoxicity Prevention by Mirtazapine. Pharm. Rep. 2012, 64, 594–602. [CrossRef] 104. Sharma, S.K.; Goyal, N. Protective Effect of Heliotropium Eichwaldi against Cisplatin-Induced Nephrotoxicity in Mice. Zhong Xi Yi Jie He Xue Bao 2012, 10, 555–560. [CrossRef] 105. Sherif, I.O. Amelioration of Cisplatin-Induced Nephrotoxicity in Rats by Triterpenoid Saponin of Terminalia Arjuna. Clin. Exp. Nephrol. 2015, 19, 591–597. [CrossRef] 106. Shi, H.-H.; Wang, C.-C.; Guo, Y.; Xue, C.-H.; Zhang, T.-T.; Wang, Y.-M. DHA-PC Protects Kidneys against Cisplatin-Induced Toxicity and Its Underlying Mechanisms in Mice. Food Funct. 2019, 10, 1571–1581. [CrossRef] 107. Topcu-Tarladacalisir, Y.; Sapmaz-Metin, M.; Karaca, T. Curcumin Counteracts Cisplatin-Induced Nephrotoxicity by Preventing Renal Tubular Cell Apoptosis. Ren. Fail. 2016, 38, 1741–1748. [CrossRef] p p F.; Han, X.-Y.; Sun, Y.-S.; Zhang, L.-X.; Liu, W.; Liu, X.-X.; Li, W.; Liu, Y.-Y. Kidney Protection Effect of Ginsenos 108. 127. Santandreu, F.M.; Roca, P.; Oliver, J. Uncoupling Protein-2 Knockdown Mediates the Cytotoxic Effects of Cisplatin. Free Radic. Biol. Med. 2010, 49, 658–666. [CrossRef] References Cisplatin Preferentially Binds Mitochondrial DNA and Voltage-Dependent Anion Channel Protein in the Mitochondrial Membrane of Head and Neck Squamous Cell Carcinoma: Possible Role in Apoptosis. Clin. Cancer Res. 2006, 12, 5817–5825. [CrossRef] 20 of 21 20 of 21 Antioxidants 2021, 10, 1355 129. Custódio, J.B.A.; Cardoso, C.M.P.; Santos, M.S.; Almeida, L.M.; Vicente, J.A.F.; Fernandes, M.A.S. Cisplatin Impairs Rat Liver Mitochondrial Functions by Inducing Changes on Membrane Ion Permeability: Prevention by Thiol Group Protecting Agents. Toxicology 2009, 259, 18–24. [CrossRef] gy [ ] 130. Larosa, V.; Remacle, C. Insights into the Respiratory Chain and Oxidative Stress. Biosci. Rep. 2018, gy 130. Larosa, V.; Remacle, C. Insights into the Respiratory Chain and Oxidative Stress. Biosci. Rep. 2018, 38, BSR20171492. [CrossRef] 131. Stöckl, P.; Zankl, C.; Hütter, E.; Unterluggauer, H.; Laun, P.; Heeren, G.; Bogengruber, E.; Herndler-Brandstetter, D.; Breitenbach, M.; Jansen-Dürr, P. Partial Uncoupling of Oxidative Phosphorylation Induces Premature Senescence in Human Fibroblasts and Yeast Mother Cells. Free Radic. Biol. Med. 2007, 43, 947–958. [CrossRef] [PubMed] 132. Šileikyt˙e, J.; Forte, M. The Mitochondrial Permeability Transition in Mitochondrial Disorders. Oxidative Med. Cell. Longev. 2019, 2019. [CrossRef] 133. Chernyak, B.V. Redox Regulation of the Mitochondrial Permeability Transition Pore. Biosci. Rep. 1997, 17, 293–302. [CrossRef] 134. Kowaltowski, A.J.; Castilho, R.F.; Vercesi, A.E. Mitochondrial Permeability Transition and Oxidative Stress. FEBS Lett. 2001, 495, 12–15. [CrossRef] 133. Chernyak, B.V. Redox Regulation of the Mitochondrial Permeability Transition Pore. Biosci. Rep. 19 133. Chernyak, B.V. Redox Regulation of the Mitochondrial Permeability Transition Pore. Biosci. Rep. 1997, 17, 293–302. [CrossRef] 134 K lt ki A J C tilh R F V i A E Mit h d i l P bilit T iti d O id ti St FEBS L tt 2001 495 133. Chernyak, B.V. Redox Regulation of the Mitochondrial Permeability Transition Pore. Biosci. Rep. 1997, 17, 293–302. [CrossRef] 134. Kowaltowski, A.J.; Castilho, R.F.; Vercesi, A.E. Mitochondrial Permeability Transition and Oxidative Stress. FEBS Lett. 2001, 495, 134. Kowaltowski, A.J.; Castilho, R.F.; Vercesi, A.E. Mitochondrial Permeability Transition and Oxidativ 12–15. [CrossRef] 135. Batandier, C.; Leverve, X.; Fontaine, E. Opening of the Mitochondrial Permeability Transition Pore Induces Reactive Oxygen Species Production at the Level of the Respiratory Chain Complex I. J. Biol. Chem. 2004, 279, 17197–17204. [CrossRef] 1 Q l C M h M k l d A Lé S F A M P H b k M E d f O d S d M h 136. References Kadikoylu, G.; Bolaman, Z.; Demir, S.; Balkaya, M.; Akalin, N.; Enli, Y. The Effects of Desferrioxamine on Cisplatin-Induced Lipid Peroxidation and the Activities of Antioxidant Enzymes in Rat Kidneys. Hum. Exp. Toxicol. 2004, 23, 29–34. [CrossRef] 144. Kadikoylu, G.; Bolaman, Z.; Demir, S.; Balkaya, M.; Akalin, N.; Enli, Y. The Effects of Desferrioxamine on Cisplatin-Induced Lipid Peroxidation and the Activities of Antioxidant Enzymes in Rat Kidneys. Hum. Exp. Toxicol. 2004, 23, 29–34. [CrossRef] 145. Makris, K.; Spanou, L. Acute Kidney Injury: Deﬁnition, Pathophysiology and Clinical Phenotypes. Clin. Biochem. Rev. 2016, 37, 85–98. [PubMed] y y 145. Makris, K.; Spanou, L. Acute Kidney Injury: Deﬁnition, Pathophysiology and Clinical Phenotypes. Clin. Biochem. Rev. 2016, 37, 85–98. [PubMed] [ ] o, C.; Kellum, J.A.; Haase, M. Subclinical AKI Is Still AKI. Crit. Care 2012, 16, 313. [CrossRef] [PubMed] C.; Kellum, J.A.; Haase, M. Subclinical AKI Is Still AKI 147. Murray, P.T.; Mehta, R.L.; Shaw, A.; Ronco, C.; Endre, Z.; Kellum, J.A.; Chawla, L.S.; Cruz, D.; Ince, C.; Okusa, M.D. Potential Use of Biomarkers in Acute Kidney Injury: Report and Summary of Recommendations from the 10th Acute Dialysis Quality Initiative Consensus Conference. Kidney Int. 2014, 85, 513–521. [CrossRef] 148. Fattah, H.; Layton, A.; Vallon, V. How Do Kidneys Adapt to a Deﬁcit or Loss in Nephron Number? Physiology 2019, 34, 189–197. [CrossRef] [PubMed] 149. Hamzeh, M.; Amiri, F.T.; Beklar, S.Y.; Hosseinimehr, S.J. Nephroprotective Effect of Cerium Oxide Nanoparticles on Cyclophosphamide-Induced Nephrotoxicity via Anti-Apoptotic and Antioxidant Properties in BALB/c Mice. Marmara Pharm. J. 2018, 22. [CrossRef] 150. Hirst, S.M.; Karakoti, A.S.; Tyler, R.D.; Sriranganathan, N.; Seal, S.; Reilly, C.M. Anti-Inﬂammatory Properties of Cerium Oxide Nanoparticles. Small 2009, 5, 2848–2856. [CrossRef] [PubMed] 151. Xu, M.-X.; Zhu, Y.-F.; Chang, H.-F.; Liang, Y. Nanoceria Restrains PM2.5-Induced Metabolic Disorder and Hypothalamus Inﬂammation by Inhibition of Astrocytes Activation Related NF-KB Pathway in Nrf2 Deﬁcient Mice. Free Radic. Biol. Med. 2016, 99, 259–272. [CrossRef] ellomo, R. Erythropoietin (EPO) in Acute Kidney Injury. Ann. Intensive Care 2011, 1, 3. [CrossRef] [PubMed] 152. Moore, E.; Bellomo, R. Erythropoietin (EPO) in Acute Kidney Injury. Ann. Intensive Care 2011, 1, 3. 152. Moore, E.; Bellomo, R. Erythropoietin (EPO) in Acute Kidney Injury. Ann. Intensive Care 2011, 1, 3. [CrossRef] [PubMed] 153. Abe, T.; Isaka, Y.; Imamura, R.; Kakuta, Y.; Okumi, M.; Yazawa, K.; Ichimaru, N.; Tsuda, H.; Nonomura, N.; Takahara, S.; et al. Carbamylated Erythropoietin Ameliorates Cyclosporine Nephropathy without Stimulating Erythropoiesis. Cell Transplant. References Quoilin, C.; Mouithys-Mickalad, A.; Lécart, S.; Fontaine-Aupart, M.P.; Hoebeke, M. Evidence of Oxidative Stress and Mitochon- drial Respiratory Chain Dysfunction in an in Vitro Model of Sepsis-Induced Kidney Injury. Biochim. Biophys. Acta Bioenerg. 2014, 1837, 1790–1800. [CrossRef] 137. Sastre, J.; Pallardö, F.V.; Viña, J. Mitochondrial Oxidative Stress Plays a Key Role in Aging and Apoptosis. Iubmb Life 2000, 49, 427–435. 138. Chang, B.; Nishikawa, M.; Sato, E.; Utsumi, K.; Inoue, M. L-Carnitine Inhibits Cisplatin-Induced Injury of the Kidney and Small Intestine. Arch. Biochem. Biophys. 2002, 405, 55–64. [CrossRef] 139. Murata, T.; Hibasami, H.; Maekawa, S.; Tagawa, T.; Nakashima, K. Preferential Binding of Cisplatin to Mitochondrial DNA and Suppression of ATP Generation in Human Malignant Melanoma Cells. Biochem. Int. 1990, 20, 949–955. [PubMed] 140. Olivero, O.A.; Semino, C.; Kassim, A.; Lopez-Larraza, D.M.; Poirier, M.C. Preferential Binding of Cisplatin to Mitochondrial DNA f Chi H t O C ll M t t R L tt 1995 346 221 230 [C R f] 139. Murata, T.; Hibasami, H.; Maekawa, S.; Tagawa, T.; Nakashima, K. Preferential Binding of Cisplatin to Mitochondrial DNA and Suppression of ATP Generation in Human Malignant Melanoma Cells. Biochem. Int. 1990, 20, 949–955. [PubMed] 140. Olivero, O.A.; Semino, C.; Kassim, A.; Lopez-Larraza, D.M.; Poirier, M.C. Preferential Binding of Cisplatin to Mitochondrial DNA of Chinese Hamster Ovary Cells. Mutat. Res. Lett. 1995, 346, 221–230. [CrossRef] y 141. Greggi Antunes, L.M.; D’arc, C.; Darin, J.; Bianchi, M.D.L.P. Protective Effects of Vitamin C against Cispl city and Lipid Peroxidation in Adult Rats: A Dose-Dependent Study. Pharmacol. Res. 2000, 41, 405–411 141. Greggi Antunes, L.M.; D’arc, C.; Darin, J.; Bianchi, M.D.L.P. Protective Effects of Vitamin C against Cisplatin-Induced Nephrotoxi- city and Lipid Peroxidation in Adult Rats: A Dose-Dependent Study. Pharmacol. Res. 2000, 41, 405–411. [CrossRef] [PubMed] 142. Hannemann, J.; Duwe, J.; Baumann, K. Iron- and Ascorbic Acid-Induced Lipid Peroxidation in Renal Microsomes Isolated from 142. Hannemann, J.; Duwe, J.; Baumann, K. Iron- and Ascorbic Acid-Induced Lipid Peroxidation in Renal M Rats Treated with Platinum Compounds. Cancer Chemother. Pharmacol. 1991, 28, 427–433. [CrossRef] [P Treated with Platinum Compounds. Cancer Chemother. Pharmacol. 1991, 28, 427–433. [CrossRef] [PubMed] Rats Treated with Platinum Compounds. Cancer Chemother. Pharmacol. 1991, 28, 427–433. [CrossRef] [PubMed] 143. Yasuyuki, S.; Takahiro, S.; Yoshio, T. Mechanism of the Increase in Lipid Peroxide Induced by Cisplatin in the Kidneys of Rats. Toxicol. Lett. 1992, 62, 293–300. [CrossRef] 144. References 2012, 21, 571–580. [CrossRef] 154. Pallet, N.; Rabant, M.; Legendre, C.; Martinez, F.; Choukroun, G. The Nephroprotective Properties of Recombinant Human Erythropoietin in Kidney Transplantation: Experimental Facts and Clinical Proofs. Am. J. Transplant. 2012, 12, 3184–3190. [CrossRef] [PubMed] 155. Koca, U.; Süntar, I.P.; Keles, H.; Yesilada, E.; Akkol, E.K. In Vivo Anti-Inﬂammatory and Wound Healing Activities of Centaurea Iberica Trev. Ex Spreng. J. Ethnopharmacol. 2009, 126, 551–556. [CrossRef] [PubMed] 21 of 21 21 of 21 Antioxidants 2021, 10, 1355 156. Hussain, T.; Tan, B.; Yin, Y.; Blachier, F.; Tossou, M.C.B.; Rahu, N. Oxidative Stress and Inﬂammation: What Polyphenols Can Do for Us? Oxidative Med. Cell. Longev. 2016, 2016, e7432797. [CrossRef] [PubMed] g 157. Lugrin, J.; Rosenblatt-Velin, N.; Parapanov, R.; Liaudet, L. The Role of Oxidative Stress during Inﬂammatory Processes. Biol. Chem. 2014, 395, 203–230. [CrossRef] [PubMed] 158. Kannan, K.; Jain, S.K. K. Oxidative Stress and Apoptosis. Pathophysiology 2000, 7, 153–163. [CrossR , ; J , p p p y gy , , [ ] 159. Casanova, A.G.; Hernández-Sánchez, M.T.; López-Hernández, F.J.; Martínez-Salgado, C.; Prieto, M.; Vicente-Vicente, L.; Morales, A.I. Systematic Review and Meta-Analysis of the Efﬁcacy of Clinically Tested Protectants of Cisplatin Nephrotoxicity. Eur. J. Clin. Pharmacol. 2020, 76, 23–33. [CrossRef] [PubMed] 160. Sanchez-Gonzalez, P.D.; Lopez-Hernandez, F.J.; Perez-Barriocanal, F.; Morales, A.I.; Lopez-Novoa, J.M. Quercetin Reduces Cisplatin Nephrotoxicity in Rats without Compromising Its Anti-Tumour Activity. Nephrol. Dial. Transplant. 2011, 26, 3484–3495. [CrossRef] 161. Sánchez-González, P.D.; López-Hernández, F.J.; Dueñas, M.; Prieto, M.; Sánchez-López, E.; Thomale, J.; Ruiz-Ortega, M.; López- Novoa, J.M.; Morales, A.I. Differential Effect of Quercetin on Cisplatin-Induced Toxicity in Kidney and Tumor Tissues. Food Chem. Toxicol. 2017, 107, 226–236. [CrossRef] [PubMed] 162. Muldoon, L.L.; Wu, Y.J.; Pagel, M.A.; Neuwelt, E.A. N-Acetylcysteine Chemoprotection without Decreased Cisplatin Antitumor Efﬁcacy in Pediatric Tumor Models. J. Neuro Oncol. 2015, 121, 433–440. [CrossRef] [PubMed] 163. Hirosawa, A.; Niitani, H.; Hayashibara, K.; Tsuboi, E. Effects of Sodium Thiosulfate in Combination Therapy of Cis- Dichlorodiammineplatinum and Vindesine. Cancer Chemother. Pharmacol. 1989, 23, 255–258. [CrossRef] [PubMed] 164. Visacri, M.B.; Quintanilha, J.C.F.; de Sousa, V.M.; Amaral, L.S.; de F.L. Rosiane, Ambrósio; Calonga, L.; Curi, S.F.B.B.; Mayra, M.F.; Chone, C.T.; Altemani, J.M.C.; et al. Can Acetylcysteine Ameliorate Cisplatin-Induced Toxicities and Oxidative Stress without Decreasing Antitumor Efﬁcacy? A Randomized, Double-Blind, Placebo-Controlled Trial Involving Patients with Head and Neck Cancer. Cancer Med. 2019, 8, 2020–2030. [CrossRef] [PubMed]
https://openalex.org/W2125969590	https://hess.copernicus.org/articles/18/5239/2014/hess-18-5239-2014.pdf	English	null	Assessing winter cover crop nutrient uptake efficiency using a water quality simulation model	Hydrology and earth system sciences	2,014	cc-by	12,164	Correspondence to: I.-Y. Yeo (iyeo@umd.edu) Received: 30 September 2013 – Published in Hydrol. Earth Syst. Sci. Discuss.: 21 November 2013 Revised: 25 September 2014 – Accepted: 14 October 2014 – Published: 16 December 2014 Received: 30 September 2013 – Published in Hydrol. Earth Syst. Sci. Discuss.: 21 November 2013 Revised: 25 September 2014 – Accepted: 14 October 2014 – Published: 16 December 2014 Abstract. Winter cover crops are an effective conservation management practice with potential to improve water quality. and delivery to the waterways. Without winter cover crops, annual nitrate loading from agricultural lands was approx- imately 14 kg ha−1, but decreased to 4.6–10.1 kg ha−1 with cover crops resulting in a reduction rate of 27–67 % at the watershed scale. Rye was the most effective species, with a potential to reduce nitrate leaching by up to 93 % with early planting at the ﬁeld scale. Early planting of cover crops (∼30 days of additional growing days) was crucial, as it low- ered nitrate export by an additional ∼2 kg ha−1 when com- pared to late planting scenarios. The effectiveness of cover cropping increased with increasing extent of cover crop im- plementation. Agricultural ﬁelds with well-drained soils and those that were more frequently used to grow corn had a higher potential for nitrate leaching and export to the wa- terways. This study supports the effective implementation of cover crop programs, in part by helping to target critical pol- lution source areas for cover crop implementation. management practice with potential to improve water quality. Throughout the Chesapeake Bay watershed (CBW), which is located in the mid-Atlantic US, winter cover crop use has been emphasized, and federal and state cost-share programs are available to farmers to subsidize the cost of cover crop establishment. The objective of this study was to assess the long-term effect of planting winter cover crops to improve water quality at the watershed scale (∼50 km2) and to iden- tify critical source areas of high nitrate export. A physically based watershed simulation model, Soil and Water Assess- ment Tool (SWAT), was calibrated and validated using water quality monitoring data to simulate hydrological processes and agricultural nutrient cycling over the period of 1990– 2000. Correspondence to: I.-Y. Yeo (iyeo@umd.edu) To accurately simulate winter cover crop biomass in re- lation to growing conditions, a new approach was developed to further calibrate plant growth parameters that control the leaf area development curve using multitemporal satellite- based measurements of species-speciﬁc winter cover crop performance. Multiple SWAT scenarios were developed to obtain baseline information on nitrate loading without win- ter cover crops and to investigate how nitrate loading could change under different winter cover crop planting scenar- ios, including different species, planting dates, and imple- mentation areas. The simulation results indicate that win- ter cover crops have a negligible impact on the water bud- get but signiﬁcantly reduce nitrate leaching to groundwater 1 Introduction The Chesapeake Bay (CB) is the largest and most produc- tive estuary in the US, supporting more than 3600 species of plants and animals (CEC, 2000). It is an international as well as a national asset. The importance of CB has been recog- nized by its designation as a Ramsar site of international im- portance (Gardner and Davidson, 2011). However, the bay’s Assessing winter cover crop nutrient uptake efﬁciency using a water quality simulation model I.-Y. Yeo1,4, S. Lee1, A. M. Sadeghi2, P. C. Beeson, W. D. Hively3, G. W. McCarty2, and M. W. Lang1 1Department of Geographical Sciences, University of Maryland, College Park, MD 20742, USA 2US Department of Agriculture – Agricultural Research Service, Hydrology and Remote Sensing Laboratory, Beltsville, MD 20705, USA 3U.S. Geological Survey, Eastern Geographic Science Center, Reston, VA 20192, USA 4School of Engineering, The University of Newcastle, Callaghan NSW 2308, Australia Formerly at: US Department of Agriculture – Agricultural Research Service, Hydrology and Remote Sensing Laboratory, Beltsville, MD 20705, USA Correspondence to: I.-Y. Yeo (iyeo@umd.edu) I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency To accurately simulate the growth of winter cover crops and their nutrient uptake and nitrate reduction efﬁciencies, we have developed a new approach to calibrate model parameters that control winter cover crop biomass, resulting in model estimates that closely approximate observed values. This study provided important information for decision making to effectively implement winter cover crop programs and to target critical pollution source areas for future BMP implementation. It is particularly important to implement best management practices (BMPs) on agricultural lands in the coastal plain in order to improve water quality in the Chesapeake Bay. Nitro- gen exports from agricultural lands are signiﬁcantly higher than those for other land uses in the coastal plain of the CBW (Jordan et al., 1997; Fisher et al., 2010; Reckhow et al., 2011). Fisher et al. (2010) discussed that N export in- creases by a factor of ∼10 as agriculture increases from 40 to 90 % of land use within coastal plain watersheds. Jordan et al. (1997) showed that N was exported from cropland at a rate of 18 kg N ha−1 year−1, 7 times higher than the rate from other land uses in the coastal Plain. High nitrate exports from coastal plain watersheds have intensiﬁed CB water quality problems, due in part to short hydraulic distances (Reckhow et al., 2011). The implementation of winter cover crops as a best man- agement practice on agricultural lands has been recognized as one of the most important conservation practices being used in the CBW (Chesapeake Bay Commission, 2004). Winter cover crops can sequester residual N after the harvest of summer crops, reducing nitrate leaching to groundwater and delivery to waterways by surface runoff (Hively et al., 2009), and can also reduce the loss of sediment and phospho- rus from agricultural lands. Therefore, federal and state gov- ernments have established cost-share programs to promote winter cover cropping practices (MDA, 2012). However, the overall efﬁciency of cover crops for reducing nitrate load- ings has not been fully evaluated. The inﬂuence of BMPs, such as winter cover crops, on nitrate ﬂux to streams has not been measured in situ at scales larger than ﬁeld, because of the substantial residence time of leached N in ground- water and the difﬁculty of monitoring over long time peri- ods (McCarty et al., 2008). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5240 ecosystems have been greatly degraded. The Chesapeake Bay watershed (CBW) extends over 165 759 km2 and cov- ers parts of New York, Pennsylvania, Maryland, Delaware, West Virginia, Virginia and the District of Columbia. Nearly 16 million people reside in the CBW, and its population is increasing rapidly, leading to accelerated land use and land cover change. The high ratio of watershed area to estuary water surface (14 : 1) ampliﬁes the inﬂuence of human mod- iﬁcations, and excessive nutrient and sediment runoff has led to eutrophication (Kemp et al., 2005; Cerco and Noel, 2007). High nitrogen (N) input to the bay is the foremost water qual- ity concern (Boesch et al., 2001). In the CBW, groundwater contributes more than half of total annual streamﬂow, and groundwater nitrate loads account for approximately half of the total annual N load of streams entering the bay (Phillips et al., 1999). Nitrate leached to the groundwater has substan- tial residence time on the order of 5–40 years (McCarty et al., 2008; Meals et al., 2010). production and N uptake efﬁciency at the landscape scale. However, the catchment-scale beneﬁts of winter cover crop to improve water quality have not been fully understood. As the nutrient uptake and nitrate reduction efﬁciencies of winter cover crops are primarily dependent upon cover crop biomass (Malhi et al., 2006; Hively et al., 2009), it is cru- cial to simulate plant growth accurately. The accurate sim- ulation of the plant growth would require ﬁeld-based infor- mation and an improved calibration method to carefully ac- count for the climate, soil characteristics, and site-speciﬁc nutrient management. Furthermore, the effectiveness of nu- trient management practices, such as winter cover crops, has not been fully explored for coastal agricultural watersheds in the study region due to the challenge of accurately simu- lating hydrologic and nutrient cycling in lowland areas with high groundwater–surface water interaction (Lee et al., 2000; Sadeghi et al., 2007; Sexton et al., 2010; Lam et al., 2012). This study utilized a physically based watershed model, Soil and Water Assessment Tool (SWAT) (Arnold and Fohrer, 2005), to simulate hydrological processes and nitro- gen cycling for an agricultural watershed in the coastal plain of the CBW. We examined the long-term impact (∼10 years) of winter cover crops on the water budget and nitrate loadings under multiple cover crop implementation scenarios (e.g., species, timing and area planted). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency A few ﬁeld studies have demon- strated cover crop nitrate reduction efﬁciencies at the ﬁeld scale (e.g., Shipley et al., 1992; Staver and Brinsﬁeld, 2000). Hively et al. (2009) used satellite remote sensing images and ﬁeld sampling data to estimate winter cover crop biomass Published by Copernicus Publications on behalf of the European Geosciences Union. Published by Copernicus Publications on behalf of the European Geosciences Union. I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5241 Figure 2. Characteristics of the study site (German Branch wa- tershed): land cover, elevation, and hydrologic soil group. Note: (1) Miscellaneous land cover indicates agricultural lands used for minor crops, vegetables, and fruits; (2) hydrologic soil group (HSG) is characterized as follows: Type A – well drained soils with 7.6–11.4 mm hr−1 (0.3–0.45 inch hr−1) water inﬁltration rate; Type B – moderately well drained soils with 3.8–7.6 mm hr−1 (0.15–0.30 inch hr−1) water inﬁltration rate; Type C – moderately poorly drained soils with 1.3–3.8 mm hr−1 (0.05–0.15 inch hr−1) water inﬁltration rate; Type D – poorly drained soils with 0– 1.3 mm hr−1 (0-00.05 inch hr−1) water inﬁltration rate; (3) the land cover map shown is obtained from 2008 National Cropland Data Layer (NCDL). The time series NCDL maps (not shown here) in- dicate the areas grown with corn/soybean rotation are similar to the areas grown with soybean/corn rotation. Figure 1. Geographical location of the study area (German Branch watershed, with the size of 50 km2). Figure 2. Characteristics of the study site (German Branch wa- tershed): land cover, elevation, and hydrologic soil group. Note: (1) Miscellaneous land cover indicates agricultural lands used for minor crops, vegetables, and fruits; (2) hydrologic soil group (HSG) is characterized as follows: Type A – well drained soils with 7.6–11.4 mm hr−1 (0.3–0.45 inch hr−1) water inﬁltration rate; Type B – moderately well drained soils with 3.8–7.6 mm hr−1 (0.15–0.30 inch hr−1) water inﬁltration rate; Type C – moderately poorly drained soils with 1.3–3.8 mm hr−1 (0.05–0.15 inch hr−1) water inﬁltration rate; Type D – poorly drained soils with 0– 1.3 mm hr−1 (0-00.05 inch hr−1) water inﬁltration rate; (3) the land cover map shown is obtained from 2008 National Cropland Data Layer (NCDL). The time series NCDL maps (not shown here) in- dicate the areas grown with corn/soybean rotation are similar to the areas grown with soybean/corn rotation. Figure 1. Geographical location of the study area (German Branch watershed, with the size of 50 km2). to maintainused in this study. Daily climate records on water quality for the mid-Atlantic region of the US (McCarty et al., 2008). a temperate, humid climate with an average annual precipi- tation of 120 cm year−1 (Ator et al., 2005). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency Precipitation is evenly distributed throughout the year, and approximately 50 % of annual precipitation recharges groundwater or enters streams via surface ﬂow, while the remaining precipitation is lost to the atmosphere via evapotranspiration (Ator et al., 2005). 2.2 SWAT model: model description, data, calibration, and validation. SWAT was used to simulate the effects of winter cover crops on nitrate uptake with multiple cover crop scenarios over the period of 1990–2000. The model simulation was run for the entire watershed (including forested, row croplands, and non-row croplands), and changes in both water budgets and nitrate loads to receiving waters under multiple scenarios were compared with baseline conditions (no cover crops) at the ﬁeld and/or watershed scales. The overall modeling ap- proach is presented in Fig. 3. Since cover crop N reduction efﬁciency is controlled by winter cover crop biomass (Malhi et al., 2006), we developed a new method to calibrate plant growth parameters that control leaf area development to pro- duce simulation outputs close to observed values (discussed in Sect. 2.2.4). The Choptank River watershed has been identiﬁed as an “impaired” water body by the US Environmental Protection Agency (US EPA) under Section 303(d) of the Clean Wa- ter Act due to excessive nutrients and sediments, and nutri- ent runoff from agricultural land has been identiﬁed as the main contributor of water pollution (McCarty et al., 2008). Since 1980, substantial efforts have been made to monitor water quality in the Choptank River watershed to establish baseline information on nutrient loadings from agricultural watersheds. Water quality in the GB watershed was inten- sively monitored between 1990 and 1995 as part of the Tar- geted Watershed project, a multiagency state initiative (Jor- dan et al., 1997; Primrose et al., 1997). In 2004, the Chop- tank River watershed was selected to become part of the US Department of Agriculture (USDA) Conservation Effects Assessment Project (CEAP), which evaluates the effective- ness of various agricultural conservation practices designed 2.1 Description of the study site This study was undertaken in the German Branch (GB) wa- tershed, located within the CBW. The GB is a third-order coastal plain stream, located within the non-tidal zone of the Choptank River basin (Fig. 1). Its drainage area is ap- proximately 50 km2 and its land use is dominated by agri- culture (∼72 %) and forest (∼27 %) (Fig. 2). Agricultural lands are evenly split between corn and soybean cropping. The study site is relatively ﬂat with elevations ranging from 1 to 26 m above sea level. Most of the soils are moderately well-drained (hydrologic soil group (HSG) B) or moder- ately poorly drained (HSG C). Soil groups B and C cover 52 and 35 % of the study area, respectively. Well-drained (HSG A) and poorly drained (HSG D) soils account for less than 1 and 14 %, respectively, of the study area. Figure 2 presents information on land use, hydrologic soil types, and topography of the study site. The area is characterized by www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 2.3 Description of SWAT model SWAT is a continuous, physically based semidistributed wa- tershed process model. SWAT simulation runs on a daily time step. SWAT includes and enhances modeling capabilities of 2.4 Data and input preparation Table 1 presents the list of data and other relevant in- formation used in this study. Daily climate records on precipitation and temperature were obtained from the Na- tional Oceanic Atmospheric Administration (NOAA) Na- tional Climate Data Center (NCDC) (Royal Oak, Station ID: USC00187806). Daily solar radiation, relative humidity, wind speed, and missing precipitation and temperature in- formation were derived using SWAT’s built-in weather gen- erator (Neitsch et al., 2011). Monthly streamﬂow and water quality information over the period of 1990–1995 was ob- tained from Jordan et al. (1997). Annual estimates of nitrate loads by subwatershed areas within GB watershed were pro- vided by Primrose et al. (1997). Figure 3. Schematic diagram of modeling procedure. Note: This shows the overall modeling procedure of the presented study and summarizes what simulation results are compared at the various spatial scales. HLZ (High Loading Zones) refers to those agricul- tural ﬁelds (HRUs) with high nitrate export potential. The geospatial data set needed to run SWAT simulations includes digital elevation models (DEM), hydrologic soil types, and land cover/land use. A lidar-based 2 m DEM, processed to add artiﬁcial drainage ditches by the USDA ARS at Beltsville, Maryland (Lang et al., 2012), was used to extract topographic information. The DEM was used to delineate the drainage area, subdivide the study area into smaller modeling units, and deﬁne the stream network. Soil information was obtained from the Soil Survey Geographi- cal Database (SSURGO) available from the USDA Natural Resources Conservation Service (NRCS). a number of different models previously developed by the USDA Agricultural Research Service (ARS) and the US EPA. Arnold and Fohrer (2005) discuss the capabilities of SWAT in detail. Technical documents on physical processes implemented in SWAT, input requirements, and explanation of output variables are available online (Neitsch et al., 2011). The key physical processes in SWAT relevant to this research are brieﬂy discussed below. A map of land use was prepared based on the com- prehensive analysis of existing land use maps, including the US Geological Survey’s National Land Cover Database of 1992, 2001, and 2006, the USDA National Agriculture Statistics Service (NASS) National Cropland Data Layer (NCDL) of 2002, 2008, 2009, and 2010 (Boryan et al., 2011), and a high-resolution land use map developed from 1998 National Aerial Photography Program (NAPP) digital or- thophoto quad imagery (Sexton et al., 2010). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5242 Figure 3. Schematic diagram of modeling procedure. Note: This shows the overall modeling procedure of the presented study and summarizes what simulation results are compared at the various spatial scales. HLZ (High Loading Zones) refers to those agricul- tural ﬁelds (HRUs) with high nitrate export potential. www.hydrol-earth-syst-sci.net/18/5239/2014/ www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 2.4 Data and input preparation These maps indicated a consistent pattern of land use distribution over the last 2 decades with little change. The spatial distribution of major croplands (e.g., soybean and corns) (Fig. 2) was determined using 2008 NCDL. As the 2-year rotations of corn–soybean or soybean–corn were common practice and agricultural lands were used evenly for both crops, the place- ment of the crop rotations was simpliﬁed to alternate the lo- cations of corn and soybean croplands every year using the 2008 NCDL as a base map. While the placement of crop ro- tations between various years would vary, it was not possible to obtain the spatial distribution of major croplands for each simulation year. In addition, time series cropland patterns ob- served from recent NCDL maps seem to support this gener- alized crop rotation pattern of interchanging the locations of corn and soybean ﬁelds. The main components of SWAT include weather, hydrol- ogy, sedimentation, soil temperature, crop growth, nutrients, pesticide, pathogens, and land management (Neitsch et al., 2011). In SWAT, a watershed is subdivided into smaller spa- tial modeling units, subwatersheds and hydrologic response units (HRUs). A HRU is the smallest spatial unit used for ﬁeld-scale processes within the model. HRU is characterized by homogeneous land cover, soil type, and slope. The over- all hydrologic balance as well as nutrient cycling is simu- lated for each HRU, summed to the subwatershed level, and then routed through stream channels to the watershed out- let. In the SWAT model, a modiﬁcation of the Soil Conser- vation Service (SCS) curve number (CN) method was used to simulate surface runoff for all land cover types including row crops, forests, and non-row croplands. The CN method determines runoff based on land use, the soil’s permeability, and antecedent soil water conditions. The transformation and transport of nitrogen between several organic and inorganic pools are simulated within a HRU as a function of nutrient cycles. Simulated loss of N can occur by surface runoff in solution and by eroded sediment and crop uptake. It can also take place in percolation below the root zone, in lateral sub- surface ﬂow, and by volatilization to the atmosphere. Detailed agronomic management information was col- lected in the ﬁeld, as well as through literature reviews and interviews with farmers and extension agents. Modeled agri- cultural practices and management reﬂects actual practices (i.e., no winter cover crop practice, utilizing conservation Hydrol. Earth Syst. Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 2.4 Data and input preparation Sci., 18, 5239–5253, 2014 www.hydrol-earth-syst-sci.net/18/5239/2014/ I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5 Table 1. List of data used in this study. Data Source Description Year DEM MD-DNR Lidar-based 2 m resolution 2006 USDA-NASS Land use map based on cropland data layers 2008 USGS National Land Cover Database 1992, 2002, 2006 Land use USDA-ARS at Beltsville Land use map developed through on-screen digitizing using National Aerial Photography Program (NAPP) digital orthophoto quad imagery (Sexton et al., 2012) 1998 Soils USDA-NRCS Soil Survey Geographic database 2012 Climate NCDC Daily precipitation and temperature 1990–2010 Streamﬂow Jordan et al. (1997) Monthly streamﬂow 1990–1995 Water quality Winter cover crop Biomass Jordan et al. (1997) Hively et al. (2009) Monthly nitrate Winter cover crop biomass estimated from ﬁeld survey and satellite imageries 1990–1995 2005–2006 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5243 eters were calibrated sequentially in order of their sensitivity as reported by Sexton et al. (2010). The calibration was run in a batch and the model performance statistics (discussed below) were computed for each run. We chose the parameter values that produce the best statistical outputs while meet- ing the model performance criteria as discussed by Moriasi et al. (2007). To assess longer-term effects, the model sim- ulations were performed over the period of 1992–2000. We used ArcSWAT 2009 with the 582 version of the executable ﬁle in the ArcGIS 9.3.1 interface. tillage without irrigation) in the study region during the time of water quality monitoring (Sadeghi et al., 2007), and the guidelines for winter cover crop implementation practices were developed by the Maryland Department of Agriculture (MDA) cover crop program. The GB watershed was subdivided into 29 sub-basins based on tributary drainage areas. Within each sub-basin, the superimposing of similar land uses and soil type generated a total of 402 HRUs with 283 classiﬁed as agricultural HRUs. The average size of HRUs ranged from 0.2 to 118.6 ha, with an average size of 11.8 ha and a standard deviation of 13.0 ha. Accuracy of the model calibration was assessed with three statistical model performance measures: the Nash– Sutcliffe efﬁciency coefﬁcient (NSE), root mean squared er- ror (RMSE)-standard deviation ratio (RSR), and percent bias (PBIAS) (Moriasi et al., 2007). They are deﬁned as follows: www.hydrol-earth-syst-sci.net/18/5239/2014/ 2.5 Calibration and validation of SWAT model Although SWAT simulations were calculated on a daily ba- sis, the calibration and validation were performed using the monthly water quality record available from the monitor- ing station located at the study watershed outlet. The cali- bration was performed manually under the baseline scenario with the 2-year crop rotations, following the standard pro- cedure outlined in the SWAT user’s manual (Winchell et al., 2011). The key parameters and their allowable ranges were identiﬁed using the sensitivity analysis performed by Sex- ton et al. (2010) and previous studies (Table 2). The sim- ulations included a 2-year warm-up period (1990–1991) to establish the initial conditions. Model calibration was done using the next 2 years of water quality records (1992–1993), and the remaining records were used for validation (1994– 1995). This short period of spin up and calibration could limit the model’s capability to capture the effects of interannual variability of weather on streamﬂow and nitrate. The calibra- tion was done as follows. We ﬁrst adjusted the parameters related to the streamﬂow and then for nitrate, by making a small change in their allowable ranges (Table 2). The param- NSE = 1 −   nP i=1 (Oi −Si)2 nP i=1 (Oi −O)2  , (1) (1) RSR = RMSE STDEVobs =   s nP i=1 (Oi −Si)2 s nP i=1 (Oi −O)2   , (2) PBIAS =   nP i=1 (Oi −Si) × 100 nP i=1 Oi  , (3) Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 RSR = RMSE STDEVobs =   s nP i=1 (Oi −Si)2 s nP i=1 (Oi −O)2   , (2) PBIAS =   nP i=1 (Oi −Si) × 100 nP i=1 Oi  , (3) Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 (2) PBIAS =   nP i=1 (Oi −Si) × 100 nP i=1 Oi  , (3) (3) Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 www.hydrol-earth-syst-sci.net/18/5239/2014/ I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5244 Table 2. List of calibrated parameters. Simulation Calibrated Parameter module Description Range value Reference* CN2 Flow Curve number −20 to +20 % −16 % Zhang et al. (2008) ESCO Flow Soil evaporation compensation factor 0–1 1.000 Kang et al. (2006) SURLAG Flow Surface runoff lag coefﬁcient 0–10 1 Zhang et al. Table 2. List of calibrated parameters. Table 2. List of calibrated parameters. 2.5 Calibration and validation of SWAT model (2008) ALPHA_BF Flow Base ﬂow recession constant (1/days) 0–1 0.045 Meng et al. (2010) GW_DELAY Flow Delay time for aquifer recharge (days) 0–50 26 Meng et al. (2010) CH_K2 Flow Effective hydraulic conductivity (mm h−1) 0–150 2 Zhang et al. (2008) CH_N2 Flow Manning coefﬁcient 0.02–0.1 0.038 Meng et al. (2010) NPERCO Nitrogen Nitrogen percolation coefﬁcient 0.01–1 1 Meng et al. (2010) N_UPDIS Nitrogen Nitrogen uptake distribution parameter 5–50 50 Saleh and Du (2004) ANION_EXCL Nitrogen Fraction of porosity from which anions are ex- cluded 0.1–0.7 0.405 Meng et al. (2010) ERORGN Nitrogen Organic N enrichment ratio for loading with sediment 0–5 4.97 Meng et al. (2010) BIOMIX Nitrogen Biological mixing efﬁciency 0.01–1.0 0.01 Chu et al. (2004) LAIMX1 LAI Fraction of the maximum leaf area index corre- sponding to the ﬁrst point on the leaf area de- velopment curve – 0.01 (Wheat) 0.02 (Barley) 0.12 (Rye) Hively et al. (2009) LAIMX2 LAI Fraction of the maximum leaf area index corre- sponding to the second point – 0.14 (Wheat) 0.31 (Barley) 0.35 (Rye) Hively et al. (2009) Note: the ranges of parameters were adapted from existing literature (noted as Reference*). LAIMX1 and LAIMX2 were estimated using the regression method based on biomass estimates reported in Hively et al. (2009) and the simulation outputs from the crop growth module of SWAT (see details in Sect. 2.2.3). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency over the course of the simulation period to avoid days with substantial precipitation falling immediately prior to winter cover planting. Note that the harvest date of summer crops under the baseline was set for 15 October to make the model results from the baseline more comparable to the early and late cover crop scenarios by setting the harvesting date in between them. Actual practices and historical statistics indi- cate that early planting was generally allowed for corn only, as soybean requires later harvest in the Choptank River re- gion. MDA’s county level statistics over 2006–2011 showed that winter cover crops were generally planted later follow- ing soybean (in general, after mid-October), while two-thirds of cover crop implementation occurred prior to mid-October after corn. This difference could be due to late harvesting to allow for double planted soybean crops. In this study, early planting scenarios were considered to be more active con- servative agricultural practices than late planting scenarios. Therefore, early planting scenarios were set to apply the early planting date at 100 % where it could be applicable (i.e., corn ﬁelds), while the remaining ﬁelds (i.e., soybean ﬁelds) were assumed to be treated with 100 % of late plantings. As a result, these scenarios include 50 % of cover cropping with early planting on cornﬁelds and the remaining 50 % with late planting on soybean ﬁelds, as both crop types have roughly an equal share of total croplands. Due to this mixed effect, the nitrate removal efﬁciency by different planting dates could not be fully assessed at the watershed scale, but evaluated at the ﬁeld scale. difference in heat units counted during two different time pe- riods. Heat units are based on the accumulated number of growing days that have a daily temperature above the base temperature. Below the base temperature, no plant growth should occur. Using this information, we then were able to relate simu- lated LAI values to the reported biomass estimates and heat units. These LAI values and the corresponding heat units were then normalized by the maximum LAI and total poten- tial heat units required for plant maturity, and the relation- ship between these two normalized values (fractional LAI and heat units) was ﬁtted using a simple regression model. This ﬁtted model was extrapolated to identify two LAI pa- rameter values (Table 2) required to adjust the leaf area de- velopment curve in the SWAT model. 2.7 Assessing the effectiveness of winter cover crops with multiple scenarios We assessed the potential effects of winter cover crops on nitrate removal at the ﬁeld and watershed scales under multi- ple implementation scenarios. Details of these scenarios are presented in Table 3. The MDA Cover Crop Program offers a varying cost share according to winter cover crop plant- ing species and cutoff planting dates. Following the program guidelines and county-level statistics of winter cover crop implementation (MDA, 2012), we constructed multiple sce- narios relevant to regional cover crop practices with three major cover crop species – i.e., barley (Hordeum vulgare L.), rye (Secale cereale L.), and wheat (Triticum aestivum L.) – and two planting date categories (early/late). Additional cover crop scenarios were developed to assess their effective- ness by varying extent of cover crop implementation. The av- erage nitrate export was assessed at the ﬁeld scale based on the simulation output over the period of 1992–2000 under the baseline scenario (i.e., no cover crop). Then, all agricultural HRUs were sorted by nitrate loading and equally subdivided into ﬁve groups. Each group was then introduced incremen- tally for cover crop implementation, in order from the highest to the lowest nitrate loading. 2.6 Calibration of plant growth parameters where Oi are observed and Si are simulated data, O is ob- served mean values, and n equals the number of observations. The values of those statistical measures were compared to the model evaluation criteria set for various water quality param- eters (Moriasi et al., 2007). Cover crop plant growth parameters were calibrated to more realistically simulate cover crop growth during winter at the ﬁeld scale. Speciﬁcally, we modiﬁed the parameters that control the leaf area development curve using biomass esti- mates provided by Hively et al. (2009). Their study reported landscape-level biomass estimates for three commonly used winter cover crops categorized by various planting dates over the period of 2005–2006 in the Choptank River region. This information was analyzed to associate winter cover crop biomass estimates with heat units. Heat units were com- puted based on the potential heat unit (PHU) theory as im- plemented in SWAT, with the daily climate record over the cover crop monitoring period (2005–2006). The crop growth module of SWAT was then run with average daily climate data over 1992–2000 using the default parameter values to provide estimates of biomass and leaf area index (LAI) by growing degree days. This assumption should not have a sig- niﬁcant effect on plant growth simulation, even if there is some interannual variability in weather conditions between the two periods. This is because the plant growth cycle in SWAT is simulated using heat unit theory, and there was little The prediction uncertainty of the model was assessed us- ing the 95 % prediction uncertainty (95 PPU), the P factor, and the R factor (Singh et al., 2014). They were computed using all simulation outputs obtained during the manual cal- ibration process. The 95 PPU bands are calculated at the 2.5 and 97.5 percentiles of the cumulative distribution of simu- lation outputs. The P factor indicates the percentage of ob- served data falling within 95 PPU band, and the R factor is the average thickness of the 95 PPU bands by the standard deviation of the observed data. The R factor can vary be- tween 0 (i.e., achievement of a small uncertainty bound) and inﬁnity, while the P factor can vary from 0 to 100 % (i.e., all observations bracketed by the prediction uncertainty) (Singh et al., 2014). Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 www.hydrol-earth-syst-sci.net/18/5239/2014/ 5245 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency www.hydrol-earth-syst-sci.net/18/5239/2014/ Table 3. List of cover crop scenarios. Table 3. List of cover crop scenarios. Scenario Cover crop species Planting timing Abbreviations 1 None N/A Baseline 2 Winter wheat Early planting (3 October) WE 3 Barley Early planting (3 October) BE 4 Rye Early planting (3 October) RE 5 Wheat Late planting (1 November) WL 6 Barley Late planting (1 November) BL 7 Rye Late planting (1 November) RL Note: early planting scenarios include 50 % of early planting on corn and 50 % of late planting on soybean. Soybean requires longer growing day, and actual practices and county statistics showed that early planting was generally allowed for corn only. Table 4. Agricultural practices and management scheduling for the baseline and cover crop scenarios. 3.1 SWAT calibration and validation The simulated results of monthly streamﬂows and nitrate were compared with the observed data for both the calibra- tion and validation periods. Table 2 provides the list of the adjusted parameter values after model calibration. Overall, Fig. 4 shows good agreement between measured and simu- lated monthly discharge of streamﬂow and nitrate. It illus- trates the 95 PPU (the shaded region) of the SWAT simu- lation model with the monthly observed and the best sim- ulated streamﬂows and nitrates. The 95 PPU of streamﬂow seems to quantify most uncertainties as the interval includes most of the measured data. However, the 95 PPU of nitrate does not seem to represent all the uncertainty, particularly for the low-ﬂow season when most of the simulated streamﬂows are not in good agreement with the observed streamﬂows. This could be caused by the limitations of SWAT itself and the large errors associated with calibration. The calibration was conducted over a short period and this could limit the capability of the calibrated model to capture the effects of weather variability on streamﬂow and nitrate. In addition, the nitrate load calculated based on the ﬁeld sampling of nitrate stream concentration (i.e., the observed nitrate load) could Table 4 summarizes agricultural practices and scheduling used for different scenarios. There was no difference between baseline and cover crop scenarios during the growing sea- son. The croplands were managed with the typical 2-year corn–soybean or soybean–corn rotation, and fertilizer was only applied to corn cropping in the beginning of the grow- ing season, due to its high demand for nutrients to support growth and yield. Instead of winter fallow, cover crop sce- narios assumed placement of winter cover crops. The cover crops were planted after harvesting of summer crops either in the beginning of October (early planting) or November (late planting), and were chemically killed at the beginning of the following growing season (early April). The speciﬁc dates (3 October and 1 November) of cover crop planting were set according to MDA guidelines, with slight adjustment www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 Table 3. List of cover crop scenarios. Baseline scenario Year Corn–soybean rotation Soybean–corn rotation First year 12 Apr – poultry manure; 4942 kg ha−1 (4413 lb/ac) 20 May – soybean plant: no-till 27 Apr – poultry manure; 2471 kg ha−1 (2206 lb/ac) 15 Oct – soybean harvest 30 April – corn plant: no-till 15 Jun – sidedress 30 % UAN; 112 kg ha−1 (100 lb/ac) 15 Oct – corn harvest Second year 20 May – soybean plant: no-till 12 Apr – poultry manure; 4942 kg ha−1 (4413 lb/ac) 15 Oct – soybean harvest 27 Apr – poultry manure; 2471 kg ha−1 (2206 lb/ac) 30 Apr – corn plant: no-till 15 Jun – sidedress 30 % UAN; 112 kg ha−1 (100 lb/ac) 15 Oct – corn harvest Cover crop scenario Year Corn–soybean rotation Soybean–corn rotation First year 12 Apr – poultry manure; 4942 kg ha−1 (4413 lb/ac) 20 May – soybean plant: no-till 27 Apr – poultry manure; 2471 kg ha−1 (2206 lb/ac) 30 Oct – soybean harvesting 30 Apr – corn plant: no-till 1 Nov – cover crop planting 15 Jun – sidedress 30 % UAN; 112 kg ha−1 (100 lb/ac) 1 & 30 Oct – corn harvesting 3 Oct & 1 Nov – cover crops planting Second year 1 Apr – chemically kill cover crops 1 Apr – chemically kill cover crops 20 May – soybean plant: no-till 12 April – poultry manure; 4942 kg ha−1 (4413 lb/ac) 30 Oct – soybean harvesting 27 April – poultry manure; 2471 kg ha−1 (2206 lb/ac) 1 Nov – cover crop planting 30 April – corn plant: no-till 15 Jun – sidedress 30 % UAN; 112 kg ha−1 (100 lb/ac) 1 & 30 Oct – corn harvesting 3 Oct & 1 Nov – cover crop planting Note: the typical N content for poultry manure is 2.8 % (Glancey et al., 2012). Table 4. Agricultural practices and management scheduling for the baseline and cover crop scenarios. Note: the typical N content for poultry manure is 2.8 % (Glancey et al., 2012). captured if a lower level of prediction interval (e.g., 90 %) is chosen. The nitrate simulation results produced a much smaller P factor value than the streamﬂow, indicating much greater uncertainty. However, the R factor value of nitrate is smaller than that of streamﬂow, indicating the 95 PPU band for the nitrate is narrower (Table 5). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5246 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5247 Table 5. Model performance measures for streamﬂow and nitrate. ce measures for streamﬂow and nitrate. Variable Period RSR NSE P bias (%) P factor R factor Flow Calibration 0.50c 0.74b 7.0c 0.75 0.94 Validation 0.52b 0.72b −2.9c 0.62 0.83 Nitrate Calibration 0.55b 0.68b −3.4c 0.50 0.67 Validation 0.69a 0.50a −15.6c 0.29 0.62 Note: performance rating a indicates satisfactory, b good, c very good. The performance rating criteria are adapted from Moriasi et al. (2009) and these statistics are computed based on the monthly water quality record. Note: performance rating a indicates satisfactory, b good, c very good. The performance rating criteria are adapted from Moriasi et al. (2009) and these statistics are computed based on the monthly water quality record. Figure 4. Observed and simulated monthly streamﬂows and nitrate loads during the monitoring period (1992–1995) at the watershed scale. in 1994. Unlike the simulation output, a high peak in stream- ﬂow and consequently in nitrate loading was observed in Au- gust. This relatively high ﬂow and nitrate were somewhat unusual, as the weather record for this site did not show any dramatic change in precipitation during August of 1994 com- pared to the previous years. However, the reported stream- ﬂow in August of 1994 was much higher than observations from other years. In addition, the streamﬂow record from an adjacent watershed, with similar characteristics and size, did not produce high peak values for streamﬂow during the same period. This difference could perhaps be explained due to un- expected agricultural practices, localized thunderstorms that did not occur at the weather station and nearby watershed, or human/measurement errors, although the exact cause of such error could not be determined. The SWAT simulation provided considerably improved results compared to previ- ous studies conducted in the study area (Lee et al., 2000; Sadeghi et al., 2007; Sexton et al., 2010). These improve- ments may be due to different model choice (Niraula et al., 2013), the recent update of the SWAT model to more accu- rately predict nitrate in groundwater (USDA-ARS, 2012; Seo et al., 2014), and use of more accurate higher spatial resolu- tion DEMs (Chaplot, 2005; Chaubey et al., 2005). Accurate simulation of winter cover crop growth and biomass at various stages of production is crucial to accu- rately estimating the potential of winter cover crop to uptake residual N and reduce nitrate loading. Table 3. List of cover crop scenarios. be overestimated for the low ﬂow season, if it is not based on sufﬁcient coverage and consistency within the data set (e.g., continuous on-site measurements). The P factor values for streamﬂow ranges between 0.62 and 0.75 (as shown in Ta- ble 5), but most observed data outside the 95 PPU are not far off from this shaded region. These values could be well be overestimated for the low ﬂow season, if it is not based on sufﬁcient coverage and consistency within the data set (e.g., continuous on-site measurements). The P factor values for streamﬂow ranges between 0.62 and 0.75 (as shown in Ta- ble 5), but most observed data outside the 95 PPU are not far off from this shaded region. These values could be well www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency Table 5. Model performance measures for streamﬂow and nitrate. Variable Period RSR NSE P bias (%) P factor R factor Flow Calibration 0.50c 0.74b 7.0c 0.75 0.94 Validation 0.52b 0.72b −2.9c 0.62 0.83 Nitrate Calibration 0.55b 0.68b −3.4c 0.50 0.67 Validation 0.69a 0.50a −15.6c 0.29 0.62 Note: performance rating a indicates satisfactory, b good, c very good. The performance rating criteria are adapted from Moriasi et al. (2009) and these statistics are computed based on the monthly water quality record. I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency www.hydrol-earth-syst-sci.net/18/5239/2014/ I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency The winter cover crop program was implemented in 2005 at this site and, there- fore, no data were available to validate predicted winter cover crop biomass over the period of 1992–2000. However, we are conﬁdent in our biomass simulation, as the simulated 8- year averaged winter cover crop biomass estimates obtained at the HRU scale were comparable to the range of cover crop biomass reported by Hively et al. (2009). It is to be noted that without calibration, cover crop growth was simulated at a much faster growth rate, and the growth trend over win- ter months did not match ﬁeld data as reported in Hively et al. (2009). This study calculated above-ground winter cover crop biomass with a range of planting dates, based on ﬁeld survey and satellite images acquired over the period of 2005– 2006. For example, the modeled growth rate of rye before calibration was substantially lower in the early growth stage, Figure 4. Observed and simulated monthly streamﬂows and nitrate loads during the monitoring period (1992–1995) at the watershed scale. Table 5 also presents a summary of model performance measures and their accuracy ratings based on the statisti- cal evaluation guidelines reported by Moriasi et al. (2007). These performance measures are calculated based on a monthly water quality record. Overall, the model perfor- mance rating for streamﬂow and nitrate loads exceeded the “satisfactory” rating in both the calibration and validation periods. Model simulation results for streamﬂow were more congruent with the observed values than for nitrate, but the pattern of simulated nitrate was similar to the trend of simu- lated streamﬂow. Also, simulation results for the calibration period were in better agreement with the observed values, compared to the validation period. The largest discrepancy between simulated and measured streamﬂow and nitrate was www.hydrol-earth-syst-sci.net/18/5239/2014/ 3.2 Multiple scenarios analysis y p p g The simulation results indicate that rye is the most effec- tive cover crop at reducing nitrate loads. Rye is well adapted for use as a winter cover crop due to its rapid growth and win- ter hardiness, and these characteristics enabled rye to con- sume a larger amount of excessive nitrogen than other crops (Shipley et al., 1992; Clark, 2007; Hively et al., 2009). Bar- ley is a cool-season crop and develops a strong root system during the winter season. Barley exhibits better nutrient up- take capacity than wheat (Malhi et al., 2006; Clark, 2007). Our simulation results were consistent with previous studies. As shown in Fig. 5, rye grows faster than other winter cover crops particularly in the early growth stage, taking up higher levels of nitrate. Compared to the baseline scenario, rye re- moved more than 67 % of nitrate with early planting, and 54 % with late plating (Fig. 6). Barley had a nitrate reduction rate of 57 % and winter wheat 41 % with early planting, but this removal efﬁciency drops to 38 % for barley and 27 % for winter wheat with late planting (Fig. 6). Figure 6 illustrates that late planted rye was nearly as effective as early planted barley and more effective than early planted winter wheat. Winter cover crops had little impact on catchment hydrology but a profound effect on nitrate exports. Figure 6 presents 9-year average annual mean streamﬂow, annual evapotran- spiration, and annual nitrate loads, under baseline and mul- tiple cover crop scenarios. As reported from previous stud- ies (Kaspar et al., 2007; Islam et al., 2006), the inclusion of a winter cover crop reduced streamﬂows only slightly (< 10 %). Similarly, our study found streamﬂow reductions of less than 8 %. Winter cover cropping reduced stream- ﬂow from 8.5 to 7.8 m3 s−1 (RE, rye early) and 8.4 m3 s−1 (WL, wheat late), and increased evapotranspiration from 667 to 673 mm (WL) and 710 mm (RE), in comparison to the baseline scenario. While the effects of winter vegetation on evapotranspiration were relatively low, any water loss due to evapotranspiration could be offset as cover cropping usually increases soil saturation by increasing water inﬁltration ca- pacity (Dabney, 1998; Islam et al., 2006). 3.2 Multiple scenarios analysis Because the study site typically exhibits maximum streamﬂow during winter with rising groundwater levels (Fisher et al., 2010), the rel- ative difference in streamﬂows due to winter cover crops re- mained small. Rye cover crops caused the most changes to the hydrologic budget followed by barley and winter wheat cover crops. Early planting scenarios produced slightly lower streamﬂow and higher evapotranspiration, compared to those with the later planting date. y y p Simulated nitrate removal efﬁciency was greatly affected by different levels of cover crop implementation as shown in Fig. 7. As expected, removal efﬁciency increased with in- creasing coverage of cover crop implementation, though the slope of removal efﬁciency slightly decreased at the 60 % extent. This ﬁnding seems to indicate that the nitrate reduc- tion rate does not increase linearly with increasing coverage, but its relative efﬁciency could decrease after the coverage of cover crop implementation exceeds 50 % of the croplands. While this ﬁnding seems to be reasonable, further ﬁeld-based studies are needed to verify this ﬁnding. It was noted that 60 % cover crop coverage with an early planting date would reduce more nitrate than 100 % cover crop coverage with late planting, emphasizing the importance of early cover crop planting as indicated by other studies (Ritter et al., 1998; Hively et al., 2009). Unlike its small hydrologic effect, winter cover cropping greatly reduced nitrate loads and there were large differences in nitrate loads by planting species and dates. Annual ni- trate loads with cover crop scenarios ranged from 4.6 (RE) to 10.1 kg ha−1 (WL). The difference in nitrate loadings un- der different cover crop scenarios ranged from 1.3 (when RE was compared to BE, barley early) to 5.5 kg ha−1 (when RE was compared to WL). If the comparison of the removal efﬁ- ciency was made within species, early cover cropping (3 Oc- tober) lowered annual nitrate loads by 1.8 (rye and winter wheat) to 2.7 (barley) kg ha−1, compared to late cover crop- ping (1 November). When compared with the baseline sce- nario (13.9 kg ha−1), the cover crop scenarios reduced nitrate loads by 27 (WL)–67 % (RE) at the watershed scale. This ﬁnding compared well with the results of previous studies that reported the importance of early planting date (Ritter et al., 1998; Feyereisen et al., 2006; Hively et al., 2009). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5248 producing much less biomass than observed values. Fig- ure 5 shows the agreement between measured and simulated biomass estimates after calibration, at the ﬁeld (HRU) scale. Note that the simulated estimates of cover crop biomass were at the upper end of the reported values, as the simulation out- put included both above- and below-ground biomass. seasons and warmer conditions (Baggs et al., 2000). Similar research in Minnesota also demonstrated that winter cover crops planted 45 days earlier reduced 6.5 kg N ha−1 more ni- trogen than late planting (Feyereisen et al., 2006). Our simu- lation results are slightly lower than these published values, due to fewer growing days (∼30 days). The earlier planting occurred ∼30 days prior to the late planting. www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 3.2 Multiple scenarios analysis Shorter day lengths and lower temperatures could also limit the growth of cover crop biomass during the winter sea- son. Therefore, earlier planting could increase the amount of nitrogen uptake by cover crops because of longer growing The effects of cover cropping were further assessed by quantifying the amount of nitrate transported from agricul- tural ﬁelds by different delivery pathways to waterways (sur- face runoff, lateral ﬂow, and shallow groundwater) and ni- trate leached to deep groundwater. Figure 8 presents nitrate loads per unit area leaving agricultural ﬁelds during the win- ter fallow period (October–March). The effectiveness of win- ter cover cropping to reduce nitrate leaching is particularly noticeable, as reported by earlier studies (McCraacken et al., 1994; Brandi-Dohrn et al., 1997; Francis et al., 1998; Bergstrom and Jokela, 2001; Rinnofner et al., 2008). At the ﬁeld scale, the seasonal average of nitrate leaching (shown www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5249 Figure 5. Estimation of winter cover crop biomass during the winter fallow period. Note: This ﬁgure presents monthly average total biomass (both above- and below-ground biomass) over the simulation period for three planting species obtained at the ﬁeld (HRU) scale. The vertical dotted line represents the range of above-ground biomass estimates due to different growing/planting days from Hively et al. (2009). The simulated total biomass lies at the upper end of above ground biomass estimates. Figure 5. Estimation of winter cover crop biomass during the winter fallow period. Note: This ﬁgure presents monthly average total biomass (both above- and below-ground biomass) over the simulation period for three planting species obtained at the ﬁeld (HRU) scale. The vertical dotted line represents the range of above-ground biomass estimates due to different growing/planting days from Hively et al. (2009). The simulated total biomass lies at the upper end of above ground biomass estimates. Figure 6. The 9-year average streamﬂow, actual evapotranspira- tion (ET), and nitrate loads at watershed scale under multiple cover crop scenarios. Note: Error bar (vertical line) represents standard deviation. The numeric value in parentheses, (), indicates reduc- tion rate (RR). RR is calculated by taking the relative difference in simulation outputs from the baseline and cover crop scenarios [RR = (Baseline −Cover crop Scenario) / Baseline]. Figure 8. 3.2 Multiple scenarios analysis The 8-year average nitrate leaching and delivery to wa- terways during winter fallow assessed at the ﬁeld scale under multi- ple cover crop scenarios. Note: DPs (Direct pathways) refers to the amount of nitrate transported from agricultural ﬁelds (HRUs) to wa- terways by surface ﬂow, lateral ﬂow, and groundwater; L is nitrate leaching to groundwater. The numeric value in parentheses, (), indi- cates reduction rate (RR). As the growth period of winter cover crop covers from October to March, results presented here were based on the eight years of simulation from October 1992 to March 2000. Figure 8. The 8-year average nitrate leaching and delivery to wa- terways during winter fallow assessed at the ﬁeld scale under multi- ple cover crop scenarios. Note: DPs (Direct pathways) refers to the amount of nitrate transported from agricultural ﬁelds (HRUs) to wa- terways by surface ﬂow, lateral ﬂow, and groundwater; L is nitrate leaching to groundwater. The numeric value in parentheses, (), indi- cates reduction rate (RR). As the growth period of winter cover crop covers from October to March, results presented here were based on the eight years of simulation from October 1992 to March 2000. Figure 6. The 9-year average streamﬂow, actual evapotranspira- tion (ET), and nitrate loads at watershed scale under multiple cover crop scenarios. Note: Error bar (vertical line) represents standard deviation. The numeric value in parentheses, (), indicates reduc- tion rate (RR). RR is calculated by taking the relative difference in simulation outputs from the baseline and cover crop scenarios [RR = (Baseline −Cover crop Scenario) / Baseline]. shallow groundwater (referred to as DPs, direct pathways, in Fig. 8) was greatly reduced from 2.9 to 10.7 kg ha−1 with cover crop scenarios, a reduction rate of 25–80 %. Similar to the watershed-scale analysis, rye with an early planting date produced the most effective result at the ﬁeld scale with the highest reduction rate both through direct pathways and leaching. Figure 7. Nitrate reduction rates by varying degree of cover crop implementation at the ﬁeld scale. 3.3 Geospatial analysis to identify high nitrate loading areas The 9-year annual and monthly nitrate loads from agricul- tural ﬁelds (HRU) simulated under the baseline scenario were analyzed to pinpoint those areas with a high poten- tial for nitrate loadings and better understand the character- istics and variability of these high loading zones. We clas- siﬁed all agricultural HRUs into ﬁve classes according to different levels of nitrate export potential. Nitrate export po- tential was computed by summing up nitrate transported by direct pathways and leaching to groundwater. We observed consistent spatial patterns in nitrate loadings at the inter- annual and monthly timescale. Figure 9 illustrates the ge- ographical distribution of nutrient loadings from all agri- cultural HRUs based on the 9-year annual and monthly Figure 7. Nitrate reduction rates by varying degree of cover crop implementation at the ﬁeld scale. as “L” in Fig. 8) over the winter fallow period (October– March) without cover crops was estimated as 43 kg ha−1. With winter cover crops, nitrate leaching decreased to 3.0– 32.0 kg ha−1, depending on planting species and timing, re- sulting in a reduction rate of 26–93 %, compared to base- line values. In addition, the amount of nitrate transported from ﬁelds to waterways by surface runoff, lateral ﬂow, or www.hydrol-earth-syst-sci.net/18/5239/2014/ I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5250 Figure 9. The spatial distribution of nitrate export potential from agricultural ﬁelds. Note: Nitrate export potential was computed by adding the annual or monthly averaged amount of nitrate leaching to the groundwater (L) and leaving to the streams by surface runoff, lateral ﬂow, and groundwater (DPs) from the 9-year simulation results. Estimated nitrate loads from the HRUs were classiﬁed into ﬁve groups. In the legend M. High refers to Moderately High and M. Low Moderately Low. The HRUs within the black circle indicates outliers with extremely high nitrate loadings. This area is characterized by poorly drained hydric soil (“Urban land”) and consistently produces extremely high nitrate loadings throughout years and seasons. The white area is non-agricultural land as shown in Fig. 2. Figure 9. The spatial distribution of nitrate export potential from agricultural ﬁelds. Note: Nitrate export potential was computed by adding the annual or monthly averaged amount of nitrate leaching to the groundwater (L) and leaving to the streams by surface runoff, lateral ﬂow, and groundwater (DPs) from the 9-year simulation results. Estimated nitrate loads from the HRUs were classiﬁed into ﬁve groups. In the legend M. High refers to Moderately High and M. Low Moderately Low. The HRUs within the black circle indicates outliers with extremely high nitrate loadings. This area is characterized by poorly drained hydric soil (“Urban land”) and consistently produces extremely high nitrate loadings throughout years and seasons. The white area is non-agricultural land as shown in Fig. 2. that nitrate exports vary based on edaphic and agronomic characteristics of the croplands upon which crops are planted. Therefore, it is important to develop management guidelines to encourage optimal planting species, timing, and locations to achieve enhanced water quality beneﬁts. This study suggests that early planted rye is the most effective cover crop practice, with the potential to reduce nitrate load- ing by 67 % over the baseline at the watershed scale. We hy- pothesize that the relatively high nitrate removal efﬁciency of early planted rye is due to the more rapid growth rate of rye, especially in the early growth stage, compared to other species. As expected, nitrate removal efﬁciency increased signiﬁcantly with early planting of all species and increasing cover crop implementation. I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency The study also illustrates that lo- cations of high nitrate export were generally associated with moderately well-drained soils and agricultural ﬁelds more frequently used for corn. Therefore, it would be important to prioritize winter cover crop application with early planted rye for those areas with well-drained soils used for corn pro- duction. average simulation results from selected months. Those se- lected months were chosen considering seasonal characteris- tics of climate and hydrology as well as the timing of agricul- tural practices and scheduling that may produce differences in nitrate loadings (e.g., high precipitation and groundwater ﬂow in March/April, killing winter cover crop and fertilizer application in April, and cover crop application in Novem- ber). The location of high nitrate loading areas was generally as- sociated with moderately well-drained soils and agricultural ﬁelds more frequently used for corn over the simulation pe- riod. Nitrate leaching dominated the total nitrate loads from the ﬁelds (i.e., potential for nitrate export), as it outweighed nitrate transport by direct pathways (as shown in Fig. 8). We hypothesize that areas with moderately well-drained soils al- lowed high nitrate leaching due to their high inﬁltration ca- pacity (Fig. 2). Because of the high nitrogen demand for corn growth and yield, corn cropping requires a consider- able amount of fertilizer application during the early growth stage, while soybean does not require any fertilizer applica- tion (Table 4). Consequently, nitrate export from agricultural ﬁelds more frequently used for corn over the simulation pe- riod was signiﬁcantly greater than those used for soybean, as reported by Kaspar et al. (2012). Therefore, it would be important to prioritize winter cover cropping application for those areas with well-drained soils used for corn production. This study also provides a new approach to calibrate win- ter cover crop growth parameters. Growth parameters for winter cover crops need to be carefully calibrated for shorter day lengths and lower temperatures during the winter, to provide an accurate estimation of the nutrient uptake efﬁ- ciency of cover crops. Unfortunately, at present there are lim- ited data available on winter cover crop growth and biomass estimation at the ﬁeld or landscape scales. However, this data limitation is expected to be resolved in the future, as the planting of winter cover crops becomes more common and monitoring programs are enhanced through the avail- ability of no- or low-cost time series of remotely sensed data (e.g., Landsat). I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency With multiyear cover crop biomass and www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency growth data, the methodology presented in this paper could be extended to better calibrate growth parameters and val- idate winter cover crop biomass, improving the accuracy of SWAT in estimating nitrate removal efﬁciency by winter cover crops. Chaplot, V.: Impact of DEM mesh size and soil map scale on SWAT runoff, sediment, and NO3-N loads predictions, J. Hydrol., 312, 207–222, 2005. Chaubey, I., Cotter, A. S., Costello, T. A., and Soerens, T. S.: Effect of DEM data resolution on SWAT output uncertainty, Hydrol. Process., 19, 621–628, 2005. Chesapeake Bay Commission: Cost-effective strategies for the bay: smart investments for nutrient and sediment reduction, Chesa- peake Bay Commission, Annapolis, MD, 2004. Acknowledgements. This research was funded by the National Aeronautics and Space Administration (NASA) Land Cover and Land Use Change (LCLUC) Program, 2011 University of Maryland Behavioral & Social Sciences (BSOS) Dean’s Research Initiative, US Geological Survey (USGS) Climate and Land Use Change Program (CLU), and US Department of Agriculture (USDA) Conservation Effects Assessment Project (CEAP). The suggestion and comments made by the reviewers and the managing editor of the journal greatly improved our manuscript and they were much appreciated. Acknowledgements. This research was funded by the National Aeronautics and Space Administration (NASA) Land Cover and Land Use Change (LCLUC) Program, 2011 University of Maryland Behavioral & Social Sciences (BSOS) Dean’s Research Initiative, US Geological Survey (USGS) Climate and Land Use Change Program (CLU), and US Department of Agriculture (USDA) Conservation Effects Assessment Project (CEAP). The suggestion and comments made by the reviewers and the managing editor of the journal greatly improved our manuscript and they were much appreciated. peake Bay Commission, Annapolis, MD, 2004. Chu, T. W., Shirmohammadi, A., Montas, H., and Sadeghi, A.: Evaluation of the SWAT model’s sediment and nutrient compo- nents in the piedmont physiographic region of Maryland, Trans. ASAE, 47, 1523–1538, 2004. Clark, A.: Managing cover crops proﬁtably, 3rd Ed., Handbook Se- ries Book 9, Sustainable Agriculture Network, Beltsville, MD, 244 pp., 2007. Dabney, S. M.: Cover crop impacts on watershed hydrology, J. Soil Water Conserv., 53, 207–213, 1998. Disclaimer. The USDA is an equal opportunity provider and employer. Any use of trade, ﬁrm, or product names is for descrip- tive purposes only and does not imply endorsement by the US Government. Feyereisen, G. W., Wilson, B. N., Sands, G. R., Strock, J. S., and Porter, P. References Arnold, J. G. and Fohrer, N.: SWAT2000: current capabilities and research opportunities in applied watershed modelling, Hydrol. Process., 19, 563–572, 2005. Francis, G. S., Bartley, K. M., and Tabley, F. J.: The effect of win- ter cover crop management on nitrate leaching losses and crop growth, J. Agr. Sci., 131, 299–308, 1998. Ator, S. W., Denver J. M., Krantz, D. E., Newell, W. L., and Mar- tucci, S. K.: A Surﬁcial Hydrogeologic Framework for the Mid- Atlantic Coastal Plain, US Geological Survey Professional Paper 1680, Reston, Virginia, 44 pp., 2005. Gardner, R. and Davidson, N.: The Ramsar Convention, in: Wet- lands, edited by: LePage, B. A., Springer Netherlands, 189–203, 2011. Glancey, J., Brown, B., Davis, M., Towle, L., Timmons, J., and Nel- son, J.: Comparison of Methods for Estimating Poultry Manure Nutrient Generation in the Chesapeake Bay Watershed, avail- able at: http://www.csgeast.org/2012annualmeeting/documents/ Glancey.pdf (last access: 25 September 2014), 2012. Baggs, E. M., Watson, C. A., and Rees, R. M.: The fate of nitrogen from incorporated cover crop and green manure residues, Nutr. Cycl. Agroecosys., 56, 153–163, 2000. Bergstrom, L. F. and Jokela, W. E.: Ryegrass cover crop effects on nitrate leaching in spring barley fertilized with (NH4)-N-15 (NO3)-N-15, J. Environ. Qual., 30, 1659–1667, 2001. Hively, W. D., Lang, M., McCarty, G. W., Keppler, J., Sadeghi, A., and McConnell, L. L.: Using satellite remote sensing to estimate winter cover crop nutrient uptake efﬁciency, J. Soil Water Con- serv., 64, 303–313, 2009. Boesch, D. F., Brinsﬁeld, R. B., and Magnien, R. E.: Chesapeake Bay eutrophication: Scientiﬁc understanding, ecosystem restora- tion, and challenges for agriculture, J. Environ. Qual., 30, 303– 320, 2001. Islam, N., Wallender, W. W., Mitchell, J., Wicks, S., and Howitt, R. E.: A comprehensive experimental study with mathematical modeling to investigate the affects of cropping practices on water balance variables, Agr. Water Manage., 82, 129–147, 2006. Boryan, C., Yang, Z., and Di, L.: Deriving 2011 cultivated land cover data sets using USDA National Agricultural Statistics Ser- vice historic Cropland Data Layers, Geoscience and Remote Sensing Symposium (IGARSS), IEEE International, 6297–6300, 2011. Jordan, T. E., Correll, D. L., and Weller, D. E.: Effects of agricul- ture on discharges of nutrients from coastal plain watersheds of Chesapeake Bay, J. Environ. Qual., 26, 836–848, 1997. Brandi-Dohrn, F. M., Dick, R. P., Hess, M., Kauffman, S. M., Hemphill, D. D., and Selker, J. S.: Nitrate leaching under a cereal rye cover crop, J. Environ. I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency M.: Potential for a rye cover crop to reduce ni- trate loss in southwestern Minnesota, Agron. J., 98, 1416–1426, doi:10.2134/agronj2005.0134, 2006. Fisher, T. R., Jordan, T. E., Staver, K. W., Gustafson, A. B., Koskelo, A. I., Fox, R. J., Sutton, A. J., Kana, T., Beckert, K. A., Stone, J. P., McCarty, G., and Lang, M.: The Choptank Basin in tran- sition: intensifying agriculture, slow urbanization, and estuarine eutrophication, in: Coastal Lagoons: critical habitats of environ- mental change, edited by: Kennish, M. J. and Paerl, H. W., CRC Press, 135–165, 2010. Edited by: N. Romano 4 Conclusions This study demonstrates the effectiveness of winter cover crops for reducing nitrate loads and shows that nitrate re- moval efﬁciency varies greatly by species, timing, and ex- tent of winter cover crop implementation. It also illustrates www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 5251 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency M.: Winter cover crops as a best management practice for reducing nitrogen leaching, J. Contam. Hydrol., 34, 1–15, 1998. Malhi, S., Johnston, A., Schoenau, J., Wang, Z., and Vera, C.: Sea- sonal biomass accumulation and nutrient uptake of wheat, barley and oat on a Black Chernozem Soil in Saskatchewan, Can. J. Plant Sci., 86, 1005–1014, 2006. Sadeghi, A., Yoon, K., Graff, C., McCarty, G., McConnell, L., Shir- mohammadi, A., Hively, D., and Sefton, K.: Assessing the Per- formance of SWAT and AnnAGNPS Models in a Coastal Plain Watershed, Choptank River, Maryland, Proceedings of the 2007 ASABE Annual International Meeting, St. Joseph, MI, 17–20 June 2007, Paper No. 072032, 2007. McCarty, G. W., McConnell, L. L., Hapernan, C. J., Sadeghi, A., Graff, C., Hively, W. D., Lang, M. W., Fisher, T. R., Jordan, T., Rice, C. P., Codling, E. E., Whitall, D., Lynn, A., Keppler, J., and Fogel, M. L.: Water quality and conservation practice effects in the Choptank River watershed, J. Soil Water Conserv., 63, 461– 474, doi:10.2489/jswc.63.6.461, 2008. Saleh, A. and Du, B.: Evaluation of SWAT and HSPF within BASINS program for the Upper North Bosque River watershed in Central Texas, Trans. ASAE, 47, 1039–1049, 2004. McCracken, D. V., Smith, M. S., Grove, J. H., Mackown, C. T., and Blevins, R. L.: Nitrate Leaching as Inﬂuenced by Cover Crop- ping and Nitrogen-Source, Soil Sci. Soc. Am. J., 58, 1476–1483, 1994. Seo, M., Yen, H., Kim, M-.K., and Jeong, J.: Transferability of SWAT Models between SWAT2009 and SWAT2012, J. Environ. Qual., 43, 869–880, doi:10.2134/jeq2013.11.0450, 2014. Sexton, A. M., Sadeghi, A. M., Zhang, X., Srinivasan, R., and Shir- mohammadi, A.: Using Nexrad and Rain Gauge Precipitation Data for Hydrologic Calibration of SWAT in a Northeastern Wa- tershed, Trans. ASABE, 53, 1501–1510, 2010. MDA (Maryland Department of Agriculture) Ofﬁce of Re- source Conservation: Keep Maryland Under Cover: Plant Cover Crops, available at: http://http://www.mda.state.md.us/resource_ conservation/Documents/2012_CoverCrop.pdf (last access: 25 September 2013), 2012. Shipley, P. R., Meisinger, J. J., and Decker, A. M.: Conserving Residual Corn Fertilizer Nitrogen with Winter Cover Crops, Agron. J., 84, 869–876, 1992. Meals, D. W., Dressing, S. A., and Davenport, T. E.: Lag Time in Water Quality Response to Best Management Practices: A Re- view, J. Environ. Qual., 39, 85–96, doi:10.2134/Jeq2009.0108, 2010. Singh, A., Imtiyaz, M., Isaac, R. K., and Denis, D. I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency 5252 tion with SWAT and GWLF, Ecol. Modelling, 268. 123–133, doi:10.1016/j.ecolmodel.2013.08.007, 2013. Kaspar, T. C., Jaynes, D. B., Parkin, T. B., Moorman, T. B., and Singer, J. W.: Effectiveness of oat and rye cover crops in reducing nitrate losses in drainage water, Agr. Water Manage., 110, 25–33, doi:10.1016/j.agwat.2012.03.010, 2012. tion with SWAT and GWLF, Ecol. Modelling, 268. 123–133, doi:10.1016/j.ecolmodel.2013.08.007, 2013. Phillips, S. W., Focazio, M. J., and Bachman, L. J.: Discharge, ni- trate load, and residence time of ground water in the Chesapeake Bay watershed, Washington, DC, US Geological Survey, Fact Sheet FS-150-99, 1999. j g Kemp, W. M., Boynton, W. R., Adolf, J. E., Boesch, D. F., Boicourt, W. C., Brush, G., Cornwell, J. C., Fisher, T. R., Glibert, P. M., Hagy, J. D., Harding, L. W., Houde, E. D., Kimmel, D. G., Miller, W. D., Newell, R. I. E., Roman, M. R., Smith, E. M., and Steven- son, J. C.: Eutrophication of Chesapeake Bay: historical trends and ecological interactions, Mar. Ecol. Prog. Ser., 303, 1–29, 2005. Primrose, N. L., Millard, C. J., McCoy, J. L., Dobson, M. G., Sturm, P. E., Bowen, S. E., and Windschitl, R. J.: German Branch Tar- geted Watershed Project: Biotic and Water Quality Monitoring Evaluation Report 1990 through 1995, Maryland Department of Natural Resources, Annapolis, MD, 1997. Lam, Q. D., Schmalz, B., and Fohrer, N.: Assessing the spatial and temporal variations of water quality in lowland areas, Northern Germany, J. Hydrol., 438, 137–147, 2012. Reckhow, K. H., Norris, P. E., Budell, R. J., Di Toro, D. M., Gal- loway, J. N., Greening, H., Sharpley, A. N., Shirmhhammadi, A., and Stacey, P. E.: Achieving Nutrient and Sediment Reduction Goals in the Chesapeake Bay: An Evaluation of Program Strate- gies and Implementation, The National Academies Press, Wash- ington, DC, 2011. Lang, M., McDonough, O., McCarty, G., Oesterling, R., and Wilen, B.: Enhanced Detection of Wetland-Stream Connectivity Using LiDAR, Wetlands, 32, 461–473, 2012. Rinnofner, T., Friedel, J. K., de Kruijff, R., Pietsch, G., and Freyer, B.: Effect of catch crops on N dynamics and following crops in organic farming, Agron. Sustain. Dev., 28, 551–558, 2008. Lee, K. Y., Fisher, T. R., Jordan, T. E., Correll, D. L., and Weller, D. E.: Modeling the hydrochemistry of the Choptank River Basin using GWLF and Arc/Info: 1. Model calibration and validation, Biogeochemistry, 49, 143–173, doi:10.1023/A:1006375530844, 2000. Ritter, W. F., Scarborough, R. W., and Chirnside, A. E. References Qual., 26, 181–188, 1997. Kang, M. S., Park, S. W., Lee, J. J., and Yoo, K. H.: Applying SWAT for TMDL programs to a small watershed containing rice paddy ﬁelds, Agr. Water Manage., 79, 72–92, 2006. CEC (Chesapeake Executive Council): Chesapeake 2000 agree- ment, Chesapeake Bay Program, Annapolis, MD, 2000. Kaspar, T. C., Jaynes, D. B., Parkin, T. B., and Moorman, T. B.: Rye Cover Crop and Gamagrass Strip Effects on NO Concentration and Load in Tile Drainage, J. Environ. Qual., 36, 1503–1511, doi:10.2134/jeq2006.0468, 2007. Cerco, C. F. and Noel, M. R.: Can oyster restoration reverse cultural eutrophication in Chesapeake Bay?, Estuar. Coast., 30, 331–343, 2007. www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 Zhang, X., Srinivasan, R., and Van Liew, M.: Multi-Site Calibration of the Swat Model for Hydrologic Modeling, Trans. ASABE, 51, 2039–2049, 2008. Winchell, M., Srinivasan, R., Di Luzio, M., and Arnold, J. G.: Arc- SWAT interface for SWAT2009. User’s Guide, Texas A&M Uni- versity Press, College Station, TX, 2011. Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency M.: Assess- ing the performance and uncertainty analysis of the SWAT and RBNN models for simulation of sediment yield in the Nagwa watershed, India, Hydrolog. Sci. J., 59, 351–364, doi:10.1080/02626667.2013.872787, 2014. Meng, H., Sexton, A. M., Maddox, M. C., Sood, A., Brown, C. W., Ferraro, R. R., and Murtugudde, R.: Modeling Rappahannock River Basin Using Swat – Pilot for Chesapeake Bay Watershed, Appl. Eng. Agric., 26, 795–805, 2010. Staver, K. W. and Brinsﬁeld, R. B.: Evaluating changes in sub- surface nitrogen discharge from an agricultural watershed into Chesapeake Bay after implementation of a groundwater protec- tion strategy, Final report to Maryland Department of Natural Re- sources, Section 319, University of Maryland, College of Agri- culture and Natural Resources, Agricultural Experiment Station, Wye Research and Education Center, Queenstown, MD, 2000. Moriasi, D. N., Arnold, J. G., Van Liew, M. W., Bingner, R. L., Harmel, R. D., and Veith, T. L.: Model evaluation guidelines for systematic quantiﬁcation of accuracy in watershed simulations, Trans. ASABE, 50, 885–900, 2007. Neitsch, S. L., Arnold, J. G., Kiniry, J. R., and Williams, J. R.: Soil and Water Assessment Tool. Theoretical Documentation; Ver- sion 2009, Texas Water Resources Institute Technical Report No. 406, Texas A&M University System, College Station, TX, 2011. USDA-ARS (Agricultural Research Service) Soil & Water Re- search Laboratory: SWAT changes in Revision 481 and Revi- sion 535, available at: http://swat.tamu.edu/media/51071/swat_ changes_481_535.pdf (last access: 25 September 2013), 2012. Niraula, R., Kalin, L., Srivastava, P., and Anderson, C. J.: Identifying critical source areas of nonpoint source pollu- www.hydrol-earth-syst-sci.net/18/5239/2014/ Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 5253 I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency www.hydrol-earth-syst-sci.net/18/5239/2014/ I.-Y. Yeo et al.: Assessing winter cover crop nutrient uptake efﬁciency Hydrol. Earth Syst. Sci., 18, 5239–5253, 2014 www.hydrol-earth-syst-sci.net/18/5239/2014/
https://openalex.org/W4224983750	https://www.nature.com/articles/s41598-022-10744-5.pdf	English	null	Application of Mac-2 binding protein glycosylation isomer as a non-invasive biomarker for probing liver disease	Scientific reports	2,022	cc-by	7,047	Kawin Tangvoraphonkchai1, Tanita Suttichaimongkol1, Churairat Kularbkaew2, Prakasit Sangaimwibool2 & Wattana Sukeepaisarnjaroen1 Kawin Tangvoraphonkchai1, Tanita Suttichaimongkol1, Churairat Kularbkaew2, Prakasit Sangaimwibool2 & Wattana Sukeepaisarnjaroen1 Liver disease remains a major critical challenge in Thailand due to viral hepatitis. Clinical management requires close monitoring of liver fibrosis severity. Non-invasive testing is an attractive method for probing of disease progression. Mac-2 binding protein glycosylation isomer (M2BPGi) is a novel serum marker for fibrosis staging. The current study evaluates the marker among healthy donors and hepatitis C (HCV) patients. 100 HCV subjects were evaluated by liver biopsy. These patients had varying fibrosis severity based on METAVIR scores. Healthy donors were confirmed based on normal liver functions tests. Comparisons of M2BPGi levels among different study groups were performed and the effectiveness was evaluated using receiver operating characteristics (ROC) curves. Using liver biopsy as the reference standard, median M2BPGi levels in HCV cases were 0.74, 1.38 and 2.88 COI for F0-1, F2 and > F3 cases respectively. In healthy donors, the baseline values ranged 0.1–0.24 COI and statistically lower than liver disease cases profiled using M2BPGi. ROC analysis demonstrated superior results for M2BPGi levels among diseased populations and healthy controls. AUROC was determined at 0.983. Comparing with other non-invasive tests, M2BPGi showed a positive linear trend that indicated a strong match to existing methodologies. M2BPGi addresses a critical need in the management of liver disease by providing straightforward means to probe fibrosis severity. In this study, we found significant differences between hepatitis C and healthy subjects and established the background level in healthy donors. Chronic liver disease progresses to liver fibrosis, cirrhosis, and finally hepatocellular carcinoma. Every year, chronic liver disease causes 2 million deaths globally, of which one million are due to cirrhosis and one million are due to viral hepatitis and hepatocellular carcinoma1 while liver cancer is the sixth most commonly diagnosed cancer and the fourth leading cause of cancer deaths (8.2% of cancer deaths) worldwide in 20182. In Thailand, there are approximately 600,000 hepatitis C (HCV) patients3 that are a cause of concern for complications and cancer. In addition, the prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing in Asia due to increasing prevalence of obesity. This will inevitably lead to an increase in the number of patients with cirrhosis and cancer4. Hence, the economic burden of liver diseases such as HCV5 and non-alcoholic steatohepatitis (NASH)6 are high in Thailand. www.nature.com/scientificreports www.nature.com/scientificreports 1Division of Gastroenterology, Department of Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. 2Department of Pathology, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. email: kawin_tang@hotmail.com Kawin Tangvoraphonkchai1, Tanita Suttichaimongkol1, Churairat Kularbkaew2, Prakasit Sangaimwibool2 & Wattana Sukeepaisarnjaroen1 The study design to investigate the usefulness of M2BPGi in Thai liver disease patients. Figure 1. The study design to investigate the usefulness of M2BPGi in Thai liver disease patients. equipment, are time-consuming, and have low specificity and sensitivity. Therefore, they may not be useful in resource-poor settings where liver disease burden is high. It is also difficult to integrate into existing healthcare screening practices since patients will need additional time on site to perform such tests.i g p p p Mac-2 binding protein glycosylation isomer (M2BPGi) is a biomarker that was identified in patients with liver fibrosis11. In addition, studies have shown that M2BPGi is correlated with liver fibrosis stages12, is a predictor of lenvatinib tolerability and response in hepatocellular carcinoma13, regression of liver fibrosis after treatment in HCV14, is associated with advanced liver fibrosis in diabetes15, is correlated with liver stiffness16, and is a bio- marker for survival after HCV eradication17, demonstrating the usefulness of this novel biomarker. M2BPGi is also a reimbursable clinical test in Japan but limited data is available in other Asia Pacific countries as highlighted in the 2016 Asian-Pacific Association for the Study of the Liver (APASL) consensus guidelines. Further studies in comparison to other non-invasive modalities were also recommended by the panel. In this study, we aim to establish clear clinical relevance of M2BPGi in the Thai population. Building upon prior M2BPGi research work, we established and examined M2BPGi levels within HCV patients in comparison to healthy people in Thailand. These determine the usefulness of the biomarker in the diagnosis of liver disease. In addition, the comparison results with other non-invasive alternatives showed interesting trends that were not previously highlighted. Kawin Tangvoraphonkchai1, Tanita Suttichaimongkol1, Churairat Kularbkaew2, Prakasit Sangaimwibool2 & Wattana Sukeepaisarnjaroen1 ( ) gh Early diagnosis and treatment are of paramount importance in stopping the progression of liver disease to cirrhosis and liver cancer. Chronic HCV can be treated with direct-acting antivirals (DAAs) such as sofosbu- vir/velpatasvir and glecaprevir/pibrentasvir. Current guidelines recommend DAAs for chronic HCV treatment regardless of fibrosis stage. The control of viral loads is effective with such treatments, but liver diseases have manifested in these chronic patients that need to be addressed. Hence, liver fibrosis staging is important in the management of patients with chronic liver disease. At present, the gold standard for liver fibrosis staging is a liver biopsy. However, liver biopsy has its limitations, such as high cost, sampling error, inter-observer varia- tions, and the risk of complications7. It is also not feasible to perform liver biopsies repeatedly for monitoring of disease progression. Other methods for liver fibrosis staging include transient elastography8, acoustic radiation force impulse9, serum biomarkers such as hyaluronic acid and lincRNA-p2110. These methods require expensive 1Division of Gastroenterology, Department of Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. 2Department of Pathology, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. email: kawin_tang@hotmail.com Scientific Reports \| (2022) 12:6757 \| https://doi.org/10.1038/s41598-022-10744-5 www.nature.com/scientificreports/ Chronic HCV paents who underwent a liver biopsy (N = 100) Exclusion criteria Paents with HIV co-infecon Paents with contraindicaons to liver biopsy: paents do not cooperate, paents with risk bleeding cannot be solved by providing blood components (INR> 1.5, platelet <80,000)., Paents with skin infecons in the right ribs, Paents with obstrucve jaundice, Paents with ascites, Paents suspected liver nodules. Paents with uncontrolled underlying diseases, such as heart failure, renal failure, diabetes, hyper or hypothyroidism, alcoholism, hypertension, ischemic heart diseases, COPD, major depressive disorder, and BMI ≥ 25 kg/m2. Paents with consumpon of alcohol within six months or use of addicve drugs. Paents do not stop anplatelet at least 10 days. Pregnancy. Organ transplant. Paents undergo chemotherapy. Paents with acute hemolysis or extra hepac cholestasis. Paents with chronic hemolysis or Gilbert's syndrome. Liver biopsy M2BPGi and other biomarker test Figure 1. The study design to investigate the usefulness of M2BPGi in Thai liver disease patients. Chronic HCV paents who underwent a liver biopsy (N = 100) Figure 1. Materials and methods Characteristics Values Male:Female 67:33 Age (years) 51.0 (8.3) Underlying disease None 77 (77) Diabetes mellitus 11 (11) Hypertension 12 (12) Hemoglobin (g/dl) 13.8 (1.4) Platelet count (103/mm3) 192.5 (56.5) Creatinine (mg/dl) 0.9 (0.2) Albumin (g/dl) 4.4 (0.4) Total bilirubin (mg/dl) 0.7 (0.3) ALT (U/L) 73.2 (54.8) AST (U/L) 63.6 (43.3) ALP (U/L) 90.0 (36.8) INR 1.0 (0.1) HCV Viral load (IU/ml) 4,032,315 (5,919,394) HCV genotypes 1A 15 (15) 1B 17 (17) 3 40 (40) 6 28 (28) Fibrosis score (METAVIR) F0 7 (7) F1 28 (28) F2 33 (33) F3 17 (17) F4 15 (15) Liver assessment APRI 1.18 (0.95) FIB4 2.27 (1.37) FibroIndex 1.76 (0.57) Fibroscan (kPA) 13.39 (11.85) Fibrotest 0.53 (0.28) M2BPGi (COI) 1.97 (2.07) T bl 1 B li h t i ti f th 100 HC Characteristics Values Male:Female 67:33 Age (years) 51.0 (8.3) Underlying disease None 77 (77) Diabetes mellitus 11 (11) Hypertension 12 (12) Hemoglobin (g/dl) 13.8 (1.4) Platelet count (103/mm3) 192.5 (56.5) Creatinine (mg/dl) 0.9 (0.2) Albumin (g/dl) 4.4 (0.4) Total bilirubin (mg/dl) 0.7 (0.3) ALT (U/L) 73.2 (54.8) AST (U/L) 63.6 (43.3) ALP (U/L) 90.0 (36.8) INR 1.0 (0.1) HCV Viral load (IU/ml) 4,032,315 (5,919,394) HCV genotypes 1A 15 (15) 1B 17 (17) 3 40 (40) 6 28 (28) Fibrosis score (METAVIR) F0 7 (7) F1 28 (28) F2 33 (33) F3 17 (17) F4 15 (15) Liver assessment APRI 1.18 (0.95) FIB4 2.27 (1.37) FibroIndex 1.76 (0.57) Fibroscan (kPA) 13.39 (11.85) Fibrotest 0.53 (0.28) M2BPGi (COI) 1.97 (2.07) Table 1. Baseline characteristics of the 100 HCV patients. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Table 1. Baseline characteristics of the 100 HCV patients. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. batched processed at a rate of 200 tests per hour on the HISCL 5000. From manufacturer’s claim, results above 1.0 COI indicate a risk of liver fibrosis. batched processed at a rate of 200 tests per hour on the HISCL 5000. From manufacturer’s claim, results above 1.0 COI indicate a risk of liver fibrosis. Fibroscan of study subjects. Materials and methods Study design and patient demographics. 100 subjects with HCV were recruited for this study between June 2017 and December 2019. These patients have mixed liver fibrosis stages based on METAVIR scores. This study is a prospective, single center diagnostic study with approval by the Khon Kaen University Ethics commit- tee for Human Research based on the Declaration of Helsinki and the ICH Good Clinical Practice Guidelines. (HE591548). Informed consent was obtained from all study subjects. Liver tissues were used as the comparator as this method is the reference gold standard. Other commonly used non-invasive liver disease testing modali- ties were concurrently collected from these groups of patients. Inclusion criteria for HCV patients were for treat- ment naïve cases and of mono-infection. The list of exclusion criteria are shown in Fig. 1. 63 Healthy volunteers were used as controls in this study and liver function in these volunteers was verified by hepatic function tests (Table 1), serology (HBsAg, Anti-HBs, Anti-HBc, Anti-HCV, Anti-HIV), and ultrasound. M2BGPi biomarker measurements. Serum M2BPGi levels were measured on the HISCL 5000 auto- mated immunoassay analyzer (Sysmex Corporation, Japan). 10 μL of sample were used and M2BPGi levels were measured by a sandwich immunoassay. Each reaction took 17 min and M2BPGi was expressed as the cut-off index (COI) and calibrated using the manufacturer’s calibrators. Residual samples from routine blood sampling were used and remaining serum samples stored at −20 °C prior to processing on the instrument. Samples were https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ Table 1. Baseline characteristics of the 100 HCV patients. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Materials and methods Characteristics Values Male:Female 67:33 Age (years) 51.0 (8.3) Underlying disease None 77 (77) Diabetes mellitus 11 (11) Hypertension 12 (12) Hemoglobin (g/dl) 13.8 (1.4) Platelet count (103/mm3) 192.5 (56.5) Creatinine (mg/dl) 0.9 (0.2) Albumin (g/dl) 4.4 (0.4) Total bilirubin (mg/dl) 0.7 (0.3) ALT (U/L) 73.2 (54.8) AST (U/L) 63.6 (43.3) ALP (U/L) 90.0 (36.8) INR 1.0 (0.1) HCV Viral load (IU/ml) 4,032,315 (5,919,394) HCV genotypes 1A 15 (15) 1B 17 (17) 3 40 (40) 6 28 (28) Fibrosis score (METAVIR) F0 7 (7) F1 28 (28) F2 33 (33) F3 17 (17) F4 15 (15) Liver assessment APRI 1.18 (0.95) FIB4 2.27 (1.37) FibroIndex 1.76 (0.57) Fibroscan (kPA) 13.39 (11.85) Fibrotest 0.53 (0.28) M2BPGi (COI) 1.97 (2.07) Table 1. Baseline characteristics of the 100 HCV patients. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkalin phosphatase, INR = international normalized ratio. Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. dine 25 mg intravenously before liver biopsy. Percutaneous liver biopsy was performed with a 1.6 mm diameter Trucut needle with a length of at least 2.5 cm. All liver biopsies were at least 8 portal triads and a length ≥ 2.5 cm. All specimens were fixed in formalin, embedded in paraffin, cut, and stained with hematoxylin and eosin. All liver histology interpretations were performed by two pathologists independently (K. Churairat and S. Praka- sit) who had extensive experience in reading biopsied liver samples. Disagreements between pathologists were resolved through discussion and mutual decision about results. Reports were based on the METAVIR scoring system, which specifies a fibrosis score from 0 to 4. dine 25 mg intravenously before liver biopsy. Percutaneous liver biopsy was performed with a 1.6 mm diameter Trucut needle with a length of at least 2.5 cm. All liver biopsies were at least 8 portal triads and a length ≥ 2.5 cm. All specimens were fixed in formalin, embedded in paraffin, cut, and stained with hematoxylin and eosin. All liver histology interpretations were performed by two pathologists independently (K. Churairat and S. Praka- sit) who had extensive experience in reading biopsied liver samples. Disagreements between pathologists were resolved through discussion and mutual decision about results. Reports were based on the METAVIR scoring system, which specifies a fibrosis score from 0 to 4. Statistical analysis. All variables were expressed as mean ± SD unless otherwise defined. The Kruskal– Wallis one-way analysis of variance (ANOVA) was used to compare multiple independent groups as we cannot ascertain normality from our dataset. The Mann–Whitney test was used to compare if there existed a differ- ence in the dependent variable for two independent groups. Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. P values < 0.05 were considered statistically signifi- cant in these investigations. Comparative analysis among healthy controls and diseased patients was computed using Receiver operating characteristics (ROC) curves and plotted to determine the area under ROC (AUROC) curves. These analyses were conducted to compare HCV patients with varying degree of liver fibrosis as well as with healthy control data. Spearman’s rank-order correlations were used to determine the correlations between M2BPGi measurements and other non-invasive liver fibrosis markers. Statistical analyses were performed with STATA® 10.1 software (StataCorp, USA). Materials and methods All patients had liver stiffness measured by FibroScan® 502 touch (Echosens, France) using the M or XL probe according to patients’ body mass index (XL probe for body mass index ≥ 25). This was conducted by experienced nurses (who had conducted more than 100 cases each) on the same day liver biopsy was performed on study subjects (NPO of more than 8 h on day of procedure). The report showed E score (liver stiffness) to interpret liver fibrosis stage. Liver disease biomarker tests. 20 cc of blood specimens were collected from each test subject includ- ing healthy controls. For liver biopsy requirements and consistency, subjects were told to abstain from food 8 h before (NPO > 8 h). Blood samples were used in various laboratory assays to evaluate the comparative indirect liver fibrosis markers such as Fibrotest (LiverFact, Fibronostics Singapore) that defined significant liver fibrosis (F2 METAVIR score) with scores > 0.48, APRI (AST to platelet ratio index), Forns index and FIB-4 (The Fibro- sis-4 Index). Blood tests were also necessary for basic liver function test analysis and preparation for liver biopsy. Liver biopsy. Liver biopsy was performed on patients who did not have contraindication for the procedure. Patients were laid on their backs and given NSS intravenously at 120 ml/hr. Ultrasound was done to check for liver parenchyma and to mark the area for liver biopsy. These patients were given midazolam 2.5 mg and pethi- https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ Table 2. Baseline characteristics of the 63 normal volunteers. Data are expressed as mean (standard deviation) or number (%). ALT = alanine aminotransferase, AST = aspartate aminotransferase, ALP = alkaline phosphatase, INR = international normalized ratio. Characteristics Values Male:Female 17:46 Age (years) 48.11 (11.06) Body Mass Index (kg/m2) 21.62 (2.07) Hemoglobin (g/dl) 12.67 (1.44) Platelet count (103/mm3) 282.97 (93.87) Creatinine (mg/dl) 0.8 (0.16) Cholesterol (mg/dl) 175.94 (19.72) Albumin (g/dl) 4.37 (0.21) Total bilirubin (mg/dl) 0.54 (0.12) ALT (U/L) 17.79 (5.48) AST (U/L) 20.67 (4.3) ALP (U/L) 64.54 (15.81) INR 0.96 (0.34) Results i b Patient background data. Table 1 presents the patient subject demographics. A total of 100 patients were enrolled with none of them were excluded. Among them, 67 were male (67%) and 33 female (33%), and were all HCV positive. The average patient age was 51.0 ± 8.3 years. Genotyping were performed and shown to be pre- dominantly genotypes 3 and 6 (genotype 1a, 1b, 3 and 6 in 15 (15%), 17 (17%), 40 (40%) and 28 (28%) patients, respectively). Figure 1 shows the study design to investigate the usefulness of M2BPGi in Thai liver disease patients. Patients were selection for mono-infection cases only and excluded if they had HIV co-infections, una- ble to perform liver biopsy, had uncontrolled underlying conditions, pregnant cases or patients on other thera- pies. Patients were treatment naive at recruitment and mean viral load among these patients were 4,032,315 IU/ mL. We performed liver biopsies in all 100 patients, and fibrosis stages were F0 (7, 7%), F1 (28, 28%), F2 (33, 33%), F3 (17, 17%) and F4 (15, 15%). An additional 63 healthy controls were considered, all of which had LFTs within normal reference intervals and normal ultrasound upper abdomen (Table 2). Low baseline M2BPGi levels among healthy subjects. Baseline M2BPGi levels were established through healthy donor measurements. Figure 2 shows the spread of results among all 63 datapoints. Median M2BPGi levels were 0.1 COI and the range was observed to be between 0.1 to 0.24 COI, well below the recom- mended cutoff of 1.0 COI that indicates risk of liver disease. Further examination of the dataset showed that results were skewed towards the left, with majority of datapoints below 0.25 COI in healthy Thai subjects. This baseline dataset allows a direct comparison to diseased patients to establish suitable cutoffs for disease assess- ment. Positive correlation with other liver fibrosis markers. Raw data values of M2BPGi were correlated to commonly used non-invasive techniques for liver disease assessment. This is presented in Fig. 3. These associa- tions included comparisons with AST (Figure A); GGT (Figure B); APRI (Figure C); FIB-4 (Figure D); Fibroin- dex (Figure E); Forns Index (Figure F); Fibroscan (Figure G); and Fibrotest (Figure H). Results i b Spearman’s rho among https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 M2BPGi (COI) Frequency 0.10 0.15 0.20 0.25 0 10 20 30 M2BPGi (COI) Frequency Minimum 25% Percentile Median 75% Percentile Maximum 0.1 0.1 0.1 0.17 0.24 Figure 2. M2BPGi levels among healthy volunteers which is significantly below the cutoff (1.0 COI) recommended by the manufacturer for positive results. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 M2BPGi (COI) Frequency 0.10 0.15 0.20 0.25 0 10 20 30 M2BPGi (COI) Frequency Minimum 25% Percentile Median 75% Percentile Maximum 0.1 0.1 0.1 0.17 0.24 Figure 2. M2BPGi levels among healthy volunteers which is significantly below the cutoff (1.0 COI) recommended by the manufacturer for positive results. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 M2BPGi (COI) Frequency 0.10 0.15 0.20 0.25 0 10 20 30 M2BPGi (COI) Frequency Minimum 25% Percentile Median 75% Percentile Maximum 0.1 0.1 0.1 0.17 0.24 Figure 2. M2BPGi levels among healthy volunteers which is significantly below the cutoff (1.0 COI) recommended by the manufacturer for positive results. Figure 3. Comparison of M2BPGi levels among other non-invasive measures of liver fibrosis assessment. (A) Comparison with AST, (B) GGT, (C) APRI, (D) FIB-4, (E) FibroIndex, (F) Forns Index, (G) Fibroscan and (H) Fibrotest. Results showed positive correlations using a Spearman’s rank order correlation. Figure 3. Comparison of M2BPGi levels among other non-invasive measures of liver fibrosis assessment. (A) Comparison with AST, (B) GGT, (C) APRI, (D) FIB-4, (E) FibroIndex, (F) Forns Index, (G) Fibroscan and (H) Fibrotest. Results showed positive correlations using a Spearman’s rank order correlation. these parameters ranged from 0.564 to 0.686, which indicated positive correlations of M2BPGi measurements to each of the liver disease assessment methods. Mean M2BPGi levels obtained was 1.98 ± 2.06 COI that spans from patients from F0 (no fibrosis) to F4 (cirrhosis) conditions. In three of such comparisons (Fibroindex, Forns index and Fibrotest), we observed that the comparator assays plateaued approximately after M2BPGi levels at 3.0 COI. The results highlight potential clinical utility to utilize M2BPGi measurements in staging liver fibrosis cases. these parameters ranged from 0.564 to 0.686, which indicated positive correlations of M2BPGi measurements to each of the liver disease assessment methods. Discussion Non-invasive tests for measuring liver fibrosis stage are attracting more and more attention. This technology is particularly useful in detecting liver fibrosis in remote regions with high incidence of HCV, such as Thailand. The HCV seroprevalence is 0.9%, which is 760,000–1,475,000 anti-HCV positive patients5. In addition, more than half the patients (52.2 to 62.5%) had advanced liver fibrosis (F3 and F4) in areas with high (12%) and average (2%) HCV viremia prevalence5. This shows that many patients in Thailand are not treated until advanced liver fibrosis, which results in greater morbidity and mortality.i i g y y A bottleneck in the treatment of such patients is identification of HCV patients and determining their liver fibrosis stages as most cases occur in rural regions of Thailand. The advent of DAAs has tremendously changed the management of chronic HCV and in the foreseeable future, liver fibrosis will be better controlled in this group. Reaching this group of patients is crucial to drive complete HCV elimination and integrative care for liver disease will be beneficial to prevent progression into advance stages. M2BPGi is attractive from this point of view as blood sampling would have been collected for serology testing of anti HCV, and can be subsequently used in positive cases for fibrosis assessment. Our results demonstrated clear distinct differences among HCV patients with varying degree of liver fibrosis manifestations and healthy subjects. This clear distinction allows early warning flags for risk that can be addressed by referral these patients to hepatologists. In addition, M2BPGi levels are correlated with other markers of liver fibrosis and ultrasound. This is consistent with the results of previous studies in Vietnam18, Thailand19, Japan20, and Egypt21. However, these studies used surrogate markers of liver fibrosis instead of liver biopsy, which is the gold standard. The present study is the first one to show a correlation between M2BPGi levels and liver fibrosis as assessed by liver biopsy outside of Japan, which increases its reliability and demonstrates the utility of using M2BPGi as a surrogate marker for liver fibrosis.fih y y g gi Since 2012, the National Health Security Office of Thailand has included PEGylated-interferon (PEG-IFN) based therapy for HCV into the Universal Health Coverage program. However, this policy has constraints such as the stringent screening eligibility criteria that consist of requirements related to severity and progression of liver damage. Results i b Mean M2BPGi levels obtained was 1.98 ± 2.06 COI that spans from patients from F0 (no fibrosis) to F4 (cirrhosis) conditions. In three of such comparisons (Fibroindex, Forns index and Fibrotest), we observed that the comparator assays plateaued approximately after M2BPGi levels at 3.0 COI. The results highlight potential clinical utility to utilize M2BPGi measurements in staging liver fibrosis cases. Estimation of diagnostic accuracy in chronic hepatitis C cases. We attempted to assess the diag- nostic accuracy of M2BPGi in comparisons with healthy subjects. A ROC analysis was performed as shown in Fig. 4A. M2BPGi levels were vastly discriminatory against chronic HCV cases with AUROC of 0.983 (95% CI https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ 0 20 40 60 80 100 0 20 40 60 80 100 100% - Specificity% Sensitivity% AUROC 0.983 95% CI 0.968 - 0.998 A B Figure 4. Receiver operating characteristics (ROC) curves analysis comparing (A) HCV study cohort (AUROC = 0.983) with controls and (B) early liver fibrosis cases with moderate (F2), or severe (F3) and cirrhotic (F4) cases. AUROC of 0.73 and 0.84 obtained for F0-1 against F2 and ≥ F3 cases respectively. 0 20 40 60 80 100 0 20 40 60 80 100 100% - Specificity% Sensitivity% AUROC 0.983 95% CI 0.968 - 0.998 A A B Figure 4. Receiver operating characteristics (ROC) curves analysis comparing (A) HCV study cohort (AUROC = 0.983) with controls and (B) early liver fibrosis cases with moderate (F2), or severe (F3) and (F4) cases. AUROC of 0.73 and 0.84 obtained for F0-1 against F2 and ≥ F3 cases respectively. of 0.968 to 0.998). Mean values of AST and ALT (Table 1), which are part of basic liver function assessments are significantly elevated (p < 0.001) in the patient cohort compared with the upper limit of normal (ULN). Fur- ther discriminatory analyses were conducted among HCV groups of patients of varying degree of liver fibrosis (Fig. 4B). We observe AUROC of 0.73 and 0.84 for patient groups F0-1 against F2 and ≥ F3 cases respectively. These results highlight the excellent sensitivity and specificity of such tests in differentiating various degree of liver disease. In addressing suitable liver fibrosis cutoffs using M2BPGi and associating with histopathology results, we categorized HCV patients into different fibrosis severities as shown in Fig. 5 similar to what was conducted using the ROC analysis. Results i b Liver biopsy data provided definitive assessment of patient’s condition using current gold standard methodology. Median M2BPGi cutoffs of F0-1, F2 and > F3 cases were 0.74 COI, 1.38 COI and 2.88 COI respectively. A Kruskal Wallis test showed statistical significance between different study groups (p < 0.0001). Further assessments using non-parametric Mann–Whitney analyses (one tail) indicated statistical significance for higher values of > F3 cases against F2 (p < 0.01) and F0-1 against healthy controls (p < 0.01). This provided strong evidence to segregate each cohort using M2BPGi measurements, especially addressing early disease (F0-1 cases) and detecting patients with severe fibrosis or cirrhosis (> F3 cases). Discussion Therefore, the use of M2BPGi level to determine liver damage could circumvent this constraint and allow more HCV patients to be treated. The decreasing cost of DAAs may help increase treatment availability to more patients in Thailand. In fact, there is a need for a national plan for identification and eradication of HCV, as evidenced by a report by the Coalition to Eradicate Viral Hepatitis in Asia Pacific in 2015. In that whitepaper, https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ A F0- a lack of epidemiological data is cited as one of the barriers for elimination of HCV3. The promotion of M2BPGi as a liver fibrosis method can help in the identification of liver fibrosis patients and aid in HCV elimination Our Healthy F0-1 F2 >F3 0 1 2 3 4 5 Study Groups M2BPGi (C.O.I.) ✱✱ ✱✱✱ A F0- F F3/ B Figure 5. Comparison of M2BPGi levels with Metavir scoring from histopathological evaluation in liver biopsy of patients. (A) Representative histological images of F0-1 (no fibrosis), F2 (enlargement of the portal tract with rare septa formation) and F3/4 case (cirrhosis). (B) Corresponding average levels in each study group. SEM plotted. Kruskal Wallis one-way ANOVA shows significant differences among study groups. A Mann–Whitney test between patients with significant liver fibrosis and severe fibrosis/cirrhosis showed statistical significance (p = 0.005) and between healthy and early disease (p < 0.0001). A A F0- F F3/ F F0- B Healthy F0-1 F2 >F3 0 1 2 3 4 5 Study Groups M2BPGi (C.O.I.) ✱✱ ✱✱✱ Healthy F0-1 F2 >F3 0 1 2 3 4 5 Study Groups M2BPGi (C.O.I.) ✱✱ ✱✱✱ Figure 5. Comparison of M2BPGi levels with Metavir scoring from histopathological evaluation in liver biopsy of patients. (A) Representative histological images of F0-1 (no fibrosis), F2 (enlargement of the portal tract with rare septa formation) and F3/4 case (cirrhosis). (B) Corresponding average levels in each study group. SEM plotted. Kruskal Wallis one-way ANOVA shows significant differences among study groups. A Mann–Whitney test between patients with significant liver fibrosis and severe fibrosis/cirrhosis showed statistical significance (p = 0.005) and between healthy and early disease (p < 0.0001). Figure 5. Comparison of M2BPGi levels with Metavir scoring from histopathological evaluation in liver biopsy of patients. (A) Representative histological images of F0-1 (no fibrosis), F2 (enlargement of the portal tract with rare septa formation) and F3/4 case (cirrhosis). Discussion (B) Corresponding average levels in each study group. SEM plotted. Kruskal Wallis one-way ANOVA shows significant differences among study groups. A Mann–Whitney test between patients with significant liver fibrosis and severe fibrosis/cirrhosis showed statistical significance (p = 0.005) and between healthy and early disease (p < 0.0001). a lack of epidemiological data is cited as one of the barriers for elimination of HCV3. The promotion of M2BPGi as a liver fibrosis method can help in the identification of liver fibrosis patients and aid in HCV elimination. Our results indicated possible uses in detecting early disease as significantly different M2BPGi levels were observed comparing early liver fibrosis cases with healthy controls. Identifying early cases link patients to care immediately that will greatly improve clinical outcomes. M2BPGi assay provides a straightforward means to profile hepatitis patients using the same blood sample without further adding on to patient’s burden. In identifying severe or cirrhotic cases, the distinctions are clear with median M2BPGi levels in > F3 cases to more than 2 times that of F2 cases. As we observed in the comparisons for Fibroindex, Forns Index and Fibrotest, these tests saturated for cases with high M2BPGi levels. The clear advantage for M2BPGi is the larger dynamic range to profile late-stage patients. This potentially addresses disease monitoring cases that aims to control the progression of liver disease. a lack of epidemiological data is cited as one of the barriers for elimination of HCV3. The promotion of M2BPGi as a liver fibrosis method can help in the identification of liver fibrosis patients and aid in HCV elimination. Our results indicated possible uses in detecting early disease as significantly different M2BPGi levels were observed comparing early liver fibrosis cases with healthy controls. Identifying early cases link patients to care immediately that will greatly improve clinical outcomes. M2BPGi assay provides a straightforward means to profile hepatitis patients using the same blood sample without further adding on to patient’s burden. In identifying severe or cirrhotic cases, the distinctions are clear with median M2BPGi levels in > F3 cases to more than 2 times that of F2 cases. As we observed in the comparisons for Fibroindex, Forns Index and Fibrotest, these tests saturated for cases with high M2BPGi levels. The clear advantage for M2BPGi is the larger dynamic range to profile late-stage patients. This potentially addresses disease monitoring cases that aims to control the progression of liver disease. References 1. Asrani, S. K., Devarbhavi, H., Eaton, J. & Kamath, P. S. Burden of liver diseases in the world. J. Hepatol. 70, 151–171. https://doi org/10.1016/j.jhep.2018.09.014 (2019). 1. Asrani, S. K., Devarbhavi, H., Eaton, J. & Kamath, P. S. Burden of liver diseases in the world. J. Hepatol. 70, 151–171. https://doi. org/10.1016/j.jhep.2018.09.014 (2019). g j j 2. Bray, F. et al. GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin 2018, 394. https://doi.org/10.3322/caac.21492 (2018). p g 3. Wait, S. et al. Hepatitis B and hepatitis C in southeast and southern Asia: Challenges for governments. Lancet Gastroenterol. Hepatol 1, 248–255. https://doi.org/10.1016/S2468-1253(16)30031-0 (2016). p g 4. Koh, J. C. et al. Asian consensus on the relationship between obesity and gastrointestinal and liver diseases. J. Gastroenterol. Hepatol. 31, 1405–1413. https://doi.org/10.1111/jgh.13385 (2016). p g 4. Koh, J. C. et al. Asian consensus on the relationship between obesity and gastrointestinal and liver diseases. J. Gastroenterol. Hepatol. 31, 1405–1413. https://doi.org/10.1111/jgh.13385 (2016). p g jg 5. Wasitthankasem, R. et al. Liver disease burden and required treatment expenditures for hepatitis C virus (HCV) infection in Thailand: Implications for HCV elimination in the new therapeutic era, a population-based study. PLoS ONE 13, e0196301. https:// doi.org/10.1371/journal.pone.0196301 (2018).iih 5. Wasitthankasem, R. et al. Liver disease burden and required treatment expenditures for hepatitis C virus (HCV) infection in Thailand: Implications for HCV elimination in the new therapeutic era, a population-based study. PLoS ONE 13, e0196301. https:// doi.org/10.1371/journal.pone.0196301 (2018).iih g j p ( ) 6. Phisalprapa, P. et al. Economic burden of non-alcoholic steatohepatitis with significant fibrosis in Thailand. BMC Gastroenterol. 21, 135. https://doi.org/10.1186/s12876-021-01720-w (2021). g j p 6. Phisalprapa, P. et al. Economic burden of non-alcoholic steatohepatitis with significant fibrosis in Thailand. BMC Gastroenterol. 21, 135. https://doi.org/10.1186/s12876-021-01720-w (2021). 21, 135. https://doi.org/10.1186/s12876-021-01720-w (2021). p g 7. Regev, A. et al. Sampling error and intraobserver variation in liver biopsy in pat. Am. J. Gastroenterol. 97, 2614–2618. https://doi org/10.1111/j.1572-0241.2002.06038.x (2002).i g j 8. Sandrin, L. et al. Transient elastography: A new noninvasive method for assessment of hepatic fibrosis. Ultrasound Med. Biol. 29 1705–1713. https://doi.org/10.1016/j.ultrasmedbio.2003.07.001 (2003).l g j 8. Sandrin, L. et al. Transient elastography: A new noninvasive method for assessment of hepatic fibrosis. Ultrasound Med. Biol. 29, 1705–1713. https://doi.org/10.1016/j.ultrasmedbio.2003.07.001 (2003).l p g j 9. Chen, C. C. et al. www.nature.com/scientificreports/ www.nature.com/scientificreports/ In fact, studies found that liver fibrosis regression occurs when DAAs are used for HCV treatment14,19,21 and M2BPGi is a marker for patient survival after treatment with DAAs17. One paper even found that M2BPGi has a better performance than APRI in monitoring liver fibrosis19 while another paper found that M2BPGi shows better performance than other markers in assessing liver fibrosis in HCV patients22. Most studies on M2BPGi levels and HCV were on patients treated with DAAs. However, a study from Japan showed that M2BPGi level was equally useful when HCV patients were treated with IFN-based therapies23, which are mostly used in Thailand. q y p p yh One limitation of this study is that it is a single-center study involving two cohorts (HCV and healthy sub- jects) and a larger, multicenter study is required for verification of the results as Thailand is a large country with different regions. In addition, the subjects in this study were Thai and the results may not be extrapolated to our ethnicities. In fact, most of the studies performed on M2BPGi were in Asian populations and more studies should be conducted in other races to determine the optimal cutoff indexes for different liver diseases in these populations. Another limitation is our assessment of healthy controls which are rudimentary consisting of liver function tests and ultrasound. More stringent criteria may be applied in other countries, but our current data shows M2BPGi levels are significantly lower than HCV patients. This cutoff as demonstrate in the ROC analysis show good sensitivity and specificity to rapidly identify possible liver fibrosis cases. Conclusion Our study results showed that M2BPGi levels can be used to differentiate between HCV patients and healthy subjects and determine fibrosis stage in HCV patients. The present study is the first to show a correlation between M2BPGi level and liver fibrosis stage as assessed by biopsy other than the limited dataset from Japan. This tech- nique can be used to track regression of liver fibrosis during HCV treatment, which is useful in resource-poor settings where prevalence of HCV and advanced liver fibrosis are high. Received: 27 September 2021; Accepted: 8 April 2022 References Diagnostic accuracy of acoustic radiation force impulse (ARFI) and Wisteria floribunda agglutinin-positive Mac-2-binding protein (WFA(+)-M2BP) in patients with chronic liver disease. Med. Sci. Monit. 25, 7169–7174. https://doi.org/ 10.12659/MSM.916533 (2019).i 9. Chen, C. C. et al. Diagnostic accuracy of acoustic radiation force impulse (ARFI) and Wisteria floribunda agglutinin-positive Mac-2-binding protein (WFA(+)-M2BP) in patients with chronic liver disease. Med. Sci. Monit. 25, 7169–7174. https://doi.org/ 10.12659/MSM.916533 (2019).i 10. Yu, F. et al. Serum lincRNA-p21 as a potential biomarker of liver fibrosis in chronic hepatitis B patients. J. Viral Hepat. 24, 580–588. https://doi.org/10.1111/jvh.12680 (2017).i 1. Kuno, A. et al. A serum “sweet-doughnut” protein facilitates fibrosis evaluation and therapy assessment in patients with vira hepatitis. Sci. Rep. 3, 1065. https://doi.org/10.1038/srep01065 (2013).l p p p g p 12. Toshima, T. et al. A novel serum marker, glycosylated Wisteria floribunda agglutinin-positive Mac-2 binding protein (WFA(+)- M2BP), for assessing liver fibrosis. J Gastroenterol. 50, 76–84. https://doi.org/10.1007/s00535-014-0946-y (2015). gi g y 3. Eso, Y. et al. Mac-2 binding protein glycosylation isomer predicts tolerability and clinical outcome of lenvatinib therapy for hepa- tocellular carcinoma. J. Hepatobil. Pancreat Sci. https://doi.org/10.1002/jhbp.954 (2021).i i 13. Eso, Y. et al. Mac-2 binding protein glycosylation isomer predicts tolerability and clinical tocellular carcinoma. J. Hepatobil. Pancreat Sci. https://doi.org/10.1002/jhbp.954 (2021).i 14. Nozaki, A. et al. Sofosbuvir-based therapies associated with regression of liver fibrosis in patients with hepatitis C virus infec A prospective observational study. Medicine 100, e25110. https://doi.org/10.1097/MD.0000000000025110 (2021). p p y p g 15. Park, H., Jun, D. W., Park, H. K. & Kim, M. Clinical implications of serum Mac-2-binding protein glycan isomer as a novel bio- marker of advanced hepatic fibrosis in diabetes. Ann. Transl. Med. 8, 1583. https://doi.org/10.21037/atm-20-5216 (2020).f y g 15. Park, H., Jun, D. W., Park, H. K. & Kim, M. Clinical implications of serum Mac-2-binding protein glycan isomer as a novel 5. Park, H., Jun, D. W., Park, H. K. & Kim, M. Clinical implications of serum Mac-2-binding protein glycan isomer as a novel bio- marker of advanced hepatic fibrosis in diabetes. Ann. Transl. Med. 8, 1583. https://doi.org/10.21037/atm-20-5216 (2020).f i 6. Mak, L. Y. et al. Correlation of serum Mac-2-binding protein glycosylation isomer (M2BPGi) and liver stiffness in chronic hepatitis B infection. Hepatol. Int. 13, 148–156. https://doi.org/10.1007/s12072-019-09928-5 (2019).t p p g ( ) 17. Nakagawa, M. et al. Mac-2 binding protein glycosylation isomer as a novel predictive biomarker for patient survival after hepatitis C virus eradication by DAAs. J. Gastroenterol. Discussion https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ 23. Nagata, H. et al. Serial measurement of Wisteria floribunda agglutinin positive Mac-2-binding protein is useful for predicting liver fibrosis and the development of hepatocellular carcinoma in chronic hepatitis C patients treated with IFN-based and IFN-free therapy. Hepatol. Int. 10, 956–964. https://doi.org/10.1007/s12072-016-9754-1 (2016). Acknowledgementsh g The authors thank the patients for their participation, the Faculty of Medicine, Khon Kaen University for Fund- ing Support (Grant Number IN60328), The Gastroenterological Association of Thailand, Sysmex Asia pacific for sponsor M2BPGi test, and Digital Diagnostic Asia Pte. Ltd. for sponsor Fibrotest/Actitest test. Competing interests h p g The authors declare no competing interests. Author contributions K.T. wrote the main manuscript text and prepared figures.C.K. and P.S. were the liver pathologist for read the ver histology.All authors reviewed the manuscript. References 55, 990–999. https://doi.org/10.1007/s00535-020-01715-6 (2020). y p g 18. Pham, T. T. T., Ho, D. T. & Nguyen, T. Usefulness of Mac-2 binding protein glycosylation isomer in non-invasive pro disease in the Vietnamese population. World J. Hepatol. 12, 220–228. https://doi.org/10.4254/wjh.v12.i5.220 (2020).i 19. Chuaypen, N. et al. Liver fibrosis improvement assessed by magnetic resonance elastography and Mac-2-binding protein glycosyla- tion isomer in patients with hepatitis C virus infection receiving direct-acting antivirals. Hepatol Res. 51, 528–537. https://doi.org/ 10.1111/hepr.13630 (2021).ii p 20. Ueda, N. et al. Liver fibrosis assessments using FibroScan, virtual-touch tissue quantification, the FIB-4 index, and mac-2 binding protein glycosylation isomer levels compared with pathological findings of liver resection specimens in patients with hepatitis C infection. BMC Gastroenterol. 20, 314. https://doi.org/10.1186/s12876-020-01459-w (2020).i 20. Ueda, N. et al. Liver fibrosis assessments using FibroScan, virtual-touch tissue quantification, the FIB-4 index, and mac-2 binding protein glycosylation isomer levels compared with pathological findings of liver resection specimens in patients with hepatitis C infection. BMC Gastroenterol. 20, 314. https://doi.org/10.1186/s12876-020-01459-w (2020).i p g 1. Saleh, S. A., Salama, M. M., Alhusseini, M. M. & Mohamed, G. A. M2BPGi for assessing liver fibrosis in patients with hepatitis C treated with direct-acting antivirals. World J. Gastroenterol. 26, 2864–2876. https://doi.org/10.3748/wjg.v26.i21.2864 (2020). l f d h b ( ) l f l fib h h treated with direct-acting antivirals. World J. Gastroenterol. 26, 2864–2876. https://doi.org/10.3748/wjg.v26.i21.2864 (2020). 22. Xu, H. et al. Accuracy of M2BPGi, compared with Fibro Scan(R), in analysis of liver fibrosis in patients with hepatitis C. BMC Gastroenterol. 17, 62. https://doi.org/10.1186/s12876-017-0618-5 (2017). g J p g jg ( ) 22. Xu, H. et al. Accuracy of M2BPGi, compared with Fibro Scan(R), in analysis of liver fibrosis in patients with hepatitis C. BMC Gastroenterol. 17, 62. https://doi.org/10.1186/s12876-017-0618-5 (2017). https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \| Additional information Correspondence and requests for materials should be addressed to K.T. Correspondence and requests for materials should be addressed to K.T. Reprints and permissions information is available at www.nature.com/reprints. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2022 https://doi.org/10.1038/s41598-022-10744-5 Scientific Reports \| (2022) 12:6757 \|
https://openalex.org/W2472828463	https://research.tees.ac.uk/ws/files/4182839/615695%20a.pdf	English	null	Affective Interaction with a Virtual Character Through an fNIRS Brain-Computer Interface	Frontiers in computational neuroscience	2,016	cc-by	10,897	Edited by: Eduardo Fernandez, University Miguel Hernandez, Spain Edited by: Eduardo Fernandez, University Miguel Hernandez, Spain Reviewed by: John E. Lewis, University of Ottawa, Canada Petia D. Koprinkova-Hristova, Bulgarian Academy of Sciences, Bulgaria Reviewed by: John E. Lewis, University of Ottawa, Canada Petia D. Koprinkova-Hristova, Bulgarian Academy of Sciences, Bulgaria Affective Interaction with a Virtual Character Through an fNIRS Brain-Computer Interface Gabor Aranyi 1, Florian Pecune 2, Fred Charles 1, Catherine Pelachaud 2 and Marc Cavazza 3* Gabor Aranyi 1, Florian Pecune 2, Fred Charles 1, Catherine Pelachaud 2 and Marc Cavazza 3* 1 School of Computing, Teesside University, Middlesbrough, UK, 2 CNRS - LTCI, Telecom ParisTech, Paris, France, 3 School of Engineering and Digital Arts, University of Kent, Canterbury, Kent, UK Affective brain-computer interfaces (BCI) harness Neuroscience knowledge to develop affective interaction from ﬁrst principles. In this article, we explore affective engagement with a virtual agent through Neurofeedback (NF). We report an experiment where subjects engage with a virtual agent by expressing positive attitudes towards her under a NF paradigm. We use for affective input the asymmetric activity in the dorsolateral prefrontal cortex (DL-PFC), which has been previously found to be related to the high-level affective-motivational dimension of approach/avoidance. The magnitude of left-asymmetric DL-PFC activity, measured using functional near infrared spectroscopy (fNIRS) and treated as a proxy for approach, is mapped onto a control mechanism for the virtual agent’s facial expressions, in which action units (AUs) are activated through a neural network. We carried out an experiment with 18 subjects, which demonstrated that subjects are able to successfully engage with the virtual agent by controlling their mental disposition through NF, and that they perceived the agent’s responses as realistic and consistent with their projected mental disposition. This interaction paradigm is particularly relevant in the case of affective BCI as it facilitates the volitional activation of speciﬁc areas normally not under conscious control. Overall, our contribution reconciles a model of affect derived from brain metabolic data with an ecologically valid, yet computationally controllable, virtual affective communication environment. Received: 25 March 2016 Accepted: 27 June 2016 Published: 12 July 2016 Received: 25 March 2016 Accepted: 27 June 2016 Published: 12 July 2016 ORIGINAL RESEARCH published: 12 July 2016 doi: 10.3389/fncom.2016.00070 INTRODUCTION AND RATIONALE Correspondence: Marc Cavazza m.o.cavazza@kent.ac.uk Correspondence: Marc Cavazza m.o.cavazza@kent.ac.uk One major challenge for brain-computer interfaces (BCI) is to be based on sound computational models grounded in Neuroscience findings. Recent years have seen an interest in developing affective BCI (Mühl et al., 2014) to support a wide range of interactive systems, as an extension of work in affective computing. However, unlike with imagery-based or motor-oriented BCI, the anatomical localization of emotional responses is often elusive (Lindquist et al., 2012), hampering the design of neuroscience-inspired computational models. From a similar perspective, Mattout (2012) has advocated the use of social neuroscience signals to develop improved BCI which would go beyond the traditional command paradigm of conversion of signal into action, and support more sophisticated, life like interactions. PREVIOUS AND RELATED WORK One of the earliest works using neuroimaging to study the perception of virtual agents is from Schilbach et al. (2006). They investigated neural correlates of observed social interactions with virtual agents using fMRI, the behaviors of virtual agents being designed to be perceived as either socially relevant or arbitrary. Their main finding was that perception of social interactions (defined as ‘‘adequate facial expressions’’ for the virtual agent) resulted in an activation of the ventromedial PFC (VM-PFC), which was not observed when the agents expressions were arbitrary. They attribute this activation to the VM-PFC’s involvement in the early processing of social meaning, also citing its role in joint attention as well as in guiding approach and withdrawal. In more recent work, they have explored an interactive setting for joint attention based on mutual gaze between the user and a virtual agent (Schilbach et al., 2010), showing reward-related neurocircuitry to be activated during the initiation of joint attention. We have designed an asymmetric situation, in which the user interacts through BCI alone, whilst the virtual agent uses realistic non-verbal behavior. This setting is intended to facilitate the controllability of the experiment, as well as the principled use of background knowledge in affective neuroscience and affective psychology to support the computational approach. Citation: Aranyi G, Pecune F, Charles F, Pelachaud C and Cavazza M (2016) Affective Interaction with a Virtual Character Through an fNIRS Brain-Computer Interface. Front. Comput. Neurosci. 10:70. doi: 10.3389/fncom.2016.00070 In this article, we report on an experimental framework and investigate affective BCI in a synthetic but realistic situation, in which the user interacts with a virtual agent endowed with July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 1 Aranyi et al. Affective Interaction Through an fNIRS BCI facial expressions. Our objective is to study under which conditions users can engage with a virtual agent that displays realistic non-verbal behavior, using only their mental disposition towards the agent. Hence it aims at reconciling some limited but corroborated neuroscience data about mental disposition with what is perhaps the largest body of work in affective computing, facial expressions. Following the definitions of rapport according to Gratch et al. (2007), in our setting the positive signal from the user towards the agent corresponds to the intensity of approach, and the positive response from the agent to the evolution of non-verbal behavior and facial expressions (combination of focus of attention and positively valenced expressions). The establishment of rapport being seen as the concomitance of positive signals on both sides in an interaction setting, it is thus well captured by the experimental design. In this framework, we are creating a new vision of rapport (Gratch et al., 2007), which we claim to be more controllable than empathy (Light et al., 2009), requiring generally a stronger context and background, or alignment (Menenti et al., 2012) in which verbal communication is required on both sides. We have opted for a Neurofeedback (NF) paradigm, which is appropriate to the closed-loop setting required to investigate rapport. In this system, the input signal is constituted by prefrontal cortex (PFC) asymmetry measured through functional near infrared spectroscopy (fNIRS; capturing the mental disposition towards the agent), and the feedback channel is constituted by the virtual agent’s non-verbal behavior, facial expressions in particular, responding to the perceived disposition. Other supporting elements include the fact that NF facilitates BCI input where the signal is not under direct volitional control, and that PFC asymmetry has been demonstrated to be amenable to NF control (Rosenfeld et al., 1995; Aranyi et al., 2015b). Citation: While previous studies have related the perception of rapport to PFC signals (Schilbach et al., 2006; Gordon et al., 2014), it is important to note that here, we are not directly equating rapport to PFC asymmetry. Instead, we show that rapport can emerge, from the user perspective, from the perception of a positively valenced response to her own engagement or mental disposition. We posit that a computational model of approach can support a visually realistic engagement with an agent displaying realistic affective behavior. This will be measured through the combination of successful NF in this interactive context with subjective experience questionnaires. Frontiers in Computational Neuroscience \| www.frontiersin.org SYSTEM OVERVIEW Agents display their emotional states through their verbal and non-verbal behaviors. While earlier models focused on the six prototypical expressions of emotion, latest models allow modeling a large variety of facial expressions as a combination of the expressions of emotions (Albrecht et al., 2005) as sequences of multimodal signals (Niewiadomski et al., 2011), or even as a blend of expressions (Niewiadomski and Pelachaud, 2007). Models may rely on pure combinatorial approaches where the expression of an emotion is the result of an algebraic combination of other expressions (Albrecht et al., 2005) of regions on the face (Bui et al., 2001; Niewiadomski and Pelachaud, 2007); other models have relied on corpora annotation (Niewiadomski et al., To support our experiments with virtual agent’s control from a single affective dimension, we designed a complete software platform, which is presented in Figure 1. From the user’s perspective, the virtual agent behaves autonomously as a response to what it perceives as the user’s mental disposition towards itself. Users are instructed to express positive feelings towards the agent in order to capture its interest. This should in turn result in the agent responding to the user with an expression matching the perceived interest in both valence and intensity. FIGURE 1 \| System overview. Brain signals are collected through functional near infrared spectroscopy (fNIRS) system (A) where left-most and right-most channels are processed to generate a left-asymmetry score (1). During the View epoch (2), the left-asymmetry values are used to deﬁne the Min and Max bounds (3) to be used during the Neurofeedback (NF) epoch where the real-time left-asymmetry scores (4) are normalized (B) before being used as single input (5) to the virtual agent’s facial expressions action units (AUs) and body action parameters under the neural network’s control (C). FIGURE 1 \| System overview. Brain signals are collected through functional near infrared spectroscopy (fNIRS) system (A) where left-most and right-most channels are processed to generate a left-asymmetry score (1). During the View epoch (2), the left-asymmetry values are used to deﬁne the Min and Max bounds (3) to be used during the Neurofeedback (NF) epoch where the real-time left-asymmetry scores (4) are normalized (B) before being used as single input (5) to the FIGURE 1 \| System overview. Brain signals are collected through functional near infrared spectroscopy (fNIRS) system (A) where left-most and right-most channels are processed to generate a left-asymmetry score (1). BCI: PREFRONTAL ASYMMETRY AND fNIRS Despite difficulties in identifying neural substrates for major categories of emotions (Lindquist et al., 2012), a large body of work has demonstrated a strong correlation between PFC activity and the affective-motivational dimension of approach/withdrawal (Davidson, 1992). Furthermore, approach has been shown to underpin higher-level emotions such as empathy (Light et al., 2009; Gilroy et al., 2013) and anger (Harmon-Jones, 2003). Nozawa and Kondo (2009) have monitored the activity of the PFC using fNIRS during controlled mouse-based interaction with a virtual agent and found a surge of activity predominantly in the DL-PFC, which they interpreted in attentional terms only. Benbasat et al. (2010) reported a neuroimaging study of how virtual agents induce social presence as a function of their appearance (gender, ethnicity) using fMRI, with the anterior paracingulate cortex as a region of interest. In previous work, we have demonstrated that both anger and empathy could be been successfully used in a BCI context to interact with virtual agents (Gilroy et al., 2013; Aranyi et al., 2015b). Our objective here is to bridge the gap with affective expression and improve the joint analysis of objective and subjective users’ responses. Hence, as a component of rapport, approach is the positive element of input on the user’s side. PFC asymmetry can be measured through various signals including electroencephalography (EEG) (Davidson, 1992), functional magnetic resonance imaging (fMRI) (Zotev et al., 2013) and fNIRS (Doi et al., 2013). We have opted for fNIRS because it offers better signal stability and robustness to motion artifacts and also because PFC activity is readily accessible to fNIRS measurement, in particular the dorsolateral PFC (DL-PFC) through lateral optodes (Ayaz et al., 2011). Our own work, while clearly following on results from Schilbach et al. (2006, 2010), aims at deriving BCI techniques supporting social interaction with virtual agents rather than investigating the neural correlates of human-agent interaction. BCI input bypasses traditional expressive mechanisms of human subjects (facial expressions, gaze), while virtual agents use a range of facial expressions reflecting the user’s perceived level of engagement. We concentrate on the DL-PFC as a region of interest because of its relation to approach (Spielberg et al., 2012) Frontiers in Computational Neuroscience \| www.frontiersin.org July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 2 Affective Interaction Through an fNIRS BCI Aranyi et al. and our previous experience with fNIRS-based NF (Aranyi et al., 2015a,b). BCI: PREFRONTAL ASYMMETRY AND fNIRS While our installation is not fully ecological in terms of communication, by relying on users’ mental disposition it trades naturalness for controllability and sustained interaction. 2011) where videos are carefully annotated to extract multimodal signals expressing emotions. Others have applied a perceptual approach (Grammer and Oberzaucher, 2006; Etemad et al., 2015) where human users are asked to create expressions of the virtual agents for given emotions. Previous research in embodied conversational agents has investigated both empathic agents supporting the user in learning situations (Prendinger and Ishizuka, 2005) and agents that elicit empathy from users. The use of physiological signals has been experimented in both contexts (Prendinger and Ishizuka, 2005; Gilroy et al., 2013). Gilroy et al. (2013) have described the use of EEG prefrontal asymmetry to capture empathy towards a virtual agent during an interactive narrative, but have not explored the agent’s expressions; instead they used a simplified feedback mechanism based only on color saturation. Huang et al. (2011) also report that virtual agents can create rapport during interactions with human users by generating proper verbal and non-verbal behaviors. This provides a setting whereby we assume that generated non-verbal behaviors (i.e., facial expressions) are the reflection of the human user, which thus creates a relationship between the user’s input signals and the resulting visual feedback generated by the virtual agent. SYSTEM OVERVIEW During the View epoch (2), the left-asymmetry values are used to deﬁne the Min and Max bounds (3) to be used during the Neurofeedback (NF) epoch where the real-time left-asymmetry scores (4) are normalized (B) before being used as single input (5) to the virtual agent’s facial expressions action units (AUs) and body action parameters under the neural network’s control (C). Frontiers in Computational Neuroscience \| www.frontiersin.org July 2016 \| Volume 10 \| Article 70 3 Affective Interaction Through an fNIRS BCI Aranyi et al. Users’ attitudes are captured through their levels of PFC asymmetry, using fNIRS to measure differences in activity in left vs. right PFC (Figure 1A). This is based on extensive work relating the approach/withdrawal affective dimension to PFC asymmetry, measured through EEG (Davidson et al., 1990) as well as MRI (Zotev et al., 2014). We have adopted fNIRS because of its lesser sensitivity to artifacts compared to EEG and the fact that it is well-suited to capture activity in the PFC, in particular its dorsolateral region (Doi et al., 2013). Furthermore, experience gained when studying PFC asymmetry through fMRI (Cavazza et al., 2014) could provide guidance to design fNIRS experiments. et al., 2002). An expression is the combination of several AUs, where each AU corresponds to a muscular contraction. To design behaviors conveying an appropriate positive emotion, thus expressing engagement, we based our work on Ochs et al. (2016). Our combinations of AUs are slightly different than the ones described in Schilbach et al. (2006): we did not use winking (AU46), but added cheek rising (AU6) in combination with smiles, which is universally related to the expression of joy (Ekman et al., 2002). Moreover, we added jaw-dropping (AU26) to the mouth opening to model an expression of delight. The following AUs (AU1, AU2, AU5, AU12 and AU25) are used in both Shilbach’s work and ours. For body animations, our agent follows the MPEG-4 standard which defines BAPs. Each BAP represents the rotation angle of one body joint along one axis. From the virtual agent’s perspective, the main challenge is to design an appropriate control process that would interpret and respond appropriately to the level of approach expressed mentally by the user. The system uses a network- like control representation to relate a single input to an array of action units (AUs) and body animation parameters (BAPs) that provide low-level control for the agent animation (Figure 1C). VIRTUAL AGENT Designing Agent’s Non-Verbal Behaviors To characterize the virtual agent’s range of non-verbal behaviors, we considered four levels of engagement, ranging from disengagement to full engagement. We designed non-verbal behaviors for our virtual agent corresponding to these four levels (see Figure 2). Gaze is one of the most prominent indicators of engagement. Indeed, gaze avoidance can be perceived as disengagement (Doherty-Sneddon et al., 2002), while mutual gaze is attributed to engagement (Burgoon et al., 1984). Moreover, empathic engagement can be achieved through affective interactions (Hall et al., 2005). Frequently expressing positively valenced emotions denote a high level of engagement, while negative expressions usually indicate disengagement. SYSTEM OVERVIEW Our neural network behaves more like a graphical model with activation propagation than a traditional neural network with learning. It is designed to fine-tune control over a large number of variables with multiple control parameters (for further details, see Charles et al., 2015). Altogether, eight expressions were created for this experiment, following the literature (further details provided in Table 1): • Low level of engagement: three behaviors depict a disengaged agent: (1, Neg1) The agent gazes away from the user, slightly tilts its head to the right, and shows a lip pout. (2, Neg2) The agent gazes completely down. (3, Neg3) The agent slightly turns its body away from the user and lowers its lips’ corners. The facial expressions generated used AU15, AU17, AU18, AU23 and AU24 for the mouth; and AU43, AU62 and AU64 for the eyes of the agent. Body movements generated used vc1roll, vc1tilt and vc1torsion for the cerebral vertebra and vl1torsion for the lumbar vertebra. An overview of the operation of the system is as follows: the system captures PFC asymmetry in real-time and produces a matching virtual agent’s response. The PFC asymmetry baseline, which is subject-dependent, is acquired through an epoch during which subjects watch the virtual agent (displaying a neutral attitude) while carrying out a simple mental counting task (Figure 1B). During active engagement by subjects, PFC asymmetry is measured and its increase over the baseline is interpreted as the intensity of the approach towards the agent. Finally, the value is passed to the control network, which will generate matching virtual agent’s behavior, interrupting idle behavior and producing appropriate facial expressions. • Below average level of engagement: (4, Neutral) The agent shows a neutral facial expression with its gaze and body oriented towards the user. • Above average level of engagement: two behaviors portray a mildly positive attitude, expressing happiness. (5, MildPos1) The agent keeps its gaze directed at the user and displays a smile of mild intensity. (6, MildPos2) The agent slightly tilts its head on the left, looks at the user, and displays a smile of mild intensity. The facial expressions generated used AU12, AU25 and AU26 for the mouth; and AU5 for the eyes of the agent. Body movement generated used vc1roll for the cerebral vertebra. • High level of engagement: two behaviors match with the agent’s engagement as it displays delight. SYSTEM OVERVIEW (7, HighPos1) Its smile widens, crow’s feet appear around the eyes as the cheeks are raised and its head slightly tilts backwards. (8, HighPos2) The agent’s smile widens, it slightly opens its mouth and raises its eyebrows. The facial expressions generated used AU1 and AU2 for the eyebrows; AU6 for the cheeks; AU12, AU25 and AU26 for the mouth; and AU5 for the eyes of the agent. Body movement generated used vc1tilt for the cerebral vertebra. Frontiers in Computational Neuroscience \| www.frontiersin.org Designing Natural Transitions Between Behaviors To ensure the provision of continuous natural feedback to the user, the different behaviors are mapped onto one other. We developed transitions between behaviors representing different levels of engagement. Each behavior corresponds to a particular The facial expressions of the virtual agent are defined in terms of AUs of facial action coding system (FACS) (Ekman July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 4 Aranyi et al. Affective Interaction Through an fNIRS BCI FIGURE 2 \| Progression of the agent’s behaviors according to its level of engagement. Frontiers in Computational Neuroscience \| www.frontiersin.org Evaluating the Perceived Valence of Non-Verbal Behaviors The purpose of our evaluation was to test the following hypothesis (H1): the valence of the agents’ non-verbal behaviors is appropriately perceived. We generated eight videos. In each video, we captured the transition from the neutral behavior to one of the eight non- verbal behaviors we designed (see Figure 2). After the transition, the agent maintained its behavior for almost 2 s. All the videos were generated with the same procedure: same appearance of the virtual agent, same position of the camera, same duration of the animation (approximately 4 s). All the videos last exactly 2.3 s and were made the same way: the agent keeps a neutral expression for 0.15 s, then expresses a non-verbal behavior for 2 s and finally maintains a neutral expression for 0.15 s. There was no sound in the videos. We asked subjects to watch each of the eight videos through a web study. After watching each video, subjects rated the valence of the feeling expressed by the agent using 7-point labeled Likert scales. To avoid carryover effects during the evaluation, we used a Latin square design to generate the order of the videos. We recruited a total of 16 participants via mailing lists (56% male). Most of the subjects were French (75%), and between 25 and 35 years old (82%). FIGURE 3 \| Valence ratings associated with AU and BAP combinations. Note that the four facial-expression categories (negative, neutral, mildly positive, and highly positive) were rated in the intended order. criterion adjusted for three comparisons: p < 0.017): (1) highest- rated negative vs. neutral; (2) neutral vs. lowest-rated mild positive; and (3) highest-rated mild positive vs. lowest-rated high positive. Because the parametric assumption of normality was violated for half of the AU and BAP combinations, we report pairwise related-samples comparisons using Wilcoxon Signed Ranks Test. In order to test H1, we ran a one-way repeated measures analysis of variance (ANOVA) with perceived valence as dependent variable. The within-subjects factor was AU and BAP combinations (with eight levels; Figure 2). There was a statistically significant effect of AU and BAP combinations on perceived valence, F(7,105) = 53.91, p < 0.001, η2 is 0.78 (large). Figure 3 shows average valence ratings for each AU and BAP combination with 95% confidence interval (CI). Engagement level Negative Neutral Mild positive High positive AU family Eyebrows AUl (Inner brow raiser) AU2 (Outer brow raiser) Eyes AU43 (Eyes closed) AU5 (Upper lid raiser) AU5 (Upper lid raiser) AU62 (Eyes right) AU64 (Eyes down) AU61 (Eyes left) Cheeks AU6 (Cheek raiser) Mouth AU23 (Lip tightener) AU12 (Lip corner puller) AU12 (Lip corner puller) AU24 (Lip pressor) AU25 (Lips part) AU25 (Lips part) AU18 (Lip pucker) AU26 (Jaw drop) AU26 (Jaw drop) AU15 (Lip corner depressor) AU17 (Chin raiser) BAP vc1roll (Cervical vertebra roll) vc1roll (Cervical vertebra roll) vc1tilt (Cervical vertebra tilt) vc1tiIt (Cervical vertebra tilt) vc1torsion (Cervical vertebra torsion) vl1torsion (Lumbar vertebra torsion) TABLE 1 \| Details of the action units (AUs) and body movement deﬁned to encode the facial expressions and behaviors TABLE 1 \| Details of the action units (AUs) and body movement deﬁned to encode the facial expressions and behaviors for the each level of engagement. Us) and body movement deﬁned to encode the facial expressions and behaviors for the each level of engagement. FIGURE 3 \| Valence ratings associated with AU and BAP combinations. Note that the four facial-expression categories (negative, neutral, mildly positive, and highly positive) were rated in the intended order. FIGURE 3 \| Valence ratings associated with AU and BAP combinations. Note that the four facial-expression categories (negative, neutral, mildly positive, and highly positive) were rated in the intended order. FIGURE 2 \| Progression of the agent’s behaviors according to its level of engagement. time. Hence, the agent displays a new behavior whenever the user maintains a particular level of left-asymmetry for around 2 s. combination of AUs and BAPs. To avoid uncanny animations (e.g., when the agent’s expression freezes) and to obtain smooth transitions between two behaviors, we designed an interpolation function (based on Huang and Pelachaud, 2012) so that the previous behavior fades away while the new one slowly appears, resulting in a blended behavior. To validate the behaviors displayed by the agent, therefore to ensure giving an adequate feedback to the user, we conducted two evaluation studies. In the first one, we asked participants to evaluate each of the agent’s non-verbal behaviors according to their valence (negative or positive) and their naturalness. In the second one, we asked participants to rank four different behaviors using the same valence scale, from the least positive to the most positive. Having different behaviors for the same level of engagement avoids the uncanny repetition of similar expressions and increases realism. The variation of behaviors also ensures that the agent does not freeze when it remains at the same level for a long July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 5 Affective Interaction Through an fNIRS BCI Aranyi et al. TABLE 1 \| Details of the action units (AUs) and body movement deﬁned to encode the facial expressions and behaviors for the each level of engagement. Frontiers in Computational Neuroscience \| www.frontiersin.org Evaluating the Perceived Valence of Non-Verbal Behaviors To assess whether the four categories of agent behavior (negative, neutral, mildly positive, and highly positive) were perceived reliably differently, we performed three post hoc pairwise comparisons using Bonferroni correction (one-tailed p < 0.05 Comparison 1: subjects rated the valence of neutral agent expression (Neutral; M = 4.31, SD = 0.79) significantly higher than that of the highest-rated negative agent expression (Neg1; M = 3.38, SD = 0.89), T = 12, z = −2.66, p (one-tailed) = 0.004, r = 0.66 (large). Comparison 2: subjects rated the valence of the lowest-rated mild positive agent expression (MildPos2: M = 5.56, SD = 0.81) significantly higher than the neutral agent July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 6 Aranyi et al. Affective Interaction Through an fNIRS BCI We selected the HbO signal to provide real-time feedback, which has been applied successfully in research involving affect- related manipulation in the DL-PFC (Tuscan et al., 2013), and in the context of approach/withdrawal-related experimental manipulation (Morinaga et al., 2007). Furthermore, HbO has been reported to be more sensitive to changes in cerebral blood flow than HbR and total (HbT) hemoglobin changes (Hoshi, 2003), and it is characterized with a higher signal- to-noise ration then HbR and HbT (Strangman et al., 2002). Since we used HbO for real-time mapping to the feedback channel, we report post hoc analyses based on the same signal for consistency. expression, T = 11, z = −3.04, p (one-tailed) = 0.001, r = 0.76 (large). Comparison 3: subjects rated the valence of the lowest- rated high positive agent expression (HighPos 2; M = 6.38, SD = 0.72) significantly higher than that of the highest-rated mild positive agent expression (MildPos1; M = 5.69, SD = 0.87), T = 3, z = −2.37, p (one-tailed) = 0.009, r = 0.59 (large). H1 was supported: users could differentiate between the different facial expressions of the agent’s non-verbal behaviors, the valence ratings were in line with the intended order, and the differences between the four main categories of the agent’s facial expressions were characterized with a large effect size. To further investigate if the non-verbal behavior of the agent is interpreted as intended, we conducted a second evaluation addressing the following hypothesis (H2): the non-verbal behaviors validated in the previous evaluation are correctly ranked according to their valence. Evaluating the Perceived Valence of Non-Verbal Behaviors We operationalized BCI input based on asymmetric functional activation in the DL-PFC. A metric of asymmetry was derived real-time by the experimental software by averaging HbO values across the four channels located over the left and right DL-PFC, respectively, then subtracting average Right from average Left. This value, updated with the same 2 Hz frequency of the sampling rate, reflects inter-hemispheric difference in HbO change in micromolar units (µMol/L). We kept the same videos as for our first evaluation and we designed a drag and drop interface allowing users to rank the behaviors expressed in the videos according to perceived valence. To avoid carryover effects during the evaluation, we used a Latin square design to generate the order of the combinations. We recruited a total of 36 subjects via mailing lists (61% male). Most of the subjects were French (61%), and between 25 and 35 years old (81%). The collection of fNIRS data was organized into blocks, each consisting of a sequence of epochs (short time periods with a specific task). We describe the structure of blocks and the experimental tasks during each epoch in the next section. HbO values on each channel were extracted using time synchronization markers. To compensate for the approximately 7 s delay in the hemodynamic response (Bunce et al., 2006), we excluded data from the first 7 s in each epoch and included data from the first 7 s after the completion of each epoch (Figure 4; Aranyi et al., 2015a). We coded the ranking of the behaviors into numerical valence values and computed the mean valence score of each behavior. The results confirmed the ones obtained during the first evaluation: the eight behaviors could be clustered into different groups according to their valence. We checked whether each of the eight behaviors was ranked according to our previous findings. Most of them were correctly ranked (M ≥90%) except for one negative behavior (Neg1: M = 67.36) and the neutral behavior (Neutral: M = 85.88). The behavior Neg1 representing a pouting behavior was sometimes considered more positive than the neutral behavior. This can be explained as the pouting AU is not directly related with negatively valenced emotions, unlike the AU used for the behaviors Neg2 and Neg3. We can also hypothesize that Pouting expressions may be found culturally acceptable rather than negative, especially considering the subjects’ age range. Overall, the results supported H2. Subjects, Procedure, and Feedback Mapping Eighteen english-speaking adult subjects participated in the experiment (eight female, mean age = 35.11 years, SD = 11.25, range: [21; 60]). Data from one subject was discarded due to technical problems during data collection. Subjects were right-handed and were not treated for psychiatric conditions. The study was approved by Teesside University’s Research Ethics Committee. Subjects provided written consent prior to participation and received an online retailer voucher equivalent to $30 upon completing the experimental protocol. MATERIALS AND METHODS We followed the recommendations of Solovey et al. (2009) for the use of fNIRS in a HCI setting. Subjects seated approximately 47′′ (120 cm) away from a 24′′ monitor displaying experimental instructions and stimuli (including visual feedback during NF) in a dimly-lit, quiet room in a comfortable chair to minimize movement artifacts, with the fNIRS sensor positioned on their forehead, covered with non-transparent material to prevent ambient light from reaching the sensors. Subjects were instructed to refrain from talking, frowning and moving their limbs during fNIRS data collection periods within the protocol. Additionally, we applied sliding-window motion artifact rejection (SMAR; Ayaz et al., 2010); each channel used for calculating the asymmetry metric was inspected post hoc to identify motion artifacts during NF. For post hoc analyses, raw data were low-pass BCI Input p We used an fNIR400 Optical Brain Imaging Station by Biopac Systems, with a 16-channel sensor and a fixed 2.5 cm source- detector separation for BCI input (for channel locations, see Ruocco et al., 2014). This device measures intensity changes in two wavelengths (730 nm and 850 nm) to calculate changes in oxygenated and deoxygenated hemoglobin concentration (HbO and HbR, respectively) using the modified Beer-Lambert law (see Ayaz, 2010). Raw data and HbO values were collected with 2 Hz sampling frequency using software provided by the device manufacturer (COBI Studio and FnirSoft v4.1), and was sent to a remote client experimental software over TCP/IP (using FnirSoft DAQ Tools). July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 7 Affective Interaction Through an fNIRS BCI Aranyi et al. FIGURE 4 \| Protocol design, including setting the baseline of the fNIRS system, as well as windowing the data collection during the View and NF epochs to account for the delay in hemodynamic response. FIGURE 4 \| Protocol design, including setting the baseline of the fNIRS system, as well as windowing the data collection during the View and NF epochs to account for the delay in hemodynamic response. filtered using a finite impulse response filter with order 20 and 0.1 Hz cut-off frequency (Ayaz et al., 2010). We included the View epoch in each block as a reference (with the mental arithmetic task unrelated to asymmetry) to support feedback mapping for NF within the same block, in the following way. We defined the threshold (i.e., minimum asymmetry value resulting in feedback) during NF based on asymmetry values collected during View within the same block (Figure 5). The minimum point (Min) for mapping was defined as the mean of asymmetry values during the View epoch plus 1.28 times their SD. Assuming normally distributed asymmetry scores, this threshold would result in no feedback for 90% of the asymmetry values during View. This approach to determine NF threshold is consistent with the original one of Rosenfeld et al. (1995) for EEG-based frontal-asymmetry NF. To determine the maximum point (Max) for mapping (i.e., the asymmetry value resulting in feedback with maximum magnitude), we added the variation range of asymmetry values during View to the threshold Min we defined above. BCI Input Outliers (values outside three SDs from the mean) were removed for calculating the threshold and range in order to prevent extreme values, likely resulting from movement artifacts, exerting an unduly influence on NF mapping. Asymmetry values within the range [Min; Max] during NF were mapped linearly onto the virtual agent’s facial expression, with the same 2 Hz frequency as the acquisition of asymmetry values. Following instructions and a practice block, each participant completed eight identical blocks, consisting of three epochs: Rest, View, and NF. Block design and the timing of epochs are presented in Figure 4 (also see Charles et al., 2015). p g During Rest, subjects were instructed to look at a crosshair at the center of a gray screen, to try to clear their head of thoughts and relax. HbO for each channel within a block was calculated with respect to a baseline measured during the last 10 s of this Rest epoch. During View, subjects were instructed to look at the virtual agent while carrying out a mental arithmetic task: counting backwards from 500 by increments of an integer provided before the epoch in a visual prompt lasting 3 s. Backward counting is the most commonly used mental arithmetic task in fNIRS research (Naseer and Hong, 2015); it was included to control for unwanted mental processes while looking at the virtual agent (see Sarkheil et al., 2015). Subjects were informed that the agent would be unresponsive to their brain input during View. During NF, subjects were instructed to attempt to cheer up the virtual agent using their thoughts. NF was separated from View by a 10 s Rest epoch, and preceded by a 3 s prompt to start engaging with the virtual agent through (positive) thinking. Subjects were told that the agent would be responsive to their brain input during NF, and to expect a few seconds delay in the agent’s reactions due to the nature of the brain input used in the study. Since, due to this delay, the effectiveness of any cognitive strategy could not be seen immediately, we also asked subjects to be patient and apply a strategy consistently throughout a block. They were also shown how a happy agent would look like (i.e., the target state) before the practice block preceding data collection. Frontiers in Computational Neuroscience \| www.frontiersin.org RESULTS AND DISCUSSION FIGURE 6 \| The distribution of block success across subjects. After data screening and filtering, we identified 136 valid blocks for analysis (eight blocks were collected from each 17 subjects). We defined block success as a statistically significant increase in average asymmetry during NF compared to the View epoch within the same block. Block success was determined post hoc using independent t-test with bootstrapping resampling method (1000 samples, 95% CI) with epoch type (View/NF) as independent variable and asymmetry scores as dependent variable, applying Bonferroni correction on a subject basis to control false discovery rate (one-tailed p threshold of 0.0125). (Note that the entire NF epoch was considered to determine block success, not just above-threshold asymmetry values.) Based on this criterion, 70 out of the total 136 blocks (51%) were successful. For a detailed discussion of the design of experimental blocks to support the use of this success criterion, see Aranyi et al. (2015a). Based on their NF performance, we classified subjects as respondents (completed at least one block successfully) and non- respondents (completed no successful blocks). Fifteen of the seventeen valid subjects (88%) were classified as respondents. Additionally, the two non-respondents were not simply failed on the statistical criterion; they have not crossed the feedback threshold at all during the experiment (i.e., received no feedback whatsoever). Eleven out of the 15 respondents (73%) completed at least half of the blocks successfully (Median = 4). No subject completed all blocks successfully (Figure 6). This success rate on the subject basis is identical to what we observed in a separate study applying the same NF protocol (Charles et al., 2015). we used to achieve sufficient statistical power to detect changes in asymmetry. Therefore, we also calculated effect-size metrics (r and Cohen’s d) to characterize the magnitude of NF success, rather than solely relying on statistical significance depending on sampling frequency. The effect-size r is interpreted similarly to the correlation coefficient, its value is constrained between 0 and 1, and according to Cohen (1988) conventions, r = 0.10 corresponds to small, 0.30 to medium, and 0.50 to large effect-size. The smallest r we could detect was 0.18 (small), and average r in successful blocks was 0.75 (large). Furthermore, the distribution of r in successful blocks was negatively skewed (Median The dichotomous (NF success/fail) statistical success criterion described above does not contain information about the magnitude of asymmetry up-regulation during successful NF. BCI Input We were deliberately vague with instructions regarding how to cheer up the agent in order to allow subjects to develop their own strategies. NF was followed by a final Rest epoch lasting 7 s. After completing each block, subjects were asked to describe the strategy they used during NF in broad terms. Additionally, they rated the extent to which the virtual agent’s facial expressions were appropriate to their thought contents during NF on a 1–7 scale. This rating was included to estimate the subjective level of alignment between the subject’s thoughts and the agent’s expressive behavior in each block. Note that View and NF epochs were matched for length and for visual stimulus, with the exception that during NF, subjects could influence the agent’s expressions. Since subjects’ mental activity was controlled during View with a mental arithmetic task theoretically unrelated to DL-PFC asymmetry, increase in asymmetry can be attributed to changes mental activity between the View and NF epochs. This way, using a reference epoch provides a control condition within each block. Rather than providing absolute levels of HbO and HbR, the collected values reflect changes relative to a baseline (Ayaz, 2010), they can be difficult to compare across subjects (Sakatani et al., 2013), and the magnitude of HbO changes can differ substantially across blocks within the same subjects. However, the above strategy for determining feedback threshold and range on a per-block basis mitigates these issues and promotes the comparability of NF epochs both within and between subjects. July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 8 Aranyi et al. Affective Interaction Through an fNIRS BCI FIGURE 5 \| Example of a successful block, where asymmetry during NF is signiﬁcantly larger than during View. FIGURE 5 \| Example of a successful block, where asymmetry during NF is signiﬁcantly larger than during View. Frontiers in Computational Neuroscience \| www.frontiersin.org RESULTS AND DISCUSSION Additionally, the experimental protocol was designed considering the fixed 2 Hz sampling rate of the fNIRS system July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 9 Aranyi et al. Affective Interaction Through an fNIRS BCI FIGURE 7 \| Distribution of the effect-size measure r in successful blocks. FIGURE 8 \| Asymmetry change from View to NF in non-successful and successful blocks. Successful blocks are characterized with a marked asymmetry increase during NF, while non-successful ones are characterized with a slight decrease. FIGURE 8 \| Asymmetry change from View to NF in non-successful and successful blocks. Successful blocks are characterized with a marked asymmetry increase during NF, while non-successful ones are characterized FIGURE 7 \| Distribution of the effect-size measure r in successful blocks. FIGURE 8 \| Asymmetry change from View to NF in non-successful and successful blocks. Successful blocks are characterized with a marked asymmetry increase during NF, while non-successful ones are characterized with a slight decrease. r = 0.81), indicating that successful blocks were predominantly characterized with large asymmetry increase (Figure 7). We calculated Cohen’s d by dividing the difference between mean asymmetry during the NF and View epochs with the pooled SD of asymmetry scores. The resulting d value reflects the distance between the distributions of asymmetry scores collected during successive reference and NF epochs, which can be readily interpreted. Mean d in successful blocks was 2.58 (SD = 1.41), which corresponds to an average 20% overlap in asymmetry scores between successive View and NF epochs (assuming normally distributed values), with a 96% chance that an asymmetry value picked randomly from the NF epoch will be greater than a random asymmetry value from the View epoch (probability of superiority). The smallest effect we could detect (with 40 s epochs sampled at 2 Hz) was d = 0.35 (small), which corresponds to 86% overlap and 60% probability of superiority. This magnitude of increase in asymmetry is comparable to our previous study, where subjects up-regulated left-asymmetry by mentally expressing anger (a negatively valenced, but approach- related affect) towards a virtual agent who was previously identified as mischievous in a narrative context (see Aranyi et al., 2015b). across subjects with time (from Block 1 to 8). Since Block Success was a continuous dichotomy, we calculated the biserial correlation coefficient to assess the linear relationship between Block Success (Yes/No) and block number (1–8). RESULTS AND DISCUSSION The analysis found no practice effect, rb = 0.03, p = 0.76, ns. Figure 8 presents average asymmetry change between successive View and NF epochs. Successful blocks were characterized with a marked increase in asymmetry during NF (M = 0.36, SD = 0.29); conversely, non-successful blocks were characterized with a comparably small asymmetry decrease during NF (M = −0.22, SD = 0.23). We explored how HbO changes on the left and right sides contributed to asymmetry increase from View to NF in successful blocks by conducting a two-way within-subject ANOVA with average HbO change as dependent variable, and Epoch type (View/NF) and Side (Left/Right) as factors. The analysis revealed a significant interaction, F(1,69) = 108.76, p < 0.001, partial η2 = 0.61 (large), which indicates that HbO changed differentially on the left and right sides between consecutive View and NF epochs. This interaction is displayed in Figure 9, showing that average HbO change during View (with counting reference task) did not display lateralized difference, t(69) = 0.53, p = 0.595, n.s., r = 0.06 (small), while average HbO increase on the left size (M = 0.47, SD = 1.05) was significantly and substantially larger than on the right (M = 0.13, SD = 1.07), t(69) = 6.73, p < 0.001, r = 0.63 (large). These findings indicate that our success criterion was sensitive to detect small increases in asymmetry, while asymmetry values on average had small overlap between View and NF epochs: successful blocks were characterized with a marked increase in asymmetry during NF. Additionally, in successful blocks, the r and d effect-size measures were positively and significantly correlated with perceived alignment (the subjective rating of how appropriate were the virtual agent’s facial expressions to the subject’s thoughts during NF): r = 0.56, and r = 0.45, respectively (both p < 0.001). This provides validation from self-report data for the use of effect-size metrics to characterize the magnitude of NF success. Conversely, non-successful blocks were characterized by stability, or even a decrease in left asymmetry (Figure 8). Out of the 66 failed blocks, 15 (23%) showed no statistically significant change in asymmetry from View to NF, while 51 (77%) exhibited a significant decrease in left-asymmetry. Replicating the above two-way within-subject ANOVA on non-successful Since our experiment consisted of a single session, no training effect was expected. Frontiers in Computational Neuroscience \| www.frontiersin.org RESULTS AND DISCUSSION However, a possible practice effect would consist in a linear increase in the proportion of Block Success July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 10 Aranyi et al. Affective Interaction Through an fNIRS BCI FIGURE 9 \| Left-lateralized increase in average oxygenated hemoglobin concentration (HbO) in successful blocks. not consistent with our previous findings using a similar protocol (Aranyi et al., 2015b), where we found no lateralized changes in HbO in failed blocks. These findings show that 51 out of the 136 blocks (37.5%) collected in the experiment where characterized with a decrease in left-asymmetry while subjects reported using cognitive strategies traditionally associated with left-asymmetry up-regulation. However, we note that 13 (25%) of these blocks were actually produced by the two non-responsive subjects (i.e., those who had not completed a single successful block). It also needs to be noted that our subjects did not receive NF training spread over multiple sessions prior to the experiment (e.g., see Rosenfeld et al., 1995; Kotchoubey et al., 2002). Here, subjects’ interaction with the system was restricted to a single short session (eight blocks including a 40 s NF epoch each, plus a practice block); therefore, no learning effects could be expected, while practice effects within a single session were confounded with the effect of fatigue (ratings of task-difficulty are discussed below). We conducted post hoc pairwise comparisons to breakdown the significant interaction effect of Epoch type (View/NF) and Side (Left/Right) in successful blocks. On the left side, average HbO increase from View (M = −0.19, SD = 0.80) to NF (M = 0.47, SD = 1.05) was statistically significant, t(69) = 9.44, p < 0.001, r = 0.75 (large). On the right side, average HbO increase from View (M = −0.17, SD = 0.70) to NF (M = 0.13, SD = 1.07) was also statistically significant, but with a markedly lower effect-size, t(69) = 3.80, p < 0.001, r = 0.42 (medium). These findings indicated that although mean HbO increased on both sides during successful NF, asymmetry resulted from a more pronounced HbO increase on the left side (Figure 10 illustrates FIGURE 9 \| Left-lateralized increase in average oxygenated hemoglobin concentration (HbO) in successful blocks. blocks revealed a significant interaction of Epoch type (View/NF) and Side (Left/Right), F(1,65) = 58.11, p < 0.001, partial η2 = 0.47 (large). CONCLUSION for across successful blocks, whilst Figure 11 shows an example of a single successful NF epoch). Our results confirm that users are able to engage with virtual agents by controlling their mental disposition, through thought contents and cognitive strategies. However, this is the first complete report of a context which is behaviorally realistic on the agent side (unlike, for instance, (Gilroy et al., 2013), which used color saturation instead of realistic facial expressions), and in which BCI input is analyzed together with user experience subjective data. This bilateral increase in HbO may result from increased mental effort when switching from the counting task to NF. Subjects rated the subjective difficulty of both tasks on a 1–7 Likert scale after completing the protocol. Wilcoxon signed-rank test revealed that subjects rated the NF task (M = 5.18, SD = 1.13) more difficult than the Count task (M = 4.12, SD = 1.50), T = 20, z = 2.321, p = 0.020, r = 0.56 (large). Additionally, each subject answered six questions extracted from a generic rapport questionnaire (Gratch et al., 2007), rated on a 1–7 Likert scale, after completing the experimental protocol. The scale had satisfactory internal consistency (Cronbach’s alpha = 0.73). We calculated Spearman’s rho non-parametric correlation coefficient to explore the relationship between alignment and NF success. There was a strong and statistically significant positive relationship between perceived alignment the number of successfully completed blocks, rs = 0.62, p < 0.01. Previous work has associated rapport primarily with activation of the VM-PFC (Gupta Gordon et al., 2014), while empathy has been considered closely related to approach, whose relevant area is the DL-PFC. At the same time, VM-PFC is generally considered inaccessible to fNIRS (Doi et al., 2013), and our results have evidenced a significant correlation between self-reported magnitude of rapport (characterized as perceived alignment) and DL-PFC asymmetry. To resolve this apparent contradiction, we should note that the definition of rapport has been sometimes strongly influenced by alignment (Menenti et al., 2012) and mimicry (Lakin and Chartrand, 2003; Wang and Hamilton, 2015), in which the affective component is less prominent. Our definition of rapport is closer to empathy without identifying entirely with it, because of the lack of background information on the virtual agent and the fact that it is not expressing a pre-existing emotional state. RESULTS AND DISCUSSION In failed blocks, average HbO during View did not display lateralized difference, t(65) = −0.33, p = 0.740, ns, r = 0.04 (small), while during NF, average HbO change on the left size (M = 0.24, SD = 0.98) was significantly lower than on the right (M = 0.47, SD = 0.91), t(65) = −3.93, p < 0.001, r = 0.44 (medium). This is FIGURE 10 \| Mean and standard error of HbO across successful blocks (N = 70), for left (red) and right (blue) sides separately. The signal on the two sides overlaps completely during View, while left rises above right during NF. FIGURE 10 \| Mean and standard error of HbO across successful blocks (N = 70), for left (red) and right (blue) sides separately. The signal on the two sides overlaps completely during View, while left rises above right during NF. July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 11 Affective Interaction Through an fNIRS BCI Aranyi et al. FIGURE 11 \| Topographic snapshots of asymmetric HbO increase in the dorsolateral prefrontal cortex (DL-PFC) during a successful NF epoch. symmetric HbO increase in the dorsolateral prefrontal cortex (DL-PFC) during a successful NF epoch. FIGURE 11 \| Topographic snapshots of asymmetric HbO increase in the dorsolateral prefrontal cortex (DL-PFC) during a successful NF epoch. FIGURE 11 \| Topographic snapshots of asymmetric HbO increase in the dorsolateral prefrontal cortex (DL-PFC) duri REFERENCES Charles, F., Pecune, F., Aranyi, G., Pelachaud, C., and Cavazza, M. (2015). ‘‘ECA control using a single affective user dimension,’’ in Proceedings of the 2015 ACM on International Conference on Multimodal Interaction (Seattle, WA), 183–190. Albrecht, I., Schröder, M., Haber, J., and Seidel, H. (2005). Mixed feelings: expression of non-basic emotions in a muscle-based talking head. Virtual Real. 8, 201–212. doi: 10.1007/s10055-005-0153-5 Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. 2nd Edn. New Jersey, NJ: Hillsdale. Aranyi, G., Cavazza, M., and Charles, F. (2015a). ‘‘Using fNIRS for prefrontal- asymmetry neurofeedback: methods and challenges,’’ in Symbiotic Interaction, eds B. Blankertz, G. Jacucci, L. Gamberini, A. Spagnolli and J. Freeman (Switzerland: Springer International Publishing), 7–20. Davidson, R. J. (1992). Anterior cerebral asymmetry and the nature of emotion. Brain Cogn. 20, 125–151. doi: 10.1016/0278-2626(92)90065-t Davidson, R. J., Ekman, P., Saron, C. D., Senulis, J. A., and Friesen, W. V. (1990). Approach-withdrawal and cerebral asymmetry: emotional expression and brain physiology:. J. Pers. Soc. Psychol. 58, 330–341. doi: 10.1037/0022- 3514.58.2.330 Aranyi, G., Charles, F., and Cavazza, M. (2015b). ‘‘Anger-based BCI using fNIRS neurofeedback,’’ in Proceedings of the 28th Annual ACM Symposium on User Interface Software and Technology (UIST ’15) (ACM, New York, NY), 511–521. Doherty-Sneddon, G., Bruce, V., Bonner, L., Longbotham, S., and Doyle, C. (2002). Development of gaze aversion as disengagement from visual information. Dev. Psychol. 38, 438–445. doi: 10.1037/0012-1649.38. 3.438 Ayaz, H. (2010). Functional Near Infrared Spectroscopy Based Brain Computer Interface. Philadelphia, PA: PhD thesis, Drexel University. Ayaz, H., Izzetoglu, M., Shewokis, P. A., and Onaral, B. (2010). Sliding-window motion artifact rejection for functional near-infrared spectroscopy. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2010, 6567–6570. doi: 10.1109/IEMBS.2010. 5627113 Doi, H., Nishitani, S., and Shinohara, K. (2013). NIRS as a tool for assaying emotional function in the prefrontal cortex. Front. Hum. Neurosci. 7:770. doi: 10.3389/fnhum.2013.00770 Ayaz, H., Shewokis, P. A., Bunce, S., and Onaral, B. (2011). An optical brain computer interface for environmental control. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2011, 6327–6330. doi: 10.1109/IEMBS.2011.6091561 Ekman, P., Friesen, W. V., and Hager, J. C. (2002). Facial Action Coding System: A Technique for the Measurement of Facial Movement. Palo Alto, CA: Consulting Psychologists Press. Benbasat, I., Dimoka, A., Pavlou, P. A., and Qiu, L. (2010). ‘‘Incorporating social presence in the design of the anthropomorphic interface of recommendation agents: insights from an fMRI study,’’ in International Conference on Information Systems (St. Louis, MO), 228. FUNDING regardless of the actual cognitive strategy employed. Strictly speaking, the true ‘‘rapport’’ hypothesis is only fully realized when the subject’s cognitive strategies targets the agent itself, the issue with positive thoughts strategies not being so much the use of valence but the fact that some positive thoughts could be self-centered. The above finding is attributable to individual subjects adopting conservative cognitive strategies, rather than to an inherent limitation of the model: this can be explained in particular considering the limited NF training that subjects have been offered. In earlier work, we have demonstrated that successful interaction with virtual agents could be realized through approach alone, by using an anger paradigm (Aranyi et al., 2015b), in which approach is fully dissociated from valence (Harmon-Jones, 2003). In the context of rapport, a mixed contribution from approach and valence seems unavoidable, the decisive factor being that positive thoughts should be directed at the agent rather than self- centered. This work was supported by the European project H2020 ARIA- VALUSPA (No 645378) and the French ANR project MOCA (ANR-12-CORD-019). It was partially performed within the Labex SMART (ANR-11-LABX-65) supported by French state funds managed by the ANR within the Investissements d’Avenir programme under reference ANR-11-IDEX-0004-02. AUTHOR CONTRIBUTIONS GA: (BCI) protocol design, experimental setup, experiments, statistical analysis, write-up. FP: (ECA) software design and implementation, ECA experimental setup and experiments, write-up. FC: (BCI) software design and implementation, experiments, write-up. CP: development of research question, research expertise and guidance, write-up. MC: development of research question, research expertise and guidance, write-up. CONCLUSION Following each block, subjects were asked to describe their cognitive strategies to engage with the virtual agent, and thought contents during NF epochs (regardless of the amount of feedback they received during the NF epoch). We identified two broad categories of cognitive strategy (i.e., a set of thought contents to concentrate on during NF): direct and indirect. Direct strategies involved trying to cheer up the agent in an internal dialog, imagery involving the agent (e.g., reaching out to her), expressing empathy or otherwise engaging mentally with her. Indirect strategies involved recollection of pleasant personal memories, imagery of pleasant activity unrelated to the agent, and thinking about pleasant things (e.g., singing a song in one’s head). Chi-squared test revealed no significant association between blocks success (Success/Fail) and strategy type (Direct/Indirect), χ2 (1) = 2.52, p = 0.112, Cramer’s V = 0.14 (small). Indirect strategy is more prominent in failed blocks (41 out of 66; 62%) whilst, in successful blocks, there was an almost even split in the use of direct and indirect strategies (36 and 34, respectively). A potential limitation of these experiments is the non- dissociation between approach and valence when measuring PFC asymmetry, valence also impacting activity in the DL-PFC (Berkman and Lieberman, 2010). This can be also evidenced by the subjects’ self-reporting of cognitive strategies, which showed an even split between cognitive strategies emphasizing approach (reaching out, targeting the agent itself), and those emphasizing positive thoughts in general [often self-centered, as previously reported in other, fMRI-based PFC asymmetry NF experiments (Zotev et al., 2014)]. However, in terms of user experience, the agent’s response was perceived as appropriate July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 12 Affective Interaction Through an fNIRS BCI Aranyi et al. REFERENCES Etemad, S. A., Arya, A., Parush, A., and DiPaola, S. (2015). Perceptual validity in animation of human motion. Comput. Animat. Virtual Worlds 27, 58–71. doi: 10.1002/cav.1631 Gilroy, S. W., Porteous, J., Charles, F., Cavazza, M., Soreq, E., Raz, G., et al. (2013). ‘‘A brain-computer interface to a plan-based narrative,’’ in Proceeding IJCAI ’13 Proceedings of the 23rd International Joint Conference on Artificial Intelligence (Beijing), 1997–2005. Berkman, E. T., and Lieberman, M. D. (2010). Approaching the bad and avoiding the good: lateral prefrontal cortical asymmetry distinguishes between action and valence. J. Cogn. Neurosci. 22, 1970–1979. doi: 10.1162/jocn.2009.21317 Bui, T. D., Heylen, D., Poel, M., and Nijholt, A. (2001). ‘‘Generation of facial expressions from emotion using a fuzzy rule based system,’’ in Australian Joint Conference on Artificial Intelligence , eds D. C. M. Stumptner and M. Brooks (London, UK: Springer-Verlag), 83–94. Gordon, R. G., Tranel, D., and Duff, M. C. (2014). The physiological basis of synchronizing conversational rhythms: the role of the ventromedial prefrontal cortex. Neuropsychology 28, 624–630. doi: 10.1037/neu00 00073 Bunce, S. C., Izzetoglu, M., Izzetoglu, K., Onaral, B., and Pourrezaei, K.. (2006). Functional near-infrared spectroscopy. IEEE Eng. Med. Biol. Mag. 25, 54–62. doi: 10.1109/memb.2006.1657788 Grammer, K., and Oberzaucher, E. (2006). The reconstruction of facial expressions in embodied systems. ZiF: Mitteilungen 2, 14–31. Gratch, J., Wang, N., Gerten, J., Fast, E., and Duffy, R. (2007). ‘‘Creating rapport with virtual agents,’’ in Intelligent Virtual Agents, eds C. Pelachaud, J.-C. Martin, E. André, G. Chollet, K. Karpouzis and D. Pelé (Paris), 125–138. Burgoon, J. K., Buller, D. B., Hale, J. L., and Turck, M. A. (1984). Relational messages associated with nonverbal behaviors. Hum. Commun. Res. 10, 351–378doi: 10.1111/j.1468-2958.1984.tb00023.x Hall, L., Woods, S., Aylett, R., Newall, L., and Paiva, A. (2005). ‘‘Achieving empathic engagement through affective interaction with synthetic characters,’’ in Proceeding ACII’05 Proceedings of the First international conference on Affective Computing and Intelligent Interaction (Beijing), 731–738. Cavazza, M., Aranyi, G., Charles, F., Porteous, J., Gilroy, S., Jackont, G., et al. (2014). ‘‘Frontal alpha asymmetry neurofeedback for brain-computer interfaces,’’ in 6th International Brain-Computer Interface Conference (Graz, Austria). July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 13 Aranyi et al. Affective Interaction Through an fNIRS BCI Harmon-Jones, E. (2003). Clarifying the emotive functions of asymmetrical frontal cortical activity. Psychophysiology 40, 838–848. doi: 10.1111/1469- 8986.00121 Ruocco, A. C., Rodrigo, A. H., Lam, J., Di Domenico, S. REFERENCES I., Graves, B., and Ayaz, H. (2014). A problem-solving task specialized for functional neuroimaging: validation of the Scarborough adaptation of the Tower of London (S-TOL) using near-infrared spectroscopy. Front. Hum. Neurosci. 8:185. doi: 10. 3389/fnhum.2014.00185 Hoshi, Y. (2003). Functional near-infrared optical imaging: utility and limitations in human brain mapping. Psychophysiology 40, 511–520. doi: 10.1111/1469- 8986.00053 Sakatani, K., Takemoto, N., Tsujii, T., Yanagisawa, K., and Tsunashima, H. (2013). NIRS-based neurofeedback learning systems for controlling activity of the prefrontal cortex. Adv. Exp. Med. Biol. 789, 449–454. doi: 10.1007/978-1-4614- 7411-1_60 Huang, L., Morency, L.-P., and Gratch, J. (2011). ‘‘Virtual rapport 2.0 Proceeding IVA’11’’, Proceedings of the 10th international conference on Intelligent virtual agents (Reykjavik), 68–79. g y j Huang, J., and Pelachaud, C. (2012). ‘‘Hierarchical body animation pipeline for virtual agent,’’ in Proceedings of the 12th International Conference on Intelligent Virtual Agents (Santa-Cruz) 355–362. Sarkheil, P., Zilverstand, A., Kilian-Hütten, N., Schneider, F., Goebel, R., and Mathiak, K. (2015). fMRI feedback enhances emotion regulation as evidenced by a reduced amygdala response. Behav. Brain Res. 281, 326–332. doi: 10.1016/j. bbr.2014.11.027 Kotchoubey, B., Kübler, A., Strehl, U., Flor, H., and Birbaumer, N. (2002). Can humans perceive their brain states? Conscious. Cogn. 11, 98–113. doi: 10. 1006/ccog.2001.0535 Schilbach, L., Wilms, M., Eickhoff, S. B., Romanzetti, S., Tepest, R., Bente, G., et al. (2010). Minds made for sharing: initiating joint attention recruits reward- related neurocircuitry. J. Cogn. Neurosci. 22, 2702–2715. doi: 10.1162/jocn. 2009.21401 Lakin, J. L., and Chartrand, T. L. (2003). Using nonconscious behavioral mimicry to create affiliation and rapport. Psychol. Sci. 14, 334–339. doi: 10.10.1111/1467- 9280.14481 Schilbach, L., Wohlschlaeger, A. M., Kraemer, N. C., Newen, A., Shah, N. J., Fink, G. R., et al. (2006). Being with virtual others: neural correlates of social interaction. Neuropsychologia 44, 718–730. doi: 10.1016/j.neuropsychologia. 2005.07.017 Light, S. N., Coan, J. A., Zahn-Waxler, C., Frye, C., Goldsmith, H. H., and Davidson, R. J. (2009). Empathy is associated with dynamic change in prefrontal brain electrical activity during positive emotion in children. Child Dev. 80, 1210–1231. doi: 10.1111/j.1467-8624.2009.01326.x Solovey, E. T., Girouard, A., Chauncey, K., Hirshfield, L. M., Sassaroli, A., Zheng, F., et al. (2009). ‘‘Using fNIRS brain sensing in realistic HCI settings: experiments and guidelines,’’ in Proceedings of the 22nd Annual Acm Symposium on User Interface Software and Technology (Victoria, BC), 157–166. Lindquist, K. A., Wager, T. D., Kober, H., Bliss-Moreau, E., and Barrett, L. F. (2012). REFERENCES The brain basis of emotion: a meta-analytic review. Behav. Brain Sci. 35, 121–143. doi: 10.1017/S0140525X11000446 Mattout, J. (2012). Brain-computer interfaces: a neuroscience paradigm of social interaction? A matter of perspective. Front. Hum. Neurosci. 6:114. doi: 10. 3389/fnhum.2012.00114 Spielberg, J. M., Miller, G. A., Warren, S. L., Engels, A. S., Crocker, L. D., Banich, M. T., et al. (2012). A brain network instantiating approach and avoidance motivation. Psychophysiology 49, 1200–1214. doi: 10.1111/j.1469-8986.2012. 01443.x Menenti, L., Pickering, M. J., and Garrod, S. C. (2012). Towards a neural basis of interactive alignment in conversation. Front. Hum. Neurosci. 6:185. doi: 10. 3389/fnhum.2012.00185 Strangman, G., Culver, J. P., Thompson, J. H., and Boas, D. A. (2002). A quantitative comparison of simultaneous BOLD fMRI and NIRS recordings during functional brain activation. Neuroimage 17, 719–731. doi: 10. 1016/s1053-8119(02)91227-9 Morinaga, K., Akiyoshi, J., Matsushita, H., Ichioka, S., Tanaka, Y., Tsuru, J., et al. (2007). Anticipatory anxiety-induced changes in human lateral prefrontal cortex activity. Biol. Psychol. 74, 34–38. doi: 10.1016/j.biopsycho.2006.06.005 Tuscan, L. A., Herbert, J. D., Forman, E. M., Juarascio, A. S., Izzetoglu, M., and Schultheis, M. (2013). Exploring frontal asymmetry using functional near- infrared spectroscopy: a preliminary study of the effects of social anxiety during interaction and performance tasks. Brain Imaging Behav. 7, 140–153. doi: 10. 1007/s11682-012-9206-z Mühl, C., Allison, B., Nijholt, A., and Chanel, G. (2014). A survey of affective brain computer interfaces: principles, state-of-the-art and challenges. Brain Comput. Interf. 1, 66–84. doi: 10.1080/2326263x.2014.912881 f Naseer, N., and Hong, K. S. (2015). fNIRS-based brain-computer interfaces: a review. Front. Hum. Neurosci. 9:172. doi: 10.3389/fnhum.2015. 00172 Wang, Y., and Hamilton, A. F. de C. (2015). Anterior medial prefrontal cortex implements social priming of mimicry. Soc. Cogn. Affect. Neurosci. 10, 486–493. doi: 10.1093/scan/nsu076 Niewiadomski, R., and Pelachaud, C. (2007). ‘‘Model of facial expressions management for an embodied conversational agent,’’ in Affective Computing and Intelligent Interaction (ACII), eds A. C. R. Paiva, R. Prada and R. W. Picard (Lisbon), 12–23. Zotev, V., Phillips, R., Young, K. D., Drevets, W. C., and Bodurka, J. (2013). Prefrontal control of the amygdala during real-time fMRI neurofeedback training of emotion regulation. Plos One 8:e79184. doi: 10.1371/journal.pone. 0079184 Niewiadomski, R., Hyniewska, S. J., and Pelachaud, C. (2011). Constraint-based model for synthesis of multimodal sequential expressions of emotions. IEEE Trans. Aff. Comput. 2, 134–146. doi: 10.1109/t-affc.2011.5 Zotev, V., Phillips, R., Yuan, H., Misaki, M., and Bodurka, J. (2014). July 2016 \| Volume 10 \| Article 70 REFERENCES Self- regulation of human brain activity using simultaneous real-time fMRI and EEG neurofeedback. Neuroimage 85, 985–995. doi: 10.1016/j.neuroimage.2013. 04.126 Nozawa, T., and Kondo, T. (2009). ‘‘Effectiveness of intrinsically motivated adaptive agent for sustainable human-agent interaction,’’ in Neural Information Processing, eds C. S. Leung M. Lee and J. H. Chan (Bangkok), 179–188. Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Ochs, M., Diday, E., and Afonso, F. (2016). From the symbolic analysis of virtual faces to a smiles machine. IEEE Trans. Cybern. 46, 401–409. doi: 10. 1109/TCYB.2015.2411432 Prendinger, H., and Ishizuka, M. (2005). The empathic companion: a character- based interface that addresses users’ affective states. Int. J. Appl. Artif. Intell. 19, 267–285. doi: 10.1080/08839510590910174 Copyright © 2016 Aranyi, Pecune, Charles, Pelachaud and Cavazza. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Rosenfeld, J. P., Cha, G., Blair, T., and Gotlib, I. H. (1995). Operant (biofeedback) control of left-right frontal alpha power differences: potential neurotherapy for affective disorders. Biofeedback Self Regul. 20, 241–258. doi: 10.1007/bf01474516 July 2016 \| Volume 10 \| Article 70 Frontiers in Computational Neuroscience \| www.frontiersin.org 14
https://openalex.org/W4212996289	https://www.researchsquare.com/article/rs-100097/v1.pdf?c=1604691596000	English	null	MSCs‑derived exosomes attenuate ischemia-reperfusion brain injury and inhibit microglia apoptosis via exosomal miR-26a-5p mediated suppression of CDK6	Research Square (Research Square)	2,020	cc-by	6,721	MSCs‑derived exosomes attenuate ischemia- reperfusion brain injury and inhibit microglia apoptosis via exosomal miR-26a-5p mediated suppression of CDK6 Chang Cheng xinqiao hospital third military medical university Xiuying Chen xinqiao hospital third military medical university Yuhan Wang xinqiao hospital third military medical university Wenchao Cheng xinqiao hospital third military medical university Xuzheng Zuo xinqiao hospital thirf military medical university Weiju Tang xinqiao hospital third military medical university Wen Huang (  WenHuangChongqing@163.com ) Xinqiao Hospital third military medical university http MSCs‑derived exosomes attenuate ischemia- reperfusion brain injury and inhibit microglia apoptosis via exosomal miR-26a-5p mediated suppression of CDK6 Chang Cheng xinqiao hospital third military medical university Xiuying Chen xinqiao hospital third military medical university Yuhan Wang xinqiao hospital third military medical university Wenchao Cheng xinqiao hospital third military medical university Xuzheng Zuo xinqiao hospital thirf military medical university Weiju Tang xinqiao hospital third military medical university Wen Huang (  WenHuangChongqing@163.com ) Xinqiao Hospital third military medical university https://orcid.org/0000-0002-0822-2690 Research article Keywords: Exosomes, Mesenchymal stromal cells, miR-26a-5p, CDK6, ischemia-reperfusion inju Posted Date: November 6th, 2020 DOI: https://doi.org/10.21203/rs.3.rs-100097/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International Lic Read Full License Research article Page 1/18 Abstract Background: This study aims to explore the role of mesenchymal stromal cells (MSCs)-derived exosomes (MSCs-Exo) in the cerebral ischemia-reperfusion (I/R) injury. Methods: Exosomes were isolated from MSCs of adult C57BL/6J mice by the gradient centrifugation method. The expression of miR-26a-5p and CDK6 in MSCs-Exo and mice brain tissues were evaluated by qRT-PCR and western blot. MiR-26a-5p mimics and miR-NC were transfected into MSCs, and exosomes were isolated from the stably expressing MSCs. Then MSCs-Exo-miR-26a-5p mimics or MSCs-Exo-miR-NC was injected into mice through the tail vein, or added into medium to stimulate BV-2 cells. Cell viability was evaluated by CCK-8 assay. Cell apoptosis was detected by flow cytometry. The apoptosis in brain tissues was evaluated by TUNEL staining assay. Bioinformatics analysis and luciferase reporter assay were performed to determine the relationship between miR-26a-5p and CDK6. Results: MiR-26a-5p was downregulated and CDK6 was upregulated in MSCs-Exo of MCAO and OGD model. MSCs-Exo-miR-26a-5p mimics significantly reduced cell apoptosis of OGD-injured BV-2 cells. MSCs-Exo-miR-26a-5p mimics significantly reduced infarct volume of MCAO-induced mice. Luciferase reporter assay revealed that CDK-6 was a target of miR-26a-5p. In addition, MSCs-Exo-miR-26a-5p mimics significantly decreased the expression of CDK6 in both OGD-induced BV-2 cells and MCAO-treated mice brains. Conclusion: Our results indicated that MSCs‑Exo attenuated I/R injury in mice by inhibiting microglia apoptosis via exosomal miR-26a-5p mediated suppression of CDK6. Our study shed light on the application of MSC-Exo as a potential therapeutic tool for cerebral I/R injury. Background Ischemic stroke is becoming a major leading cause of disability and death worldwide(1). Approximately 85% of all reported strokes are because of cerebral ischemia that always occurs when an embolus or thrombus blocks the major cerebral artery and then leads to the cell death(2). The primary treatments for ischemic stroke are recanalization therapies, which are proved to replenish nutrients and oxygen, as well as remove toxic metabolites(3). The identification of new therapies or understanding of critical mediators participated in the progression of cerebral ischemic stroke is still urgent. Exosomes are nano-scale messengers carrying bio-molecular cargo of RNA, DNA, and Proteins, and stem cell secreted exosome has been reported that can regulate various autocrine and paracrine functions to alter cell micro-environment and progression in various human diseases(4). The potential clinical values of exosomes can be attributed to their characteristics including their surface markers and molecular cargoes(5), and their ability to cross the brain-blood barrier (BBB)(6), and their potential functions as mediators of the regenerative responses(7). Recently, stem cell therapy has becoming a promising therapeutic option for some neurological disorders including Stroke, Parkinson’s disease, Amyotrophic Lateral Sclerosis and Huntington’s disease(8). Because the greater benefit on tissue regeneration and Page 2/18 Page 2/18 repair than the stem cells themselves, stem cell-derived exosomes that are responsible for their therapeutic benefits have begun to be noticed. One of the most popular stem cell-derived exosomes which are widely applied in neurological diseases researches are MSCs-Exo(9, 10). However, the research about MSCs-Exo in cerebral ischemia-reperfusion (I/R) injury is lack and attracted us to pay more attention on it. MiRNAs are found in both intracellular and extracellular environments, and also detected in exosomes participated cell-cell communication(11). Increasing reports have revealed a large number of exosome- derived miRNAs in cerebral ischemic stroke. For example, exosomal miR-223 was significantly upregulated in acute ischemic stroke, and high level of exosomal miR-223 was associated with the occurrence of acute ischemic stroke and stroke severity(12). M2 microglia-derived exosomes has been showed to protect the mouse brain against I/R injury via exosomal miR-124(13). Exosomes-derived miR- 26a-5p, newly identified exosomal miRNA, has been revealed to play important roles in human diseases. Human bone MSCs-derived exosomes overexpressing miRNA-26a-5p significantly alleviate osteoarthritis through the downregulation of PTGS2(14). Dysregulation of exosomal miRNAs such as miR-26a-5p may affect the regulatory pathways associated with clopidogrel-induced liver injury(15). Background In addition, one previous study has demonstrated that miR-26a-5p is downregulated in myocardial I/R injury, and overexpression of miR-26a-5p effectively improves viability and inhibits apoptosis in cardiomyocytes upon I/R injury by inhibiting PTEN expression(16). Although exosomal miR-26a-5p play crucial functions in various diseases including I/R injury of some organs, its function and molecular mechanism in cerebral I/R injury remains unclear. CDK6 is known as a classic cell cycle kinase by forming complexes with D-type cyclins, which can phosphorylate the Retinoblastoma protein (Rb) to regulate transition from G1 to S phase(17). Previous studies have demonstrated that CDK6 is closely involved in the biological processes including cell proliferation, apoptosis and the transition from G1 to S phase in cerebral ischemic stroke(18, 19). Interestingly, Huang et al. found that miR-26a-5p inhibits the growth of breast cancer cells by downregulating the expression of CDK6(20), suggesting that CDK6 might be a direct or indirect downstream gene. Meanwhile, a miRNA profiling and bioinformatics prediction performed by Canturk et at. indicated that the CDK6 was predicted to be a target of miR-26a-5p, and the axis was significantly associated with the progression of bladder cancer(21), indicating that CDK6 might be a direct target of miR-26a-5p. In the present study, we explored the role of exosomal miR-26a-5p/CDK6 axis in cerebral I/R injury. Taken together, our study demonstrated that MSCs‑derived exosomes overexpressing miR-26a-5p effectively reduced cell apoptosis of OGD-treated microglia cell line BV-2 cells in vitro, and also attenuated MCAO- induced infarct volume in mice, suggesting that exosomal miR-26a-5p might be a potential therapeutic target for cerebral ischemic stroke. Characterization of MSCs-derived Exosomes For transmission electron microscopy (TEM), isolated exosomes were fixed with 1% glutaraldehyde, then a drop of fixed exosomes was spotted onto a formvar/carbon-coated grid and negatively stained with 3% aqueous phosphotungstic acid for 1 min. Subsequently, MSCs-derived exosomes were observed under TEM (Hitachi, Tokyo, Japan, SU-8010). For flow cytometry analysis of exosomes, MSCs-derived exosomes were mixed with 3 μm aldehyde/sulfate latex beads (Invitrogen, Batch Num: 979383) for 10 min with continuous rotation. 1M glycine in PBS containing 2% BSA was added into the mixture to stop the reaction. Beads coated with exosomes were incubated with antibodies CD63-FITC (Lot: GR320523-9, Abcam), CD81-PE (Cat: MA5-17941, Invitrogen) at 37 ℃ for 25 min. Then a FCM flow cytometry (BD FACSalibur) was used to detect the mesenchymal markers. Methods Page 3/18 Isolation of MSCs-derived Exosomes Mesenchymal stem cells (MSCs) were isolated from six adult C57BL/6J mice by the whole bone marrow adherence method as previously described(22). 2×106 MSCs were cultured in 100 mm dishes, MSCs were washed with PBS and kept in fetal bovine serum (FBS)-free L-DMEM medium (Gibco, CA, USA) for 48 h when cells reached nearly 80 ~ 90% confluence. Then the supernatant was collected and subject to sequential centrifugation to obtain the exosomes according to a previous study(23). The precipitated exosomes were stored at − 80 ℃. Animal model of cerebral ischemia-reperfusion (I/R) A total of 30 adult C57BL/6J mice (8 weeks old and 250 g in weight) were provided from Anima Center of Third Military Medical University (Chongqing, China). All animal producers were approved by the Institutional Animal Care and Use Committee of Anima Center of Third Military Medical University (Chongqing,China).. Mice with cerebral ischemia was induced by right middle cerebral artery occlusion (MCAO) as described previously with minor modifications(24). Reperfusion was performed by withdrawing the suture 1 h after MCAO. For sham group, mice accepted the same operation except MCAO procedure. 2 h after reperfusion, 200 μL/mice MSCs-exosomes-miR-NC or MSCs-exosomes-miR-26a-5p mimics (RiboBio, Guangzhou, China) were immediately injected through the tail vein. The mice in the control group were given an equal volume of normal saline (n = 6 in each group). All mice were divided into four groups: Sham group, MCAO group, MCAO + MSCs-exosome-miR-NC, and MCAO + MSCs- exosome-miR-26a-5p mimics group. MSCs of OGD model MSCs was cultured with in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, USA) supplemented with 10% (v/v) fetal bovine serum (FBS, Biological Industries, USA), 100 U/mL penicillin and 100 μg/mL streptomycin (Gibco, USA) at 37°C under 5% CO2. To mimics the ischemia condition in vitro, cells were firstly induced by oxygen-glucose deprivation (OGD) by using deoxygenated glucose-free DMEM medium in an incubator with 95% N2, 5% CO2 for 1, 2 and 4 h, then cells were transferred into normal condition for an additional 24 h for re-oxygenation. Cells treated without OGD were used as the control group. Cell transfection MSCs were transfected with 50 nM miR-26a-5p mimics or miR-NC (RiboBio, Guangzhou, China) by using Lipofectamine® 2000 Transfection Reagent (Gibco Life Technologies) according to the manufacturer’s instructions. The sequences used in this study as follows: miR-26a-5p mimics: 5′- UUCAAGUAAUCCAGGAUAGGCU-3′; miR-NC: 5′-UUCUCCGAACGUGUCACGUTT-3′. When needed, the exosomes were isolated from MSCs transfected with miR-26a-5p mimics or miR-NC and used for the treatment of microglia cell line BV-2 cells. Determination of infarct size After intraperitoneal injection of 3% sodium pentobarbital, the brains of mice in different groups were removed, placed in 4% PFA overnight and then fully dehydrated in 30% sucrose for 2 days. The brain tissue was then cut into coronal sections with approximately 2 mm in thickness, and then quickly incubated in 2% TTC solution (Sigma) at 37°C for 15 min. Subsequently, the sections were fixed by 4% paraformaldehyde and photographed under a ×100 optical microscope (Olympus Corporation). The Page 4/18 Page 4/18 Page 4/18 image analysis software Image J 1.43 (National Institutes of Health) was used to evaluate the relative infarct percentage according to the following equation: infarct size = 100% × (infarcted volume/total brain volume). image analysis software Image J 1.43 (National Institutes of Health) was used to evaluate the relative infarct percentage according to the following equation: infarct size = 100% × (infarcted volume/total brain volume). The treatment of microglia To simulate an in vivo environment of MCAO in microglia, BV-2 cells (Bioleaf, Shanghai, China) were induced by OGD/R treatment as similarly as the MSCs of OGD model above. Then 200 μg/mL MSCs- derived exosomes with overexpression of miR-26a-5p mimics or miR-NC were added into the culture medium to explore the effect of MSCs-Exo on microglia function. CCK-8 assay Cell viability was evaluated by using a Cell Counting Kit-8 (CCK-8, Dojindo Molecular Technologies, Gaithersburg, MD). In brief, MSCs were seeded into 96-well plates overnight. After the treatment, 10 ul of CCK-8 reagent was added to each well at 24, 48, 72 and 96 h and then incubated for another 4 h. The absorbance at 450 nm was detected with a microplate reader. The absorbance was detected at 450 nm with a microplate reader. Apoptosis analysis Cell apoptosis was analyzed by using a commercialized Annexin V-FITC/PI Apoptosis Detection Kit (YEASEN, Shanghai) according to the manufacturer’s protocols. Briefly, BV-2 cells were washed with PBS, and re-suspended in 100 ul of binding buffer, and 5 ul of Annexin V-FITC and 10 ul of propidium iodide (PI) were gently mixed and added into cell suspension, then incubated in the dark for 15 min. Finally, cell apoptosis was detected by flow cytometry (BD FACS Canto II, USA). Luciferase reporter assay The putative binding sites between miR-26a-5p and 3’-UTR of CDK6 were predicted by starBase (http://starbase.sysu.edu.cn/). The mutant-type (MUT) of 3’-UTR (CDK6- MUT) and wild type (WT) of 3’- UTR (CDK6-WT) were amplified and cloned into the pmirGLO dual luciferase reporter vector (Promega, Madison, WI, USA). Then the luciferase reporter plasmids were co-transfected with miR-26a-5p mimics or Page 5/18 Page 5/18 miR-NC into 293T cells by using Lipofectamine® 2000 Transfection Reagent. 48 h fater transfection, cells were lysed and the relative luciferase activity was detected by the dual-luciferase reporter gene assay (Promega). miR-NC into 293T cells by using Lipofectamine® 2000 Transfection Reagent. 48 h fater transfection, cells were lysed and the relative luciferase activity was detected by the dual-luciferase reporter gene assay (Promega). Western blot Total protein of mice brains or cultured cells was isolated by using RIPA lysis buffer (Beyotime Institute of Biotechnology). Approximately equal amounts of protein were separated by 10% SDS-PAGE and transferred onto PVDF membranes (Millipore). After blocking with 5% skim milk, the membranes were incubated with primary antibodies including CD9, CD63, CD81, HSP70, CDK6, cleaved Caspase3 (c- Caspase3), cleaved PARP, and GAPDH antibodies (diluted into 1:1000; Cell Signaling Technology, USA) overnight at 4°C. On the next day, the membranes were incubated with HRP-labeled secondary antibody at room temperature for 1 h. After washing with TBS-T, the protein bands were visualized by ECL reagent and band intensity of targets was quantified by Image-Pro Plus 6.0 software (Media Cybernetic). RNA extraction and qRT-PCR analysis Total RNA was extracted from cultured cells or mice brains by using Trizol reagent (Invitrogen). Single- strand cDNA was synthesized using a universal cDNA synthesis kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The expression of targets was tested with a fast real-time PCR system (7900 HT, ABI, Foster City, CA) by using a SYBR Green master mix (Qiagen). The relative expression change of targets was analyzed by the 2-ΔΔCt method with GAPDH and U6 as the internal references. The primers used as follows: miR-26a-5p: forward: 5′-GACGGTACCTTGTCCCTGAATGTAACTCG-3′ reverse: 5′- GTTCTCGAGAAAGCAGTCCCAGCCTAAA-3′; U6: forward: 5′-CTCGCTTCGGCAGCACA-3′, reverse: 5′- AACGCTTCACGAATTTGCGT-3′; GAPDH forward: 5′-CAAGGTCATCCATGACAACTTTG-3′, reverse: 5′- GTCCACCACCCTGTTGCTGTAG-3′. Statistical analysis All data were presented as mean ± SD, and each experiment was repeated three times. Statistical analysis was performed by using GraphPad Prism 6.0 software. Groups comparison was performed with two tailed Student’s t test (two groups) or one-way analysis of variance (ANOVA; multiple groups). P < 0.05 was considered to be significant. In situ detection of fragmented DNA (TUNEL assay) Brains tissues were collected and the apoptosis in vivo was evaluated TUNEL staining kit (YEASEN, Shanghai) as the manufacturer’s instructions. TUNEL positive brain cells were counted under a fluorescence microscope. Specially, cell nucleus dyed green were considered to be apoptotic cells, and the rate of apoptosis (%) was calculated as the percentage of TUNEL positive cell nucleus in 5 random fields for each sample. Page 6/18 Page 6/18 Characterization of MSCs‑derived exosomes To ensure the reliability of subsequent experiments, we firstly detected the characterization of MSCs‑derived exosomes. Under the transmission electronic microscope (TCM), MSCs-derived exosomes showed widespread, derangement distribution, and conglobation in some areas, and the vesicles exhibited typical exosome morphology with a dimension varying from 30 to 150 nm (Fig. 1A). Flow cytometry indicated that CD63 and CD81 positive rates were 30.84% and 43.75%, respectively (Fig. 1B), suggesting that MSCs-exosomes isolated were of certain purity. Meanwhile, the expression of exosomal markers was evaluated by western blot, and the results showed that the exosomal markers including CD9, CD63, CD81 and HSP70 were all expressed in MSCs and MSCs-derived exosomes, and their levels in MSCs-derived exosomes were higher than that in MSCs (Fig. 1C). These results suggested that the isolated exosomes could be better used for the subsequent experiments. MiR-26a-5p was downregulated and CDK6 was upregulated in MSCs-derived exosomes of MCAO and OGD model MiR-26a-5p was downregulated and CDK6 was upregulated in MSCs-derived exosomes of MCAO and Next, we explore the expression of miR-26a-5p and CDK6 in exosomes of cerebral I/R in vitro and in vivo. Compared with sham group, the expression of miR-26a-5p was significantly downregulated both in MSCs-derived exosomes (p < 0.01, Fig. 2A) and brains tissues (p < 0.01, Fig. 2B) of MCAO-induced mice. Similarly, the level of mIR-26a-5p in the exosomes of OGD-induced MSCs was also downregulated compared with control group (without OGD treatment) in a time dependent manner (p < 0.05, Fig. 2C). In addition, the protein level of CDK6 was increased in MSCs-derived exosomes of MCAO-induced mice compared with that of sham group (p < 0.01, Fig. 2D). Meanwhile, the expression of CDK6 was also upregulated in the exosomes of OGD-induced MSCs compared with control group (p < 0.05, Fig. 2E). These results revealed that miR-26a-5p was downregulated and CDK6 was upregulated in MSCs-derived exosomes of MCAO and OGD model. MSCs-Exo-miR-26a-5p mimic attenuated ischemia-reperfusion injury in MCAO/R model We then explore the neuroprotective effect of MSCs-Exo-miR-26a-5p on I/R injury in vivo through injection with 200 μL/mice MSCs-exosomes-miR-NC or MSCs-exosomes-miR-26a-5p mimics by the tail vein after 2 h for MCAO/R treatment. The expression of miR-26a-5p in brain tissues was firstly detected by qRT-PCR and the results showed that MCAO/R treatment significantly reduced miR-26a-5p level compared with sham operation (p < 0.001), and compared with MSCs-Exo-miR-NC group, MSCs-Exo-miR-26a-5p mimics obviously attenuated MCAO/R-induced decrease of miR-26a-5p (p < 0.01) (Fig. 4A). The results of TTC staining (Fig. 4B) showed that MCAO/R treatment significantly enhanced infarct volume compared with sham operation (p < 0.001), and compared with Exo-miR-NC group, Exo-miR-26a-5p mimics markedly reduced infarct percentage (p < 0.01). Meanwhile, cell apoptosis in brain tissues was determined by TUNEL staining and the results indicated that TUNEL positive cells were increased after MCAO/R treatment compared with sham operation (p < 0.001), and Exo-miR-26a-5p mimics obviously reduced TUNEL positive cells of rat brains after MCAO/R treatment compared with Exo-miR-NC group (p < 0.01) (Fig. 4C). In addition, the expression of apoptosis-related proteins in brain tissues was also evaluated by western blot and the results showed that MCAO/R-treatment significantly increased the expression of c- caspase 3 and c-PARP compared with sham operation, and compared with Exo-miR-NC group, Exo-miR- 26a-5p mimics obviously reduced the level of c-caspase 3 and c-PARP in rat brains induced by MCAO/R. These results suggested that MSCs-Exo-miR-26a-5p mimics attenuated ischemia-reperfusion injury in MCAO/R model in vivo. MiR-26a-5p mimics reversed the effects of MSCs-derived exosomes in reducing cell apoptosis of OGD- injured microglia To further explore the effect of exosomal miR-26a-5p on microglia function, BV-2 cells were treated OGD for the given time, and then treated with 200 μg/mL MSCs-Exo-miR‐26a-5p mimics or MSCs-Exo-miR-NC. CCK-8 assay indicated that OGD treatment significantly reduced cell viability compared with control Page 7/18 Page 7/18 group (p < 0.01), and OGD-treated BV-2 cells cultured with MSCs-Exo-miR‐26a-5p showed greater cell viability than MSC-Exo-miR-NC group (p < 0.05, Fig. 3A). Meanwhile, OGD treatment significantly exacerbated the apoptosis rate of BV-2 cells compared with control group (p < 0.001), and the elevated apoptosis of BV-2 cells stimulated by OGD were obviously reduced by MSCs-Exos-miR-26a-5p mimics compared with MSCs-Exo-miR-NC group (p < 0.01) (Fig. 3B). In addition, the expression of apoptosis- related proteins was evaluated by western blot, and the results showed that the expression of c-caspase and c-PARP was increased in BV-2 cells after OGD treatment compared with control group (p < 0.001), and the effect was significantly reversed by MSCs-Exo-miR-26a-5p mimics compared with MSCs-Exo- miR-NC group (p < 0.05) (Fig. 3C). These results indicated that overexpression of miR-26a-5p could reverse the effects of MSCs-derived exosomes in reducing cell apoptosis of OGD-injured microglia. The effect of miR-26a-5p was partially mediated by CDK6 To further determine the molecular mechanism of miR-26a-5p, we analyzed the sequence of the 3′-UTR of the human CDK6 gene. We found that there was a putative binding site between miR-26a-5p and 3′-UTR of CDK6 by using Targetscan software (http://www.targetscan.org) (Fig. 5A). Then miR-26a-5p mimics or miR-NC was transfected into 293T cells and qRT-PCR showed that miR-26a-5p mimics significantly increased the miR-26a-5p level compared with miR-NC (p < 0.01, Fig. 5B). Meanwhile, luciferase reporter assay indicated that miR-26a-5p mimics obviously reduced the relative luciferase activity of CDK6-WT compared with miR-NC (p < 0.01), while had no change on CDK6-MUT (Fig. 5C). Next, miR-26a-5p mimics Page 8/18 Page 8/18 or miR-NC was transfected into BV-2 cells and we found that miR-26a-5p mimics significantly decreased the expression of CDK6 in BV-2 cells at both mRNA level (p < 0.01) and protein level (p < 0.01) compared with miR-NC (Fig. 5D and E). Moreover, we found that OGD/R treatment significantly increased CDK6 level compared with control group (p < 0.01), and Exo-miR-26a-5p mimics obviously reduced CDK6 level in BV- 2 cells compared with Exo-miR-NC group (p < 0.05) (Fig. 5F). In addition, MCAO/R treatment significantly increased CDK6 level in mice brains compared with sham operation (p < 0.01), and Exo-miR-26a-5p mimics markedly reduced CDK6 level in mice brains induced by MCAO/R treatment compared with Exo- miR-NC group (p < 0.05) (Fig. 5G). These data suggested that MSCs-derived exosomes with miR-26a-5p overexpression attenuated I/R brain injury through targeting CDK6. Discussion In this study, we explore the role of exosomal miR-26a-5p in cerebral I/R injury, and found that miR-26a-5p was downregulated in MSCs-derived exosomes of MCAO and OGD model. Moreover, MSCs-Exo-miR-26a- 5p mimics effectively reduced cell apoptosis of BV-2 cells submitted to OGD/R treatment, and also reduced infarct volume of mice submitted to MCAO/R treatment by elevating the expression of CDK6. Our results provided that MSCs‑Exo overexpressing miR-26a-5p could effectively attenuate I/R brain injury through targeting CDK6. A series of miRNAs have been found to be loaded by exosomes of different types of cells, and have been showed to play crucial roles in neuron injury(25). For example, miRNA-181a overexpression in MSCs- derived exosomes suppress inflammatory response after myocardial I/R injury(26). Cortical neuron- derived exosomal miRNA-181c-3p inhibits neuro-inflammation by downregulating CXCL1 in astrocytes of a rat model with ischemic brain injury(27). Exosomes derived from miR-138-5p-overexpressing bone marrow-derived mesenchymal stem cells confer neuroprotection to astrocytes following ischemic stroke via inhibition of LCN2(25). Exosome-shuttled miR-92b-3p from ischemic preconditioned astrocytes protects neurons against oxygen and glucose deprivation(28). Exosomal miR-26a-5p level has been identified to be decreased in hUCMSCs, and exosomal miR-26b-5p from hUCMSCs could repress M1 polarization of microglia by targeting CH25H to inactivate the TLR pathway, then finally relieve nerve injury after cerebral I/R(29). In this study, we found that miR-26a-5p was downregulated in MSCs-derived exosomes of MCAO and OGD model. To explore the effect of exosomal miR-26a-5p in cerebral I/R injury, miR-26a-5p overexpressing exosomes were isolated from MSCs and treated with BV-2 cells or injected with mice submitted MCAO treatment. Moreover, MSCs-Exo-miR-26a-5p mimics effectively reduced cell apoptosis of BV-2 cells submitted to OGD/R treatment, and also reduced infarct volume of mice submitted to MCAO/R treatment. These results indicated the protective effect of exosomal miR-25a-5p in cerebral I/R injury, which might be regarded as a novel therapeutic target for cerebral ischemic stroke. Here, we found that CDK6 was upregulated in MSCs-derived exosomes of MCAO and OGD model, suggesting the essential role of CDK6 in cerebral I/R injury. Previous studies have reported that CDK6 could serve as a target of miRNAs to participate in neuron injury. MiR-99a overexpression inhibits H2O2 Page 9/18 Page 9/18 induced G1/S phase transition along with a significant low level of CDK6 in neuro-2a cells(30). MiR-424 has been identified to protects from permanent focal cerebral ischemia injury in mice through targeting CDK6 to inhibit microglia activation (31). Discussion However, the regulatory network of CDK6 involved in exosomal miRNAs has not been well studied. cerebral I/R injury, bioinformatics prediction was performed and suggested that CDK6 might a direct target of exosomal miR-26a-5p. Then luciferase reporter assay further confirmed the correlation between exosome miR-26a-5- and CDK6. In addition, MSCs-Exo-miR-26a- 5p mimics obviously reduced CDK6 level in BV-2 cells after OGD/R treatment, and also reduced CDK6 level in mice brains induced by MCAO/R treatment. All these data suggested that the protective effect of MSCs-Exo-miR-26a-5p mimics on I/R injury both in vitro and in vivo might be mediated by CDK6, providing a new highlight of exosomal miRNAs and cell proliferation-related proteins involved in cerebral I/R injury. Our study suggests that EMSCs-derived exosomes overexpressing miR-26a-5p might be applied for the personalized treatment against cerebral ischemic stroke. However, there was a limitation existing in the study which was determined in the subsequent experiments. That was, whether high level of CDK6 reversed the protective effect of MSCs-Exo-miR-26a- 5p mimics on cerebral I/R injury. Ethics approval and consent to participate Ethics approval and consent to participate Not applicable. Conclusion In summary, our results demonstrated that MSCs‑derived exosomes could effectively attenuate I/R injury in vivo and inhibit microglia apoptosis in vitro by exosomal miR-26a-5p mediated downregulation of CDK6, suggesting that miR-26a-5p might be a novel therapeutic target for cerebral I/R injury. Consent for publication All authors give the consent for publication Acknowledgements Not applicable. Funding This work was supported by the National Natural Science Foundation of China (Grant No. 81571129). This work was supported by the National Natural Science Foundation of China (Grant No. 81571129). Availability of data and materials Availability of data and materials Availability of data and materials Page 10/18 Any additional information related to this study is available from the author for correspondence upon reasonable request. Authors’ contributions WH conceived and designed the experiments. CC, XC, YW, WC, XZ and WT performed the experiments. CC analyzed the data and wrote the manuscript. All authors read and approved the final manuscrip Competing interests The authors declare that they have no competing interests The authors declare that they have no competing interests References de Freitas RCC, et al. (2017) Modulation of miR-26a-5p and miR-15b-5p Exosomal Expression Associated with Clopidogrel-Induced Hepatotoxicity in HepG2 Cells. Frontiers in pharmacology 8: 906. 16. Xing X, Guo S. (2020) miR-26a-5p protects against myocardial ischemia/reperfusion injury by regulating the PTEN/PI3K/AKT signaling pathway 53: e9106. 16. Xing X, Guo S. (2020) miR-26a-5p protects against myocardial ischemia/reperfusion injury by regulating the PTEN/PI3K/AKT signaling pathway 53: e9106. 17. Tigan AS, Bellutti F, Kollmann K, Tebb G, Sexl V. (2016) CDK6-a review of the past and a glimpse into the future: from cell-cycle control to transcriptional regulation. Oncogene 35: 3083-3091. 17. Tigan AS, Bellutti F, Kollmann K, Tebb G, Sexl V. (2016) CDK6-a review of the past and a glimpse into the future: from cell-cycle control to transcriptional regulation. Oncogene 35: 3083-3091. 18. Li G, et al. (2018) Diagnostic and Immunosuppressive Potential of Elevated Mir-424 Levels in Circulating Immune Cells of Ischemic Stroke Patients. Aging and disease 9: 172-181. 18. Li G, et al. (2018) Diagnostic and Immunosuppressive Potential of Elevated Mir-424 Levels in Circulating Immune Cells of Ischemic Stroke Patients. Aging and disease 9: 172-181. 19. Demyanenko S, Uzdensky A. (2017) Profiling of Signaling Proteins in Penumbra After Focal Photothrombotic Infarct in the Rat Brain Cortex. Molecular neurobiology 54: 6839-6856. 19. Demyanenko S, Uzdensky A. (2017) Profiling of Signaling Proteins in Penumbra After Focal Photothrombotic Infarct in the Rat Brain Cortex. Molecular neurobiology 54: 6839-6856. 20. Huang ZM, et al. (2019) MicroRNA-26a-5p inhibits breast cancer cell growth by suppressing RNF6 expression. The Kaohsiung journal of medical sciences 35: 467-473. 20. Huang ZM, et al. (2019) MicroRNA-26a-5p inhibits breast cancer cell growth by suppressing RNF6 expression. The Kaohsiung journal of medical sciences 35: 467-473. 21. Canturk KM, et al. (2014) Investigation of key miRNAs and target genes in bladder cancer using miRNA profiling and bioinformatic tools. Molecular biology reports 41: 8127-8135. 21. Canturk KM, et al. (2014) Investigation of key miRNAs and target genes in bladder cancer using miRNA profiling and bioinformatic tools. Molecular biology reports 41: 8127-8135. 22. Maria AT, et al. (2016) Human adipose mesenchymal stem cells as potent anti-fibrosis therapy for systemic sclerosis. Journal of autoimmunity 70: 31-39. 22. Maria AT, et al. (2016) Human adipose mesenchymal stem cells as potent anti-fibrosis therapy for systemic sclerosis. Journal of autoimmunity 70: 31-39. 23. Théry C, Amigorena S, Raposo G, Clayton A. References 1. Barthels D, Das H. (2020) Current advances in ischemic stroke research and therapies. Biochimica e biophysica acta. Molecular basis of disease 1866: 165260. 2. Powers WJ. (2020) Acute Ischemic Stroke. N Engl J Med 383: 252-260. 2. Powers WJ. (2020) Acute Ischemic Stroke. N Engl J Med 383: 252-260. 3. Phipps MS, Cronin CA. (2020) Management of acute ischemic stroke. BMJ (Clinical research ed.) 368: l6983. 3. Phipps MS, Cronin CA. (2020) Management of acute ischemic stroke. BMJ (Clinical research ed.) 368: l6983. 4. Sharma A. (2018) Role of stem cell derived exosomes in tumor biology. International journal of cancer 142: 1086-1092. 5. Dorayappan KDP, Wallbillich JJ, Cohn DE, Selvendiran K. (2016) The biological significance and clinical applications of exosomes in ovarian cancer. Gynecologic oncology 142: 199-205. 6. Kalani A, Tyagi A, Tyagi N. (2014) Exosomes: mediators of neurodegeneration, neuroprotection and therapeutics. Molecular neurobiology 49: 590-600. 7. Xin H, et al. (2012) Exosome-mediated transfer of miR-133b from multipotent mesenchymal strom cells to neural cells contributes to neurite outgrowth. Stem cells (Dayton, Ohio) 30: 1556-1564. 8. Yoo J, Kim HS, Hwang DY. (2013) Stem cells as promising therapeutic options for neurological disorders. Journal of cellular biochemistry 114: 743-753. 9. Luarte A, Bátiz LF, Wyneken U, Lafourcade C. (2016) Potential Therapies by Stem Cell-Derived Exosomes in CNS Diseases: Focusing on the Neurogenic Niche. Stem cells international 2016: 5736059. 10. Marote A, Teixeira FG, Mendes-Pinheiro B, Salgado AJ. (2016) MSCs-Derived Exosomes: Cell- Secreted Nanovesicles with Regenerative Potential. Frontiers in pharmacology 7: 231. 11. Zhang J, et al. (2015) Exosome and exosomal microRNA: trafficking, sorting, and function. Genomics, proteomics & bioinformatics 13: 17-24. 12. Chen Y, et al. (2017) Increased Circulating Exosomal miRNA-223 Is Associated with Acute Ischemic Stroke. Frontiers in neurology 8: 57. Page 11/18 Page 11/18 13. Song Y, et al. (2019) M2 microglia-derived exosomes protect the mouse brain from ischemia- reperfusion injury via exosomal miR-124. Theranostics 9: 2910-2923. 14. Jin Z, Ren J, Qi S. (2020) Human bone mesenchymal stem cells-derived exosomes overexpressing microRNA-26a-5p alleviate osteoarthritis via down-regulation of PTGS2. International immunopharmacology 78: 105946. 14. Jin Z, Ren J, Qi S. (2020) Human bone mesenchymal stem cells-derived exosomes overexpressing microRNA-26a-5p alleviate osteoarthritis via down-regulation of PTGS2. International immunopharmacology 78: 105946. 15. de Freitas RCC, et al. (2017) Modulation of miR-26a-5p and miR-15b-5p Exosomal Expression Associated with Clopidogrel-Induced Hepatotoxicity in HepG2 Cells. Frontiers in pharmacology 8: 906. 15. References (2013) MiRNA-424 protects against permanent focal cerebral ischemia injury in mice involving suppressing microglia activation. Stroke 44: 1706-1713. References (2006) Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Current protocols in cell biology Chapter 3: Unit 3.22. 23. Théry C, Amigorena S, Raposo G, Clayton A. (2006) Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Current protocols in cell biology Chapter 3: Unit 3.22. 24. Yang G, et al. (1994) Human copper-zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia. Stroke 25: 165-170. 24. Yang G, et al. (1994) Human copper-zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia. Stroke 25: 165-170. 25. Deng Y, et al. (2019) Exosomes derived from microRNA-138-5p-overexpressing bone marrow-derived mesenchymal stem cells confer neuroprotection to astrocytes following ischemic stroke via inhibition of LCN2. Journal of biological engineering 13: 71. 25. Deng Y, et al. (2019) Exosomes derived from microRNA-138-5p-overexpressing bone marrow-derived mesenchymal stem cells confer neuroprotection to astrocytes following ischemic stroke via inhibition of LCN2. Journal of biological engineering 13: 71. 26. Wei Z, et al. (2019) miRNA-181a over-expression in mesenchymal stem cell-derived exosomes influenced inflammatory response after myocardial ischemia-reperfusion injury. Life sciences 232: 116632. 26. Wei Z, et al. (2019) miRNA-181a over-expression in mesenchymal stem cell-derived exosomes influenced inflammatory response after myocardial ischemia-reperfusion injury. Life sciences 232: 116632. 27. Song H, et al. (2019) Cortical Neuron-Derived Exosomal MicroRNA-181c-3p Inhibits Neuroinflammation by Downregulating CXCL1 in Astrocytes of a Rat Model with Ischemic Brain Injury. Neuroimmunomodulation 26: 217-233. 27. Song H, et al. (2019) Cortical Neuron-Derived Exosomal MicroRNA-181c-3p Inhibits Neuroinflammation by Downregulating CXCL1 in Astrocytes of a Rat Model with Ischemic Brain Injury. Neuroimmunomodulation 26: 217-233. 28. Xu L, et al. (2019) Exosome-shuttled miR-92b-3p from ischemic preconditioned astrocytes protects neurons against oxygen and glucose deprivation. Brain research 1717: 66-73. 28. Xu L, et al. (2019) Exosome-shuttled miR-92b-3p from ischemic preconditioned astrocytes protects neurons against oxygen and glucose deprivation. Brain research 1717: 66-73. Page 12/18 Page 12/18 29. Li G, et al. (2020) Exosomes-carried microRNA-26b-5p regulates microglia M1 polarization after cerebral ischemia/reperfusion. Cell cycle (Georgetown, Tex.) 19: 1022-1035. 30. Tao Z, et al. (2015) Neuroprotective effect of microRNA-99a against focal cerebral ischemia- reperfusion injury in mice. Journal of the neurological sciences 355: 113-119. 30. Tao Z, et al. (2015) Neuroprotective effect of microRNA-99a against focal cerebral ischemia- reperfusion injury in mice. Journal of the neurological sciences 355: 113-119. 31. Zhao H, et al. 29. Li G, et al. (2020) Exosomes-carried microRNA-26b-5p regulates microglia M1 polarization after cerebral ischemia/reperfusion. Cell cycle (Georgetown, Tex.) 19: 1022-1035. Figures Figures Figure 1 Characterization of MSCs‑derived exosomes. (A) Morphological change of MSCs-derived exosomes by TEM. (B) FACS analysis of exosome surface markers CD63 and CD81 in MSCs-derived exosomes. (C) The protein expression of exosome surface markers (CD9, CD63, CD81, and HSP70) in MSCs and MSCs derived exosomes was evaluated by western blot. N = 3. Figure 1 Characterization of MSCs‑derived exosomes. (A) Morphological change of MSCs-derived exosomes by TEM. (B) FACS analysis of exosome surface markers CD63 and CD81 in MSCs-derived exosomes. (C) The protein expression of exosome surface markers (CD9, CD63, CD81, and HSP70) in MSCs and MSC derived exosomes was evaluated by western blot. N = 3. Figure 1 Characterization of MSCs‑derived exosomes. (A) Morphological change of MSCs-derived exosomes by TEM. (B) FACS analysis of exosome surface markers CD63 and CD81 in MSCs-derived exosomes. (C) The protein expression of exosome surface markers (CD9, CD63, CD81, and HSP70) in MSCs and MSCs- derived exosomes was evaluated by western blot. N = 3. Page 13/18 Page 13/18 Figure 2 MiR-26a-5p was downregulated and CDK6 was upregulated in MSCs-derived exosomes of MCAO and OGD model. (A and B) The mRNA expression of miR-26a-5p in MSCs-derived exosomes (A) and brain tissues (B) of MCAO-induced mice was evaluated by qRT-PCR (n = 6). (C) The level of miR-26a-5p in exosomes of OGD-treated MSCs at indicated time was evaluated by qRT-PCR (n = 3). (D) The protein expression of CDK6 in MSCs-derived exosomes of MCAO-induced mice was evaluated by western blot (n = 6). (E) The protein expression of CDK6 in exosomes of OGD-treated MSCs at indicated time was evaluated by western blot (n = 3). * p < 0.05, ** p < 0.01. Figure 2 MiR-26a-5p was downregulated and CDK6 was upregulated in MSCs-derived exosomes of MCAO and OGD model. (A and B) The mRNA expression of miR-26a-5p in MSCs-derived exosomes (A) and brain tissues (B) of MCAO-induced mice was evaluated by qRT-PCR (n = 6). (C) The level of miR-26a-5p in exosomes of OGD-treated MSCs at indicated time was evaluated by qRT-PCR (n = 3). (D) The protein expression of CDK6 in MSCs-derived exosomes of MCAO-induced mice was evaluated by western blot (n = 6). (E) The protein expression of CDK6 in exosomes of OGD-treated MSCs at indicated time was evaluated by western blot (n = 3). * p < 0.05, ** p < 0.01. Page 14/18 Page 14/18 Figure 3 MiR-26a-5p reversed the effects of MSCs-derived exosomes in reducing cell apoptosis of OGD-injured microglia. BV-2 cells were treated OGD for the given time, and then treated with 200 μg/mL Exo-miR‐26a- 5p mimics or Exo-miR-NC. (A) Cell viability was evaluated by CCK-8 assay. (B) Cell apoptosis was detected by flow cytometry. (C) The protein expression of apoptosis-related makers was evaluated by western blot. N = 3. * p < 0.05, p < 0.01, * p < 0.001. Figure 3 MiR-26a-5p reversed the effects of MSCs-derived exosomes in reducing cell apoptosis of OGD-injured microglia. BV-2 cells were treated OGD for the given time, and then treated with 200 μg/mL Exo-miR‐26a- 5p mimics or Exo-miR-NC. (A) Cell viability was evaluated by CCK-8 assay. (B) Cell apoptosis was detected by flow cytometry. (C) The protein expression of apoptosis-related makers was evaluated by western blot. N = 3. * p < 0.05, p < 0.01, * p < 0.001. Page 15/18 Figure 4 MSCs-Exo-miR-26a-5p mimic attenuated ischemia-reperfusion injury in MCAO/R model. M treated with MCAO/R, and injected with 200 μL/mice MSCs-exosomes-miR-NC or MSCs-e 26a-5p mimics. (A) The mRNA level of miR-26a-5p in brain tissues was evaluated qRT-PC Representative brain sections of I/R induced mice were detected by TTC staining. (C) TUN Figure 4 MSCs-Exo-miR-26a-5p mimic attenuated ischemia-reperfusion injury in MCAO/R model. Mice were treated with MCAO/R, and injected with 200 μL/mice MSCs-exosomes-miR-NC or MSCs-exosomes-miR- 26a-5p mimics. (A) The mRNA level of miR-26a-5p in brain tissues was evaluated qRT-PCR. (B) Representative brain sections of I/R induced mice were detected by TTC staining. (C) TUNEL staining in Figure 4 MSCs-Exo-miR-26a-5p mimic attenuated ischemia-reperfusion injury in MCAO/R model. Mice were treated with MCAO/R, and injected with 200 μL/mice MSCs-exosomes-miR-NC or MSCs-exosomes-miR- 26a-5p mimics. (A) The mRNA level of miR-26a-5p in brain tissues was evaluated qRT-PCR. (B) Representative brain sections of I/R induced mice were detected by TTC staining. (C) TUNEL staining in Page 16/18 Page 16/18 brain tissues. (D) The protein level of apoptosis-related markers in brain tissues was detected by western blot. N = 6, * p < 0.05, p < 0.01, * p < 0.001. blot. N = 6, * p < 0.05, p < 0.01, * p < 0.001. Figure 5 Page 17/18 Figure 5 The effect of miR-26a-5p was partially mediated by CDK6. (A) The putative binding site between miR-26a- 5p and CDK6 was predicted by Targetscan. (B) 293T cells were transfected with miR-26a-5p mimics or miR-NC, and the mRNA level of miR-26a-5p was detected by qRT-PCR. (C) 293T cells were co-transfected Figure 5 Page 17/18 The effect of miR-26a-5p was partially mediated by CDK6. (A) The putative binding site between miR-26a- 5p and CDK6 was predicted by Targetscan. (B) 293T cells were transfected with miR-26a-5p mimics or miR-NC, and the mRNA level of miR-26a-5p was detected by qRT-PCR. Figure 3 (C) 293T cells were co-transfected Page 17/18 Page 17/18 with miR-26a-5p mimics or miR-NC and Luc-CDK6-WT or Luc-CDK6-MUT, and the relative luciferase reporter activity was evaluated by dual luciferase reporter system. (D and E) BV-2 cells were transfected with miR-26a-5p or miR-NC, and the expression of CDK6 was detected by qRT-PCR (D) and western blot (E). (F) BV-2 cells were treated by OGD/R, and then treated with 200 μg/mL Exo-miR-26a-5p mimics or Exo-miR-NC, and the protein level of CDK6 was evaluated by western blot. (G) Mice were induced by MCAO/R, and then injected with 200 μL/mice Exo-miR-NC or Exo-miR-26a-5p mimics. The protein level of CDK6 in brain tissues was detected by western blot. N = 6. * p < 0.05 ** p < 0.01. with miR-26a-5p mimics or miR-NC and Luc-CDK6-WT or Luc-CDK6-MUT, and the relative luciferase reporter activity was evaluated by dual luciferase reporter system. (D and E) BV-2 cells were transfected with miR-26a-5p or miR-NC, and the expression of CDK6 was detected by qRT-PCR (D) and western blot (E). (F) BV-2 cells were treated by OGD/R, and then treated with 200 μg/mL Exo-miR-26a-5p mimics or Exo-miR-NC, and the protein level of CDK6 was evaluated by western blot. (G) Mice were induced by MCAO/R, and then injected with 200 μL/mice Exo-miR-NC or Exo-miR-26a-5p mimics. The protein level of CDK6 in brain tissues was detected by western blot. N = 6. * p < 0.05 ** p < 0.01. with miR-26a-5p mimics or miR-NC and Luc-CDK6-WT or Luc-CDK6-MUT, and the relative luciferase reporter activity was evaluated by dual luciferase reporter system. (D and E) BV-2 cells were transfected with miR-26a-5p or miR-NC, and the expression of CDK6 was detected by qRT-PCR (D) and western blot (E). (F) BV-2 cells were treated by OGD/R, and then treated with 200 μg/mL Exo-miR-26a-5p mimics or Exo-miR-NC, and the protein level of CDK6 was evaluated by western blot. (G) Mice were induced by MCAO/R, and then injected with 200 μL/mice Exo-miR-NC or Exo-miR-26a-5p mimics. The protein level of CDK6 in brain tissues was detected by western blot. N = 6. * p < 0.05 ** p < 0.01. Page 18/18 Page 18/18
https://openalex.org/W3210091192	https://europepmc.org/articles/pmc8568287?pdf=render	English	null	Sex-dependent development of Kras-induced anal squamous cell carcinoma in mice	PloS one	2,021	cc-by	12,561	PLOS ONE RESEARCH ARTICLE Editor: Murray Korc, Indiana University School of Medicine, UNITED STATES Editor: Murray Korc, Indiana University School of Medicine, UNITED STATES Copyright: © 2021 Walcheck et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: SRK Received all funding grant number IRG-15-213-51 This work was supported by American Cancer Society https://www.cancer.org/ No (The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript). Morgan T. WalcheckID1, Kristina A. Matkowskyj2,3,4, Anne Turco5, Simon Blaine-Sauer1, Manabu Nukaya5, Jessica Noel1, Oline K. Ronnekleiv6, Sean M. Ronnekleiv-Kelly1,2,5* 1 McArdle Laboratory for Cancer Research, University of Wisconsin, Madison, WI, United States of America, 2 University of Wisconsin Carbone Cancer Center, Madison, WI, United States of America, 3 Department of Pathology and Laboratory Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States of America, 4 William S. Middleton Memorial Veterans Hospital, Madison, WI, United States of America, 5 Division of Surgical Oncology, Department of Surgery, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States of America, 6 Department of Chemical Physiology and Biochemistry, Oregon Health & Sciences University, Portland, OR, United States of America a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * ronnekleiv-kelly@surgery.wisc.edu Abstract Anal squamous cell carcinoma (SCC) will be diagnosed in an estimated 9,080 adults in the United States this year, and rates have been rising over the last several decades. Most peo- ple that develop anal SCC have associated human papillomavirus (HPV) infection (~85– 95%), with approximately 5–15% of anal SCC cases occurring in HPV-negative patients from unknown etiology. This study identified and characterized the Kras-driven, female sex hormone-dependent development of anal squamous cell carcinoma (SCC) in the LSL- KrasG12D; Pdx1-Cre (KC) mouse model that is not dependent on papillomavirus infection. One hundred percent of female KC mice develop anal SCC, while no male KC mice develop tumors. Both male and female KC anal tissue express Pdx1 and Cre-recombinase mRNA, and the activated mutant KrasG12D gene. Although the driver gene mutation KrasG12D is present in anus of both sexes, only female KC mice develop Kras-mutant induced anal SCC. To understand the sex-dependent differences, KC male mice were castrated and KC female mice were ovariectomized. Castrated KC males displayed an unchanged phenotype with no anal tumor formation. In contrast, ovariectomized KC females demonstrated a marked reduction in anal SCC development, with only 15% developing anal SCC. Finally, exogenous administration of estrogen rescued the tumor development in ovariectomized KC female mice and induced tumor development in castrated KC males. These results con- firm that the anal SCC is estrogen mediated. The delineation of the role of female sex hor- mones in mediating mutant Kras to drive anal SCC pathogenesis highlights a subtype of anal SCC that is independent of papillomavirus infection. These findings may have clinical applicability for the papillomavirus-negative subset of anal SCC patients that typically respond poorly to standard of care chemoradiation. * ronnekleiv-kelly@surgery.wisc.edu OPEN ACCESS Citation: Walcheck MT, Matkowskyj KA, Turco A, Blaine-Sauer S, Nukaya M, Noel J, et al. (2021) Sex-dependent development of Kras-induced anal squamous cell carcinoma in mice. PLoS ONE 16(11): e0259245. https://doi.org/10.1371/journal. pone.0259245 Sex-dependent development of Kras-induced anal squamous cell carcinoma in mice Morgan T. WalcheckID1, Kristina A. Matkowskyj2,3,4, Anne Turco5, Simon Blaine-Sauer1, Manabu Nukaya5, Jessica Noel1, Oline K. Ronnekleiv6, Sean M. Ronnekleiv-Kelly1,2,5* PLOS ONE PLOS ONE Introduction In 2021, an estimated 9,090 adults will be diagnosed with anal squamous cell carcinoma (SCC) n the United States, and anal SCC has been increasing in incidence over the last several Competing interests: The authors have declared that no competing interests exist. 1 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC decades [1, 2]. Anal SCC is typically associated with human papillomavirus (HPV) infection (~85–95%) [2, 3], yet approximately 5–15% of anal SCC cases occur in HPV-negative patients with unknown etiology [4, 5]. Unfortunately, patients with HPV-negative anal SCC are signifi- cantly less responsive to standard of care chemoradiation [5], and have a worse prognosis than HPV-positive anal SCC [6]. This study presents a novel etiology for HPV-negative anal SCC development driven by mutant Kras. In human anal cancer, mutant Kras is identified in 10% of HPV-negative anal SCC samples [7]. Despite this association, to our knowledge, the present study is the first to identify this cor- relation in a pre-clinical model. The mutant Kras-driven development of anal SCC was detected in a genetically engineered mouse model (GEMM) traditionally used in the investiga- tion of pancreatic ductal adenocarcinoma (PDAC). This mouse harbors a Kras-mutation (Kras G12D) in cells expressing Cre-recombinase from pancreatic and duodenal homeobox 1 (Pdx1) promoter (KC mice: Lox-stop-lox KrasG12D/+; Pdx1-Cre) [8]. In this study, we found that Pdx1 expression and consequent Cre-recombinase expression in the anal epithelium caused activation of the KrasG12D gene in the anal epithelium and tumor development. Fur- ther, we observed that only female mice developed anal SCC suggesting a sex-hormone depen- dent interaction with Kras G12D that triggers tumor formation. Therefore, we sought to understand the sex-dependent development of anal SCC in KC mice. Activated KrasG12D was present in the anal tissue of both sexes of KC mice, suggesting both sexes have the potential to develop Kras-mutant anal SCC. To ascertain why only female KC mice develop tumors, we ovariectomized females and castrated males to eliminate endoge- nous sex hormones production in the mice and found ovariectomized females displayed sig- nificantly reduced anal tumor development, signifying female sex hormone dependence. In turn, ovariectomized and castrated KC mice dosed with estrogen resulted in tumor develop- ment in both KC female and KC male mice, respectively, indicating the anal tumor develop- ment is estrogen mediated. Introduction This novel phenotype shows a female sex hormone dependent pathogenesis of Kras-mutant anal SCC that is independent of HPV infection. Given that 2–5% of anal SCC overall and 10% of HPV negative anal SCC harbor Kras-mutations [7], these find- ings may have therapeutic implications for this subset of patients. Lastly, the sex-based differ- ence highlights the importance of characterizing both sexes in pre-clinical studies. Animals All animal studies were conducted according to an approved protocol (M005959) by the Univer- sity of Wisconsin School of Medicine and Public Health (UW SMPH) Institutional Animal Care and Use Committee (IACUC). Mice were housed in an Assessment and Accreditation of Labo- ratory Animal Care (AALAC) accredited selective pathogen-free facility (UW Medical Sciences Center) on corncob bedding with chow diet (Mouse diet 9F 5020; PMI Nutrition International), and water ad libitum. The Lox-Stop-Lox (LSL) KrasG12D (B6.129S4-Kras tm4Tyj/J #008179), Pdx1-Cre (B6.FVB-Tg(Pdx1-cre)6Tuv/J) and Ai14 (B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J #007914) mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and housed under identical conditions. All mice listed are congenic on a C57BL/6J background (backcrossing > 15 generations). The LSLKrasG12D and Pdx1-Cre mice were bred to develop LSL-KrasG12D; Pdx1-Cre (KC) mice. The Ai14, Pdx1-Cre and LSLKrasG12D were bred to develop Rosa26LSL-tdTomato; LSLKrasG12D; Pdx1-Cre (AiKC) mice. Genotyping was performed according to Jackson Laboratory’s protocols (Cre: Protocol #21298, Kras: Protocol #29388 and Ai14: Pro- tocol #29436). Original observations were performed in 16 male KC and 14 female KC mice and 83 control genotypes. Based on stark sex-dependence, we calculated that 10 KC male, 10 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 2 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC KC female, and 12 control mice were needed for castration / ovariectomy studies to detect a 50% change in tumor formation by Fisher Exact test and alpha of 0.05. We then calculated that 12 KC female mice would be sufficient for the E2 dosing studies (6 E2 dosed and 6 sham con- trols) as well as 12 KC male mice (6 E2 dosed and 6 sham controls). Finally, we used 6 AiKC mice to visually confirm the location of Pdx1-Cre (projected location of mutant Kras expres- sion) in the anal tissue. The health and well-being of the mice were monitored closely by research and veterinary staff. Mice that showed signs of distress such as disheveled coat, hunched posture, rapid weight loss, lack of feeding or lack of defecation were immediately euthanized. During the experiment process, one castrated KC male mouse was euthanized due to decline in health and one castrated KC male mouse, one E2 dosed ovariectomized KC female, one sham dosed castrated KC male and two E2 dosed castrated KC male mice were found deceased of uncertain circumstances before the study end point. These mice were not included in the results. Histology KC mouse tissues (anus and pancreas) were fixed in 10% buffered formalin for 48 hours. Serial 4 μm sections from paraffin-embedded tissues were mounted on charged slides. Hematoxylin and eosin (H&E) was performed by the Experimental Animal Pathology Lab (EAPL) at the Uni- versity of Wisconsin-Madison. The histology was evaluated by a certified pathologist (KAM). mutation construct Activated KrasG12D refers to the successful Cre-mediated excision of the Lox-Stop sequence, allowing for transcription of the mutant Kras allele. To determine the tissue specific activation of the KrasG12D mutation, we followed the standard method first published by Hingorani [8, 9] and further utilized by other groups working with this Lox–Stop–Lox conditional Kras mouse strain [10–12]. Genomic DNA was isolated from tail, pancreas, anus and anal tumor from KC mice. The DNA was then amplified using polymerase chain reaction (PCR) with the following probes: 5’-GGGTAGGTGTTGGGATAGCTG-3’ (OL8403) and 5’-CCGAATTCAGTGACTAC AGATGTACAGAG-3’ (OL8404) with conditions previously published [11]. These primers amplified a 325 bp band corresponding to the activated KrasG12D mutant allele and a 285 bp band corresponding to the WT allele. Tumor studies The study endpoint (age 9 months) was selected based on existing data evaluating and report- ing on male KC mice at this age [8]. At 9 months, mice were euthanized and underwent cervi- cal dislocation followed by midline laparotomy for solid organ assessment. The anus was also removed for pathologic analysis. A board-certified surgical pathologist with subspecialty train- ing in gastrointestinal pathology (KAM) who was blinded to the mouse genotype and sex eval- uated the histologic sections. Animals Mice were euthanized through CO2 asphyxiation. DNA recovery from H&E stained formalin fixed paraffin embedded (FFPE) samples The anal tissue was scraped from H&E stained slides using a sterile blade [13]. The deparaffini- zation and genome DNA extraction from H&E stained anal tissues was performed according to manufacturer’s instructions using ReliaPrep FFPE gDNA MiniPrep System (Promega, Madison, WI). 3 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Immunohistochemistry (IHC) for tdTomato IHC staining for red fluorescence protein (tdTomato) was performed by the Experimental Ani- mal Pathology Lab (EAPL) at the University of Wisconsin-Madison. For IHC staining, sections were deparaffinized in xylenes and hydrated through graded alcohols to distilled water. Antigens were retrieved using citrate buffer pH 6.0 (10 mM Citric Acid, 0.05% tween 20). Endogenous per- oxidase was blocked with 0.3% H2O2 in PBS for 10 minutes at room temperature and blocking of non-specific binding was performed using 10% goat serum. Sections were incubated with rabbit anti-RFP antibody (600-401-379, Rockland Inc, Pottstown, PA) (1:1600) overnight at 4˚C. After rinsing, sections were incubated with ImmPRESS goat anti-rabbit IgG (MP-7451, Vector Labora- tories, Burlingame, CA) for 30 minutes at room temperature. Detection was performed using DAB substrate kit (8059S, Cell Signaling Technology, Danvers, MA). Samples were counter- stained using Mayer’s hematoxylin (MHS32, Millipore-Sigma, St. Louis, MO) for one minute. RNAScope in situ hybridization MmuPV1 detection was performed using the RNAscope 2.5 HD Assay-Brown kit (Advanced Cell Diagnostics, Newark, CA; 322300) and probe to MmuPV1 E4 (473281) as previously described [14]. NSG mouse anal tissues that were infected with MmuPV1 or mock infected [14] were included as positive and negative controls, respectively. Estrogen receptor alpha immunofluorescence Formalin-fixed (10% formalin), paraffin-imbedded mouse tissue sections mounted on Super- frost Plus glass slides (Fisher Scientific, Pittsburgh, PA), were deparaffinized with Xylene (3 x 5 min), and rehydrated in descending concentrations of ethanol as follows: 2 x 10 min each in 100%, 95%, 70%, and 50% ethanol followed by two washes in deionized water for 5 min each and a final wash in phosphate buffer (PB; 0.1 M phosphate buffer, pH 7.4) solution for 10–15 min. Sections were pretreated with normal donkey serum solution (3% donkey serum, 0.3% Triton-X 100 in PB, pH 7.4) for 30 min at room temperature and then washed briefly in PB before being incubated for 48 hrs at 4˚ C with an estrogen receptor α (ERα) rabbit antibody (C1355) diluted 1:5000. Thereafter the sections were rinsed in PB and next incubated with bio- tinylated donkey anti-rabbit IgG (1:500) for 2 hours at room temperature, another wash for 30 min in PB and then reacted with streptavidin Alexa Fluor 594 (1:2500) for 3 hrs. Both primary and secondary antisera were diluted in Tris-(hydroxymethyl)aminomethane (0.5%; Sigma- Aldrich) in phosphate buffer containing 0.7% seaweed gelatin (Sigma-Aldrich), 0.5% Triton X-100 (Sigma-Aldrich), and 3% BSA (Sigma-Aldrich), adjusted to pH 7.6. Adjacent sections were treated equally, but without the ERα antibody for control purposes. After a final rinse overnight in PB, the sections were cover-slipped with gelvatol containing the anti-fading agent 1,4-diazabicyclo(2,2)octane (DABCO; Sigma-Aldrich; 50 mg/ml). Sections were screened and photographed using a Nikon E800 fluorescent microscope (Eclipse E800; Nikon Instruments, Melville, NY) equipped with a fiber illuminator (Intensilight C-HGFI; Nikon Instruments) and a high-definition digital microscope camera head (DS-Fi1; Nikon Instruments) interfaced with a PC-based camera controller (DS-U3; Nikon Instruments). It should be noted that the C1355 ERα antibody has been documented to be specific for ERα in rat and mouse tissues and does not recognize ERβ [15]. Castration and ovariectomy To evaluate the sex differences of anal SCC development in KC mice, male WT and male KC mice were castrated; meanwhile, female WT and female KC mice were ovariectomized at 6–7 weeks of age. Mice were anesthetized with isoflurane inhalation throughout the surgery. Slow- release buprenorphine was used as an analgesic for mice undergoing this surgery. The hair from the surgical area was removed with clippers, and the surgical area was sterilized with an iodine scrub. Under sterile conditions and using sterilized tools, the testis and ovaries were removed from male and female mice, respectively [18, 19]. To remove the testis, gentle pres- sure was applied to the abdomen to push the testis into the scrotal sac [18]. A short 10mm mid- line incision was made through the skin in the middle of the abdomen [18]. The testis were located, gently pulled out through that incision along with the epididymal fat pad and carefully removed via cauterization [18]. To remove the ovaries, two incisions were made: short dorsal midline incisions parallel to and on either side of the spine [19]. The ovaries were located and dissected free from attachments [19]. All incisions were sutured, wound clipped and sterile glue applied (vetbond). RNA isolation Immediately following cervical dislocation and resection of the organs, specimens (pancreas and anus) allocated for RNA isolation were placed into RNAlater (ThermoFisher Scientific, Waltham, MA). The RNA isolation commenced immediately using Qiazol lysis and homoge- nization using a tissue homogenizer (Brinkmann Instruments, Model PT 10/35, 110 Volts, 6 Amps, 60 Hz). RNA was isolated using the Qiagen RNeasy Kit (Qiagen, Hilden, Germany). The extracted RNA was quantified using a spectrophotometer (ClarioStar Plate Reader, BMG LABTECH, Ortenberg Germany) and diluted to 50 ng/μL. Electrophoresis of the purified RNA was performed with the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA), and each sample demonstrated an RNA Integrity Number (RIN) of 7.5 or higher, indicative of high- quality RNA. DNA recovery from FFPE tissues and MmuPV1 detection by PCR DNA was isolated from two formalin fixed paraffin embedded slides per sample as previously described [14]. PCR was performed using primers specific to the MmuPV1 genome in the L1 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 4 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC region (F: 5’-GGAAGGAGAGAGCAAGTGTATG-3’, R: 5’-GGGTTTGGTGTGTTGGTTTG- 3’) and analyzed via agarose gel. region (F: 5’-GGAAGGAGAGAGCAAGTGTATG-3’, R: 5’-GGGTTTGGTGTGTTGGTTTG- 3’) and analyzed via agarose gel. Quantitative reverse transcription PCR The qPCR was done as previously described [16]. Briefly, 500 ng of RNA was reverse tran- scribed using the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher, Waltham, Ma) per manufacturer protocol. The qPCR was performed on the Thermo Fisher QuantStudio 7 (Thermo Fisher, Waltham, Ma). All reactions were run in triplicate. Results were analyzed using the delta-delta CT method [17]. Expression levels were calculated relative to the average of the C57BL/6J female mice (baseline) or the average of the KC females. The reference group was labeled on each graph. The following TaqMan1 probes were used: Cre (Enterobacteria phage P1 cyclization recombinase, Mr00635245_cn), Pdx1 (pancreatic and duodenal homeo- box 1, Mm00565835_cn) and the house keeping gene Hprt (hypoxanthine guanine phosphori- bosyl transferase, Mm03024075_m1) (Thermo Fisher, Waltham, Ma). 17-beta estradiol silastic capsule preparation and administration KC mice were castrated or ovariectomized at age 6–7 weeks. The first 17-beta estradiol silastic capsule (or sham implant) was implanted 14 days later, and replaced every 4 weeks up to age 9 months. Silastic tubing (Silastic Laboratory Tubing, 1.58 mm inside diameter × 3.18 mm PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 5 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC outside diameter, catalog no. 508–008, Dow Corning) was cut to 4.8mm. The tubing was sealed at one end with medical grade adhesive. The 4.8mm of the capsules was filled with 17-beta estradiol (E2) (17-beta estradiol, 99% pure, catalog no. 50-28-2). Silastic capsules were sealed with silastic medical adhesive, type A (product no. A-100, Dow Corning, pur- chased from Factor II). 17-beta Estradiol-filled silastic capsules have been shown to effectively increase estrogen levels in C57BL/6J mice when implanted as previously described [20, 21]. Before implantation the capsules were soaked in sterile saline overnight at 37˚C. Mice were anesthetized with isoflurane for silastic capsule implantation and given slow-release buprenor- phine (0.5 mg/kg sc). The back was shaved using clippers and sterilized with iodine scrub. An incision was made on the caudal aspect of the back just to the right of midline. Capsules were inserted parallel to the spine, and the incision was closed with wound clips. Statistical analysis Comparisons of tumor development between groups was accomplished using the Fisher’s Exact test. The qPCR data was analyzed using an unpaired, two-tailed t-test with Welch’s cor- rection to evaluate possible expression differences of Pdx1 and Cre in the sample groups. Data was considered significant with a p-value <0.05. Development of anal tumors in KC mice Of the 30 KC mice (16 KC males and 14 KC females) initially evaluated, 16.7% (5/30) devel- oped pancreatic ductal adenocarcinoma (PDAC), which is consistent with previously pub- lished incidence in KC mice by age 9 months (S1A Fig) [8]. There were no statistically significant differences between development of PDAC precursor lesions (PanIN-1, PanIN-2, or PanIN-3) or PDAC in male and female KC mice (S1B Fig). Furthermore, 66.7% (20/30) of the KC mice possessed at least one external tumor on the body surface. Concordant with previ- ous publications, roughly 36% of KC mice developed a facial lesion identified as facial papil- loma [8, 22]. Notably, anal tumors were also observed and identified as invasive anal squamous cell carcinoma (anal SCC) on histopathologic analysis (Fig 1). The tumors became macroscopically visible after 5 months, were clearly evident by 6 months, and of significant size by 9 months (Fig 2). Mice with anal tumors displayed no increase in lethality, with normal mobility and weight gain up until the time of euthanasia (age 9 months). All anal SCC tumors were located at or just distal to the anorectal junction. All the tumors were characterized as grade 1. The neoplastic cells were well differentiated and easily recognized as squamous epithe- lium, infiltrating within a desmoplastic stroma with focal keratinization. Anal SCCs were local- ized to the anus, with no evidence of metastasis to distant organs. The pancreas, stomach, small intestine, colon, spleen, thymus, lungs and liver underwent gross analysis, but no evi- dence of metastasis from anal SCC tumors were found. Additionally, the stomach, spleen, pan- creas, small intestine, and liver underwent histopathologic analysis with no evidence of metastatic spread. Only KC mice developed tumors (i.e. only mice possessing activated Kras- mutation), while age-matched male and female control mice did not develop external tumors (Table 1). https://doi.org/10.1371/journal.pone.0259245.g001 Sex significantly influences anal SCC tumor development in KC mice A significantly higher rate of external tumors was observed in female KC mice compared to male KC mice (100% vs 38%, p = 0.005). While there was no difference in the incidence of facial papilloma between the female and male KC mice (29% vs 38%, p = 0.65), there was a PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 6 / 22 Fig 1. Anal tumor development in KC mice. At age 9 months, anorectal tissue was excised for histological analysis. In male (A) and female (B) C57BL/6J wild type mice, normal anorectal histology was present. Additionally, 9-month-old KC male mice (C) demonstrated normal anorectal histology. In contrast, large perianal tumors PLOS ONE Sex-differences in mutant-Kras induced anal SCC PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 1. Anal tumor development in KC mice. At age 9 months, anorectal tissue was excised for histological analysis. In male (A) and female (B) C57BL/6J wild type mice, normal anorectal histology was present. Additionally, 9-month-old KC male mice (C) demonstrated normal anorectal histology. In contrast, large perianal tumors were grossly evident in 9-month-old female KC mice with invasive anal SCC present on histologic examination. This is indicated by the bracket (D). Scale bars equal 200 μm https://doi.org/10.1371/journal.pone.0259245.g001 stark difference in anal SCC incidence, which occurred exclusively in female KC mice (100% vs 0%, p = 0.00001) (Table 1). In addition to the sex-dependency of anal SCC development, there was complete penetrance (n = 14/14) of anal SCC in female KC mice. These findings were confirmed by histopathologic analysis, where excised anal tissue from wild type males, wild type females, and KC males demonstrated normal microscopic anal histology in compari- son to the anal SCC seen in KC females (Fig 1). These findings demonstrate anal SCC develop- ment is dependent on sex. stark difference in anal SCC incidence, which occurred exclusively in female KC mice (100% vs 0%, p = 0.00001) (Table 1). In addition to the sex-dependency of anal SCC development, there was complete penetrance (n = 14/14) of anal SCC in female KC mice. These findings were confirmed by histopathologic analysis, where excised anal tissue from wild type males, wild type females, and KC males demonstrated normal microscopic anal histology in compari- son to the anal SCC seen in KC females (Fig 1). These findings demonstrate anal SCC develop- ment is dependent on sex. PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Anal carcinogenesis in KC mice was not due to papillomavirus infection Mouse papillomavirus (MmuPV1) has been associated with the development of anal disease and cancer in mice [14, 23]. The animal facility where the mice are housed is routinely screened for MmuPV1 and it has not been detected in our colony, and immunocompetent C57BL/6J mice are known to rapidly clear MmuPV1 before tumor development occurs [23]. Furthermore, the anal tumors that developed in the KC mice were negative for characteristic features of papillomavirus-induced anal SCC during histopathologic evaluation. The overlying squamous mucosa did not exhibit koilocytosis, binucleation or raisanoid nuclei to suggest viral cytopathic effect from histopathology of KC mouse anus (Fig 3A) [24]. Representative anal SCC tumors in KC mice were evaluated for MmuPV1 viral transcripts using RNAScope and for MmuPV1 DNA using PCR [14]. No MmuPV1 signal was detected by RNAScope (Fig 3A), and no MmuPV1 DNA was detected within the anal tumors of the KC mice (Fig 3B). Together, these data show that the anal SCC in this study was not driven by papillomavirus infection. KrasG12D mutation is present in anal tissue of KC mice. The absence of anal SCC in any control mice combined with the presence of anal tumors only in KC mice indicated Kras- mutation was likely responsible for the observed anal SCC. To confirm, we assessed anal tissue for expression of Pdx1 mRNA, Cre mRNA, and activated KrasG12D mutation (genomic DNA) in both male and female mice. In the KC model, the Pdx1 promoter mediates the expression of Cre recombinase in both male and female anal tissues. We compared Pdx1 mRNA levels in male and female WT and KC mouse anus at nine months of age (Fig 4A and 4B), and found similar levels of Pdx1 expression amongst the WT male and female mice (Fig 4A). The Table 1. Sex-dependent incidence of skin and anal lesions. Sex significantly influences anal SCC tumor development in KC mice 7 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 2. Anal tumor progression over time. In female KC mice, anal tumors are visible as an area of congestion at 4 months of age, with mild erythema around the anal region. By age 5 months, an anal tumor is generally evident. By 9 months of age, the anal tumors are significant in size. Despite the large size, the mice are able to maintain weight, consume food, and defecate normally. No mice experienced obstructive symptoms. h //d i /10 1371/j l 0259245 002 Fig 2. Anal tumor progression over time. In female KC mice, anal tumors are visible as an area of congestion at 4 months of age, with mild erythema around the anal region. By age 5 months, an anal tumor is generally evident. By 9 months of age, the anal tumors are significant in size. Despite the large size, the mice are able to maintain weight, consume food, and defecate normally. No mice experienced obstructive symptoms. https://doi.org/10.1371/journal.pone.0259245.g002 https://doi.org/10.1371/journal.pone.0259245.g002 Anal carcinogenesis in KC mice was not due to papillomavirus infection Group Mouse Facial Papilloma Anal SCC Control C57Bl6 Male (n = 12) 0 / 12 (0%) 0 / 12 (0%) C57Bl6 Female (n = 11) 0 / 11 (0%) 0 / 11 (0%) KrasG12D Male (n = 18) 0 / 18 (0%) 0 / 18 (0%) KrasG12D Female (n = 9) 0 / 9 (0%) 0 / 9 (0%) Pdx1-Cre Male (n = 10) 0 / 10 (0%) 0 / 10 (0%) Pdx1-Cre Female (n = 23) 0 / 23 (0%) 0 / 23 (0%) KC Mice KC Male (n = 16) 6 / 16 (38%) 0 / 16 (0%) KC Female (n = 14) 4 / 14 (29%) 14 / 14 (100%) https://doi org/10 1371/journal pone 0259245 t001 Table 1. Sex-dependent incidence of skin and anal lesions. https://doi.org/10.1371/journal.pone.0259245.t001 8 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 3. Anal squamous cell carcinomas in KC mice were negative for MmuPV1 infection. (A) No viral signatures were detected in representative anal tumors from KC mice stained via RNAScope ISH with a probe specific to the MmuPV1 E4 region of the genome. MmuPV1-infected and mock-infected Nod-scid IL2Rγnull (NSG) mouse anal tissues were included as positive and negative controls, respectively. Scale bars equal 100 μm. (B) DNA was recovered from FFPE slides of representative KC female anal squamous cell carcinomas and MmuPV1-infected and mock-infected anal tissues, and PCR was performed using primers specific to the MmuPV1 genome. KC lesions were negative for MmuPV1 DNA. Fig 3. Anal squamous cell carcinomas in KC mice were negative for MmuPV1 infection. (A) No viral signatures were detected in representative anal tumors from KC mice stained via RNAScope ISH with a probe specific to the MmuPV1 E4 region of the genome. MmuPV1-infected and mock-infected Nod-scid IL2Rγnull (NSG) mouse anal tissues were included as positive and negative controls, respectively. Scale bars equal 100 μm. (B) DNA was recovered from FFPE slides of representative KC female anal squamous cell carcinomas and MmuPV1-infected and mock-infected anal tissues, and PCR was performed using primers specific to the MmuPV1 genome. KC lesions were negative for MmuPV1 DNA. https://doi org/10 1371/journal pone 0259245 g003 Fig 3. Anal squamous cell carcinomas in KC mice were negative for MmuPV1 infection. Anal carcinogenesis in KC mice was not due to papillomavirus infection The male KC anal tissue appeared grossly and histologically normal, yet genomic DNA isolated from whole anal tissue demonstrated the same activated KrasG12D mutation (Fig 6, S2 Fig). As a follow-up, the anus from KC mice at the earliest time point allowable (age 6–8 weeks) was excised and genomic DNA isolated (S3 Fig). Activated KrasG12D-mutation was present in both sexes indicating early expression of the oncogene. This was concordant with the clear evidence of tdTomato (Pdx1/Cre) expression at age 12 weeks in the male and female AiKC mice. Interestingly, we were unable to detect a mutant-Kras band (or only a faint band) in one male KC mouse (Fig 6), which was likely due to detection error presence of Pdx1 in anal tissue has also been identified in prior investigations [25]. At nine months, expression of Pdx1 in KC female anus / anal tumor was significantly higher than KC male anus (Fig 4B). We concordantly found Cre-recombinase expression was significantly higher in KC females compared to KC males (Fig 4C). WT mice did not express Cre-recombi- nase due to the absence of Pdx1-Cre transgene. The difference in Pdx1 and Cre mRNA expres- sion at nine months of age was likely related to evaluation of KC female anus / tumor (tumor tissue harboring more Pdx1-Cre expressing cells) versus non-tumor anal tissue of males. Thus, we assessed Pdx1 and Cre expression at five months, prior to onset of macroscopic tumor and found no differences in levels of Cre and Pdx1 expression between the groups of mice (Fig 4D– 4F). To substantiate our expression data, we crossed the KC mice to Ai14 mice to generate the AiKC ‘marker’ mouse model (Rosa26LSL-tdTomato; LSLKrasG12D; Pdx1-Cre). AiKC mice harbor the LSL-tdTomato red fluorescent gene in the Rosa26 locus, and in the presence of Cre-recom- binase, the stop sequence is excised allowing for expression of tdTomato protein, localizing Pdx1 and Cre expression and serving as a marker for expression of activated KrasG12D gene. Both male and female AiKC mice (Fig 5) displayed tdTomato in the anal canal epithelium con- firming Pdx1 and Cre expression and providing an expected localization for mutant KrasG12D expression. Concordantly, isolation of genomic DNA from anal tissue of male and female KC mice demonstrated the activated KrasG12D mutation (Fig 6 and S2 Fig), which was absent in WT mice. Anal carcinogenesis in KC mice was not due to papillomavirus infection (A) No viral signatures were detected in representative anal tumors from KC mice stained via RNAScope ISH with a probe specific to the MmuPV1 E4 region of the genome. MmuPV1-infected and mock-infected Nod-scid IL2Rγnull (NSG) mouse anal tissues were included as positive and negative controls, respectively Scale bars equal 100 μm. (B) DNA was recovered from FFPE slides of representative KC female anal squamous cell carcinomas and MmuPV1-infected and mock-infected anal tissues, and PCR was performed using primers specific to the MmuPV1 genome. KC lesions were negative for MmuPV1 DNA. https://doi.org/10.1371/journal.pone.0259245.g003 presence of Pdx1 in anal tissue has also been identified in prior investigations [25]. At nine months, expression of Pdx1 in KC female anus / anal tumor was significantly higher than KC male anus (Fig 4B). We concordantly found Cre-recombinase expression was significantly higher in KC females compared to KC males (Fig 4C). WT mice did not express Cre-recombi- nase due to the absence of Pdx1-Cre transgene. The difference in Pdx1 and Cre mRNA expres- sion at nine months of age was likely related to evaluation of KC female anus / tumor (tumor tissue harboring more Pdx1-Cre expressing cells) versus non-tumor anal tissue of males. Thus, we assessed Pdx1 and Cre expression at five months, prior to onset of macroscopic tumor and found no differences in levels of Cre and Pdx1 expression between the groups of mice (Fig 4D– 4F). To substantiate our expression data, we crossed the KC mice to Ai14 mice to generate the AiKC ‘marker’ mouse model (Rosa26LSL-tdTomato; LSLKrasG12D; Pdx1-Cre). AiKC mice harbor the LSL-tdTomato red fluorescent gene in the Rosa26 locus, and in the presence of Cre-recom- binase, the stop sequence is excised allowing for expression of tdTomato protein, localizing Pdx1 and Cre expression and serving as a marker for expression of activated KrasG12D gene. Both male and female AiKC mice (Fig 5) displayed tdTomato in the anal canal epithelium con- firming Pdx1 and Cre expression and providing an expected localization for mutant KrasG12D expression. Concordantly, isolation of genomic DNA from anal tissue of male and female KC mice demonstrated the activated KrasG12D mutation (Fig 6 and S2 Fig), which was absent in WT mice. When excising the anus from female KC mice, the specimen was removed en bloc with the large anal tumor, and DNA isolation revealed clear presence of the activated KrasG12D mutation (Fig 6, S2 Fig). PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Anal carcinogenesis in KC mice was not due to papillomavirus infection When excising the anus from female KC mice, the specimen was removed en bloc with the large anal tumor, and DNA isolation revealed clear presence of the activated KrasG12D mutation (Fig 6, S2 Fig). The male KC anal tissue appeared grossly and histologically normal, yet genomic DNA isolated from whole anal tissue demonstrated the same activated KrasG12D mutation (Fig 6, S2 Fig). As a follow-up, the anus from KC mice at the earliest time point allowable (age 6–8 weeks) was excised and genomic DNA isolated (S3 Fig). Activated KrasG12D-mutation was present in both sexes indicating early expression of the oncogene. This was concordant with the clear evidence of tdTomato (Pdx1/Cre) expression at age 12 weeks in the male and female AiKC mice. Interestingly, we were unable to detect a mutant-Kras band (or only a faint band) in one male KC mouse (Fig 6), which was likely due to detection error PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 9 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 4. Pdx1 expression and Cre-recombinase expression in mouse anus. Anal tissue was excised from male and female wild type mice and male and female KC mice at 9 and 5 months of age. mRNA was isolated and RT-PCR performed to evaluate for Pdx1 and Cre-recombinase expression. Male and female C57BL/6J WT mice demonstrate Pdx1 expression in anal tissue at both 5 and 9 months, as did male and female KC mice (A, B, D, E). At 9 months age, KC females express significantly higher amounts of Cre mRNA (C) due to the presence of tumor tissue. At 5 months, male and female KC mice express similar amounts of Pdx1 and Cre mRNA (E, F). https://doi.org/10.1371/journal.pone.0259245.g004 Fig 4. Pdx1 expression and Cre-recombinase expression in mouse anus. Anal tissue was excised from male and female wild type mice and male and female KC mice at 9 and 5 months of age. mRNA was isolated and RT-PCR performed to evaluate for Pdx1 and Cre-recombinase expression. Male and female C57BL/6J WT mice demonstrate Pdx1 expression in anal tissue at both 5 and 9 months, as did male and female KC mice (A, B, D, E). At 9 months age, KC females express significantly higher amounts of Cre mRNA (C) due to the presence of tumor tissue. At 5 months, male and female KC mice express similar amounts of Pdx1 and Cre mRNA (E, F). PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Anal carcinogenesis in KC mice was not due to papillomavirus infection https://doi.org/10.1371/journal.pone.0259245.g004 https://doi.org/10.1371/journal.pone.0259245.g004 https://doi.org/10.1371/journal.pone.0259245.g004 from the excised anal samples. Thus, to confirm, we used FFPE sections generated from the 9-month male KC mice (cohort used to evaluate for tumor formation) to specifically evaluate genomic DNA from the anal canal epithelium. These sections should harbor the activated KrasG12D mutation based on results of tdTomato staining (AiKC mice), and notably, we found that all male mice express mutant KrasG12D (S4 Fig). Together, these data demonstrate the presence of activated KrasG12D mutation in both male and female KC anus, yet an absence of tumor formation in male KC mice. Sex hormone dependence of anal SCC formation. To discern why only female KC mice develop anal tumors despite the presence of activated KrasG12D mutation in both male and female KC mice, we assessed sex hormone dependence. The roles of male and female sex hor- mones in the development of these tumors were judged by castration of male mice (n = 11) and ovariectomy of female mice (n = 13) at 6–7 weeks of age, according to standardized proto- col [18, 19]. Castration dramatically lowers the amount of testosterone that is produced in male mice [18] and, similarly, ovariectomy significantly lowers the amount of estrogen/proges- terone produced in female mice [19, 26]. Castrated KC males displayed an unchanged pheno- type compared to intact KC males, with none of the mice developing anal SCC (0/11 KC castrated males vs 0/16 KC intact males, p = 1) (Table 2). In contrast, ovariectomized KC female mice exhibited a striking change compared to intact KC female mice, with only 15% of the ovariectomized cohort developing macroscopic anal lesions (2/13 ovariectomized KC females vs 14/14 intact KC females (P<0.0001)) (Table 2). This was confirmed on microscopic PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 10 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 5. Tdtomato expression indicates Pdx1-Cre expression and mutant Kras expression in both male and female mice. Anal tissue was excised from 3 month old Ai14; LSL-KrasG12D; Pdx1-Cre (AiKC) female (A, B, C) and male (D, E, F) mice, fixed and frozen embedded and sectioned for H&E analysis (A and D). Additional adjacent sections were prepared for immunohistochemistry (IHC) to identify tdTomato protein at 4x and 10x (B, C, E and F). Scale bars equal 100 μm. Anal carcinogenesis in KC mice was not due to papillomavirus infection Group Mouse Facial Papilloma Anal SCC Castrated/Ovariectomized Controls C57Bl6 Male (n = 2) 0 / 2 (0%) 0 / 2 (0%) C57Bl6 Female (n = 2) 0 / 2 (0%) 0 / 2 (0%) KrasG12D Male (n = 2) 0 / 2 (0%) 0 / 2 (0%) KrasG12D Female (n = 2) 0 / 2 (0%) 0 / 2 (0%) Pdx1-Cre Male (n = 2) 0 / 2 (0%) 0 / 2 (0%) Pdx1-Cre Female (n = 2) 0 / 2 (0%) 0 / 2 (0%) Castrated/Ovariectomized KC Mice Castrated KC Male (n = 11) 4 / 11 (33%) 0 / 11 (0%) Ovariectomized KC Female (n = 13) 4 / 13 (30%) 2 / 13 (15%) https://doi.org/10.1371/journal.pone.0259245.t002 Castrated/Ovariectomized KC Mice analysis, in which the eleven ovariectomized KC female mice without macroscopic tumors demonstrated normal anal histology (i.e. no microscopic tumors or dysplasia) (Fig 7). This remarkable finding indicates that female sex hormones are crucial for KrasG12D-driven anal SCC development in KC mice. analysis, in which the eleven ovariectomized KC female mice without macroscopic tumors demonstrated normal anal histology (i.e. no microscopic tumors or dysplasia) (Fig 7). This remarkable finding indicates that female sex hormones are crucial for KrasG12D-driven anal SCC development in KC mice. Estrogen dependence of anal SCC formation. Estrogen has a known role in the develop- ment of several tumors [27–32] and a correlation with Kras-mutant cancer [29–32]. Thus, we tested whether the anal SCC tumor development in a papillomavirus-negative context was driven by estrogen. Ovariectomized and castrated KC mice were dosed with physiologic levels of 17-β-estradiol (N = 5 for ovariectomized and N = 4 for castrated) or sham dosed as a control (N = 6 for ovariectomized and N = 5 for castrated). To confirm successful E2 administration, uterine weights were assessed (S5 Fig) [21]. E2 dosed females should remain in proestrus and thus have normal uterine weights while the sham dosed mice will have significantly lower uter- ine weights (S5A Fig) [21]. This standard approach enables accurate determination of estrogen reduction as opposed to a single time point (serum) which can vary substantially even in wild type (intact) mice [21]. Sham dosed mice demonstrated significant decrease in uterine weights confirming successful reduction in estrogen levels, while all E2 dosed female KC mice pos- sessed normal uterine weights (i.e. intact) revealing appropriate and sufficient exogenous estrogen administration (S5A Fig). Anal carcinogenesis in KC mice was not due to papillomavirus infection In the sham-dosed ovariectomized KC female group only 33% (2/6) developed a tumor (Table 3), concordant with results seen in the untreated ovariec- tomized mice (33% vs 15%, p-value = 0.56) (Table 2). Meanwhile, in the beta-estradiol (E2) dosed ovariectomized KC female mice, 100% (5/5) developed macroscopically visible anal tumors by 4 months of age (Table 3), ‘rescuing’ the tumor phenotype and again demonstrating stark contrast to ovariectomized mice (100% vs 15%, p-value = 0.001). In the sham-dosed cas- trated KC male group, 0% (0/5) developed a tumor (Table 3), consistent with the results seen in the untreated castrated KC male mice (0/5 vs 0/11, p-value = 1). In contrast, 75% (3/4) of E2-dosed castrated KC male mice developed anal SCC that was macroscopically visible by 8 months of age and confirmed on histopathologic analysis (Fig 8). This remarkable and signifi- cant increase in tumor formation in E2-dosed KC males (75% vs 0%, p-value = 0.0088), when coalesced with the novel data from KC females, demonstrates an estrogen mediated sex- dependent development of Kras-mutant anal SCC. Anal carcinogenesis in KC mice was not due to papillomavirus infection Positive IHC signal for tdTomato reveals the location of Cre expression (Pdx1-Cre), which serves as a marker for the location of mutant KrasG12D expression. Fig 5. Tdtomato expression indicates Pdx1-Cre expression and mutant Kras expression in both male and female mice. Anal tissue was excised from 3 month old Ai14; LSL-KrasG12D; Pdx1-Cre (AiKC) female (A, B, C) and male (D, E, F) mice, fixed and frozen embedded and sectioned for H&E analysis (A and D). Additional adjacent sections were prepared for immunohistochemistry (IHC) to identify tdTomato protein at 4x and 10x (B, C, E and F). Scale bars equal 100 μm. Positive IHC signal for tdTomato reveals the location of Cre expression (Pdx1-Cre), which serves as a marker for the location of mutant KrasG12D expression. https://doi.org/10.1371/journal.pone.0259245.g005 Fig 6. Activated KrasG12D is present in the anal SCC anal tumor tissue. The activated KrasG12D mutation is detected in the female anal tumor tissue, and the male anal tissues. Activated Kras refers to the successful removal of the Lox- stop-Lox codon preceding the KrasG12D. The activated mutation is present in the pancreas due to extensive Pdx1 expression, and absent in tail samples, which lack Pdx1 expression. Positive and negative controls are shown in S2 Fig. https://doi.org/10.1371/journal.pone.0259245.g006 Fig 6. Activated KrasG12D is present in the anal SCC anal tumor tissue. The activated KrasG12D mutation is detected in the female anal tumor tissue, and the male anal tissues. Activated Kras refers to the successful removal of the Lox- stop-Lox codon preceding the KrasG12D. The activated mutation is present in the pancreas due to extensive Pdx1 expression, and absent in tail samples, which lack Pdx1 expression. Positive and negative controls are shown in S2 Fig. https://doi org/10 1371/journal pone 0259245 g006 https://doi.org/10.1371/journal.pone.0259245.g006 11 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Table 2. Sex-dependent incidence of skin and anal lesions after castration or ovariectomy. Estrogen receptor present in male and female anus and female anal tumors Estrogen signaling is mediated through two distinct receptors, ERα and ERβ [33]. Estrogen signaling through ERα has been shown to increase cellular proliferation, particularly within the mammary gland and uterus, while ERβ has been shown to counteract the proliferative effects of ERα [33]. Given the proliferation-inducing role of ERα, we expected that the anal PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 12 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 7. Castrated and ovariectomized KC mice and assessment for anal tumor. Male KC mice were castrated and female KC mice were ovariectomized at 6–7 weeks of age, and the anal tissue evaluated at age 9 months. Macroscopic examination revealed no abnormalities in castrated KC males and most ovariectomized KC females (representative images, A and C). Concordantly, histopathologic examination revealed no microscopic tumor formation in male KC mice or macroscopically normal ovariectomized KC female mice (representative images B and D). Scale bars equal 100 μm. https://doi.org/10.1371/journal.pone.0259245.g007 Fig 7. Castrated and ovariectomized KC mice and assessment for anal tumor. Male KC mice were castrated and female KC mice were ovariectomized at 6–7 weeks of age, and the anal tissue evaluated at age 9 months. Macroscopic examination revealed no abnormalities in castrated KC males and most ovariectomized KC females (representative images, A and C). Concordantly, histopathologic examination revealed no microscopic tumor formation in male KC mice or macroscopically normal ovariectomized KC female mice (representative images B and D). Scale bars equal 100 μm. https://doi.org/10.1371/journal.pone.0259245.g007 https://doi.org/10.1371/journal.pone.0259245.g007 Table 3. Incidence of anal SCC in beta-estradiol dosed KC mice. Group Mouse Anal SCC Intact KC Mice Intact KC Male (n = 16) 0 / 16 (0%) Intact KC Female (n = 14) 14 / 14 (100%) Castrated/Ovariectomized KC Mice Castrated KC Male (n = 11) 0 / 11 (0%) Ovariectomized KC Female (n = 13) 2 / 13 (15%) Dosed and Castrated/Ovariectomized KC Mice Sham Dosed Castrated KC Male (n = 5) 0 / 5 (0%) Beta-Estradiol Dosed Castrated KC Male (n = 4) 3 / 4 (75%) Sham Dosed Ovariectomized KC Female (n = 6) 2 / 6 (33%) Beta-Estradiol Dosed Ovariectomized KC Female (n = 6) 5 / 5 (100%) https://doi.org/10.1371/journal.pone.0259245.t003 Table 3. Incidence of anal SCC in beta-estradiol dosed KC mice. Table 3. Incidence of anal SCC in beta-estradiol dosed KC mice. e 3. Incidence of anal SCC in beta-estradiol dosed KC mice. Estrogen receptor present in male and female anus and female anal tumors PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 13 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 8. Estrogen dosed castrated KC male develop anal SCC. Castrated KC male mice and ovariectomized KC female mice were treated with 17-beta estradiol (E2) o sham (sesame oil). Sham dosed KC males did not develop anal SCC (A), while E2 dosing induced anal SCC in KC males (B, shown by bracket). Similarly, sham dosed KC females did not develop anal SCC (C) while E2 dosing rescued the anal SCC phenotype in KC females (D, shown by bracket). Scale bars equal 200 μm. https://doi.org/10.1371/journal.pone.0259245.g008 PLOS ONE Sex-differences in mutant-Kras induced anal SC Fig 8. Estrogen dosed castrated KC male develop anal SCC. Castrated KC male mice and ovariectomized KC female mice were treated with 17-beta estradiol (E2) o sham (sesame oil). Sham dosed KC males did not develop anal SCC (A), while E2 dosing induced anal SCC in KC males (B, shown by bracket). Similarly, sham dosed KC females did not develop anal SCC (C) while E2 dosing rescued the anal SCC phenotype in KC females (D shown by bracket) Scale bars equal 200 μm Fig 8. Estrogen dosed castrated KC male develop anal SCC. Castrated KC male mice and ovariectomized KC female mice were treated with 17-beta estradiol (E2) or sham (sesame oil). Sham dosed KC males did not develop anal SCC (A), while E2 dosing induced anal SCC in KC males (B, shown by bracket). Similarly, sham dosed KC females did not develop anal SCC (C) while E2 dosing rescued the anal SCC phenotype in KC females (D, shown by bracket). Scale bars equal 200 μm. https://doi.org/10.1371/journal.pone.0259245.g008 epithelium and anal tumors would display ERα expression. To investigate, we performed fluo- rescent IHC using an antibody specific for ERα [15] in the male and female anus. In particular, we assessed ERα in intact KC females and found robust presence of ERα in the anal tissue including the tumor (Fig 9). Moreover, we analyzed normal anus of ovariectomized KC females (no tumor) and KC males and again identified strong presence of ERα in the anal tis- sue, indicating exogenous estrogen in male (E2 dosed) and endogenous estrogen in female KC mice can bind receptor to induce tumor formation (i.e. receptor is not just expressed in tumor). PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Discussion In this study, we identified that LSL-KrasG12D; Pdx1-Cre (KC) mice showed female-specific anal SCC development. Although this is a highly utilized genetically engineered mouse model in the study of PDAC, this novel phenotype has likely been overlooked for several reasons. While earlier studies of KC mice included only males, more recent investigations have included both male and female Kras-mutant mice to assess the development of PDAC. How- ever, these studies focused on concomitant genetic mutations (e.g., Trp53, Ink4a/Arf) [9, 34] or the influence of environmental changes (e.g., high-fat diet) [35] in addition to the Pdx1-Cre driven KrasG12D-mutation, which facilitate the onset of PDAC and consequent death at an early age (roughly four months for LSL-KrasG12D/+; Trp53-/-; Pdx1-Cre mice in our laboratory). Thus, because these tumors were only identifiable starting at 5–6 months age, studies with combination genetic mutations or environmental changes that caused earlier evaluation / demise in male and female KC mice may have conceivably missed the onset of anal SCC growth in female mice. Approximately 85–95% of anal SCC development in humans is due to HPV infection, while the remaining 5–15% of cases occur from an unknown etiology [4]. Although mice can be infected in the anal tract with the mouse papillomavirus MmuPV1 [14, 23], the KC mice in this study did not develop anal SCC as a result of MmuPV1 or HPV infection. Immunocompe- tent mice, such as the KC mice, have been shown to rapidly clear MmuPV1 in the anal tract [23]. Furthermore, there was no evidence of papillomavirus in the colony of this study, on his- topathologic analysis, using RNA Scope or via PCR analysis of the anal SCC that developed within the KC mice. This is an important finding when considering that roughly 5–15% of anal SCC patients do not harbor papillomavirus as the underlying etiology and suggests this is a potential model of interest in studying the etiology of non-papillomavirus induced anal SCC. In this study, the development of anal SCC in the KC mice is due to the activation of the KrasG12D mutation in the anal tissue, which subsequently drives tumor formation. In the KC mouse model, the activation of the KrasG12D mutation is controlled by cells expressing Cre- recombinase from the Pdx-1 promoter [8]. Estrogen receptor present in male and female anus and female anal tumors Together, these results suggest that estrogen binding to ERα may act to potentiate mutant-KrasG12D induced development of anal SCC in KC mice (Fig 9). PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 14 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Fig 9. Estrogen Receptor alpha (ERα) is present in the KC anal tissue. The presence of ERα was detected using immunofluorescence IHC. ERα was found to be present in the anal SCC of intact KC female mice (I), in the anal epithelium of ovariectomized KC female anus lacking tumor formation (II) and in the KC male anus (III). Panels A and B show the fluorescent staining of the receptor at 10x and 40x respectively. Panel C in each group shows the tissue with only the use of the secondary antibody confirming no off target staining. Group IV shows ERα staining of mouse arcuate nucleus as the positive control [15]. https://doi.org/10.1371/journal.pone.0259245.g009 Fig 9. Estrogen Receptor alpha (ERα) is present in the KC anal tissue. The presence of ERα was detected using immunofluorescence IHC. ERα was found to be present in the anal SCC of intact KC female mice (I), in the anal epithelium of ovariectomized KC female anus lacking tumor formation (II) and in the KC male anus (III). Panels A and B show the fluorescent staining of the receptor at 10x and 40x respectively. Panel C in each group shows the tissue with only the use of the secondary antibody confirming no off target staining. Group IV shows ERα staining of mouse arcuate nucleus as the positive control [15]. https://doi.org/10.1371/journal.pone.0259245.g009 15 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Discussion Castration did not alter the anal phenotype of male KC mice suggesting that the lack of testos- terone does not modify the development of anal SCC. In contrast, only two out of thirteen ovariectomized mice developed anal SCC suggesting the tumor development was almost entirely dependent on female physiologic levels of estrogen/progesterone. To further evaluate the involvement of the female sex hormones, we dosed ovariectomized KC female mice and castrated KC male mice with estrogen to see if this would result in the tumor phenotype. We chose to evaluate estrogen because of the strong correlation with other Kras-driven cancers. For example, a study done by Hammond et al. [29] used LSL-KrasG12D mice (K mice in that study) to investigate the sex-differences seen in the development of lung adenocarcinomas. The methodology in this study included ovariectomy in female K mice fol- lowed by activation of the KrasG12D -mutation through intra-nasal exposure of an adenoviral vector expressing Cre recombinase (AdeCre). The authors found a significant reduction in lung tumor burden (quantity and size) compared to intact females. Concordant with the cur- rent study, they successfully rescued the phenotype through estrogen administration using silastic capsules [29]. Furthermore, studies have shown estrogen mediates the development of mutant-Kras-driven endometrial cancer, ovarian cancer and vaginal SCC [30–32]. It has been shown that ERα is present in 4% of human anal SCC samples [27] and that estrogen is essential for activating cell proliferation of human epithelial SCC cell lines [36]. Thus, from these data, it is feasible that estrogens can ultimately influence Kras-induced non-papillomavirus anal SCC development, resulting in the sex-dependent development of anal SCC phenotype in female KC mice. Notably, in our study, ovariectomized KC mice that were dosed with physio- logic levels of beta-estradiol (E2) developed anal SCC at 4 months of age, ‘rescuing’ the tumor phenotype. Furthermore, E2 dosed castrated KC male mice (equivalent dose as females) also developed anal SCC, albeit with a relatively delayed macroscopic onset (8 months) compared to E2 dosed females. This delayed onset despite equivalent E2 dosing may be due to differences in the number of ERα expressing cells or amount of ERα present in anal tissue, and will be evaluated in future studies. Regardless, the data presented strongly suggests that the develop- ment of anal SCC in KC mice is Kras-driven and estrogen mediated. Discussion Although only female KC mice develop anal SCC, both sexes of KC mice were shown to have equivalent expression of Pdx1 and Cre in the anal epithelium at age 5 months. Furthermore, the expression of tdTomato was clearly evident in both male and female AiKC mouse anus, which reveals the location of Pdx1-mediated excision of lox-stop and consequent tdTomato expression (surrogate for location of Kras-mutation). Along with the observation that only KC mice formed anal SCC, this data shows that the devel- opment of anal SCC is driven by the expression of Pdx-1 and Cre in the anal tissue resulting in activation of the mutated KrasG12D gene. While we are not the first group to describe Pdx1 expression outside the pancreas leading to the development of SCC tumors [25], to our knowl- edge, we are the first group to identify both Pdx1 and activated Kras-mutation in anal tissue of KC mice and resultant anal SCC formation. Furthermore, the evidence presented disputes the possibility that sex-bias differences in anal tumor formation was simply due to the absence of Pdx1, Cre or activated KrasG12D expression in male anal tissue. Notably, the activated Kras-mutation did not appear to be present (or very faintly present) in all male anal samples retrieved by whole-excision and analyzed by PCR (one sample of the 9 month cohort). This was likely due to detection error when isolating DNA from whole male anus (i.e. hair follicles, glands, skin, colon), which prompted PCR analysis of genomic DNA from FFPE anus samples (containing anal epithelium) to confirm the presence of activated Kras-mutation in KC male anus. Importantly, the mice in the current study were back-crossed 15 times into the C57BL/6J background. It is conceivable that this highly pure background could be contributing to the unique appearance of anal SCC. 16 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC To understand the female sex predilection for development of anal SCC, we evaluated the roles of sex hormones, as these are likely candidates contributing towards the observed pheno- type. We castrated male mice to achieve significant reduction in circulating testosterone and ovariectomized female mice to reduce systemic production of female sex hormones. This stan- dard approach is the most accurate method to determine sex-hormone dependence [18–21]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Discussion It is important to note that following reduction of estrogen (ovariectomy), 15% (2/13) of female KC mice and 2/6 (33%) sham dosed female KC mice still developed tumors. To confirm that our ovariectomized KC female mice did experience a significant reduction in circulating estrogen, and that the E2 dosed mice possessed sufficient levels of circulating estrogen, we used a standardized approach of uterine weights. This methodology is more accurate than ‘sin- gle time point’ levels of estrogen in circulating blood, due to the substantial variation of circu- lating estrogen in normal females [21]. In contrast, uterine weights reflect the steady levels of estrogen stimulation over an extended period. These techniques helped to confirm successful reduction of (ovariectomy) and rescue of (E2 dosed) circulating estrogen. Although our study revealed a clear correlation between estrogen and tumor formation, and a dramatic change in tumor phenotype with reduction of estrogen, it remains unclear why few ovariectomized KC mice still developed tumors; this may have been related to tumor initiation prior to the onset of ovariectomy. In follow up studies, we will use ERα knock out mice (B6.129P2-Esr1tm1Ksk/J) crossed to KC mice to investigate whether innate absence of estrogen ability to bind ERα pre- vents tumor formation. It is also currently unknown how estrogen and mutant-Kras synergis- tically drive anal tumor formation. Interestingly, one group has developed a non-HPV model PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 17 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC of anal carcinogenesis, using tamoxifen-inducible deletions of Pten and Tgfbr1 [37]. The authors found that anal SCC development was contingent upon STAT3 activation [37]. Can- cers in other organ systems (e.g. pancreas cancer, lung cancer) have shown an interdepen- dence between KrasG12D mutation and heightened STAT3 activity [38, 39], and it has also been shown that estradiol functions to increase STAT3 activation [40]. For example, estradiol was shown to increase STAT3 activation in female rat brain which results in neuroprotection against ischemic brain injury [40]. The association between estrogen and STAT3 activation along with the association between STAT3 activity and mutant KrasG12D-induced cancer for- mation suggests a possible mechanism behind the phenotype of sex-dependent anal SCC development in KC mice. Subsequent analyses will aim to clarify these questions and study limitations, as well as focus on elucidating the specific underlying mechanism by which estro- gen enhances Kras-mutant anal SCC development. Discussion Our study clearly shows the sex-dependent development of anal SCC is tied to presence of physiologic levels of estrogen in female mice and characterizes a clinically relevant subtype of anal SCC. The finding that the Kras-mutation is largely dependent upon estrogen to induce tumor development is an exciting result that may have direct clinical applicability for patients with non-HPV anal SCC that have poorly understood pathogenesis and are known to exhibit resistance to standard of care therapy [5]. Additionally, with the previously unidentified obser- vation of Pdx1-driven Kras-mutation present in anal tissue of KC mice, the novel phenotype described in this study may also provide a new mouse model for evaluation of the non-papillo- mavirus subtype of anal SCC. Supporting information S1 Fig. The incidence of pancreas pathology of male and female mice. A) The incidence of pancreatic precursor lesions (PanIN-1, PanIN-2, PanIN-3) and PDAC in all the KC mice match what was previously reported in this model. B) There is no statistically significant differ- ences in development of pancreatic pathology between male (n = 16) and female KC (n = 14) mice. Both males and females show a 100% incidence of PanIN-1 (p-value = 1). PanIN-1, PanIN-2, PanIN-3 and PDAC male vs female incidence with their p-vales are as follows: 100% vs 100% (p = 1), 6.25% vs 7.15% (P > 0.99), 0% vs 21.43% (P = 0.09) and 6.25% vs 28.57% (P = 0.16). S2 Fig. Genotyping for mutated KrasG12D gene in 9 month old male and female KC mice. Full gel images of mutated Kras genotyping in 9 month old male and female anus (i), pancreas (ii) and tail (iii) next to the cropped image used in the manuscript text (A). The ‘+’ and ‘-‘ indi- cate the positive and negative control bands on each gel. The positive control is DNA from LSL Kras/+; Mx-1 cre/+ mouse bone marrow containing the activated KrasG12D gene (Control (+)) and the negative control is DNA from the pancreas of a C57BL/6J mice (Control (-)). (TIF) S3 Fig. Genotyping for mutated KrasG12D gene in 8 week old male and female KC mice. A) Full gel images taken of the PCR product. This shows the male KC anus with the activated mutant Kras gene at an early age (8 weeks old). Despite the active mutant Kras gene being present, no males develop anal SCC. The full gel images for the anus (i), pancreas (ii) and tail (iii) are to the right of the full image. The ‘+’ and ‘-‘ indicate the positive and negative control bands on each gel. The positive control is DNA from LSL Kras/+; Mx-1 cre/+ mouse bone mar- row containing the activated KrasG12D gene (Control (+)) and the negative control is DNA S3 Fig. Genotyping for mutated KrasG12D gene in 8 week old male and female KC mice. A) Full gel images taken of the PCR product. This shows the male KC anus with the activated mutant Kras gene at an early age (8 weeks old). Despite the active mutant Kras gene being present, no males develop anal SCC. PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Author Contributions Conceptualization: Sean M. Ronnekleiv-Kelly. Data curation: Anne Turco, Simon Blaine-Sauer, Oline K. Ronnekleiv. Data curation: Anne Turco, Simon Blaine-Sauer, Oline K. Ronnekleiv. Formal analysis: Morgan T. Walcheck, Oline K. Ronnekleiv. Formal analysis: Morgan T. Walcheck, Oline K. Ronnekleiv. Funding acquisition: Sean M. Ronnekleiv-Kelly. Investigation: Morgan T. Walcheck, Kristina A. Matkowskyj, Anne Turco, Simon Blaine- Sauer, Manabu Nukaya, Oline K. Ronnekleiv. Methodology: Morgan T. Walcheck, Anne Turco, Simon Blaine-Sauer, Manabu Nukaya, Jes- sica Noel, Oline K. Ronnekleiv, Sean M. Ronnekleiv-Kelly. Resources: Sean M. Ronnekleiv-Kelly. Visualization: Morgan T. Walcheck, Kristina A. Matkowskyj. Writing – original draft: Morgan T. Walcheck, Sean M. Ronnekleiv-Kelly. Writing – review & editing: Kristina A. Matkowskyj, Anne Turco, Simon Blaine-Sauer, Man- abu Nukaya, Jessica Noel, Oline K. Ronnekleiv. Acknowledgments The authors would like to thank the Experimental Animal Pathology Lab (EAPL) for use of its facilities and services. The authors would also like to thank Dr. Jing Zhang for providing posi- tive control DNA for activated Kras-mutation from the bone marrow of LSL Kras/+; Mx-1 cre/ + mice, Dr. Paul Lambert for input regarding MmuPV1-mediated anal tumor formation, and Dr. Chad Vezina for insight into investigating sex-hormone driven cancers. We also thank Ms. Martha A. Bosch for technical assistance. Supporting information The full gel images for the anus (i), pancreas (ii) and tail (iii) are to the right of the full image. The ‘+’ and ‘-‘ indicate the positive and negative control bands on each gel. The positive control is DNA from LSL Kras/+; Mx-1 cre/+ mouse bone mar- row containing the activated KrasG12D gene (Control (+)) and the negative control is DNA PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 18 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC from the pancreas of a C57BL/6J mice (Control (-)). (TIF) S4 Fig. Anus from male KC mice displays the activated KrasG12D mutation. Genomic DNA was isolated from the available FFPE samples from age 9 month male KC mice and analyzed using PCR for the activated KrasG12D mutation. All male anus showed the presence of the acti- vated KrasG12D mutation within the anal tissue (A). The full gel is shown in panel B. (TIF) from the pancreas of a C57BL/6J mice (Control (-)). (TIF) S4 Fig. Anus from male KC mice displays the activated KrasG12D mutation. Genomic DNA was isolated from the available FFPE samples from age 9 month male KC mice and analyzed using PCR for the activated KrasG12D mutation. All male anus showed the presence of the acti- vated KrasG12D mutation within the anal tissue (A). The full gel is shown in panel B. (TIF) S5 Fig. E2 dosed females show increased uterine weights consistent with E2 dosing. The E2 dosed female mice have a significantly increased uterine weight compared to the sham dosed mice indicating successful E2 administration. (TIF) S1 Raw images. (PDF) (TIF) S5 Fig. E2 dosed females show increased uterine weights consistent with E2 dosing. The E2 dosed female mice have a significantly increased uterine weight compared to the sham dosed mice indicating successful E2 administration. (TIF) S1 Raw images. (PDF) S5 Fig. E2 dosed females show increased uterine weights consistent with E2 dosing. The E2 dosed female mice have a significantly increased uterine weight compared to the sham dosed mice indicating successful E2 administration. (TIF) S5 Fig. E2 dosed females show increased uterine weights consistent with E2 dosing. The E2 dosed female mice have a significantly increased uterine weight compared to the sham dosed mice indicating successful E2 administration. (TIF) S5 Fig. E2 dosed females show increased uterine weights consistent with E2 dosing. The E2 dosed female mice have a significantly increased uterine weight compared to the sham dosed mice indicating successful E2 administration. (TIF) References 1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer Statistics, 2021. CA Cancer J Clin. 2021; 71: 7–33. https://doi.org/10.3322/caac.21654 PMID: 33433946 2. Osborne MC. Anal squamous cell carcinoma: An evolution in disease and management. World J Gas- troenterol. 2014; 20: 13052. https://doi.org/10.3748/wjg.v20.i36.13052 PMID: 25278699 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 19 / 22 PLOS ONE Sex-differences in mutant-Kras induced anal SCC 3. Baricevic I, He X, Chakrabarty B, Oliver AW, Bailey C, Summers J, et al. High-sensitivity human papil- loma virus genotyping reveals near universal positivity in anal squamous cell carcinoma: Different impli- cations for vaccine prevention and prognosis. Eur J Cancer. 2015; 51: 776–785. https://doi.org/10. 1016/j.ejca.2015.01.058 PMID: 25702585 4. Bernardi M-P, Ngan SY, Michael M, Lynch AC, Heriot AG, Ramsay RG, et al. Molecular biology of anal squamous cell carcinoma: implications for future research and clinical intervention. Lancet Oncol. 2015; 16: e611–e621. https://doi.org/10.1016/S1470-2045(15)00292-2 PMID: 26678214 5. Meulendijks D, Tomasoa NB, Dewit L, Smits PHM, Bakker R, van Velthuysen M-LF, et al. HPV-negative squamous cell carcinoma of the anal canal is unresponsive to standard treatment and frequently carries disruptive mutations in TP53. Br J Cancer. 2015; 112: 1358–1366. https://doi.org/10.1038/bjc.2015.20 PMID: 25871546 6. Morris VK, Rashid A, Rodriguez-Bigas M, Das P, Chang G, Ohinata A, et al. Clinicopathologic Features Associated With Human Papillomavirus/p16 in Patients With Metastatic Squamous Cell Carcinoma of the Anal Canal. The Oncologist. 2015; 20: 1247–1252. https://doi.org/10.1634/theoncologist.2015- 0091 PMID: 26382740 7. Chung JH, Sanford E, Johnson A, Klempner SJ, Schrock AB, Palma NA, et al. Comprehensive genomic profiling of anal squamous cell carcinoma reveals distinct genomically defined classes. Ann Oncol. 2016; 27: 1336–1341. https://doi.org/10.1093/annonc/mdw152 PMID: 27052656 8. Hingorani SR, Iii EFP, Maitra A, Rajapakse V, King C, Jacobetz MA, et al. Preinvasive and invasive duc- tal pancreatic cancer and its early detection in the mouse. CANCER CELL. 2003; 14. https://doi.org/10. 1016/s1535-6108(03)00309-x PMID: 14706336 9. Hingorani SR, Wang L, Multani AS, Combs C, Deramaudt TB, Hruban RH, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal ade- nocarcinoma in mice. Cancer Cell. 2005; 7: 469–483. https://doi.org/10.1016/j.ccr.2005.04.023 PMID: 15894267 10. Jackson EL. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev. 2001; 15: 3243–3248. https://doi.org/10.1101/gad.943001 PMID: 11751630 11. Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S, et al. Endogenous oncogenic K- rasG12D stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell. References https://doi.org/10.1038/nm.3646 PMID: 25326799 23. Cladel NM, Budgeon LR, Balogh KK, Cooper TK, Brendle SA, Christensen ND, et al. Mouse papilloma- virus infection persists in mucosal tissues of an immunocompetent mouse strain and progresses to can- cer. Sci Rep. 2017; 7: 16932. https://doi.org/10.1038/s41598-017-17089-4 PMID: 29208932 24. Darragh TM, Colgan TJ, Thomas Cox J, Heller DS, Henry MR, Luff RD, et al. The Lower Anogenital Squamous Terminology Standardization Project for HPV-associated Lesions: Background and Con- sensus Recommendations From the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology. Int J Gynecol Pathol. 2013; 32: 76–115. https://doi.org/10.1097/ PGP.0b013e31826916c7 PMID: 23202792 25. Mazur PK, Gru¨ner BM, Nakhai H, Sipos B, Zimber-Strobl U, Strobl LJ, et al. Identification of Epidermal Pdx1 Expression Discloses Different Roles of Notch1 and Notch2 in Murine KrasG12D-Induced Skin Carcinogenesis In Vivo. Bauer JA, editor. PLoS ONE. 2010; 5: e13578. https://doi.org/10.1371/journal. pone.0013578 PMID: 21042537 26. Koebele SV, Bimonte-Nelson HA. Modeling menopause: The utility of rodents in translational behav- ioral endocrinology research. Maturitas. 2016; 87: 5–17. https://doi.org/10.1016/j.maturitas.2016.01. 015 PMID: 27013283 27. Wong NACS. Oestrogen receptor can be expressed by normal and dysplastic anal squamous epithelia but only rarely by anal squamous cell carcinoma. Histopathology. 2018; 73: 1039–1040. https://doi.org/ 10.1111/his.13720 PMID: 30062743 28. Zhang D, Ku J, Yi Y, Zhang J, Liu R, Tang N. The prognostic values of estrogen receptor alpha and beta in patients with gastroesophageal cancer: A meta-analysis. Medicine (Baltimore). 2019; 98: e17954. https://doi.org/10.1097/MD.0000000000017954 PMID: 31725654 29. Hammoud Z, Tan B, Badve S, Bigsby RM. Estrogen promotes tumor progression in a genetically defined mouse model of lung adenocarcinoma. Endocr Relat Cancer. 2008; 15: 475–483. https://doi. org/10.1677/ERC-08-0002 PMID: 18509000 30. Ring KL, Yates MS, Schmandt R, Onstad M, Zhang Q, Celestino J, et al. Endometrial Cancers With Activating KRas Mutations Have Activated Estrogen Signaling and Paradoxical Response to MEK Inhi- bition: Int J Gynecol Cancer. 2017; 27: 854–862. https://doi.org/10.1097/IGC.0000000000000960 PMID: 28498246 31. Kato S, McFall T, Takahashi K, Bamel K, Ikeda S, Eskander RN, et al. KRAS -Mutated, Estrogen Receptor-Positive Low-Grade Serous Ovarian Cancer: Unraveling an Exceptional Response Mystery. The Oncologist. 2021; 26. https://doi.org/10.1002/onco.13702 PMID: 33528846 32. Wang X, Prac¸a MSL, Wendel JRH, Emerson RE, DeMayo FJ, Lydon JP, et al. Vaginal Squamous Cell Carcinoma Develops in Mice with Conditional Arid1a Loss and Gain of Oncogenic Kras Driven by Pro- gesterone Receptor Cre. Am J Pathol. 2021; 191: 1281–1291. https://doi.org/10.1016/j.ajpath.2021.03. 013 PMID: 33882289 33. References 2004; 5: 375–387. https://doi.org/10.1016/s1535-6108(04)00085-6 PMID: 15093544 12. Tien JC-Y, Chugh S, Goodrum AE, Cheng Y, Mannan R, Zhang Y, et al. AGO2 promotes tumor pro- gression in KRAS-driven mouse models of non–small cell lung cancer. Proc Natl Acad Sci. 2021; 118: e2026104118. https://doi.org/10.1073/pnas.2026104118 PMID: 33972443 13. Ramesh PS, Madegowda V, Kumar S, Narasimha S, R PS, Manoli NN, et al. DNA extraction from archived hematoxylin and eosin-stained tissue slides for downstream molecular analysis. World J Meth- odol. 2019; 9: 32–43. https://doi.org/10.5662/wjm.v9.i3.32 PMID: 31799154 14. Blaine-Sauer S, Shin M-K, Matkowskyj K, Ward-Shaw E, Lambert PF. A novel model for papillomavi- rus-mediated anal disease and cancer using the mouse papillomavirus. mBio. https://doi.org/10.1128/ mBio.01611-21 PMID: 34281391 15. Friend KE, Resnick EM, Ang LW, Shupnik MA. Specific modulation of estrogen receptor mRNA iso- forms in rat pituitary throughout the estrous cycle and in response to steroid hormones. Mol Cell Endo- crinol. 1997; 131: 147–155. https://doi.org/10.1016/s0303-7207(97)00098-1 PMID: 9296373 16. Baker BB, Yee JS, Meyer DN, Yang D, Baker TR. Histological and Transcriptomic Changes in Male Zebrafish Testes Due to Early Life Exposure to Low Level 2,3,7,8-Tetrachlorodibenzo- p -Dioxin. Zebra- fish. 2016; 13: 413–423. https://doi.org/10.1089/zeb.2016.1275 PMID: 27618130 17. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative CT method. Nat Protoc. 2008; 3: 1101–1108. https://doi.org/10.1038/nprot.2008.73 PMID: 18546601 18. Valkenburg KC, Amend SR, Pienta KJ. Murine Prosttae Micro-dissection and Surgical Castration. J Vis Exp. 2016; 53984. https://doi.org/10.3791/53984 PMID: 27213557 19. Souza V.R., Mendes E., Casaro M., Antiorio A.T.F.B., Oliveira F.A., Ferreira C.M. Description of Ovari- ectomy Protocol in Mice.) Pre-Clinical Models Methods in Molecular Biology. New York, NY: Humana Press; 20. Van Steenbrugge GJ, Groen M, De Jong FH, Schroeder FH. The use of steroid-containing silastic implants in male nude mice: Plasma hormone levels and the effect of implantation on the weights of the ventral prostate and seminal vesicles. The Prostate. 1984; 5: 639–647. https://doi.org/10.1002/pros. 2990050610 PMID: 6494059 21. Becker JB, Arnold AP, Berkley KJ, Blaustein JD, Eckel LA, Hampson E, et al. Strategies and Methods for Research on Sex Differences in Brain and Behavior. Endocrinology. 2005; 146: 1650–1673. https:// doi.org/10.1210/en.2004-1142 PMID: 15618360 20 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC 22. Scho¨nhuber N, Seidler B, Schuck K, Veltkamp C, Schachtler C, Zukowska M, et al. A next-generation dual-recombinase system for time- and host-specific targeting of pancreatic cancer. Nat Med. 2014; 20: 1340–1347. 40. Dziennis S, Jia T, Ronnekleiv OK, Hurn PD, Alkayed NJ. Role of Signal Transducer and Activator of Transcription-3 in Estradiol-Mediated Neuroprotection. J Neurosci. 2007; 27: 7268–7274. https://doi. org/10.1523/JNEUROSCI.1558-07.2007 PMID: 17611279 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 Sex-differences in mutant-Kras induced anal SCC References Paterni I, Granchi C, Katzenellenbogen JA, Minutolo F. Estrogen receptors alpha (ERα) and beta (ERβ): Subtype-selective ligands and clinical potential. Steroids. 2014; 90: 13–29. https://doi.org/10. 1016/j.steroids.2014.06.012 PMID: 24971815 34. Aguirre AJ. Activated Kras and Ink4a/Arf deficiency cooperate to produce metastatic pancreatic ductal adenocarcinoma. Genes Dev. 2003; 17: 3112–3126. https://doi.org/10.1101/gad.1158703 PMID: 14681207 35. Chang H-H, Moro A, Takakura K, Su H-Y, Mo A, Nakanishi M, et al. Incidence of pancreatic cancer is dramatically increased by a high fat, high calorie diet in KrasG12D mice. Rooman I, editor. PLOS ONE. 2017; 12: e0184455. https://doi.org/10.1371/journal.pone.0184455 PMID: 28886117 36. Kato K, Ueoka Y, Kato K, Hachiya T, Nishida J, Wake N. Contribution of enhanced transcriptional acti- vation by ER to [12Val] K-Ras mediated NIH3T3 cell transformation. Oncogene. 1997; 15: 3037–3046. https://doi.org/10.1038/sj.onc.1201497 PMID: 9444952 37. Bu L-L, Li Y-C, Yu G-T, Liu J-F, Deng W-W, Zhang W-F, et al. Targeting phosphorylation of STAT3 delays tumor growth in HPV-negative anal squamous cell carcinoma mouse model. Sci Rep. 2017; 7: 6629. https://doi.org/10.1038/s41598-017-06643-9 PMID: 28747781 38. Grabner B, Schramek D, Mueller KM, Moll HP, Svinka J, Hoffmann T, et al. Disruption of STAT3 signal- ling promotes KRAS-induced lung tumorigenesis. Nat Commun. 2015; 6: 6285. https://doi.org/10.1038/ ncomms7285 PMID: 25734337 39. D’Amico S, Shi J, Martin BL, Crawford HC, Petrenko O, Reich NC. STAT3 is a master regulator of epi- thelial identity and KRAS-driven tumorigenesis. Genes Dev. 2018; 32: 1175–1187. https://doi.org/10. 1101/gad.311852.118 PMID: 30135074 21 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0259245 November 4, 2021 November 4, 2021 PLOS ONE Sex-differences in mutant-Kras induced anal SCC Sex-differences in mutant-Kras induced anal SCC 22 / 22
https://openalex.org/W1575452751	https://ccforum.biomedcentral.com/track/pdf/10.1186/s13054-015-0948-z	English	null	‘Biomarking’ infection during continuous renal replacement therapy: still relevant?	Critical care	2,015	cc-by	638	* Correspondence: patrick.honore@az.vub.ac.be ICU Department, Universitair Ziekenhuis Brussel, Vrije Universiteit Brussel University, 101, Laarbeeklaan, 1090 Jette, Brussels, Belgium © 2015 Honore et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Competing interests h h d l h The authors declare that they have no competing interests. The authors declare that they have no competing interests. ‘Biomarking’ infection during continuous renal replacement therapy: still relevant? Patrick M Honore*, Rita Jacobs, Inne Hendrickx, Elisabeth De Waele, Viola Van Gorp and Herbe See related research by Park et al., http://ccforum.com/content/18/6/640 See related research by Park et al., http://ccforum.com/content/18/6/640 Abbreviations We greatly appreciated the recent research article by Park and colleagues in Critical Care showing that pro- calcitonin (PCT) is not superior to C-reactive protein (CRP) as a marker for infection in renal impairment [1]. Additionally, we would like to comment on the validity of CRP and PCT measurement during continuous renal replacement therapy (CRRT). Both acute-phase proteins are indeed effectively cleared by CRRT. CRP is predom- inantly present as a monomer (mCRP) in the blood [2] and is removed by all forms of CRRT because its molecu- lar weight (22-25 kDa) lies below the cutoff permeability limits of all classic dialysis membranes [3]. Most of the PCT mass is easily eliminated by convection [4]. However, although mCRP and PCT are adequately filtered, substan- tial amounts of both are adsorbed on the dialysis mem- brane [3, 4]. Therein lies a clinically relevant but poorly recognized problem! In fact, highly adsorptive dialysis membranes are increasingly applied to CRRT in many in- tensive care units worldwide. The use of such membranes will inherently accentuate mCRP and PCT removal. As a result, plasma levels of both biomarkers risk becoming falsely low during CRRT and thereby losing all potential to help clinicians diagnose or evaluate infection. The recently introduced sepsis biomarkers brain natriuretic peptide and N-terminal pro-brain natriuretic peptide have been proclaimed to be superior to CRP and PCT, but whether they perform better under CRRT is doubtful because their molecular weights (3.5 and 8.5 kDa, respectively) already predict highly effective clearance by both high- and low- flux membranes [5]. CRP: C-reactive protein; CRRT: Continuous renal replacement therapy; mCRP: Monomeric C-reactive protein; PCT: Procalcitonin. Honore et al. Critical Care (2015) 19:232 DOI 10.1186/s13054-015-0948-z Open Access Open Access References 1. Park JH, Kim DH, Jong HR, Kim MJ, Jung SH, Lee JH, et al. Clinical relevance of procalcitonin and C-reactive protein as infection markers in renal impairment: a cross-sectional study. Crit Care. 2014;18:640. 1. Park JH, Kim DH, Jong HR, Kim MJ, Jung SH, Lee JH, et al. Clinical relevance of procalcitonin and C-reactive protein as infection markers in renal impairment: a cross-sectional study. Crit Care. 2014;18:640. y 2. Taylor KE, van den Berg CW. Structural and functional comparison of native pentameric, denatured monomeric and biotinylated C-reactive protein. Immunology. 2007;120:404–11. gy 3. Honore PM, Jacobs R, De Waele E, Van Gorp V, Spapen HD. Biomarkers of inflammation during continuous renal replacement therapy: sensors, players, or targets ? Blood Purif. 2014;38:102–3. 4. Dahaba AA, Elawady GA, Rehak PH, List WF. Procalcitonin and proinflammatory cytokine clearance during continuous venovenous haemofiltration in septic patients. Anaesth Intensive Care. 2002;30:269–74. 5. Wahl HG, Graf S, Renz H, Fassbinder W. Elimination of the cardiac natriuretic peptides B-type natriuretic peptide (BNP) and N-terminal proBNP by hemodialysis. Clin Chem. 2004;50:1071–4. haemofiltration in septic patients. Anaesth Intensive Care. 2002;30:269–74. 5. Wahl HG, Graf S, Renz H, Fassbinder W. Elimination of the cardiac natriuretic peptides B-type natriuretic peptide (BNP) and N-terminal proBNP by hemodialysis. Clin Chem. 2004;50:1071–4.
https://openalex.org/W4384455659	https://journal.access-bg.org/journalfiles/journal/issue-4-3-2023/challenges_of_sustainable_ecotourism_and_its_impact_on_economy-georgian case.pdf	English	null	Challenges of sustainable ecotourism and its impact on economy (Georgian case)	Access	2,023	cc-by	3,914	ABSTRACT This article discusses main aspects of sustainable ecotourism, as the economic field (environmental protection, economy, and social well-being), which plays an important role in increasing sustainability of the world economy and is one of the strongest priority powers. The role of the state for the sustainable development of sustainable ecotourism, specific programs for the development of national sustainable tourism are analyzed and the expected risks that may arise in this process are evaluated (environmental protection, economic and social aspects). Also, statistical data analyzes the impact of tourism on the economy and future tourism sustainability Objectives: Define the role of sustainable ecotourism to ensure sustainability of the society, analyze and assess its impact on the economy. Methods/Approach: Scientific analysis methods such as deduction, analysis and synthesis, analysing graphic data, dynamic research, forecasting, and others. Results: The research showed that analysing development of ecotourism sector and sustainable development, it is focusing on economical, ecological, social and cultural factors. The results of the study highlight that ecotourism is an integral part of sustainable tourism development. Conclusions: The paper examines the mechanisms and activities ssustainable development and ssustainable tourism, as an economic sector, are of crucial importance for many developing countries, and as an economic phenomenon, they cover many aspects of current issues of society's development. Keywords: sustainable ecotourism, sustainable economic development, business environment JEL classification: M000, O180, O440, Q570 Paper type: Case Study JEL classification: M000, O180, O440, Q570 Paper type: Case Study JEL classification: M000, O180, O440, Q570 Paper type: Case Study JEL classification: M000, O180, O440, Q570 Paper type: Case Study Citation: Mushkudiani, Z., Chkhirodze, D., Tevdoradze, S. (2023). Challenges of sustainable ecotourism and its impact on economy (Georgian case). Access to science, business, innovation in the digital economy, ACCESS Press, 4(3), 400- 408, https://doi.org/10.46656/access.2023.4.3(5) Zurab Mushkudiani1, Darejan Chkhirodze2, Sopiko Tevdoradze3 1), 3, Guram Tavartkiladze Teaching University, Tbilisi, Georgia 2) Kutaisi University, Kutaisi, Georgia e-mails: 1 zurab.mushkudiani@gttu.edu.ge, 2darejan.chkhirodze@unik.edu.ge, 3sofo.tevdoradze@gttu.edu.ge Received: 13, Feb 2023; Accepted: 13 March 2023 Online Published: 16 July 2023 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) ISSN 2683-1007 (Online) INTRODUCTION Due to various economic, social and political processes of globalization tourism is the leading economic sector in many countries. The strategic approach should be focused mainly on building the desired image of the destination ... (Konstantinova, 2022). Ecotourism is one of the most quickly advancing areas of global tourism. Ecotourism is based on the fundamental principles of sustainable tourism, therefore ecotourism is an integral part of sustainable tourism (Dorokhov et al., 2023; Lazarova et al., 2023; Petrova, Buzko & Dyachenko, 2018). 400 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ ISSN 2683-1007 (Online) ISSN 2683-1007 (Online) https://journal.access-bg.org/ Sustainable development- system of social development, which considering the interests of economic development and environmental protection of the community, ensures human well-being, the increase in the quality of life and the right of future generations to enjoy natural resources and the environment protected from irreversible quantitative and qualitative changes to the extent possible (Brıassoulis, 2002). Sustainable development implies a form of economic growth, which ensures social well-being is short, medium, and most importantly long-term. It is based on the principle that today’s needs shall be satisfied in a manner which does not threaten future generations. Development of a sustainable environment establishes conditions for long-term economic development, considering environmental protection issues (Odinokova et al., 2018). In 1995 Copenhagen world summit on sustainable development has highlighted combat against social segregation, and need for public health protection. Biological resources should be used in the manner and speed, which will not result in degradation of biological diversity and will be preserved considering the needs of future generation. 2030 Agenda for sustainable development (includes 17 goals - sustainable development goals (SDGs), and 169 objectives for achievement of these goals) is to be adopted within 15 years to ensure the solution of the most important tasks for the well-being of all countries and people of the world, which means: ending extreme poverty, preventing climate change and environmental degradation, reducing inequality and injustice. The World Tourism Organization actively participates in the realization of the provisions of the sustainable development of tourism, which are established in the document “Agenda for Travel and Tourism Sectors in the XXI Century". ACCESS Journal: J Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ The idea of ecotourism includes the following elements: 1. Natural environment; 2. Ecological and cultural sustainability; 3. Education; 4. Ensuring local and national income; increased interest in ecotourism is facilitated by growing public desire to visit and experience the natural environment and to adopt minimal impact practices to preserve the quality of the environment visited. Today, as increased number of travelers seek to visit untouched nature and cultural heritage, ecotourism is experiencing significant growth. Sustainable ecotourism shall: - ensure the optimal use of natural resources (which are the main element of tourism development), promote the correct management of ecological processes and thereby preserve and protect natural resources and biodiversity; - respect local social-cultural characteristics, preserve cultural heritage and traditions; - ensure fair distribution of socio-economic benefits for all parties involved in the process - employment, profit, social security and thereby contribute to reducing the level of poverty. - ensure fair distribution of socio-economic benefits for all parties involved in the process - employment, profit, social security and thereby contribute to reducing the level of poverty. In September 2017, Georgia hosted the 67th conference of AIEST (founded in 1951), the most influential international association of scientific and practical tourism experts, whose goal was to improve the world's sustainable tourism industry based on the analysis of the latest tourism trends and developments expected in this area, as well as forecasting challenges and finding ways to solve them. During the conference, the sustainability of tourism business in Georgia and related challenges and business models of sustainable development of tourism were also discussed. According to the international experts, sustainable development of Georgia implies balancing two business models. On the one hand there is a society-oriented business model, which is focused on the small and medium size businesses, and on the other hand, there is a cooperation- oriented business model, which is focused on resort development, for example ski resorts and recreation centres. The experts believe that, balancing these two business models is required for development of sustainable tourism in Georgia. INTRODUCTION Based on this document, the “Silk Road Campaign” is being launched, and many countries including Georgia are part of it. In August 2002, at the World Summit on Sustainability Development in Johannesburg addressed Sustainable Tourism, the program of the World Tourism Organization UNWTO and UNCTAD was approved - a number of recommendations for facilitation of sustainable economy in developed and developing countries were shared. Sustainable tourism - Sustainable Tourism-Eliminating Poverty (ST-EP)- this program aims at achieving two main goals: sustainable development of tourism, and its role in elimination of poverty, and sustainable economic development of less developed and developing countries. Ecotourism is the type of travel, which not only serves to preservation on ecosystem, but also respects the interests of local population (Goeldner, 2010). Main principles of ecotourism are as follows: minimization of physical, social, behavioural and psychological impacts; raising awareness and respecting environment and culture; ensuring positive experience for local population, as well as for tourists; direct financial benefits for protection and preservation of the nature; ensuring financial benefit for local population and private industry; providing memorable experience to the tourists (Kotler, Makens, 2017) , which will support the host country in raising awareness on political, natural and social climate; respecting rights and believes of the local population, recognition, protection and cooperation; establishment, construction and exploitation of facilities with minimum impact on the environment (Dorobantu & Nıstoreanu, 2012). 401 ISSN 2683-1007 (Online) METHODOLOGY The article is based on the scientific literature, by analysing ecotourism as an integral part of sustainable tourism development, general research methods were used - systematic, logical and comparative analysis of scientific literature, synthesis. In this literature search the main search criteria were ecotourism, sustainable development, which were searched separately and combined. ACCESS Journal: It shall be noted that, the cooperation between the private and public sectors in Georgia is currently governed by the state entities, which is why the interesting innovative concepts presented by associations and non-governmental organizations in the field of tourism are rarely taken into account. In this regard European practice shows that, cooperation between the private and public sector is important, and attempts of such cooperation are observed in Georgia as well (Abuseridze et al., 2022). This process also supports the development of society, which in turn is crucial for the creation of authentic and reliable touristic products (Vadachkoria & Okrotsvaradze, 2011). Development of sustainable society is one of the key factors for development of sustainable tourism. It is noteworthy that, tourism as strategic field in mentioned in three target categories, in particular: goal 8 – to promote sustained, inclusive, and sustainable economic growth, full and productive employment and decent work for all; goal 12 – to ensure sustainable consumption and production patterns; goal 14 – to conserve and sustainably use the oceans, seas and marine resources for sustainable development purposes. All the above- mentioned indicate the importance of tourism as a universal field for the further development of our planet. For the sustainable development of tourism, the Paris Agreement of December 12, 2015, within the framework 402 ISSN 2683-1007 (Online) ACCESS Journal: Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ of the United Nations Framework Convention on Climate Change should also be considered, as climate change is one of the main challenges for the development of the tourism sector in the long term. The target of the goal eight is also separation of economic growth and environmental degradation- sustainable consumption and production- a ten-year program; implementation of “Global Jobs Pact” (ILO), and others. The objective 9.1 implies development of quality, reliable, sustainable, and resilient infrastructure. For example, the "Kakheti Regional Development Project 2012-16" or the planned construction of waste processing facilities correspond to the above-mentioned objective, and the Ministry of Regional Development and Infrastructure is responsible for these projects. However, the role of tourism in sustainable development of the economy is not limited to these three indicators, as it is directly or indirectly related to other sustainable development goals (SDGs). Relevance of the subject We have studied, analysed, and assessed economic indicators in cases of stability and instability (pandemic), as well as impact on economy. We have studied, analysed, and assessed economic indicators in cases of stability and instability (pandemic), as well as impact on economy. Table 1. Income from international travels (thousand US Dollars) 2014 2015 2016 2017 2018 2019 2020 2021 2022 Total 1 787 140 1 868 479 2 110 709 2 704 340 3 222 074 3 268 654 541 687 1 244 942 3 516 634 Including:          I quarter 315 461 305 180 342 590 425 673 550 730 578 442 427 699 53 551 393 708 II quarter 429 690 409 470 497 441 629 854 809 969 877 626 28 842 246 123 748 822 III quarter 649 775 742 849 813 161 1 078 274 1 210 561 1 126 771 42 534 565 981 1 374 653 IV quarter 392 215 410 979 457 517 570 538 650 814 685 814 42 611 379 287 999 451 Source: Author’s illustration Table 1. Income from international travels (thousand US Dollars) In the third quarter of 2022, the average monthly number of local visitors aged 15 and older on the territory of Georgia was determined by 1,286.8 thousand, and the average monthly number of visits made by them to the territory of the country amounted to 1,470.8 thousand. In the mentioned period, the average monthly number of visitors is 6.7% less than the similar indicator of the third quarter of 2021. 403 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ Figure 1. In the III quarters of 2021-2022, the number of departures of Georgian resident travelers and the number of visits made by Georgian resident visitors abroad. Source: National Statistics Office of Georgia 177,4 693,4 159,2 531,8 0 100 200 300 400 500 600 700 800 IIIquater 2021 III quater, 2022 ათასი Axis Title Excursions of resident travelers of Georgia Visits made by resident visitors of Georgia ISSN 2683-1007 (Online) Figure 1. In the III quarters of 2021-2022, the number of departures of Georgian resident travelers and the number of visits made by Georgian resident visitors abroad. Relevance of the subject Source: National Statistics Office of Georgia In the third quarter of 2022, the average monthly number of touristic visits by Georgian residents to the territory of Georgia was 774.7 thousand, which is 4.8 percent less than the corresponding period of the previous year. The diagram 2 shows the distribution of the number of visits made by resident visitors of Georgia based on the type of visit (III quarter, 2022). Number of visits of the Georgian citizens based on the data on visits during the III Quarter of 2021-2022. According to the data of 2022, during the third quarter, 1,693,400 Georgian residents traveled abroad, which is 3.9 times more than the indicator of the corresponding period of the previous year. The majority of visits- 46.3 percent were made by travelers aging 31-50. Distribution of travels of Georgian residents by age groups during the III quarters of 2021-2022 The number of visits abroad by Georgian residents amounted to 531.8 thousand, which is 3.3 times more than the indicator of the corresponding period of 2021. 404 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org Figure 2. Distribution of trips of Georgian resident travelers by age groups in the III quarters of 2021-2022 Source: National Statistics Office of Georgia 7,9 44,4 82,6 40 2,5 24 162,5 320,8 170,4 15,8 0 50 100 150 200 250 300 350 0-14 15-30 31-50 51-70 71+ Axis Title Distribution of trips of Georgian resident travelers by age groups in the III quarters of 2021-2022 III quater, 2021 წელი III quater, 2022 წელი ISSN 2683-1007 (Online) Figure 2. Distribution of trips of Georgian resident travelers by age groups in the III quarters of 2021-2022 Source: National Statistics Office of Georgia Based on the statistical data provided above, we can say that sustainable ecotourism, as the economic sphere plays a decisive role for our county and as an economic phenomenon covers many aspects of current issues related to society development. By analyzing international and local trends, we can conclude that sustainable development of tourism decreases the level of migration from the mountainous areas. It helps increase the volume of local agricultural products in tourist areas and reduces the influx of imported goods. Relevance of the subject Along with increased number of the tourists, the state is implementing effective measures in order to preserve cultural heritage and protect natural treatment resources in touristic areas, as well as to protect it from depletion and destruction, and to avoid disasters. Georgia is an inspired tourist destination, which has huge tourist potential, and for preservation purposes it is crucial for Georgian legislation on tourism to take into account sustainable tourism development preconditions for following decades, and regulate the destination management organization. The legislation to regulate the cultural heritage, and nature protection, as well as conservation, legislation to regulate tourism payment infrastructure is also required. CONCLUSION In conclusion, we would like to point out that sustainable development and sustainable tourism, as an economic sector, are of crucial importance for many developing countries, and as an economic phenomenon, they cover many aspects of current issues of society's development. It can generate increased employment opportunities 405 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ ISSN 2683-1007 (Online) https://journal.access-bg.org/ and large cash flows for local businesses. The level of tourism sustainability (including ecological tourism), constant growth of its economic importance was deemed as one of the major factors for sustainable economic development by the world leaders. Countries have different strategies and approaches to development to sustainable development of the tourism; There is a lot of experience in the sustainable development of tourism in the world, and in recent years’ integration of sustainable tourism on business, location, political organization, and association level has been observed. Establishing communication between the tourist locations and the industries is the greatest challenge around the world. World experience has proved that it is necessary to fundamentally change the rules of unsustainable production and consumption, for which developed countries shall ensure scientific-technological and innovative capacity building of the developing countries through the financial and technical assistance, and for this purposes the world's leading organizations are intensively implementing various levels of international legislative acts and programs and their implementation in practice. Nationalization process in each country, including Georgia has shown that, the study on tourism potential and expectations of the local population shall be carried out. The impact of tourism development on the economy, natural and social environment shall be studied thoroughly, and preventive measures shall be established in order to avoid negative impact on economy, environment, and population. The following shall be improved in Georgia: - existing database; - development strategy plan; - safety measures (accidents in mountainous resorts), engagement of qualified personnel; - safety measures (accidents in mountainous resorts), engagement of qualified personnel; The level of professional services does not meet the requirements set out in the Sustainable Development Goals; new methods are applied accordingly. The need for modern cultural and historical heritage websites that support lighter but fully functional mobile versions, facilitating and facilitating their access from portable devices is undeniable (Stefanov & Stefanova, 2012). CONCLUSION The use of free smartphone applications is relevant, providing users with information about historical and cultural attractions while traveling (Stefanov & Stefanova, 2013). Sharing international experience in the tourism sector, economic sustainability, and well-defined economic outcomes, preventive measures- identification of effective means for resolution of these problems will support transformation of sustainable tourism into the driver of economic development. Conflicts of interests The author declares no conflict of interest. The author declares no conflict of interest. Institutional Review Board statement: not applicable Institutional Review Board statement: not applicable Informed Consent Statement: not applicable Date Availability Statement: The data presented in this study are available on request from the corresponding author. The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy issues. Author Contributions: Author Contributions: Conceptualization, S.M, D.Ch.; methodology, Z.M.; formal analysis, S.M.; investigation, D.Ch. Z.M; project administration, D.Ch; data curation, S.M.; resources, Z.M..; supervision, S.M.; validation, Z.M. D.Ch; writing - original draft preparation, S.M. D.Ch.; writing - review and editing, Z.M. All authors have read and agreed to the published version of the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: not applicable 406 ACCESS Journal: Access to Science, Business, Innovation in Digital Economy ISSN 2683-1007 (Online) 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ Institutional Review Board statement: not applicable Informed Consent Statement: not applicable Date Availability Statement: The data presented in this study are available on request from the corresponding ACCESS Journal: Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) References by Pavlov R., Stanchev P., Sofia: Bulgarian Academy of Sciences, Institute of Mathematics and Informatics, 2012, II/2012, рp. 93 – 102, Print ISSN: 1314-4006, Online-ISSN: 2535-0366 Stefanov, T., M. Stefanova. (2013). Mobile Applications for Cultural and Historical Places. 3rd International Conference on Digital Presentation and Preservation of Cultural and Scientific Heritage, DiPP 2013, Veliko Tarnovo, September 2013. Ed. by Pavlov R., Stanchev P., Sofia: Bulgarian Academy of Sciences, Institute of Mathematics and Informatics, 2013, III/2013, рp. 227 – 234, Print ISSN: 1314-4006, Online-ISSN: 2535-0366 adachkoria M., Okrotsvaradze A. (2011) Tourism and Hospitality Management. Tbilisi, Georgia Vadachkoria M., Okrotsvaradze A. (2011) Tourism and Hospitality Management. Tbilisi, Georgia Zurab MUSHKUDIANI. PhD, Leading researcher, head of BBA/MBA programs and Professor at Guram Tavartkiladze Tbilisi Teaching University, Georgian National University, Batumi Navigation Teaching University and International Teaching University of Management and Communication Alterbridge. Research interests: Management, Finance and Education. Doctor of Technical Sciences, Professor at Kutaisi University, Georgia. Doctor of Technical Sciences, Professor at Kutaisi University, Georgia. Research interests: Social psychology, Ecology, Business ethic and environmental education. She has supervised numerous masters theses at Kutaisi University She has been taking participation in different international and local conferences. ORCID ID: 0000-0002-3711-8331 References Abuseridze, G., Petrova, M., Zahars, V., Tumalavicius, V. (2022). Transformation of Georgia’s trade policy strategy: from frailty to sustainability. Access to science, business, innovation in digital economy, ACCESS Press, 3(1): 43- 52. https://doi.org/10.46656/access.2022.3.1(4) Brıassoulis, H. (2002). Sustainable tourism and the question of the commons. Analysis of Tourism Research 29(4):1065- 1085 Dorobantu, M.R.& Nıstoreanu, P. (2012). Rural Tourism and Ecotourism – The Main Priorities in Sustainable Development Orientations of Rural Local Communities in Romania. Economy Transdisciplinary Cognition, 15(1): 259-266 Dorokhov, O.; Malyarets, L.; Ukrainski, K.; Petrova, M.; Yevstrat, D.; Aliyeva, A. (2023). Consumer expectations and real experiences: case of Ukrainian tourists in Turkey. Access to science, business, innovation in digital economy, ACCESS Press, 4(1), 102-114, https://doi.org/10.46656/access.2023.4.1(8) Ecotourism Strategy for Georgia 2016. Retrieved from https://www.responseandability.com/index.php/files/63/Project- results/63/2020--Ecotourism-Strategie-for-Georgia-English.pdf .(accessed: December 2022) Georgian National Tourism Administration (2023). Retrieved from: https://gnta.ge/statistics/ (accessed: January, 2023) Goeldner C. R. J. (2010). Tourism pprinciples, practices, philosophies, the United States of America Konstantinova, Y. (2022). Cultural tourism development in the town of Byala - opportunities for innovation. Access to science, business, innovation in digital economy, ACCESS Press, 3(1): 67-80. https://doi.org/10.46656/access.2022.3.1(6) Kotler B. Makens B. (2017). Marketing for Hospitality and Tourism, the UK Kotler B. Makens B. (2017). Marketing for Hospitality and Tourism, the UK Lazarova, E.; Pavlov, P.; Petrova, M.; Shalbayeva, S. (2023). Analysis and Assessment of Infrastructural Potential in Rural Territories. Economics. Ecology. Socium, 7, 1-14. https://doi.org/10.31520/2616-7107/2023.7.1-1 National Statistics Office of Georgia (2023). Retrieved from https://www.geostat.ge/en/modules/categories/100/tourism-statistics (accessed: December 2022) Odinokova, T., Bozhinova, M., Petrova, M. (2018). Promotion of Innovative Entrepreneurship Under Sustainable Development. E3S Web Conferences Volume 41, 04015 (2018). IIMS, Kemerovo, Russian Federation, 2018, https://doi.org/10.1051/e3sconf/20184104015. Petrova, M., Buzko, I., Dyachenko, Yu. (2018). Cognitive, Intelligence Technologies and Economical Foundations of Teaching of International Economical Relations and Tourism. 17th International Scientific Conference Engineering for Rural Development, 23.-25.05.2018. Jelgava, Latvia, pp. 1102-1106, https://doi.org/10.22616/erdev2018.17.n170 Stefanov, T., M. Stefanova. (2012). Mobile Access to Cultural and Historical Heritage Web Pages. 2nd International Conference on Digital Presentation and Preservation of Cultural and Scientific Heritage, DiPP 2012, Veliko 407 ISSN 2683-1007 (Online) ACCESS Journal: Access to Science, Business, Innovation in Digital Economy 2023, 4(3), 400-408, https://doi.org/10.46656/access.2023.4.3(5) https://journal.access-bg.org/ https://journal.access-bg.org/ Tarnovo, September 2012. Ed. by Pavlov R., Stanchev P., Sofia: Bulgarian Academy of Sciences, Institute of Mathematics and Informatics, 2012, II/2012, рp. 93 – 102, Print ISSN: 1314-4006, Online-ISSN: 2535-0366 Tarnovo, September 2012. Ed. Sopiko TEVODRADZE Doctor of Business Administration, Vice-rector, Professor at Guram Tavartkiladze Tbilisi Teaching and Georgian National University. Research interests: Tourism, Marketing, Management, Business and Economic. She has taken part in different scientific and educational conferences. ORCID ID: 0000-0002-1668-051X This work is licensed under the Creative Commons Attribution International License (CC BY 408
https://openalex.org/W4367692513	https://link.springer.com/content/pdf/10.1007/s11250-023-03574-8.pdf	English	null	Genomic variability and population structure of six Colombian cattle breeds	Tropical animal health and production	2,023	cc-by	6,246	Abstract Analyses of the genetic diversity of indigenous cattle are essential for implementing conservation programs, promoting their sustainable use and maintaining productive advantages offered by these breeds in local conditions. The aim of this study was to investigate the genetic diversity and population structure of six Colombian cattle breeds: Blanco Orejinegro (BON), Costeño con Cuernos (CCC), Romosinuano (ROM), Sanmartinero (SAM), Casanareño (CAS), and Hartón del Valle (HDV). Two additional breed groups were included for comparison: Zebu (CEB) and a crossbreed of Colombian cattle breeds × Zebu. Genetic diversity within breeds was analyzed using expected heterozygosity (He), inbreeding coefficient (f), and runs of homozygosity (ROH). Population structure was assessed using model-based clustering (ADMIXTURE) and principal components analysis (PCA). Zebu cattle showed the lowest genetic diversity (He = 0.240). Breeds with the highest genetic diversity level were HDV and BON (He = 0.350 and 0.340, respectively). Inbreeding was lower for Colombian cattle breeds ranging between 0.005 and 0.045. Overall, the largest average genetic distance was found among the group of Colombian cattle breeds and Zebu, while the smallest was found between ROM and CCC. Model-based clustering revealed some level of admixture among HDV and CAS cattle which is consistent with their recent history. The results of the present study provide a useful insight on the genetic structure of Colombian cattle breeds. Keywords Admixture · Beadchip · Bovine · Genetic variability · SNP Genomic variability and population structure of six Colombian cattle breeds Rodrigo Martinez1 · Diego Bejarano1 · Julián Ramírez1 · Ricardo Ocampo1 · Nelson Polanco1 · Juan Esteban Perez1 · Héctor Guillermo Onofre1 · Juan Felipe Rocha1 Received: 3 September 2022 / Accepted: 29 March 2023 / Published online: 3 May 2023 © The Author(s) 2023 Received: 3 September 2022 / Accepted: 29 March 2023 / Published online: 3 May 2023 © The Author(s) 2023 Tropical Animal Health and Production (2023) 55:185 https://doi.org/10.1007/s11250-023-03574-8 Tropical Animal Health and Production (2023) 55:185 https://doi.org/10.1007/s11250-023-03574-8 REGULAR ARTICLES * Ricardo Ocampo rocampo@agrosavia.co 1 Corporación Colombiana de Investigación Agropecuaria – Agrosavia, Mosquera, Cundinamarca, Colombia Introduction low-density panels of microsatellite markers. However, thou- sands of single nucleotide polymorphisms (SNP) evenly dis- tributed along the genome are currently available to perform genome-wide population analyses in livestock populations (Decker et al. 2014). Modern cattle descend from independently domesticated lin- eages of Taurine and Indicine cattle, which diverged more than 200,000 years ago, with numerous breeds having hybrid ancestry between these different lineages (Murray et al. 2010). Regarding Indicine cattle, Colombia has one of the biggest Zebu populations in the world. These animals have been used in crossbreeding mating systems with Colombian creole cat- tle, which were originally brought to America by the Span- iards five centuries ago (Martinez et al. 2012). Population structure analysis and breed diversity have provided insight into the origin and evolution of these cattle (Ocampo et al. 2021a, b), and the genetic structure in Colombian breeds such as Blanco Orejinegro (BON) (Martinez et al. 2013) by using SNP genotyping has become an important tool for ani- mal breeding, with different methods available to identify population stratification. These new genotyping techniques offer the possibility to perform thorough assessments of the genetic structure and the relationships between cattle populations. This improves previous population genetic parameter reports like the fixation index (FST) or inbreeding coefficients, usually estimated from just few genetic markers (Corbin et al. 2012). FST is a fixation index that evaluates genetic diversity of animal populations and allows determin- ing the relative genetic distance between these populations (Brito et al. 2017). It is estimated from allele frequencies and their variance, and it has also been used to identify loci under selection (Flori et al. 2009). Inbreeding coeffi- cients on the other hand, which have been calculated from (0123 1 Tropical Animal Health and Production (2023) 55:185 185 Page 2 of 8 185 Page 2 of 8 185 Pa (n = 61) and ROM × Zebu (n = 19) were obtained from commercial herds. pedigree records, are now estimated along with coancestry coefficients by using high-density SNP arrays. Genome-wide estimates of these parameters have higher accuracy as they represent the actual genome percentage that is homozygous (inbreeding) or the actual genome percentage shared by two individuals (coancestry), while pedigree-based estimates are only expectations of such percentages. Moreover, genome- wide estimates can include relationships due to very distant common ancestors, which are not captured by pedigree- based estimates (Cortes et al. 2019). Population This study included a total of 2182 individuals of Blanco Orejinegro (BON, n = 658), Romosinuano (ROM, n = 569), Costeño con Cuernos (CCC, n = 464), San Martinero (SAM, n = 293), Casanareño (CAS, n = 50), and Hartón del Valle (HDV, n = 34). These animals belong to the Sis- tema de Bancos de Germoplasma de la Nación para la Ali- mentación y la Agricultura (SBGNAA; System of National Germplasm Banks for Food and Agriculture of Colombia), which the Colombian Corporation for Agricultural Research (AGROSAVIA) guard and manage. A mating strategy (Nomura and Yonezawa 1996) has been implemented to con- trol inbreeding in these creole cattle populations kept in situ. It involves the splitting of breed groups into several families, with mating males being moved periodically to neighboring families in a circular pattern. Data from the purebred Zebu (CEB, n = 34) and the crossbred populations BON × Zebu Introduction Creole cattle populations were located in research centers of AGROSAVIA across the country. There was no need to apply for approval from the Animal Care and Use Com- mittee for the present study since data was extracted from existing herd books and the biological samples had already been obtained for previous projects and kept in an in vitro germplasm bank. DNA samples and genotyping These population parameters are useful to implement dif- ferent genome wide evaluation studies (Cortes et al. 2019). Genomic selection causes a larger reduction in inbreed- ing rates per generation than traditional pedigree best lin- ear unbiased prediction (BLUP) selection; however, it is not inbreeding free. Therefore, measuring and controlling inbreeding is necessary (Ocampo et al. 2019). Consequently, the understanding of the genetic structure across populations is important to achieve genetic progress by using genome- wide association studies and genomic selection strategies (Kijas et al. 2009). Furthermore, genetic diversity and pop- ulation structure studies among cattle breeds are essential for genetic improvement and for the understanding of their adaptation and the use and conservation of these animal resources (Ocampo et al. 2021a, b). Frozen blood samples from each animal were available from the in vitro bank of SBGNAA. Most of the DNA was obtained from these samples, and in some cases, DNA extraction was carried out from semen straws. Genomic DNA was obtained with a commercial extrac- tion kit (MoBio Laboratories, Inc., Carlsbad, CA, USA). DNA quantification was made by a spectrophotometric analysis at 260 nm using a NanoDrop ND-2000 spectropho- tometer (Thermo Fischer Scientific Inc., USA) and diluted to 50 ng/μL. Subsequently, DNA quality was evaluated using the same equipment to estimate the absorbance ratio at 260 and 280 nm. Genotyping was performed with the Illumina Bovine SNP50 BeadChip (Illumina, Lincoln, NE), using a HiScan array scanner in the Molecular Genetics Laboratory of AGROSAVIAand following established protocols (Matukumalli et al. 2009). The dataset only included animals with call rates > 98% and SNP with > 95%. SNPs with genotyping errors or with unknown chromosomal positions were excluded after quality control procedures were performed. Likewise, SNPs with call rates < 95%, minor allele frequency < 0.025 or significant departure from the Hardy Weinberg equilibrium within breeds (P < 0.01), were also excluded from the analysis. Finally, the analysis included 44496 autosomal SNPs with an average marker density of 56.4 kb. Considering this, the aim of this study was to investi- gate the genetic diversity and population structure within and among seven Colombian cattle breeds using the BovineSNP50 BeadChip to explore the inclusion of these cattle populations in future genomic selection programs. Genetic variability and genetic cluster analysis The expected heterozygosity (He) and inbreeding coeffi- cient (F) were calculated with the genome and het proce- dures of PLINK v.1.07 software (Purcell et al. 2007). Runs of homozygosity (ROH) were also calculated using PLINK but adjusting the parameters according to published recom- mendations (Bosse et al. 2012) (–homozyg-density 1000, –homozyg-window-het 1, –homozyg-kb 10, –homozyg- window-snp 20). Cleaning of the dataset and population structure parameters such as inbreeding, heterozygosity, and fixation indexes were determined with PLINK (Purcell et al. 2007). A principal component analysis (PCA) using multi- dimensional scaling (Purcell et al. 2007) was additionally 1 3 3 Page 3 of 8 Page 3 of 8 185 Tropical Animal Health and Production (2023) 55:185 185 carried out with the same software, and an identity by state (IBS) matrix was constructed. observed for Zebu (0.155 ± 0.170) along with the highest homozygosity (0.75 ± 0.05). ROH were low for all creole breeds, with values ranging between 0.011 ± 0.01 for CAS and 0.015 ± 0.01 for SAM. The highest ROH value was obtained for Zebu (0.023 ± 0.015); however, it was consider- ably lower than the inbreeding value estimated for this breed.f Furthermore, the ADMIXTURE software was used to implement an unsupervised clustering on the 44,496 autoso- mal SNPs in all animal datasets. A 100-fold cross-validation was performed to establish the most likely number of clus- ters (K). Then, the K value with the lowest cross-validation error was chosen as the best K value. A stacked bar plot of these results was obtained with the R© software (http:// cran.r-project.org). FST values between breeds were used to infer population structure using a neighbor-joining (NJ) clustering method, and the relationship tree was constructed using the MEGA software version 5.0 (Kumar et al. 2018). According to the PCA, there was a clear differentia- tion among groups. The first three principal components explained 14.19% of the variance (6.7%, 4.26%, and 3.23%, respectively). Moreover, creole breeds were separated into four groups, with the BON cattle displaying the highest variation (blue) and located relatively far from the other creole breeds in PC1 (Fig. 1). Furthermore, CAS (yellow) and HDV (cyan) breeds were found close to the centroid in both axes and located between SAM (red) and CCC (green) breeds. Genetic variability and genetic cluster analysis On the other hand, Zebu (grey) was found clearly detached from the creole breeds, while the group of cross- bred animals (black) included for this analysis was located between Taurine and Indicine breeds on PC2, but clearly differentiated from the pure breeds. Few individuals of CAS (yellow), BON (lilac), and crossbred groups (black) were observed near the Zebu cluster. Results A joint analysis based on all SNP datasets generated for 2181 individuals from seven different populations (BON, CCC, CAS, CEB, ROM, HDV, and SAM) was carried out to evaluate the genetic variability within and among Colombian Creole cattle breeds, and to explore possible mixture with Zebu animals. Cross validation (CV) values markedly decreased from 0.621 at K = 1 to 0.537 at K = 5 with a lower variation found in CV values from K = 6 to K = 10 (Fig. 2). The optimum number of clusters (K = 7) was identified with a CV value of 0.532. When K reached values of 4, some creole cattle groups showed differentiation and were represented by well- defined clusters for CCC, ROM, SAM, and BON. The CEB cluster showed a clear differentiation at K = 5 and similarly through K = 6 and K = 7. However, CAS and HDV indi- viduals did not show a strong pure origin, a result that was consistent across all K values (Fig. 3). The SNP average heterozygosity ranged between 0.310 for SAM and 0.350 for HDV (Table 1), and all breeds showed high heterozygosity values compared to Zebu, which had the lowest values (0.24 ± 0.04). ROH values were lower than F values for all breeds except BON. The highest mean number of ROH per individual (MNROH) and the highest sum of all ROH segments per individual (SROH) were found in ROM, while the lowest MNROH and the lowest SROH were observed in HDV. Data of observed versus expected number of homozygous genotypes was used to estimate the inbreeding coefficient. Its values ranged between 0.018 ± 0.13 for ROM and 0.056 ± 0.01 for BON, with similar values obtained across all the creole breeds. Conversely, a high inbreeding coefficient was Additionally, Fst divergences calculated between pop- ulations and the neighbor-joining phylogenetic trees are shown in Fig. 4. Phylogenetic trees were constructed for K = 4, K = 5, K = 6, and K = 7. Results In all scenarios, Zebu 1 3 Table 1 Genetic variability in six Colombian creole cattle breeds and an indicine breed a Expected heterozygosity b Inbreeding coefficient c Runs of homozygosity d Mean number of ROH per individual e Sum of all ROH segments per individual (Mb) Breed Hea Homozygosity Fb ROHc MNROH d SROH e BON 0.340 ± 0.03 0.650 ± 0.03 0.056 ± 0.01 0.013 ± 0.01 9.390 ± 4.24 117.780 ± 66.85 ROM 0.320 ± 0.04 0.670 ± 0.04 0.018 ± 0.13 0.013 ± 0.01 15.320 ± 7.03 184.280 ± 96.7 CCC 0.320 ± 0.04 0.670 ± 0.04 0.031 ± 0.12 0.013 ± 0.01 13.190 ± 5.47 163.100 ± 89.52 SM 0.310 ± 0.04 0.680 ± 0.04 0.036 ± 0.12 0.015 ± 0.01 9.670 ± 5.04 111.220 ± 73.06 CAS 0.330 ± 0.02 0.660 ± 0.02 0.045 ± 0.06 0.011 ± 0.01 6.880 ± 5.72 104.210 ± 117.50 HDV 0.350 ± 0.04 0.640 ± 0.04 0.039 ± 0.11 0.013 ± 0.01 6.820 ± 6.05 90.890 ± 87.70 Zebu 0.240 ± 0.04 0.750 ± 0.05 0.155 ± 0.17 0.023 ± 0.02 11.550 ± 6.32 119.020 ± 81.39 a Expected heterozygosity b Inbreeding coefficient c Runs of homozygosity d Mean number of ROH per individual e Sum of all ROH segments per individual (Mb) Breed Hea Homozygosity Fb ROHc MNROH d SROH e BON 0.340 ± 0.03 0.650 ± 0.03 0.056 ± 0.01 0.013 ± 0.01 9.390 ± 4.24 117.780 ± 66.85 ROM 0.320 ± 0.04 0.670 ± 0.04 0.018 ± 0.13 0.013 ± 0.01 15.320 ± 7.03 184.280 ± 96.7 CCC 0.320 ± 0.04 0.670 ± 0.04 0.031 ± 0.12 0.013 ± 0.01 13.190 ± 5.47 163.100 ± 89.52 SM 0.310 ± 0.04 0.680 ± 0.04 0.036 ± 0.12 0.015 ± 0.01 9.670 ± 5.04 111.220 ± 73.06 CAS 0.330 ± 0.02 0.660 ± 0.02 0.045 ± 0.06 0.011 ± 0.01 6.880 ± 5.72 104.210 ± 117.50 HDV 0.350 ± 0.04 0.640 ± 0.04 0.039 ± 0.11 0.013 ± 0.01 6.820 ± 6.05 90.890 ± 87.70 Zebu 0.240 ± 0.04 0.750 ± 0.05 0.155 ± 0.17 0.023 ± 0.02 11.550 ± 6.32 119.020 ± 81.39 3 Tropical Animal Health and Production (2023) 55:185 185 Page 4 of 8 185 Page 4 of 8 Fig. 1 Principal components analysis (PCA) of allele frequency vari- ation in Colombian creole cattle. derived from a node that diverges into both the branch that contains HDV and the common node for ROM and CCC. Fig. 2 Cross validation error values obtained with the admixture anal- ysis in six creole cattle breeds and Zebu Results A Principal component analysis in Colombian creole cattle compared to Zebu breed. B Principal compo- nent analysis for several Colombian cattle breeds. BON, Blanco ore- jinegro (lilac); CCC, Costeño con cuernos (green); ROMO, Romo- sinuano (blue); SAM, Sanmartinero (red); CAS, Casanareño (yellow); HDV, Hartón del Valle (cyan), Zebu (grey), crossbred animals (black) Fig. 1 Principal components analysis (PCA) of allele frequency vari- ation in Colombian creole cattle. A Principal component analysis in Colombian creole cattle compared to Zebu breed. B Principal compo- nent analysis for several Colombian cattle breeds. BON, Blanco ore- jinegro (lilac); CCC, Costeño con cuernos (green); ROMO, Romo- sinuano (blue); SAM, Sanmartinero (red); CAS, Casanareño (yellow); HDV, Hartón del Valle (cyan), Zebu (grey), crossbred animals (black) derived from a node that diverges into both the branch that contains HDV and the common node for ROM and CCC. Discussion A whole dataset of SNP genotypes for six different Colom- bian creole cattle breeds was obtained to assess levels of inbreeding and genetic variability in these populations. Estimated heterozygosity was high for all creole breeds (> 0.31), while the lowest value was observed in Zebu. These genetic diversity levels are higher than those reported by Makina et al. (2016) in indigenous African cattle, with levels of expected heterozygosity that ranged from 0.24 in Afri- kaner cattle to 0.30 in the Drakensberg breed. Other breeds included in that study such as Angus and Holstein had the highest levels of gene diversity (He = 0.31). Likewise, McTavish et al. (2013) reported an average heterozygosity of 0.27 in Taurine breeds, while breeds of Indicine origin had an average heterozygosity of 0.16. In that study, the ROM Taurine breed had an average heterozygosity of 0.27. Fig. 2 Cross validation error values obtained with the admixture anal- ysis in six creole cattle breeds and Zebu and Colombian creole cattle breeds were in two different branches. The phylogenetic tree obtained with K = 6 and K = 7 shows a common node for ROM and CCC, and a common node for CAS and SAM. When K = 7, BON is 1 3 Tropical Animal Health and Production (2023) 55:185 Page 5 of 8 185 Page 5 of 8 185 Fig. 3 Admixture clustering using six creole cattle breeds and Zebu from K = 2 to K = 7 Fig. 3 Admixture clustering using six creole cattle breeds and Zebu from K = 2 to K = 7 heterozygosity erosion. On the other hand, and as discussed by Ben Jemaa et al. (2018), the bias contained in the Bovine SNP50 BeadChip caused by the inclusion of SNP mostly derived from sequences obtained from European cattle breeds might lead to a biased genetic divergence estimation between more distantly related populations, as the one found between Colombian creole breeds and Zebu cattle. Neverthe- less, Gautier et al. (2010) found that hybrid populations had higher polymorphism levels than their population of origin, including hybrids resulting from crosses between European Taurines and Zebu cattle. Thus, the BovineSNP50 would adequately describe the genetic variability of the Colombian creole breeds, which have a Taurine origin, and this would possibly not underestimate polymorphism levels of crosses between these cattle and breeds of Indicine origin. Discussion Furthermore, studies of genomic divergence using high- density BeadChips with more than 750,000 SNP (Porto- Neto et al. 2013) also confirmed this trend, with average heterozygosity values of 0.21 and 0.29 for Zebu and Taurine cattle, respectively. This lower genetic diversity observed in Zebu might be due to the effect of SNP ascertainment bias. As many other genotyping arrays, the Illumina Bovine SNP50 BeadChip contains biased sets of pre-ascertained SNP, causing an under representation of fixed genetic dif- ferences between taurine and indicine cattle, or leading to passing over sites that are polymorphic only in indicine cat- tle (McTavish et al. 2015), which in turn would affect the calculation of heterozygosity in these breeds. A possible explanation of the higher genetic diversity of the Taurine breeds analyzed in this study could be the lower levels of selection pressure applied to the Colombian creole cattle, preventing these populations from undergoing Inbreeding levels estimated for Colombian creole cat- tle were low (0.056 ± 0.01 to 0.018 ± 0.). Inbreeding 1 1 3 3 3 Tropical Animal Health and Production (2023) 55:185 185 Page 6 of 8 Fig. 4 FST distances between six Colombian creole cattle and Zebu breeds. BON, Blanco orejinegro; CCC, Costeño con cuernos; ROMO, Romosinuano; SAM, Sanmartinero; CAS, Casanareño; HDV, Hartón del Valle; CEB, Zebu et al. 2012). Conversely, McTavish and Hillis (2015) found much higher values for molecular homozygosity and simi- larity coefficients compared with inbreeding and coancestry coefficients. Likewise, a study in Lidia cattle breed (Cortés et al. 2019) showed a significantly higher mean ROH-based inbreeding coefficients (FROH) for 50 k SNP chip data (≈ 0.21–0.26) than the pedigree inbreeding coefficient (FPED) (0.13); this might be explained by the fact that pedigree captures only relatively recent inbreeding since pedigree recording started. Different factors such as the minimum length of ROH, minimum number of SNP involved, or even the density of the beadchip used for genotyping make dif- ficult the comparison of ROH and its statistics (MNROH and SROH) with other studies. Further analysis of ROH along with other statistics for each class of ROH length, the relationship between these statistics, and the coverage and distribution of ROH across chromosomes, is beyond the scope of this paper; however, they must be studied in depth for each Colombian breed to have a better understanding of the inbreeding his- tory of these populations. Fig. Discussion Conversely, high inbreeding levels were observed in Zebu (0.155 ± 0.170), which also showed the highest homozygo- sity value (0.75 ± 0.05). A high inbreeding in Zebu might be caused by its high historic selection pressure for productive traits and by a low effective population size. On the other hand, inbreeding levels can change drastically according to the history of the population analyzed. For instance, the study published by Makina et al. (2016) showed very low levels of inbreeding for Afrikaner (0.004) and for Drakens- berger (-0.002) cattle; however, allele frequencies may not be a good inbreeding estimate, although a change over time of this parameter in a particular population might be more useful. Hence, inbreeding levels should be monitored with a regular frequency in Colombian creole breeds, even when they show low inbreeding levels. coefficients for these breeds had been calculated using pedi- gree information (Ocampo et al. 2021a, b; 2020). Further- more, Martinez et al. (2012) found values between 0.018 for BON and 0.122 for ROM. The lower inbreeding lev- els found in ROM might be attributed to the low selection pressure these animals have undergone under the conserva- tion program they have been kept for more than 20 years, which includes a circular mating system to control rates of inbreeding (Martinez et al. 2012). Additionally, the effective population size per generation for this breed (Martinez et al. 2012) is consistent with a desired effective population size for conservation of small populations (Meuwissen 2009). Conversely, high inbreeding levels were observed in Zebu (0.155 ± 0.170), which also showed the highest homozygo- sity value (0.75 ± 0.05). A high inbreeding in Zebu might be caused by its high historic selection pressure for productive traits and by a low effective population size. On the other hand, inbreeding levels can change drastically according to the history of the population analyzed. For instance, the study published by Makina et al. (2016) showed very low levels of inbreeding for Afrikaner (0.004) and for Drakens- berger (-0.002) cattle; however, allele frequencies may not be a good inbreeding estimate, although a change over time of this parameter in a particular population might be more useful. Hence, inbreeding levels should be monitored with a regular frequency in Colombian creole breeds, even when they show low inbreeding levels. Genetic structure patterns were inferred through admix- ture analysis ignoring breed membership. Discussion 4 FST distances between six Colombian creole cattle and Zebu breeds. BON, Blanco orejinegro; CCC, Costeño con cuernos; ROMO, Romosinuano; SAM, Sanmartinero; CAS, Casanareño; HDV, Hartón del Valle; CEB, Zebu Clustering animals based on the genetic PCA analysis showed a clear differentiation between Taurine and Indicine cattle, but also among Colombian creole breeds. Results demonstrated that most of the individuals analyzed are purebred, as the estimated breed proportion assigned to a single animal was greater than 0.9 for a particular breed. However, HDV and CAS breeds showed considerable levels of admixture. Also, Porto-Neto et al. (2013) found well- defined clusters of Taurine (Angus, Charolais, and Holstein) and Indicine (Gir, Guzera, and Nelore) cattle, which was correlated with a lower pair-wise FST observed for Zebu breeds in comparison to Taurine breeds. The proportion of variance explained by each component in this study (PC1 = 6.8 and PC2 = 4.3) is similar to that obtained by Rodriguez-Ramilo et al. (2015) (PC1 = 6.1 and PC2 = 3.7), with the first component separating the Zebu Brahman from the Taurine breeds and the second component subdividing in turn the Taurine breeds as in the PCA carried out by Porto-Neto et al. (2013) and the one performed in our study. Likewise, the clusters partly overlapped those of other breeds located nearby, which was particularly true for the CAS and HDV breeds. On the other hand, the crossbred individuals fell between the Taurine and Indicine cattle clusters on the PC2 axis. Similarly, the crossbred cattle included in the study of Zhang et al. (2015) was also placed in the middle of the axis between the parental breeds, i.e., indigenous African cattle and European Bos Taurus breeds. coefficients for these breeds had been calculated using pedi- gree information (Ocampo et al. 2021a, b; 2020). Further- more, Martinez et al. (2012) found values between 0.018 for BON and 0.122 for ROM. The lower inbreeding lev- els found in ROM might be attributed to the low selection pressure these animals have undergone under the conserva- tion program they have been kept for more than 20 years, which includes a circular mating system to control rates of inbreeding (Martinez et al. 2012). Additionally, the effective population size per generation for this breed (Martinez et al. 2012) is consistent with a desired effective population size for conservation of small populations (Meuwissen 2009). 1 3 Conclusions The existence of six distinct Colombian Creole cattle popu- lations was demonstrated with the use of a genomic analy- sis. These populations have high genetic diversity and low inbreeding levels compared with Zebu. However, a regular evaluation of these parameters is recommended as the popu- lation structures change over time. Finally, all Colombian creole cattle breeds must be essential and viable objects for conservation, since they are genetically unique and have particular characteristics that include its high capacity to adapt to the difficult conditions of the Colombian tropics, surviving under conditions of low food and water availabil- ity, and still showing high fertility and resistance to several pathogens. Conflict of interest The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Discussion Results revealed a well-defined structure for animals from the same breed cluster, while introgression signals were only found in two creole cattle breeds, showing that some animals did not have a pure origin while four creole cattle were Taurine, probably Moreover, ROH values were lower than the inbreeding coefficients calculated in Colombian creole cattle using genomic information, and also lower than inbreeding levels previously estimated from pedigree information (Martinez 1 3 1 3 Tropical Animal Health and Production (2023) 55:185 Page 7 of 8 185 collaborated with previous sampling sessions and gathering genea- logical information. collaborated with previous sampling sessions and gathering genea- logical information. with two different clusters and related with their historical origin. Geographically, the first cluster was related to the north coast of Colombia (CCC and ROM) and the second cluster to herds brought from the oriental savannas (SAM) or from the southern part of the country (BON). Likewise, Ocampo et al. (2021a, b) also found similar results using microsatellite markers in Creole cattle to estimate genetic distances.i Author contribution All THE authors contributed to the conception and design of the study. Material preparation, data collection, and anal- ysis were performed by all authors. All the authors read and approved the final manuscript. Funding Open Access funding provided by Colombia Consortium This work was supported by the research project “Six Creole cattle breeds characterized at the molecular level using information from 24 microsatellite markers and new SNP-type variants” through Sistema de Bancos de Germoplasma de la Nación para la Alimentación y la Agricultura (SBGNAA) using public funds. As expected, the fixation index FST was higher between the Colombian creole cattle and Zebu compared to the value obtained among creole breeds (11–18%), indicating a lower genetic structure differentiation in the latter. Similarly, Campos et al. (2017) observed a FST of 12% among locally adapted taurine Brazilian breeds, while much lower values were reported in Zebu breeds (5.33%). A similar genetic differentiation was observed by Weerasinghe et al. (2013) between African and European breeds (15%). Data availability The datasets generated or analyzed during the cur- rent study, or both, are available from the corresponding author upon reasonable request. Declarations In addition, New World cattle such as Texas Longhorns, Corriente, and ROM breeds have showed evidence of hav- ing an admixed ancestry between African hybrid cattle and European cattle. However, the Indicine genomic component found in those breeds might be due to a recent introgres- sion with Indicine cattle in the New World, rather than to an ancient admixture (Edea et al. 2013). In our study, there was no evidence of any admixed ancestry with Indicine breeds for most Colombian breeds. However, HDV and CAS had considerable levels of admixture, including admixed ances- try with Indicine breeds. This might be explained by their particular breeding histories. Unlike HDV and CAS, breeds such as ROM, BON, CCC, and SAM have been subject to breeding programs that aim to maintain their genetic diver- sity over time. Ethics approval This work had the endorsement of the ethics com- mittee of Corporación Colombiana de Investigación Agropecuaria AGROSAVIA. Collection of blood samples was carried out by veteri- narians adhering to the regulations and guidelines on animal husbandry and welfare. Consent for publication All the authors consent to publish the manu- script. References Ben Jemaa, S.L., Rahal, O., Gaouar, S.B.S., Mastrangelo, S., Boussaha, M. and Ciani, E., 2018. Genomic characterization of Algerian Guelmoise cattle and their genetic relationship with other North African populations inferred from SNP genotyping arrays Live- stock Science, 217, 19-25 Bosse, M., Megens, H.J., Madsen, O., Paudel, Y., Frantz, L.A. and Schook L.B., 2012. Regions of homozygosity in the porcine genome: consequence of demography and the recombination landscape PLoS Genet, 8(11), e1003100 Brito, L.F., Kijas, J.W., Ventura, R.V., Sargolzaei, M., Porto-Neto, L. R., Cánovas, A., Feng, Z., Jafarikia, M. and Schenkel, F. S., 2017. Genetic diversity and signatures of selection in various goat breeds revealed by genome-wide SNP markers BMC Genomics, 18(1), 229 Acknowledgements We thank our colleagues from the different research centers of the Corporación Colombiana de Investigación Agripecuaria (AGROSAVIA) who provided support for maintain- ing the in vivo germplasm bank of Colombian cattle breeds and Campos, B. M., do Carmo, A. S., do Egito, A. A., da Mariante, A. S., do Albuquerque, M., de Gouveia, J., Malhado, C., Verardo, L. L., 1 3 3 185 Page 8 of 8 Tropical Animal Health and Production (2023) 55:185 185 independent domestication events Proc Natl Acad Sci U S A, 110(15), 1398-1406 da Silva, M. and Carneiro, P., 2017. Genetic diversity, population structure, and correlations between locally adapted zebu and tau- rine breeds in Brazil using SNP markers Tropical Animal Health and Production, 49(8), 1677–1680 da Silva, M. and Carneiro, P., 2017. Genetic diversity, population structure, and correlations between locally adapted zebu and tau- rine breeds in Brazil using SNP markers Tropical Animal Health and Production, 49(8), 1677–1680 McTavish, E.J. and Hillis, D.M., 2015. How do SNP ascertainment schemes and population demographics affect inferences about population history? BMC Genomics, 16(1), 266 Corbin, L. J., Liu A., Bishop S. and Woolliams, J., 2012. Estimation of historical effective population size using linkage disequilib- ria with marker data Journal of Animal Breeding and Genetics, 129(4), 257–270 p p y Meuwissen, T.H.E., 2009. Accuracy of breeding values of “unrelated” individuals predicted by dense SNP genotyping Genet Sel Evol, 41(1), 35 Murray, C., Huerta-Sanchez, E., Casey, F and Bradley, D.G., 2010. Cattle demographic history modelled from autosomal sequence variation. Philos Trans R Soc Lond B Biol Sci, 365, 2531–2539 Cortés, O., Eusebi, P., Dunner, S., Sevane, N., and Cañón, J., 2019. References Comparison of diversity parameters from SNP, microsatellites and pedigree records in the Lidia cattle breed Livestock Science, 219, 80-85 Nomura, T. and Yonezawa, K., 1996. A comparison of four systems of group mating for avoiding inbreeding Genet Sel Evol, 28(2), 141-159 Decker, J.E., McKay, S.D., Rolf, M.M., Kim, J.W., Molina Alcalá, A., Sonstegard, T.S., Hanotte, O., Götherström, A., Seabury, C.M., Praharani, L., Babar, M.E., Correia de Almeida Regitano, L., Yildiz, M.A., Heaton, M.P., Liu, W.S., Lei, C.Z., Reecy, J.M., Saif-Ur-Rehman, M., Schnabel, R.D. and Taylor, J.F., 2014. Worldwide Patterns of Ancestry, Divergence, and Admixture in Domesticated Cattle PLoS Genetics, 10 Ocampo, R.J., Tobón, J.A., Martínez, P.Y., Ramirez, E.J. and Lucero, C.E., 2019. Análisis de diversidad genética en cerdo criollo san Pedreño utilizando datos de pedigrí Ecosistemas y Recursos Agro- pecuarios, 6(17), 333-341 Ocampo, R.J., Ramirez, J., Lopera, J., Restrepo, G. and Gallego, J.L., 2020. Genetic diversity assessed by pedigree analysis in the blanco orejinegro (Bon) cattle breed population from the colom- bian germplasm bank Chilean Journal of Agricultural & Animal Sciences, 36(1), 69-77 Edea, Z., Dadi, H., Kim, S.W., Dessie, T., Lee, T., Kim, H., Kim, J.J. and Kim, K.S., 2013. Genetic diversity, population structure and relationships in indigenous cattle populations of Ethiopia and Korean Hanwoo breeds using SNP markers Frontiers in genetics, 4, 35f Ocampo, R.J., Bejarano, D.H. and Onofre, H.G., 2021a. Estructura pobla- cional del ganado criollo colombiano Sanmartinero por análisis de pedigrí Livestock Research for Rural Development, 33(10), 1-12 Flori, L., Fritz, S., Jaffrézic, F., Boussaha, M., Gut, I., Heath, S., Foul- ley, J.L. and Gautier, M., 2009. The genome response to artificial selection: a case study in dairy cattle PLoS ONE, 4(8), e6595 Ocampo, R.J., Rocha, J.F. and Martínez, R.A., 2021b. Assessment of genetic diversity and population structure of Colombian Creole cattle using microsatellites Tropical Animal Health and Produc- tion, 53, 122 Gautier, M., Laloë, D. and Moazami-Goudarzi, K., 2010. Insights into the genetic history of French cattle from dense SNP data on 47 worldwide breeds PLoS ONE, 5(9), e13038 Porto-Neto, L.R., Sonstegard, T.S., Liu, G.E., Bickhart, D.M., Vb, M., Silva, D., Machado, M.A., Utsunomiya, Y.T., Garcia, J.F., Gondro, C. and Van Tassell, C.P., 2013. Genomic divergence of zebu and taurine cattle identified through high-density SNP geno- typing BMC Genomics 14. Kijas, J.W., Townley, D., Dalrymple, B.P., Heaton, M.P., Maddox, J.F., McGrath, A., and Wilson, P., 2009. References A genome wide survey of SNP variation reveals the genetic structure of sheep breeds PLoS One, 4(3), e4668 Kumar, S., Stecher, G., Li, M., Knyaz, C. and Tamura, K., 2018. MEGA X: Molecular evolutionary genetics analysis across com- puting platforms Molecular Biology and Evolution, 35 Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M.A., Bender, D., Maller, J., Sklar, P., De Bakker, P.I., Daly, M.J. and Sham, P.C., 2007. PLINK: a tool set for whole-genome associa- tion and population-based linkage analyses American Journal of Human Genetics, 81, 559–575 Makina, S.O., Whitacre, L.K., Decker, J.E., Taylor, J.F., Macneil, M.D., Scholtz, M.M., Van Marle-Köster, E., Muchadeyi, F.C., Makgahl- ela, M.L. and Maiwashe, A., 2016. Insight into the genetic com- position of South African Sanga cattle using SNP data from cattle breeds worldwide Genetics Selection Evolution, 48, 88 Rodríguez-Ramilo, S.T., Fernández, J., Toro, M.A., Hernández, D., Villanueva, B. and Lund, M., 2015. Genome-Wide Estimates of Coancestry, Inbreeding and Effective Population Size in the Span- ish Holstein Population PLoS One, 10(4), e0124157. Martínez, R., Vásquez, R., Gallego, J., Gómez, Y., Moreno, F., Fernán- dez, J., Tobon, J., Neira, J., Córdoba, S., Maldonado, J., Trujillo, L., Pedraza, A., Rocha, J., García, D., Onofre, G., Pérez, J., Cañón, J., Lucero, C., Lopera, J. and Quiceno, J., 2012. Eficiencia pro- ductiva de la raza BON en el trópico Colombiano In:, Eficiencia productiva de la raza BON en el trópico Colombiano, (Corpoica: Bogotá), 7–16 Weerasinghe, W.M.S.P., Gondro, C., Jeyaruban, M.G., Mwai, O., Mujibi, D.F. and Gibson, J.P., 2013. The prediction of genetic structure of East African smallholder dairy cattle Proc Assoc Advmt Anim Breed Genet, 163, 166 Zhang, Q., Calus, M.P.L., Guldbrandtsen, B., Lund, M.S. and Sahana, G., 2015. Estimation of inbreeding using pedigree, 50k SNP chip genotypes and full sequence data in three cattle breeds BMC Genet, 16, 88 Martínez, R.A., Llinás, A.P. and Rocha, J.F., 2013 Genetic variability in Blanco Orejinegro breed cattle populations in Colombia Genet Mol Res, 12(2), 1083–1094 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Matukumalli, L.K., Lawley, C.T., Schnabel, R.D., Taylor, J.F., Allan, M.F., Heaton, M.P. and O'Connell J., 2009. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Development and characterization of a high density SNP genotyping assay for cat- tle PLoS One, 4(4), e5350 McTavish, E.J., Decker, J.E., Schnabel, R.D., Taylor, J.F. and Hillis, D.M., 2013. New World cattle show ancestry from multiple 1 3
https://openalex.org/W4205473889	https://www.scielo.br/j/aabc/a/mL3kMqSjGhQhSTgV3YcGtgy/?lang=en&format=pdf	English	null	Gender-related differences in the modulation of anthropometric, biochemical, and immune markers by physical activity in hypertensive and diabetic individuals	Anais da Academia Brasileira de Ciências	2,021	cc-by	7,162	DÉBORAH C. DOS SANTOS, ISRAEL S. RIBEIRO, DIEGO PATRICK S. LOPES, ÍTALO S. PEREIRA, DENISAR P. SANTOS, IGOR P.R. MUNIZ, FERNANDA M. DA SILVA LIMA, ANDRESSA ANTÔNIA S. COSTA, DANIELA DA SILVA P. SOUZA, STÉFANO PEDRO DE MELO CALADO, CAROLINE V. GONÇALVES, MARIA POLIANA L. GALANTINI, DANILO N. LOPES, ROBSON A.A. DA SILVA & GILVANÉIA S. SANTOS Abstract: Systemic arterial hypertension (SAH) and type 2 diabetes mellitus (T2DM) compose the two major noncommunicable chronic infl ammatory diseases. Physical activity has been shown as a promising complementary approach to control the systemic infl ammation. However, it is still unclear whether this modulation is gender- dependent. The objective of this study was evaluate the gender-related infl uence of physical activity on the infl ammatory response and biochemical profi le of individuals with SAH and T2DM. An international physical activity questionnaire was applied to 376 individuals diagnosed with SAH and T2DM in order to access their exercises routine and was evaluated the infl uence of physical activity in biochemical, anthropometrical, and immunological markers involved in these disorders in men and women. Even though active individuals have exhibited lower serum levels of IL-1β, IFN-γ, TNF-α, and IL-17A, the ratios between IL-10 and all infl ammatory cytokines were higher in men than in women. Physically active individuals also demonstrated increased HDL/LDL and HDL/ VLDL ratios. Moreover, multiple correlations revealed that in active women both IL-10 and TNF-α serum levels positively correlate with fasting glucose levels, and were negatively associated with HDL levels. Our fi ndings suggest that gender-related differences dictate a distinct crosstalk between infl ammatory and biochemical markers in physically active individuals. Key words: Systemic arterial hypertension, gender, inflammation, physical activity, type 2 diabetes mellitus. An Acad Bras Cienc (2021) 93(Suppl. 4): e20201905 DOI 10.1590/0001-3765202120201905 Anais da Academia Brasileira de Ciências \| Annals of the Brazilian Academy of Sciences Printed ISSN 0001-3765 I Online ISSN 1678-2690 www.scielo.br/aabc \| www.fb.com/aabcjournal An Acad Bras Cienc (2021) 93(Suppl. 4): e20201905 DOI 10.1590/0001-3765202120201905 Anais da Academia Brasileira de Ciências \| Annals of the Brazilian Academy of Sciences Printed ISSN 0001-3765 I Online ISSN 1678-2690 www.scielo.br/aabc \| www.fb.com/aabcjournal Gender-related differences in the modulation of anthropometric, biochemical, and immune markers by physical activity in hypertensive and diabetic individuals DÉBORAH C. DOS SANTOS, ISRAEL S. RIBEIRO, DIEGO PATRICK S. LOPES, ÍTALO S. PEREIRA, DENISAR P. SANTOS, IGOR P.R. MUNIZ, FERNANDA M. DA SILVA LIMA, ANDRESSA ANTÔNIA S. COSTA, DANIELA DA SILVA P. SOUZA, STÉFANO PEDRO DE MELO CALADO, CAROLINE V. GONÇALVES, MARIA POLIANA L. GALANTINI, DANILO N. LOPES, ROBSON A.A. DA SILVA & GILVANÉIA S. SANTOS INTRODUCTION and diabetes mellitus are the most common pathologies, causing a lower quality of life to these individuals, besides higher assistance costs, and lower life expectancy (Marengoni et al. 2008, Prince et al. 2015). Studies have also been demonstrating an augment in prevalence of chronic noncommunicable diseases in young adults, raising the need to include them in population studies (Chen et al. 2011, Kissela Over the years, chronic noncommunicable diseases have been assuming an important role in health-related concerns, and nowadays, they represent themajor affecting disorders for the world population (Prince et al. 2015). It is known that, in elderlies, cardiovascular diseases, systemic arterial hypertension, cancer, An Acad Bras Cienc (2021) 93(Suppl. 4) GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. et al. 2012). Overall, they present more severe prognoses as they grow older (Smajlović 2015). testosterone and that were supplemented with estradiol at levels similar to those of women in the luteal phase had completely different inflammatory responses (Timmons et al. 2006). This fact demonstrates the importance of studies that consider different parameters and multifactorial analyzes, in order to obtain a real understanding of gender differences and how they influence the applied therapeutic platforms. Therefore, in this study, we sought to evaluate the gender-related influence of physical activity on the inflammatory response and biochemical profile of individuals with SAH and T2DM. Systemic arterial hypertension (SAH) stands out as a chronic multifactorial disease, being one of the main causes of cardiovascular diseases and stroke (Nag & Ghosh 2013). It affects about 25% of the world’s adult population and it is estimated that by 2025 this picture will rise up to 29% (Mittal & Singh 2010). SAH is usually associated with other chronic conditions such as type 2 diabetes mellitus (T2DM), a metabolic disorder characterized by the inability of insulin in performing its functions (Cheung & Li 2012). Both conditions have their pathogenesis related to chronic low-grade inflammation, defined by the release of pro-inflammatory cytokines such as TNF-α and low levels of regulatory cytokines such as IL-10 (Aroor et al. 2012, Zhong et al. 2014). Researches have reported that, besides immunological factor, body composition can infer the health-disease state of individuals (Sagun et al. 2014, Ribeiro et al. 2019). Ethical considerations This project was approved by the Human Research Ethics Committee of Federal University of Bahia under the protocol number 27439114.4.0000.5556. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 2 \| 13 Study population The practice of physical activity has been proposed as an important therapeutic approach for diabetic and hypertensive individuals (Hu et al. 2004, Trachta et al. 2014). It is already known that exercise induces HDLc production, an important protective molecule in this condition (Zhou et al. 2014). Furthermore, works have revealed that people who exercise have beneficial anti-inflammatory modulation, even in overweight and obese individuals (Lopes et al. 2016). However, men and women have physiological differences that are crucial in inflammatory modulation between different genders, which is an important variable that must be considered when carrying out studies in the biomedical and health-related areas (Wizemann & Pardue 2001). Such differences are not limited to hormonal aspects. This is clear in studies in which men with reduced levels of This study comprised the evaluation of 376 adult individuals of both genders, aging between 30 to 80 yearsold, that had been diagnosed with SAH and T2DM and assisted by Family Health Units (USF) of Vitória da Conquista, Bahia, Brazil. Data collection was performed in the Health Units, participation was voluntary, and the sample size was defined by convenience. The assessment initially consisted of an explanation of research subject and scope, and afterwards, those who agree in participating had to signed the Informed Consent Form and/or performed finger print identification. Later, the semi-structured questionnaire on socioeconomic data and physical activity practice was applied and anthropometric evaluation, analysis of the estimation of body composition by bioimpedance, blood pressure measurement, An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 2 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. and blood collection for biochemical and cytokines measurements were carried out. For bioimpedance data to be as reliable as possible, some recommendations were suggested to participants and they weredo not consume alcoholic beverages and/or containing caffeine, fast for four hours, do not perform physical activity the day before the test, drink water two hours before the examination and urinate at least thirty minutes before performing the bioimpedance device. Arterial pressure assessment Blood pressure was measured through an automatic digital pressure device (G-TECH brand, model BP3AA1-1), approved by the Brazilian Society of Cardiology, with individuals sitting comfortably for at least 5 minutes (feet flat on the floor), in a silent environment with the right arm positioned at the level of the heart. For purpose of data analysis, mean values obtained through two blood pressure measurements, carried out at intervals of approximately 10 minutes, were used. Dosages of biochemical markers Previously aware of an 8-hour fasting, five milliliters of peripheral blood were collected in dry vacuum tubes from the volunteers. The tubes were centrifuged at 300g for 10 minutes and the plasma stored in Eppendorf tubes in freezer at -70ºC until the time of the experiments for biochemical and immunological analyzes. Criteria for inclusion or exclusion The inclusion criterionminimum age of 30 and maximum of 80 years oldwere applied to the volunteers, who had not physical activityimpairment or musculoskeletal injuries, and who had been exercising for at least three months. This information was acquired through a sociodemographic questionnaire, self-reported by the participants. The exclusion criteria were to have been diagnosed of autoimmune diseases and/or use of anti-inflammatory drugs. Anthropometric evaluation Anthropometric measures of weight and height, body mass index (BMI), waist circumference (WC), and WC/height ratio were collected to characterize participants’ nutritional status. Weight determination was accomplished using a digital weight balance coupled to bioimpedance (Plenna, SP, Brazil) previously calibrated in the Nutritional Assessment Laboratory of Federal University of Bahia. Bioimpedance measurements were performed providing the fat mass, lean mass, bone mass, body water, and basal metabolism data. Kits from the manufacturer Bioclin, Minas Gerais, Brazil and manufacturer’s specifications for the kits monoreagent cholesterol (K083), enzymatic HDL (K015), monoreagent triglycerides (K117) and monoreagent glucose (K088) were strictly followed. Sample classification For this study, the cutoff point used for patients practicing physical activity (PA) was thereference values indicated by the IPAQ (Matsudo et al. 2001). Active patients were classified as practicing physical activity in a frequency greater than or equal to one hundred and fifty minutes per week (≥150 minutes/week) taking into account recreational, sports, physical exercise, and leisure activities. Sedentary patients were classified as people who did not exercise the volume of weekly physical activities to be classified as active. Statistical analysis In this research, physically active men presented lower waist circumference/height ratio (WC/ Height), lower BMI, and more lean mass than sedentary men, and active and sedentary women (Table I). Remarkably, sedentary men presented less fat mass than physically active and sedentary womenand more lean mass than sedentary women. Data normal distribution was verified by Shapiro-Wilk test. When the variables did not present a normal distribution, the comparisons between the active and sedentary groups were performed by the non-parametric Mann Whitney test. For data that presented normal distribution, Student’s t-test (parametric test) was then used. To study the associations among the parameters in this study a correlation matrix was plotted applying the R-package corrplot 0.84 based on Spearman rank correlations with ordering of variables following first principal component analysis (Wei & Simko 2013). Interestingly, we were able to detect adifference between the mean systolic blood pressureof sedentary and active women. Such difference was not observed between active and sedentary men (Table I). Characterization of the most used drugs by the study population Among the most common medicines used by the study population, oral antihypertensives, such as hydrochlorothiazide (54%) and losartan potassium (50%) were the major representatives. The main oral hypoglycemic drugs were metformin (67%) and glibenclamide (27%). The statin class drug simvastatin was prescribed for 37% of the study sample. There was no statistical difference between men and women groups concerning common medications. Quantification of cytokines Cytokines concentration (IL-1β, TNF-α, IL-10, IL-17A and IFN-γ) were quantified by ELISA (Enzyme- Linked Immunosorbent Assay) according to An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 3 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. Quantikine® manufacturers’ recommendations of the respective kits (R&D Systems). Physically active men and women presented a distinct profile of biochemical markers than sedentary men and women Statistical analysis was done using GraphPad- Prisma Version 6.0 software (GraphPad Software, CA, USA). Numerical variables were expressed as mean ± standard deviation and the differences considered statistically significant were applied when the value of p <0.05, with a confidence interval of 95%. We were able to detect a higher concentration of HDL alongside with lower values for LDL, VLDL, glucose, triglycerides, and total cholesterol (Table I) in the serum of individuals (men and women) who practiced physical activity. Furthermore, through the HDL cholesterol and LDL or VLDL fractions ratios, it was observed that individuals engaged in physical activity exhibited higher ratio values than the sedentary ones. Active men with diabetes and hypertension presented less inflammation than active women Individuals serum inflammatory profile was assessed. The inflammation markers profile acquired here has shown to be in consonance with the biochemical data, as the inflammatory cytokines serum concentration (represented by IL-1β, TNF-α, IL-17 and IFN-γ) were lower, together with a higher level of IL-10 in individuals An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 4 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. Table I. Characterization of anthropometric, blood pressure data, biochemical and cytokine markers in diabetics and hypertensive individuals. Parameters Physically active men (n=40) Sedentary men (n= 54) Physically active women (n=98) Sedentary women (n=184) WC/Height 0.5582 0.6322 0.6181 0.6761 SD: 0.058 § ‡ ¶ SD:0.062 † SD:0.069 † SD:0.449 † BMI (kg/m²) 25.66 29.35 27.95 30.89 SD: 3.545 § ‡ ¶ SD: 4.597 † SD: 3.673 † SD: 6.138 † Leanmass (%) 27.89 12.8 14.58 11.56 SD: 6.538 § ‡ ¶ SD: 1.874 †¶ SD: 5.248 † SD: 2.727 †§ Fat mass (%) 13.02 29.11 31.41 34.19 SD: 7.413 § ‡ ¶ SD: 3.966 †‡ ¶ SD: 5.307 †§ SD: 4.937 †§ MSP (mmHg) 143.3 153.3 140.7 154.9 SD: 12.73 ¶ SD: 22.92 ‡ SD: 10.94 §¶ SD: 18.39 †‡ Cholesterol (mg/dL) 192.9 233 195.9 233.3 SD:10.13 §¶ SD: 36.16 †‡ SD: 12.65 §¶ SD: 40.76 †‡ Triglycerides (mg/ dL) 123.9 196.2 131.1 196.2 SD: 20.92 §¶ SD: 35.3 †‡ SD: 21.47 §¶ SD: 34.86 †‡ Glucose (mg/dL) 125.1 195.1 123.2 191.9 SD: 20.01 §¶ SD: 29.65 †‡ SD: 17.08 §¶ SD: 31.23 †‡ HDL (mg/dL) 59.49 37.63 60 37.63 SD: 9.65 §¶ SD: 10.96 †‡ SD: 8.743 §¶ SD: 6.651 †‡ LDL (mg/dL) 110 156 109.4 153.8 SD:8.862 §¶ SD: 32.01 †‡ SD: 14.84 §¶ SD: 44.4 †‡ VLDL (mg/dL) 24.79 39.24 26.19 39.24 SD: 4.184 §¶ SD: 7.06 †‡ SD: 4.311 §¶ SD: 6.972 †‡ HDL/LDL ratio 0.5498 0.2533 0.5462 0.2555 SD:0.103 §¶ SD:0.098 †‡ SD:0.087 §¶ SD:0.0769 †‡ HDL/VLDL ratio 2.470 0.9726 2.364 0.9862 SD:0.588 §¶ SD:0.299 †‡ SD:0.577 §¶ SD:0.241 †‡ tion of anthropometric, blood pressure data, biochemical and cytokine markers in diabetics dividuals. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 5 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. Active men with diabetes and hypertension presented less inflammation than active women IL-17 (ng/mL) 0.5993 2.207 0.494 2.019 SD: 1.223 §¶ SD: 2.099 †‡ SD: 0.666 §¶ SD: 2.637 †‡ IL-1β (pg/mL) 15.5 35.94 20.65 32.34 SD: 13.76 §¶ SD: 20.54 †‡ SD: 27.03 §¶ SD: 25.5 †‡ TNF-α (ng/mL) 0.6044 3.142 0.4886 2.829 SD: 0.857 §¶ SD: 2.123 †‡ SD: 0.776 §¶ SD: 1.809 †‡ IL-10 (pg/mL) 145 20.77 101.2 30.37 SD:143.4 §¶ SD: 53.61 †‡ SD:128.2 §¶ SD:61.68 †‡ IFN-γ (pg/mL) 2.806 6.118 3.253 6.267 SD:3.601 §¶ SD:7.194 †‡ SD:3.554 §¶ SD:8.208 †‡ IL-10/IL-17A 3731 103.2 1187 169.3 SD:555 §¶ SD:303.0 †‡ SD:3517 §¶ SD:663.0 †‡ IL-10/IL-1β 7.503 3.444 20.31 2.019 SD: 13.65 §¶ SD: 9.656 †‡ SD: 49.48 §¶ SD:4.743 †‡ IL-10/IFN-γ 51.61 16.63 34.36 14.45 SD: 99.22 §¶ SD: 37.02 † SD: 60.87 ¶ SD: 33.47 †‡ IL-10/TNF-α 175.9 26.36 95.33 42.00 SD: 413.7 §¶ SD: 33.93 †‡ SD: 140.6 §¶ SD: 232.0 †‡ Table I. Continuation. Table I. Continuation. Physically active men n=40; Sedentary men n=54; Physically active women n=98;Sedentary women n=184. Differences were considered significant when p value was less than 0.05 and, if detected, and was illustrated by: † (When there are differences with the physically active men group); § (When there are differences with the sedentary men group); ‡ (When there are differences with the physically active women group); ¶ (When there are differences with thesedentary women group). SD: Standard deviation; WC: Waist circumference; BMI: Body mass index; MSP: mean blood systolic pressure; HDL: High density lipoprotein;LDL: Low density lipoprotein;VLDL: Very low density lipoprotein; 17A: Interleukin 17A;IL-1β: Interleukin 1 beta;TNF-α: Tumor necrosis factor alpha;IL-10: Interleukin 10;IFN-γ: Interferon gamma;mmHg: millimeters ofmercury;Kg: kilogram; kg/m2: Kilograms per square meter; %:Percent; mg/dL: Milligram per deciliter; ng/mL: Nanogram per milliliter; pg/mL: Picogram per milliliter. Physically active men n=40; Sedentary men n=54; Physically active women n=98;Sedentary women n=184. Differences were considered significant when p value was less than 0.05 and, if detected, and was illustrated by: † (When there are differences with the physically active men group); § (When there are differences with the sedentary men group); ‡ (When there are differences with the physically active women group); ¶ (When there are differences with thesedentary women group). Physically active men n=40; Sedentary men n=54; Physically active women n=98;Sedentary women n=184. Differences were considered significant when p value was less than 0.05 and, if detected, and was illustrated by: † (When there are differences with the physically active men group); § (When there are differences with the sedentary men group); ‡ (When there are differences with the physically active women group); ¶ (When there are differences with thesedentary women group). SD: Standard deviation; WC: Waist circumference; BMI: Body mass index; MSP: mean blood systolic pressure; HDL: High density lipoprotein;LDL: Low density lipoprotein;VLDL: Very low density lipoprotein; 17A: Interleukin 17A;IL-1β: Interleukin 1 beta;TNF-α: Tumor necrosis factor alpha;IL-10: Interleukin 10;IFN-γ: Interferon gamma;mmHg: millimeters ofmercury;Kg: kilogram; kg/m2: Kilograms per square meter; %:Percent; mg/dL: Milligram per deciliter; ng/mL: Nanogram per milliliter; pg/mL: Picogram per milliliter. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 6 \| 13 Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers This research provides relevant information on the influence of gender on the body’s responsiveness to physical activity, especially with regard to modulation in body composition and inflammatory balance in individuals with SAH and T2DM. In this work, although there are no statistical differences regarding drug treatment between men and women, specific differences were observed between genders. The lipid profile showed no gender-related differences in changes in concentrations due to physical activity. However, in sedentary women, these parameters seem to be more directly associated with modulations in body composition than in sedentary men, an important fact since these parameters are risk factors for T2DM and SAH. Active men showed a better anti-inflammatory balance when compared to active women. In addition, there were punctual differences in the way in which some cytokines such as IL- 10 relate to the other parameters in a gender- dependent manner. Therefore, these results suggest that gender is related to a differential modulation of physical activity not only in the serum concentrations of biochemical and inflammatory markers, but also in the way they interact with each other and with the different systems in the physiological microenvironment in individuals with SAH and T2DM. Along with the changes in serum concentrations of biochemical and immune markers, we also observed that each assessed group presented a distinct multiple correlation profile. People engaged in physical activity, for instance, had positive correlations between their mean blood systolic pressure (MSP) and fat mass, BMI, and their waist circumference/height ratio. Thus, both physically active men and women showed similar anthropometric profile. A negative correlation between the serum inflammatory marker TNF-α and HDL was noticed in all groups. Hypertensive and diabetic women, regardless of physical activity, displayed a strong direct correlation between blood glucose levels and TNF-α, likewise, IL-10 serum levels correlated with sera glucose levels in physically active women. In this group, a strong negative correlation (p>0,001) between HDL and glucose is remarkable, however, a strong negative correlation (p>0,001) between HDL and IL-10 was observed. Concerning to their synergy, IL-17 and TNF-α were positively correlated in men’s blood sera, while IL-1β and TNF-α were positively correlated in all physically active groups. Sedentary men and women presented distinct triglyceride and VLDL correlations (Figure 1). In sedentary women, triglycerides and VLDL positively correlated with fat mass and were negatively associated with lean mass. GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 7 \| 13 Active men with diabetes and hypertension presented less inflammation than active women SD: Standard deviation; WC: Waist circumference; BMI: Body mass index; MSP: mean blood systolic pressure; HDL: High density lipoprotein;LDL: Low density lipoprotein;VLDL: Very low density lipoprotein; 17A: Interleukin 17A;IL-1β: Interleukin 1 beta;TNF-α: Tumor necrosis factor alpha;IL-10: Interleukin 10;IFN-γ: Interferon gamma;mmHg: millimeters ofmercury;Kg: kilogram; kg/m2: Kilograms per square meter; %:Percent; mg/dL: Milligram per deciliter; ng/mL: Nanogram per milliliter; pg/mL: Picogram per milliliter. committed tophysical activity, for both men and women (Table I). physically active men seemed to have a much higher anti-inflammatory/inflammatory cytokine ratio in serum than physically active women, as IL-10/IL-17A ratio were 3.1 times higher in physically active men, followed by an increase in IL-10/TNF-α ratio (1.8x) and IL-10/IFN-γ ratio (1.5x). We also sought to compare the ratio between IL-10 and IL-1β or TNF-α or IL-17 or IFN-γ in active or sedentary men and women. Men and women engaged inexercises exhibited higher ratio between IL-10 and all inflammatory cytokines previously mentioned (Table I). Nonetheless, An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 6 \| 13 Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers Notably, both triglycerides and VLDL correlated with BMI in sedentary women and men, positively and negatively respectively. Overall, sedentary individuals exhibited a greater number of correlations between their sera biochemical markers than the group in exercises. It could be explained by the reduced serum concentrations of these markers in physically active people (Table I). Body mass, usually expressed by the percentage of fat and lean mass, is an important predictor of the individual’s health status and can be used as a predictor of chronic diseases, being essential the healthy control of these parameters (Sagun et al. 2014). In this study, we observed that hypertensive and diabetic men, regardless of physical activity, had lower values of fat mass than women did. This could be related to the fact that even active women presented a relevant overweight BMI. It is well An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 7 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. Figure 1. Correlations between anthropometric, blood pressure data, biochemical and immunological markers. a: Physically active men (n=40); b: Sedentary men (n=54); c: Physically active women (n=98); d: Sedentary women (n=184). Positive correlations are indicated in blue, and reverse correlations areindicated in red. Darker color tones indicate higher r values. pvalues are indicated based on Spearman's rank correlation tests (* p< .05; p< .01; * p<.001;**** p< .0001). Variable grouping is based on principal component analysis using the R packet corrplot.WC: Waist circumference; BMI: Body mass index; MSP: Mean systolic pressure. Figure 1. Correlations between anthropometric, blood pressure data, biochemical and immunological markers. a: Physically active men (n=40); b: Sedentary men (n=54); c: Physically active women (n=98); d: Sedentary women (n=184). Positive correlations are indicated in blue, and reverse correlations areindicated in red. Darker color tones indicate higher r values. pvalues are indicated based on Spearman's rank correlation tests (* p< .05; p< .01; * p<.001;**** p< .0001). Variable grouping is based on principal component analysis using the R packet corrplot.WC: Waist circumference; BMI: Body mass index; MSP: Mean systolic pressure. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 8 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. established that estrogen plays an important role in regulating body fat in women (Ahmadi et al. 2018). Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers However, studies show that in obese women, many protective effects of estrogen are lost (McGill et al. 2011). This explains the fact that, despite being overweight, active women had better immunological and biochemical parameters when compared to sedentary women. In addition, women are generally more responsive to physical activity, even leisure time, to control fat mass, when compared to men (Ball et al. 2001). Thus, regardless of differences in the body fat accumulation profile, active women are more protected than sedentary men or women. In this way, we demonstrate that men exhibit distinct profiles of correlations between serum lipoproteins than those in women. Inflammatory cytokines are soluble markers that play an important role in the pathology of chronic noncommunicable diseases such as T2DM and SAH. A number of studies involving the application of physical activity have already reported the reduction of resting serum concentrations of inflammatory cytokines such as TNF-α, IL-6 and IL-1β (Moon et al. 2017). Particularly in obese and overweight individuals, the practice of physical exercise (e.g., walking, running and jogging) contributed to the reduction of resting serum IL-1β levels probably by reducing fat mass (Olson et al. 2007). In fact, regardless of sex, individuals classified here as physically active (≥150 minutes/week of physical activity) had lower restingserum concentrations of all inflammatory markers (IL-1β, TNF-α, IFN-γ and IL- 17). However, studies show that increased serum concentrations of pro-inflammatory molecules are predictive of frailty incidents in women, but not men (Gale et al. 2013). Therefore, although gender is not a determinant in the serum level of the analyzed proinflammatory cytokines, it is possible that their responsiveness to them is different between the genders. In this way, more studies are needed to clarify this issue. We have not been able to detect lower values of fat mass in body composition of women in exercise, even though their inflammatory serum cytokines concentrations were lower than the sedentary ones. This picture illustrates the multifactorial nature of physical activity influence on the production and serum concentration of inflammatory cytokines. Increased levels of glycaemia have been reported as a factor that stimulates the production of TNF-α by peripheral monocytes, then leading to spontaneous release of this inflammatory cytokine (Gonzalez et al. 2012). An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 9 \| 13 Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers This fact highlights its clinical In a research carried out in Brazil with 15,105 participants of both genders, physically active or not (physically active ≥ 150 minutes of physical activity/week of moderate physical activity), it was possible to observe that individuals who used to exercise exhibited lower serum levels of LDL, VLDL and triglycerides, accompanied by a higher concentration of serum HDL (Da Silva et al. 2016). Here, we were able to observe a reproduction of this result along with the detection of lower serum glucose values for individuals in physical activity. Moreover, this difference does not seem to be influenced by gender. Besides that, physically active individuals, regardless of sex, exhibited a stronger positive correlation between total and HDL cholesterol. It is well-defined that men and women exhibit a different lipoprotein metabolism throughout life, estrogen seems to influence the homeostasis of production and thereby serum concentration of lipoproteins in women (Soares et al. 2009). Here, sedentary and physically active women exhibited positive correlation between serum cholesterol and LDL. Physically active men, on the other hand, did not exhibit any correlation between LDL and cholesterol, only presenting a weak correlation between serum cholesterol levels and VLDL. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 9 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. relevance in T2DM pathogenesis, especially in women, where we were able to encounter a positive correlation between fasting glycaemia and TNF-α serum levels, in both sedentary and physically active individuals. Interleukin-17A is an essential mucosal cytokine (Barin et al. 2016). Nevertheless, in this context, might act as an inflammation potentiator, contributing to intense vascular inflammation especially by its synergic action with TNF-α, through the induction of further production of inflammatory cytokines and chemokines, as well as the expression of receptors, such as TNF receptor (Veldhoen 2017). Interleukin-10 is an anti-inflammatory cytokine that exercises essential control over biochemical and immunological parameters, having an important protective role in diseases such as T2DM and SAH (Jung & Choi 2014, Ribeiro et al. 2019). Its serum concentration is associated with increased skeletal muscle sensitivity to insulin and inhibition of inflammation in adipose tissue. Thus, it can contribute to glycemic control and regulation of the systemic inflammatory state associated with the pathogenesis of chronic non-communicable diseases (Hong et al. 2009). Interestingly, in this study, IL-10 showed different correlation patterns between active women and the other groups. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 10 \| 13 Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers In these women, from the observed correlations, IL-10 seems to participate more strongly in glycemic control. Such a pattern was not observed in active or sedentary men. Studies have shown that stimulation of the estrogen receptor modulates the secretion of IL-10 by inflammatory cells (Itoga et al. 2015). Furthermore, with increasing age, serum concentrations of IL-10 decrease more markedly in men than in women, showing that it varies in different ways depending on the individual’s gender (Hirokawa et al. 2013). relevance in T2DM pathogenesis, especially in women, where we were able to encounter a positive correlation between fasting glycaemia and TNF-α serum levels, in both sedentary and physically active individuals. Interleukin-17A is an essential mucosal cytokine (Barin et al. 2016). Nevertheless, in this context, might act as an inflammation potentiator, contributing to intense vascular inflammation especially by its synergic action with TNF-α, through the induction of further production of inflammatory cytokines and chemokines, as well as the expression of receptors, such as TNF receptor (Veldhoen 2017). is a potent inflammation modulator that directly inhibits the synthesis of inflammatory markers such as IL-1α, IL-1β, TNF-α, IL-8 and MIP-1α (Nimmo et al. 2013). In our study, we observed that in physically active individuals there was a higher serum concentration of IL-10. In fact, physical activity is able to stimulate an increase in serum levels of IL-6, followed by an increase in serum levels of resting IL-10 (Petersen & Pedersen 2005). Given its regulatory importance in maintaining inflammatory homeostasis, the relationship between serum concentrations of IL-10 with TNF-α and IL-17 was analyzed, and it was observed that, regardless of gender, physically active individuals exhibited higher proportions among IL-10 and markers. Thus, this molecule can and should contribute to the control of the chronic systemic inflammatory microenvironment of T2DM and SAH. However, although the IL-10 / TNF-α and IL-10 / IL-17A ratios seem superior in men, Ribeiro et al. (2019) observed that IL-17A and IL-10 have a relevant role associated with the modulation of risk factors inflammatory, biochemical and anthropometric in elderly women with SAH and T2DM. Thus, we postulate that the inflammatory modulation performed by IL-10/IL-17A associated with the pathophysiology of SAH and T2DM may be related to gender, but also to age. Acknowledgments This work was supported by the Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB) and by Universidade Federal da Bahia – Instituto Multidisciplinar em Saúde - Campus Anísio Teixeira (UFBA-IMS-CAT) (Prodoc UFBA 04/2014). Some students involved in this study received Master and Doctoral scholarships from Fundação de Amparo a Pesquisa do Estado da Bahia (FAPESB), from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil. HONG EG ET AL. 2009. Interleukin-10 prevents diet-induced insulin resistance by attenuating macrophage and cytokine response in skeletal muscle. Diabetes 58(11): 2525-2535. HU G, LINDSTRÖM J, VALLE TT, ERIKSSON JG, JOUSILAHTI P, SILVENTOINEN K, QIAO E & TOUMILEHTO J. 2004. Physical Activity, Body Mass Index, and Risk of Type 2 Diabetes in Patients with Normal or Impaired Glucose Regulation. Arch Intern Med 164(8): 892-896. ITOGA M ET AL. 2015. G-protein-coupled estrogen receptor agonist suppresses airway inflammation in a mouse model of asthma through IL-10. PLoS ONE 10(3): e0123210. GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. genders, it is possible that, depending on the gender, this modulation perpetrated through different control routes. Such knowledges are essential since once the main differences between the genders are identified, it is possible to determine personalized treatment methodologies, respecting the individual’s particularities and enabling better results. However, further studies are needed involving parameters such as analysis of serum androgens and age in hypertensive and diabetic individuals, using models of amplified correlation analysis. Only in this way will it be possible to achieve a greater understanding of the complex relationship between physical activity and gender in the biochemical, inflammatory and anthropometric markers associated with the pathophysiology of hypertension and diabetes. BARIN JG ET AL. 2016. Collaborative Interferon-γ and Interleukin-17 Signaling Protects the Oral Mucosa from Staphylococcus aureus. Am J Pathol 186(9): 2337-2352. BARIN JG ET AL. 2016. Collaborative Interferon-γ and Interleukin-17 Signaling Protects the Oral Mucosa from Staphylococcus aureus. Am J Pathol 186(9): 2337-2352. CHEN L, MAGLIANO DJ & ZIMMET PZ. 2011. The worldwide epidemiology of type 2 diabetes mellitus—present and future perspectives. Nat Rev Endocrinol 8(4): 228-236. CHEUNG BMY & LI C. 2012. Diabetes and Hypertension: Is There a Common Metabolic Pathway? CurrAtheroscler Rep 14(2): 160-166. DA SILVA RC, DINIZ MFHS, ALVIM S, VIDIGAL PG, FEDELI LMG & BARRETO SM. 2016. Physical Activity and Lipid Profile in the ELSA-Brasil Study. Arq Bras Cardiol 107(1): 10-19. DA SILVA RC, DINIZ MFHS, ALVIM S, VIDIGAL PG, FEDELI LMG & BARRETO SM. 2016. Physical Activity and Lipid Profile in the ELSA-Brasil Study. Arq Bras Cardiol 107(1): 10-19. GALE CR, BAYLIS D, COOPER C & SAYER AA. 2013. Inflammatory marker sand incident frailty in men and women: the English Longitudinal Study of Ageing. AGE 35: 2493-2501. GONZALEZ Y, HERRERA MT, SOLDEVILA G, GARCIA-GARCIA L, FABIÁN G, PÉREZ-ARMENDARIZ EM, BOBADILLA K, GUZMÁN-BELTRÁN S, SADA E & TORRES M. 2012. High glucose concentrations induce TNF-α production through the down-regulation of CD33 in primary human monocytes. BMC Immunol 13(1): 19. HIROKAWA K, UTSUYAMA M, HAYASHI Y, KITAGAWA M, MAKINODAN T & FULOP T. 2013. Slower immune system aging in women versus men in the Japanese population. Immun Ageing 10: 19. Men and women exhibited distinct patterns of multiple correlations among biochemical, anthropometric, and immune markers Interleukin-10 is an anti-inflammatory cytokine that exercises essential control over biochemical and immunological parameters, having an important protective role in diseases such as T2DM and SAH (Jung & Choi 2014, Ribeiro et al. 2019). Its serum concentration is associated with increased skeletal muscle sensitivity to insulin and inhibition of inflammation in adipose tissue. Thus, it can contribute to glycemic control and regulation of the systemic inflammatory state associated with the pathogenesis of chronic non-communicable diseases (Hong et al. 2009). Interestingly, in this study, IL-10 showed different correlation patterns between active women and the other groups. In these women, from the observed correlations, IL-10 seems to participate more strongly in glycemic control. Such a pattern was not observed in active or sedentary men. Studies have shown that stimulation of the estrogen receptor modulates the secretion of IL-10 by inflammatory cells (Itoga et al. 2015). Furthermore, with increasing age, serum concentrations of IL-10 decrease more markedly in men than in women, showing that it varies in different ways depending on the individual’s gender (Hirokawa et al. 2013). However, further studies are needed to assess whether glycemic control related to IL-10 has differences between genders. In addition, IL-10 All together, our findings indicate that there are gender-related differences in the modulation of anthropometric and biochemical characteristics by physical activity in hypertensive and diabetic individuals. Although exercise is clearly beneficial for reducing lipoproteins associated with increased cardiovascular risk, multiple correlation analyzes indicate a complex and gender-related relationship between physical activity and these lipoprotein markers. In addition, these gender-related correlations are also observed for inflammatory markers. However, despite being beneficial in both An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 10 \| 13 REFERENCES AHMADI S, ESHRAGHIAN MR, HEDAYATI M & PISHVA H. 2018. Relationship between estrogen and body composition, energy, and endocrine factors in obese women with normal and low REE. Steroids 130: 31-35. JUNG U & CHOI MS. 2014. Obesity and its metabolic complications: the role of adipokines and the relationship between obesity, inflammation, insulin resistance, dyslipidemia and nonalcoholic fatty liver disease. Int J Mol Sci 15: 6184-6223. AROOR AR, MANDAVIA CH & SOWERS JR. 2012. Insulin Resistance and Heart Failure: Molecular Mechanisms. Heart Fail Clin 8(4): 609-617. AROOR AR, MANDAVIA CH & SOWERS JR. 2012. Insulin Resistance and Heart Failure: Molecular Mechanisms. Heart Fail Clin 8(4): 609-617. KISSELA BM ET AL. 2012. Age at stroke: Temporal trends in stroke incidence in a large, biracial population. Neurology 79(17): 1782-1787. KISSELA BM ET AL. 2012. Age at stroke: Temporal trends in stroke incidence in a large, biracial population. Neurology 79(17): 1782-1787. BALL K, OWEN N, SALMON J, BAUMAN A & GORE CJ. 2001. Associations of physical activity with body weight and fat in men and women. Int J Obes 25: 914-919. LOPES WA, LEITE N, DA SILVA LR, BRUNELLI DT, GÁSPARI AF, RANDOMINSKI RB, CHACON-MIKAHIL MPT & CAVAGLIERI CR. 2016. An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 11 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. measurements to predict metabolic syndrome. Clinics 69(5): 347-353. Effects of 12 weeks of combined training without caloric restriction on inflammatory markers in overweight girls. J Sports Sci 34(20): 1902-1912. Effects of 12 weeks of combined training without caloric restriction on inflammatory markers in overweight girls. J Sports Sci 34(20): 1902-1912. SMAJLOVIĆ D. 2015. Strokes in young adults: Epidemiology and prevention. Vasc Health Risk Manag 11: 157-164. MARENGONI A, WINBLAD B, KARP A & FRATIGLIONI L. 2008. Prevalence of chronic diseases and multimorbidity among the elderly population in Sweden. Am J Public Health 98(7): 1198-1200. SOARES GM, VIEIRA CS, DE PAULA MARTINS W, DOS REIS RM, DE SÁ MFS & FERRIANI RA. 2009. Metabolic and cardiovascular impact of oral contraceptives in polycystic ovary syndrome. Int J Clin Pract 63(1): 160-169. MATSUDO S, ARAÚJO T, MATSUDO V, ANDRADE D, ANDRADE E, OLIVEIRA LC & BRAGGION G. 2001. International physical activity questionnaire (IPAQ): study of validity and reliability in Brazil. Rev Bras Ativ Fís Saúde 6: 5-18. TIMMONS BW, HAMADEH MJ & TARNOPOLSKY MA. 2006. REFERENCES No effect of short-term 17β-estradiol supplementation in healthy men on systemic inflammatory responses to exercise. Am J Physiol Regul Integr Comp Physiol 291(2): R285-R290. MCGILL JB, PETERSON LR, HERRERO P, SAEED IM, RECKLEIN C, COGGAN AR, DEMOSS AJ, SCHECHTMAN KB, DENCE CS & GROPLER RJ. 2011. Potentiation of abnormalities in myocardial metabolism with the development of diabetes in women with obesity and insulin resistance. J Nucl Cardiol 18: 421-429. TRACHTA P ET AL. 2014. Three months of regular aerobic exercise in patients with obesity improve systemic subclinical inflammation without major influence on blood pressure and endocrine production of subcutaneous fat. Physiol Res 63(2Suppl): S299-S308. MITTAL BV & SINGH AK. 2010. Hypertension in the developing world: challenges and opportunities. Am J Kidney Dis 55(3): 590-598. VELDHOEN M. 2017. Interleukin 17 is a chief orchestrator of immunity. Nat Immunol 18(6): 612-621. MOON S, SCHMIDT M, SMIRNOVA IV, COLGROVE Y & LIU W. 2017. Qigong exercise may reduce serum TNF-α levels and improve sleep in people with Parkinson’s disease: A pilot study. Medicines 4(2): 23. WEI T & SIMKO V. 2013. Corrplot: Visualization of a correlation matrix. R package version 0.73 230(231): 11. WIZEMANN TM & PARDUE ML. 2001. Exploring the Biological Contributions to Human Health. Does Sex Matter? Washington (DC): National Academy Press, 288 p. NAG T & GHOSH A. 2013. Cardiovascular disease risk factors in Asian Indian population: A systematic review. J Cardiovasc Dis Res 4(4): 222-228. ZHONG J ET AL. 2014. T-cell costimulation protects obesity- induced adipose inflammation and insulin resistance. Diabetes 63(4): 1289-1302. NIMMO MA, LEGGATE M, VIANA JL & KING JA. 2013. The effect of physical activity on mediators of inflammation. Diabetes Obes Metab 15(3 Suppl): 51-60. ZHOU H, YANG C, DONG C, GUO Z, HU X, XU Y & ZHOU Z. 2014. Predictive Power for Type 2 Diabetes Mellitus using Dynamic Change of Metabolic Syndrome, Dynamic Change of Fasting Plasma Glucose, Metabolic Syndrome and Fasting Plasma Glucose. Iran J Public Health 43(4): 432-440. OLSON TP, DENGEL DR, LEON AS & SCHMITZ KH. 2007. Changes in inflammatory biomarkers following one-year of moderate resistance training in overweight women. Int J Obes 31(6): 996-1003. PETERSEN AMW & PEDERSEN BK. 2005. The anti-inflammatory effect of exercise. J Appl Physiol 98(4): 1154-1162. Correspondence to: Gilvanéia S. Santos E-mail: gilsufba@gmail.com FERNANDA M. DA SILVA LIMA1 https://orcid.org/0000-0001-9108-7831 E-mail: gilsufba@gmail.com How to cite PRINCE MJ, WU F, GUO Y, ROBLEDO LMG, O’DONNELL M, SULLIVAN R & YUSUF S. 2015. The burden of disease in older people and implications for health policy and practice. The Lancet 385(9967): 549-562. DOS SANTOS DC ET AL. 2021. Gender-related differences in the modulation of anthropometric, biochemical, and immune markers by physical activity in hypertensive and diabetic individuals. An Acad Bras Cienc 93: e20201905. DOI 10.1590/0001-3765202120201905. RIBEIRO IS ET AL. 2019.Association between body composition and inflammation: A central role of IL-17 and IL-10 in diabetic and hypertensive elderly women. Exp Gerontol 127: 110734. RIBEIRO IS ET AL. 2019.Association between body composition and inflammation: A central role of IL-17 and IL-10 in diabetic and hypertensive elderly women. Exp Gerontol 127: 110734. Manuscript received on December 10, 2020; accepted for publication on September 6, 2021 DÉBORAH C. DOS SANTOS1 https://orcid.org/0000-0003-4880-6014 DÉBORAH C. DOS SANTOS1 https://orcid.org/0000-0003-4880-6014 SAGUN G, OGUZ A, KARAGOZ E, FILIZER AT, TAMER G & MESCI B. 2014. Application of alternative anthropometric SAGUN G, OGUZ A, KARAGOZ E, FILIZER AT, TAMER G & MESCI B. 2014. Application of alternative anthropometric An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 12 \| 13 An Acad Bras Cienc (2021) 93(Suppl. 4) e20201905 13 \| 13 GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY GENDER MODULATES RESPONSE TO PHYSICAL ACTIVITY DÉBORAH C. DOS SANTOS et al. DIEGO PATRICK S. LOPES1 https://orcid.org/0000-0002-5075-6872 1Instituto Multidisciplinar em Saúde, Campus Anísio Teixeira, Universidade Federal da Bahia, Rua Hormindo Barros, 58, Bairro Candeias, 45.029-094 Vitória da Conquista, BA, Brazil 2Universidade Federal do Sul da Bahia, Campus Paulo Freire, Praça Joana Angélica, 250, São José, 45988-058 Teixeira de Freitas, BA, Brazil 1Instituto Multidisciplinar em Saúde, Campus Anísio Teixeira, Universidade Federal da Bahia, Rua Hormindo Barros, 58, Bairro Candeias, 45.029-094 Vitória da Conquista, BA, Brazil 2Universidade Federal do Sul da Bahia, Campus Paulo Freire, Praça Joana Angélica, 250, São José, 45988-058 Teixeira de Freitas, BA, Brazil ÍTALO S. PEREIRA1 https://orcid.org/0000-0001-5697-5673 DENISAR P. SANTOS1 https://orcid.org/0000-0002-3930-9510 IGOR P.R. MUNIZ1 https://orcid.org/0000-0003-1842-6432 ISRAEL S. RIBEIRO1,2 https://orcid.org/0000-0002-1262-1387 DIEGO PATRICK S. LOPES1 https://orcid.org/0000-0002-9211-5758 ÍTALO S. PEREIRA1 https://orcid.org/0000-0001-5697-5673 DENISAR P. SANTOS1 https://orcid.org/0000-0002-3930-9510 IGOR P.R. MUNIZ1 https://orcid.org/0000-0003-1842-6432 FERNANDA M. DA SILVA LIMA1 https://orcid.org/0000-0001-9108-7831 ANDRESSA ANTÔNIA S. COSTA1 https://orcid.org/0000-0003-0079-7016 DANIELA DA SILVA P. SOUZA1 https://orcid.org/0000-0001-6050-433X STÉFANO PEDRO DE MELO CALADO https://orcid.org/0000-0003-3458-1328 CAROLINE V. GONÇALVES1 https://orcid.org/0000-0003-0895-1583 MARIA POLIANA L. GALANTINI1 https://orcid.org/0000-0001-9423-5334 DANILO N. LOPES1 https://orcid.org/0000-0001-9473-9268 ISRAEL S. RIBEIRO1,2 https://orcid.org/0000-0002-1262-1387 Author contributions ANDRESSA ANTÔNIA S. COSTA1 https://orcid.org/0000-0003-0079-7016 Déborah C. Santos, Israel S. Ribeiro, Diego P.S. Lopes and Ítalo S. Pereira: conception and design, acquisition of data, analysis and interpretation of data and manuscript writing. Denisar P. Santos, Igor P.R. Muniz, Fernanda M.S. Lima, Andressa A.S. Costa, Daniela S.P. Souza, Stéfano P.M. Calado, Caroline V. Gonçalves and Maria P.L. Galantini: acquisition of data, analysis and interpretation of data. Danilo N. Lopes, Robson A.A. da Silva and Gilvanéia S. Santos: conception and design, analysis and interpretation of data. DANIELA DA SILVA P. SOUZA1 https://orcid.org/0000-0001-6050-433X STÉFANO PEDRO DE MELO CALADO1 https://orcid.org/0000-0003-3458-1328
https://openalex.org/W3114427539	https://zenodo.org/records/4607382/files/2020_Ranavolo%20et%20al.%20Critical%20Issues%20and%20Imminent%20Challenges%20in%20the%20Use%20of%20sEMG%20in%20Return-To-Work%20Rehabilitation%20of%20Patients%20Affected%20by%20Neurologic.pdf	English	null	Critical Issues and Imminent Challenges in the Use of sEMG in Return-To-Work Rehabilitation of Patients Affected by Neurological Disorders in the Epoch of Human–Robot Collaborative Technologies	Frontiers in neurology	2,020	cc-by	8,399	PERSPECTIVE published: 22 December 2020 doi: 10.3389/fneur.2020.572069 PERSPECTIVE Keywords: return-to-work rehabilitation, surface electromyography, instrumental-based biomechanical risk assessment, exoskeletons control, manual material handling monitoring Critical Issues and Imminent Challenges in the Use of sEMG in Return-To-Work Rehabilitation of Patients Affected by Neurological Disorders in the Epoch of Human–Robot Collaborative Technologies Alberto Ranavolo 1, Mariano Serrao 2,3 and Francesco Draicchio 1 1 Department of Occupational and Environmental Medicine, Epidemiology and Hygiene, INAIL, Rome, Italy, 2 Department of Medical and Surgical Sciences and Biotechnologies, Sapienza University of Rome, Rome, Italy, 3 Movement Analysis LAB, Policlinico Italia, Rome, Italy Patients affected by neurological pathologies with motor disorders when they are of working age have to cope with problems related to employability, difﬁculties in working, and premature work interruption. It has been demonstrated that suitable job accommodation plans play a beneﬁcial role in the overall quality of life of pathological subjects. A well-designed return-to-work program should consider several recent innovations in the clinical and ergonomic ﬁelds. One of the instrument-based methods used to monitor the effectiveness of ergonomic interventions is surface electromyography (sEMG), a multi-channel, non-invasive, wireless, wearable tool, which allows in-depth analysis of motor coordination mechanisms. Although the scientiﬁc literature in this ﬁeld is extensive, its use remains signiﬁcantly underexploited and the state-of-the-art technology lags expectations. This is mainly attributable to technical and methodological (electrode-skin impedance, noise, electrode location, size, conﬁguration and distance, presence of crosstalk signals, comfort issues, selection of appropriate sensor setup, sEMG amplitude normalization, deﬁnition of correct sEMG-related outcomes and normative data) and cultural limitations. The technical and methodological problems are being resolved or minimized also thanks to the possibility of using reference books and tutorials. Cultural limitations are identiﬁed in the traditional use of qualitative approaches at the expense of quantitative measurement-based monitoring methods to design and assess ergonomic interventions and train operators. To bridge the gap between the return-to-work rehabilitation and other disciplines, several teaching courses, accompanied by further electrodes and instrumentations development, should be designed at all Bachelor, Master and PhD of Science levels to enhance the best skills available among physiotherapists, occupational health and safety technicians and ergonomists. Edited by: William Zev Rymer, Shirley Ryan AbilityLab, United States Reviewed by: Roberto Merletti, Politecnico di Torino, Italy Leif Sandsjö, University of Gothenburg, Sweden Correspondence: Alberto Ranavolo a.ranavolo@inail.it Edited by: William Zev Rymer, Shirley Ryan AbilityLab, United States Reviewed by: Roberto Merletti, Politecnico di Torino, Italy Leif Sandsjö, University of Gothenburg, Sweden Correspondence: Alberto Ranavolo a.ranavolo@inail.it Correspondence: Alberto Ranavolo a.ranavolo@inail.it Specialty section: This article was submitted to Neurorehabilitation, a section of the journal Frontiers in Neurology Received: 12 June 2020 Accepted: 30 November 2020 Published: 22 December 2020 Specialty section: This article was submitted to Neurorehabilitation, a section of the journal Frontiers in Neurology INTRODUCTION pulling, repetitive movements, and patients handling activities (32–39). However, they remain underused in return-to-work rehabilitation plans for people with neurological pathologies with motor disorders. Bipolar and multi-channel (high-density) surface electromyography (sEMG and HDsEMG) represent non-invasive physiological approaches, which enable greater comprehension of the upper limb, lower limb, and trunk muscle behaviors during the execution of movement (1–4). Diﬀerential bipolar sEMG signals, currently acquired using miniaturized wireless sensors attached to the skin, represent the spatio-temporal summation of all the motor unit action potentials, which propagate from the innervation zones to the tendon regions along the muscle ﬁbers closest to the skin. In view of the complexity of these interference signals, simple indices facilitate an appropriate and complete analysis of the electrical activity of muscles. Simple indices provide information on “when” and “how much” the muscles are electrically active during the execution of both isometric and dynamic activities, and include the following parameters: People aﬀected by neurological pathologies need to be integrated/reintegrated into their workplaces because their motor disease symptoms appeared when they were of working age, reducing their working capacity (40, 41) and employability (42). Recent studies have proven that avoiding early exit from employment plays a beneﬁcial and key role in the overall quality of life of people aﬀected by neurological disorders (41, 42). Clinicians manage their patients’ premature work interruption (43, 44) by designing appropriate traditional and innovative pharmacological, surgical, and rehabilitation treatments, such as robotic rehabilitation, virtual reality, and neuromodulation (45– 49). Furthermore, job accommodation plans are being enriched with new ergonomic options, such as work task rehabilitation and workplace interventions (50–52). In fact, the fourth industrial revolution has recently opened new occupational scenarios within which key human-robot collaborative (HRC) technologies, such as exoskeletons (Figure 1) and cobots, assist workers in their workplaces. Cobots can be deﬁned as reconﬁgurable collaborative robots able to interact with workers within a shared space and to respond to the worker intentions and task variations in a timely manner while simultaneously oﬄoading them from external loadings, and keep them in task- optimum and ergonomic working conditions. Small-medium enterprises can currently use collaborative technologies allowing ﬂexible and ergonomic workplaces, which can adapt to the characteristics of workers with neurological disorders (53). Citation: Ranavolo A, Serrao M and Draicchio F (2020) Critical Issues and Imminent Challenges in the Use of sEMG in Return-To-Work Rehabilitation of Patients Affected by Neurological Disorders in the Epoch of Human–Robot Collaborative Technologies. Front. Neurol. 11:572069. doi: 10.3389/fneur.2020.572069 December 2020 \| Volume 11 \| Article 572069 Frontiers in Neurology \| www.frontiersin.org 1 sEMG in Return-To-Work Rehabilitation Ranavolo et al. Frontiers in Neurology \| www.frontiersin.org INTRODUCTION Quite recently, the European Union’s Horizon 2020 research and innovation program funded the SOPHIA (Socio-physical Interaction Skills for Cooperative Human-Robot Systems in Agile Production) project to establish and achieve, among others, the goal of validating the HRC technologies developed under the aegis of the project in the healthcare sector and in return-to-work rehabilitation of patients aﬀected by neurological disorders. In particular, the European consortium is developing myoelectric HRC interfaces to study how new hybrid work environments can ﬂexibly adapt to the human physical states and needs, thereby contributing to improvements in ergonomic interventions (53). Furthermore, the project has the aims, for prevention, to design training plans of professionals specialized in the workers motor performance measurement by using sEMG- based approaches, develop miniaturized wearable devices to monitor human-motor variables and render haptic stimuli to speciﬁc areas of the worker’s body and develop new standards for adaptation of work environments and biomechanical risk assessment in collaborative manufacturing scenarios. Within this context, it is quite evident that sEMG can play a crucial role in complex vocational reintegration programs in classifying residual motor functions, assessing pre-post-rehabilitation and ergonomic interventions, and controlling wearable robotics. The professionals trained for the sEMG in return-to-work programs should be physiotherapists, occupational health and safety technicians and ergonomists which should operate in a multidisciplinary team also constituted by neurologists, - amplitude indices (5, 6) ◦maximum, ◦averaged rectiﬁed value, ◦root mean square, and - muscle activation timings (7). - muscle activation timings (7). Furthermore, sEMG-based algorithms enable greater in-depth comprehension of the muscle coordination mechanisms adopted by the central nervous system (CNS) by estimating the following: - simultaneous activation of several muscles or muscle groups (co-activation), a mechanism adopted by the CNS to stabilize joints, upper and lower limbs, and the spine (8–11), - myoelectric manifestation of muscle fatigue, estimated by measuring the decrease in ﬁber conduction velocity (12, 13), which is reﬂected in an amplitude increase and spectral compression over time (14–16), and - locomotor coordination (17–26), analyzed to comprehend how CNS lesions of neurological subjects with motor deﬁcits inﬂuence plasticity and modular control of muscle patterns (27). It has been demonstrated that the CNS can linearly combine, with diﬀerent weights, a limited number of basic functions called primitives or muscle synergies, to implement several motor tasks. During steady-state walking and running, ﬁve and four primitives, respectively, account for muscle activity (28–30). High-density sEMG recordings are performed using high-density surface grids placed on the skin to evaluate the online spatial distribution of the sEMG activity and estimate the discharge times of several motor units by using decomposition algorithms (2, 31). Despite the many advantages of sEMG and HDsEMG, these instrumental tools are largely underexploited and their application lags expectations in the ﬁelds of ergonomics and occupational medicine. These tools have begun to be applied in the prevention of work-related musculoskeletal disorders, a set of painful inﬂammatory and degenerative conditions aﬀecting the joints, spinal dizcs, cartilage, muscles, tendons, ligaments and peripheral nerves, caused by manual lifting, pushing and December 2020 \| Volume 11 \| Article 572069 2 sEMG in Return-To-Work Rehabilitation Ranavolo et al. recruited in the study and in accordance with the local legislation and institutional requirements. FIGURE 1 \| (A) Wearable wireless sEMG sensors placed bilaterally over the erector spinae (ES), latissimus dorsi (LD), upper trapezius (UT) and posterior deltoids (PD) muscles to assess the efﬁcacy of a passive exoskeleton during the execution of an overhead screwing activity. (B) sEMG envelopes of right ES, LD, UT and PD muscles of a representative subject without (red traces) and with (green traces) the use of a passive exoskeleton. Degenerative and acquired neurological diseases, including neuropathies, multiple sclerosis, stroke, spastic paraplegia, cerebellar ataxia, dystonia, traumatic spine and brain lesions, and encephalitis, are disorders, which can aﬀect the motor function during working age and severely limit the autonomy and eﬃciency of workers (42, 54–58). Application of sEMG in Monitoring Occupational health and safety technicians have an academic degree oﬀered at Bachelor of Science level in the same ﬁeld of physiotherapists (health professions sciences). Instead, professionals from diﬀerent disciplines, for example occupational physicians, who aspire to practice the profession of ergonomist, need a degree that is not conferred by universities. In Europe, the Center for the Registration of the European Ergonomists (CREE) is the professional certiﬁcation supported by European ergonomics associations and recognized by the International Ergonomics Association (IEA) which provides the title of “Eur.Erg.” and allows to practice the profession of ergonomist in 47 countries. When a worker aﬀected by a neurological pathology with motor disorders is reintegrated at work, an exhaustive assessment of his/her residual motor function is of primary importance to design and/or adapt his/her workplace well. Additionally, it is necessary to verify and monitor the eﬃcacy of these ergonomic interventions over time. Although no consistent studies have speciﬁcally investigated the sEMG of patients with neurological disorders in the workplace context, there are several reports on the residual muscle function assessment of patients with neurological disorders [(10), for review see (59, 60)]. Many muscle activation measures and indices exploring several aspects of motor control have been proposed for patients with neurological diseases: co- contraction/co-activation of single-joint or multi-joint muscles [(9, 23, 61), for review see (62)], spatio-temporal modular muscle activation (19–22, 27), muscle activation asymmetry (63), time-frequency coherences between joint muscle signals during dynamic contractions to detect the spasticity in the upper limbs (64), and muscle fatigue (65, 66). Furthermore, it is possible to obtain from these studies a set of functional measures to consider for work reintegration while simultaneously taking into account the uniqueness of the motor deﬁcit speciﬁc to each disease. The above premise serves as a rationale for identifying and discussing the barriers to the coherent and widespread use of sEMG in work integration/reintegration. This article represents the perspective, under the aegis of the SOPHIA project, of the Laboratory of Ergonomics and Physiology of the Italian Institute for Insurance against Accidents at Work (INAIL), a public non-proﬁt entity aimed at facilitating the return-to-work of people with motor disorders. - muscle activation timings (7). The motor impairment of workers aﬀected by neurological diseases may encompass several motor domains, including hand function, balance, and locomotion, resulting in considerable onus on the society in terms of reduced work productivity and cost. The main purpose of pharmacological, surgical, and rehabilitation treatments must be to improve the motor performance, autonomy, and daily lives of patients, thereby oﬀering them the possibility of returning to work and optimizing their work capability. Combined with kinematic and kinetic measurements, sEMG is currently widely used in research laboratories by movement scientists and still little in clinical routine by health operators to classify quantitatively the nature and degree of motor dysfunction, analyze the complex relationship between the primary deﬁcit and the adaptive and compensatory mechanisms, categorize patients based on their speciﬁc neurological disease, and ﬁnally monitor pre-post-treatment. Importantly, as sEMG essentially investigates the ﬁnal output of motor commands, it can quantify the residual motor function, which can theoretically be monitored continuously in workplace adaptation and integrated into HRC technologies. FIGURE 1 \| (A) Wearable wireless sEMG sensors placed bilaterally over the erector spinae (ES), latissimus dorsi (LD), upper trapezius (UT) and posterior deltoids (PD) muscles to assess the efﬁcacy of a passive exoskeleton during the execution of an overhead screwing activity. (B) sEMG envelopes of right ES, LD, UT and PD muscles of a representative subject without (red traces) and with (green traces) the use of a passive exoskeleton. occupational physicians, physiatrists, biomedical engineers and movement scientists. We believe that no new professions should be created. AN OVERVIEW OF SEMG USE IN RETURN-TO-WORK OF PATIENTS WITH NEUROLOGICAL DISORDERS Some long-lasting degenerative neurological diseases, such as cerebellar ataxia and spastic paraplegia, often begin at a young age and can persist for the entire duration of a subject’s working life. This is of interest in the context of work-related rehabilitation because the majority of patients consider themselves capable of Ethical review and approval and written informed consent were not required for the study because no human participants were December 2020 \| Volume 11 \| Article 572069 Frontiers in Neurology \| www.frontiersin.org 3 Ranavolo et al. sEMG in Return-To-Work Rehabilitation working, and approximately 78% of non-working patients seek employment (42). Furthermore, workers with cerebellar ataxia show low or average-to-low job stress-related risk (42). by neural information obtained from the decomposition of HDsEMG (72). by neural information obtained from the decomposition of HDsEMG (72). g j The use of sEMG analysis in patients with degenerative cerebellar ataxia (for instance, spinocerebellar ataxia) reveals a series of muscle activation abnormalities (67). Speciﬁcally, patients with spinocerebellar ataxia show increased amplitude and duration of sEMG bursts in both the upper and lower (21, 22) limbs, with signiﬁcant diﬀerences in muscle activation timing (21, 22). Further evidence reveals increases in both single- joint antagonist muscle co-activation (61) and multi-joint multi- muscle co-activation (68). All these abnormalities are related to the severity of the disease and balance features, suggesting they could be exploited as potential biomarkers for the work- related biomechanical risk evaluation and workplace adaptation monitoring of these patients. One possible approach is the planning of assistive devices for workplace adaptation, such as the use of supportive elastic suits (69). These soft wearable devices can improve movement stability and reduce the need to co- activate muscles. Consequently, they can lower the associated energy costs and tissue-overuse injuries owing to excessive compression and shear forces at, for instance, the L5-S1 joint of the spine. BARRIERS TO SEMG USE IN RETURN-TO-WORK OF PATIENTS WITH NEUROLOGICAL DISEASES The critical issues hindering the widespread adoption of sEMG in return-to-work programs are mainly attributable to technical, methodological and cultural limitations. The technical issues are attributable to both monitoring and control functions. With regard to monitoring, the most critical technical aspects are strongly associated with the sEMG technique: - electrode-skin impedance, noise, and electrode contact stability; - while the methodological aspects are associated to: - problems linked to electrode location, size, conﬁguration, and distance; - presence of crosstalk signals (16). - placement of sEMG electrodes for long hours; - selection of the right sensor setup on the base of the neurological pathology and manual handling activity to be investigated; It has been reported that patients with degenerative spastic paraplegia (for instance, hereditary paraplegia) show increased global (23) and segmental lower limb muscle co-activation (9), which correlate positively with disease severity, degree of spasticity, and energetic costs (9). In addition, when mapping the simultaneous activities of a large number of muscles during walking onto the anatomical rostrocaudal location of the motor neuron pools (70), patients with spastic paraplegia show an abnormal spread of muscle activation during gait, initially involving the sacral segments and, at more severe stages, the lumbar segments (20). - management the sEMG amplitude normalization; - deﬁnition of appropriate sEMG-related outcomes and normative data. Fortunately, the eﬀect of these critical issues on the sEMG signal quality can be reduced with the aid of authoritative reference books and tutorials. In particular, the European Recommendations for Surface Electromyography (83), which needs to be updated, and the Atlas of Muscle Innervation Zones (84) are recommended as guides for the use of sEMG together with recent tutorials and consensus papers (85–87). A knowledge of the contents of these texts and tutorials makes users aware of the current limitations of the sEMG approach given its ability to monitor only a limited number of superﬁcial muscles. Despite this, the technical and methodological limitations of the sEMG approach can be minimized and do not justify its non-usage in return-to-work rehabilitation plans. In addition, new wearable sensors and electronic smart devices such as smartphones and tablets allow simple monitoring of the worker at the workplace. Wearable sensors do not interfere with the typical movements performed by workers owing to their miniaturization and wireless communication protocols. In addition, multi-channel sEMG systems are available in wireless versions despite their high number of channels and high data rate. BARRIERS TO SEMG USE IN RETURN-TO-WORK OF PATIENTS WITH NEUROLOGICAL DISEASES The combination of sensor networks and intelligent work environments provides real-time estimation of physiological parameters, enabling direct feedback to workers who are monitored directly and constantly at the workplace. Real-time monitoring is additionally useful for providing acoustic, visual, and vibro-tactile stimuli to workers (53, 88) executing manual handling tasks in awkward postures or requiring signiﬁcant physical eﬀort, or when muscle fatigue sets in. Frontiers in Neurology \| www.frontiersin.org Application of sEMG in Controlling HRC Technologies Human-robot collaborative technologies, particularly exoskeletons, have proven signiﬁcantly useful in the rehabilitation programs of patients aﬀected by neurological diseases. Physiological parameters with sEMG play a key role in monitoring muscle activation amplitude and fatigue in the design of innovative active exoskeleton controller systems and assessment of their eﬀectiveness on motor performance (71). p g p Wearable sensors do not interfere with the typical movements performed by workers owing to their miniaturization and wireless communication protocols. In addition, multi-channel sEMG systems are available in wireless versions despite their high number of channels and high data rate. The combination of sensor networks and intelligent work environments provides real-time estimation of physiological parameters, enabling direct feedback to workers who are monitored directly and constantly at the workplace. Real-time monitoring is additionally useful for providing acoustic, visual, and vibro-tactile stimuli to workers (53, 88) executing manual handling tasks in awkward postures or requiring signiﬁcant physical eﬀort, or when muscle fatigue sets in. Recent improvements in the measurement and real-time classiﬁcation of myoelectric signals have already facilitated the use of sEMG in man-machine interfaces for controlling prostheses and orthoses among other devices (72–76). Very recently, human-exoskeleton interfaces were developed for the purpose of rehabilitation to support physically weak and disabled people in performing several motor activities of daily living, such as walking. These interfaces were developed based on new technical and mathematical approaches, including new sEMG signal-processing procedures (77–82). The performance of these man-machine interfaces is quickly improving owing to neuro-musculoskeletal models driven December 2020 \| Volume 11 \| Article 572069 4 sEMG in Return-To-Work Rehabilitation Ranavolo et al. Ranavolo et al. is evidence of poor knowledge about adaptation of workers with neurological pathologies. As regard the sEMG use in work environment, each manual handling activity represent a risk of onset of well-classiﬁed diseases aﬀecting the musculoskeletal system. For instance, lifting activities imply high compression and shear forces at L5-S1 joints (strongly correlated with muscle co-activation) with a signiﬁcant involvement of erector spinae and rectus abdominus muscles, while handling low loads at high frequency imply neck and upper limbs muscles fatigue. For this reason the choice of channels and the rationale for selecting them should be task-guided. The lack of quantitative approaches suggests that, except in rare cases, sEMG is not taught in the Ph.D. programs of universities and occupational medicine specialization schools. DISCUSSION Although sEMG is considered the most informative instrument for muscle monitoring when wearable robots are used, its use poses a number of challenges. One such challenge, which we believe is on the verge of being addressed, pertains to the use of exoskeletons for active rehabilitation therapies. It pertains to the optimization of appropriate smart algorithms to detect patients’ intentions and allow exoskeletons to act in synergy with them (93, 94). The challenge is to enable symbiotic physical HRC by incorporating accurate subject-speciﬁc computer models of each individual’s neuro-musculoskeletal system to enable appropriate anticipation mechanisms. This is crucial to estimate muscle and joint stiﬀness accurately to determine the onset of excessive rigidity, which may be related to fatigue or negative compensatory strategies (95). Without doubt, in job integration/reintegration, the sEMG approach has yet to be adopted and the critical issues associated with it are managed with diﬃculty. In fact, while reasonable workplace accommodation and disability employment issues are being historically and widely addressed by the governments of the industrialized world, the adoption of instrument-based quantitative assessments of ergonomic interventions has so far been disregarded, owing to cultural barriers, which lead to a preference for qualitative approaches. Myocontrol should be increasingly based on HDsEMG to increase spatial resolution with respect to low-density sEMG and to improve the accuracy of motor workers’ intentions recognition. Moreover, machine learning approaches such as artiﬁcial neural networks should be used to evaluate the capacity of workers with neurological diseases for myocontrol. A clear testimony of the presence of this educational gap is the worldwide social policies for persons with disabilities (90, 91), such as the European Disability Strategy (2010– 2020) and the directives 89/654/EEC, 2000/78/EC, and 2000/78/EC are the corresponding policies. Another example is that of the Job Accommodation Network, a facility of the United States Department of Labor’s Oﬃce of Disability Employment Policy, which provides a valuable strategy for the inclusion of people with neurological disabilities (92). Nevertheless, only half of the population with disabilities has been accommodated well in terms of workplace design and there In a recent study (96), HDsEMG was used to test the ability of participants with Duchenne muscular dystrophy (DMD) to produce repeatable HDsEMG patterns, which were unexpectedly comparable with those of healthy participants, suggesting a clear potential for the myocontrol of wearable exoskeletons. Application of sEMG in Controlling HRC Technologies Furthermore, the domestic and international chapters of the Human Factors and Ergonomics Society do not actively promote meetings, conferences, and events to address the challenges of sEMG use. Finally, the reference society for sEMG, the International Society of Electrophysiology and Kinesiology, has thus far not dedicated speciﬁc training programs regarding job accommodation. All the above mentioned training activities would typically be successful to address methodological issues, as increased knowledge alone cannot overcome technical limitations. g g The criticisms related to the use of sEMG to control collaborative wearable trunk and upper limb devices designed to assist people with neurological disabilities are attributable to the algorithms used in human-robot interfaces. These algorithms are used for pattern recognition and classiﬁcation of patients’ movement intentions. Only a few years ago, the performance of sEMG-based interfaces had not reached accuracies acceptable for widespread commercial use (74). Accuracy was limited by the high inter-subject variability, which required subjective calibrations and training. Fortunately, these interfaces, including machine-learning algorithms, have since been signiﬁcantly improved, enabling the acceptable optimization of HRC control mostly for people with severe upper and lower limb disabilities (89). The enormous potential of the sEMG and HDsEMG approaches and their very limited use in return-to-work programs represent a real paradox. It is diﬃcult to determine why sEMG is underused. Perhaps, the most likely reason is that the training provided to professionals is based on more qualitative rather than quantitative approaches, and the transition from one approach to the other is evidently diﬃcult. Professionals in the ﬁeld have to be trained to understand the extremely variable abnormalities of workers suﬀering from neurological pathologies and to associate an appropriate return-to-work plan with them. Another technical limitation of sEMG-based interfaces for use in HRC technology control is the fact that most wearable assistive devices use traditional control tools, such as bipolar sEMG, to record antagonist muscle activities. This low spatial sampling implies that a maximum of one degree of freedom (DoF) can be controlled. Furthermore, managing up to two DoFs requires slow, sequential, and unintuitive control. The related limited functionality in conjunction with the extensive training required of neurological subjects led to the high rejection rates of these technologies (72). Currently, classiﬁcation and regression approaches outperform traditional control tools in controlling complex motor activities in terms of speed and accuracy, providing a promising method for advanced myoelectric control (72). Frontiers in Neurology \| www.frontiersin.org December 2020 \| Volume 11 \| Article 572069 DISCUSSION There are other challenges that are speciﬁc to application of sEMG in return-to-work environment according to the speciﬁc motor impairment characterizing the diﬀerent neurological diseases. For instance, muscle fatigue, which is a common feature of several neurological diseases (i.e., multiple sclerosis, stroke, muscle dystrophy) results in altered motor unit recruitment and decreased maximal voluntary motor unit ﬁring rate that can be detected by sEMG monitoring. A speciﬁc scenario could be to adapt the workplace and to modify the work-task according to the subjects’ abnormal fatiguing performance, by assisting the workers with devices and/or reducing the amount and the duration of the work-related activity. Furthermore, an individualized rehabilitative program could be planned to improve the impact of the work-task on the fatigability in a long-term period. pp These mandatory educational opportunities must capitalize on the skills of the leaders and innovators of sEMG to serve physiotherapists, health and safety technicians, and ergonomists by providing them with qualiﬁed training focused on the management of the monitoring and HRC technologies and instrumental recordings. In particular, these professionals should know the physiology of sEMG signals, electrodes placement, software and hardware for acquisition. Furthermore, they are expected to know basic and some more complex concepts of signal processing, do a general visual interpretation and be able to generate reports that can be interpreted by the other team members. The multidisciplinary team should plan, implement and evaluate the return-to-work program in both the clinical and occupational environment. It should be considered that when patients with neurological disorders are involved, the key professionals leading the team should be neurologists in the health sector and occupational physicians at workplace. INAIL is organized throughout the national territory in order to fully manage the job integration/reintegration process but this service is also oﬀered by consultant companies. Rehabilitation engineers should be given greater options to work with patients and physiotherapists, health and safety technicians end ergonomists. We believe that no other hybrid professions should be created. The contents must be based ﬁrst on elementary concepts [which are illustrated in free online materials, such as those available at [99]] and, second, on integrated approaches promoting the culture and acceptance of instrument-based quantitative methodologies. All these actions should be taken as early as possible by guiding workers with chronic neurological disorders to return to work and stay in work with well- managed occupational safety and health interventions. DISCUSSION High- density sEMG can be applied to analyze the altered motor control of people with DMD and potentially interface them with assistive wearable robots. In addition, non-invasive decoding of individual December 2020 \| Volume 11 \| Article 572069 5 sEMG in Return-To-Work Rehabilitation Ranavolo et al. abovementioned activities may additionally yield tangible savings for businesses and national health and welfare systems. α-motor neuron activation may represent a new option for the design of real-time closed-loop control applications, such as transcutaneous and epidural electrical stimulations. Obviously, the teaching activities directed to strategic professions alone cannot solve the problem of under-use of sEMG in return-to-work plans. In fact, they must be accompanied by further electrodes and instrumentations development especially from the perspective of emerging artiﬁcial intelligence that may be encapsulate sEMG knowledge. Such tools may provide the professional with information to act on, thereby reducing the current “art” aspect of using EMG to something more in everyone’s hands to use and exploit. To bridge the gap between the return-to-work concept and other disciplines, several educational activities should be developed to enhance and apply the best skills available in rehabilitation engineering, physiotherapy, occupational therapy, and ergonomics. For a few years, INAIL has been promoting a Ph.D. program together with the Sapienza University of Roma titled “Kinematic, Kinetic, and Electromyographic Characterization of Motor Disabilities and Biomechanical Overload Risk Management for Job Reintegration,” a Masters’ course titled “New Methodologies for the Evaluation and Management of Biomechanical Risk and Criteria and Methods for the Adaptation of Workplaces,” and several training courses regarding the role of sEMG in occupational medicine and ergonomics. The participation, although the events were not free and not recognized as continuing education, has always been conspicuous with a number of participants (occupational health and safety technicians, physiotherapists, ergonomists, occupational physicians, rehabilitation engineers and movement scientists) that has always reached the maximum allowed limit (30 in the case of training courses). The authors of this article begin to observe a ﬁrst positive impact of these initiatives on the need that operators in the ﬁeld have in using the sEMG approach. In addition to the motor impairments, cognitive and speech impairments are major contributors that strongly impact on the returning to work. Important aspects to consider are to understand how patients with cognitive and speech problems can adapt to the sEMG monitoring and how they can be assisted by using speciﬁc sEMG technology (96). Frontiers in Neurology \| www.frontiersin.org REFERENCES (1990) 69:1810–20. doi: 10.1152/jappl.1990.69.5.1810 6. Merletti R, Knaﬂitz M, De Luca CJ. Myoelectric manifestations of fatigue in voluntary and electrically elicited contractions. J Appl Physiol. (1990) 69:1810–20. doi: 10.1152/jappl.1990.69.5.1810 19. Martino G, Ivanenko Y, Serrao M, Ranavolo A, Draicchio F, Casali C, et al. Locomotor coordination in patients with Hereditary Spastic Paraplegia. J Electromyogr Kinesiol. (2019) 45:61–9. doi: 10.1016/j.jelekin.2019. 02.006 7. Torres-Oviedo G, Ting LH. Muscle synergies characterizing human postural responses. J Neurophysiol. (2007) 98:2144–56. doi: 10.1152/jn.01360.2006 8. Varrecchia T, Rinaldi M, Serrao M, Draicchio F, Conte C, Conforto S, et al. Global lower limb muscle coactivation during walking at diﬀerent speeds: Relationship between spatio-temporal, kinematic, kinetic, and energetic parameters. J Electromyogr Kinesiol. (2018);43:148–57. doi: 10.1016/j.jelekin.2018.09.012 20. Martino G, Ivanenko Y, Serrao M, Ranavolo A, Draicchio F, Rinaldi M, et al. Diﬀerential changes in the spinal segmental locomotor output in Hereditary Spastic Paraplegia. Clin Neurophysiol. (2018) 129:516–25. doi: 10.1016/j.clinph.2017.11.028 21. Martino G, Ivanenko YP, d’Avella A, Serrao M, Ranavolo A, Draicchio F, et al. Neuromuscular adjustments of gait associated with unstable conditions. J Neurophysiol. (2015) 114:2867–82. doi: 10.1152/jn.00029.2015 9. Rinaldi M, Ranavolo A, Conforto S, Martino G, Draicchio F, Conte C, et al. Increased lower limb muscle coactivation reduces gait performance and increases metabolic cost in patients with hereditary spastic paraparesis. Clin Biomech (Bristol, Avon). (2017) 48:63–72. doi: 10.1016/j.clinbiomech.2017.07.013 22. Martino G, Ivanenko YP, Serrao M, A Ranavolo A, d’Avella A, Draicchio F, et al. Locomotor patterns in cerebellar ataxia. J Neurophysiol. (2014) 112:2810– 21. doi: 10.1152/jn.00275.2014 10. Rosa MC, Marques A, Demain S, Metcalf CD, Rodrigues J. Methodologies to assess muscle co-contraction during gait in people with neurological impairment—a systematic literature review. J Electromyogr Kinesiol. (2014) 24:179–91. doi: 10.1016/j.jelekin.2013.11.003 23. Serrao M, Rinaldi M, Ranavolo A, Lacquaniti F, Martino G, Leonardi L, et al. Gait patterns in patients with hereditary spastic paraparesis. PLoS ONE. (2016) 11:e0164623. doi: 10.1371/journal.pone.0164623 24. Lacquaniti F, Ivanenko YP, d’Avella A, Zelik KE, Zago M. Evolutionary and developmental modules. Front Comput Neurosci. (2013) 7:61. doi: 10.3389/fncom.2013.00061 11. Tatarelli A, Serrao M, Varrecchia T, Fiori L, Draicchio F, Silvetti A, et al. Global muscle coactivation of the sound limb in gait of people with transfemoral and transtibial amputation. Sensors. (2020) 20:E2543. doi: 10.3390/s20092543 12. Farina D, Merletti R. A novel approach for estimating muscle ﬁber conduction velocity by spatial and temporal ﬁltering of surface EMG signals. IEEE Trans Biomed Eng. (2003) 50:1340–51. doi: 10.1109/TBME.2003.819847 25. DISCUSSION The The abnormal muscle co-activation is another example of common problem identiﬁable by sEMG in several neurological disorders (e.g., cerebellar ataxia, Parkinson disease, multiple sclerosis, stroke) which can be detected by measuring the simultaneous time-varying sEMG signal in many muscles. It is known that patients with balance disorders increase the muscle co-activation to control their walking instability in the attempt to stiﬀen the body segments. Unfortunately, this compensatory mechanism has some negative eﬀects, such as increased metabolic cost and risk of cartilage degeneration. A speciﬁc scenario for these workers is to plan an appropriate workplace rehabilitation to improve their balance, to stabilize the body segments and to reduce the need to increase the muscle co-activation. In conclusion, sEMG should be used in job integration plans to: - classify the nature and degree of residual motor function in order to design/adapt the workplace; - assess the eﬃcacy of work-task rehabilitation and ergonomic interventions; g - control new assistive technologies such as collaborative robots; - evaluate the biomechanical risk during the execution of manual handling activities; - plan a preventive rehabilitation program to prevent injury. December 2020 \| Volume 11 \| Article 572069 6 Ranavolo et al. sEMG in Return-To-Work Rehabilitation DATA AVAILABILITY STATEMENT The research presented in this article was carried out as part of the SOPHIA project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 871237. The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. ETHICS STATEMENT Including the sEMG approach in work integration/reintegration oﬀers the possibility of designing programs based on the residual motor abilities of the worker and adapting his/her workplace. This allows to consider a wide range of workers with several neurological pathologies and diﬀerent levels of severity but without a complete inability to perform activities of daily life. Ethical review and approval and written informed consent were not required for the study as per local legislation and institutional requirements. AUTHOR CONTRIBUTIONS To make this possible, Bachelor, Master and PhD programs should be promoted, or at least, supervised and monitored, by scientiﬁc societies in the ﬁelds of physiotherapy, ergonomics, occupational medicine, biomechanics, electrophysiology and kinesiology, and should include continuing education courses on the use of sEMG speciﬁcally oriented to teachers in these ﬁelds. AR, MS, and FD planned the manuscript, performed the literature searches, wrote the text and revised the manuscript. All authors contributed to the article and approved the submitted version. REFERENCES 13. Farina D, Merletti R. Methods for estimating muscle ﬁbre conduction velocity from surface electromyographic signals. Med Biol Eng Comput. (2004) 42:432– 45. doi: 10.1007/BF02350984 1. Merletti R, Botter A, Troiano A, Merlo E, Minetto MA. Technology and instrumentation for detection and conditioning of the surface electromyographic signal: state of the art. Clin Biomech. (2009) 24:122–34. doi: 10.1016/j.clinbiomech.2008.08.006 1. Merletti R, Botter A, Troiano A, Merlo E, Minetto MA. Technology and instrumentation for detection and conditioning of the surface electromyographic signal: state of the art. Clin Biomech. (2009) 24:122–34. doi: 10.1016/j.clinbiomech.2008.08.006 14. Cifrek M, Medved V, Tonkovi´c S, Ostoji´c S. Surface EMG based muscle fatigue evaluation in biomechanics. Clin Biomech. (2009) 24:327–40. doi: 10.1016/j.clinbiomech.2009.01.010 2. Farina D, Negro F, Muceli S, Enoka RM. Principles of Motor Unit Physiology Evolve With Advances in Technology. Physiology (Bethesda). (2016) 31:83–94. doi: 10.1152/physiol.00040.2015 15. Merletti R, Farina D. Surface Electromyography: Physiology, Engineering, and Applications. John Wiley & Sons (2016). 2. Farina D, Negro F, Muceli S, Enoka RM. Principles of Motor Unit Physiology Evolve With Advances in Technology. Physiology (Bethesda). (2016) 31:83–94. doi: 10.1152/physiol.00040.2015 16. Merletti R, Parker P. Electromyography—Physiology, Engineering and Noninvasive Applications. 1st ed. Hoboken, New Jersey: John Wiley & Sons, Inc. (2004). 3. Merletti R, Farina D, Gazzoni M. The linear electrode array: a useful tool with many applications. J Electromyogr Kinesiol. (2003) 13:37–47. doi: 10.1016/s1050-6411(02)00082-2 3. Merletti R, Farina D, Gazzoni M. The linear electrode array: a useful tool with many applications. J Electromyogr Kinesiol. (2003) 13:37–47. doi: 10.1016/s1050-6411(02)00082-2 17. Janshen L, Santuz A, Ekizos A, Arampatzis A. Fuzziness of muscle synergies in patients with multiple sclerosis indicates increased robustness of motor control during walking. Sci Rep. (2020) 10:7249. doi: 10.1038/s41598-020-63788-w 4. Zwarts MJ, Stegeman DF. Multichannel surface EMG: basic aspects and clinical utility. Muscle Nerve. (2003) 28:1–17. doi: 10.1002/mus.10358 4. Zwarts MJ, Stegeman DF. Multichannel surface EMG: basic aspects and clinical utility. Muscle Nerve. (2003) 28:1–17. doi: 10.1002/mus.10358 5. Hägg GM, Luttmann A, Jäger M. Methodologies for evaluating electromyographic ﬁeld data in ergonomics. J Electromyogr Kinesiol. (2000) 10:301–12. doi: 10.1016/s1050-6411(00)00022-5 18. De Marchis C, Ranaldi S, Serrao M, Ranavolo A, Draicchio F, Lacquaniti F, et al. Modular motor control of the sound limb in gait of people with trans-femoral amputation. J Neuroeng Rehabil. (2019) 16:132. doi: 10.1186/s12984-019-0616-7 6. Merletti R, Knaﬂitz M, De Luca CJ. Myoelectric manifestations of fatigue in voluntary and electrically elicited contractions. J Appl Physiol. REFERENCES J Occup Rehabil. (2019) 29:175–93. doi: 10.1007/s10926-018-9777-7 32. Ranavolo A, Mari S, Conte C, Serrao M, Silvetti A, Iavicoli S, et al. A new muscle co-activation index for biomechanical load evaluation in work activities. Ergonomics. (2015) 58:966–79. doi: 10.1080/00140139.2014.991764 53. Ajoudani A, Albrecht P, Bianchi M, Cherubini A, Del Ferraro S, Fraisse P, et al. Smart collaborative systems for enabling ﬂexible and ergonomic work practices. IEEE Robot Autom Mag. (2020) 27:169–76. doi: 10.1109/MRA.2020.2985344 33. Ranavolo A, Varrecchia T, Rinaldi M, Silvetti A, Serrao M, Conforto S, et al. Mechanical lifting energy consumption in work activities designed by means of the “revised NIOSH lifting equation.” Ind Health. (2017) 55:444–54. doi: 10.2486/indhealth.2017-0075 54. Vinci P, Serrao M, Millul A, Deidda A, De Santis F, Capici S, et al. Quality of life in patients with charcot-marie-tooth disease. Neurology. (2005) 65:922–4. doi: 10.1212/01.wnl.0000176062.44360.49 34. Ranavolo A, Draicchio F, Varrecchia T, Silvetti A, Iavicoli S. Erratum, Alberto R. et al. Wearable monitoring devices for biomechanical risk assessment at work: current status and future challenges—a systematic review. Int J Environ Res Public Health. (2018) 15:2569. doi: 10.3390/ijerph15112569 55. Doogan C, Playford ED. Supporting work for people with multiple sclerosis. Mult Scler. (2014) 20:646–50. doi: 10.1177/1352458514523499 35. Ranavolo A, Varrecchia T, Iavicoli S, Marchesi A, Rinaldi M, Serrao M, et al. “Surface electromyography for risk assessment in work activities designed using the “revised NIOSH lifting equation””. Int J Ind Ergon. (2018) 68:34– 45.doi: 10.1016/j.ergon.2018.06.003 56. Raggi A, Covelli V, Schiavolin S, Scaratti C, Leonardi M, Willems M. Work-related problems in multiple sclerosis: a literature review on its associates and determinants. Disabil Rehabil. (2016) 38:936–44. doi: 10.3109/09638288.2015.1070295 57. Hendricks DJ, Sampson E. Accommodating individuals with traumatic brain injury: an analysis of employer-initiated cases handled by the Job Accommodation Network. Work. (2017) 58:29–4. doi: 10.3233/WOR-172594 36. Ranavolo A, Chini G, Silvetti A, Mari S, Serrao M, Draicchio F. Myoelectric manifestation of muscle fatigue in repetitive work detected by means of miniaturized sEMG sensors. Int J Occup Saf Ergon. (2018) 24:464–74. doi: 10.1080/10803548.2017.1357867 58. van der Kemp J, Kruithof WJ, Nijboer TCW, van Bennekom CAM, van Heugten C, Visser-Meily JMA. Return to work after mild-to-moderate stroke: work satisfaction and predictive factors. Neuropsychol Rehabil. (2019) 29:638– 53. doi: 10.1080/09602011.2017.1313746 37. Alberto R, Draicchio F, Varrecchia T, Silvetti A, Iavicoli S. Wearable monitoring devices for biomechanical risk assessment at work: current status and future challenges–a systematic review. REFERENCES Int J Environ Res Public Health. (2018) 15:2001. doi: 10.3390/ijerph15092001 59. Wimalaratna HS, Tooley MA, Churchill E, Preece AW, Morgan HM. Quantitative surface EMG in the diagnosis of neuromuscular disorders. Electromyogr Clin Neurophysiol. (2002) 42:167–74. 38. Varrecchia T, De Marchis C, Draicchio F, Schmid M, Conforto S, Ranavolo A. Lifting activity assessment using kinematic features and neural networks. Appl Sci. (2020) 10:1989.doi: 10.3390/app10061989 60. Labarre-Vila A. Electromyographie de surface et fonction musculaire en pathologie [Assessment of muscle function in pathology with surface electrode EMG]. Rev Neurol. (2006) 162:459–65. doi: 10.1016/s0035-3787(06)75037-8 39. Varrecchia T, De Marchis C, Rinaldi M, Draicchio F, Serrao M, Schmid M, et al. Lifting activity assessment using surface electromyographic features and neural networks. Int J Ind Ergon. (2018) 66:1–9.doi: 10.1016/j.ergon.2018.02.003 61. Mari S, Serrao M, Casali C, Conte C, Martino G, Ranavolo A, et al. Lower limb antagonist muscle co-activation and its relationship with gait parameters in cerebellar ataxia. Cerebellum. (2014) 13:226–36. doi: 10.1007/s12311-013-0533-4 40. Klockgether T. Sporadic ataxia with adult onset: classiﬁcation and diagnostic criteria. Lancet Neurol. (2010) 9:94–104. doi: 10.1016/S1474-4422(09) 70305-9 62. Latash ML. Muscle coactivation: deﬁnitions, mechanisms, and functions. J Neurophysiol. (2018) 120:88–104. doi: 10.1152/jn.00084.2018 41. Miller A, Dishon S. Health-related quality of life in multiple sclerosis: The impact of disability, gender and employment status. Qual Life Res. (2006) 15:259–71. doi: 10.1007/s11136-005-0891-6 63. Castagneri C, Agostini V, Rosati S, Balestra G, Knaﬂitz M. Asymmetry index in muscle activations. IEEE Trans Neural Syst Rehabil Eng. (2019) 27:772–9. doi: 10.1109/TNSRE.2019.2903687 42. Ranavolo A, Serrao M, Varrecchia T, Casali C, Filla A, Roca A, Silvetti A, et al. The Working Life of People with Degenerative Cerebellar Ataxia. Cerebellum. (2019) 18:910–21. doi: 10.1007/s12311-019-01065-x 64. Becker S, von Werder SCFA, Lassek AK, Disselhorst-Klug C. Time-frequency coherence of categorized sEMG data during dynamic contractions of biceps, triceps, and brachioradialis as an approach for spasticity detection. Med Biol Eng Comput. (2019) 57:703–13. doi: 10.1007/s11517-018-1911-3 43. Toldrá RC, Santos MC. People with disabilities in the labor market: facilitators and barriers. Work. (2013) 45:553–63. doi: 10.3233/WOR-131641 44. Honan CA, Brown RF, Hine DW, Vowels L, Wollin JA, Simmons RD, et al. The multiple sclerosis work diﬃculties questionnaire. Mult Scler. (2012) 18:871–80. doi: 10.1177/1352458511431724 65. Morrison S, Kavanagh J, Obst SJ, Irwin J, Haseler LJ. The eﬀects of unilateral muscle fatigue on bilateral physiological tremor. Exp Brain Res. (2005) 167:609–21. doi: 10.1007/s00221-005-0050-x 45. Karamians R, Proﬃtt R, Kline D, Gauthier LV. REFERENCES d’Avella A, Lacquaniti F. Control of reaching movements by muscle synergy combinations. Front Comput Neurosci. (2013) 7:42. doi: 10.3389/fncom.2013.00042 December 2020 \| Volume 11 \| Article 572069 Frontiers in Neurology \| www.frontiersin.org 7 Ranavolo et al. sEMG in Return-To-Work Rehabilitation 26. Ivanenko YP, Cappellini G, Dominici N, Poppele RE, Lacquaniti F. Coordination of locomotion with voluntary movements in humans. J Neurosci. (2005) 25:7238–53. doi: 10.1523/JNEUROSCI.1327-05.2005 46. Triegaardt J, Han TS, Sada C, Sharma S, Sharma P. The role of virtual reality on outcomes in rehabilitation of Parkinson’s disease: meta-analysis and systematic review in 1031 participants. Neurol Sci. (2020) 41:529–36. doi: 10.1007/s10072-019-04144-3 27. Ivanenko YP, Cappellini G, Solopova IA, Grishin AA, Maclellan MJ, Poppele RE, et al. Plasticity and modular control of locomotor patterns in neurological disorders with motor deﬁcits. Front Comput Neurosci. (2013) 7:123. doi: 10.3389/fncom.2013.00123 47. Naro A, Leo A, Russo M, Casella C, Buda A, Crespantini A, et al. Breakthroughs in the spasticity management: Are non- pharmacological treatments the future? J Clin Neurosci. (2017) 39:16–27. doi: 10.1016/j.jocn.2017.02.044 28. Janshen L, Santuz A, Ekizos A, Arampatzis A. Modular control during incline and level walking in humans. J Exp Biol. (2017) 220(Pt 5):807–13. doi: 10.1242/jeb.148957 48. Iosa M, Morone G, Cherubini A, Paolucci S. The three laws of neurorobotics: a review on what neurorehabilitation robots should do for patients and clinicians. J Med Biol Eng. (2016) 36:1–11. doi: 10.1007/s40846-016-0115-2 29. Santuz A, Ekizos A, Janshen L, Mersmann F, Bohm S, Baltzopoulos V, et al. Modular control of human movement during running: an open access data set. Front Physiol. (2018) 9:1509. doi: 10.3389/fphys.2018.01509 49. Lewis PM, Thomson RH, Rosenfeld JV, Fitzgerald PB. Brain neuromodulation techniques: a review. Neuroscientist. (2016) 22:406–21. doi: 10.1177/1073858416646707 30. Ivanenko YP, Poppele RE, Lacquaniti F. Five basic muscle activation patterns account for muscle activity during human locomotion. J Physiol. (2004) 556(Pt 1):267–82. doi: 10.1113/jphysiol.2003.057174 50. Ashley KD, Lee LT, Heaton K. Return to work among stroke survivors. Workplace Health Saf. (2019) 67:87–94. doi: 10.1177/2165079918812483 31. Negro F, Muceli S, Castronovo AM, Holobar A, Farina D. Multi-channel intramuscular and surface EMG decomposition by convolutive blind source separation. J Neural Eng. (2016) 13:026027. doi: 10.1088/1741-2560/13/2/026027 51. Skamagki G, King A, Duncan M, Wåhlin C. A systematic review on workplace interventions to manage chronic musculoskeletal conditions. Physiother Res Int. (2018) 23:e1738. doi: 10.1002/pri.1738 52. Hoosain M, de Klerk S, Burger M. Workplace-based rehabilitation of upper limb conditions: a systematic review. REFERENCES Eﬀectiveness of virtual reality- and gaming-based interventions for upper extremity rehabilitation poststroke: a meta-analysis. Arch Phys Med Rehabil. (2020) 101:885–96. doi: 10.1016/j.apmr.2019.10.195 66. Menotti F, Bazzucchi I, Felici F, Damiani A, Gori MC, Macaluso A. Neuromuscular function after muscle fatigue in Charcot-Marie-Tooth type 1A patients. Muscle Nerve. (2012) 46:434–9. doi: 10.1002/mus.23366 December 2020 \| Volume 11 \| Article 572069 Frontiers in Neurology \| www.frontiersin.org 8 Ranavolo et al. sEMG in Return-To-Work Rehabilitation 67. Bodranghien F, Bastian A, Casali C, Hallett M, Louis ED, Manto M, et al. Consensus paper: revisiting the symptoms and signs of cerebellar syndrome. Cerebellum. (2016) 15:369–91. doi: 10.1007/s12311-015-0687-3 84. Barbero M, Merletti R, Rainoldi A. Atlas of Muscle Innervation Zones: Understanding Surface Electromyography and its Applications. New York, NY: Springer (2012). 68. Fiori L, Ranavolo A, Varrecchia T, Tatarelli A, Conte C, Draicchio F, et al. Impairment of global lower limb muscle coactivation during walking in cerebellar ataxias. Cerebellum. (2020) 19:583–96. doi: 10.1007/s12311-020-01142-6 85. Besomi M, Hodges PW, Van Dieën J, Carson RG, Clancy EA, Disselhorst- Klug, et al. Consensus for experimental design in electromyography (CEDE) project: electrode selection matrix. J Electromyogr Kinesiol. (2019) 48:128–44. doi: 10.1016/j.jelekin.2019.07.008 j j 86. Merletti R, Muceli S. Tutorial. Surface EMG detection in space and time: best practices. J Electromyogr Kinesiol. (2019) 49:102363. doi: 10.1016/j.jelekin.2019.102363 69. Serrao M, Casali C, Ranavolo A, Mari S, Conte C, Chini G, et al. Use of dynamic movement orthoses to improve gait stability and trunk control in ataxic patients. Eur J Phys Rehabil Med. (2017) 53:735–43. doi: 10.23736/S1973-9087.17.04480-X 87. Merletti R, Cerone GL. Tutorial. Surface EMG detection, conditioning and pre-processing: best practices. J Electromyogr Kinesiol. (2020) 54:102440. doi: 10.1016/j.jelekin.2020.102440 70. Ivanenko YP, Poppele RE, Lacquaniti F. Spinal cord maps of spatiotemporal alpha-motoneuron activation in humans walking at diﬀerent speeds. J Neurophysiol. (2006) 95:602–18. doi: 10.1152/jn.00767.2005 88. Ajoudani A, Godfrey SB, Bianchi M, Catalano MG, Grioli G, Tsagarakis N, et al. Exploring teleimpedance and tactile feedback for intuitive control of the pisa/iit softhand. IEEE Trans Haptics. (2014) 7:203–15. doi: 10.1109/TOH.2014.2309142 71. de Looze MP, Bosch T, Krause F, Stadler KS, O’Sullivan LW. Exoskeletons for industrial application and their potential eﬀects on physical work load. Ergonomics. (2016) 59:671–81. doi: 10.1080/00140139.2015.1081988 72. Kapelner T, Sartori M, Negro F, Farina D. Neuro-musculoskeletal mapping for man-machine interfacing. Sci Rep. (2020) 10:5834. doi: 10.1038/s41598-020-62773-7 89. Gao B, Wei C, Ma H, Yang S, Ma X, Zhang S. REFERENCES Real-time evaluation of the signal processing of sEMG used in limb exoskeleton rehabilitation system. Appl Bionics Biomech. (2018) 2018:1391032. doi: 10.1155/2018/1391032 90. Krahn GL. WHO World report on disability: a review. Disabil Health J. (2011) 4:141–2. doi: 10.1016/j.dhjo.2011.05.001 73. Kapelner T, Vujaklija I, Jiang N, Negro F, Aszmann OC, Principe J, et al. Predicting wrist kinematics from motor unit discharge timings for the control of active prostheses. J Neuroeng Rehabil. (2019) 16:47. doi: 10.1186/s12984-019-0516-x 91. de Jong PR. Sickness, disability and work: breaking the barriers—a synthesis of ﬁndings across OECD countries—OECD. Int Rev für Soz Sicherheit. (2011) 64:115–7. doi: 10.1111/j.1752-1726.2011.01406.x\| 74. Lobov S, Krilova N, Kastalskiy I, Kazantsev V, Makarov VA. Latent factors limiting the performance of sEMG-interfaces. Sensors. (2018) 18:1122. doi: 10.3390/s18041122 92. Copaci D, Serrano D, Moreno L, Blanco D. A high-level control algorithm based on sEMG signalling for an elbow joint SMA exoskeleton. Sensors. (2018) 18:2522. doi: 10.3390/s18082522 75. Roche AD, Rehbaum H, Farina D, Aszmann OC. Prosthetic myoelectric control strategies: a clinical perspective. Curr Surg Rep. (2014) 2:44. doi: 10.1007/s40137-013-0044-8 93. Suplino LO, Sommer LF, Forner-Cordero A. EMG-based control in a test platform for exoskeleton with one degree of freedom. Conf Proc IEEE Eng Med Biol Soc. (2019) 5366–9. doi: 10.1109/EMBC.2019.8856836 76. Hahne JM, Biessmann F, Jiang N, Rehbaum H, Farina D, Meinecke FC, et al. Linear and nonlinear regression techniques for simultaneous and proportional myoelectric control. IEEE Trans Neural Syst Rehabil Eng. (2014) 22:269–79. doi: 10.1109/TNSRE.2014.2305520 94. Sartori M, Maculan M, Pizzolato C, Reggiani M, Farina D. Modeling and simulating the neuromuscular mechanisms regulating ankle and knee joint stiﬀness during human locomotion. J Neurophysiol. (2015) 114:2509–27. doi: 10.1152/jn.00989.2014 77. Gordon KE, Kinnaird CR, Ferris DP. Locomotor adaptation to a soleus EMG-controlled antagonistic exoskeleton. J Neurophysiol. (2013) 109:1804– 14. doi: 10.1152/jn.01128.2011 95. Nizamis K, Rijken NHM, van Middelaar R, Neto J, Koopman BFJM, Sartori M. Characterization of forearm muscle activation in duchenne muscular dystrophy via high-density electromyography: a case study on the implications for myoelectric control. Front Neurol. (2020) 11:231. doi: 10.3389/fneur.2020.00231 78. Kiguchi K, Hayashi Y. An EMG-based control for an upper-limb power-assist exoskeleton robot. IEEE Trans Syst Man Cybern B Cybern. (2012) 42:1064–71. doi: 10.1109/TSMCB.2012.2185843 79. Yagi E, Harada D, Kobayashi M. Upper-limb power assist control for agriculture load lifting. Int J Autom Technol. (2009) 3:716–22. doi: 10.20965/ijat.2009p0716 96. Janke M, Diener L. EMG-to-speech: direct generation of speech from facial electromyographic signals. Frontiers in Neurology \| www.frontiersin.org REFERENCES IEEE/ACM Trans Audio Speech Lang Proc. (2017) 12:2375–85. doi: 10.1109/TASLP.2017.27 38568 80. Su H, Li Z, Li G, Yang C. EMG-based neural network control of an upper-limb power-assist exoskeleton robot. In: Guo C, Hou ZG, Zeng Z, editors. Advances in Neural Networks—ISNN 2013. Berlin, Heidelberg: Springer (2013). Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 81. Khokhar ZO, Xiao ZG, Menon C. Surface EMG pattern recognition for real-time control of a wrist exoskeleton. Biomed Eng Online. (2010) 9:41. doi: 10.1186/1475-925X-9-41 82. Gopura RARC, Kiguchi K, Li Y. SUEFUL-7: a 7DOF upper-limb exoskeleton robot with muscle-model-oriented EMG-based control. In: IEEE/RSJ International Conference on Intelligent Robots and Systems. St. Louis, MO (2009). Copyright © 2020 Ranavolo, Serrao and Draicchio. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 83. Hermens HJ, Freriks B. European Recommendations for Surface Electromyography: Results of the SENIAM Project. Enschede: Roessingh Research and Development (1999). December 2020 \| Volume 11 \| Article 572069 Frontiers in Neurology \| www.frontiersin.org 9
https://openalex.org/W3091264322	https://link.springer.com/content/pdf/10.1007%2F978-3-030-42472-5_9.pdf	English	null	European Mortality Forecasts: Are the Targets Still Moving?	The Springer series on demographic methods and population analysis	2,020	cc-by	5,994	Chapter 9 European Mortality Forecasts: Are the Targets Still Moving? Nico Keilman and Sigve Kristoffersen N. Keilman () · S. Kristoffersen Department of Economics, University of Oslo, Oslo, Norway e-mail: nico.keilman@econ.uio.no; sigvekr@student.ikos.uio.no © The Author(s) 2020 S. Mazzuco, N. Keilman (eds.), Developments in Demographic Forecasting, The Springer Series on Demographic Methods and Population Analysis 49, https://doi.org/10.1007/978-3-030-42472-5_9 9.1 Introduction and Problem Formulation Many statistical agencies routinely produce population forecasts, and revise these forecasts when new data become available, or when current demographic trends indicate that an update is necessary. When the forecaster strongly revises, from one forecast round to the next one, a forecast for a certain target year (for instance the life expectancy in 2050), this indicates large uncertainty connected to mortality predictions. The aim of this chapter is to shed more light on the uncertainty in mortality forecasts, by analysing the extent to which life expectancy predictions for 2030 and 2050 were revised in subsequent rounds of population forecasts published by statistical agencies in selected countries. It updates and extends earlier work that focused on United Nations and Eurostat forecasts published between 1994 and 2004 (Keilman et al. 2008). There the conclusion was that life expectancy forecasts for 18 European countries for the year 2050 had been revised upwards systematically, by around 2 years on average during the 10-year publication period. A recent analysis based on ofﬁcial population forecasts for Norway published in the period 1999– 2018 led to the same conclusion (Keilman 2018). Here we will show that the period of upward revisions seems to have ended for some European countries. To predict the life expectancy for some future year appears to be similar to aiming at a moving target (Lee 1980). The forecaster tries to hit the value as well as she can, but we cannot expect that the ﬁrst attempt will be successful. Next, there is a new attempt, but while the riﬂe was reloaded, the target appears to have moved upwards. This may go on for some forecast rounds. However, sometimes we notice hardly 179 180 N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen any revision from one forecast round to the next – in some cases, we even see a downward revision. First, we illustrate this process with life expectancy assumptions for 2030 and 2050 included in ofﬁcial population forecasts of Austria, Denmark, the Netherlands, Norway, Sweden, and the United Kingdom. These countries were selected because the statistical agencies revise their population forecasts every 2 or 3 years. In addi- tion, we show life expectancy assumptions for Japan, which is leading international trends in longevity. Next, we try to explain the systematic revisions by theories of anchoring (Tversky and Kahneman 1974; Kahneman 2011) and assumption drag (Ascher 1978). 9.2 Findings Many methods have been used in the recent past to forecast mortality. Booth and Tickle (2008) give an extensive review. Most methods use some form of extrapo- lation: one assumes that the future trends in key parameters are a continuation of trends from the past. The key parameters could be age-speciﬁc mortality rates or the parameters in an underlying model. Some scholars have developed formal models for analysing current mortality trends in which risk factors and behavioural variables are linked to mortality at various ages, but such explanatory models are very rare in ofﬁcial demographic forecasts (the model employed by Statistics Netherlands is an exception; see below), for a number of reasons. These include the poor predictive performance of the models and the fact that future trends in explanatory variables (smoking, food habits, health care etc.) are as difﬁcult to assess as future trends in mortality itself. See Bengtsson and Keilman (2019) for a recent overview. Concerning the mortality forecasts presented here, the statistical agencies of Denmark, Japan, Netherlands, Norway, and Sweden use the Lee-Carter model (Lee and Carter 1992), or variations of it. The model variant used by Statistics Netherlands has two distinctive features: the role of smoking is explicitly modelled, and current trends in other countries than the Netherlands are included. The latter feature reduces the risk of extrapolating national idiosyncratic mortality trends. Mortality forecasts for Austria and the United Kingdom are based on assumed rates of decline in age-speciﬁc mortality rates in the future. The Lee-Carter model assumes that a set of age-speciﬁc mortality rates observed for a number of years can be summarized in three sets of parameters. The ﬁrst is a general age pattern of age-speciﬁc mortality, with one parameter value for each age. The second is a period index, with one parameter value for each year. The period index reﬂects falling mortality over time. However, the decrease is not the same for each age, and therefore the model contains an additional set of age-speciﬁc parameters, which modify the period index for each age. When used for projecting future mortality, one extrapolates the period index to future years, while keeping the two sets of age-speciﬁc parameters constant. Predicted age-speciﬁc mortality rates for a certain year can be summarized into a prediction for the life expectancy at birth 9 European Mortality Forecasts: Are the Targets Still Moving? 181 9 (LE) for that year. 9.2 Findings The model has been criticized for under-projecting long-term life expectancies (and even short-term life expectancies when using long time series with historical mortality rates); see Stoeldraijer et al. (2018), and the references therein. During some years, the LE increased faster than in other years. Therefore, it is difﬁcult to select a certain period that can be thought to be representative for the future. Moreover, the non-linear nature of the model tends to slow down the increase in predicted LE. The result is a concave curve that eventually shows a tendency towards “ﬂattening out” in the longer term. Extrapolation of mortality based on constant rates of decline in age-speciﬁc mor- tality also leads to a concave curve for the LE as a function of time. A proportional improvement in mortality makes less and less difference in the expectation of life (Keyﬁtz and Caswell 2005, 81). In what follows, we will focus on LE-values for men and women for 2030 and for 2050. We would like to stress that the LE is not the primary mortality indicator deliberately set to some value by the statistical agencies. Rather, it summarizes extrapolated age-speciﬁc mortality rates that were set either directly (Austria, the UK) or indirectly (through the Lee-Carter model and its parameters; see above). We acknowledge that many different age patterns of mortality can lead to the same value of the LE – yet we focus on the latter measure because it is a simple and straightforward indicator for checking the plausibility of assumptions on future mortality. 9.2.1 Descriptive Findings 9.2 Life expectancy predictions for Denmark, Norway, and Japan for the year 2030, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark Norway Japan Predicted life expectancy 78 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Fig. 9.2 Life expectancy predictions for Denmark, Norway, and Japan for the year 2030, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom (2051) 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom (2051) Predicted life expectancy Predicted life expectancy Fig. 9.3 Life expectancy predictions for Austria, the Netherlands, Sweden, and the United Kingdom around 2050, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom (2051) Predicted life expectancy 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom (2051) Predicted life expectancy Fig. 9.3 Life expectancy predictions for Austria, the Netherlands, Sweden, and the United Kingdom around 2050, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) The graphs show a more or less systematic upward revision of LE-values from one forecast round to the next. For the case of Denmark, the upward trend appears to have ended around 2013. In the forecasts computed from 2013 onwards, there seems to be agreement about an LE for 2050 around 86 years for men and 88 years for women. For the other two countries, the forecasters show increased optimism in the sense that assumed LE-values were adjusted upwards in subsequent forecasts, although the revisions are not as strong as those for Denmark are during the period before 2013. 9.2.1 Descriptive Findings Figure 9.1 plots assumed values for the LE in 2049/2050 for men and women in a series of forecasts for the populations in Denmark, Japan, and Norway. The assumptions refer to ofﬁcial forecasts made by statistical agencies in the three countries during the period 2000–2018. The data come from various sources, as listed in the Appendix. 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark 2049 Norway 2050 Japan 2050 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Predicted life expectancy Predicted life expectancy Jump-off year of forecast Denmark 2049 Norway 2050 Japan 2050 Fig. 9.1 Life expectancy predictions for Denmark, Norway, and Japan around 2050, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark 2049 Norway 2050 Japan 2050 Predicted life expectancy 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Predicted life expectancy Jump-off year of forecast Denmark 2049 Norway 2050 Japan 2050 78 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast 78 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Fig. 9.1 Life expectancy predictions for Denmark, Norway, and Japan around 2050, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) Fig. 9.1 Life expectancy predictions for Denmark, Norway, and Japan around 2050, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) 182 N. Keilman and S. Kristoffersen 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark Norway Japan Predicted life expectancy 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark Norway Japan 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Denmark Norway Japan Predicted life expectancy Predicted life expectancy Fig. 9.2.1 Descriptive Findings One has to be a bit cautious concerning the LE of Japanese women, because we have only a few data points, and the upward revision from the 2010- forecast to the 2015-forecast is very modest. The patterns that emerge for 2049/2050 in Fig. 9.1 are very similar to those for the year 2030 in the three countries; see Fig. 9.2. However, there is one exception: the 2030 predictions for Danish men computed between 2015 and 2018 show minor downward corrections. The “target” appears to move in opposite direction, compared to forecasts published before 2015. Figures 9.3 and 9.4 show downward revisions in predicted LEs for 2030 and 2050 in four other countries: Austria, the Netherlands, Sweden, and the United Kingdom. The predictions for Austria appear to be the ﬁrst ones for which upward revisions came to a halt: for both target years 2030 and 2050, this is visible starting 9 European Mortality Forecasts: Are the Targets Still Moving? 183 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom Predicted life expectancy Predicted life expectancy 78 80 82 84 86 88 90 92 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Jump-off year of forecast Austria Netherlands Sweden United Kingdom Predicted life expectancy Fig. 9.4 Life expectancy predictions for Austria, the Netherlands, Sweden, and the United Kingdom for the year 2030, forecasts prepared between 2000 and 2018. Left panel: men. Right panel: women. (Source: See Appendix) Fig. 9.5 Life expectancy at birth. (Source: See Appendix) 75 77 79 81 83 85 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 y c n a tc e p x e efil d e v r e s b O Calendar year Austria Netherlands Sweden United Kingdom men women in 2007. Other countries followed a few years later. The cases of Sweden and the UK stand out with strong downward revisions in the last forecast, compared to the previous one. In the forecast of 2018, the 2050 LE-prediction for Swedish women was 0.55 years lower than the corresponding value in the forecast of 2017. 9.2.1 Descriptive Findings For men and women in the UK, the 2050 predictions for LE fell by a whole year between 2014 and 2016, which makes a downward slope of half a year of life per calendar year. These revisions are of similar magnitude as those for Austria between 2015 and 2016 (−0.58 years). Also, note that LE-assumptions in Figs. 9.1, 9.2, 9.3 and 9.4 seem to converge over time, with much larger differences between countries for forecasts computed in the ﬁrst decade of the century than in later forecasts. An obvious question is whether the patterns shown in Figs. 9.3 and 9.4 are related to trends in actually observed LEs for recent years. Figure 9.5 may shed some led on this. We note that the upward trend in LE has weakened in all four countries in recent years, perhaps with the exception of men in Sweden. Thus, a possible explanation of the ﬂat or even decreasing trends in predicted LE in Figs. 9.3 and 9.4 might be the fact that increases in actual LE tend to slow down, at least for Austria, Netherlands, and the United Kingdom. In other words, forecasters are possibly strongly guided by trends in the current value of the LE, when they predict the LE for future years. In Sect. 9.3, we will suggest psychological explanations for these ﬁndings. 184 N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen Some evidence for an association between observed and predicted trends can be found in the justiﬁcations that statistical agencies give for the downward revisions. ONS (2017) writes, for the case of the United Kingdom, “ . . . actual life expectancy has increased less than projected since mid-2014; this means that the life expectancy values for 2016 are lower, and also reduces the rate of increase in subsequent years.” Statistics Netherlands justiﬁes the downward revision by referring to the unfavourable mortality development in the last months of 2016 and the limited decrease in mortality in the ﬁrst 8 months of 2017. At the same time, relatively low mortality in 2014 (and a rather high LE that year) led to high values for predicted LEs in 2030 and 2050 in the 2015-based forecast, in particular for women. This effect disappeared in later forecasts (Stoeldraijer et al. 2017). 9.3 Possible Explanations: Assumption Drag and Anchoring Why did population forecasters in the countries analysed here so often revise their views on people’s length of life in an upwards direction? Or, to put it in terms of the simple model of Sect. 9.2.2: why did mortality forecasters under-predict so often the pace of annual LE-improvement? According to Pison (2018), French forecasters did not anticipate the sharp drop, after the Second World War, in adult mortality, old-age mortality in particular. There is no reason to assume that the situation was different in the seven countries analysed here until the beginning of this century. The decline in cardiovascular mortality explains much of the drop in adult mortality during the past 50 years. Falling numbers of cancer deaths contribute also. Forecasters did not foresee this decline, and relied heavily upon observed trends. Longevity improved only slowly during the 1950s and early 1960s, in particular for men. In some countries, there was even a stagnation or a decline. Examples are Denmark, the Netherlands, Norway, and Sweden. Therefore, forecasters assumed that the LE would increase very little in the immediate future, and that it would soon reach a maximum value (“ceiling”, or “limit”; see Oeppen and Vaupel 2001). Indeed, statistical agencies in ﬁve of our countries used such a ceiling: Austria (until the 1990-based forecast, in which mortality was kept constant after 2015), Denmark (forecast of 1997, constant after 2012), Norway (forecast of 1990, constant after 2010), Netherlands (forecast of 1995, constant after 2010), and Sweden (forecast of 1994, constant after 2025). During the 1990s, however, the forecasters in these countries dropped the idea of a ceiling, and started to extrapolate a much longer increase in future LE, although the slope was not steep enough. French forecasters used an LE-ceiling up to the forecast published in 1986, but gave up this idea starting with the forecast published in 1995 (Pison 2018). 9 European Mortality Forecasts: Are the Targets Still Moving? 9 European Mortality Forecasts: Are the Targets Still Moving? Finally, assume that the increase in actual life expectance slows down, or even stagnates, whereas the extrapolations still follow a straight line with slope be. Then the difference (b −be) may become negative, which implies a lower life expectancy forecast for year T compared to the previous forecast. Note that the straight-line assumptions formulated above are not crucial for the qualitative results. As long as average annual increases over relevant time intervals are b and be for actual and extrapolated trends, respectively, we will see upward revisions for the predicted life expectancy in year T whenever the actual life expectancy improves faster than the extrapolated one (b > be). 9.2.2 A Simple Model The process can be formalized as follows. For simplicity, we assume linearity both for observed and for extrapolated life expectancy trajectories, but with different slopes. Consider a time interval [t0,T], where t0 is a certain year in the past, and T is some future year (“target year”). A forecaster has data on actual life expectancy values LE(t) for the time interval [t0, t1] and is faced with the task of predicting the life expectancy LE(t) to year T, starting from the jump-off year t1. Assume that actual life expectancy LE(t) follows a straight line with slope b > 0 on [t0, T]. Assume further that the extrapolated trajectory is a straight line on [t1, T] with slope be > 0. Then the predicted life expectancy in year T, resulting from the prediction with jump-off year t1, is LE1 (T) = LE (t1) + (T −t1) .be. An updated forecast is made in year t2 > t1. The new extrapolation starts from LE (t2) = LE (t1) + (t2 −t1) .b. The revised prediction for year T is now LE2 (T) = LE (t2) + (T −t2) . be = LE (t1) + (t2 −t1) . b + (T −t2) . be The revised forecast LE2(T) differs from the previous forecast LE1(T) by an amount of LE2 (T) −LE1 (T) = (t2 −t1) . b + (T −t2) . be −(T −t1) . be= (t2−t1) . (b−be) . First, assume that be < b. The extrapolated life expectancy falls short compared to the actual life expectancy by an amount of (b- be) annually. When the inter-forecast period is (t2 −t1) years, the new life expectancy forecast for year T is higher than the previous one by (t2 −t1).(b −be) years. This is the situation in Figs. 9.1 and 9.2. Next, assume that life expectancy is extrapolated with the correct slope (be = b). Then the new forecast for year T is the same as the previous one: LE2 (T)−LE1 (T). Much of the data in Figs. 9.3 and 9.4 reﬂect this pattern. Much of the data in Figs. 9.3 and 9.4 reﬂect this pattern. 185 9.3.1 Assumption Drag Forty years ago, Ascher (1978) analysed fertility forecasts in developed countries and noted that forecasters tend to rely strongly on recently observed data; they give N. Keilman and S. Kristoffersen 186 less weight to the long-term trend. Figure 9.5 suggests that this “assumption drag” might hold for mortality, too: forecasters in Austria, the Netherlands, Sweden, and the UK revised assumed LE-values for 2030 and 2050 downwards, because they relied strongly on a weak upward trend of observed LEs in recent years. Here, “assumption drag” is to be understood as the maintenance of incorrect assumptions after their validity has been contradicted by the data. Why this practice? First, there might be a tendency among demographers to agree on incorrect assumptions because of socially validated beliefs, for example that there must be an upper limit to longevity, or a lower limit to fertility. Such a consensus makes it easier to reject conﬂicting evidence, such as new research results or data errors. Second, the complexity of advanced methods can mean that the results achieved are outdated, because all data are collected and processed and the high costs of advanced methods can mean that the forecasts simply tend to copy the underlying assumptions from a previous round. Let us assume Ascher’s assumption drag applies to mortality, too. The simple model of Sect. 9.2.2 states that it is primarily the slope in the LE between the jump-off year of the forecast and the year 2030/2050 that is under-predicted, not so much the level. Following this line of thought, Ascher’s theory of assumption drag applies to improvements in the LE, rather than LE levels. The consequence may very well be that in future population forecasts, the downward revisions in Figs. 9.3 and 9.4 will come to a halt and that more or less stable patterns will emerge. This is more likely for 2030 than for 2050. After all, the closer we get to a certain target year, the easier it becomes to predict the LE for that year. Obviously, there is one additional important assumption underlying these speculations, namely that the long-term trend in LE expectancy is deﬁnitely upward, and that any periods of stagnation are only temporary. 9.3.2 Anchoring The anchoring effect is one of the most solid tested phenomena in the world of experimental psychology. Tversky and Kahneman (1974; see also Kahneman 2011) discovered a cognitive bias, which takes place when we consider a particular value of an unknown quantity before estimating such quantity. The value we have considered or that has been shown to us before, strongly determines the estimate we are going to make, which will always be relatively close to that previous value, which is called the anchor. Once the anchor has been established, we evaluate whether it is high or low and then we adjust our estimate to that amount. This mental process ﬁnishes early, because we are not sure of the real amount. Therefore, our estimation is not usually far from the anchor. Thus, the idea of an adjust-and-anchor heuristic as a strategy for estimating uncertain quantities is as follows. Start from an anchoring number, assess whether it is too high or too low, and gradually adjust your estimate by mentally “moving” from the anchor. The adjustment typically ends European Mortality Forecasts: Are the Targets Still Moving? 9 187 9 prematurely, because people stop when they are no longer certain that they should move farther. We can use the theory of anchoring to explain the patterns that we see in Figs. 9.1, 9.2, 9.3 and 9.4. To ﬁx ideas, consider a forecast made every 3 years; let us say in 2012, 2015, and 2018. A forecaster confronted with the task of extrapolating LE between 2012 and 2030 uses recently observed values as an anchor. In spite of the fact that historical values have increased more or less linearly at a certain pace, a simple straight-line extrapolation with the same slope would move the prediction for 2030 too far away from the anchor value, and the forecaster decides to extrapolate with smaller annual improvements than historically. This may be a straight line, or, a decelerating (concave) curve. The next forecast round starts from the LE observed for 2015, and moves the complete extrapolated line or curve upwards. This is in essence the process described by the model in Sect. 9.2.2. Because the extrapolations do not increase fast enough, the new prediction for 2030 is higher than the old one for the same year. The whole procedure is repeated for 2018, and the result is an even higher LE-prediction for 2030. 9.3.2 Anchoring Figure 9.6 illustrates this process for the case of the United Kingdom. Fig. 9.6 Actual and projected period expectation of life at birth (EOLB), males, United Kingdom, 1966 to 2030, selected projections. Fig. 9.6 Actual and projected period expectation of life at birth (EOLB), males, United Kingdom, 1966 to 2030, selected projections. Fig. 9.6 Actual and projected period expectation of life at birth (EOLB), males, United Kingdom, 1966 to 2030, selected projections. 188 N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen N. Keilman and S. Kristoffersen Between 1985 and 2012, the Ofﬁce for National Statistics (ONS) did not extrapolate the LE according to a straight line, but used a concave curve. As argued in Sect. 9.2, not only extrapolations based on proportionate changes in age-speciﬁc mortality, but also those based on the Lee-Carter model will result in LE-improvements that diminish over time. In Sect. 9.2.2, we demonstrated that even with straight-line extrapolations, we would observe systematic upward revisions of predicted LEs for a certain target year if the slope of the extrapolation were less steep than that of actual values. This was the case for ONS-forecasts between 1971 and 1981 in Fig. 9.6. g The discussion so far attempts to explain the patterns in Figs. 9.1 and 9.2, where LE-predictions are systematically revised upwards. However, we can also use the theory of anchoring behaviour to explain downward revisions as in Figs. 9.3 and 9.4. When actual LE stagnates, the anchoring effect becomes stronger, and the extrapolations in the previous round of forecasts are considered too steep. As a result, the revised extrapolation curve is ﬂatter than the original one, leading to a revised 2030-prediction that is close to the value in the previous round. This may explain the patterns we see for Danish men and women after 2011 in Figs. 9.1 and 9.2, and for Austrian men and women for forecasts with jump-off years between 2009 and 2015. Very strong anchoring may even lead to a downward revision; cf. the cases of Sweden and the UK in particular. Kahneman (2011) notes that there are situations in which anchoring appears reasonable. People who are asked difﬁcult questions clutch at straws, and the anchor is a plausible straw. To predict long-term trends in mortality is clearly difﬁcult. Therefore, it is reasonable to use actual mortality trends as anchors. 9.3.2 Anchoring Yet one may wonder if forecasters, once being aware of the anchoring effect when formulating forecast assumptions, will learn from the errors they made in the past? 9.4 Conclusions Acknowledgements We acknowledge gratefully the help of Alexander Hanika (Statistics Austria) and Annika Klintefelt (Statistics Denmark) in collecting data on historical forecasts for the two countries. 9.4 Conclusions Life expectancy predictions for a certain target year (for instance, 2030, or 2050) computed by statistical agencies in some countries during the past decade have been revised upwards frequently. We noticed this in ofﬁcial LE-predictions for Denmark, Japan, and Norway. However, for a number of other countries (viz. Austria, the Netherlands, Sweden, the United Kingdom), such upward revisions are no longer visible. The LE-adjustments for 2030 and 2050 appear to be very small – they are even negative in the most recent forecasts for these countries. This means that in the current forecast, the forecaster is less optimistic about the LE in the target year than she was in the previous forecast. One possible explanation is that actual LE did not improve much, perhaps even stagnated, during the period between two forecasts. The patterns described here, illustrated by Figs. 9.1, 9.2, 9.3 and 9.4, are compatible with a situation in which the real (but unknown) LE until 2030 or 2050 improves faster than the predicted LE. We referred to two psychological factors that can be used to explain these patterns. The ﬁrst one is an assumption drag, a term European Mortality Forecasts: Are the Targets Still Moving? 189 9 ﬁrst coined by Ascher in 1978 in connection with fertility forecasts in developed countries in the 1960s, which tended to be far too high. The assumption drag involves a psychological mechanism according to which forecasters rely heavily on recently observed data, whereas they give less weight to long-term trends. The second psychological mechanism that one may use to explain upward and downward revisions of the LE in a series of population forecasts is an anchoring effect, discovered by Tversky and Kahneman. When a forecaster has to predict an unknown and uncertain quantity, he will start from a known value (the anchor), and predict a value that is close to that value. The process with upward or downward revisions of predicted LE for a certain year in the future resembles the behaviour of a hunter, who aims at a moving target. Sometimes the target moves up (upward revision of the LE), sometimes down (downward revision). However, a simple model based on linear extrapolations of the LE suggests that upward revisions result simply from the fact that extrapolated LE does not improve as fast as actual LE. Downward revisions may be the result of a temporary stagnation of LE-improvement. 9 European Mortality Forecasts: Are the Targets Still Moving? 191 Statistics Sweden (2005). https://www.scb.se/statistik/BE/BE0401/2005A01/ BE0401_2005A01_SM_BE18SM050 1.pdf. Accessed: October 2018. Statistics Sweden (2005). https://www.scb.se/statistik/BE/BE0401/2005A01/ BE0401_2005A01_SM_BE18SM050 1.pdf. Accessed: October 2018. p Statistics Sweden (2006). http://www.scb.se/statistik/_publikationer/BE0401_ 2006I50_BR_BE51ST0602.pdf. Accessed: October 2018. Statistics Sweden (2018). Sveriges framtida befolkning 2018–2070. https://www. scb.se/hitta-statistik/statistik-efter-amne/befolkning/befolkningsframskrivningar/ befolkningsframskrivningar/pong/publikationer/sveriges-framtida-befolkning- 20182070/. Accessed: October 2018. Statistics Sweden (2007). http://www.statistikdatabasen.scb.se/goto/en/ssd/ PrognosLivslangd04. Accessed: October 2018. Statistics Sweden (2008). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefPrognosLivslangd. Accessed: October 2018. Statistics Sweden (2009). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefPrognosLivslang09. Accessed: October 2018. Statistics Sweden (2010). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2010. Accessed: October 2018. Statistics Sweden (2011). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2011. Accessed: October 2018. Statistics Sweden (2012). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2012. Accessed: October 2018. g g Statistics Sweden (2013). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2013. Accessed: October 2018. Statistics Sweden (2014). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2014. Accessed: October 2018. Statistics Sweden (2014). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2014. Accessed: October 2018. Statistics Sweden (2015). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2015. Accessed: October 2018. Statistics Sweden (2015). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2015. Accessed: October 2018. g g Statistics Sweden (2016). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2016. Accessed: October 2018. g g Statistics Sweden (2016). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2016. Accessed: October 2018. g g Statistics Sweden (2017). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2017. Accessed: October 2018. Statistics Sweden (2017). http://www.statistikdatabasen.scb.se/goto/en/ssd/ BefProgLivslangd2017. Accessed: October 2018. Appendix: Data Sources Frank Hansen, M., Stephensen, P. (2013). Danmarks Fremtidige Befolkning: Befolkningsfremskrivning 2013. Alexander Hanika (2019) Personal communication. National Institute of Population and Social Security Research – IPSS (2012). http://www.ipss.go.jp/syoushika/tohkei/newest04/h4_2.html. Accessed: October 2018. IPSS (2007). http://www.ipss.go.jp/syoushika/tohkei/suikei07/houkoku/katei/ 11-5.xls. Accessed: October 2018. IPSS (2007). http://www.ipss.go.jp/syoushika/tohkei/suikei07/houkoku/katei/ 11-5.xls. Accessed: October 2018. IPSS. http://www.ipss.go.jp/syoushika/tohkei/Mokuji/1_Japan/J_Detail_14.asp? fname=1_katei/1-2.htm&title1=%82P%81D%89%BC%92%E8%92l%95%5C& title2=%95%5C%82P%81%7C%82Q%81D%89%BC%92%E8%82%B3%82 %EA%82%BD%95%BD%8B%CF%8E%F5%96%BD%81i%8Fo%90%B6%8E IPSS. http://www.ipss.go.jp/syoushika/tohkei/Mokuji/1_Japan/J_Detail_14.asp? fname=1_katei/1-2.htm&title1=%82P%81D%89%BC%92%E8%92l%95%5C& title2=%95%5C%82P%81%7C%82Q%81D%89%BC%92%E8%82%B3%82 t t e %95%5C%8 %8 %7C%8 Q%8 %89% C%9 % 8%8 % 3%8 %EA%82%BD%95%BD%8B%CF%8E%F5%96%BD%81i%8Fo%90%B6%8E %9E%82%CC%95%BD%8B%CF%97%5D%96%BD%81j%82%CC%90%84 %88%DA. Accessed: October 2018. Q %EA%82%BD%95%BD%8B%CF%8E%F5%96%BD%81i%8Fo%90%B6%8E %9E%82%CC%95%BD%8B%CF%97%5D%96%BD%81j%82%CC%90%84 %88%DA. Accessed: October 2018. IPSS. http://www.ipss.go.jp/ppzenkoku/e/zenkoku_e2017/g_images_e/pp29gt 0402e.ﬁles/sheet001.htm. Accessed: October 2018. IPSS. http://www.ipss.go.jp/ppzenkoku/e/zenkoku_e2017/g_images_e/pp29gt 0402e.ﬁles/sheet001.htm. Accessed: October 2018. Ofﬁce for National Statistics – ONS (2001). https://webarchive.nationalarchives. gov.uk/20160106011038/; http://www.ons.gov.uk/ons/rel/npp/national-population- projections-historic-series/2000-based-projections/index.html. Accessed: October 2018. 190 N. Keilman and S. Kristoffersen ONS (2003). https://webarchive.nationalarchives.gov.uk/20160106011038/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections-historic-series/ 2002-based-projections/index.html. Accessed: October 2018. ONS (2003). https://webarchive.nationalarchives.gov.uk/20160106011038/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections-historic-series/ 2002-based-projections/index.html. Accessed: October 2018. ONS (2005). https://webarchive.nationalarchives.gov.uk/20160106011038/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections-historic-series/ 2004-based-projections/index.html. Accessed: October 2018. ONS (2007). https://webarchive.nationalarchives.gov.uk/20160105223341/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2006-based- projections/index.html. Accessed: October 2018. ONS (2009). https://webarchive.nationalarchives.gov.uk/20160105223341/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2008-based- projections/index.html. Accessed: October 2018. ONS (2011). https://webarchive.nationalarchives.gov.uk/20160105223341/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2010-based- projections/index.html. Accessed: October 2018. ONS (2013). https://webarchive.nationalarchives.gov.uk/20160105223341/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2012-based- projections/index.html. Accessed: October 2018. ONS (2015). https://webarchive.nationalarchives.gov.uk/20160105223341/; http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2014-based- projections/index.html. Accessed: October 2018. ONS (2017). National Population Projections: 2016-based statistical bulletin. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/ populationprojections/bulletins/nationalpopulationprojections/2016basedstatistical bulletin. Accessed: October 2018. ONS (2018). https://www.ons.gov.uk/peoplepopulationandcommunity/births deathsandmarriages/lifeexpectancies/datasets/nationallifetablesunitedkingdom referencetables. Accessed: October 2018. Kaneko, R., Ishikawa, A., Ishii, F., Sasai, T., Iwasawa, M., Mita, F. and Moriizumi, R. (2006). Population projections for Japan: 2006–2055 outline of results, methods, and assumptions. Statistics Denmark (2019). Population Projections for Denmark. https://www. dst.dk/en/Statistik/emner/befolkning-og-valg/befolkning-og-befolkningsfrems krivning/befolkningsfremskrivning. Accessed: January 2019. Statistics Denmark. https://www.dst.dk/da/Statistik/Publikationer/StE/statistiske- efterretninger-emner?psi=486. Accessed: October 2018. Statistics Netherlands (2019). https://opendata.cbs.nl/#/CBS/nl/navigatieScherm/ zoeken?searchKeywords=*&page=1&year%5B%5D=Prognose. Accessed: June 2019. Statistics Sweden (2002). https://www.scb.se/statistik/BE/BE0401/2003M00/ BE18SM0201.pdf. Accessed: October 2018. Statistics Sweden (2002). https://www.scb.se/statistik/BE/BE0401/2003M00/ BE18SM0201.pdf. Accessed: October 2018. p Statistics Sweden (2003). http://www.scb.se/statistik/BE/BE0401/2003I50/ BE51ST0304.pdf. Accessed: October 2018. p Statistics Sweden (2003). http://www.scb.se/statistik/BE/BE0401/2003I50/ BE51ST0304.pdf. Accessed: October 2018. Statistics Sweden (2004). http://www.scb.se/statistik/BE/BE0401/2003M00/ BE0401_2004A01_SM_BE18SM0401.pdf. Accessed: October 2018. Statistics Sweden (2004). http://www.scb.se/statistik/BE/BE0401/2003M00/ BE0401_2004A01_SM_BE18SM0401.pdf. Accessed: October 2018. 9 European Mortality Forecasts: Are the Targets Still Moving? References Ascher, W. (1978). Forecasting: An appraisal for policy makers and planners. Baltimore: John Hopkins University Press. Bengtsson, T., & Keilman, N. (Eds.). (2019). Old and new perspectives on mortality forecasting Cham: Springer. Booth, H., & Tickle, L. (2008). Mortality modelling and forecasting: A review of methods. Annals of Actuarial Science, 3(1 & 2), 3–43. Kahneman, D. (2011). Thinking, fast and slow. London: Penguin Books. Keilman, N. (2018). Increasing (but Insufﬁcient?) Optimism about future life expectancy. NIUSSP 15 October 2018. Keilman, N., Cruijsen, H., & Alho, J. (2008). Changing views of future demographic trends. In J. Alho, S. Hougaard Jensen, & J. Lassila (Eds.), Uncertain demographics and ﬁscal sustainability (pp. 11–33). Cambridge: Cambridge University Press. 192 N. Keilman and S. Kristoffersen Keyﬁtz, N., & Caswell, H. (2005). Applied mathematical demography (3rd ed.). New York: Springer. Lee, R. (1980). Aiming at a moving target: Period fertility and changing reproductive goals Population Studies, 34(2), 205–226. Lee, R. D., & Carter, L. R. (1992). Modeling and forecasting US mortality. Journal of the American Statistical Association, 87(419), 659–671. ( ) Oeppen, J., & Vaupel, J. (2001). Broken limits to life expectancy. Science, 296(5570), 1029–1031 Oeppen, J., & Vaupel, J. (2001). Broken limits to life expectancy. Science, 296(5570), 1029–1031. Pison, G. (2018). 1968-2018: Four demographic surprises in France over the last 50 years. Population & Societies, 553, 1–6. Stoeldraijer, L., Van Duin, C., & Huisman, C. (2017). Bevolkingsprognose 2017–2060: 18,4 miljoen inwoners in 2060 (Population forecast 2017–2060: 18.4 million inhabitants in 2060). Statistische Trends, December 2017. Stoeldraijer, L., Van Duin, C., Van Wissen, L., & Janssen, F. (2018). Comparing strategies for matching mortality forecasts to the most recently observed data: Exploring the trade-off between accuracy and robustness. Genus, 74(16), 1–20. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science 185(4147), 1124–1131. Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. References If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
https://openalex.org/W3206587422	https://idus.us.es/bitstream/11441/139260/1/Muscular%20Echovariation%20as%20a%20New%20Biomarker%20for%20the%20Classification%20of%20Soleus%20Muscle%20Pathology%20A%20Cross-Sectional%20Study.pdf	English	null	Muscular Echovariation as a New Biomarker for the Classification of Soleus Muscle Pathology: A Cross-Sectional Study	Diagnostics	2,021	cc-by	6,237	Citation: De-la-Cruz-Torres, B.; Romero-Morales, C. Muscular Echovariation as a New Biomarker for the Classiﬁcation of Soleus Muscle Pathology: A Cross-Sectional Study. Diagnostics 2021, 11, 1884. https:// doi.org/10.3390/diagnostics11101884 Keywords: ultrasound; soleus injury; diagnosis; echovariation; biomarker Academic Editors: César Fernández De Las Peñas and Francisco Alburquerque-Sendín diagnostics diagnostics Article Muscular Echovariation as a New Biomarker for the Classiﬁcation of Soleus Muscle Pathology: A Cross-Sectional Study Blanca De-la-Cruz-Torres 1,* and Carlos Romero-Morales 2 Blanca De-la-Cruz-Torres 1 1 Department of Physiotherapy, University of Seville, Avicena Street, 41009 Seville, Spain 2 Faculty of Sport Sciences, Universidad Europea, Villaviciosa de Odón, 28670 Madrid, Spain; carlos.romero@universidadeuropea.es * C d b @ 1 Department of Physiotherapy, University of Seville, Avicena Street, 41009 Seville, Spain 2 Faculty of Sport Sciences, Universidad Europea, Villaviciosa de Odón, 28670 Madrid, Spain; carlos.romero@universidadeuropea.es * Correspondence: bcruz@us.es p * Correspondence: bcruz@us.es Abstract: Background: Soleus injury is one of the most common soft tissue tears during sport activities. Current classiﬁcations of muscle tears are based on symptoms and tear size and they do not contribute suitable evidence-based treatment protocols. The objective of this study was to analyze the most frequent echotexture ﬁndings of patients with soleus muscle injury, located in the central intramuscular tendon (IMT), and healthy people to determine whether they behave differently and to propose an ultrasound (US)-based classiﬁcation. Methods: eighty-four athletes, who played in sport activities comprising lower limbs. Echotexture characteristics of soleus muscle were reviewed for 84 subjects. They were divided based on the muscle echogenicity in three groups (Injury Type 1 group, Injury type 2 group and healthy group). Echointensity (EI) and Echovariation (EV) were taken in all groups like quantitative US variable. Results. The Injury Type 1 group was identiﬁed by a hypoechoic area and characterized by a higher EV; and Injury Type 2 group was identiﬁed by a ﬁbrotic area and characterized by a lower EV. The echogenic pattern of healthy people obtained an intermediate value of EV between both injured soleus types. Conclusions. EV may be useful to classify different types of soleus muscle pathology according to the echogenicity pattern. An innovative proposed US-based classiﬁcation system for soleus tears may be used to guide treatment decisions for patients with central tendon injury of soleus muscle. Citation: De-la-Cruz-Torres, B.; Romero-Morales, C. Muscular Echovariation as a New Biomarker for the Classiﬁcation of Soleus Muscle Pathology: A Cross-Sectional Study. Diagnostics 2021, 11, 1884. https:// doi.org/10.3390/diagnostics11101884 Academic Editors: César Fernández De Las Peñas and Francisco Alburquerque-Sendín Received: 13 September 2021 Accepted: 10 October 2021 Published: 12 October 2021 1. Introduction Soleus pathologies are one of the most popular soft tissue tears in sports modalities and they are much more common than gastrocnemius pathologies [1,2]. Current studies describe an intramuscular tendon (IMT) and two aponeuroses (lateral and medial) in the soleus muscle [3,4]. Although the IMT may present variability between individuals, it is positioned in the central part of the muscle and has a relevant role like an attaching place of the muscle ﬁbers and being partly of the Achilles tendon [5]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. From a topographical point of view, ﬁve locations in the soleus muscle were recently identiﬁed like possible lesional sites [6]: two myofascial areas (posterior and anterior tears) and three musculotendinous junction points (proximal medial, proximal lateral and distal central tendon tears). Clinically, IMT injury is the most prevalent, being very common in sporting populations, such as dancers or soccer players [5,7]. This injury is characterized by several symptoms that the athlete describes as overload during sports activity, stiffness or lack of elasticity during stretching. The IMT ruptures may create a hypertrophic intramuscular connective tissue scar in the muscle [8]. In addition, they could be considered as non-limiting injuries for sports performance. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Although several physiological parameters have been used as biomarkers in muscle damage [9], their quantiﬁcation was expensive and time-consuming. Describing a muscle https://www.mdpi.com/journal/diagnostics Diagnostics 2021, 11, 1884. https://doi.org/10.3390/diagnostics11101884 Diagnostics 2021, 11, 1884 2 of 9 ultrasound biomarker that allows clinicians to diagnose a muscle injury and determine the characteristics of the injury state will be a relevant advance. ultrasound biomarker that allows clinicians to diagnose a muscle injury and determine the characteristics of the injury state will be a relevant advance. j y Muscle ultrasonography is an easily accessible, low-cost and repeatable tool that provides the image analysis for the classiﬁcation of muscle tears by clinicians. In 2017, Loizides et al. [10] afﬁrmed that US may be considered as the ﬁrst-line imaging modality to diagnosis and follow-up of musculoskeletal injuries. The main echotexture parameters are echointensity (EI) and echovariation (EV). 2.2. Ethical Considerations The study was approved by the Ethics Committee of University Hospital Virgen Macarena-Virgen del Rocio (0663-N-19), respecting the Declaration of Helsinki throughout the study. Before being a participant in this study, all the subjects signed a written informed consent form. 2. Materials and Methods 2.1. Design A cross-sectional study was carried out in a sports population according to the Strengthening the Reporting of Observational Studies in Epidemiology Statement (STROBE) recommendations [23]. 2.3. Sample Size Calculation GPower software has been employed for the sample size calculation by the difference between 3 groups for a One-way ANOVA model with a total sample of 12 subjects of a pilot trial with a mean of: Injury Type 1 group = 39.18; Injury Type 2 group = 30.01 and healthy group = 39.19 for the EI variable. Thus, with an effect size of 0.35, an α error of 0.05, a power of 0.80 and a number of 3 groups the total sample determined was 84 subjects. 1. Introduction The EI is considered as the pixel intensity quantiﬁcation of an ultrasound image and the EV is determined by the relation between standard deviation (SD) and the EI. Both parameters may express different physiologi- cal conditions and their employment has been raised lately in order to obtain a better knowledge of healthy and injured muscle characteristics. Therefore, EI has been used to analyze the muscle quality in healthy populations [11,12] and in patients with degenerative muscle pathologies [13–15]. Echointensity changes have been considered as a disruption of normal connective tissue muscle in both degenerative [16,17] and acute conditions [18]. However, the role of the EI parameter as muscle biomarkers in damage status is still limited. For a better interpretation of the muscle echogenicity, EV has been currently employed in combination with EI. To our knowledge, EV has been used in patients with plantar fascia pathology [19], chronic lumbopelvic pain [20], dancers with asymptomatic Achilles tendon [21] and individuals with amyotrophic lateral sclerosis [22]. Regarding muscle injuries in sports people, De-la-Cruz-Torres et al. [7] found that central IMT injury of soleus muscle is characterized by a minor EI. However, they found no differences in EV between healthy subjects and patients with soleus pathology. Authors hypothesize that the analysis of EV in soleus pathology should be more exhaustive and differentiate between types of injuries, characterized by presenting hypoechoic areas or ﬁbrotic areas. Therefore, the main goal of this study was to analyze the most frequent echotexture ﬁndings of patients with soleus muscle injury, located in the IMT, and healthy people to determine whether they behave differently and to propose an ultrasound (US)-based classiﬁcation system for soleus injuries. A possible classiﬁcation of soleus injuries may contribute suitable evidence-based treatment protocols. 2.4. Participants Echotexture features of the soleus muscle were reviewed for 84 subjects. They were divided based on the muscle echogenicity in three groups: Injury Type 1 group, healthy group (Figure 1), Injury Type 1 and Injury Type 2 (Figure 2). As in previous studies [7,24], a central IMT tear in the soleus muscle was identiﬁed when patients reported having symp- Diagnostics 2021, 11, 1884 3 of 9 toms for at least six months, and they established a self-rated progressive stiffness value greater than or equal to 4 points during sports exercises or during the soleus stretching, measured using the numeric rating scale (NRS) [25]. The enrollment was carried out in a private clinic specialized in sports injuries. First of all, researchers ﬁlled out the clinical history of the subjects. The presence of a central IMT injury of the soleus muscle was determined according to the following criteria: clinical history of the individuals, clinical evaluation performed by an experienced therapist and US examination carried out by an US-imaging specialist clinician with more than 15 years of experience. symptoms for at least six months, and they established a self-rated progressive stiffness value greater than or equal to 4 points during sports exercises or during the soleus stretching, measured using the numeric rating scale (NRS) [25]. The enrollment was carried out in a private clinic specialized in sports injuries. First of all, researchers filled out the clinical history of the subjects. The presence of a central IMT injury of the soleus muscle was determined according to the following criteria: clinical history of the individuals, clinical evaluation performed by an experienced therapist and US examination carried out by an US-imaging specialist clinician with more than 15 years of experience. symptoms for at least six months, and they established a self-rated progressive stiffness value greater than or equal to 4 points during sports exercises or during the soleus stretching, measured using the numeric rating scale (NRS) [25]. The enrollment was carried out in a private clinic specialized in sports injuries. First of all, researchers filled out the clinical history of the subjects. The presence of a central IMT injury of the soleus muscle was determined according to the following criteria: clinical history of the individuals, clinical evaluation performed by an experienced therapist and US examination carried out by an US-imaging specialist clinician with more than 15 years of experience. Figure 1. 2.4. Participants Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT central intramuscular tendon Figure 1. Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT, central intramuscular tendon. Figure 1. Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT, central intramuscular tendon. Figure 1. Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT central intramuscular tendon Figure 1. Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT, central intramuscular tendon. Figure 1. Ultrasound image of soleus muscle in healthy athletes. It shows a normal central IMT. Left panel: anatomic diagram of soleus muscle; Right panel: ultrasound image of the soleus muscle; Abbreviations: IMT, central intramuscular tendon. (A) (B) Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by fibrotic imagen; Left panel: anatomic diagram (A) (B) Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by fibrotic imagen; Left panel: anatomic diagram Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by ﬁbrotic imagen; Left panel: anatomic diagram of soleus muscle; Middle panel: ultrasound image with marked lesion area (yellow line); Right panel: ultrasound image of the soleus muscle. (A) (A) (A) (A) (A) (A) (B) (B) B) (B) Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. 2.4. Participants (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by fibrotic imagen; Left panel: anatomic diagram Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by fibrotic imagen; Left panel: anatomic diagram Figure 2. Ultrasound image of soleus muscle in athletes with injured soleus muscle. (A) Injury Type 1 group, shows an injured central IMT characterized by hypoechoic imagen; and (B) Injury Type 2 group, shows an injured central IMT characterized by ﬁbrotic imagen; Left panel: anatomic diagram of soleus muscle; Middle panel: ultrasound image with marked lesion area (yellow line); Right panel: ultrasound image of the soleus muscle. 4 of 9 Diagnostics 2021, 11, 1884 The selection criteria for the patients were (a) be over 18 years old; (b) athletes who played in sports activities comprising lower limbs (for example, soccer, running, rugby); (b) at least ﬁve years practicing the sport; (c) training at least two hours, four days a week; (d) non-acute lower limb pathology at least six months ago; (e) non low back musculoskeletal disorders. Furthermore, inclusion criteria to be included for the healthy group was being a healthy athlete, and for the pathological groups, having a clinical diagnosis of a central IMT injury in the soleus muscle, characterized by muscle stiffness and lack of elasticity when stretching the muscle and corroborated by an experienced clinician. 2.6. Ultrasonography All US imaging assessment was performed by two qualiﬁed clinicians (over 15 years of experience: B.d.l.C.T and M.C.T), who were blind to the subjects’ muscle status. Each clinician, individually and separately, analyzed the images and then the results were provided to the person who analyzed the data. A high-deﬁnition US with a 6–15 MHz linear transducer (S7, General Electrical Healthcare, Chicago, IL, USA) was used to achieve all US images. To the classiﬁcation of participants, US evaluation was developed in prone position with the feet hanging over the end of the table. Furthermore, a US probe was placed in transverse plane at the central IMT in order to analyze the echogenic model. Ultrasound images were acquired and registered with a depth of 4 cm, 12 MHz of frequency, 51 points of gain, neutral position of the time gain compensation, 2 focus located at 1 and 2 cm each one, respectively, in the lesional area to remain stable during the assessments. The angle of transducer was adapted until the best muscle image was visualized in each image to decrease the existence of artifacts. Researchers captured three images with a resolution of 820 × 614 pixels with 256 gray-scale levels and were saved as .TIFF ﬁles without compression or losses. 2.5. Clinical Variables Sociodemographic and clinical variables were registered, such as age, height, weight, body mass index (BMI), sex, dominance side and injured side (Table 1). The pain level at palpation in the injured area and gradual stiffness during the soleus stretching during sport activities were measured using NRS (0 point, no symptoms; 10 points, worse symptoms). Table 1. Sociodemographic data of the sample. Data Total Sample (n = 84) Injury Type 1 Group (n = 28) Injury Type 2 Group (n = 28) Healthy Group (n = 28) p Value Age, y 26.86 ± 8.85 25.57 ± 7.23 26.28 ± 10.62 28.75 ± 8.36 0.375 Weight, kg 68.71 ± 10.47 68.81 ± 12.84 65.14 ± 8.41 64.39 ± 9.87 0.395 Height, m 1.69 ± 0.08 1.70 ± 0.10 1.67 ± 0.62 1.71 ± 0.87 0.283 BMI, kg/m2 23.46 ± 3.00 23.82 ± 3.18 23.38 ± 3.88 23.19 ± 1.94 0.745 Gender (F/M) 45/39 16/12 15/13 14/14 - Dominance side (R/L) 76/8 25/3 26/2 25/3 - Injured side (R/L) 42/32 17/11 16/12 19/9 - Abbreviations: F, female; M, male; L, left; R, right. Table 1. Sociodemographic data of the sample. Table 1. Sociodemographic data of the sample. 2.6. Ultrasonography 2.7. Image Analysis Echointensity and echovariation were taken in all groups like quantitative US variables. The imaging study was performed ofﬂine with the ImageJ software (Bethesda, MD, USA) by a blinded tester. According to previous study [7], a range of interest (ROI) of 64 × 64 with an 8-bit gray scale using the ROI Manager tool was executed to extract the pixel distribution histogram. The ROI was considered as the muscle area without bone and fascia with the best reﬂection (Figure 1). Subsequently, echotexture values were captured from the histogram where EI was considered as the mean value of the gray-scale distribution of the pixels, and EV was described as the relation between SD and the mean of pixel distribution Diagnostics 2021, 11, 1884 5 of 9 (EV = σ/µ × 100), where σ is the SD of the image intensities and µ is the mean value of intensity in each ROI of the three images previously recorded. (EV = σ/µ × 100), where σ is the SD of the image intensities and µ is the mean value of intensity in each ROI of the three images previously recorded. Authors were aware that the muscle echotexture depends on ROI characteristics (for example, the size, shape and location) [26]. To reduce probable mistakes, ROI size was the same in all participants and the location was the one that best identiﬁes the muscle characteristics. Intra-rater reliability in the ROI selection has been reported in a previous study with the same dataset [7] In addition, inter-rater reliability, using the intraclass correlation coefﬁcient, was established in a collection of 15 images as sufﬁcient for clinical evaluation. 2.8. Statistics Analysis SPSS software v.26 (IBM SPSS Statistics; IBM, New York, NY, USA) was used for the statistics. The Kolmogorov-Smirnov test was used to check the normality assumption. A descriptive analysis was carried out for all the subjects separately between groups. One-way ANOVA was employed to assess the homogeneity of the sample and to assess the differences between groups for each variable. In addition, Tukey correction post hoc analyses were developed. Moreover, a multivariate analysis was developed using a linear regression to predict the inﬂuence of the descriptive data and group. The dependent variables were EI and EV and the independent variables were group, age, weight, height and BMI. The level of signiﬁcance was set at p < 0.05 with an α error of 0.05 (95% conﬁdence interval) and a desired power of 80% (β error of 0.2). 3. Results Sociodemographic data did not show signiﬁcant differences (p > 0.05) between groups (Table 1). According to Table 2, statistically signiﬁcant differences (p < 0.05) were shown for between-groups comparison for the echotexture variables (EI and EV). Regarding Table 3, post hoc comparison showed statistically signiﬁcant differences between Injury Type 1 group and Injury Type 2 group, Injury Type 1 group and Healthy group, Healthy group and Injury Type 2 group for EI variable. In addition, EV variable showed signiﬁcant differences (p < 0.05) for Injury Type 1 group and Injury Type 2 group, Injury Type 1 group and Healthy group, Healthy group and Injury Type 2 group. Moreover, inter-reliability values for the EI (ICC = 0.901) and EV (ICC = 0.912) were considered excellent. Table 2. One-way ANOVA for the Numerical Rating Scale (NRS) during sport activity and at palpation in soleus injury, Echointensity (EI), and Echovariation (EV) variables. Data Injury Type 1 Group Injury Type 2 Group Healthy Group p Value NRS during activity (points) 5.39 ± 1.64 5.89 ± 1.49 N/A N/A NRS at palpation (points) 5.33 ± 1.34 6.03 ± 1.34 N/A N/A Echointensity (EI) 19.64 ± 7.38 48.62 ± 8.83 64.53 ± 10.51 (178.8) 0.001 Echovariation (EV) 53.21 ± 19.23 22.54 ± 9.08 32.93 ± 7.36 (40.34) 0.001 for the Numerical Rating Scale (NRS) during sport activity and at palpation in soleus injury, ovariation (EV) variables. Table 2. One-way ANOVA for the Numerical Rating Scale (NRS) during sport activity and at palpation Echointensity (EI), and Echovariation (EV) variables. Data Injury Type 1 Group Injury Type 2 Group Healthy Group p Value NRS during activity (points) 5.39 ± 1.64 5.89 ± 1.49 N/A N/A NRS at palpation (points) 5.33 ± 1.34 6.03 ± 1.34 N/A N/A Echointensity (EI) 19.64 ± 7.38 48.62 ± 8.83 64.53 ± 10.51 (178.8) 0.001 Echovariation (EV) 53.21 ± 19.23 22.54 ± 9.08 32.93 ± 7.36 (40.34) 0.001 According to the linear regression analysis (Table 4), the prediction model for EI (R2 = 0.816) was determined by group (absence or presence of plantar fasciitis) and weight. For EV prediction model (R2 = 0.243) was determined by group. The rest of the independent variables did not report signiﬁcant differences between the case and control groups. Diagnostics 2021, 11, 1884 6 of 9 Table 3. 3. Results Bonferroni correction for Numerical Rating Scale (NRS) during sport activity and at palpation in soleus injury, Echointensity (EI), and Echovariation (EV) variables. Data Group Group Mean Difference (95% CI Minimum–Maximum) p Value NRS during activity (points) Injury Type 1 Injury Type 2 −0.500 (−1.35–0.28) 0.316 Injury Type 1 Healthy 5.392 (4.61–6.25) 0.001 Healthy Injury Type 2 −5.89 (5.14–6.78) 0.001 NRS at palpation (points) Injury Type 1 Injury Type 2 −0.696 (−1.24–0.14) 0.052 Injury Type 1 Healthy 5.339 (5.23–6.62) 0.001 Healthy Injury Type 2 −6.035 (−6.07–−4.68) 0.001 Echointensity (EI) Injury Type 1 Injury Type 2 −28.976 (−34.72—-−23.11) 0.001 Injury Type 1 Healthy −44.887 (−51.90–−40.30) 0.001 Healthy Injury Type 2 15.911 (11.38–22.99) 0.001 Echovariation (EV) Injury Type 1 Injury Type 2 30.673 (22.63–43.44) 0.001 Injury Type 1 Healthy 20.279 (13.62–34.44) 0.001 Healthy Injury Type 2 10.393 (−1.39–19.41) 0.010 Table 3. Bonferroni correction for Numerical Rating Scale (NRS) during sport activity and at palpation in soleus injury, Echointensity (EI), and Echovariation (EV) variables. Table 4. Multivariate predictive analysis for EI and EV variables for patients with plantar fasciitis and controls. ble 4. Multivariate predictive analysis for EI and EV variables for patients with plantar fasciitis d l Parameter Model Beta Coefﬁcient Model R2 EI 96.914 0.816 22.561 Group 0.874 ‡ −59.737 * Weight (kg) −0.249 ‡ EV 7.371 −12.173 * Group −0.464 ‡ 0.243 Abbreviations: EI, echointensity; EV, echovariation. * Multiplay: Group (control = 0; Plantar fasciitis = 1); ‡ p-value < 0.001 for a 95% conﬁdence interval was shown. 4. Discussion The main ﬁnding of the present study was to provide a better understanding and new insights about different injured soleus types located in the IMT by ultrasound parameters. In this study, an echotexture classiﬁcation of injuries affecting the IMT of the soleus muscle is proposed, based on ﬁndings in the sports population. The classiﬁcation may be useful in the clinical setting for the diagnosis, follow up and prevention of musculoskeletal injuries. Speciﬁcally, the results show that EV could be a muscle biomarker in athletes with soleus pathology. According to the echogenic pattern, the classiﬁcation of soleus tears that authors propose is: Injury Type 1, identiﬁed by a hypoechoic area and characterized by a higher EV; and Injury Type 2, identiﬁed by a ﬁbrotic area and characterized by a lower EV. The echogenic pattern of healthy people should be an intermediate value between both injured soleus types. An intra and inter-rater reliability test was carried out with the objective to increase the reliability of the scores of the assessments. They showed a high reliability in all parameters for this population. Considering US as the ﬁrst-line imaging modality to diagnosis and follow-up of musculoskeletal injuries [10], it was not necessary to perform magnetic resonance imaging like other studies [27]. Nowadays, in the ﬁeld of sports, there is an urgent need to improve the interpretation of ultrasound images to establish accuracy diagnoses and apply the best treatment options, from a regenerative therapeutic point of view [28,29]. Hence, there is a huge enthusiasm in knowing soleus muscle features and having a better knowledge of pathological characteristics [6,7,30,31]. Based on the special characteristics of IMT tears, these injuries are often not easily recognized, both by therapists and athletes. Indeed, in many scenarios, soleus tears does not require athletes to stop training or playing games [8]. According to our data, authors suggest that the analysis of echotexture variables, such as EV, may be applied as a complementary test for the management of soleus injuries. The results described in this research are in line with previous studies that hypothesize the possibility to obtain a muscle biomarker with ultrasonography. In fact, the EV had been analyzed in different muscles of patients with amyotrophic lateral [19] and athletes Diagnostics 2021, 11, 1884 7 of 9 with chronic lumbopelvic pain [21]. 4.1. Clinical Applications The most relevant clinical implication of this study was the proposal of the EV as a muscle biomarker for the classiﬁcation of soleus injured muscle, based on echogenic patterns. Hence, EV may be used in future studies for the prevention as well as for the management and monitoring of muscle tears. This muscle biomarker may even determine the most effective treatment options for the recovery process. More research on treatment options are needed. It is crucial to highlight that the proposed US-based classiﬁcation is complementary to other muscle injury classiﬁcations. 4. Discussion To the best of our knowledge, there are limited studies about the relationship between EV and different classiﬁcations of muscle injuries in sport ﬁeld. Previously, De-la-Cruz-Torres et al. [7] compared the EV of patients with IMT injury of soleus muscle and healthy people, but they found no differences. Authors suggest that this could be because it is needed to determine the injury status to determine its echogenic pattern. Pathologies with the same symptomatology and lesional mechanism may be characterized by a different histology and, therefore, the diagnosis and treatment may be different. Authors hypothesize that IMT tears of the soleus muscle can be considered (Figure 2, Table 3): on the one hand, a re-injury characterized by a hypoechoic image, compatible with edema, and a higher EV (Injury type 1); and on the other hand, a ﬁbrotic injury characterized by a ﬁbrotic image, compatible with ﬁbrosis, and a lower EV (Injury type 2). In this line, future studies are also needed to better explain the characteristics of the pathological features on injured muscle. Knowledge of basic anatomic and histology in combination with pathology imaging are required. 4.2. Limitations and Future Lines The present study had several limitations. In this study, only ultrasound was used as a diagnostic tool. Comparing different diagnostic strategies for muscle injuries, such as magnetic resonance imaging or laboratory tests, could help to better understand the pathological characteristics of muscle injuries. Furthermore, to establish the role of EV as a diagnostic or follow-up biomarker, this echotexture analysis must be applied throughout the entire recovery process and after applying the treatments. Furthermore, future research may be considered in other muscle and soft tissues disturbances. 5. Conclusions Authors suggest that EV may be useful to classify different types of soleus muscle pathology according to the echogenicity pattern. An innovative proposed US-based clas- siﬁcation system for soleus tears may be used to guide treatment decisions for patients with central tendon injury of soleus muscle. The proposed classiﬁcation may be easily performed in the clinical setting with the aim to improve the diagnosis, management and monitoring their return to sport after soleus muscle injury. Author Contributions: Conceptualization, B.D.-l.-C.-T., and C.R.-M.; data curation, B.D.-l.-C.-T.; methodology, B.D.-l.-C.-T., and C.R.-M.; formal analysis, C.R.-M.; investigation, B.D.-l.-C.-T., and C.R.-M.; writing—original draft preparation, B.D.-l.-C.-T., and C.R.-M.; writing—review and editing, B.D.-l.-C.-T., and C.R.-M. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Funding: This research received no external funding. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of University Hospital Virgen Macarena-Virgen del Rocio, (protocol code 0663-N-19; approval date 27 July 2019). Informed Consent Statement: Written informed consent has been obtained from the patients to publish this paper. Data Availability Statement: Data sharing is not applicable to this article. Data Availability Statement: Data sharing is not applicable to this article. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 8 of 9 Diagnostics 2021, 11, 1884 References 1. Green, B.; Lin, M.; Schache, A.G.; McClelland, J.A.; Semciw, A.; Rotstein, A.; Cook, J.; Pizzari, T. Calf muscle strain injuries in elite Australian Football players: A descriptive epidemiological evaluation. Scand. J. Med. Sci. Sports 2019, 30, 174–184. [CrossRef] 2. Brukner, P.D.; Connell, D. ‘Serious thigh muscle strains’: Beware the intramuscular tendon which plays an important role in difﬁcult hamstring and quadriceps muscle strains. Br. J. Sports Med. 2015, 50, 205–208. [CrossRef] [PubMed] g q p 3. Koulouris, G.; Ting, A.Y.I.; Jhamb, A.; Connell, D.; Kavanagh, E.C. Magnetic resonance imaging ﬁndi muscle complex. Skelet. Radiol. 2007, 36, 921–927. [CrossRef] p 4. Courthaliac, C.; Weilbacher, H. Imaging of painful calf in athletes. J. Radiol. 2007, 88, 200–208. [CrossR 5. ´Smigielski, R. Management of Partial Tears of the Gastro-Soleus Complex. Clin. Sports Med. 2008, 27, 2 4. Courthaliac, C.; Weilbacher, H. Imaging of painful calf in athletes. J. Radiol. 2007, 88, 200–208. [CrossRef] 5. ´Smigielski, R. Management of Partial Tears of the Gastro-Soleus Complex. Clin. Sports Med. 2008, 27, 219–229. [CrossRef] g g p p 6. Balius, R.; Alomar, X.; Rodas, G.; Miguel-Pérez, M.; Pedret, C.; Dobado, M.C.; Blasi, J.; Koulouris, G. The soleus muscle: MRI, anatomic and histologic ﬁndings in cadavers with clinical correlation of strain injury distribution. Skelet. Radiol. 2012, 42, 521–530. [CrossRef] 7. De-La-Cruz-Torres, B.; Navarro-Flores, E.; López-López, D.; Romero-Morales, C. Ultrasound Imaging Evaluation of Textural Features in Athletes with Soleus Pathology—A Novel Case-Control Study. Int. J. Environ. Res. Public Health 2021, 18, 1983. [CrossRef] ukner, P.; Cook, J.L.; Purdam, C.R. Does the intramuscular tendon act like a free tendon? Br. J. Sports Med. 2 rossRef] 8. Brukner, P.; Cook, J.L.; Purdam, C.R. Does the intramuscular tendon act like a free tendon? Br. J. Sports Med. 2018, 52, 1227–1228. [CrossRef] 9. Kilic-Erkek, O.; Caner, V.; Abban-Mete, G.; Baris, I.C.; Bor-Kucukatay, M. Determination of the p damage and regeneration in response to acute and long-term swimming exercise in mice. Life Sci. 2 9. Kilic-Erkek, O.; Caner, V.; Abban-Mete, G.; Baris, I.C.; Bor-Kucukatay, M. Determination of the pathways of potential muscle damage and regeneration in response to acute and long-term swimming exercise in mice. Life Sci. 2021, 272, 119265. [CrossRef] 10. Loizides, A.; Gruber, H.; Peer, S.; Plaikner, M. Muscular injuries of athletes: Importance of ultrasound. Radiologe 2017, 57, 1019–1028. References J. Am. Med. Dir. Assoc. 2017, 18, 290–300. [CrossRef] 16. Pillen, S.; Tak, R.O.; Zwarts, M.J.; Lammens, M.; Verrijp, K.N.; Arts, I.M.; van der Laak, J.A.; Hoogerbrugge, P.M.; van Engelen, B.G.; Verrips, A. Skeletal Muscle Ultrasound: Correlation Between Fibrous Tissue and Echo Intensity. Ultrasound Med. Biol. 2009, 35, 443–446. [CrossRef] [PubMed] 17. Martínez-Payá, J.J.; Ríos-Díaz, J.; Medina-Mirapeix, F.; Vázquez-Costa, J.F.; Del Baño-Aledo, M.E. Monitoring Progression of Amyotrophic Lateral Sclerosis Using Ultrasound Morpho-Textural Muscle Biomarkers: A Pilot Study. Ultrasound. Med. Biol. 2018, 44, 102–109. [CrossRef] [PubMed] 18. Yitzchaki, N.; Zhu, W.G.; Kuehne, T.E.; Vasenina, E.; Dankel, S.J.; Buckner, S.L. An examination of changes in skeletal muscle thickness, echo intensity, strength and soreness following resistance exercise. Clin. Physiol. Funct. Imaging 2020, 40, 238–244. [CrossRef] 19. Ríos-Díaz, J.; Martínez-Payá, J.J.; del Baño-Aledo, M.E.; de Groot-Ferrando, A.; Botía-Castillo, P.; Fernández-Rodríguez, D. Sonoelastography of Plantar Fascia: Reproducibility and Pattern Description in Healthy Subjects and Symptomatic Subjects. Ultrasound Med. Biol. 2015, 41, 2605–2613. [CrossRef] 20. Almazán-Polo, J.; López-López, D.; Romero-Morales, C.; Rodríguez-Sanz, D.; Becerro-De-Bengoa-Vallejo, R.; Losa-Iglesias, M.E.; Bravo-Aguilar, M.; Calvo-Lobo, C. Quantitative Ultrasound Imaging Differences in Multiﬁdus and Thoracolumbar Fasciae between Athletes with and without Chronic Lumbopelvic Pain: A Case-Control Study. J. Clin. Med. 2020, 9, 2647. [CrossRef] 21. De-La-Cruz-Torres, B.; Barrera-García-Martín, I.; Almazán-Polo, J.; Jaén-Crespo, G.; Romero-Morales, C. Ultrasound imaging evaluation of structural and textural features in asymptomatic achilles tendons in pre-professional dancers: A cross-sectional study. Phys. Ther. Sport 2020, 44, 85–91. [CrossRef] [PubMed] y y p 22. Martínez-Payá, J.J.; Del Baño-Aledo, M.E.; Ríos-Díaz, J.; Tembl-Ferrairó, J.I.; Vázquez-Costa, J.F.; Medina-Mirapeix, F. Muscular Echovariation: A New Biomarker in Amyotrophic Lateral Sclerosis. Ultrasound. Med. Biol. 2017, 43, 1153–1162. [CrossRef] 22. Martínez-Payá, J.J.; Del Baño-Aledo, M.E.; Ríos-Díaz, J.; Tembl-Ferrairó, J.I.; Vázquez-Costa, J.F.; Medina-Mirapeix, F. Muscular Echovariation: A New Biomarker in Amyotrophic Lateral Sclerosis. Ultrasound. Med. Biol. 2017, 43, 1153–1162. [CrossRef] 23. Von Elm, E.; Altman, D.G.; Egger, M.; Pocock, S.J.; Gøtzsche, P.C.; Vandenbroucke, J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. Int. J. Surg. 2014, 12, 1495–1499. [CrossRef] y p 23. Von Elm, E.; Altman, D.G.; Egger, M.; Pocock, S.J.; Gøtzsche, P.C.; Vandenbroucke, J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. Int. J. Surg. 2014, 12, 1495–1499. [CrossRef] 24. De-La-Cruz-Torres, B.; Barrera-García-Martín, I.; Valera-Garrido, F.; Minaya-Muñoz, F.; Romero-Morales, C. 24. De-La-Cruz-Torres, B.; Barrera-García-Martín, I.; Valera-Garrido, F.; Minaya-Muñoz, F.; Romero-Morales, C. Ultrasound-Guided Percutaneous Needle Electrolysis in Dancers with Chronic Soleus Injury: A Randomized Clinical Trial. Evidence-Based Complement. Altern. Med. 2020, 2020, 1–8. [CrossRef] References [CrossRef] ge ege e o espo se o c e o g e s g e e c se ce f S 0 , , 9 65 [C oss e ] 10. Loizides, A.; Gruber, H.; Peer, S.; Plaikner, M. Muscular injuries of athletes: Importance of ultrasound. Radiologe 2017, 57, 1019–1028. [CrossRef] 11. Fukumoto, Y.; Ikezoe, T.; Yamada, Y.; Tsukagoshi, R.; Nakamura, M.; Mori, N.; Kimura, M.; Ichihashi, N. Skeletal muscle quality assessed from echo intensity is associated with muscle strength of middle-aged and elderly persons. Graefe’s Arch. Clin. Exp. Ophthalmol. 2011, 112, 1519–1525. [CrossRef] [PubMed] 12. Wong, V.; Abe, T.; Chatakondi, R.N.; Bell, Z.W.; Spitz, R.W.; Dankel, S.J.; Loenneke, J.P. The inﬂuence of biological sex and cuff width on muscle swelling, echo intensity, and the fatigue response to blood ﬂow restricted exercise. J. Sports Sci. 2019, 37, 1865–1873. [CrossRef] 13. Martínez-Payá, J.J.; Ríos-Díaz, J.; Del Baño-Aledo, M.E.; Tembl-Ferrairó, J.I.; Vazquez-Costa, J.F.; Medina- Muscle Ultrasonography Using Textural Analysis in Amyotrophic Lateral Sclerosis. Ultrason. Imaging 201 13. Martínez-Payá, J.J.; Ríos-Díaz, J.; Del Baño-Aledo, M.E.; Tembl-Ferrairó, J.I.; Vazquez-Costa, J.F.; Medina-Mirapeix, F. Quantitative Muscle Ultrasonography Using Textural Analysis in Amyotrophic Lateral Sclerosis. Ultrason. Imaging 2017, 39, 357–368. [CrossRef] 14. Arts, I.M.; Overeem, S.; Pillen, S.; Kleine, B.U.; Boekestein, W.A.; Zwarts, M.J.; Schelhaas, H.J. Muscle ultrasonography: A diagnostic tool for amyotrophic lateral sclerosis Clin Neurophysiol 2012 123 1662 1667 [CrossRef] [PubMed] g p y g y y p g g 14. Arts, I.M.; Overeem, S.; Pillen, S.; Kleine, B.U.; Boekestein, W.A.; Zwarts, M.J.; Schelhaas, H.J. Muscle ultrasonography: A diagnostic tool for amyotrophic lateral sclerosis. Clin. Neurophysiol. 2012, 123, 1662–1667. [CrossRef] [PubMed] ts, M ; Ove ee , S ; e , S ; e e, U ; oekeste , W ; wa ts, M J ; Sc e aas, J Musc e u t aso og ap y diagnostic tool for amyotrophic lateral sclerosis. Clin. Neurophysiol. 2012, 123, 1662–1667. [CrossRef] [PubMed] 15. Ticinesi, A.; Meschi, T.; Narici, M.V.; Lauretani, F.; Maggio, M. Muscle Ultrasound and Sarcopenia in Older Individuals: A Clinical Perspective J Am Med Dir Assoc 2017 18 290–300 [CrossRef] g y p p y 15. Ticinesi, A.; Meschi, T.; Narici, M.V.; Lauretani, F.; Maggio, M. Muscle Ultrasound and Sarcopenia in Older Individuals: A Clinical Perspective. J. Am. Med. Dir. Assoc. 2017, 18, 290–300. [CrossRef] 15. Ticinesi, A.; Meschi, T.; Narici, M.V.; Lauretani, F.; Maggio, M. Muscle Ultrasound and Sarcopenia in Old Perspective. References Ultrasound-Guided Percutaneous Needle Electrolysis in Dancers with Chronic Soleus Injury: A Randomized Clinical Trial. Evidence-Based Complement. Altern. Med. 2020, 2020, 1–8. [CrossRef] 9 of 9 Diagnostics 2021, 11, 1884 25. Downie, W.W.; Leatham, P.A.; Rhind, V.M.; Wright, V.; Branco, J.A.; Anderson, J.A. Studies with pain rating scales. Ann. Rheum. Dis. 1978, 37, 378–381. [CrossRef] 26. Caresio, C.; Molinari, F.; Emanuel, G.; Minetto, M.A. Muscle echointensity: Reliability and conditioning factors. Clin. Physiol. Funct. Imaging 2014, 35, 393–403. [CrossRef] g g 27. Pedret, C.; Balius, R.; Blasi, M.; Dávila, F.; Aramendi, J.F.; Masci, L.; De La Fuente, J. Ultrasound classiﬁcation of medial gastrocnemious injuries. Scand. J. Med. Sci. Sports 2020, 30, 2456–2465. [CrossRef] g j 28. Indino, C.; D’Ambrosi, R.; Usuelli, F.G. Biologics in the Treatment of Achilles Tendon Pathologies. Foot Ankle Clin. 2019, 24, 471–493. [CrossRef] [PubMed] 29. Sánchez-Sánchez, J.L.; Calderón-Díez, L.; Herrero-Turrión, J.; Méndez-Sánchez, R.; Arias-Buría, J.L.; Fernández-De-Las-Peñas, C. Changes in Gene Expression Associated with Collagen Regeneration and Remodeling of Extracellular Matrix after Percutaneous Electrolysis on Collagenase-Induced Achilles Tendinopathy in an Experimental Animal Model: A Pilot Study. J. Clin. Med. 2020, 9, 3316. [CrossRef] 30. Giacchino, M.; Caresio, C.; Gorji, N.; Molinari, F.; Massazza, G.; Minetto, M. Quantitative analysis of patellar tendon size and structure in asymptomatic professional players: Sonographic study. Muscle Ligaments Tendons. J. 2019, 7, 449. [CrossRef] l l b l d 31. Romero-Morales, C.; Calvo-Lobo, C.; Navarro-Flores, E.; Mazoteras-Pardo, V.; García-Bermejo, P.; López-López, D.; Martínez- Jiménez, E.; De-La-Cruz-Torres, B. M-mode Ultrasound Examination of Soleus Muscle in Healthy Subjects: Intra- and Inter-Rater Reliability Study. Healthcare 2020, 8, 555. [CrossRef] [PubMed]
https://openalex.org/W2971237683	https://revistas.up.edu.pe/index.php/forseti/article/download/1087/1241	Spanish; Castilian	null	Negociando con Ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla	Forseti	2,018	cc-by	11,612	Forseti. Revista de Derecho. Edición de aniversario, Lima, DERUP, 2018, pp 57-83 ** Bachiller en Derecho por la Universidad del Pacífico. Asociada de las áreas de Venture Capital y Fusiones y Adquisiciones del Estudio Rebaza, Alcázar y & De Las Casas. * Abogada por la Universidad de Lima y LL.M. por Stanford University, reconocida cuna del emprendimiento tecnológico. La autora es actualmente socia del Estudio Rebaza, Alcázar & De Las Casas en el que lidera el área de Venture Capital y ha participado en gran parte de las principales transacciones realizadas en el eco- sistema emprendedor peruano, representando a inversionistas ángeles, family offices, fondos de inversión de venture capital, así como también a emprendedores. Asimismo, Alexandra es socia del Área de Fusiones y Adquisiciones. ** Bachiller en Derecho por la Universidad del Pacífico. Asociada de las áreas de Venture Capital y Fusiones y Adquisiciones del Estudio Rebaza, Alcázar y & De Las Casas. * Abogada por la Universidad de Lima y LL.M. por Stanford University, reconocida cuna del emprendimiento tecnológico. La autora es actualmente socia del Estudio Rebaza, Alcázar & De Las Casas en el que lidera el área de Venture Capital y ha participado en gran parte de las principales transacciones realizadas en el eco- sistema emprendedor peruano, representando a inversionistas ángeles, family offices, fondos de inversión de venture capital, así como también a emprendedores. Asimismo, Alexandra es socia del Área de Fusiones y Adquisiciones. I. Introducción Las pymes (que responden a la abreviatura de pequeñas y micro empresas) siempre han existido en nuestro país, aunque, en la actua- lidad, es cada vez más frecuente escuchar el término startup. Si bien hace tan sólo unos años atrás éste era un término desconocido para la mayoría de peruanos, la palabra startup se ha abierto paso en nuestro vocabulario, lecturas y conversaciones. Evoca en nosotros pro- ductos de alta tecnología, jóvenes cumpliendo sus sueños de triunfar y convertirse en referentes creando sus propias empresas en lugar de trabajar para otros, así como rondas de inversión de varios millones. Algo que antes parecía lejano y que muchos consideraban que sólo se trataba de una moda, o que sólo tenía sentido en lugares apartados como Silicon Valley, Tel Aviv o Bangalore, es hoy una realidad en nuestro país, donde durante los últimos años se ha creado todo un ecosistema emprendedor que aglutina no sólo a las propias startups, sino a un importante número de actores relacionados: emprendedores, incubadoras, aceleradoras, espacios de coworking, entre otros, inclu- yendo a los famosos inversionistas, con un papel cada vez más activo. Pero, para utilizar con propiedad el término startup ¿qué empresas deben ser consideradas realmente dentro de esta categoría? Una startup es aquella empresa que se encuentra en edad temprana o es de reciente creación y presenta grandes posibilidades de crecimiento. A diferencia de una pyme, una startup se caracteriza por ser un negocio escalable y por su capacidad de crecer de una forma mucho más rápida y eficiente. Asimismo, una startup tiene una fuerte relación con la tecnología y se caracteriza por desarrollar negocios con ideas muy innovadoras para satisfacer las necesidades o problemas de su público objetivo, desarrollando para ello soluciones tecnológicas que satisfagan las necesidades identificadas. Es así que el carácter innovador y disruptivo de las startups las enfrenta a múltiples retos desde el punto de vista legal a lo largo de su ciclo de vida, que van desde la protección de la idea innovadora, la regulación de la relación entre los socios fundadores (conocidos en inglés como founders), la contratación de personal clave, la redacción de acuerdos comerciales con terceros para el desarrollo y monetiza- ción del negocio, el levantamiento de capital a través de la realización de rondas de inversión, etc. Negociando con Ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Alexandra Orbezo Grahammer * Josefina Arana Baldeón ** Resumen.- Las autoras presentan diversos aspectos legales relaciona- dos con las startups, específicamente, la hoja de términos y condicio- nes que el emprendedor utiliza en negociaciones con inversionistas en las rondas de financiamiento. Además, analizan distintas formas de levantamiento de capital orientadas a inversionistas ángeles, implementadas en otras jurisdicciones o que se han adaptado a la legislación local. Abstract.- The authors present legal aspects of startups, specifica- lly the term sheet that the entrepreneur uses in negotiations with investors in the equity financing rounds. They also analyze the different forms of capital raising focused on angel investors, which have been implemented in other jurisdictions or have managed to adapt to local legislation. Ѳeamdra r0eo oseCma rama Edición de Aniversario I. Introducción Es por ello que las disciplinas tradicio- nales del derecho se enfrentan a la obligación de reconfigurarse ante las nuevas realidades a las que se enfrentan las startups, los empren- dedores e inversionistas. El rol del abogado comprende el reto de dar 58 58 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla respuesta a situaciones jurídicamente no reguladas y de adaptar al derecho peruano fórmulas jurídicas comunes en otros países con legislaciones más acondicionadas a las necesidades del ecosistema emprendedor, pero que no existen como tales en el Perú. Asimismo, el Derecho peruano, en sus distintas ramas, exige también un ejercicio de revisión y reformulación ante estos nuevos retos. En este contexto, el presente artículo tiene por objetivo presentar una introducción a los aspectos legales clave vinculados a la hoja de términos y condiciones (conocida en inglés como term sheet), que constituye el primer documento legal que el emprendedor negocia con los inversionistas en las rondas de financiamiento de capital, enfocándose específicamente en aquel term sheet que es negociado en las rondas de capital semilla con inversionistas ángeles, y que es celebrado bajo ley peruana. Adicionalmente, a lo largo del desarrollo del presente artículo iremos analizando algunas de las formas legales a través de las cuales se vienen adaptado a la legislación local estruc- turas o figuras jurídicas utilizadas en otras jurisdicciones en el marco de la implementación de las rondas de levantamiento de capital dirigidas a inversionistas ángeles. Consideramos que los emprendedores e inversionistas ángeles esta- rán mejor preparados para obtener y entregar capital, respectiva- mente, si ambas partes entienden el proceso, y conocen de manera anticipada los términos de la transacción y las potenciales comple- jidades que surgirán en el camino. Y parte importante del rol del abogado es la de educar y proveer la asistencia legal necesaria para que el entendimiento y conocimiento antes señalados tengan lugar. Finalmente, antes de pasar a desarrollar el contenido del presente artículo, es importante anotar que como en todo sector, en el ámbito de las startups también existe un vocabulario o terminología propia o de uso habitual que resulta importante conocer para moverse con cierta fluidez dentro del entorno emprendedor. I. Introducción Con el objetivo de tener una noción básica de dicho vocabulario y comprender mejor la información proporcionada en el presente artículo, iremos a lo largo del desarrollo del mismo, repasando aquella terminología básica del ecosistema de los startups que consideramos resulta útil manejar. III. El capital semilla El capital semilla representa aquellas inversiones que se realizan en una startup que está en su primera fase o fase temprana, es decir, en empresas que aún están validando su modelo de negocio y, por tanto, traccionando sus primeras ventas y clientes. Son operaciones de alto riesgo en las que usualmente intervienen las 3Fs (que responden a la abreviatura en inglés de friends, family and fools, traducido al español como amigos, familiares y tontos) y los inversionistas ángeles. La primera ronda de financiamiento de una startup es uno de los eventos más importantes dentro de su ciclo de vida, ya que establece el marco para futuros financiamientos. II. Fuentes de financiamiento Luego de que una startup se encuentra constituida y operando, una de sus continuas tareas es la de levantar los fondos suficientes para financiar su negocio en cada una de las etapas de su desarrollo. 59 59 Ѳeamdra r0eo oseCma rama Edición de Aniversario Dado que en una etapa temprana las startups no son empresas lo suficientemente grandes, ni cuentan con activos fijos de valor signifi- cativo, difícilmente éstas podrán acceder a un financiamiento tradi- cional a través del mercado de capitales o a través de bancos u otras entidades de crédito. Como consecuencia de ello, los emprendedores deberán salir al mer - cado a levantar fondos de distintos tipos de inversionistas a cambio de una participación en el capital social de la startup, con el objetivo de que ésta logre desarrollar su negocio, aumente su valor y una vez madurada la inversión, se genere un evento de liquidez (conocido en inglés como liquidity event) a través del cual el inversionista se retire obteniendo un retorno elevado. Las principales estrategias de salida (conocida en inglés como exit) que se plantean para una inversión de este tipo son la venta o fusión de la startup con un inversionista estra- tégico, o la realización de una oferta pública primaria de las acciones de la compañía. El perfil del inversionista y el tipo de instrumento de inversión ade- cuado para cada ronda de financiamiento depende sustancialmente de la fase del ciclo de vida en la que se encuentre la startup. En este contexto, los abogados juegan un papel importante en educar a los emprendedores sobre la forma de financiamiento que resulta más apropiada para cada una de las etapas de desarrollo en la que se encuentre su startup. a. Friends, family and fools El tipo de financiamiento del friends, family and fools o simplemente, friends and family, ocurre en una etapa relativamente temprana en 60 60 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla la vida de la startup, usualmente luego de que los socios fundadores han ya invertido todo el capital propio que les es posible. Los inver- sionistas que componen el financiamiento del friends and family pertenecen al entorno cercano de los socios fundadores (familiares, amigos, entre otros), y buscan ayudar a estos últimos en el desarrollo de su proyecto de emprendimiento a través de la infusión de dinero en efectivo en sus startups. Este dinero por lo general es necesario para finalizar el prototipo del producto o servicio que brindará la startup y dar el salto al mercado con el producto mínimo viable. Esta fase de financiamiento usualmente se concreta a través de la suscripción de acciones comunes con derecho a voto, sin ningún tipo de derecho especial a favor del inversionista. Los inversionistas de la ronda del friends and family son quienes muy probablemente se diluirán significativamente en las siguientes rondas de financia- miento, conforme vayan ingresando inversionistas más instituciona- les, puesto que no realizarán nuevos desembolsos de capital. b. Inversionistas ángeles El capital aportado por los inversionistas ángeles cubre la brecha del financiamiento de una startup entre los fondos levantados del friends and family como parte del capital semilla y el que se levantará más adelante de los fondos de inversión de capital riesgo (conocidos en inglés como venture capital funds o VCs) quienes invierten en una etapa más avanzada de desarrollo de la startup. Inicialmente los inversionistas ángeles eran de hecho familiares y amigos de los emprendedores, o individuos con alto patrimonio neto individual (conocidos en inglés como high net worth individuals), pero no necesariamente inversionistas sofisticados. A través de los años, esta categoría de inversionistas ha mutado para convertirse en una red más formal de individuos con alto patrimonio neto indivi- dual, quienes son más sofisticados y se organizan en redes, grupos o clubes de inversionistas ángeles para compartir esfuerzos y unir sus capitales de inversión, y están más involucrados de manera directa en la supervisión de sus inversiones. Esta categoría en algunos casos incluye individuos que anteriormente emprendieron y que tuvieron exitosos exits en sus propias startups, motivo por el cual cuentan con un patrimonio significativo para invertir. Asimismo, conforme el mercado de inversionistas ángeles ha ido madurando, los tipos y sofisticación de los instrumentos de inversión también han evolucionado. La manera más sencilla (pero no por ello necesariamente la única utilizada) de estructurar la inversión de un 61 61 Ѳeamdra r0eo oseCma rama Edición de Aniversario inversionista ángel es a través de la figura legal de una deuda con- vertible en acciones. La deuda convertible en acciones en jurisdicciones como los Estados Unidos de América consiste típicamente en la suscripción de un note purchase agreement (que es el contrato de compraventa para la adquisi- ción del instrumento de deuda convertible en acciones) y un converti- ble promissory note (traducido al español como nota convertible, que es el instrumento en sí que representa la deuda convertible en acciones). Esta estructura permite a la startup (vendedora de la nota convertible) el levantamiento de capital. El inversionista ángel que adquiere la nota convertible invierte en la startup a través de la compra de dicho instrumento. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla • Luego de un tiempo, la startup levanta capital de un fondo de venture capital en una Serie A, ronda de inversión que usualmente tiene lugar pasado el umbral de rentabilidad (conocido en inglés como break-even) que es el punto de inflexión donde los ingresos de la operación de la startup son superiores a los gastos. Imaginemos que la valorización de la startup antes de dicha ronda es de US$ 10,000,000 y el precio de suscripción por acción fijado para la misma es de US$ 5.00. Si aplicáramos el descuento de 20%, el precio de con- versión por acción que regirá para el inversionista ángel será de US$ 4.00 (US$ 5.00 multiplicado por (1-20%). Como consecuencia de ello, el inversionista ángel recibiría un total de 6,250 acciones Serie A (que resultan de dividir su inversión de US$ 25,000 en la nota convertible entre el precio de conversión por acción de US$ 4.00). • Luego de un tiempo, la startup levanta capital de un fondo de venture capital en una Serie A, ronda de inversión que usualmente tiene lugar pasado el umbral de rentabilidad (conocido en inglés como break-even) que es el punto de inflexión donde los ingresos de la operación de la startup son superiores a los gastos. Imaginemos que la valorización de la startup antes de dicha ronda es de US$ 10,000,000 y el precio de suscripción por acción fijado para la misma es de US$ 5.00. Si aplicáramos el descuento de 20%, el precio de con- versión por acción que regirá para el inversionista ángel será de US$ 4.00 (US$ 5.00 multiplicado por (1-20%). Como consecuencia de ello, el inversionista ángel recibiría un total de 6,250 acciones Serie A (que resultan de dividir su inversión de US$ 25,000 en la nota convertible entre el precio de conversión por acción de US$ 4.00). Por el contrario, si aplicáramos para el inversionista ángel el valua- tion cap de US$ 5,000,000 en reemplazo de la valorización de la startup fijada en US$ 10,000,000, el precio de conversión por acción que regirá para el inversionista ángel será de US$ 2.50 por acción (que resultan de multiplicar el precio de suscripción por acción de US$ 5.00 por el resultado de la diferencia entre el valuation cap y la valorización de la startup previa a la ronda (US$ 5,000,000 entre US$ 10,000,000)). b. Inversionistas ángeles Bajo esta estructura, la nota convertible se convertirá automática- mente en el mismo tipo de acciones que sean emitidas en la siguiente ronda de financiamiento que cumpla ciertos umbrales mínimos que, por ejemplo, pueden haber sido establecidos en función al monto total recaudado o al tipo de inversionista que participa en dicha ronda. En este sentido, las acciones que recibirán los titulares de las notas con- vertibles tendrán los mismos derechos y condiciones acordados para la ronda de financiamiento que gatilla la conversión, los cuales son negociadas entre el inversionista que lidera dicha ronda (lead investor) y el emprendedor. Usualmente, el monto total del principal más los intereses devenga- dos por la nota convertible, de ser éste el caso, serán convertidos en acciones aplicando para efectos de dicha conversión el mismo precio de suscripción por acción fijado para los nuevos inversionistas que participan en la ronda de financiamiento que gatilla la conversión, menos la aplicación de un descuento, y/o en algunos casos también, la aplicación de un monto límite máximo de valorización (conocido en inglés como valuation cap) previamente negociados, lo que resulte en un número mayor de acciones para el inversionista ángel tenedor de la nota convertible. Dicho descuento y valuation cap, de ser el caso, premia al inversionista ángel por tomar un mayor nivel de riesgo al haber invertido en la startup en una etapa más temprana de la que un fondo de inversión de capital de riesgo estaría dispuesto a hacerlo. A continuación, un ejemplo numérico que busca reflejar la estructura antes señalada, en el cual, para efectos prácticos de cálculo, hemos omitido tomar en consideración cualquier interés devengado: • Imaginemos una nota convertible por la que el inversionista ángel desembolsó la suma de US$ 25,000 y respecto a la cual se pactó un valuation cap de US$ 5,000,000 y un descuento de 20%. • Imaginemos una nota convertible por la que el inversionista ángel desembolsó la suma de US$ 25,000 y respecto a la cual se pactó un valuation cap de US$ 5,000,000 y un descuento de 20%. 62 62 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Si bien las notas convertibles permiten que las primeras rondas de levantamiento de capital semilla sean más sencillas y rápidas puesto que evitan el tener que negociar en dicho momento la valorización de la startup permitiendo a las partes concentrarse únicamente en la negociación del descuento y/o del valuation cap, de ser el caso, tienen como desventaja que son un instrumento al cual le resultan aplicables las normas que regulan el otorgamiento y adquisición de deuda (por ejemplo, el registro contable como un pasivo en el balance de la startup, el devengo de intereses, el plazo máximo de repago, entre otros). Es así que otro instrumento ha sido desarrollado en los Estados Unidos de América para esta etapa de financiamiento de capital semilla. Hablamos del SAFE (que responde a la abreviatura en inglés de simple agreement for future equity). Su estructura es similar a la de una nota convertible en el sentido que el inversionista ángel desembolsa una suma de dinero para suscribir el SAFE a cambio de un derecho a obtener en el futuro una participación en el capital social de la startup. El inversionista ángel recibe las acciones futuras cuando la startup realiza la primera ronda de financiamiento que cumpla las condicio- nes pactadas (por ejemplo, que sea la primera ronda en la que se haya acordado una valorización), o cuando ocurra un evento de liquidez (como podría ser la venta total o parcial de la startup o un cambio de control), lo que suceda primero. La ventaja es que, a diferencia de la nota convertible, el SAFE no es un instrumento de deuda y, por lo tanto, no tiene que ser registrado contablemente como un pasivo de la startup. Se asemeja más a una opción de suscripción de acciones en la que el precio es pagado de manera anticipada. Ello permite que el SAFE no genere intereses y que no establezca una fecha de venci- miento, por lo cual, en muchas jurisdicciones de los Estados Unidos de América, no se encuentra sujeto a la regulación aplicable a la deuda. Una figura legal que viene siendo utilizada en el ecosistema local, que busca adecuar a la legislación peruana las estructuras antes descritas, es el préstamo capitalizable. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Como consecuencia de ello, el inversionista ángel recibiría un total de 10,000 acciones Serie A (que resultan de dividir la inversión de US$ 25,000 en la nota convertible entre el precio de conversión por acción de US$ 2.50). En este sentido, finalmente el inversionista ángel recibirá un total de 10,000 acciones Serie A luego de aplicar el valuation cap puesto que ello resulta en un número mayor de acciones al que recibiría si el precio de conversión a ser aplicado fuera el precio de suscripción pagado por el fondo de venture capital, incluso luego de aplicar el descuento del 20%. • Si por el contrario, la startup levanta capital de un fondo de venture capital en una Serie A, con una valorización de la startup antes de dicha ronda fijada en US$ 6,000,000 y con el mismo precio de sus- cripción por acción de US$ 5.00 (fijado en el ejemplo anterior), el precio de conversión por acción, si aplicáramos el descuento de 20%, será de US$ 4.00 (al igual que en el ejemplo anterior). Pero si aplicáramos para el inversionista ángel el valuation cap de US$ 5,000,000 en reemplazo de la valorización de la startup fijada en US$ 6,000,000, el precio de conversión por acción que regirá para el in- versionista ángel sería de US$ 4.1667 por acción (US$ 5.00 multipli- cado por (US$ 5,000,000 entre US$ 6,000,000). Como consecuencia de ello, aplicará como precio de conversión el resultante de aplicar 63 63 Ѳeamdra r0eo oseCma rama Edición de Aniversario el descuento de 20% (y no el valuation cap), por ser el menor precio por acción, lo que resultará en un mayor número de acciones Serie A para el inversionista ángel tenedor de la nota convertible. el descuento de 20% (y no el valuation cap), por ser el menor precio por acción, lo que resultará en un mayor número de acciones Serie A para el inversionista ángel tenedor de la nota convertible. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla En este sentido, la startup y el inversionista ángel suscriben un contrato de préstamo en virtud del cual el principal e intereses (si estos hubiesen sido pactados) se convertirán automáticamente en el mismo tipo de acciones que sean emitidas en la siguiente ronda de financiamiento que cumpla las condiciones pactadas por las partes. Adicionalmente, el inversionista tiene el derecho de convertir el préstamo en caso ocurra un evento de liquidez entre la celebración del préstamo capitalizable y la ronda de 64 64 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla financiamiento. En caso de no ejercer dicho derecho de conversión, el inversionista ángel puede optar por requerir el repago del préstamo. Asimismo, en caso de liquidación de la startup, el inversionista ángel tendría un derecho de cobro preferente respecto al de los accionistas. En el Perú es perfectamente válido pactar, desde el punto de vista contractual, que el préstamo capitalizable no devengará intereses, a fin de asemejarlo a la estructura del SAFE. Al respecto, es importante tomar en consideración, desde el punto de vista tributario, que no se aplicarán los intereses presuntos para efectos de la determinación de un impuesto a la renta de cargo del inversionista ángel que otorgó el préstamo siempre que: (i) el inversionista ángel no califique como parte vinculada económicamente a la startup1; (ii) no se haya regis- trado en la contabilidad de la startup intereses a favor del inversio- nista ángel (y ello pueda ser acreditado en los libros contables de la startup); y (iv) que se haya pactado expresamente en el contrato de préstamo capitalizable que no se devengarán intereses. Por el contrario, en caso el inversionista ángel que otorgó el préstamo sí califique como parte vinculada económicamente a la startup, resultarán aplicables, para fines tributarios, intereses presuntos para efectos de la determinación de un impuesto a la renta de cargo del inversionista ángel, los mismos que deberán ser calculados conforme a las normas de precios de transferencia previstas en la legislación del Impuesto a la Renta, resultando aplicable una tasa del Impuesto a la Renta que variará dependiendo de la condición del inversionista ángel que actúa como prestamista (persona natural o persona jurídica vinculada). Para los inversionistas ángeles, existen ventajas y desventajas respecto a la estructura de la nota convertible como instrumento de inversión. 1 A fin de determinar si existe vinculación para efectos tributarios se debe tomar en cuenta lo previsto en el artículo 24° del Reglamento de la Ley del Impuesto a la Renta. Entre los criterios de vinculación que dicha norma establece se encuentra el caso en el cual una persona natural o jurídica posea más del 30% del capital de otra persona jurídica, direc- tamente o por intermedio de un tercero. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Dentro de las ventajas, debemos recordar el hecho que el inversionista ángel recibirá el mismo tipo de acciones que las que serán emitidas al fondo de venture capital en la ronda de financiamiento que gatille la conversión de la nota convertible. En este sentido, al permitir a los fondos de inversión de venture capital negociar los términos y condi- ciones de las acciones en el marco de un financiamiento que involucra un importe mucho más alto de inversión que el aporte realizado por el propio inversionista ángel en la startup, es mucho más probable que ello resulte en más y mejores derechos de los que el inversionista 65 65 Ѳeamdra r0eo oseCma rama Edición de Aniversario ángel hubiera podido obtener directamente, debido a que los fondos de venture capital están en una mejor posición de negociación que un inversionista ángel pequeño. La desventaja para los inversionistas ángeles es que la negociación de la valorización de la startup ocurre en una etapa posterior dentro del ciclo de vida de la startup, luego de que la misma ha tenido ya la posi- bilidad de usar los fondos levantados de los inversionistas ángeles para alcanzar suficiente tracción e hitos que le permitirán negociar una valorización mayor. Es en este punto que el descuento para la conversión de los inversionistas ángeles, juego un rol muy importante. Sin perjuicio de lo antes indicado, cada vez con mayor frecuencia los inversionistas ángeles están demandando recibir participación en el capital de la startup (equity) desde el día uno en el que desembolsan su inversión, así como una valorización que les resulte más favorable que la que probablemente recibirían bajo una estructura de deuda convertible en acciones (aun tomando en consideración la aplicación de un descuento o valuation cap). Los inversionistas ángeles se están dando cuenta que podrían estar actuando en contra de sus propios intereses al proveer los fondos que incrementarán la valorización de la startup, y tener ello como resultado el hecho de recibir una menor participación en el capital que la que hubieran obtenido si hubieran negociado la valorización al momento en que proveyeron los fondos2. 2 RUBINSTEIN, Frederic A. y ROTH, Audrey M. Inside the Minds. Deal Strategies for Venture q Capital and Private Equity Lawyers. The Life Cycle of a Venture-Backed Company. Estados Unidos de América, Aspatore Books, 2006, p. 29. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Respecto a los derechos especiales que recibirán los inversionistas ángeles como parte de sus acciones, es importante que los abogados alerten a los emprendedores y a los inversionistas ángeles de qué forma y en qué medida, cualquiera de los términos otorgados a favor de los inversionistas ángeles en una ronda temprana como es la de capital semilla puede dificultar el levantamiento de las rondas de financiamiento en etapas más avanzadas o afectar las discusiones futuras sobre valorización. Como consecuencia de ello, deben ser un número razonable de derechos que otorguen a la startup la libertad de otorgar más derechos a inversionistas que ingresen en rondas de financiamiento posteriores sin tener que pedir a los inversionistas ángeles que renuncien a los derechos que ya se les habían conferido. c. Fondos de venture capital Los fondos de venture capital son fondos de inversión que se forman con los aportes de un grupo de personas o entidades denominados 66 66 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla partícipes, para que una sociedad gestora se encargue de su adminis- tración a través de la inversión en startups, en búsqueda de rentabili- dad. Entre dichos partícipes usualmente encontramos instituciones financieras, bancos de inversión, corporaciones, family offices, fondos de pensiones, y compañías de seguros. Si bien la mayoría de los fondos de venture capital no invierte en star- tups que se encuentren en etapas muy tempranas, existen al gunos que sí realizan inversiones de capital semilla, pero son la excep- ción. Ello se debe a que el tamaño de los patrimonios administra- dos por dichos fondos se ha elevado tanto en los últimos años, lo que ha vuelto muy complejo e incluso inmanejable la posibilidad de invertir en startups con montos de inversión o tickets muy pequeños por ronda. Esto obligaría a los fondos de venture capital a tener que invertir en un número muy elevado de startups para poder colocar la totalidad de los fondos administrados, y demandaría además que la sociedad gestora tenga que dedicar demasiada atención a cada una de dichas inversiones, por un retorno que individualmente no sería muy alto. V. El term sheet El term sheet es un documento importante en la ronda de financia- miento, en la medida que es señal del compromiso firme del inversio- nista ángel de realizar una inversión, y del hecho que desea proceder a realizar un due diligence y negociar y suscribir la documentación legal definitiva para la formalización de la inversión. El term sheet cubrirá todos los aspectos importantes del financia- miento: términos económicos tales como la valorización asignada a la startup; temas vinculados al control de la misma tales como la composición del Directorio y qué tipo de supermayoría o derechos de veto tendrán los inversionistas; y derechos post-cierre de los inver- sionistas, tales como el derecho a participar en futuras rondas de financiamiento y derechos de información, los mismos que incluyen el derecho a recibir información financiera de la startup de manera periódica. El term sheet en la mayoría de los casos establecerá que es de carácter no es vinculante para las partes que lo suscriben, excepto por ciertos términos específicos que sí lo serán, como es el caso de la confiden- cialidad y la ley aplicable. Sin perjuicio de ser no vinculante, el term sheet es por lejos el documento más importante a ser negociado con los inversionistas. Casi la totalidad de los temas relevantes serán cubiertos en el term sheet, dejando temas menores pendientes de ser discutidos en los documentos definitivos de la ronda de financia- miento que serán suscritos con posterioridad. El emprendedor debe considerar el term sheet como la base de su relación con el inversio- nista, y asegurarse de darle la gran atención que requiere. Existen muchas opiniones sobre el uso y la extensión que deben tener los term sheets. Una primera aproximación es tener una versión abreviada de term sheet en la cual únicamente los puntos más impor- tantes de la transacción estén cubiertos. De esta manera, los seguido- res de dicha aproximación argumentan que las partes involucradas (emprendedores e inversionistas) pueden enfocarse en los asuntos mayores y dejar los puntos menores o auxiliares a los abogados al momento en que negocien la documentación definitiva de la ronda de financiamiento. Los seguidores de esta aproximación argumen- tan que tener una versión corta del term sheet facilitará y permitirá alcanzar la firma del mismo con mayor rapidez, y muchas veces los emprendedores los prefieren justo por eso. IV. Proceso de la ronda de financiamiento El proceso de una ronda de financiamiento de capital semilla dirigido a inversionistas ángeles incluye la preparación y negociación de un term sheet, seguido de un proceso de revisión de la compañía (due diligence) relativamente breve, y finalmente la preparación y negocia- ción de un conjunto de documentos legales definitivos para formali- zar la inversión. Entre dichos documentos definitivos usualmente encontramos un acuerdo de inversión y un convenio de accionistas. El acuerdo de inversión es un contrato privado que establece el mecanismo legal para la suscripción de acciones por parte del inversionista ángel, incluyendo los detalles de los términos comerciales acordados, tales como precio de suscripción, clase y número de acciones, entre otros. Por su parte, el convenio de accionistas define los derechos y obliga- ciones de los accionistas y busca regular la relación de ellos entre sí, y también entre ellos y la startup. Como hemos señalado en la sección introductoria, en el presenta artículo nos centraremos en analizar los aspectos legales clave en la negociación del term sheet. 67 67 67 67 Ѳeamdra r0eo oseCma rama Edición de Aniversario V. El term sheet Otra aproximación en relación al term sheet es la de tener una versión extendida del mismo, donde prácticamente todos los asuntos que 68 68 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla deban ser negociados entre las partes sean puestos sobre la mesa, a fin de que la elaboración y negociación de los documentos definitivos pueda ser más rápida y sencilla. Una desventaja de contar con una versión corta del term sheet es que éste dejará muchos asuntos para ser resueltos en la etapa de la negociación de la documentación definitiva, y si dichos asuntos no son resueltos, las partes pasarán tiempo extra e incurrirán en gastos legales adicionales que podrían haber sido evitados si se hubiera optado por contar con una versión larga del term sheet. En nuestra experiencia, es usualmente más beneficioso tanto para el inversionista ángel como para el emprendedor, tener un term sheet extendido y completo, lo cual permitirá a las partes tener desde el día uno una mayor claridad y mejor entendimiento respecto a los términos y condiciones en que se realiza la inversión, y mitigar futuros problemas en la etapa de redacción y negociación de la docu- mentación definitiva. VI. Contenido del term sheet Como ya se mencionó en la sección III anterior, el term sheet describe de manera resumida y sucinta los aspectos clave negociados entre los emprendedores y el inversionista ángel, los cuales se desarrollarán posteriormente, con mayor detalle, en los documentos definitivos. Al respecto, es importante para los emprendedores que el otorga- miento de los derechos económicos y políticos a favor de los inver- sionistas se dé de manera gradual para que la startup tenga un mayor campo de negociación en rondas de financiamiento subsecuentes y aquéllas que fueran a ocurrir en el largo plazo. Pero antes de entrar al detalle de cuáles son los principales derechos y obligaciones que son usualmente negociados entre emprendedo- res e inversionistas ángeles en rondas de financiamiento de capital semilla, creo que resulta fundamental tener en cuenta un punto central. Y este punto central radica justamente en la importancia de que los términos que se pacten en el term sheet no generen una desa- lineación entre los intereses de los emprendedores y los intereses de los inversionistas ángeles. Por un lado, el inversionista ángel debe recordar que ésta es una inversión de riesgo. No hay duda de que los inversionistas ángeles deben buscar ciertas protecciones. No obstante ello, es importante 69 69 Ѳeamdra r0eo oseCma rama Edición de Aniversario que los inversionistas ángeles tomen consciencia de que un sobre proteccionismo exagerado puede llegar eventualmente a generar esta desalineación de intereses de la que hablamos, y terminar per- judicando el desarrollo del negocio, y por lo tanto, terminar perjudi- cándose a sí mismos. Y esto se traduce en dos niveles. Uno de ellos es el económico, si con el afán del inversionista ángel de asegurarse un retorno, se pactan términos y condiciones que tengan como con- secuencia que el inversionista termine obteniendo una ganancia preferente de tal magnitud, que ello pueda conllevar a no dejar un remanente que repartir a los emprendedores. El otro nivel es el control. Una startup busca el desarrollo de una nueva idea de negocio que ha surgido de los emprendedores, y corresponde a ellos estar a cargo de su desarrollo, razón por la cual los emprendedores buscan evitar a toda costa la pérdida del control de la gestión de la startup, o ser diluidos significativamente vendiendo mucho de su participación a una valorización muy baja. VI. Contenido del term sheet De otro lado, los emprendedores deben comprender que no resulta un proceso fácil para el inversionista ángel, el entregar su dinero a los emprendedores –en su mayoría bastante jóvenes–, con quienes muchas veces sólo se han tenido unas cuantas reuniones. Es por ello que no debería resultar una sorpresa para los emprendedores que el capital sea entregado por los inversionistas ángeles bajo ciertas pro- tecciones. Como consecuencia de lo antes señalado, resulta importante tener un term sheet equilibrado que contribuya a construir una relación sólida entre los emprendedores y el inversionista ángel, e incrementar las posibilidades de éxito para ambas partes. Cuando éstas pueden trabajar alineados hacia una meta común, hay una mayor posibilidad de éxito para el emprendedor y, por ende, un mejor retorno para el inversionista ángel que lo respaldó. Conforme con lo antes señalado, la estructura del contenido del term sheet negociado entre los inversionistas ángeles y los emprendedores se divide usualmente en dos grandes bloques: los términos y condi- ciones económicos; y los términos y condiciones sobre control. 3 MANZANERA ESCRIBANO, Antonio. Finanzas para emprendedores (1a ed.). Barcelona, Deusto, 2010, p. 137. Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla a. Términos económicos A través de los términos económicos, los inversionistas ángeles buscan asegurarse una tajada lo suficientemente grande de la torta a fin de que el hecho de llevar adelante su inversión sea algo que valga la pena, sobre la base de un riesgo controlado. 70 70 i. Valorización La valorización de la startup es generalmente el punto de partida de la negociación entre el emprendedor y el inversionista ángel y resulta ser un punto crucial para ambos en el proceso de levantamiento de capital. Mientras mayor sea la valorización, mayor será el precio por acción que deberá pagar el inversionista, y menor será el número de acciones (y porcentaje de participación accionaria en la startup) que será adquirida por el inversionista ángel a partir de su inversión de capital; y, como consecuencia de ello, menor será la dilución que afectará al emprendedor. Igual de importante resulta establecer el monto de capital que se levantará en la ronda de financiamiento. A veces resulta preferible para el emprendedor levantar una menor cantidad de dinero en una etapa temprana cuando el inversionista probablemente insistirá en asignar una valorización baja a la startup, y proyectarse a que volverá a levantar una siguiente ronda de financiamiento una vez que la startup haya logrado el progreso suficiente que le permita asegurar que conseguirá negociar una valorización mayor con los futuros inversionistas. Finalmente, otro punto relacionado a los antes señalados que también resulta importante en la negociación, es conocer el número total de acciones que la startup ha fijado destinar para la entrega de acciones a favor de colaboradores clave (por ejemplo, a través del otorgamiento de opciones de suscripción sobre acciones, conocidas en inglés como stock options). En esta línea, hay dos conceptos que son negociados entre las partes. El valor pre-money, que es el valor de la startup antes de la entrada del inversionista y el valor post-money, que es el valor de la startup luego del cierre de la ronda de inversión3. En ese sentido, el monto a ser levantado en la ronda de inversión más el pre-money valuation da como resultado el post-money valuation. Por ejemplo, si los inversio- nistas en una ronda planean invertir conjuntamente un total de US$ 5,000,000 en donde el pre-money valuation de la startup ha sido acordado en US$ 15,000,000, ello significa que el post-money valuation será de US$ 20,000,000, y los inversionistas esperan adquirir, en conjunto, al cierre de la transacción, una participación accionaria total de 25% (resultado de dividir US$ 5,000,000 entre US$ 20,000,000). 71 71 Ѳeamdra r0eo oseCma rama Edición de Aniversario La valorización es un punto negociable y no existe una única fórmula o metodología correcta para calcularla. i. Valorización La valorización de startups suele ser compleja debido a su falta de datos históricos que permitan utilizarse para el análisis financiero (flujo de caja, rotación de inven- tarios, plazos de cobro, entre otros). Asimismo, la posibilidad de que un evento ocurra o no (evento binario) dificulta la valorización de una startup. Un ejemplo de ello es que un emprendimiento desarrolle una idea de negocio para crear una píldora que cure la diabetes. Si su creación es aprobada por la entidad competente, la startup generará mucho dinero, pero en caso no sea aprobada, la idea de negocio y el desarrollo de la píldora no valdrán nada4. Sin perjuicio de que cada startup es única, y como consecuencia de ello, también su proceso de valorización, existen ciertos aspectos claves que se toman usualmente en consideración para la determina- ción de la valorización: la experiencia y éxitos pasados de los empren- dedores, el tamaño de la oportunidad de mercado, la tecnología ya desarrollada y de propiedad de la startup, su tracción inicial (ingresos, alianzas estratégicas, satisfacción de clientes, publicidad favorable, etc.), el nivel de avance respecto a la obtención de un producto mínimo viable, el modelo de negocios, valorizaciones de compañías compara- bles, el entorno económico general, entre otros. 4 Ibid, p. 141. ii. Liquidación preferente Este es el segundo término económico más importante luego del precio, dado que establece la prioridad en el reparto del dinero dis- ponible en el marco de un evento de liquidez, el cual podría darse a través de la fusión o venta de la startup o la venta de la mayoría de las acciones de la startup. Como regla general los inversionistas siempre tienen una preferen- cia en el orden de cobro frente a los emprendedores. Sin embargo, el derecho de liquidación preferente va más allá, puesto que establece otros aspectos clave como el derecho que tendrá el inversionista ángel a recibir, antes que cualquier importe sea pagado a los empren- dedores y el resto de inversionistas pertenecientes a otras rondas, un importe equivalente a un múltiplo de retorno aplicado sobre su inversión original. Hoy en día el múltiplo estándar en el mercado es de una vez (1x) el monto de la inversión original o simplemente el monto de la inversión original. Establecer este derecho cobra importancia en el eventual caso de que la startup se venda a un valor 72 72 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla inferior al capital total invertido en la misma5. El derecho de liquida- ción preferente puede también establecer el derecho del inversionista ángel a participar de la ganancia junto con los emprendedores, bajo los mismos términos y condiciones, a prorrata de su participación accionaria, una vez que se le haya pagado su inversión original. Esta- blecer este derecho cobra importancia en el eventual caso de que la startup se venda generando un gran retorno. Por ejemplo, bajo un esquema de derecho de liquidación preferente de 1x, si la valorización pre-money de la ronda de inversión es de US$ 5,000,000, y un inversionista ángel invierte US$ 5,000,000 en la startup (con lo cual adquiere acciones que representan el 50%), y la startup es vendida por un precio de US$ 50,000,000, dicho precio sería distri- buido de la siguiente manera: • Si adicionalmente el derecho de liquidación preferente fuera parti- cipativo, los primeros US$ 5,000,000 irían a pagar al inversionista, y el remanente de US$ 45,000,000 será dividido 50-50 (en función a la prorrata de participación accionaria en la startup). En este sentido, los emprendedores recibirían US$ 22,500,000, y el inver- sionista ángel recibiría un total de US$ 27,500,000. 5 FELD, Brad y MENDELSON, Jason. Venture Deals (3a ed.). Estados Unidos de América, Wiley & Sons Ltd, 2016, p. 45. ii. Liquidación preferente • Si el derecho de liquidación preferente no fuera participativo, el inversionista ángel optaría por recibir las ganancias en los mismos términos y condiciones que los emprendedores (puesto que recibiría US$ 25,000,000 en función a su prorrata de 50%) en lugar de optar por ejercer su derecho de preferencia, el mismo que únicamente le asignaría los US$ 5,000,000 de su inversión inicial. iii. Dividendos No es muy usual encontrar en el term sheet un pacto de pago de divi- dendos en una etapa temprana como lo es la del capital semilla, en la medida que es altamente improbable que una startup arroje utilidades, y menos aún, que tenga la caja para pagar dividendos. Sin perjuicio de ello, en caso de encontrarse regulado, probablemente este pacto esta- blecerá que se devengará un dividendo preferente de carácter acumu- lativo en favor del inversionista ángel, calculado como un porcentaje aplicado sobre el importe de su inversión. En realidad, lo que determi- nará la inclusión de dicho pacto es la existencia de un dividendo pre- ferente que se irá acumulando cada año (al margen de que la startup no genere utilidades), el que finalmente terminará siendo pagado en 73 73 Ѳeamdra r0eo oseCma rama Edición de Aniversario caso de un evento de liquidez, momento en el cual el inversionista ángel recibirá ese dividendo adicional junto con la porción del precio de venta que le corresponda. Este derecho cobra mayor importancia en los casos en los que el precio obtenido en el evento de liquidez no sea muy alto, pero la inversión del inversionista sí lo haya sido. Al respecto, la legislación peruana, a través del artículo 83° de la Ley General de Sociedades, Ley N° 26887 (LGS), permite la creación de clases de acciones que tengan derecho a un dividendo preferente, incluso acumulativo, existiendo como única limitación, el hecho que sólo podrán ser pagados cuando existan utilidades distribuibles. En este sentido, la LGS permite que dicho dividendo sea acumulable para el caso en que no existan utilidades en uno o más años o éstas no alcancen a cubrir el rendimiento establecido. Como consecuencia de ello, es perfectamente factible para la startup pactar un dividendo preferente acumulativo a favor de la clase de acciones que recibirán los inversionistas ángeles, cuyo pago estará sujeto a la existencia de utilidades o a la ocurrencia de un evento de liquidez, en los términos señalados en la presente sección. iv. Derechos anti-dilución A medida que la startup reciba más inversionistas, aquellos que se convirtieron en accionistas de la startup con anterioridad al ingreso de los nuevos inversionistas, se diluirán en la proporción que repre- sente el nuevo capital obtenido en esa nueva ronda de inversión posterior. Si dicha nueva ronda es realizada a un precio mayor por acción que aquel que fue pagado por un inversionista que suscribió acciones con anterioridad, éste se diluirá aritméticamente mas no económicamente. Esto quiere decir que las acciones del inversionista podrán haber pasado a representar un menor porcentaje del capital social, pero el valor de mercado de dichas acciones se ha revaluado en función al mayor precio por acción pagado en la nueva ronda. Asimismo, la startup de la cual es accionista el inversionista tendrá una mayor caja en el balance. Lo especial a tener en cuenta al analizar la negociación de este derecho económico es que puede existir un escenario en el cual un nuevo inversionista suscriba acciones a un menor precio que aquél que fue pagado por un inversionista que suscribió acciones en una oportunidad anterior. Los derechos anti-dilución que pueden ser incluidos en el term sheet permiten obtener protección ante la situación antes descrita. Este derecho establece que el precio por acción que fue pagado por un 74 74 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla inversionista que suscribió acciones en una oportunidad anterior, será automática y retroactivamente «disminuido» en una menor o mayor proporción (según la forma en que haya sido pactado el derecho anti-dilución en el term sheet) sobre la base de una fórmula. Esto tendrá como consecuencia que el inversionista deberá ser titular de un mayor número de acciones que el originalmente recibido por el importe de la inversión que realizó. Una modalidad de fórmula no tan usada por ser altamente dura con el emprendedor (conocida en inglés como full ratched-based antidilution clause) es aquélla que establece que todas las acciones del inversio- nista serán ajustadas con la «disminución» de precio, sin tomar en consideración el importe de capital levantado en la ronda de financia- miento posterior que gatilló el derecho anti-dilución. iv. Derechos anti-dilución Otra modalidad de fórmula que es más frecuentemente utilizada (conocida en inglés como weighted average antidilution clause) es aquella en la cual sí se toma en cuenta, para efectos del cálculo, el tamaño de la ronda de financiamiento que gatilló el derecho anti-dilución. En este sentido, se toman en cuenta tanto el importe de capital levantado y el número de acciones emitidas a un mayor precio en la ronda de financiamiento anterior; como el importe de capital levantado y el número de acciones emitidas a un menor precio en la ronda de financiamiento posterior. Como consecuencia de ello, el ajuste con «disminución» del precio que fue pagado por un inversionista que suscribió acciones con anteriori- dad, refleja de manera más cercana y proporcional el nivel de perjuicio ocasionado originalmente a dicho inversionista con la dilución. Si bien la primera modalidad de derecho anti-dilución parece más beneficiosa para el inversionista ángel, ello no es tan cierto. En la medida que los efectos de los derechos anti-dilución son finalmente sufridos por los emprendedores (quienes reducen su porcentaje de participación accionaria cuando dichos derechos anti-dilución son ejercidos por el inversionista ángel), si dichos derechos anti-dilución son tan severos que terminan diluyendo significativamente a los em- prendedores, estos podrían tener un incentivo para intentar evitar, de cualquier forma, el tener que cerrar una ronda de inversión a un menor precio por acción. Ello puede tener como consecuencia que el emprendedor termine aceptando un inversionista inadecuado para la startup pero que sí esté dispuesto a pagar un precio mayor por acción. O si por el contrario, los emprendedores no logran evitar el tener que levantar una ronda de inversión a un menor precio, y como consecuencia de ello, los derechos anti-dilución son ejercidos por los inversionistas, los emprendedores pueden perder el incentivo de tra- bajar para la startup, ya que al haberse diluido significativamente, nunca terminarán beneficiándose de un retorno importante. 75 75 Ѳeamdra r0eo oseCma rama Edición de Aniversario a. Términos sobre control En esta sección hablaremos sobre los términos y condiciones relacio- nados al control de la startup que se deben tener en cuenta al momento de negociar el term sheet. Los inversionistas buscan obtener determinados niveles de control a fin de conocer regularmente el estado de la startup así como el uso del capital que aportaron. i. Derecho de veto Las acciones que son entregadas a los inversionistas ángeles pueden conferir o no derechos de voto, o alternativamente, derechos de voto limitados, aplicables únicamente para la toma de ciertas decisiones fundamentales respecto de la startup. En el caso de las acciones con derechos de voto irrestrictos emitidas a favor de los inversionistas ángeles, la diferencia respecto de las acciones comunes que son de propiedad de los emprendedores radica en los derechos preferentes que se les confiere a las acciones preferentes, los cuales usualmente incluyen los derechos que vienen siendo descritos en la presente sección VI. En jurisdicciones como los Estados Unidos de América, usualmente las startups emiten acciones preferentes con derecho a voto o con derecho a voto limitados a favor de los inversionistas, en los cuales los derechos preferentes del inversionista son inherentes a la clase y serie de acción preferente emitida. Conforme se celebren rondas de inversión sucesivas, se irán emitiendo series distintas de acciones preferentes, cuyos derechos deben ser iguales dentro de una misma serie, pero pueden diferir respecto de las series anteriores y posteriores. Los emprendedores, así como los colaboradores clave, mantienen la titularidad de acciones comunes con derecho a voto, pero sin los derechos preferentes que corresponden a los inversionistas. En otras jurisdicciones como es el caso de Perú y España, las startups emiten acciones comunes tanto para los emprendedores como para los inversionistas, y los derechos especiales que se otorgan a los inversionistas se formalizan y regulan a través de la celebración de un convenio de accionistas. Muchas veces ello se debe a que los emprendedores prefieren evitar el tener que cumplir con formali- dades legales adicionales que pueden resultar tediosas, costosas y generar demoras al momento de tener que aprobar ciertos acuerdos cuando se tienen distintas clases de acciones emitidas. Un ejemplo de ello es la formalidad legal de tener que convocar a juntas espe- 76 76 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla ciales de accionistas que sesionan de manera separada a la junta general ordinaria, cuando se requiera adoptar acuerdos que puedan afectar los derechos particulares de alguna de las diferentes clases de acciones y, en algunos casos, incluso el tener que contar con la participación de aquellos accionistas titulares de los derechos prefe- rentes que se suprimen y modifican. i. Derecho de veto En este sentido, en el Perú, los inversionistas ángeles que suscriban acciones comunes con derecho a voto de la startup tendrán, adicional- mente a los derechos preferentes que puedan ser adquiridos a través del convenio de accionistas a suscribirse como parte de la ronda de financiamiento, cuando menos, los siguientes derechos detallados en el artículo 95° de la LGS: • Participación en el reparto de utilidades. • Fiscalizar la gestión de la startup. • Ser preferido para la suscripción de acciones en caso de una nueva emisión de acciones. • Ser preferido para suscribir obligaciones u otros títulos converti- bles o con derecho a ser convertidos en acciones. • Separarse de la sociedad en los casos previstos en la ley y/o estatuto de la startup. Debido a que la participación accionaria de los inversionistas ángeles es minoritaria, y usualmente en las etapas de capital semilla los emprendedores retienen la mayoría accionaria, como regla general las decisiones podrán ser aprobadas por estos últimos a nivel de junta general de accionistas, sin necesidad de contar con el voto favorable de los inversionistas ángeles. Es en este sentido que los inversionis- tas ángeles usualmente negocian como medida de protección, que se les confieran derechos de «veto» respecto de las decisiones estratégi- cas sobre el manejo de la startup. El referido derecho de veto puede ser implementado a través de distintos mecanismos. Por ejemplo, a través de la inclusión de una provisión en el convenio de accionistas y en el estatuto de la startup que exija un quórum y mayoría más alta o supermayoría para la adopción de dichas decisiones estratégicas, los cuales los emprendedores no puedan alcanzar sin la participa- ción y voto favorable de una parte de los inversionistas ángeles. Otra posibilidad es incluir adicionalmente en dicha provisión la condición de que ese voto adicional de los inversionistas ángeles, represente cuando menos al 50% de ellos. Dichos mecanismos de supermayoría 77 77 Ѳeamdra r0eo oseCma rama Edición de Aniversario otorgan representación a los inversionistas ángeles, cuya presencia en las reuniones y voto favorable se hacen indispensables para que la startup pueda adoptar decisiones válidas y, de esta manera, aseguran que su intervención y opinión sean tomadas en cuenta en el proceso decisorio de la startup. i. Derecho de veto Debe evitarse otorgar este derecho de veto con nombre y apellido a favor de un inversionista específico, en la medida que ello no será bien recibido por los futuros inversionistas, perjudicando la posibilidad de la startup de levantar rondas subse- cuentes de capital. Dentro de las decisiones estratégicas sujetas a la supermayoría usual- mente se encuentran la modificación de estatuto, endeudamiento, emisión de nuevas acciones (que tengan por ejemplo los mismos derechos, o derechos superiores a los que actualmente tienen los inversionistas), venta de activos, venta de la startup, recompra de acciones, modificación del número de directores, distribución de dividendos, licenciamiento de los activos de propiedad intelectual, la disolución y liquidación de la startup, entre otros. Es debatible si una ronda de inversión en una etapa temprana como la del capital semilla debería contar con los derechos de superma- yoría antes señalados, y en todo caso, qué decisiones fundamenta- les deberán estar sujetas a estos. Sin embargo, ello dependerá del poder de negociación, sofisticación, y nivel de involucramiento de los emprendedores vis-a-vis los inversionistas ángeles. ii. Directorio Muchas veces los emprendedores subestiman la importancia del directorio como parte del gobierno de las startups, y lo consideran como una mera formalidad. Sin embargo, ello está totalmente alejado de la realidad, ya que el Directorio es el órgano encargado del pla- neamiento del negocio, de la gestión y representación legal necesa- rias para la administración de la startup. En este sentido, muchas veces los emprendedores cometen el error de colocar como directores a personas cercanas, incluso en rondas muy tempranas como la de friends and family, buscando darles cierta tran- quilidad a los participantes en dicha ronda. Sin embargo, ello puede limitar en el futuro la posibilidad de la startup de levantar nuevas rondas de capital, puesto que los nuevos inversionistas (que serán más sofisticados y cuya inversión será más significativa en importe), pedirán la recomposición del directorio, ya que requerirán tener la posibilidad de poder nombrar a algún miembro del directorio como mecanismo de control sobre la startup a fin de proteger su inversión. 78 78 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla Es por ello que resulta más recomendable que durante la etapa tem- pra na la startup cuente con un directorio con el mínimo número de directores6. Dicho directorio debería estar inicialmente compuesto por los propios emprendedores (de preferencia aquellos que también ocupen los cargos de gerente general o algún otro cargo ejecutivo dentro de la startup) y quizás un director independiente que provea esa experiencia adicional en un área específica con la que no cuentan los emprendedores. Conforme vayan levantando capital, el número de directores podrá irse expandiendo, para dar cabida a inversionis- tas más sofisticados que necesariamente requerirán participación en el directorio como condición para invertir (por ejemplo, inversionis- tas ángeles sofisticados o agrupados, y los fondos de venture capital). En la medida que los inversionistas tendrán una participación mi- noritaria en el directorio, muchas veces se negocia la inclusión de derechos de supermayoría en términos similares a los regulados para la junta general, que resultarán aplicables para la adopción de decisio- nes fundamentales a nivel de directorio. En caso no se les otorgue un puesto en el directorio, muchas veces los inversionistas ángeles negociarán el derecho a tener un sitio como «observadores». 6 El número de directores debe ser un número fijo o variable entre un número mínimo y máximo. En ningún caso, el número de directores deberá ser inferior a tres, de acuerdo a lo establecido en el artículo 156° de la LGS. ii. Directorio Dicha condición les confiere el derecho a asistir a las sesiones de directorio con derecho a voz pero sin derecho a voto, así como el derecho a recibir la información financiera y de otra clase que es otorgada a los directores. La composición del directorio estará finalmente sujeta a negociación, entrando a tallar el monto invertido, el número de inversionistas par- ticipantes en la ronda (y en la startup en general), el nivel de control buscado, y el nivel de comodidad de los emprendedores de conferir dicho espacio. iii. Derechos de información Los inversionistas ángeles usualmente requerirán establecer un de recho que les permita obtener cierta información financiera, y de otra índole, así como el derecho a revisar los distintos libros y regis- tros de la startup. El term sheet establecerá para ello, por ejemplo, la obligación de la startup de proveer los estados financieros de manera anual, trimestral y/o mensual, así como el presupuesto anual o plan de negocios. 79 79 Ѳeamdra r0eo oseCma rama Edición de Aniversario Al respecto, en el Perú los inversionistas tendrán, adicionalmente a los derechos de información adicionales que puedan ser incluidos en el term sheet, aquellos derechos mínimos o esenciales de informa- ción ya previstos por la LGS para todo accionista de una sociedad anónima peruana. Es importante tener en consideración que los derechos de información son derechos que ayudan a generar una buena relación con los inver- sionistas y su otorgamiento no representa mayor dificultad para los emprendedores, razón por lo cual no debería requerir mayor nego- ciación. Sin embargo, es importante tener en cuenta, que los inver- sionistas no deberán exigir la inclusión de derechos de información que requieran una producción de documentos que pueda resultar demasiado trabajosa y demande mucho tiempo de los emprendedo- res, puesto que ello podría desconcentrarlos de la gestión propia de la startup. iv. Derecho de suscripción preferente, derechos de adquisición preferente y derecho de venta conjunta Todos estos derechos otorgan cierto control a los inversionistas respecto a transacciones u actos que involucren el capital social de la startup. El derecho de suscripción preferente está ligado a la adquisición de nuevas acciones emitidas por la propia startup, puesto que otorga al inversionista el derecho a participar en futuras rondas de financia- miento. En este sentido, si a la startup le va bien, los inversionistas actuales tendrán el derecho de incrementar su inversión. En el caso de jurisdicciones como los Estados Unidos de América, este derecho de suscripción preferente tiene que ser otorgado expresamente para existir, y en algunos casos no es otorgado en etapas tempranas como las rondas de capital semilla, y de ser otorgado, usualmente es otorgado a inversionistas ángeles que suscriban una participación no menor a la mínima establecida para tal efecto (conocidos en inglés como major investors), o se otorga sujeto a ciertas restricciones (por ejemplo, el hecho que pueda quedar sin efecto en caso no haya sido usado por los inversionistas en cada una de las rondas subsecuentes). Bajo las leyes peruanas, todos los accionistas de una sociedad tienen por defecto este derecho de suscripción preferente, encontrándose reconocido en el artículo 207° de la LGS, el cual establece el derecho de los accionistas a suscribir a prorrata de su participación acciona- ria las acciones que se creen en el marco de aumentos de capital por nuevos aportes. 80 80 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla El derecho de adquisición preferente y el derecho de venta conjunta (conocido en inglés como co-sale right o tag-along right) están por el contrario ligados a una venta secundaria de acciones, transferidas por el emprendedor (o por accionistas mayoritarios), y son derechos que van de la mano. En virtud al derecho de adquisición preferente, para que un empren- dedor o un accionista mayoritario (que no tenga una restricción temporal para vender sus acciones) pueda vender o transferir por otro medio sus acciones a un tercero, primero deberá ofrecerlas a los inversionistas de la startup quienes podrán adquirir dichas acciones de manera preferente. Adicionalmente, usualmente se establece que en caso este derecho preferente no sea ejercido por los inversionistas, dicho derecho de preferencia podrá ser ejercido por la propia startup. iv. Derecho de suscripción preferente, derechos de adquisición preferente y derecho de venta conjunta Ello permite a los inversionistas incrementar su participación accionaria y evitar que terceros desconocidos ingresen como accionistas de la startup. Bajo las leyes peruanas, todos los accionistas de una sociedad anónima cerrada tienen por defecto este derecho, encontrándose reconocido en el artículo 237° de la LGS, salvo que el estatuto de la sociedad lo haya suprimido. Asimismo, dicho derecho puede ser perfectamente incorporado en el estatuto de una sociedad anónima regular. No obstante lo antes señalado, podría darse el caso que los inver- sionistas no deseen comprar las acciones adicionales, porque podría ser que a la startup no le esté yendo tan bien como se esperaba. Es para estos casos que el inversionista busca incluir adicionalmente al derecho de adquisición preferente, un derecho de venta conjunta, que otorgue a los inversionistas, ante una oferta de compra por parte de un tercero, el derecho de integrarse a dicha oferta y vender a prorrata su participación social junto con el emprendedor o accionista mayori- tario transferente, en las mismas condiciones de la oferta del tercero. Este derecho protege a los inversionistas en el caso que la perfor- mance de la startup no haya sido óptima, puesto que asegura que nadie pueda hacer líquida su participación accionaria antes que ellos. v. Derecho de arrastre Finalmente, un derecho vinculado al control sobre la posible salida de la inversión a través de un evento de liquidez, es el derecho de arrastre (conocido en inglés como drag-along right) que facilita que la venta de la startup pueda ser realizada sin demora, al evitar que los accionistas minoritarios puedan bloquear la operación de venta. 81 81 Ѳeamdra r0eo oseCma rama Edición de Aniversario Este derecho generalmente se establece en beneficio de los accionis- tas mayoritarios de la startup, mayoría que durante la etapa temprana se encuentra concentrada en manos de los emprendedores. A través del mismo se establece que ante la existencia de una oferta de compra por el 100% de las acciones de la startup, los accionistas mayoritarios o grupo controlador podrán obligar al resto de los accionistas a vender sus participaciones en los mismos términos en que estos lo están haciendo. Este derecho puede ser establecido con ciertos matices que impidan que un grupo pueda simplemente imponer al otro la venta. Para tal fin puede exigirse que la decisión de venta esté res- paldada por todos los grupos específicos de accionistas (por ejemplo, por un porcentaje que represente a la mayoría de los emprendedores tenedores de las acciones comunes y por una mayoría de los inver- sionistas, tenedores de las distintas series de acciones preferentes). VII. Conclusiones Si bien todavía nos encontramos en una etapa temprana, y aunque probablemente tardaremos aún unos pocos años más para ver sus frutos, se espera que el ecosistema emprendedor traiga, al igual que en otros países, muchos aspectos positivos para el Perú, tales como una mejora importante de la competitividad, el fomento a la creación de nueva tecnología y una contribución a la generación de empleo de alta calidad e incremento de la recaudación fiscal. Asimismo, otro factor importante del aporte de la startups al desarro- llo del país es que éstas no generan únicamente una transformación a nivel económico, sino también una transformación a nivel cultural. Y eso lo vemos reflejado en el cada vez mayor número de jóvenes peruanos que apuestan por crear sus propias compañías e intentar hacer realidad sus sueños e ideas, en lugar de trabajar para otros. El reto como país es claro. El gobierno peruano debe apostar por fortalecer el desarrollo y crecimiento del ecosistema emprendedor y promover regulaciones que la apoyen, a fin de destrabar y agilizar su actividad. Pero los abogados podemos también contribuir con el fortalecimiento del ecosistema emprendedor peruano desde nuestro rol de asesores legales. Y una forma de hacerlo es ayudando a las partes involucradas a negociar y acordar términos y condiciones justos y equitativos que contribuyan a mantener el principio básico de alineación de intereses de emprendedores e inversionistas, y fomenten como consecuencia de ello, el desarrollo exitoso de las startups. 82 82 Negociando con ángeles: términos, cláusulas y otras condiciones legales clave para la estructuración de un term sheet de levantamiento de capital semilla En este sentido, parte de la labor del abogado como asesor legal de la transacción es la de contribuir a estructurar un term sheet equi- librado para ambas partes, a fin de que los intereses de cada una de ellas se encuentren razonablemente satisfecho. Para ello, los abogados deberán entender y saber manejar las expectativas y nece- sidades de emprendedores e inversionistas, a fin de permitir a las partes convertirse en verdaderos socios en el negocio; pero también deberán ser capaces de dar respuesta a situaciones jurídicamente no reguladas y adaptar al derecho peruano fórmulas jurídicas comunes en otros países con legislaciones más avanzadas en materia de levan- tamiento de capital para el financiamiento de startups. VII. Conclusiones Esperamos que el presente artículo pueda servir como una herramienta más que contribuya a dichos fines en beneficio del ecosistema emprendedor peruano, y de todas las partes involucradas en el mismo. 83 83 83 83
W2004432879.txt	https://projecteuclid.org/journals/electronic-communications-in-probability/volume-11/issue-none/On-the-Quadratic-Wiener-Functional-Associated-with-the-Malliavin-Derivative/10.1214/ECP.v11-1174.pdf	en		On the Quadratic Wiener Functional Associated with the Malliavin Derivative of the Square Norm of Brownian Sample Path on Interval	Electronic communications in probability	2,006	cc-by	4,094	DOI: 10.1214/ECP.v11-1174 Elect. Comm. in Probab. 11 (2006), 1–10 ELECTRONIC COMMUNICATIONS in PROBABILITY ON THE QUADRATIC WIENER FUNCTIONAL ASSOCIATED WITH THE MALLIAVIN DERIVATIVE OF THE SQUARE NORM OF BROWNIAN SAMPLE PATH ON INTERVAL SETSUO TANIGUCHI1 Faculty of Mathematics, Kyushu University, Fukuoka 812-8581, Japan email: taniguch@math.kyushu-u.ac.jp Submitted 17 April 2005, accepted in final form 11 January 2006 AMS 2000 Subject classification: 60H07, 60E10, 60E05 Keywords: stochastic oscillatory integral, Brownian sample path, density function Abstract RT RT Exact expressions of the stochastic oscillatory integrals with phase function 0 ( t w(s)ds)2 dt, {w(t)}t≥0 being the 1-dimensional Brownian motion, are given. As an application, the density function of the distribution of the half of the Wiener functional is given. 1 Introduction and statement of result The study of quadratic Wiener functionals, i.e., elements in the space of Wiener chaos of order 2, goes back to Cameron-Martin [1, 2] and Lévy [8]. While a stochastic oscillatory integral with quadratic Wiener functional as phase function has a general representation via CarlemanFredholm determinant ([3, 6, 10]), in our knowledge, a few examples, where the integrals are represented with more concrete functions like the ones used by Cameron-Martin and Lévy, are available. See [1, 2, 8, 6, 10] and references therein. In this paper, we study a new quadratic Wiener functional which admits a concrete expression of stochastic oscillatory integral, and apply the expression to compute the density function of the Wiener functional. Let T > 0, W be the space of all R-valued continuous functions w on [0, T ] with w(0) = 0, and P be the Wiener measure on W. The Wiener functional investigated in this paper is ¶2 Z T µZ T w(s)ds dt, w ∈ W. q(w) = 0 t The functional q interests us because it is a key ingredient in the study of asymptotic theory on W. Namely, recall the Wiener functional Z T q0 (w) = w(t)2 dt, w ∈ W, 0 1 RESEARCH SUPPORTED IN PART BY GRANT-IN-AID FOR SCIENTIFIC RESEARCH (A) 14204010, JSPS 1 Derivative of the square norm of Brownian sample path 2 which was studied first by Cameron-Martin [1, 2, 8]. As is well-known ([15]), the stochastic oscillatory integral Z exp(ζq0 /2)δy (w(T ))dP, W where δy (w(T )) is Watanabe’s pull back of the Dirac measure δy concentrated at y ∈ R via w(T ), relates to the fundamental solution to the heat equation associated with the Schrödinger operator (1/2){(d/dx)2 +ζx2 }, which describes the quantum mechanics of harmonic oscillator. If we denote by H the Cameron-Martin subspace of W (≡ the subspace of all absolutely RT continuous h ∈ W with square integrable derivative ḣ) and set hh, giH = 0 ḣ(t)ġ(t)dt and khk2H = hh, hiH for h, g ∈ H, then it is straightforward to see that q= 1 k∇q0 k2H , 4 where ∇ denotes the Malliavin gradient. Thus q determines the stationary points of q 0 . It should be noted that, in the context of the Malliavin calculus, the set of stationary points of q0 , i.e. the set {∇q0 = 0} = {q = 0} is determined uniquely up to equivalence of quasi-surely exceptional sets. On account of the stationary phase method on finite dimensional spaces (cf.[4]), q would play R an important role in the study of asymptotic behavior of the stochastic oscillatory integral W exp(ζq0 )ψdP with amplitude function ψ (cf. [9, 11, 12], in particular [13, 14]). The aim of this paper is to show Theorem 1. (i) For sufficiently small λ > 0, the following identities hold. Z exp(λq/2) dP = ZW ½ 1 1/4 cosh(λ T ) cos(λ1/4 T ) ¾1/2 , (1) exp(λq/2)δ0 (w(T )) dP W λ1/8 =√ © ª1/2 . π sin(λ1/4 T ) cosh(λ1/4 T ) + sinh(λ1/4 T ) cos(λ1/4 T ) (2) (ii) Define θ(u; x) and pT (x) for u ∈ [0, π/2] and x ≥ 0 by θ(u; x) = ∞ X (−1) k=−∞ pT (x) = 4 πT 4 Z π/2 0 k {u + (2k + 1)π}3 e−x{u+(2k+1)π} p cosh(u + (2k + 1)π) 4 /T 4 , θ(u; x) √ du. cos u Then pT is the density function of the distribution of q/2 on R; P (q/2 ∈ dx) = pT (x)χ[0,∞) (x)dx, (3) where χ[0,∞) denotes the indicator function of [0, ∞). The assertion (i) of Theorem 1 will be shown in Section 2 and (ii) will be proved in Section 3. Derivative of the square norm of Brownian sample path 2 3 Proof of Theorem 1 (i) In this section, we shall show the identities (1) and (2). The proof is broken into several steps, each being a lemma. We first show Lemma 1. Define the Hilbert-Schmidt operator A : H → H by Ah(t) = Z t ds 0 Z T du s Z u dv 0 Z T da h(a), h ∈ H, t ∈ [0, T ]. v Then it holds that q = QA + T4 , 6 (4) where QA = (∇∗ )2 A, ∇∗ being the adjoint operator of the Malliavin gradient ∇. Moreover, A is of trace class and tr A = T 4 /6. In particular, q = QA + tr A. Proof. Due to the integration by parts on [0, T ], it is easily seen that h∇2 q, h ⊗ kiH⊗2 = 2 Z 0 T µZ T h(s)ds t ¶µZ T t ¶ k(s)ds dt = 2hAh, kiH (5) for h, k ∈ H, where H⊗2 denotes the Hilbert space of all Hilbert-Schmidt operators on H, and h·, ·iH⊗2 does its inner product. Hence ∇2 q = 2A. (6) Let C2 be the space of Wiener chaos of order 2. Since w(s)w(u) − s = w(s)2 − s + w(s){w(u) − w(s)} ∈ C2 we have that T4 =2 q− 6 Z T 0 Z T t Z for u ≥ s, T s (w(s)w(u) − s)dudsdt ∈ C2 . From this and (6), we can conclude the identity (4). Let {hn }∞ n=1 be an orthonormal basis of H, and define kt ∈ H, t ∈ [0, T ], by Z s kt (s) = (T − max{t, u})du, s ∈ [0, T ]. 0 Since RT t hn (s)ds = hkt , hn iH , due to (5), we obtain that ∞ X n=1 hAhn , hn iH = Z 0 ∞ T X n=1 hkt , hn i2H dt = Thus A is of trace class and tr A = T 4 /6. We next recall the following assertion achieved in [5, 7]. Z T 0 kkt k2H dt = T4 . 6 Derivative of the square norm of Brownian sample path 4 Lemma 2. Let U : H → H be a Hilbert-Schmidt operator admitting a decomposition U = UV + UF with a Volterra operator UV : H → H and a bounded operator UF : H → H possessing the finite-dimensional range R(UF ). (i) For sufficiently small λ ∈ R, it holds that Z © ¡ ¢ª−1/2 −(λ/2)tr UF . (7) exp(λQU /2)dP = det I − λUF (I − λUV )−1 e W (ii) Let E be a subspace of R(UF ) and {η1 , . . . , ηd } be a basis of E. Define the Wiener functional η : W → Rd by η = (∇∗ η1 , . . . , ∇∗ ηd ). Then, for sufficiently small λ ∈ R, it holds that Z exp(λQU /2)δ0 (η) dP W =p 1 (2π)d det C(η) {det(I − λU1\ (I − λUV )−1 )}−1/2 e−(λ/2)tr UF , (8) where U¡1\ = −πE U¢ V + (I − πE )UF , πE : H → H being the orthogonal projection onto E, and C(η) = hηi , ηj iH 1≤i,j≤d . Proof. The essential part of the proof can be found in [5, 7]. For the completeness, we give the proof. Due to the splitting property of the Wiener measure, it holds that Z © ª−1/2 exp(λQU /2)dP = det2 (I − λU ) , W where det2 denotes the Carleman-Fredholm determinant. For example, see [3, 7]. Observe that, for Hilbert-Schmidt operators C, D : H → H such that C is of trace class, it holds that det2 (I + C)(I + D) = det(I + C)det2 (I + D)e−tr C(I+D) . (9) Since det2 (I − λUV ) = 1, substituting C = −λUF (I − λUV )−1 and D = −λUV into (9), we obtain that det2 (I − λU ) = det(I − λUF (I − λUV )−1 )eλtr UF . Thus (7) has been shown. Put U0 = (I − πE )U (I − πE ) and U1 = πE U πE . Then it holds ([7, 12]) that Z 1 {det2 (I − λU0 )}−1/2 e−(λ/2)tr U1 . exp(λQU /2)δ0 (η) dP = p (2π)d det C(η) W Setting U \ = (I − πE )U , and substituting C = −λU1\ (I − λUV )−1 and D = −λUV into (9), we see that \ det2 (I − λU0 ) = det2 (I − λU \ ) = det(I − λU1\ (I − λUV )−1 )eλtr U1 . Since tr U1\ + tr U1 = tr UF , we obtain (8). Derivative of the square norm of Brownian sample path It is not known if, by just watching specific shape of quadratic Wiener functional, one can tell that the associated Hilbert-Schmidt operator admits a decomposition as a sum of a Volterra operator and a bounded operator with finite dimensional range. However, in our situation, we know a priori that the operator A admits such a decomposition. Namely, the Hilbert-Schmidt operator B associated with q0 admits such a decomposition ([7]). Being equal to the square of B (see Remark 1 below), so does A. The following lemma gives the concrete expression of the decomposition of A. Lemma 3. Define I, AV , AF : H → H by Z t h(s)ds, t ∈ [0, T ], Ih(t) = 0 nT2 o 1 Ih(T ) − I 3 h(T ) η1 − Ih(T )η2 , h ∈ H, 2 6 where ηj (t) = t2j−1 , t ∈ [0, T ], j = 1, 2. Then (i) A = AV +AF , (ii) AV is a Volterra operator, (iii) R(AF ) = {aη1 + bη2 \| a, b ∈ R}, (iv) tr AF = tr A, and (v) for λ > 0, it holds that Z © ª 1 t (I − λAV )−1 h(t) = ḣ(s) cosh(λ1/4 (t − s)) + cos(λ1/4 (t − s)) ds, 2 0 h ∈ H, t ∈ [0, T ]. (10) AV h = I 4 h, AF h = Proof. The assertions (i) and (ii) follow from the very definitions of A and A V . The assertion (iv) is an immediate consequence of these and Lemma 1. By the definition of A F , the inclusion R(AF ) ⊂ {aη1 + bη2 \| a, b ∈ R} is obvious. To see the converse inclusion, it suffices to notice that AF η1 = (5T 4 /24)η1 − (T 2 /12)η2 and AF η2 = (7T 6 /60)η1 − (T 4 /24)η2 . Thus (iii) has been verified. To see (v), let (I − λAV )g = h and f = I 4 g. It then holds that f (4) − λf = h, where f (n) = (d/dt)n f . This leads us to the ordinary differential equation;            f f f (0) 0 1 0 0 0 0 (1)   0 0 1 0 f (1)   0  f (1) (0) 0 d  f  =   +  ,     f (2) (0) = 0 . dt f (2)   0 0 0 1 f (2)   0  λ 0 0 0 h 0 f (3) f (3) f (3) (0) It is then easily seen that f (3) (t) = 1 2 Z t 0 h(s){cosh(λ1/4 (t − s)) + cos(λ1/4 (t − s))}ds. Since g = f (4) , this implies the identity (10). Lemma 4. The identity (1) holds. Proof. Let η1 , η2 ∈ H be as described in Lemma 3, and put fj = (I − λAV )−1 ηj , j = 1, 2. By virtue of Lemma 3, we have that λ−1/2 {cosh(λ1/4 t) − cos(λ1/4 t)}, 2 λ−1 I 3 f1 (t) = {cosh(λ1/4 t) + cos(λ1/4 t) − 2}, 2 If2 (t) = 3λ−1 {cosh(λ1/4 t) + cos(λ1/4 t) − 2}, If1 (t) = I 3 f2 (t) = 3λ−3/2 {cosh(λ1/4 t) − cos(λ1/4 t)} − 3λ−1 t2 . 5 Derivative of the square norm of Brownian sample path 6 Hence, if we set αλ = cosh(λ1/4 T ) and βλ = cos(λ1/4 T ), then (I − λAF (I − λAV )−1 )η1 o n T 2 λ1/2 1 λ1/2 (αλ − βλ ) + (αλ + βλ ) η1 + (αλ − βλ )η2 , = − 4 2 12 (I − λAF (I − λAV )−1 )η2 n 3T 2 o 1 = − (αλ + βλ ) + 3λ−1/2 (αλ − βλ ) η1 + (αλ + βλ )η2 , 2 2 Thus, by virtue of (iii), it holds that ¢ ¡ det I − λAF (I − λAV )−1 Ã ! 2 1/2 λ1/2 − T λ4 (αλ − βλ ) + 21 (αλ + βλ ) (α − β ) λ λ 12 = det = α λ βλ . 2 1 − 3T2 (αλ + βλ ) + 3λ−1/2 (αλ − βλ ) 2 (αλ + βλ ) This implies the identity (1), because Lemmas 1, 2, and 3 yield that Z exp(λq/2)dP = {det(I − λAF (I − λAV )−1 )}−1/2 . W Lemma 5. The identity (2) holds. Proof. Let ηj , j = 1, 2, be as in Lemma 3 (iii), and E = {cη1 \| c ∈ R}. Define A\1 as described in Lemma 2 with U = A, UV = AV , and UF = AF . Since πE h = (h(T )/T )η1 for any h ∈ H, we have that n 1 o T2 1 A\1 h = − I 4 h(T ) + Ih(T ) η1 − Ih(T ) η2 . T 6 6 Let f1 , f2 be as in the proof of Lemma 4. Then we see that λ−5/4 {sinh(λ1/4 t) + sin(λ1/4 t)} − λ−1 t, 2 I 4 f2 (t) = 3λ−7/4 {sinh(λ1/4 t) − sin(λ1/4 t)} − λ−1 t3 . I 4 f1 (t) = Hence, if we put σλ = sinh(λ1/4 T ) and τλ = sin(λ1/4 T ), then (I − λA\1 (I − λAV )−1 )η1 n λ−1/4 o T 2 λ1/2 λ1/2 = (σλ + τλ ) − (αλ − βλ ) η1 + (αλ − βλ )η2 , 2T 12 12 (I − λA\1 (I − λAV )−1 )η2 o n λ−3/4 T2 1 (σλ − τλ ) − (αλ + βλ ) η1 + (αλ + βλ )η2 . = T 2 2 Since R(A\1 ) ⊂ R(AF ), by Lemma 3 (ii), this yields that det(I − λA\1 (I − λAV )−1 ) Ã −1/4 T 2 λ1/2 λ 2T (σλ + τλ ) − 12 (αλ − βλ ) = det T2 λ−3/4 T (σλ − τλ ) − 2 (αλ + βλ ) = λ−1/4 {σλ βλ + τλ αλ }. 2T λ1/2 12 (αλ − βλ ) 1 2 (αλ + βλ ) ! Derivative of the square norm of Brownian sample path 7 The identity (2) follows from this, because Lemmas 1, 2, and 3 imply that Z Z exp(λq/2)δ0 (w(T ))dP = exp(λQA /2)δ0 (∇∗ η1 )dP e(λ/2)trA W W =√ 1 {det(I − λA\1 (I − λAV )−1 )}−1/2 . 2πT Remark 1. It may be interesting to see that (1) is also shown by using the infinite product expression. Namely, define B : H → H by Bh(t) = Z tZ 0 T h(u)du ds, s h ∈ H, t ∈ [0, T ]. Then there exists an orthonormal basis {hn }∞ n=0 of H so that B= ∞ µ X n=0 T (n + 12 )π ¶2 hn ⊗ h n . T (n + 21 )π ¶4 hn ⊗ h n . See [10]. Since A = B 2 , it holds that A= ∞ µ X n=0 (11) In conjunction with Lemma 1, this implies that q = QA + tr A = ∞ µ X n=0 T (n + 12 )π ¶4 (∇∗ hn )2 . Due to the splitting property of the Wiener measure, we then obtain that Z exp(λq/2)dP = W Ã = µ ∞ ½ Y 1−λ n=0 Ã ∞ ½ Y n=0 T (n + 21 )π 1+λ 1/2 µ ¶4 ¾!−1/2 T (n + 12 )π ¶2 ¾ Y µ ∞ ½ 1/2 1−λ n=0 T (n + 12 )π ¶2 ¾!−1/2 . Due to the infinite product expressions of cosh x and cos x, this implies (1). 3 Proof of Theorem 1 (ii) In this section, we shall show Theorem 1 (ii). We first describe how we realize {cosh z cos z}1/2 for complex number z. Represent z ∈ C √ as z = reiθ with r ≥ 0 and − 12 π ≤ θ < 23 π to define z = r1/2 eiθ/2 , where i2 = −1. The Derivative of the square norm of Brownian sample path 8 √ √ Riemann surface of the 2-valued function z 1/2 is realized by switching z and − z on the half line consisting of iξ, ξ < 0. Set √  cos z, if a) \|Re z\| < π2 , or    π   b) Im z > 0, − 3π  2 + 4kπ ≤ Re z < 2 + 4kπ (k ∈ Z), or π 3π G(z) = c) Im z < 0, − 2 + 4kπ ≤ Re z < 2 + 4kπ (k ∈ Z),  √  π 5π   − cos z, if a) Im z > 0, 2 + 4kπ ≤ Re z < 2 + 4kπ (k ∈ Z), or   3π b) Im z < 0, 2 + 4kπ ≤ Re z < 7π 2 + 4kπ (k ∈ Z). Then G is holomorphic on C \ {ξ \| ξ ∈ R, \|ξ\| ≥ π/2}, and realizes {cos z}1/2 . Hence G(z)G(iz) is holomorphic on D0 ≡ C \ {ξ, iξ \| ξ ∈ R, \|ξ\| ≥ π/2} and does not vanish in D0 . Recalling that cosh z = cos(iz), we write {cosh z cos z}1/2 for G(z)G(iz). We next extend the identity (1) holomorphically. Since there exists δ > 0 such that exp(δq/2) is integrable with respect to P and q ≥ 0, the mapping Z {z ∈ C \| Re z < δ} 3 z 7→ exp(zq/2)dP W is holomorphic. {cosh(zT ) cos(zT )}−1/2 being holomorphic in D0 , we can find a domain D ⊂ C such that 3 h ¯ n [ π kπ 3π kπ io ¯ D ⊃ reiθ ¯ r ≥ 0, θ ∈ + , + , and 8 2 8 2 k=0 Z 1 for every z ∈ D. exp(z 4 q/2)dP = {cosh(zT ) cos(zT )}1/2 W (12) By (11) and Lemma 1, as an easy application of the Malliavin calculus, we see that the distribution of q/2 on R admits a smooth density function pT (x) ([14, Lemma 3.1]). Since q ≥ 0, pT (x) = 0 for x ≤ 0. Hence, in what follows, we always assume that x > 0. By the inverse Fourier transformation, we have that Z Z 1 −ixt exp(itq/2)dP. (13) e I(t)dt, where I(t) = pT (x) = 2π R W For R > 0, let Γ+ (R) (resp. Γ− (R)) be the directed line segment in C starting at the origin and ending at Reiπ/8 (resp. Re−iπ/8 ). Then, parameterizing Γ± (R) by t1/4 e±iπ/8 , t ∈ [0, R4 ], we have that Z Z 4 i R 4 3 f (z )z dz = ± f (±it)dt 4 0 Γ± (R) for any piecewise continuous function f on iR, where and in the sequel, the symbol ± takes + or − simultaneously. Plugging this into (13), and then substituting (12), we obtain that 2πpT (x) = lim R→∞ ½ Z 4i 4 Γ− (R) z 3 e−xz dz {cosh(zT ) cos(zT )}1/2 Z − 4i Γ+ (R) ¾ 4 z 3 e−xz dz . {cosh(zT ) cos(zT )}1/2 (14) Derivative of the square norm of Brownian sample path 9 Thanks to the estimation that \| cosh(u + iv) cos(u + iv)\|2 ≥ sinh2 u max{cos2 u, sinh2 v}, it is a routine exercise of complex analysis to show that lim R→∞ Z 4 Γ± (R) z 3 e−xz dz {cosh(zT ) cos(zT )}1/2 = Z ∞ 0 4 u3 e−xu du. limh↓0 {cosh(uT ± ih) cos(uT ± ih)}1/2 (15) Moreover, by the definition of {cosh z cos z}1/2 , we have that lim{cosh(uT ± ih) cos(uT ± ih)}1/2 h↓0 (p cosh(uT ) cos(uT ), if − π − (± π2 ) + 4kπ ≤ uT < π − (± π2 ) + 4kπ, p = − cosh(uT ) cos(uT ), if π − (± π2 ) + 4kπ ≤ uT < 3π − (± π2 ) + 4kπ, Substitute this and (15) into (14) to see that 2πpT (x) = 8i ∞ Z X k=0 {(3π/2)+2kπ}/T {(π/2)+2kπ}/T This implies Theorem 1 (ii), because Z {(3π/2)+2kπ}/T {(π/2)+2kπ}/T = References 1 iT 4 π/2 u3 e−xu p (−1)k u3 e−xu p 4 cosh(uT ) cos(uT ) du. 4 cosh(uT ) cos(uT ) du 4 4 {v + (2k + 1)π}3 e−x{v+(2k+1)π} /T p dv cosh{v + (2k + 1)π} cos v 0 Z π/2 4 4 1 {v − (2k + 1)π}3 e−x{v−(2k+1)π} /T p − 4 dv. iT 0 cosh{v − (2k + 1)π} cos v Z [1] R. H. Cameron and W. T. Martin, The Wiener measure of Hilbert neighborhoods in the space of real continuous functions, Jour. Math. Phys. Massachusetts Inst. Technology, 23 (1944), 195 – 209. [2] R. H. Cameron and W. T. Martin, Transformations of Wiener integrals under a general class of linear transformations, Trans. Amer. Math. Soc., 58 (1945), 184–219. [3] T. Hida, Quadratic functionals of Brownian motion, Jour. Multivar. Anal. 1 (1971), 58–69. [4] L. Hörmander, The Analysis of Linear Partial Differential Operators I, 2nd ed., Springer, Berlin, 1990. [5] N. Ikeda, S. Kusuoka, and S.Manabe, Lévy’s stochastic area formula for Gaussian processes, Comm. Pure Appl. Math., 47 (1994), 329–360. Derivative of the square norm of Brownian sample path [6] N. Ikeda and S. Manabe, Asymptotic formulae for stochastic oscillatory integrals, in “Asymptotic Problems in Probability Theory: Wiener Functionals and Asymptotics”, 136– 155, Ed. by K.D.Elworthy and N.Ikeda, Longman, 1993. [7] N. Ikeda and S. Manabe, Van Vleck-Pauli formula for Wiener integrals and Jacobi fields, in “Itô’s stochastic calculus and probability theory”, Ed. by N. Ikeda, S. Watanabe, M. Fukushima, and H. Kunita, pp.141–156, Springer-Verlag, 1996. [8] P. Lévy, Wiener’s random function, and other Laplacian random functions, in “Proc. Second Berkeley Symp. Math. Stat. Prob. II”, pp.171–186, U.C. Press, Berkeley, 1950. [9] Malliavin, P. and Taniguchi, S., Analytic functions, Cauchy formula and stationary phase on a real abstract Wiener space, J. Funct. Anal. 143 (1997), 470–528. [10] H. Matsumoto and S. Taniguchi, Wiener functionals of second order and their Lévy measures, Elec. Jour. Prob. 7, No. 7, 1–30. [11] H. Sugita and S. Taniguchi, Oscillatory integrals with quadratic phase function on a real abstract Wiener space, J. Funct. Anal. 155 (1998), 229–262. [12] H. Sugita and S. Taniguchi, A remark on stochastic oscillatory integrals with respect to a pinned Wiener measure, Kyushu J. Math. 53 (1999), 151 – 162. [13] S. Taniguchi, Lévy’s stochastic area and the principle of stationary phase, J. Funct. Anal. 172 (2000), 165–176. [14] S. Taniguchi, Stochastic oscillatory integrals: Asymptotics and exact expressions for quadratic phase function , in ”Stochastic analysis and mathematical physics (SAMP/ANESTOC 2002)”, ed. R. Rebolledo, J. Resende, and J.-C. Zambrini, pp.165181, World Scientific, 2004. [15] S. Watanabe, Analysis of Wiener functionals (Malliavin calculus) and its applications to heat kernels, Ann. Prob. 15 (1987), 1–39. 10
https://openalex.org/W4301007093	https://digital.lib.washington.edu:443/researchworks/bitstream/1773/15781/1/1472-6963-7-56.pdf	English	null	Utilization and expenditures of veterans obtaining primary care in community clinics and VA medical centers: an observational cohort study	Carolina Digital Repository (University of North Carolina at Chapel Hill)	2,007	cc-by	6,946	BioMed Central BioMed Central Published: 18 April 2007 Published: 18 April 2007 BMC Health Services Research 2007, 7:56 doi:10.1186/1472-6963-7-56 This article is available from: http://www.biomedcentral.com/1472-6963/7/56 This article is available from: http://www.biomedcentral.co © 2007 Maciejewski et al; licensee BioMed Central Ltd. j This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BMC Health Services Research Open Access y Matthew L Maciejewski1,2, Mark Perkins†3, Yu-Fang Li†3,5, Michael Chapko†3,4, John C Fortney†6,7 and Chuan-Fen Liu†3,4 Address: 1Center for Health Services Research in Primary Care, Durham VA Medical Center, Department of Veterans Affairs, Durham, NC, USA, 2Division of Pharmaceutical Outcomes and Policy, School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 3Health Services Research and Development, VA Puget Sound Health Care System, Department of Veterans Affairs, Seattle, WA, USA, 4Department of Health Services, University of Washington, Seattle, WA, USA, 5School of Nursing, University of Washington, Seattle, WA, USA, 6Health Services Research and Development, Center for Mental Health and Outcomes Research, Central Arkansas Veterans Healthcare System, North Little Rock, AR, USA and 7Division of Health Services Research, Department of Psychiatry, College of Medicine, University of Arkansas for Medical Sciences, Little Rock, AR, USA Email: Matthew L Maciejewski - matt@email.unc.edu; Mark Perkins - mark.perkins@med.va.gov; Yu-Fang Li - y Michael Chapko - michael.chapko@med.va.gov; John C Fortney - FortneyJohnC@uams.edu; Chuan-Fen Liu - fliu Email: Matthew L Maciejewski* - matt@email.unc.edu; Mark Perkins - mark.perkins@med.va.gov; Yu-Fang Li - yli@u.washington.edu; Michael Chapko - michael.chapko@med.va.gov; John C Fortney - FortneyJohnC@uams.edu; Chuan-Fen Liu - fliu@u.washington.edu * Corresponding author †Equal contributors * Corresponding author †Equal contributors Received: 8 November 2006 Accepted: 18 April 2007 Methods Sample To generate the sample for this study, we had to determine which CBOCs (and VAMCs to which they were affiliated) to include, and then which veterans from these facilities to sample. CBOCs were included based on the following cri- teria: 1) CBOCs had to have congressional approval, 2) the CBOC had to be open by 10/1/98 to ensure that cost data was tracked for three years to ensure stable estimates, 3) the clinic must have enrolled 200+ veterans in FY99 for sufficient power, 4) cost data had to be trackable sepa- rately from the affiliated VAMC. There were 744 CBOCs and other outpatient clinics in operation in FY00. Only 315 of these clinics were CBOCs that required Congres- sional approval, which was the original process for desig- nating CBOCs. Another 429 outpatient clinics that have been retrospectively labeled CBOCs but did not require Congressional approval were excluded to ensure parity between the CBOCs examined in the prior study and this study. Of these 315 CBOCs, 180 were excluded because they did not meet the inclusion criteria of being in opera- tion before 10/1/98, leaving 135 CBOCs. Twenty-two of the 135 CBOCs were excluded because they cared for fewer than 200 veterans in FY99. The 113 CBOCs with sufficient sample size was reduced by five because they couldn't be found in VA administrative databases (n = 3) or were closed in FY01 (n = 2), resulting in 108 CBOCs that ml the inclusion criteria mentioned above. These 108 CBOCs were affiliated with 72 VAMCs because several VAMCs had established multiple CBOCs. Seventy-six of these 108 CBOCs were VA-staffed CBOCs, and the remaining 32 were Contract CBOCs. The VA made a major commitment to establishing CBOCs, which led to an evaluation of whether CBOC patients and veterans obtaining primary care at VA medi- cal centers (VAMC) had comparable satisfaction, quality of care, utilization and expenditures. Satisfaction and quality of care were found to be comparable between CBOC and VAMC patients [2,3]. Utilization comparisons of veterans at 38 CBOCs and 32 VAMC primary care clin- ics showed that CBOC patients had more primary care encounters, fewer specialty care encounters, and similar hospital admission rates and length of stay as VAMC patients [4]. Of the 38 CBOCs in the Fortney (2002) study, 30 were VA-staffed CBOCs and 8 were Contract CBOCs. Background Fifth, we explicitly controlled for distance to directly examine how this factor affects health care use, to isolate the effects of CBOC provider practice patterns and organ- izational structure. The purpose of this analysis was to compare VA utilization and expenditures of veterans receiving primary care at CBOCs and VAMCs in FY00 and FY01. As the VA continues to rely upon CBOCs as a pri- mary care delivery system that will improve access to care, it remains an open question as to whether CBOCs are actually fulfilling the stated goal of improving primary care access and containing costs. g The VA health care system serves 4.5 million veterans and has been establishing increasing numbers of Community- Based Outpatient Clinics (CBOCs) since 1995 to increase access to primary care for veterans [1]. CBOCs have been an essential component of the VA's transition from an inpatient care-oriented system of medical centers to an outpatient care-oriented organization that expanded pri- mary care while containing costs. The VA has established two types of CBOCs – VA-staffed CBOCs and Contract CBOCs. VA-staffed CBOCs are clinics that the VA either owns or rents and are staffed by VA providers. Contract CBOCs are private practices staffed by non-VA providers that contract with the VA to provide primary care to veter- ans. http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 BMC Health Services Research 2007, 7:56 Abstract Background: To compare VA inpatient and outpatient utilization and expenditures of veterans seeking primary care in community-based outpatient clinics (CBOCs) and VA medical centers (VAMCs) in fiscal years 2000 (FY00) and 2001. Methods: The sample included 25,092 patients who obtained primary care exclusively from 108 CBOCs in FY00, 26,936 patients who obtained primary care exclusively from 72 affiliated VAMCs in FY00, and 11,450 "crossover" patients who obtained primary care in CBOCs and VAMCs in FY00. VA utilization and expenditure data were drawn from the VA's system-wide cost accounting system. Veteran demographic characteristics and a 1999 Diagnostic Cost Group risk score were obtained from VA administrative files. Outpatient utilization (primary care, specialty care, mental health, pharmacy, radiology and laboratory) and inpatient utilization were estimated using count data models and expenditures were estimated using one-part or two-part models. The second part of two-part models was estimated using generalized linear regressions. Results: CBOC patients had a slightly more primary care visits per year than VAMC patients (p < 0.0001), but lower primary care costs (-$71, p < 0.0001). CBOC patients had lower odds of one or more specialty, mental health, ancillary visits and hospital stays per year, and fewer numbers of visits and stays if they had any and lower specialty, mental health, ancillary and inpatient expenditures (all, p < 0.0001). As a result, CBOC patients had lower total outpatient and overall expenditures than VAMC patients (p < 0.0001). Conclusion: CBOCs provided veterans improved access to primary care and other services, but expenditures were contained because CBOC patients who sought health care had fewer visits and hospital stays than comparable VAMC patients. These results suggest a more complex pattern of health care utilization and expenditures by CBOC patients than has been found in prior studies. This study also illustrates that CBOCs continue to be a critical primary care and mental health access point for veterans. Page 1 of 8 (page number not for citation purposes) (page number not for citation purposes) http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 BMC Health Services Research 2007, 7:56 patients could have no primary care visits to its affiliated CBOC, but could have visits to other unaffiliated CBOCs or VAMCs. CBOC patients were defined as veterans who had at least one visit to a CBOC and no primary care visits to the affiliated VAMC in FY00, but several CBOC patients had primary care visits at other VAMCs in FY00 not included in our sample. We defined them as CBOC patients for the purpose of this study because veterans may seek care at VAMCs not affiliated with their CBOC during vacations or other reasons. A crossover patient was a veteran who had at least one visit to a CBOC and an affil- iated VAMC primary care clinic in FY00. In the prior CBOC analyses, 12% of CBOC patients had VAMC and CBOC primary care but were included in the CBOC group [4]. Crossover patients were separately categorized from CBOC patients to examine whether their utilization and expenditures were different from VAMC and CBOC patients' use. We contrast the experience of CBOC patients as a group against the experience of VAMC patients and crossover patients in this paper, and examine the utiliza- tion and expenditure differences between Contract and VA-staffed CBOC patients in a separate paper (available from authors). The second two datasets – the FY00 and FY01 Decision Support System (DSS) Outpatient and Inpatient National Extracts – contained similar utilization and diagnostic data as OPC and PTF, but also have direct, indirect and total expenditures for each VA outpatient visit and inpa- tient hospitalization. Expenditures were inflation- adjusted using the Medical Component of the Current Price Index. We also obtained 1999 Diagnostic Cost Groups (DCG) risk scores that are routinely generated for all veterans receiving care in a given year to adjust for dif- ferential risk between patient groups, because DCG was predictive of utilization and expenditures in the prior CBOC studies [4,5]. The dependent variables for our analysis included outpa- tient and inpatient utilization and expenditure variables. Outpatient utilization and expenditures were partitioned into different types of care by clinic stop codes (e.g., pri- mary care, mental health, specialty, pharmacy, radiology, laboratory, other) and summarized into total outpatient visits and expenditures. Data The five datasets used in this study were drawn from administrative files that are maintained at the VA's central data repository, the Austin Automation Center. The first two datasets – the FY00 Outpatient Care File (OPC) and FY00 Patient Treatment File (PTF) – contained demo- graphic information, county of residence, and utilization data that were used to identify the study sample and con- trol for covariates related to health care utilization and expenditures. The outpatient and inpatient utilization files report the location of care (CBOC or VAMC), clinic stop codes that were used to identify the type of care, clinic visit date or admission and discharge dates, and diagnosis and procedure codes. http://www.biomedcentral.com/1472-6963/7/56 In the VA, lab space is physically separated from provider clinics, so veterans have to go to VA laboratories to have one or more tests done in a single lab "visit". Radiology and pharmacy "visits" work in the same way. A random sample of 250 patients was drawn from each CBOC and VAMC primary care population to obtain suf- ficient power for all analyses and frequency weights were computed for use in regression analysis. For CBOCs with 200–250 enrolled veterans, all veterans were drawn for the study sample and frequency weights for regression analysis equalled 1.0. All crossover patients receiving pri- mary care in the CBOCs and affiliated VAMCs in our sam- ple were included and were assigned frequency weights of 1.0. Patients were then excluded if they didn't survive until the end of FY01 (n = 6,754), or had missing (n = 74) or excessive (e.g., >200 miles) travel distance from their home to the closest VAMC (n = 4,688), which resulted in 23,894 patients in 108 CBOCs, 26,176 patients from the 72 affiliated VAMCs, and 11,074 crossover patients. Inpatient utilization was defined as the number of hospi- talizations, and inpatient expenditures were defined as the sum of expenditures across all hospitalizations in the same year. Demographic information used as independ- ent variables regression analysis included age, gender, race, marital status, means test-based eligibility for free care, service-related disability, and 1999 DCG risk score. We also adjusted for distance between veteran's zipcode and the veteran's usual source of care (e.g., VAMC or CBOC) to control for access issues related to VAMC and CBOC travel costs. Differential distance, defined as (Dis- tance between VAMC county and home county of resi- dence) – (Distance between CBOC county and home county of residence), was used in primary care analyses, but distance between VAMC county and home county of residence was used to analyze all other outcomes. Page 3 of 8 (page number not for citation purposes) Methods Sample The 18 VA-staffed CBOCs included in the Macie- jewski study [5] were a subset of the 30 CBOCs from the Fortney sample [4]. Comparisons of direct outpatient and inpatient expenditures of veterans at 18 VA-staffed CBOCs and 14 VAMCs showed that CBOC patients had lower specialty and total expenditures than VAMC patients, but similar primary care and inpatient expenditures [5]. In this paper, we extend the prior utilization and expend- iture analyses of CBOC and VAMC patients in five ways. First, we examined total (indirect+direct) VA costs for CBOC, VAMC and crossover patients. The VA's cost accounting system has been more rigorously imple- mented and checked for data quality to enable inclusion of indirect costs that were excluded in the prior analyses. Second, we used a much larger sample of CBOCs (108 vs. 18), VAMCs (72 vs. 14) and veterans to strengthen the generalizability of our results. Third, we included data from fiscal years (FY) 2000 and 2001 to update the com- parison. Fourth, we explicitly track "crossover" patients who obtained primary care at CBOCs and VAMCs in FY00 to assess whether this discontinuity in primary care is associated with different utilization and expenditures. In the prior study, these patients were lumped into the CBOC cohort and accounted for 12% of that group. Veterans in this study were classified into one of three mutually exclusive groups based upon where the veteran obtained primary care in FY00: VAMC patients, CBOC patients or crossover patients. VAMC patients were defined as veterans who had no primary care visits to a CBOC and at least one primary care visit to a VAMC pri- mary care clinic, including general internal medicine, ger- iatrics, women's health, or primary care. This definition of primary care stop codes was consistent with the primary care codes defined in previous studies [4,5]. VAMC Page 2 of 8 (page number not for citation purposes) Page 2 of 8 (page number not for citation purposes) Analysis To assess whether CBOCs influence the probability of care and/or the level of care, we estimate three measures of VA resource use – 1) odds of service use, 2) number of outpa- tient visits or inpatient hospitalizations, and 3) expendi- tures incurred. Ninety percent of veterans had primary care and outpatient pharmacy visits and 95% of veterans had one or more outpatient visits of all kinds, so the odds of use were not estimated for these two outcomes. Bivari- ate comparisons of odds of service were assessed via anal- ysis of variance methods and bivariate comparisons of utilization and costs were assessed by the Cuzick exten- Page 3 of 8 (page number not for citation purposes) Page 3 of 8 (page number not for citation purposes) http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 BMC Health Services Research 2007, 7:56 sion of the nonparametric Wilcoxon rank-sum test for three or more groups [6]. sion of the nonparametric Wilcoxon rank-sum test for three or more groups [6]. Since veterans were not randomly assigned to groups, we used propensity score matching with quintiles to improve the balance in the observed characteristics of veterans seen in CBOCs and VAMCs [13]. To account for the possibility that the clustering of CBOCs within VAMCs and clustering of VAMCs within Veteran Integrated Service Networks (VISNs) might impact our variance estimates, we also tested models that included VAMCs as random effects and VISNs as fixed effects. These propensity score and cluster adjustments obtained similar results to our unadjusted models, so the unadjusted models are presented. In multivariate analysis, one-part models and two-part models of utilization and costs were run. One-part mod- els were run for outcomes that had few observations with zero use (e.g., primary care, total outpatient, and overall care). Two-part models are used to estimate the odds of use and the level of use separately for those that use health care [7]. A hurdle variant of the two-part model was run for outcomes that had many observations with zero val- ues (e.g., mental health, specialty, radiology, laboratory, other outpatient, inpatient) because of the poor numeri- cal properties of count data models (e.g., zero-inflated negative binomial models) that use the same covariates in both parts of the model [8]. The odds of mental health, specialty, and ancillary (radiology, laboratory, and other) outpatient service use, as well as odds of inpatient admis- sion, were estimated using logit regressions. Unadjusted utilization and cost The odds of primary care, pharmacy and total outpatient use were significantly different between CBOC, VAMC and crossover patients, but the differences were not large (Table 2, columns 2–4). Bigger differences in the odds of use were found in specialty, mental health, ancillary care, and for inpatient care. CBOC patients had significantly lower odds of use of these services than VAMC patients, who had significantly lower odds of use than crossover patients (p < 0.001). The average number of outpatient visits and inpatient admissions per person per year were also significantly lower for CBOC patients than for VAMC patients (p < 0.001). CBOC patients had 3.3 primary care visits per year on average, while VAMC patients had 3.6 primary care visits per year and crossover patients had 5.0 primary care visits per year (p < 0.001). The same trend was found in VAMC-based specialty care visits per year (2.1 for CBOC patients vs. 4.9 for VAMC and crossover patients, p < 0.001), and total outpatient visits per year (23 visits for CBOC patients vs. 36 visits for VAMC patients vs. 40 visits for crossover patients, p < 0.001) (Table 2, columns 5–7). Expenditure models for overall care, outpatient care, pri- mary care and outpatient pharmacy were estimated using one-part generalized linear models (GLMs) with a normal distribution and log link function to restore log-normal- ity. GLMs have been shown to fit expenditures with signif- icant heteroskedasticity better than ordinary least squares [10] and have the added benefit of yielding predicted expenditures without having to retransform log costs. In a comparison of various generalized linear models and ordinary least squares from another CBOC study [11], the GLM with normal distribution and log link fit the middle three quintiles of the expenditure distribution better than OLS, based on predictive ratios [12]. Inpatient, specialty, mental health, radiology, laboratory or other outpatient expenditures for those with one or more visits were also estimated using GLMs with a normal distribution and log link function. Specifications, cluster corrections for repeated measures on each veteran, and sampling weights were standardized across all regressions. Significance was lowered from 0.05 to 0.01 to adjust for multiple compar- isons. Unadjusted VA outpatient expenditures followed the pat- terns of outpatient utilization (Table 2, columns 8–10). Patient characteristics CBOC patients are younger and more likely to be married than VAMC and crossover patients (Table 1). Two proxies of health risk – veterans' service-related disability and DCG risk score – are lower for CBOC patients than VAMC or crossover patients. A lower percentage of VAMC patients are above the means-test threshold and required to pay VA copayments than CBOC and crossover patients. Finally, CBOC patients live farther from VAMCs than VAMC and crossover patients, which results in greater dif- ferential distances (Table 1). Analysis The number of outpatient visits and number of hospitalizations were estimated using count data methods [8]. The number of mental health, specialty, and ancillary visits for those with visits, and the number of hospitalizations for those who were hospitalized, was estimated with negative binomial models. This "modified" two-part model approach allows the conventional, separate modeling of the process for generating zeros and a count data model for users that explicitly accounts for overdispersion. Poisson regression models were estimated for total outpatient visits, primary care visits, and outpatient pharmacy visits because overd- ispersion due to high utilizers was not a problem [9]. Unadjusted utilization and cost Average primary care expenditures per person per year were lowest for CBOC patients and highest for crossover patients ($381 for CBOC, $590 for VAMC and $649 for crossover patients, p < 0.001). Total outpatient expendi- tures averaged $2,092 for CBOC patients, $3,921 for Page 4 of 8 (page number not for citation purposes) Page 4 of 8 (page number not for citation purposes) BMC Health Services Research 2007, 7:56 http://www.biomedcentral.com/1472-6963/7/56 Table 1: Demographic Characteristics of CBOC, VAMC and Crossover Patients VAMC Patients CBOC Patients Crossover Patients Age 60.4 (14.3) 64.0 (13.0) 60.0 (14.2) Male (%) 94.6 96.4 92.3 Married (%) 54.1 67.8 58.9 Not Married (%) 43.7 31.7 39.8 Unknown Marital (%) 2.2 1.5 1.3 Caucasian (%) 51.3 48.0 60.2 Non-Caucasian (%) 17.3 6.8 12.1 Unknown Race (%) 31.4 45.2 27.7 Service Connection (%) 15.9 (28.0) 11.5 (23.6) 21.6 (30.9) DCG score, 1999 0.61 (1.00) 0.34 (0.64) 0.62 (0.98) Means test status (%) No means test required 36.7 30.4 46.4 Above means test 15.9 27.0 22.2 Below means test 44.7 41.8 39.9 Not applicable 2.7 0.8 1.5 Differential Distance 5.7 (33.0) 44.5 (35.6) 34.8 (37.1) Home to VAMC Distance 27.5 (31.0) 60.7 (37.7) 51.8 (39.0) Sample Size 26,176 23,894 11,074 Table 1: Demographic Characteristics of CBOC, VAMC and Crossover Patients CBOC patients had significantly lower odds of having specialty, mental health, and ancillary (radiology, labora- tory and other outpatient) visits than VAMC patients (p < 0.0001 for all). CBOC patients with use had fewer spe- cialty, mental health, outpatient pharmacy, and ancillary visits than VAMC patients that used these services. Lower CBOC utilization (for users) translated into significantly lower expenditures for these same services, which resulted in lower total outpatient expenditures compared to VAMC patients (-$923, p < 0.0001). CBOC patients were less likely to be hospitalized than VAMC patients (odds = 0.48, p < 0.0001), but those CBOC patients who were hospitalized had similar expenditures as VAMC patients who were hospitalized. As a result of significantly lower total outpatient expenditures, CBOC patients had lower VAMC patients, and $4,258 for crossover patients (p < 0.001). Average inpatient expenditures per year averaged $656 for CBOC patients, $1,867 for VAMC patients and $1,964 for crossover patients (p < 0.001). Overall expen- ditures averaged $2,748 for CBOC patients, $5,788 for VAMC patients, and $6,222 for crossover patients (p < 0.001). All outcomes are significantly different at a p-value < 0.0001; g y p Numbers in parentheses refer to Standard Deviations; All analyses were adjusted for sampling weights. p All analyses were adjusted for sampling weights. http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 BMC Health Services Research 2007, 7:56 overall expenditures (-$1551, p < 0.0001) per person per year than VAMC patients. tient ancillary care services provided at VAMCs for those veterans who received one or more services, despite the fact that the odds of using these services was higher for CBOC patients. These utilization and expenditure differ- ences held even when relative distance, demographic characteristics and patient risk were controlled [15,16]. Crossover patients who obtained primary care in CBOCs and VAMCs in FY00 had 1.3 more primary care visits and higher primary care expenditures than VAMC patients in both years (p < 0.0001). They also had greater odds of mental health and radiology visits, but lower odds of lab- oratory visits. Crossover patients with use had more out- patient pharmacy visits and expenditures ($103, p < 0.0001) than VAMC patients. Total outpatient expendi- tures and overall expenditures were significantly higher for crossover patients ($364 and $491, respectively). The results from this study differ from earlier studies that used smaller samples in 1998. CBOC-VAMC visit differ- ences were smaller in this study than in an earlier study [4] for primary care (0.1 visits vs. 1.7 visits in prior study) and specialty care (-0.4 visits vs. -1.8 in prior study), but hos- pital admissions were similar. CBOC-VAMC expenditure differences were similar for specialty care and inpatient care, but were significantly different for primary care (-$71 vs. $30 in prior study) [5]. The differences in total costs from this study were five times larger than that from the earlier study (-$1551 vs. -247 in prior study). These expenditure differences could be due to a number of fac- tors, including a larger sample size (108 vs. 18), the inclu- sion of Contract CBOCs into this sample, and more complete adjustment for confounders such as distance. Distance to a VAMC was an important determinant of uti- lization and expenditures, particularly for VAMC-based services, and this variable was not controlled in the earlier studies. Time and travel costs incurred for accessing CBOCs and VAMCs might have been an important factor contributing to lower specialty visits and expenditures of CBOC patients, because distance has shown to be an important determinant of veterans' services use in other studies [17-19]. Distance was a significant determinant of primary care uti- lization and expenditures, as well as radiology, laboratory and outpatient pharmacy utilization (p < 0.0001, not reported). http://www.biomedcentral.com/1472-6963/7/56 Greater travel distance to a VAMC was associ- ated with higher odds of being hospitalized (p < 0.01), but did not affect the level of inpatient expenditures once a veteran was hospitalized. This result is inconsistent with an earlier study that found that distance is positively asso- ciated with inpatient utilization [14]. Adjusted utilization and cost Odds ratios and predicted visit and cost differences adjusted for age, gender, race, marital status, service con- nection, distance, means-test status, and DCG risk are listed in Table 3. CBOC patients had 0.1 fewer adjusted primary care visits per year as VAMC patients, and had lower primary care expenditures (-$81, both p < 0.0001). Table 2: Unadjusted Utilization and Expenditures of CBOC, VAMC and Crossover Patients in 2000–2001 Patients with at least one Visit or Admission Number of visits or admissions Per Person Per Year Expenditures Per Patient Per Year CBOC VAMC Crossover CBOC VAMC Crossover CBOC VAMC Crossover Outpatient Primary Care 94% 93% 96% 3.3 (2.8) 3.6 (3.1) 5.0 (3.9) $381 (438) $509 (626) $649 (720) Specialty Care 48 72 75 2.1 (4.0) 4.9 (8.0) 4.9 (6.5) 413 (1706) 1013 (3568) 1077 (4098) Mental Health 13 23 27 0.8 (4.9) 2.8 (14) 2.6 (12) 102 (665) 342 (1733) 302 (1363) Pharmacy 90 90 95 -- -- -- 684 (2009) 936 (2367) 1062 (2083) Radiology 30 55 58 0.6 (1.2) 1.3 (1.9) 1.4 (2.0) 111 (717) 251 (524) 287 (556) Laboratory 68 83 83 2.2 (2.9) 3.5 (4.6) 3.7 (4.3) 119 (332) 208 (410) 238 (3473) Total Outpatient 97 97 99 23 (22) 36 (38) 40 (34) 2092 (3692) 3921 (6294) 4258 (7235) Inpatient 5 13 15 0.1 (0.4) 0.2 (0.8) 0.2 (0.8) 656 (5120) 1867 (9306) 1964 (9042) OVERALL -- -- -- -- -- -- 2748 (6936) 5788 (12351) 6222 (12701) Sample Size 23,894 26,176 11,074 23,894 26,176 11,074 23,894 26,176 11,074 n and Expenditures of CBOC, VAMC and Crossover Patients in 2000–2001 able 2: Unadjusted Utilization and Expenditures of CBOC, VAMC and Crossover Patients in 2000–2001 Page 5 of 8 (page number not for citation purposes) Discussion CBOCs have been an integral part of the VA's continuing transition into a primary care-focused health care system. These outpatient clinics were established to improve vet- eran access to primary care and contain total health care expenditures, and they appear to be providing primary care access without a commensurate increase in primary care expenditures. Overall expenditures of CBOC patients were $1,588 lower than for VAMC patients because of lower use of inpatient, specialty, mental health and outpa- This study had several limitations that must be acknowl- edged. First, veterans were not randomized to VA treat- Table 3: Adjusted Utilization and Expenditures of CBOC, VAMC and Crossover Patients in 2000–2001 Table 3: Adjusted Utilization and Expenditures of CBOC, VAMC and Crossover Pat Table 3: Adjusted Utilization and Expenditures of CBOC, VAMC and Crossover Patients in 2000–2001 Odds Ratio (for two-part models only) Difference in predicted Visits or admissions per patient per year Difference in Predicted Expenditures Per Patient Per Year CBOC Crossover CBOC Crossover CBOC Crossover Outpatient Primary Carea -- -- 0.10 (0.03)* 1.34 (0.04)* -71 (5)* 130 (6)* Specialty Careb 0.46 (0.01)* 1.10 (0.03) -1.42 (0.07)* -0.10 (0.08) -339 (33)* 60 (47) Mental Healthb 0.78 (0.02)* 1.14 (0.03)* -1.85 (0.35)* -0.60 (0.35). -282 (43)* -78 (47). Pharmacya -- -- -- -- -81 (20)* 103 (17)* Radiologyb 0.41 (0.01)* 1.05 (0.02)* -0.39 (0.02)* 0.03 (0.03) -58 (14)* 34 (8)* Laboratoryb 0.45 (0.01)* 0.90 (0.03)* -0.41 (0.04)* 0.31 (0.04)* -34 (4)* 29 (16). Reference Group is VAMC patients; *P-value < 0.0001, P-value < 0.001, P-value < 0.01; Acknowledgements This research was supported by an Investigator Initiated Research Award (IIR 20-005-3) from the Department of Veterans Affairs, Health Services Research and Development. Helpful comments from two reviewers and research assistance from Song Wang are gratefully acknowledged. Dr. Maciejewski is presently at the Center for Health Services Research in Pri- mary Care at the Durham VA Medical Center. Drs. Liu, Li, Perkins and Chapko are presently at the Northwest Center for Outcomes Research in Older Adults at the VA Puget Sound Health Care System. Dr. Fortney is presently at the Center for Mental Health and Outcomes Research at the Little Rock VA. The views expressed herein are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs, the University of North Carolina at Chapel Hill, the University of Washington, or the University of Arkansas for Medical Sciences. Authors' contributions MLM, CFL, MC and JCF conceived of the study, wrote both grant applications, designed the analyses, inter- preted the results, and drafted and revised the manuscript. YFL and MP cleaned and linked all of the data files, and conducted all of the statistical analyses. All authors read and approved the final manuscript. As the current veteran population ages and develops more chronic conditions and veterans from the Iraq Theater enter the VA system, it will be useful to examine whether relatively healthier veterans will continue to choose CBOC care or if the health profile of CBOC and VAMC patients become more similar. If the health profile of patients becomes more similar over time and the care of CBOC patients becomes more complex, CBOC-VAMC cost differences observed in this study may change. CBOCs will be challenged to maintain their current per- formance, because CBOCs may have created strong patient-provider relationships that patients valued enough to seek CBOC care in situations where specialty care would have been appropriate. CBOC patients do not have onsite access to the broad range of specialty and mental health clinics available on VAMC campuses. CBOC providers may also have differed from VAMC pro- viders in conformity with evidence-based medicine [21]. http://www.biomedcentral.com/1472-6963/7/56 In the absence of more complete quality of care and health out- come measures, it is unknown whether lower utilization and expenditures had adverse effects for veterans at CBOCs. Fourth, our CBOC sample included early entrants that had been in operation for three years by 2001, so our results may not generalize to CBOCs that have been estab- lished more recently or to the start-up phase of CBOCs. Finally, we did not have data on non-VA utilization (e.g., Medicare) to assess whether VA differences were offset by non-VA utilization and cost differences. These results rep- resent these veterans' VA experience only. has been found in prior studies, but also illustrates that CBOCs continue to be a critical primary care and mental health access point for veterans. It will be important to track CBOC performance into the future as the program matures and CBOCs add services, such as the recently mandated mental health capacity. Future investigations should examine what types of CBOC patients return to VAMCs for all of their health care needs, how CBOC and VAMC primary care clinic staffing, practice styles, management and decision technology dif- fer, what factors are associated with CBOCs being primary care centers of excellence, and how the new Medicare pre- scription drug benefit affects elderly veterans' use of CBOCs and VAMC primary care clinics. CBOCs are cur- rently in operation throughout the VA health care system and appear to be serving the VA and veterans as intended. Competing interests p g All of the authors are employees of and receive salary from the Health Services Research and Development service of the Veterans Health Administration, which administers CBOC and VAMC services to veterans. Otherwise, we have no financial or non-financial competing interests. References 1. Chapko MK, Borowsky SJ, Fortney JC, Hedeen AN, Hoegle M, Macie- jewski ML, VanDeusen Lukas C: Evaluation of the Department of Veterans Affairs community-based outpatient clinics. Med Care 2002, 40:555-560. http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 ment site, so it is possible that CBOCs experienced favorable selection and unmeasured covariates (e.g., dis- ease severity, propensity to seek primary care) related to CBOC status and resource use could have biased the effect of CBOC on utilization and expenditures [20]. We attempted to reduce unobserved confounding related to CBOC status and resource use by controlling for demo- graphic characteristics and patient risk (via the DCG risk score), distance, copayment status, marital status and service-related disability, but this confounding may not have been completely eliminated. Second, a consistent specification was used across all outcomes to simplify presentation, which may have led to specification biases that would modify our findings. Third, we had no data on health outcomes and only limited data on quality of care that showed no difference in provision of diabetic eye exams and flu shots for patients with COPD. In the absence of more complete quality of care and health out- come measures, it is unknown whether lower utilization and expenditures had adverse effects for veterans at CBOCs. Fourth, our CBOC sample included early entrants that had been in operation for three years by 2001, so our results may not generalize to CBOCs that have been estab- lished more recently or to the start-up phase of CBOCs. Finally, we did not have data on non-VA utilization (e.g., Medicare) to assess whether VA differences were offset by non-VA utilization and cost differences. These results rep- resent these veterans' VA experience only. ment site, so it is possible that CBOCs experienced favorable selection and unmeasured covariates (e.g., dis- ease severity, propensity to seek primary care) related to CBOC status and resource use could have biased the effect of CBOC on utilization and expenditures [20]. We attempted to reduce unobserved confounding related to CBOC status and resource use by controlling for demo- graphic characteristics and patient risk (via the DCG risk score), distance, copayment status, marital status and service-related disability, but this confounding may not have been completely eliminated. Second, a consistent specification was used across all outcomes to simplify presentation, which may have led to specification biases that would modify our findings. Third, we had no data on health outcomes and only limited data on quality of care that showed no difference in provision of diabetic eye exams and flu shots for patients with COPD. Discussion Total Outpatienta -- -- -- -- -923 (39) 364 (52)* Inpatientb 0.48 (0.02)* 1.08 (0.03) -0.12 (0.04)* 0.01 (0.03) -109 (577) 81 (497) OVERALLa -- -- -- -- -1551 (70)* 491 (91)* Sample Size 126,956 Variable Variable Reference Group is VAMC patients; *P-value < 0.0001, P-value < 0.001, P-value < 0.01; Utilization and expenditure regressions control for age, age-squared, female gender, age and gender interaction, marital status, race, veteran percent service connection, distance, DCG risk score, copay status and time (2000); All analyses were adjusted for sampling weights; a – one-part models of utilization and costs estimated on the entire sample and odds of use was not estimated b – two-part models of utilization and costs estimated on veterans with use > 0 and odds of use estimated on entire sample Reference Group is VAMC patients; P-value < 0.0001, P-value < 0.001, *P-value < 0.01; Utilization and expenditure regressions control for age, age-squared, female gender, age and gender interaction, marital status, race, veteran percent service connection, distance, DCG risk score, copay status and time (2000); All analyses were adjusted for sampling weights; a – one-part models of utilization and costs estimated on the entire sample and odds of use was not estimated b – two-part models of utilization and costs estimated on veterans with use > 0 and odds of use estimated on entire sample Page 6 of 8 (page number not for citation purposes) BMC Health Services Research 2007, 7:56 BMC Health Services Research 2007, 7:56 BMC Health Services Research 2007, 7:56 http://www.biomedcentral.com/1472-6963/7/56 http://www.biomedcentral.com/1472-6963/7/56 4. Fortney JC, Borowsky SJ, Hedeen AN, Maciejewski ML, Chapko MK: VA community-based outpatient clinics: access and utiliza- tion performance measures. Med Care 2002, 40:561-569. p 5. Maciejewski ML, Chapko MK, Hedeen AN, Fortney JC: VA commu- p 5. Maciejewski ML, Chapko MK, Hedeen AN, Fortney JC: VA commu- nity-based outpatient clinics: cost performance measures. Med Care 2002, 40:587-595. p 5. Maciejewski ML, Chapko MK, Hedeen AN, Fortney JC: VA commu- nity-based outpatient clinics: cost performance measures 6. Cuzick J: A method for analysing case-control studies with ordinal exposure variables. Biometrics 1985, 41:609-621. 7. Manning WG, Newhouse JP, Duan N, Keeler EB, Leibowitz A, Mar- quis MS: Health insurance and the demand for medical care: evidence from a randomized experiment. Am Econ Rev 1987, 77:251-277. 8. Cameron AC, Trivedi PK: Regression Analysis of Count Data Cam- bridge, UK: Cambridge University Press; 1998. 9. Jones AM: Health Econometrics Elsevier; 2000. 10. Manning WG, Mullahy J: Estimating log models: to transform or not to transform? J Health Econ 2001, 20:461-494. J 11. Fortney JC, Maciejewski ML, Warren JJ, Burgess JF: Does Establish- ing VHA Community Based Outpatient Clinics in Unders- erved Areas Impact Patterns of Utilization and Costs? Inquiry 2005, 42:29-42. 12. Maciejewski ML, Liu CF, Derleth A, McDonell MB, Anderson SM, Fihn SD: The Performance of Administrative and Self-Reported Measures for Risk Adjustment of VA Primary Care Patient Costs. Health Serv Res 2005, 40:887-904. 13. D'Agostino RB Jr: Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized con- trol group. Stat Med 1998, 17:2265-81. g p 14. Welch HG, Larson EB, Welch WP: Could distance be a proxy for severity-of-illness? A comparison of hospital costs in distant and local patients. Health Serv Res 1993, 28:441-458. p 15. Ash AS, Ellis RP, Pope GC, Ayanian JZ, Bates DW, Burstin H, Iezzoni LI, MacKay E, Yu W: Using diagnoses to describe populations and predict costs. Health Care Financ Rev 2000, 21:7-28. 16. Pope GC, Ellis RP, Ash AS, Liu CF, Ayanian JZ, Bates DW, Burstin H, Iezzoni LI, Ingber MJ: Principal inpatient diagnostic cost group model for Medicare risk adjustment. Health Care Financ Rev 2000, 21:93-118. 17. Burgess JF Jr, DeFiore DA: The effect of distance to VA facilities on the choice and level of utilization of VA outpatient serv- ices. Soc Sci Med 1994, 39:95-104. 18. BMC Health Services Research 2007, 7:56 LaVela SL, Smith B, Weaver FM, Miskevics SA: Geographical prox- imity and health care utilization in veterans with SCI&D in the USA. Soc Sci Med 2004, 59:2387-2399. 19. Schmitt SK, Phibbs CS, Piette JD: The influence of distance on uti- lization of outpatient mental health aftercare following inpa- tient substance abuse treatment. Addict Behav 2003, 28:1183-1192. 20. Heckman JJ: Sample Selection Bias as a Specification Error. Econometrica 1979, 47:153-161. 21. Hofer TP, Zemencuk JK, Hayward RA: When there is too much to do: how practicing physicians prioritize among recom- mended interventions. J Gen Intern Med 2004, 19:646-653. Conclusion CBOCs provided veterans improved access to primary care and other services, but expenditures were contained because CBOC patients who sought health care had fewer visits and hospital stays than comparable VAMC patients. These results suggest a more complex pattern of health care utilization and expenditures by CBOC patients than 2. Borowsky SJ, Nelson DB, Nugent SM, Bradley JL, Hamann PR, Stolee CJ, Rubins HB: Characteristics of veterans using Veterans Affairs community-based outpatient clinics. J Health Care Poor Underserved 2002, 13:334-346. 3. Hedeen AN, Heagerty PJ, Fortney JC, Borowsky SJ, Walder DJ, Chapko MK: VA community-based outpatient clinics: quality of care performance measures. Med Care 2002, 40:570-577. 3. Hedeen AN, Heagerty PJ, Fortney JC, Borowsky SJ, Walder DJ, Chapko MK: VA community-based outpatient clinics: quality of care performance measures. Med Care 2002, 40:570-577. 3. Hedeen AN, Heagerty PJ, Fortney JC, Borowsky SJ, Walder DJ, Chapko MK: VA community-based outpatient clinics: quality of care performance measures. Med Care 2002, 40:570-577. Page 7 of 8 (page number not for citation purposes) Page 7 of 8 (page number not for citation purposes) Pre-publication history The pre-publication history for this paper can be accessed here: The pre-publication history for this paper can be accessed here: Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 8 of 8 (page number not for citation purposes) http://www.biomedcentral.com/1472-6963/7/56/prepub http://www.biomedcentral.com/1472-6963/7/56/prepub Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge
https://openalex.org/W1967047840	https://zenodo.org/records/2515242/files/article.pdf	German	null	Über die Absorptions‐ und die Refraktionsmethode	Berichte der Deutschen Chemischen Gesellschaft	1,912	public-domain	8,176	127. K. Auwera: ffber die Absorptione- und die Refraktionsmethode. (Eingegangen am 18. Mirz 1912.) (Eingegangen am 18. Mirz 1912.) Die beiden soeben I) erschienenen Arbeiten von H a n t z s c h uber das in der Uberschrift genannte Thema veranlassen mich zu den fol- genden Bemerkungen. Hr. H a n t z s c h teilt in seiner ersteii Arbcit mit, daB sich mit IIilfe der A b s o r p t i o n s s p e k t r e n auBerst geringe Unterschiede im Reinheitsgrad verschiedener Fraktionen von T e r p e n - P r i p a r a t e n nachweisen lasscn. Im AnschluB darnn wirft Hr. H a n t z s c h (S. 5jS) mir vor, d d ich bei meinen spektrochemischen Unterbuchungen uber Terpene die Veranderungen, die diese K6rper bei dcr Destillation unter gewohnlichem Druck erleiclen, mit Stillschweigen iibergehe, und die sehr bekannte Zersetzlichkeit der Terpene nicht geniigend berucksichtigt habe. 1. g g g Ans dcn spektrochemiachen Arbeiten von mir untl meinen Mitarbeitern geht, glaube ich, zur Geniige hervor, daB wir uns stets nach MGglichkeit be- mcht haben, reioe Praparate zu gewinnen, und in allen erforderlichen Fillen auch die Destillation im Vakuum aber Natriurn mit oder ohne indifferenten Gasstroiii angewandt haben. Bci den Terpenen war dies nicht netig, denu Hr. H an tzsch iiberschatzt die Zersetzlickkeit dicser Verbinduogen. Durch mehrmalige Destillation unter gew6hnlichem oder vermindertem Druck erhilt man Priparate von gleicheu Eigenschaften, die sich bei meiteren Destillationen nicht mehr verindern, soweit man dies mit den bisher g e b r i u c h l i c h e n Hilfsmit teln feststellen kann. Insbesondere stimmen die Dichten und Brechungs- indices und demnach auch die spektrochemischen Konstanten der vcrschiedenen Priparate stets innerhalh der zulassigen Versuchsfehler, in der Regel sogar sehr gut, iiberein. g DaB die Kohlenwasserstoffe der T e r p e n - R e i h e bei der gewbhn- lichen Destillation nicht in merklicher Weise vergndert merdan, diirftc hberdies schon daraus hervorgehen, daB der b a t e Kenner dieses Gebietes, Wallac h, bei diesen Karpcrn fast ausschlieBlich diese Art der Destillation und nur in Ausnahmefillen die Vakuumdestillation anwendet. Das Gleiche beweisen noch weit schirfer die c a l o r i m e t r i s c h e n Be- s t i m m u n g e n meincs Kollegen W. A. Roth. 1) B. 46, 553, 559 [1912]. a) A. 385, 10G [1911]. 127. K. Auwera: ffber die Absorptione- und die Refraktionsmethode. Beispielsweise wichen dieVer- brennungswirmen dreier Priparate Limonen) die s5mtlich in gew6hn- licher Weise destilliert wortlen waren, nur urn 1-2O/oo von einander ab, wihrend bei partieller Osydation oder Polymerisation schwankende nnd vor allem zu niedrige Werte gefunden worden wiren. Der erwghnte Vorwurf von Hrn. H a n t z s c h ist daher angerechtfertigt. Dagegen trifft seine Vermutung, da4 die Refraktionsmcthode vie1 u n - empfindlicher als die Absorptionemethode sei, vollkommen zu. Dies hitte Hr. H a n t z s ch aber schon aus meinen friiheren Arbeiten nrtnehmen Icijnnen, 964 964 in tlenen ich bei jeder passenden Gelegenheit betont habc, dab die spektro- chemische Untereuchung niemals die absolute Reinheit eines Przparats ver- hiirgen kann. So schrieb ich z. B. nach einer Besprechung der spektroche- CHa in tlenen ich bei jeder passenden Gelegenheit betont habc, dab die spektro- chemische Untereuchung niemals die absolute Reinheit eines Przparats ver- hiirgen kann. So schrieb ich z. B. nach einer Besprechung der spektroche- CHa mischen Konstanten von Kohlenwasserstoffen der Formel /=-\. C KR': \-- /'. CHa BWie weit unscre Kohlenwasserstoffe ein hei tlich sind, bleibt, wie stets hei tlerartigen Snbstanzen, eine offene Frage, die sich mit .nnseren heutigen Hilfsmitteln nicht streng beantworten lafit ').a g Und kiirzlich: DFreilich ist dabei, wic stets, die Einschrankung zu machen, daB etwa vorhandene geringe Mengen isoniercr Stoffo spektrochernisch nicht erkanot werden k h n c n 3.. 2. Aul Grund der n b s o r p t i o m e t r i s c h e u U n t e r s u c h u n g e i n i g e r T e r p e n e sagt Hr. I I a n t z s c h , daB die Absorptionsmethode gegeniiber der Refraktionsmethode viele Vorzuge besitze. hngefuhrt werden ala solche: grijISere Einfanhheit uncl Anschaulichkeit, sowie griiaere Empfindlichkeit. 2. Aul Grund der n b s o r p t i o m e t r i s c h e u U n t e r s u c h u n g e i n i g e r T e r p e n e sagt Hr. I I a n t z s c h , daB die Absorptionsmethode gegeniiber der Refraktionsmethode viele Vorzuge besitze. hngefuhrt werden ala solche: grijISere Einfanhheit uncl Anschaulichkeit, sowie griiaere Empfindlichkeit. I ) B. 48, 3053 [1910]. 9 A. 387, 251 [1912]. - Dieae Arbeit war Hm. Ilantzscli Iiri fassung seiner Abhandlung noch nicht bekannt. 127. K. Auwera: ffber die Absorptione- und die Refraktionsmethode. Was den ersten Punkt betrifft, so wird bei der Beurteilung, welche Methode ei u f ac h e r ist, sahr der Urnstand mitsprechen, auf welches Verfahren der Beohachter besser eingeubt ist. Die Bestimmung der Dichte und der vier Brechungsindices, sowie die Berechnung der Mol- Refralition und -Dispersion ous diesen Daten, die durch das R o t h - E i s e n l o h rsche Hilfsbuch auBerordentlich erleichtert ist, erfordern im ganzen an Zeit je nach der Gewandtheit iind Ubung des Beob:ichters 1/g-3/4 Stunden, wenn, wie in der Regel, die Beobachtungen bei ge- wohnlicher Temperatur vorgenomrnen werden konnen, und mati auf derartige Bestimmungeu eingerichtet ist. Bei der anderen Methode mu13 das Absorptionsspelitrum photographiert, die Platte entwickelt iind schliefilich eine genaue Zeichnung der Kurve angefertigt werden. Ich bezweifle daher, ob sich eine absorptiometrische Bestimmung in kiirzerer Zeit erledigen ISiWt und einfacber ist. Jedenfalls diirfte dieser Punkt bei einem Vergleicb beider Methoden keine wesantliche Rolle spielen. Nine graphische Darstellung ist selbstredend im allgemeinen a n - s c h a u l i c h e r nls eine Zablenreihe. Ob aber im einzelnen Fall aus tlem Yergleich einer Reihe, oft komplizierter, Kurven oder aus der Gegeniiberstellung von Zahlen leichter und sicherer Schliisse aiif das Verhaltnis der untersuchten Korper gezogen werdcn konnen, i5t eine :tndere Frage, bei deren Rexntwortung Ubung und Gewohr~heit wieder mitsprecben miigen. Auch auf diesen Punkt darf daher m. E. kein groBes Gewicht gelegt werden. DaB die Absorptionsmethode in den von Hrn. H a n t z s c h gege- benen Beispielen durch g r o Be r e E m pf i n d l i c h k ei t ausgezeichnet ist, unterliegt keioem Zweifel. Indessen ware noch zu ermitteln, wie weit dies allgemein zutrifft, wie weit im besonderen Homologe oder Iso- mere mit einer verschiedenen Stellung von Alkylen sich. regelrniiI.3 ig durch ihre Absorptionsspektren unterscheiden. p p Noch wichtiger aber ist die Frage, wie weit i n dieser groSeren Empfindlichkeit der Methode ein unbedingter Vorzug zu erblicken ist. 1) SOC. 99, 1262 [1911]. Berfahte d. D. Chem. GesellaahafL Jshrg. XXXXV. 1) SOC. 99, 1262 [1911]. Berfahte d. D. Chem. GesellaahafL Jshrg. XXXXV. 64 1) SOC. 99, 1262 [1911]. Berfahte d. D. Chem. GesellaahafL Jshrg. XXXXV. Ergtinzung der spektrocbemischen Untersuchung zu bilden uad ge- rade dort helfend einzugreifen, wo diese versagt. Uber die Verteilung von Alkylen im Molekul und uber die gegenseitige Lage von Doppel- bindungen, soweit es sich nicht um den Gegensatz: konjugiert oder nicbt-konjugiert handelt, vermag die beutige Spektrochemie im allge- meinen nichts Sicheres auszusagen. Sollte ee sich herausstellen, dab die Absorptionsmethode infolge ibrer grol3eren konstitutiven Emptind- lichkeit hierzu imstande ist, so wijrde sie eine HuBerst erwuoschte Bereicberung unserer physikaliscb-c~emischen Hilfsrnittel zur Konsti- tutionserforschung darstelleo. Ob Aussicbt bierzu rorhnnden ist, dariiber lBBt sich nocb nichts sagen, denn dazu gehoren erst ausgedehnte systernatiscbe Unter- suchungen, die boffentlich von einem der auf dem Gebiete des U l t r a - v i o l e t t s tatigen Forscher in nicbt zu ferner Zeit ausgefubrt werden. So lange dies nicht gescbeben ist, bescbrlukt sich die Bedeutung der Absorptionsmetbode fur die Chemie der Terpene dararif, dal3 sie allem Anscbein nach in vielen Fjillen das empfindlichste Mittel ist, um die Rein h e i t einer Substanz festzustellen, wlhread sie uber deren K o n - s t i t u t i o n Blirilich b e s t i m m t e Ausliunft a i e die Refraktometrie n i c h t zu geben verrnng. 3. In der zweiten Arbeit beschlftigt sich Hr. H a n t z s c h rnit der Kritik, die icb an seiner absorptiometriwhen und refraktometriscben Untersuchung des Acetessigesters und dessen Derivate geiibt habel). Ich mu13 zu meinem Bedauern aul eine gro13ere Anzahl von Punkten dieser Gegeolrritik eingeben, da Hr. I l a n t z s c h eine Reihe von Be- hauptungen aufatellt, die einer Berichtigong bedurfrn. 3. In der zweiten Arbeit beschlftigt sich Hr. H a n t z s c h rnit der Kritik, die icb an seiner absorptiometriwhen und refraktometriscben Untersuchung des Acetessigesters und dessen Derivate geiibt habel). Ich mu13 zu meinem Bedauern aul eine gro13ere Anzahl von Punkten dieser Gegeolrritik eingeben, da Hr. I l a n t z s c h eine Reihe von Be- hauptungen aufatellt, die einer Berichtigong bedurfrn. Glcich im Anfang (8. 559) ifit Hro H a n t zsch ein Versehen - dns i m Laufe der Arbisit noclinirls (3. 563) wiedcrkehrt - untergelaufen, durch das der Ausgangspunkt unscrcr Kontrovcrse ginzlich rerschoben mird. flr. 127. K. Auwera: ffber die Absorptione- und die Refraktionsmethode. Mit Hilfe der weniger empfindlichen spektrochemischen Methode kann man bekanntlicb die Z a b l der D o p p e l b i n d u n g e n in beliebigen Ter- penderivaten ermitteln, vielfach e n d oc y cl isc h e und s e m i c y cli sc h e D o p p e l b i n d u n g e n unterscheiden und in der Regel aucb feststellen, ob in einem Molekiil k o n j u g i e r t e D o p p e l b i n d u n g e n irgendwelcher Art vorkommen oder nicht. Die wenigen Angaben, die bisher in dieser Beziehung iiber die Absorptionsmethode vorliegen , machen es wahr- scheiolich, daB sie die Erkennung jener Konstitutionseigentiim- lichkeiten bei weitem nicht i n gleicher Schiirfe, wenn iiberhaupt, er- mijglicht. Denn infolge des starken Einllusses sekundiiker Konstitu- tionslaktoren werden die Unterschiede zwischen den Absorptionsspek- tren der einzelnen Korpergruppen verwischt, und an ihre Stelle treten allmiihliche Ubergiinge. Tafel I1 der Hantzschschen Arbeit 1HBt dies besonders gut erkennen. Daraus folgt aber, da13 die wichtige Aufgabe, die der Relraktometrie nach dem Urteil aller Terpenforscher bei den K on s ti t u ti o n s bestimmun g e n auf diesem Gebiete zukommt, zum mindesten vorliiufig von der Absorptionsmethode nicht ubernommen, geschweige denn besser gelost werden kann. g g g A19 Beweis fuhre ich nochmals die von Hrn. H a n t z s c h bei seinen Ausfuhrungen leider nicht berucksichtigte Tatsache an, da13 Crymble, S t e w a r t , W r i g h t und Real) aus dem Absorptionsspektrum des ge- wijhnlichen T e r p i n e n s folgerten, daB ihm nicht die Formel I, son- dern 11: CH C>Q I. CH~(?~H, 11. - CH,C H, I. CH~(?~H, 11. - CH,C H, zukomme, wHbrend die chemische wie die in gleich unzweideutiger Weise die erste habeo. spektrochemische Forschung Formel als richtig erwiesen Allerdings ist es denkbar, daB die A b s o r p t i o n s m e t h o d e be- rufen sein kiinnte, auf dem Gebiete der Terpenchemie eine wertvolle 64 1) R. 44, 3523 [191 I]. 1) Im Original nicht gesperrt. 967 .BUS diesen Versuchen zieht Hantzsch den Schlnll, daB zwischen den Ergebnisssen, zu denen er durch Untersuchungen der Absorptionsspektren des Acetessigesters und dessen LBsungen gelangt war, und den Resultaten der refraktometrischen Methode kein Widerspruch besteht, sondern nur noch Dgeringe Dillerenzena verbleiben. Diese werden nder geringeren Genauigkeit der relraktometrischen Methode, zur Last gelegt . . . Jene Schlullfolge- rung') von Hantzsch mullte auf jeden, der den lnhalt der Arbeit genaner prfifte, verbltiffend wirken. Einige Zahlen zum Beweis! Aus diesen SBtzen, in Verbindung mit den dann folgenden Beispielen, geht m. E. klar und deutlich hervor, dall ich nicht Hrn. Hantzschs An- sichten fiber die geringere Genauigkeit der refraktometrischen Methode aver- blhlfenda gefunden habe, sondern seinen SchluB hiusichtlich der Bgeringeo Ditferenzena in den Ergebnissen der beiden Methoden. Dies tne ich aucb heute noch nnd erinnere daran, daB z. B. in Methglalkohol Brhhl fiir den Acetessigester die der Ketoform entsprechende Molretraktion gefunden hatte, wiihrend Hantzsch damals annahm, dall der Ester in diesem Medium zu 92 O/O a u 8 E n o 1 bestehen sollte. Wenn fibrigens Hr. Hantzsch (S. 563) darin, d d nach seiner damaligen Aullassnng der homogene Acetessigester wehr stark enolisierta sein sollte, wiihrend sich am den refraktometrischen Bestimmungen h6chstons 8-15 ''/o Enol berechnen, kein nentgegengesetztes Ergcbnisa erblickt, so will ich fiber diese Ansicht nicht streiten, wenn ich sie auch nicht teilen kann. 4. Die von mir als falsch bezeichneten Zahlen werden von Hrn. H a n t z s c h aufrecht erhalten, denn sie sollen, wie in ausfiihrlichen Darlegungen auseinandergesetzt wird, auf e i n e r r a t i o n e l l e r e n G r u n d l a g e b e r u h e n u n d t a t s i c h l i c h r i c h t i g e r sein a l s d i e von m i r berechneten Werte. Darnit sollen die von mir erhobenen Vorwtirfe auf mich zuruckfallen. 4. Die von mir als falsch bezeichneten Zahlen werden von Hrn. Hantzscli schreibt: BWesentlich ist nur, daB meioe naah Auwers wer- bliitfund wirkendena Pclilul3lolgerungen uber die geringere Geoauigkeit der Re- fraktionsmethode vollkoiiimen zii llccht beatcliena, und bemerkt gleiclizeitig tadolnd, daB ich cio bercita zweinial berichligtes Vcrsehen seincs Mitarbciters dazu benutzt habe, Bgcwissr, durch die Berichtigung erledigte, also gar nirht niuhr beateliende Difterrnzen zwisclicn den Resultaten der beiden optisclien Methotlen aiisliihrltch zu diikutieren u In kyirklichkeit mul3te icb jtme urspriioglichco Dil’kwnztm bespeclien, denn grade sie waren es, die Hrn. Hantxsch zu eincm Vcrglrich der Absorptions- und der Re- fraktionsmethotlc uud zu ciner abtlilligcn Kritik der lotzteren vevanlalten. Dieses uogiinstigo Urtcil fiber tlic spektiochemirche Methode murtle aber von Hm. Hantzsch bci jrncn Bericlitigungcn nicht zuriickgenommen. Ich fiihro Ferner a w mciner Albeit fulgeode Sitze (S. 35%) an: 1) R. 44, 3523 [191 I]. 968 Konstanten mit der wachsenden Genauigkeit der Beobachtungen USW. irn Lauf der Zeiten etwas geiindert, wie alle physikalischen oder che- mischen Konstanten dies tun. Man hat dementsprechend, als es sich herausstellte, daB die Molrefraktion nicht lediglich eine additive, sondern in rnindestens dem gleichen MaBe eine k o n s t i t u ti P e GroBe ist, Abw eichungen der gefundenen Molrefraktionen von den be- rechneten n i c h t auf eine A n d e r u n g d e s R e f r a k t i o n s w e r t e s fur CHs zuruckgefuhrt, sondern stets in irgendwelchen B e s o n d e r h e i t e n d e s Molekuls gesucht und meist auch gefunden, und so sind all- mablich die verscbiedenen spektrochemiscben Regeln entstanden. Hr. H a n t z s c h dreht die Sache gerade urn, indem er in seiner letzten Arbeit (S. 560) sagt: .\Venn man auf Unstimmigkeiten stbat, sind fir Berechnungen innerhalb derselben Kbrperklasse die ails eindeutigen, nkchstverwandten Stoffen sich ergebenden Zahlena - als Refraktioasiiquivalente fiir CHa - nzu verwendena. hlit dieser Annahme wechselnder Aquivalente fur CHa setzt sich Hr. H a n t z s c h in Widerspruch mit deru Grundprinzip der Spek- trochemie, und seine Rechnungen sind weder wach Briihlg. ) ausge- fuhrt, noch prinzipiell richtiger. Die Erklaruog fiir die acheinbaren und wirklichen Schwierig- keiten, die Hr. H a n tzsch ausfuhrlich bespricht, liegt darin, da13 Korper, die man ihren Forrneln nacb als homologe zu bezeichnen Re- wohnt ist, in ihrem chemiscben oder physikalisch-chemischen Verhalten oder in beiden nicht, selten ganz unerwartete und vorlaufig uncrklar- liche Verschieden h e i t e n auiweisen. Besonders biiufig findet man dies bei D i a l k y l d e r i v a t e n , und gerade die E s t e r d e r D i m e t h y l - und D i a t h y l - a c e t e s s i g s i u r e sind typische Beispiele dafur. I h . I) Ob die Deutung, die Hr. Hantzsch (S. 360 Anm.) dem Ansdruck mach Briihla gibt, matirlicha ist, d. h. dem allgemeinen Brauch entspricht, unterlasse ich zu beurteilen. 967 H a n t z s c h aufrecht erhalten, denn sie sollen, wie in ausfiihrlichen Darlegungen auseinandergesetzt wird, auf e i n e r r a t i o n e l l e r e n G r u n d l a g e b e r u h e n u n d t a t s i c h l i c h r i c h t i g e r sein a l s d i e von m i r berechneten Werte. Darnit sollen die von mir erhobenen Vorwtirfe auf mich zuruckfallen. In Wirklichkeit liegen die Verhaltnisse umgekehrt, denn Hr. H a n t z s c h setzt sich bei seinen Rerechnungen willkiirlich iiber all- gemein anerkannte Grundsiitze der Spektrochemie hinweg upd lHBt sicher festgestellte Tatsachen auBer acht. Die Grondlage fur die Ableitung von A t o m r e f r a k t i o n e n bildet bekanntlich die von L a n d o l t , Briihl und anderen Forschern an einem sehr umfangreichen Material festgestellte Regel , daB in b e l i e b i g en homologen Reihen die Molrefraktionen zweier auf ein- ander folgender Glieder sich um einen honstanten Betrag unter- scheiden, der somit das R e f r a k t i o n s f q u i v a l e n t d e r G r u p p e CH, darstellt. Dieser Grundwert ist von samtlichen Chemikern, die auf diesem Gebiet tiitig waren, bei allen weiteren spektrochernischen Forschungen und Rechnungen als eine ko n s t a n t e GroBe behandelt worden ; nur hat eich selbstverstandlich der absolute Wert dieser 64 968 968 H a n tz s c h s Meinung, daI3 seine Rechnungen richtiger seien, beruht aber auf dern Glauben, daB umgekehrt jene Substanzen Bechte Homo- logea ubd macbstverwandte Stoffea seien, und er erklart die gebriuch- lichen Refrnktionsaquivalente fur CHa in der Gruppe des Acetassig- esters fur unbrauchbar und unrichtig, weil sich bei ihrer Anwendung ergibt, daB der Dimethylester eine uber die Theorie urn etwa 0.3 erhiihte, der Diathylester dagegen eine hinter der Theorie urn etwa ebensoviel zuruckbleibende Molrefraktion besitzt. Das ist gewiB eine auffallende Tatsache, aber sie ist nicbt auf- fallender als die andere Tatsache, daB sich der D i UI e t h y 1 ester wesent- lich leichter verseifen 1aBt als der D i a t h y 1 ester, und sicher weniger auffallend als die weitere Tatsache, daI3 der Dimethylester durch wiiariges Ammoniak leicht und quantitativ in das zugehorige Amid verwandelt wird, wlhrend das D i a t h ylderivat bei der gleichen Be- handlung vollig unverhdert bleibt '). Uberdies hat Hr. H an tzsch selber einen weiteren interessanten Beweis dafur erbracht, da8 die beiden Ester sich auch physikalisch nicht wie gewohnliche Homologe verhalten. Er schreibt nBmlich: SDenn die oben von mir behandelte refraktometrische Abnormitit zwiachen Dimethyl- und Uiithyl-Ester zeigt sich auch in ihrer Absorption. Die Kurven der beiden Homologen sind nicht, wie in normalen Fillen, einander parallel und fast identisch, sondern verlaufen getrennt und kreuzen sich sogars Mit diesen klaren Worten bestatigt Hr. H a n t z s c h , daB die r e f r a k t o m e t r i s c h e Anomalie jener Verbindungen nichts mit der Art, wie man die Molrefraktionen berechnet, zu tun hat, sondern in d e m W e s e n dieser beiden Korper, die in ihren c h e m i s c h e n und phy sikalischen Eigenschaften so merkwiirdige Verschiedenheiten aufweisen, begrundet ist. Urn so weniger ist es zu verstehen, daB Hr. H a n t z s c h trotz der von ihm selber erkannten Abnormitiit im Ver- balten dieser Ester daran festhalt, daB gerade sie in diesem Falle zur A bleitung der Prichtigena und seinzig brauchbarena Refraktionswerte fur CHS geeignet seien. Es ist nach dem Gesagten iiberflussig, auf die von Hrn. ') Es sei bei dieser Gelegenheit bemerkt, daB auch bei anderen alky- lierten Acetessigestern spektrochemische Eigentiimlichkeiten beobachtet worden sind, doch ist dan bisher vofliegende Material z. T. noch nicht geniigend sicher. DaB auch in chemischer Beziehung diese K6rper Immer noch neue Problemc bieten, beweist die jiingste Veroffentlichung von Tafel (B. 46, 437 [1912]), in der eine ganz neiie Hypothese fiber die Konsti- 9 H. Meper, M. 28, 1 [1907]. - Vergl. such M. 27, 44 [1906]; B. 39, 198 [1906], sowie Analyse und Konstitutionsermittlung usw., 2. Auil., S. 500 [1909]. ') Es sei bei dieser Gelegenheit bemerkt, daB auch bei anderen alky- lierten Acetessigestern spektrochemische Eigentiimlichkeiten beobachtet worden sind, doch ist dan bisher vofliegende Material z. T. noch nicht geniigend sicher. DaB auch in chemischer Beziehung diese K6rper Immer noch neue Problemc bieten, beweist die jiingste Veroffentlichung von Tafel (B. 46, 437 [1912]), in der eine ganz neiie Hypothese fiber die Konsti- 9 H. Meper, M. 28, 1 [1907]. - Vergl. such M. 27, 44 [1906]; B. 39, 198 [1906], sowie Analyse und Konstitutionsermittlung usw., 2. Auil., S. 500 [1909]. 970 970 5. Im AnschluB an die Er6rterungen iiber die Berechnungsweise bemerkt Hr. H a n t z s c h , da13 auch mir Widerspriiche und Unrichtigkeiten bei meiner Kritik untcrgelaufen seien. In dem einzigen Beispiel, das Hr. H a n t z s c h dafiir anfiihrt, ist mir allerdings ein Versehen begegnet, denn ich habe am Schlusse des betreffenden Passus (S. 3527) statt ndes Diiithyl-acetessigesters. wler D ial k y 1 -acetessigesteru geschrieben. DaB ich allein den ersteren meinte, ergibt sich aus dem Zusammenhang, da ich nur von der Diiitliyl-Gruppe sprrchc! und nur den Wert, der als Mittel am deu Beobachtungen am Diiithyl - acetessigester abgeleitet ist, er6rtcre. Im iibrigen gebe ich gern zu, daB w zweckmflig gewesen wire, wenn ich daneben auch aut: das abweichende Verhalten des Dimethyl-Esters hingewiesen hattel). Aber dann hiitte ich mich aus- fiihrlicher iiber das Verhiltnis der beiden Ester (vergl. oben) verbreiten miissen, iind dies schien mir an jener Stelle nicht erforderlich, da ich im wesentlichen nur darauf hinweisen wollte, daB eine absolute Ubereinstimmung zwischen Versuch und Theorie bei spektrochemischen Beobachtungen nicht zu erwarten sei. Ubrigens habe ich beim Lesen jenes Abschnittes noch ein weiteres Ver- sehen von mir entdeckt, das hiermit bcrichtigt werde: auf Z. 17 von oben mull es statt ~ 9 x C E l a ~ n4xCHaa heil3en. 5. Im AnschluB an die Er6rterungen iiber die Berechnungsweise bemerkt Hr. H a n t z s c h , da13 auch mir Widerspriiche und Unrichtigkeiten bei meiner Kritik untcrgelaufen seien. In dem einzigen Beispiel, das Hr. H a n t z s c h dafiir anfiihrt, ist mir allerdings ein Versehen begegnet, denn ich habe am Schlusse des betreffenden Passus (S. 3527) statt ndes Diiithyl-acetessigesters. wler D ial k y 1 -acetessigesteru geschrieben. DaB ich allein den ersteren meinte, ergibt sich aus dem Zusammenhang, da ich nur von der Diiitliyl-Gruppe sprrchc! und nur den Wert, der als Mittel am deu Beobachtungen am Diiithyl - acetessigester abgeleitet ist, er6rtcre. Im iibrigen gebe ich gern zu, daB w zweckmflig gewesen wire, wenn ich daneben auch aut: das abweichende Verhalten des Dimethyl-Esters hingewiesen hattel). Aber dann hiitte ich mich aus- fiihrlicher iiber das Verhiltnis der beiden Ester (vergl. oben) verbreiten miissen, iind dies schien mir an jener Stelle nicht erforderlich, da ich im wesentlichen nur darauf hinweisen wollte, daB eine absolute Ubereinstimmung zwischen Versuch und Theorie bei spektrochemischen Beobachtungen nicht zu erwarten sei. 970 Ubrigens habe ich beim Lesen jenes Abschnittes noch ein weiteres Ver- sehen von mir entdeckt, das hiermit bcrichtigt werde: auf Z. 17 von oben mull es statt ~ 9 x C E l a ~ n4xCHaa heil3en. 6. In seiner friiheren Arbeitg berechnet Hr. H a n t z s c h die Mol-He- fraktion und - D i s p e r s i o n des O x y - c r o t o h s i u r e e s t e r s einmal aus den entsprechenden Gr613en des Athoxp-crotonsiureestcrs, ein zweites Mal aus den .ltomretraktioneo, weist auf (lie Verschiedenlieit der beiden Znhlenreihen hin iind fihrt dann fort: 6. In seiner friiheren Arbeitg berechnet Hr. H a n t z s c h die Mol-He- fraktion und - D i s p e r s i o n des O x y - c r o t o h s i u r e e s t e r s einmal aus den entsprechenden Gr613en des Athoxp-crotonsiureestcrs, ein zweites Mal aus den .ltomretraktioneo, weist auf (lie Verschiedenlieit der beiden Znhlenreihen hin iind fihrt dann fort: BFerner sei. auch anf eine andere"), vorliufig nicht erkliirliche Uil- stimmigkcits) aufmerksam gemachta. Daraus mull der Leser schkefien, daB auch jene erste Differenz von Hrn. Hantzsch als eine Unstimmigkeit mgesehen wird, zumal am SchluB des Abschnittes adie erwihntenUnstimmigkeiten~~ - andere werden nicht erwihnt - nauf die Inkonstanz der Refraktionsiquivalente im allgemeinena zuriickgcliihrt werden. Auf meinen Hinweis (S. 3526), da13 in dem angefiihrten Falle die ver- schiedenen Rerechnnngsweisen deswegen verschiedene Werte liefern, weil das eine Ma1 der exaltierende Einflull dcr Konjugation zum Ausdrnck komme, das sndere Ma1 aber vernachlisigt werde, erklart Hr. Iian tzsch jetrt (S. 563, tiition des .\cetessigesters iind seiner Alkylderivate aufgestellt wird. Rein formal wiirde die Annahme, dall z. 13. die dirlkplierten Acetessigester (fe- mische ron Keto- und Oxyd-Formen seien, eine Erklirung fiir manche - nicht f i r rlle - spektrochemische Beobachtungen an diesen KBrpern bieten. Indessen kann dieser Frage erst naher getreten werden, wenn Hr. T a f e l dic beabsichtigte genauerc Priilung seiner Hypothese ausgefiihrt hat, was ihm hoffentlich bald verg6nnt sein wird. I) Eine besondere Untersuchung iiber das spektrochemische Vcrhalten von D i a l k y l v e r b i n d u n g e n verschiedener Art habe ich schon vor lingerer Zeit hegonnen, doch ist sie wegen anderer hrbeiten noch nicht weit vorgeschritten. a) B. 43, 3065 [1910]. 3, Jm Original n i c h t gesperrt. 968 H a n t z s c h angestellten Berechnungen hier im Einzelnen einzugehen; nur sei auf den schroffen Widerspruch hingewiesen, der darin liegt, da8 Hr. H a n tzsch als Vorzug seiner Rechenweise hervorhebt, dab sie zu n o r m a l e n refraktometrischen Beziebungen awischen dem dimethyl- acetessigsauren Methyl und dem diathyl-acetessigsauren i t h y l fubren, wghrend er umgekehrt selber durch die Absorptionsmethode die tat- sachlich vorhandene refraktometrische A b n o r m i t a t dieser Korper bestatigt. Auch daB die Abweichungen der von Hrn. H a n t z s c h mit seinen willkiirlichen CHs-Werten berechneten Zahlen von den mit den normalen Werten erhaltenen nicbt v o n neuem die Unsicherheit des Fundaments der Refraktionsmethode bestatigena (S. 5G2), braucht kaum erwahnt zu werden'). 9 H. Meper, M. 28, 1 [1907]. - Vergl. such M. 27, 44 [1906]; B. 39, 198 [1906], sowie Analyse und Konstitutionsermittlung usw., 2. Auil., S. 500 [1909]. Anrn 1, daB er dies gewullt und als Kenner spektrochemischer GesetzmPOig- keiten jene Diiferenzen voransgesehen hahe. Dann ist aber nicht zu verstehen, warum Hr. H a n t z s c h die.sc Differenzen darnals als eine Unstimmigkeit hin- gestellt hat; jedenfalla mullte seine Darstellung auf Leser, die mit diesen spektrochemischen Feinheiten nicht vertraut sind, irreliihrend wirken und unbegriindete Zweifel an der Zoverlibsiukeit der refraktometrischen Methode erwecken. Bemerkt sei no&, dill Hr. H a n t z s c h in diesem Falle bei seinen Rech- nungen die Briihlschcn Rctraktions&qnivalente f i r CH1 benutzt hat; er rech- net also in derselben Arbeit mit verschicdenen Werten, ohno dies irgcndwir kenntlich zu machen. Endlich ist noch zu erwtihnen, daI3 die Berechnung der Mol-ReFraktion des Oxy-crotonsPureesters aus der des ~thOxyCrOtOnSlUreestors in der von Hrn. H a n t z s c h rngcwandten Form an und fiir bich unrichtig ist, weil beide K6rper nicht im Verhiltnis dcr Homologie zn einaoder stehen, und bei der gemihlten Berechnungsweise der Unterschicd der RefraktionsPquivalente von 0’ iind 0‘ nicbt beriicksichtigt wird 1). i. Auch die Betrachtungen, die Hr. H a n t z s c h iiber die Unsicherheit der Ableitung von Ref r a k ti one& q u i vale n ten aostellt (S. 562), treffen nicht ZU; namentlich werden hierbci nicht Stofte ohne jeden Grund ausgeschaltet, wenn auch die Autoren der Kiirze hnlber nicht immer die Griinde angegeben haben, die zur Ausschliellung dieser oder jener Bestimmungsreihen gefiihrt haben. In dem von Hrn. H a n t z s c h zitierten Beispiel bat Hr. E i s e n l o h r einige - nicht alle - A l l y l d e r i v a t e deswegen bei der Berechnung ausge- SChIOSSeii, weil begrhdeter Verdacht bestaod, da13 die betreffenden Priiparate nicht frei von Polymerisationsprodukten waren. DaB Allylderivate nicht etw4 wie Hr. H a n t z s c h zu folgern scheint, refraktometrische Anomalien anfweisen, geht aus Bestimmungen hervor, die Moosbrugger und ich kiirzlich’) ver- Bffentlicht haben, denn bei siimtlichen, von :ins untersuchten Allylverbindungen stimmten die gefundenen Mol-Rctraktionen und -Dispersionen mit den in iib- licher Weise berechneten sehr bcfriedigend iiberein. 8. Hr. H a n t z s c h schlieI3t weiter (S. ’) Ich habe dies erst beim erneuten Ifisen tler Hantzschschen Arlwit bemerkt. 970 I) Eine besondere Untersuchung iiber das spektrochemische Vcrhalten von D i a l k y l v e r b i n d u n g e n verschiedener Art habe ich schon vor lingerer Zeit hegonnen, doch ist sie wegen anderer hrbeiten noch nicht weit vorgeschritten. a) B. 43, 3065 [1910]. 3, Jm Original n i c h t gesperrt. 9 A. 387, 167 [1912]. ’) Ich habe dies erst beim erneuten Ifisen tler Hantzschschen Arlwit 9 A. 387, 167 [1912]. bemerkt. 3537 f.), dal Baly und Desch aus dem ultraviolet ten S p e k t r um d es .k t h y I-ace tessi g es t er s geschlossen hgtten, diese Substanz sei in Alkohol als reines Keton vorhanden, wghrentl Ban tzsc h aus dem gleichen Spektrum folgere, dal ein Keto-Enol-Gemisch vorliege. Hr. Hantzsch bemerkt dazu (S. 565), dal diese mangeblicha br- stehenden Differenzen in der Deutung der Resultate nder mangelnden Kennt- nis des Sachverhalts von Auwersa zur Last Fullen, denn Baly habe gar nicht aus dem Ultraviolettspektrum des Athyl-acetessigesters Sgeschlossena, dal3 dieser Ester in Alkohol als homogenes Keton gelbst sei, sondern er habe dies nwillkiirlich angenommena, aber nicht mit Recht. 9. Ich hatte darauf hingewiesen (S. 3537 f.), dal Baly und Desch aus dem ultraviolet ten S p e k t r um d es .k t h y I-ace tessi g es t er s geschlossen hgtten, diese Substanz sei in Alkohol als reines Keton vorhanden, wghrentl Ban tzsc h aus dem gleichen Spektrum folgere, dal ein Keto-Enol-Gemisch vorliege. Hr. Hantzsch bemerkt dazu (S. 565), dal diese mangeblicha br- stehenden Differenzen in der Deutung der Resultate nder mangelnden Kennt- nis des Sachverhalts von Auwersa zur Last Fullen, denn Baly habe gar nicht aus dem Ultraviolettspektrum des Athyl-acetessigesters Sgeschlossena, dal3 dieser Ester in Alkohol als homogenes Keton gelbst sei, sondern er habe dies nwillkiirlich angenommena, aber nicht mit Recht. Baly und Desch haben aus ihren spektrochemischen Untersuchungen iiber trutomere Verbindungen u. a. den SchluB gczogen, daB weder reine Keto-, noch reine Enol-Formeu einen Absorptionsstreifen im Ultraviolett baben, sondern nur Keto-Enol-Gemische. Sie fanden in dem Absorptionsspektruni jenes Esters >no trace of a baud. und folgerten demgemil: %These results pointed to the fact that the ethyl acetoacetate was entirely of the ketonic forma?. Warum dies nach Brn. Han tzsch kein SSchluBa, sondern cine wilkiirliche .4nnahmea sein soll, vermag ich nicht einzusehen; auch verstehe ich nicht, warnm die DiEferenzen in der Deutung der Resultate nnr Sangeblicha bestehen sollen 10. Fcrner wirft mir Hr. Hantzscli VOI' (S. 565), den Tatbestand be- treffeod des Natracetessigesters unrichtig dargestellt zu haben. Erstens habe ich erwshnt (S. 3538), daB zur Erklsrung der selektiven i\bsorption im Ultraviolett Baly nnd Desch, Hantzsch nnd endlich Ley drei ganz verschiedene Hypothesen aufgestellt Iiaben. An dieser Tatsrche wird dadurch nichts geirndert, dal Hr. Hantzsch erklgrt. Balys Hypothese von der Isorrhopesis definitiv widerlegt zu haben, nnd Leys Erklsrung als nicht. 563) ail3 der von Will- stiitter festgestellten Tatsacbe, daB da13 C g c l o o c t a t e t r a e n trotz seiner konjugierten Doppelbinduogen normales Brechungsvermogen be- sitzt, daB die von rnir betonte ~ZuverlHssigkeit und Genauigkeit, rnit der die Refraktionsrnethode auch die Anwesenheit von Konjugationen usw. anzeigta, tatsachlich gar nicht vorhanden sei. Hr. H a n t z s c h vergi13t dabei, rla13 ich in meinen Arbeiten immer von neuern betont habe, daB man die Spektrochernie zur Konstitutionsbestimruung nur dann verwenden knnn, *wenu alle Faktoren, die im Molekiil einer be- stirnmtcn Verbindung deren optisches Verbalten beeinflussen kiinnen, 972 972 bereits durch Untersuchung en d o g gebauter Substanzen ihrer Wir- kung nach hinreichend bekannt geworden sindq uod daB .die i;'ber- tragung von Regeln, die sich bei einer Verbindungsgruppe ergehen haben, auf eine Klasse ganz a n d e r s gebauter Substaozen ohne vor- hergehende experirnentelle Prufung nicht statthaft ist I). Und ebenso unzweideutig habe ich diesen Standpunkt am Schlusse (S. 3542) meiner von Hrn. H a n t z s c h kritisierten ,Qrbeit zum Ausdruck gebracht. Ob das Cyclooctatetraen, dieser KBrper von ganz neuem, ohne Analogie dastehendem Typus, refraktometrische Exaltatiooen aufweisen wiirde oder nicht, lieB sich ebensowenig im voraus beantworten, wie etwa die Frage, ob sich die Substanz in ibrern c h e m i s c h e n Verhalten den ungesattigten, hydroaromatischen Verbindungen oder den Benzolderi- vaten anschlieaen wurde. bereits durch Untersuchung en d o g gebauter Substanzen ihrer Wir- kung nach hinreichend bekannt geworden sindq uod daB .die i;'ber- tragung von Regeln, die sich bei einer Verbindungsgruppe ergehen haben, auf eine Klasse ganz a n d e r s gebauter Substaozen ohne vor- hergehende experirnentelle Prufung nicht statthaft ist I). Und ebenso unzweideutig habe ich diesen Standpunkt am Schlusse (S. 3542) meiner von Hrn. H a n t z s c h kritisierten ,Qrbeit zum Ausdruck gebracht. Ob das Cyclooctatetraen, dieser KBrper von ganz neuem, ohne Analogie dastehendem Typus, refraktometrische Exaltatiooen aufweisen wiirde oder nicht, lieB sich ebensowenig im voraus beantworten, wie etwa die Frage, ob sich die Substanz in ibrern c h e m i s c h e n Verhalten den ungesattigten, hydroaromatischen Verbindungen oder den Benzolderi- vaten anschlieaen wurde. 9. Ich hatte darauf hingewiesen (S. I) A u w e r e und Eisenlohr, J. pr. 121 84, 141 [1911]. a) SOC. 85, 1033 [1904). 1) In mainer friihereri Arbeit ist in der Formel ein Punkt znviel. BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer i-Form BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer cwi-Form cwi-Form cwi-Form cwi-Form 0 CHa . C,i'"-Me (H) .nachgewiesena sei, ist ein Irrtum von Hantz8ch.a Hr. Hantzsch erkllrt diese Behauptung beruhe auf einem Irrtum von mir. Das ist sehr fibemchend, denn die Zusammenfassung der Ergebnisse seiner absorptiometrischen Untersuchungen fiber den Acetessigester und desseri Derivate leitete Hr. Hantzsch mit folgendem Satz (S. 3054) ein: g ( XDurch diese Untersuchung sind also fiir den Acetessigester nnd seine Monosubstitutionsprodukte drei verschiedene Formen nachgewiesen worden, die durch zwci Gleichgewichte mit einnnder verbunden sind: Struk turisomer Va:enzisomer / - - - - - = - = - Ketoform enol-Form aci Form CHa . C : OH CH8.C.O. Mew) I --t t II 7+ I1 : . % CHa. CO CHs . COO Ca H5 HC COO Cs H5 HC COOCIH5 Dam wird, immer in gleich bestimmter und positiver Weise, er6rtert. unter welchen Bedingungen diese drei Formen existenzfnhig sind und am- dricklich gesagt, daS nnur noch die Frage offen bleibtu, ob die enol-Form vielleicht als Gleichgewicht der beiden stereoisomeren Modifikationen auftrete. g Das alles ist m. E. so klar, daB ich dem nichts hinzuzufiigen brauche. 11. Ich babe jene aci-Formen nicht als sAnalogaa gewisser hydro- aromatischer Verbindungen bezeichnet (S. 566), sondern gesagt (S. 3540), d 3 sip - aus leicht ersichtlichen Grknden - daran serinnerna: wiirden. 12. Auch habe ich nicht gesagt, da8 die Spekulationen, die Hrntzsch iiher die Natur des Acetessigesters angestellt hat, in der Luft schweben (S. 566), sondern diejenigen, die sich auf die Natur der Salze dieam Esters beziehen (S. 3540). DRS ist, namentlich im vorliegenden Fall, ein sehr weaentlicher Unterschied. 12. Auch habe ich nicht gesagt, da8 die Spekulationen, die Hrntzsch iiher die Natur des Acetessigesters angestellt hat, in der Luft schweben (S. 566), sondern diejenigen, die sich auf die Natur der Salze dieam Esters beziehen (S. 3540). DRS ist, namentlich im vorliegenden Fall, ein sehr weaentlicher Unterschied. 13. Hr. H a n t z s c h erklIirt (S. 566), sdal3 der konjugierte Acht- ring des C y c l o o c t a t e t r a e n s refraktometrisch vie1 indifferenter ist als der Sechsring des Benzolsa, und fiihrt fort: BHierdurch wird sogar wahrscheinlich, daB die refraktometrische Exaltation in Sechsringen ihr Maximum erreicht.r 13. Hr. H a n t z s c h erklIirt (S. geniigend bezeichnet. Zweitens sagte ich (S. 3530): BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer i-Form 566), sdal3 der konjugierte Acht- ring des C y c l o o c t a t e t r a e n s refraktometrisch vie1 indifferenter ist als der Sechsring des Benzolsa, und fiihrt fort: BHierdurch wird sogar wahrscheinlich, daB die refraktometrische Exaltation in Sechsringen ihr Maximum erreicht.r Wie Hr. H a n t z s c h zu diesen Beliauptuogen. kommt, ist vBllig unverstandlich. Vergleicht man niimlich die fur C y c l o o c t a t e t r a e n und Benzol gefundenen M o l r e f r a k t i o n e n (MD) rnit den Werten, die sich hierfiir mit den alten oder den neuen Atomrefraktionen b e r e c h n e n , so erhiilt man folgendes Bild: 974 Alt Neu Alt Neu Ber. fir CSH, 1% 35.24 35.08 Ber. €iir C6Hs iT 26.43 26.31 Gel. . . . . . 35.20 35.10 Gef. . . . . . 26.14') 26.14') ________ -0.04 +0.12 - 0.29 - 0.17 Bensol neist also uberhaupt keine Exaltation auF, sondern das Gegenteil ; die beiden Kohlenwasserstolfe verhalten sich also in refrak- tometrischer Hinsicht gerade umgekehrt, als es nach Hrn. H a n t z s c h der Fall sein soll. Im ubrigen ist es bekannt, daB auch die Hydrobenzolderivate riiit konjugierten endocyclischen Doppelbindungen im allgemeinen ver- haltnisniaI3ig niedrige Exaltationen besitzen, was auch gegen die Hsntzschsche Annahme spricht. 14. Der SchluBbemerkung von Hrn. H a n t z s c h (S. 566), d a b rnein Wunsch, die B e z e i c h n u n g BSpektrochemiea fur die Refrak- tionsmethode zu reservieren, sachlich nicht zu begrunden sei, stimme ich rollkommen zu, denn daB die Absorptionsmethode im w e i t e r e n S i n n e d e s W o r t e s auch eine spektrochemische Methode ist, fallt mir nicht ein bestreiten zu wollen. Lediglich xus h is t o ri s c h e n Griindeu erscheint es mir zweckmiiBig, an der e n g e r e n Begreuzung des Be- griffes Spektrochemie festzuhalten, da durch die Arheiten von B rii h t die Worte: Spektrocheniie, spelitrochemische~ Verhalten, spektro- cbsmische Metbode, spektrocbemische Konstitutionsbestimmuug usw. eine ganz bestimmte Bedeutung erhalten haben, und diese sich all- gemein eingeburgert hat. I) Mittel der Einzelwerte von Briihl (p6.13), Landolt und Jahn (Z(i.15) und Perkin sen. (26.13). - B. 87, 1066 [l894]; Ph. Ch. 10, 303 [1892]; SOC. 77, 273 [1900]. BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer i-Form zu diesem Zweck zusammen mit Eisenlo h r I ) an einigen Beispielen absichtlich gezeigt - sehr gegen den Rat und Wunsch von Briihl, der davon eine Schiidigung der Spektrochemie befurchtete - , daI3 bei unvollstiindiger Kenntnis der in Betracht kommenden Faktoreo irrtiimliche Schliisse aus refrakto- metrischen Bestimmungen gezogen werden kiinnen. hber wenn icb auf der einen Seite immer wieder betont habe, da8 in zweifelhalten Fiillen Vorsicht geboten sei, durfte ich auf der anderen Seite mit Recht darauf hinweisen , daI3 inoerhalb zshlreicher Kiirpergruppen, namentlich durch die Arbeiten von E i s e n l o h r und mir, so charak- teristische z a h l e n m a8 i g e B ezie h un g e n z w i s ch e n Kon s t i t u t ion und r e f r a k t o m e t r i s c h e n K o n s t a n t e n Sestgestellt worden sind, daB man sich hier auf vollig sicherem Boden befindet. So weit ist die Abaorptionsmethode noch nicht, denn die in Pulle vorbandenen interessanten Beobachtungen iiber das ultraviolette Spek- trum der verschiedenen Korperklassen haben ausschliefilich q uali- tativen Charakter, soweit es sich um die Abhangigkeit des Spektrums VOD der K o n s t i t u t i o n handelt. DaB weitere Forschungen v i e l l e i c h t in dieser Hinsicht Wandel schaffen werden und die Absorptions- methode zu einer wertvollen Ergilnzung der Refraktionsmetbode macheo kiinnen, bemerkte ich bereits im Eingang dieser Arbeit, doch bleibt die Erfiillung dieser Hoffnung einstweilen fraglich. Was endlich die Brauchbarkeit der beiden Methoden zur Losuog von T a u t o m e r i e - P r o b l e m e n , namentlich zur Ermittlung von Gleich- gewichtsverhaltoissen, betrifit, so habe ich bereits friiher selber erkliirt (S. 3541 f.), da8 die R e f r a k t i o n s m e t h o d e hie, gegeniiber anderen Methoden in der Regel, vor allem, wenn es sich um Liisungen handelt, erst iu z w e i t e r Linie in Frage kommen wird. I) B. 48, 827 [1910]. 3 Vgl. Hantzsch, B. 44, 1773 f. [1911]: .Die fiir die drei Schwingungs- eshlen hiernach berechneten Einzclwerte weichen allerdiogs meist erheblich von einander ab. Dime Abweichungen beruhen natiirlich auf Ungenauigkeibn BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer i-Form Eine nachtriigliche Erweiterung des Be- griffes kiinnte dsher leicht zu MiBverstandnisseu fuhren. Doch ist (lies, wie gesagt, lediglich eine ZweckmaBigkeitsErage. Auch stelle ich keineswegs in Abrede, dal3 die beiden Methoden t h eo retisc h zusammengehoren; o u r aus p r a k tisch eu Griindeu bitt ich dagegen, daB man beide .in einen TopF wirfte, 1. h. daB m m sie als gleichwertig ansieht und beispielsweise unterschiedslos von einer s p e k t r o c hemischen Konstitutionsbestimmung redet, wenn mau die eine oder die andere Methode angewendet hat. Denn daB die beiden Methoden, wie ich schrieh, in ihrem Wesen, ihrer experimen- tellen Hegriiodung und ihrer Leistungsfiihigkeit verschieden von ein- ander sind, lafit sich nicht ernstlich bestreiten. Bei meineo Untersuchungen uber die Verwertbarkeit der Spektro- chemie fiir die Zwecke der K o n s t i t u t i o n s e r m i t t l u n g bin ich von Anfang an bestrebt gewesen, die Grenzen der Methode offen iind klar 975 975 zu kennzeichnen, und habe u. a. zu diesem Zweck zusammen mit Eisenlo h r I ) an einigen Beispielen absichtlich gezeigt - sehr gegen den Rat und Wunsch von Briihl, der davon eine Schiidigung der Spektrochemie befurchtete - , daI3 bei unvollstiindiger Kenntnis der in Betracht kommenden Faktoreo irrtiimliche Schliisse aus refrakto- metrischen Bestimmungen gezogen werden kiinnen. hber wenn icb auf der einen Seite immer wieder betont habe, da8 in zweifelhalten Fiillen Vorsicht geboten sei, durfte ich auf der anderen Seite mit Recht darauf hinweisen , daI3 inoerhalb zshlreicher Kiirpergruppen, namentlich durch die Arbeiten von E i s e n l o h r und mir, so charak- teristische z a h l e n m a8 i g e B ezie h un g e n z w i s ch e n Kon s t i t u t ion und r e f r a k t o m e t r i s c h e n K o n s t a n t e n Sestgestellt worden sind, daB man sich hier auf vollig sicherem Boden befindet. zu kennzeichnen, und habe u. a. BDaB durch seine (Hm. Hantzschs) Untersuchungen die Existenz einer i-Form DaB die Anwendung der A b s o r p t i o n s m e t h o d e zur Ermittlung der quanti- tativen Zusammensetzung Ton Keto-Enol-Gemischen an sich ein interessantes und danken,swertes Unternehmen darstellt, verkenne ich keineswegs; ob und wie weit dieses Verfahren aber den anderen iiber- legen ist, bleibt noch festzustellen, denn dazu mu8 die Methode erst an einer gr613eren Anzahl von Beispielen geprfift werden, damit man den Bereich ibrer Anwendbarkeit und vor nllem den Grad ihrer Genauigkeit, der nach recht zweifelhaft erscheint ’), besser kennen 976 lernt. Man wird daher warten mussen, bis Hr. H a n t z s c h seine Untersuchungen auf weitere tautomere Verbindungen ausgedehnt hat, d e m erst an der Hand eines genfigend umfangreichen Materials wird sich ein sicheres Urteil iiber den Wert des Verfahrens an sich und im Vergleich zu den anderen Metboden gewinnen lassen. Im ubrigen miichte ich zum Schlu5 die Hoffnung aussprechen, daR Hr. H a n tzsc h auf Grund einer vorurteilalosen. eingehenden Prufung der spektrochemiscben Metbode und ihrer Leistungen zu dem gunstigeren Urteil zurfickkehren werde, das er friiher uber sie gehabt hat I), zumal an der Anderung seiner Meinung in letzter Linie nur ein unglucklicher Zufall Schuld tragt. Denn wenn Hrn. Ha n t z s c h bei der ersten Bearbeitung seiner absorptiometrischen Versuche uber den Acetessigester usw. nicbt ein wesentliches Verseben untergelaufen ware, er vielmehr sogleich die richtige Zusamrnensetzung der Keto- Enol-Gemische erkannt htiitte, wiirde Hr. H a n t z s c h schwerlich AnlaW genommen haben , die Brauchbarkeit und Genauigkeit der spektro- chemischen Methode in dern MaBe anzuzweifeln und zu bestreiten, wie er es ohne Grund leider jetzt tut. G r e i t s w ald. Chemivches Institut. (Eingegaugeu am 26. Mirn 1912.) Vor einiger Zeita) teilte ich mit, daR die Ester des o-Chlor- aceto-p - k r e s o l s bei der Behandlung mit Alkalicarbonaten Chlor- wnsaerstoff abspalten und durch eine eigentumliche Atomverschiebung in K etod e r i v at e des 4 -Me tb y 1-2- o x y-cu m a rc) n s iibergehen : y y ) g -- HGI = 0 0 co C. OH C. C O . R. HaC("'CHa. C1 ..,>\, C0.R A,\/ Die im Polgenden beschriebenen Versuche hatten den Zweck, die angenommene K o n s t i t u t i o n dieser Korper noch sicherer zu be- weisen und ihren c h e m i s c h e n C h a r a k t e r scharfer festzustellen. der optischen Methode. . . . Danach sind naturlicli auch die angeftihrten Mittelwerte nur Ann&herungswerte.a g ::Da er (der homogene Acetessigester) wegen der starken Absorption nur in sehr dtinnen Schichten bis zu 0.02 mm photographirt werden konnte, . . . ::Da er (der homogene Acetessigester) wegen der starken Absorption nur in sehr dtinnen Schichten bis zu 0.02 mm photographirt werden konnte, . . . so ist die optische Methode in diesem Falle besonders ungenau u n d irre- f h h r en d s p g p so ist die optische Methode in diesem Falle besonders ungenau u n d irre- f h h r en d s I) Vgl. B. 43, 105 [1910]. a) B. 43, 2192 [1910]. I) Vgl. B. 43, 105 [1910]. a) B. 43, 2192 [1910].
W4386560648.txt	http://jurnal.stisalhilalsigli.ac.id/index.php/jhei/article/download/94/79	en		Chief Executive Officer, Financial Leverage dan Perusahaan Manufaktur yang Terdaftar di Bursa Efek Indonesia	Jurnal Riset Hukum, Ekonomi Islam, Ekonomi, Manajemen dan Akuntansi	2,023	cc-by	3,890	Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 33 Chief Executive Officer, Financial Leverage dan Perusahaan Manufaktur yang terdaftar di Bursa Efek Indonesia Sri Winarsih Ramadana1, Mariana2* Corresponding author: marianamer02@gmail.com 1 2 Politeknik Kutaraja, Banda Aceh, Aceh, email: wirna.taryono@gmail.com STIS Al-Hilal Sigli, Aceh, email: marianamer02@gmail.com Received Date. 10 Desember 2022 ABSTRACT Revised Date. 16 Desember 2022 The purpose of this study is to determine the influence of the Accepted Date. 02 Januari 2023 The Keywords: Financial Leverage CEO Gender CEO Experience chief executive officer's gender and experience of the chief executive officer in influencing financial leverage on manufacturing companies listed on the Indonesia Stock Exchange in 2017-2021. The data analysis method in this study uses multiple linear regression to predict the influence of independent variables. The population in this research is a manufacturing company listed on the Indonesia Stock Exchange from 2017 to 2021. The sampling method in this study is the purposive sampling method. The results showed that the CEO gender affects financial leverage. This means that the gender of a CEO has a significant relationship with the use of debt. And the CEO's experience does not affect financial leverage. Because CEOs are more likely to use funds from debt and do not carefully consider their decision to use the debt. ABSTRAK Kata Kunci: Financial Leverage Gender CEO Pengalaman CEO Tujuan dari penelitian ini adalah untuk mengetahui pengaruh gender chief executive officer dan pengalaman chief executive officer berpengaruh terhadap financial leverage pada perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia tahun 2017-2021. Metode analisis data dalam penelitian ini menggunakan regresi linier berganda untuk memprediksi pengaruh variabel independent. Populasi dalam penelitain ini adalah perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia tahun 2017 sampai dengan 2021. Metode pengambilan sampel pada penelitian ini adalah metode purposive sampling. Hasil penelitian menunjukan bahwa gender CEO berpengaruh terhadap financial leverage. Artinya gender dari seorang CEO memiliki hubungan signifikan dengan penggunaan hutang. Dan pengalaman CEO tidak berpengaruh terhadap financial leverage. Hal ini dikarenakan CEO lebih cenderung menggunakan dana dari hutang dan tidak mempertimbangkan dengan matang keputusannya untuk menggunakan hutang tersebut. Pendahuluan Keputusan mengenai pendanaan suatu perusahaan berkaitan dengan proses pemilihan sumber dana yang akan digunakan untuk membiayai seluruh kegiatan operasional perusahaan. Pada dasarnya sumber dana tersebut dapat berasal dari dalam Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 34 perusahaan (sumber dana internal) maupun sumber dana dari luar perusahaan (sumber dana eksternal). Dana yang bersumber dari dalam atau internal perusahaan adalah dana yang dihasilkan sendiri oleh perusahaan (internal financing) seperti laba ditahan dan depresiasi, sedangkan dana yang bersumber dari luar atau eksternal adalah sumber dana yang berasal dari luar perusahaan (external financing) yang mana dana tersebut dapat diperoleh salah satunya dari hutang (Riyanto, 2011). Kebijakan perusahaan dalam menggunakan modal dari pinjaman berupa hutang ditinjau dari sudut pandang manajemen keuangan, merupakan suatu penerapan kebijakan financial leverage, dimana suatu perusahaan dalam membiayai kegiatan operasional dan investasinya mengunakan hutang serta menanggung beban tetap yang bertujuan untuk meningkatkan pendapatan bagi pemilik modal perusahaan. Hal ini dapat diartikan bahwa pada dasarnya financial leverage timbul karena adanya kewajiban keuangan yang sifatnya tetap (fixed financial charges) yang harus dikeluarkan oleh perusahaan. Kewajiban keuangan yang tetap ini tidak akan berubah meskipun dengan adanya perubahan tingkat keuntungan dan harus dibayar tanpa melihat sebesar apa pun tingkat keuntungan yang dicapai oleh perusahaan (Hamid et al, 2015). Implementasi kebijakan utang finansial meningkatkan nilai utang perusahaan, sedangkan penggunaan leverage memengaruhi tingkat risiko dan juga biaya modal perusahaan. Pada prinsipnya financial leverage dapat berdampak positif maupun negatif bagi perusahaan. Leverage keuangan menguntungkan ketika keuntungan yang dihasilkan lebih besar dari biaya tetap perusahaan, tetapi juga bisa berbahaya ketika keuntungan yang dihasilkan oleh perusahaan lebih kecil dari biaya tetap perusahaan. Artinya semakin besar leverage keuangan maka semakin besar pula risiko keuangan perusahaan (Ritonga et al, 2014). Ketika seorang pemimpin bisnis melakukan kesalahan dalam menentukan sumber pembiayaan yang akan digunakan, konsekuensinya sangat luas, terutama jika bisnis menggunakan terlalu banyak hutang dalam skala yang relatif besar. Selain itu, risiko yang harus ditanggung perusahaan semakin besar. Risiko potensial meliputi ketidakmampuan perusahaan untuk memenuhi semua kewajibannya dan kemungkinan tidak tercapainya laba yang diharapkan. Jika perusahaan tidak dapat membayar hutangnya, perusahaan akan menghadapi kesulitan keuangan dan bahkan mungkin bangkrut (Sawitri dan Lestari, 2015). Chief executive officer (CEO) diduga mempengaruhi financial leverage (Chen et al, 2014 dan Ting et al., 2016). Karakteristik pribadi CEO adalah faktor penting lain yang diyakini memengaruhi leverage keuangan. CEO memainkan peran penting dalam membuat keputusan terkait strategi dan alokasi sumber daya untuk mengelola perusahaan. Higherorder theory menjelaskan bahwa strategi yang dipilih oleh seorang pemimpin mencerminkan nilai-nilai kognitif dari pemimpin tersebut. Ini menentukan kemampuan CEO untuk menjelaskan situasi yang kompleks dan bagaimana situasi tersebut harus ditangani dengan tepat. Secara umum karakteristik personal chief executive officer terdiri dari tingkat overconfindence, umur, tingkat pendidikan, pengalaman, gender, jabatan, network, dan founder (Ting et al, 2015). Gender chief executive officer dan pengalaman chief executive officer merupakan beberapa karakteristik personal chief executive officer yang diduga dapat mempengaruhi pengambilan keputusan pendanaan perusahaan. Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 35 Gender chief executive officer adalah jenis kelamin dari chief executive officer yang terdiri dari pria atau wanita. Umumnya pria dinilai lebih bersifat risk taker dibandingkan dengan wanita yang lebih bersifat risk averse. Sehingga chief executive officer wanita akan cenderung memilih keputusan-keputusan dengan tingkat kemungkinan risiko yang lebih rendah dibandingkan dengan chief executive officer pria yang akan memilih keputusankeputusan dengan tingkat kemungkinan risiko lebih tinggi (Ting et al, 2015). Hasil Penelitian yang telah dilakukan oleh Chen et al (2014) menunjukkan bahwa gender chief executive officer berpengaruh terhadap financial leverage. Hal ini menggambarkan bahwa perusahaan yang dipimpin oleh chief executive officer pria relatif lebih agresif dan berani mengambil risiko dibandingkan dengan chief executive officer wanita. Chief executive officer pria akan lebih memilih pendanaan hutang yang besar untuk memuaskan permintaan atas peluang investasi karena pada dasarnya pria ingin mengambil lebih banyak tanggung jawab. Karakteristik dari personal chief executive officer selanjutnya adalah pengalaman chief executive officer. Pengalaman chief executive officer merupakan posisi jabatan yang pernah dijabat oleh chief executive officer diperusahaan lain sebelum bergabung dengan perusahaan yang diteliti saat ini (Ting et al, 2015). Pengalaman chief executive officer sangat berhubungan dengan pengambilan keputusan terkait penggunanan dana yang bersumber dari hutang. Pengalaman chief executive officer sebelumnya dapat meningkatkan pengetahuan pasar dan membantu chief executive officer dalam melihat peluang bisnis. Adanya pengalaman yang dimiliki oleh seorang chief executive officer, chief executive officer cenderung lebih berhati-hati dalam mengambil keputusan pendanaan yang terkait dengan hutang (Patzelt et al, 2008). Hasil Penelitian Ting et al (2015), menunjukkan bahwa pengalaman chief executive officer berpengaruh terhadap financial leverage. Hal ini menyiratkan bahwa semakin berpengalaman seorang chief executive officer maka chief executive officer akan lebih mempertimbangkan dengan seksama keputusannya dalam menggunakan hutang. Penelitian ini bertujuan untuk mengetahui pengaruh gender chief executive officer dan pengalaman chief executive officer berpengaruh terhadap financial leverage pada perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia. Kajian Literatur Financial Leverage Memutuskan sumber pendanaan merupakan keputusan keuangan yang sangat penting bagi perusahaan. Modal suatu perusahaan merupakan sumber pembiayaan bagi perusahaan untuk menjalankan segala aktivitasnya. Hal ini mengacu pada bagaimana perusahaan selalu berusaha untuk menjaga keseimbangan keuangannya dalam setiap operasinya atau operasi di dalam perusahaan. Leverage merupakan berapa besar modal sendiri yang dimiliki calon debitur untuk mendukung operasional perusahaan, untuk melihat sejauh mana kondisi leverage apakah masihberadapadatingkat yang wajar atau telah berada pada posisi yang mengkhawatirkan (Mariana et al., 2018; Mariana, 2022; Ramadana & Agustina, 2022). Leverage yang tinggi menunjukkan tingginya suatu usaha Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 36 yang didanai yang bersumber dari hutang. Semakin tinggi hutang yang ada menyebabkan semakin sulitnya suatu usaha untuk melunasi kewajiban dengan asumsi cash flows yang terbatas (Mariana et al., 2018). Karakteristik Personal Chief Executive Officer Chief executive officer (CEO) atau direktur utama memiliki peran yang sangat penting dalam pengambilan keputusan perusahaan. Chief executive officer bertanggung jawab dalam kelangsungan hidup dan keberhasilan perusahaan. Seorang chief executive officer tentunya mempunyai wewenang dan tanggung jawab penuh atas pengelolaan perusahaan demi kepentingan perusahaan. Peran chief executive officer dalam setiap perusahaan berbeda-beda, tergatung dari ukuran perusahaan. Dalam perusahaan skala kecil, chief executive officer memiliki banyak peran dalam perusahaan. Namun, dalam perusahaan yang berskala besar, chief executive officer lebih bertanggungjawab dengan strategi dan pengelolaan perusahaan secara keseluruhan (Ting, 2015). Chief executive officer adalah pihak yang dibayar paling tinggi dalam perusahaan dan paling banyak diekspos dibandingkan dengan eksekutif lain. Atmaja (2008) menjelaskan bahwa chief executive officer adalah seseorang yang dipercaya untuk memimpin jajaran direksi suatu perusahaan. Chief executive officer diangkat oleh dewan komisaris, dan umumnya mempunyai siklus jabatan. Bisa diangkat per lima tahun atau per sepuluh tahun tergantung kebijakan perusahaan. Pengaruh Gender Chief Executive Officer terhadap Financial Leverage Gender chief executive officer merupakan salah satu dari karakteristik personal chief executive officer yang diduga dapat mempengaruhi pengambilan keputusan pendanaan perusahaan. Difinisi gender tertuju pada jenis kelamin chief executive officer yang terdiri dari pria atau wanita. Chen et al (2014) menyatakan bahwa gender chief executive officer berpengaruh terhadap financial leverage. Hal ini memiliki makna bahwa perusahaan yang dipimpin oleh chief executive officer yang berjenis kelamin pria, keputusan yang terkait pendanaan lebih memilih pendanaan yang bersumber dari hutang. Sebalikknya, perusahaan yang dipimpin chief executive officer wanita semakin menghindari keputusan pendanaan yang bersumber dari hutang karena akan menimbulkan risiko kesulitan keuangan yang akan berujung pada kebangkrutan. Hasilnya konsisten dengan Faccio et al (2012) yang setuju bahwa perusahaan yang dijalankan oleh CEO wanita memiliki financial leverage paling rendah dari pada perusahaan yang dijalankan oleh chief executive officer pria. H1 : Gender chief executive officer berpengaruh terhadap financial leverage pada perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia. Pengaruh Pengalaman Chief Executive Officer terhadap Financial Leverage Karakteristik dari personal chief executive officer selanjutnya adalah pengalaman chief executive officer. Pengalaman chief executive officer merupakan posisi jabatan yang pernah dijabat oleh chief executive officer diperusahaan lain sebelum bergabung dengan perusahaan yang diteliti saat ini (Ting et al, 2015). Pengalaman chief executive officer 37 Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 sangat berhubungan dengan pengambilan keputasan terkait penggunanan dana yang bersumber dari hutang. Pengalaman yang dimiliki oleh chief executive officer sebelumnya dapat meningkatkatkan pengetahuan pasar dan membantu chief executive officer dalam melihat peluang. Dengan adanya pengalaman yang dimiliki oleh seorang chief executive officer, maka chief executive officer akan cenderung berhati-hati dalam mengambil keputusan pendanaan yang terkait dengan hutang (Patzelt et al, 2008). Hasil Penelitian dari Fischer and Pollock (2004), Patzelt et al (2008) dan Ting et al (2015), menunjukkan bahwa pengalaman chief executive officer berpengaruh terhadap financial leverage. Hal ini menyiratkan bahwa semakin berpengalaman seorang chief executive officer maka chief executive officer akan lebih mempertimbangkan dengan seksama keputusannya dalam menggunakan hutang. H2: Pengalaman chief executive officer berpengaruh terhadap financial leverage pada perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia. Metode Penelitian Populasi dan Sampel Populasi dalam penelitain ini adalah perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia tahun 2017 sampai dengan 2021. Metode pengambilan sampel pada penelitian ini adalah metode purposive sampling. Penentuan sampel ditentukan dengan kriteria sebagai berikut: 1. Perusahaan manufaktur yang terdaftar di BEI selama periode pengamatan tahun 20172021 dan yang melaporkan laporan tahunan secara lengkap. 2. Perusahaan yang data laporan keuangannya tersedia di www.idx.co.id selama periode pengamatan tahun 2017-2021. 3. Perusahaan yang memiliki laba selama periode pengamatan tahun 2017-2021. Operasional Variabel No Variabel 1 Financial Leverage 2 Tabel 1 Operasional Variabel Definisi Pengukuran financial leverage adalah ketika suatu perusahaan dalam membiayai kegiatan operasi dan investasinya mengunakan beban tetap (hutang) (Sudana, 2015) Skala Rasio DAR = Total Hutang Total Aset (Gomes et al, 2016) Karakteristik Personal CEO: 38 Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 No Variabel Definisi Pengukuran Skala 1.Gender CEO Gender CEO merpakan jenis kelamin CEO yang terdiri dari laki-laki atau perempuan. Gender adalah salah satu variabel yang menggambarkan karaktersitik personal CEO (Ting et al, 2015). Menggunakan varibel dummy, bernilai 1 jika CEO laki-laki dan 0 jika CEO perempuan (Ting et al, 2015). Dummy 2.Pengala man CEO Pengalaman chief executive officer merupakan posisi jabatan yang pernah dijabat oleh chief executive officer diperusahaan lain sebelum bergabung dengan perusahaan yang diteliti saat ini (Ting et al, 2015) Menggunakan varibel dummy, bernilai 1 jika CEO pernah dan 0 jika CEO tidak pernah menjabat sebagai esekutif dalam suatu perusahaan dengan tingkat misalnya, CEO, CFO, COO, atau CIO (Ting et al, 2015). Dummy Sumber: Data diolah (2022) Metode Analisis Data Metode analisis data dalam penelitian ini menggunakan regresi linier bergandauntuk mengetahui arah dari hubungan antara variabel independen dengan variabel dependen. Penarikan kesimpulan didasarkan pada nilai signifikansi sebesar 5%, baik untuk model maupun untuk pengaruh masing-masing variabel bebas (Mariana & Ibrahim, 2022). Persamaan regresi yang digunakan sebagai berikut: FLit = α0 + α1 GCit + α2 PCit + eit Keterangan: FLit α0 α1, α2 GCit PCit e = = = = = = Financial leverage perusahaan i pada periode t Konstanta (intercept) Koefisien regresi (i = 1,2) Gender CEO perusahaan i pada periode t Pengalaman CEO perusahaan i pada periode t Pengaruh variabel lain (epsilon) atau residual (error term) Sebelum dilakukan pengujian lebih lanjut terlebih dahulu dilakukan uji asumsi normal yaitu uji normalitas dan uji heteroskedasitas. Rancangan Pengujian Hipotesis Rancangan Pengujian Hipotesis Secara Bersama-sama Uji F dilakukan untuk melihat seberapa besar pengaruh semua variabel independen secara bersama-sama terhadap variabel dependen. Rancangan pengujian hipotesis adalah sebagai berikut: 39 Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 a. Jika nilai signifikan F (F Value) lebih besar dari 0,05 (α > 5%), maka dapat disimpulkan tidak ada pengaruh seluruh variabel independen terhadap variabel dependen, maka demikian H0 diterima; b. Sebaliknya jika nilai signifikan F (F Value) lebih kecil dari 0,05 (α < 5%), maka dapat disimpulkan bahwa variabel independen secara bersama-sama berpengaruh terhadap variabel dependen, artinya H0 ditolak atau Ha diterima. Untuk mengetahui seberapa besar variabel dependen dapat dijelaskan oleh variabel independen secara bersama-sama digunakan koefisiensi determinasi (R2). Nilai koefisensi determinasi berkisar antara 0 dan 1. Nilai R2 yang kecil berarti kemampuan variabel independen dalam menjelaskan variabel dependen sangat terbatas. Nilai yang mendekati 1 bermakna bahwa variabel-variabel independen akan memberikan hampir seluruh informasi yang dibutuhkan untuk memprediksi variabel dependen (Sani dan Maharani, 2013). Rancangan Pengujian Hipotesis Secara Parsial Menurut Sudjana (2005), hipotesis yang sudah dirumuskan kemudian harus diuji. Dalam pengambilan keputusan terhadap hasil uji t dapat dilihat pada tabel coefficients pada kolom sig (significance). Jika probabilitas nilai t atau signifikansi < 0,05, maka dapat dikatakan bahwa terdapat pengaruh secara parsial antara variabel bebas terhadap variabel terikat. Namun, jika probabilitas nilai t atau signifikansi > 0,05, maka dapat dikatakan bahwa tidak terdapat pengaruh secara parsial yang signifikan antara masing-masing variabel bebas terhadap variabel terikat. Hasil Penelitian dan Pembahasan Hasil Uji Normalitas Penelitian ini untuk menguji normalitas digunakan uji One Sample Kolmogorov Smirnov dengan tingkat signifikansi 5%. Hasil pengujian normalitas menunjukkan bahwa nilai residual terdistribusi secara normal dan analisis regresi layak digunakan, karena nilai residualnya (Asymp. Sig. (2-tailed)) > 0,050 yaitu sebesar 0,056. Tabel 2 One-Sample Kolmogorov-Smirnov Test N Normal Parametersa,b Most Extreme Differences Mean Std. Deviation Absolute Positive Negative Test Statistic Asymp. Sig. (2-tailed) Sumber: Data penelitian, diolah (2022). Unstandardized Residual 378 .0000000 .24320591 .069 .069 -.053 .069 .056c 40 Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 Hasil Uji Heteroskedastisitas Uji heteroskedastisitas bertujuan menguji apakah dalam model regresi terjadi ketidaksamaan variance dari residual satu pengamatan ke pengamatan yang lain (Ghozali, 2013). Hasil scatterplots terlihat bahwa titik menyebar secara acak serta tersebar baik di atas maupun di bawah angka 0 pada sumbu Y. Hal ini dapat disimpulkan bahwa tidak terjadi heterokadesitas pada model regresi sehingga model regresi layak digunakan untuk memperediksi variabel yang digunakan dalam penelitian ini. Jika hasil grafik scatterplots titik menyebar secara acak serta tersebar baik di atas maupun di bawah angka 0 pada sumbu Y. Hal ini dapat disimpulkan bahwa tidak terjadi heterokadesitas pada model regresi sehingga model regresi layak digunakan. Hasil uji heteroskedastisitas data ditunjukan dalam Gambar 1. Gambar 1. Scatterplots Hasil Pengujian Hipotesis Untuk menguji pengaruh variabel bebas karakteristik personal chief executive officer (yang terdiri dari gender CEO dan pengalaman CEO) terhadap financial leverage sebagai variable terikat dianalisis dengan menggunakan metode regresi linear berganda. Hasil regresi linear berganda disajikan dalam Tabel 3. Tabel 3. Hasil Regresi Linear Berganda Variable (Constant) GC PC Coefficient .286 .160 .046 F hitung = 5.954 Sig = 0.003 R = 0.175 R2 = 0.031 Sumber: Data Penelitian, diolah (2022) Std. Error .056 .048 .032 T 5.090 3.306 1.434 Sig .000 .001 .153 Adjusted R = 0.026 Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 41 Dari hasil yang tersaji pada Tabel 3 dapat dibangun sebuah persamaan regresi seperti berikut: FLit = 0,286+ 0,160 GCit + 0,046 PCit + eit Berdasarkan Tabel 3 Nilai koefisien determinasi sebesar 0,031 atau 3,1%. Hal ini bermakna bahwa karakteristik personal chief executive officer yang terdiri dari gender CEO dan pengalaman CEO hanya mampu menjelaskan financial leverage sebesar 3,1%. Kemampuan menjelaskan variabel independen ini kecil dikarenakan karena sisanya 96,9% dipengaruhi oleh variable yang tidak dijadikan sebagai indikator dalam penelitian. Hasil Pengujian Hipotesis Secara Secara Bersama-Sama Tabel 3 menunjukkan Nilai F value sebesar 5,954 dengan nilai signifikansi sebesar 0,003 bermakna bahwa variabel independen secara bersama-sama berpengaruh terhadap variabel dependen karena nilai signifikansi lebih kecil dari 5% atau 0,05. Hal ini menunjukan bahwa variabel karakteristik personal chief executive officer yang terdiri dari gender CEOdan karakteristik personal CEO pengalaman secara bersama-sama berpengaruh terhadap financial leverage pada perusahaan manufaktur yang terdaftar di Bursa Efek Indonesia. Hasil Pengujian Hipotesis Secara Parsial Pengaruh Gender Chief Executive Officer terhadap Financial Leverage pada perusahaan Manufaktur yang terdaftar di Bursa Efek Indonesia. Gender CEO merupakan salah satu dari karakteristik personal yang dimiliki oleh seorang CEO. Dimana nilai signifikansi untuk variabel gender CEO adalah sebesar 0,001 dengan nilai t sebesar 3,306. Nilai 0,01 tersebut lebih kecil dari α=5%. Nilai koefisien variabel gender CEO sebesar 0,160 menjelaskan bahwa jika gender CEO naik 1 (satu) persen, maka akan berakibat pada naiknya financial leverage sebesar 16 persen, dengan asumsi variabel lain konstan. Hasil tersebut menggambarkan bahwa gender CEO berpengaruh terhadap financial leverage. Artinya gender dari seorang CEO memiliki hubungan signifikan dengan penggunaan hutang. Hasil penelitian ini mendukung hipotesis yang dinyatakan sebelumnya dimana gender CEO berpengaruh terhadap financial leverage. Hal ini dikarenakan tingkat gender dalam kepemimpinan masih menjadi pertimbangan dari sebagian pemberi pinjaman. Penelitian ini didukung oleh penelitian yang dilakukan oleh Chen et al (2014) menyatakan bahwa gender CEO berpengaruh positif terhadap financial leverage. Hal ini memiliki makna bahwa perusahaan yang dipimpin oleh CEO yang berjenis kelamin pria, keputusan yang terkait pendanaan lebih memilih pendanaan yang bersumber dari hutang. Sebalikknya, perusahaan yang dipimpin CEO wanita semakin menghindari keputusan pendanaan yang bersumber dari hutang karena akan menimbulkan risiko kesulitan keuangan yang akan berujung pada kebangkrutan. Hal ini juga tidak sejalan dengan penelitian Faccio et al (2012) yang setuju bahwa perusahaan yang dijalankan oleh CEO wanita memiliki financial leverage paling rendah dari pada perusahaan yang dijalankan oleh CEO pria. Namun hasil penelitian ini berbanding terbalik dengan hasil penelitian dari Rahmadhani dan Ardhani (2017) yang memperoleh hasil dimana gender CEO tidak berpengaruh terhadap financial leverage. Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 42 Pengaruh Pengalaman Chief Executive Officer terhadap Financial Leverage pada Perusahaan Manufaktur yang Terdaftar di Bursa Efek Indonesia. Merujuk pada nilai signifikansi untuk variabel pengalaman CEO adalah sebesar 0,153 dengan nilai t sebesar 0,486. Nilai 0,153 tersebut lebih besar dari α=5% atau 0,153>0,05. Nilai koefisien variabel pengalaman CEO sebesar 0,046 menjelaskan bahwa jika pengalaman CEO naik 1 (satu) persen, maka akan berakibat pada naiknya financial leverage sebesar 4,6 persen, dengan asumsi variabel lain konstan. Berdasarkan hasil tersebut pengalaman CEO tidak berpengaruh terhadap financial leverage. Artinya pengalaman CEO tidak mempengaruhi keputusannya untuk melakukan leverage. Hal ini dikarenakan CEO lebih cenderung menggunakan dana dari hutang dan tidak mempertimbangkan dengan matang keputusannya untuk menggunakan hutang tersebut. Hasil penelitian ini menunjukan bahwa tidak mendukung hipotesis yang dinyatakan sebelumnya dimana pengalaman CEO berpengaruh terhadap financial leverage. Hasil tersebut berbanding terbalik dengan penelitian yang dilakukan oleh Fischer and Pollock (2004), Patzelt et al (2008), dan Ting et al (2015), menunjukkan bahwa karakteristik personal CEO yang ditinjau dari pengalaman CEO berpengaruh negatif terhadap financial leverage. Hal ini menyiratkan bahwa semakin berpengalaman seorang CEO maka CEO akan lebih mempertimbangkan dengan seksama keputusannya dalam menggunakan hutang. Kesimpulan 1. Gender CEO berpengaruh terhadap financial leverage. Artinya gender dari seorang CEO memiliki hubungan signifikan dengan penggunaan hutang. Hasil penelitian ini mendukung hipotesis yang dinyatakan sebelumnya dimana gender CEO berpengaruh terhadap financial leverage. Hal ini dikarenakan tingkat gender dalam kepemimpinan masih menjadi pertimbangan dari sebagian pemberi pinjaman. 2. Pengalaman CEO tidak berpengaruh terhadap financial leverage. Artinya pengalaman CEO tidak mempengaruhi keputusannya untuk melakukan leverage. Hal ini dikarenakan CEO lebih cenderung menggunakan dana dari hutang dan tidak mempertimbangkan dengan matang keputusannya untuk menggunakan hutang tersebut. Daftar Pustaka Atmaja, L. S. (2008). Manajemen keuangan. Edisi 3. Yogyakarta: Penerbit Andi. Chen, Y., Zhang, X., & Liu, Z. (2014). Manager characteristics and the choice of firm "low leverage": Evidence from China. American Journal of Industrial and Business Management, 4, 573-584. Faccio, M., Marchica, M. T., & Mura, R. (2012). Ceo gender, corporate risk-taking, and the efficiency of capital allocation. Journal of Corporate Finance, 39, 193–209. Fischer, H. M. & Pollock, T. G. (2004). Effects of social capital and power on surviving transformational change: The case of initial public offerings. Academy of Management Journal, 47(4), 463-81. Ghozali, H. I. (2013). Aplikasi Ananlisis Multivariate dengan Program IBM SPSS 21. Semarang:, Universitas Dipenogoro. Gomez, A. P., Castro, G. A., & Ortega, M. F. (2016). Determinants of leverage in Mining Companies, empirical evidence for Latin American Countries. Contaduria Y Jurnal HEI EMA, Vol. 2 No. 1, Tahun 2023 43 Administracion, 6(1), 26–40. Hamid, A., Kusrina, B. L., & Wardoyo. (2015). Pengaruh financial leverage terhadap return on equity terdaftar di Bursa Efek Indonesia. Jurnal Manajemen dan Bisnis, 19(1), 5363. Mariana, & Ibrahim, A. (2022). Determinan Cash Holding pada Perusahaan Manufaktur yang Terdaftar di Bursa Efek Indonesia Pendahuluan. HEI EMA: Jurnal Riset Hukum, Ekonomi Islam, Ekonomi, Manajemen Dan Akuntansi, 1(1), 1–13. http://jurnal.stisalhilalsigli.ac.id/index.php/jhei/article/view/28 Mariana, Nadiarsyah, & Abdullah, S. (2018). Informasi Akuntansi, Informasi NonAkuntansi, dan Keputusan Pemberian Kredit. Jurnal Reviu Akuntansi Dan Keuangan, 8(2), 177–186. https://doi.org/10.22219/jrak.v8i Mariana. (2022). Informasi Akuntansi dan Keputusan kredit. Bintang Semesta Media. Patzelt, H., Knyphausenaufseß, Z., & Nikol, P. (2008). Top management teams, business models, and performance of biotechnology ventures: An upper echelon perspective. British Journal of Management, 19(3), 205-21. Ramadana, S. W., & Agustina. (2022). Pengaruh Dividend Payment dan Leverage terhadap Cash Holding pada Perusahaan Building Construction Bursa Efek Indonesia. HEI EMA: Jurnal Riset Hukum, Ekonomi Islam, Ekonomi, Manajemen Dan Akuntansi, 1(2), 1–10.Rahmadhani, Z. I., & Ardhani, D. (2017). Semakin beragam semakin baik? isu keberagaman gender, keuangan, dan investasi perusahaan. Jurnal Keuangan dan Perbankan, 21(1), 1-13. Ritonga, M., Kertahadi, K., & Sri, M. R. (2014). Pengaruh financial leverage terhadap profitabilitas. Jurnal Administrasi Bisnis, 8(2), 1-10. Riyanto, B. (2011). Dasar-dasar pembelanjaan perusahaan. Edisi 4. Yogyakarta: BPFE. Sani, A., & Maharani, V. (2013). Metodologi penelitian manajemen sumber daya manusia (teori, kuesioner, dan analisis data). Malang: UIN Maliki Press. Sani, A., & Maharani, V. (2013). Metodologi penelitian manajemen sumber daya manusia (teori, kuesioner, dan analisis data). Malang: UIN Maliki Press. Sawitri, N. P. Y. R., & Lestari, P. V. (2015). Pengaruh risiko bisnis, ukuran perusahaan dan pertumbuhan penjualan terhadap struktur modal. E-Jurnal Manajemen Unud, 4(5), 1238-1251. Sudana, I. M. (2015). Manajemen keuangan perusahaan (Edisi 2). Jakarta: Erlangga. Ting, I. W. K., Azlinna, N., & Kweh, Q. L. (2015). Upper echelon theory revisited: The relationship between ceo personal characteristics and financial leverage decision. Procedia-Social and Behavioral Sciences, 195, 686-694.
https://openalex.org/W4366815612	https://www.medrxiv.org/content/medrxiv/early/2023/04/24/2023.04.17.23288603.full.pdf	English	null	Performance of Generative Pretrained Transformer on the National Medical Licensing Examination in Japan	medRxiv (Cold Spring Harbor Laboratory)	2,023	cc-by	14,509	. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Performance of Generative Pretrained Transformer on the National Medical 1 Licensing Examination in Japan 2 3 4 Yudai Tanaka1,2,3, Takuto Nakata1,2,3, Ko Aiga2, Takahide Etani1,4,5, Ryota Muramatsu1,3, Shun 5 Katagiri1, Hiroyuki Kawai1, Fumiya Higashino1, Masahiro Enomoto1, Masao Noda6, Mitsuhiro 6 Kometani2, Masayuki Takamura7, Takashi Yoneda2,8, Hiroaki Kakizaki9, Akihiro 7 Nomura2,7,8,10,11 8 9 10 1School of Medicine, Kanazawa University, Kanazawa, Japan 11 2Department of Health Promotion and Medicine of the Future, Kanazawa University Graduate 12 School of Medicine, Kanazawa, Japan 13 3Department of Molecular and Cellular Pathology, Kanazawa University Graduate School of 14 Medicine, Kanazawa, Japan 15 4Graduate School of Media and Governance, Keio University, Fujisawa, Japan 16 5Advanced Research Center for Human Sciences, Waseda University, Saitama, Japan 17 6Department of Otolaryngology and Head and Neck Surgery, Jichi Medical University, Tochigi, 18 Japan 19 7Department of Cardiovascular Medicine, Kanazawa University Graduate School of Medical 20 Sciences, Kanazawa, Japan 21 8College of Transdisciplinary Sciences for Innovation, Kanazawa University, Kanazawa Japan 22 9MICIN, Inc., Tokyo, Japan 23 10Frontier Institute for Tourism Science, Kanazawa University, Kanazawa, Japan 24 11CureApp Institute, Karuizawa, Japan 25 *Contributed equally 26 27 28 29 Corresponding author: 30 Akihiro Nomura, MD, PhD 31 Associate Professor 32 College of Transdisciplinary Sciences for Innovation, Kanazawa University 33 Kakuma-machi, Kanazawa, Ishikawa, 9201192, Japan 34 E-mail: anomura@med.kanazawa-u.ac.jp 35 ORCID ID: 0000-0001-6647-8240 36 Performance of Generative Pretrained Transformer on the National Medical 1 Licensing Examination in Japan 2 1 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 1 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Abstract 37 (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Abstract 37 The remarkable performance of ChatGPT, launched in November 2022, has significantly impacted 38 the field of natural language processing, inspiring the application of large language models as 39 supportive tools in clinical practice and research worldwide. Although ChatGPT recently scored 40 high on the United States Medical Licensing Examination, its performance on medical licensing 41 examinations of other nations, especially non-English speaking nations, has not been sufficiently 42 evaluated. This study assessed ChatGPT’s performance on the National Medical Licensing 43 Examination (NMLE) in Japan and compared it with the actual minimal passing rate for this exam. 44 In particular, the performances of both the GPT-3.5 and GPT-4 models were considered for the 45 comparative analysis. We initially used a model and prompt tuning set of 290 questions without 46 image data from the previous 116th NMLE (held in February 2022) to maximize the performance 47 for delivering correct answers and explanations of the questions. Thereafter, we tested the 48 performance of the best ChatGPT model (GPT-4) with tuned prompts on a dataset of 262 questions 49 without images from the latest 117th NMLE (held in February 2023). The best model with the 50 tuned prompts scored 82.7% for the essential questions and 77.2% for the basic and clinical 51 questions, both of which sufficed the minimum passing rates of 80.0% and 74.6%, respectively. 52 Simultaneously, we identified the three major factors contributing to the generation of the incorrect 53 answers—insufficient medical knowledge, information on Japan-specific medical system and 54 guidelines, and mathematical errors. In conclusion, GPT-4 powered ChatGPT with our optimally 55 tuned prompts achieved a minimum passing rate in the latest 117th NMLE in Japan. Although we 56 express strong concerns regarding the use of the current ChatGPT for medical purposes so far, 57 these artificial intelligence models may soon have the potential to serve as one of the best 58 “sidekicks” for solving medical and healthcare problems. 59 60 2 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Introduction 76 In recent decades, artificial intelligence (AI) algorithms have been widely applied in medical and 77 healthcare fields [1]. Currently, the AI algorithms available for clinical applications have been 78 developed using previous rule-based methods as well as recent machine learning (ML) methods 79 including its subfield of deep learning, promoted by the continually increasing availability of 80 computer resources and vast amount of medical data [2]. Consequently, these medical AI products 81 have been implemented to obtain targeted outputs such as the prediction of future disease risk, 82 classification as diagnostic support, or generation of various texts or images using natural 83 language processing (NLP) in medicine [1-3]. 84 NLP is an area of AI that addresses the interaction between human languages and machines 85 [4]. The major roles of NLP in medicine and healthcare include serving as supportive tools in 86 clinical practice and research [3]. Beyond the prediction of certain risk factors or clinical decision- 87 making, NLP assists physicians and researchers to efficiently extract, translate, classify and 88 analyze patients’ information and clinical-free text in electronic medical and health records, in 89 addition to dialogue generation and answering medical information [3, 4]. The performance of 90 NLP has dramatically improved following the emergence of transformer-based large language 91 models (LLMs). A transformer is a type of neural network model that employs self-attention 92 mechanism, relating multiple positions of a single sequence to compute a representation of the 93 sequence [5]. LLMs are created using advanced ML techniques, especially deep neural networks, 94 trained on enormous amounts of text data from the Internet and other sources [4]. A few notable 95 LLMs include pretrained Bidirectional Encoder Representations from Transformers (BERT) [6], 96 Language Models for Dialog Applications (LaMDA) [7], Pathway Language Model (PaLM) [8], 97 Large Language Model Meta (LLaMA) [9], and Generative Pretrained Transformer (GPT)-3 and 98 later models [10-12]. 99 Recently, InstructGPT (GPT-3.5)—a GPT model employing 175 billion parameters with 100 i d fi i d i f l i f h f db k [ ] d i di l In recent decades, artificial intelligence (AI) algorithms have been widely applied in medical and 77 healthcare fields [1]. Author summary (150 words) 64 ChatGPT's remarkable performance has inspired the use of large language models as supportive 65 tools in clinical practice and research. Although it scored well in the US Medical Licensing 66 Examination, its effectiveness in relevant examinations of non-English speaking countries remain 67 unexplored. This study assessed the performance of ChatGPT with GPT-3.5 and GPT-4 models in 68 Japan's National Medical Licensing Examination (NMLE). Initially, we used a tuning set of 290 69 questions from the 116th NMLE, and then the GPT-4 model with tuned prompts was tested on 262 70 questions from the 117th NMLE. The model scored 82.7% for essential and 77.2% for basic and 71 clinical questions, surpassing the minimum passing rates. Incorrect answers were attributed to 72 insufficient medical knowledge, Japan-specific medical system information, and mathematical 73 errors. In conclusion, GPT-4 powered ChatGPT achieved a minimum passing rate and might have 74 the potential for a valuable tool for fulfilling the needs of medical and healthcare fields. 75 3 3 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Introduction 76 Currently, the AI algorithms available for clinical applications have been 78 developed using previous rule-based methods as well as recent machine learning (ML) methods 79 including its subfield of deep learning, promoted by the continually increasing availability of 80 computer resources and vast amount of medical data [2]. Consequently, these medical AI products 81 have been implemented to obtain targeted outputs such as the prediction of future disease risk, 82 classification as diagnostic support, or generation of various texts or images using natural 83 language processing (NLP) in medicine [1-3]. 84 NLP is an area of AI that addresses the interaction between human languages and machines 85 [4]. The major roles of NLP in medicine and healthcare include serving as supportive tools in 86 clinical practice and research [3]. Beyond the prediction of certain risk factors or clinical decision- 87 making, NLP assists physicians and researchers to efficiently extract, translate, classify and 88 analyze patients’ information and clinical-free text in electronic medical and health records, in 89 addition to dialogue generation and answering medical information [3, 4]. The performance of 90 NLP has dramatically improved following the emergence of transformer-based large language 91 models (LLMs). A transformer is a type of neural network model that employs self-attention 92 mechanism, relating multiple positions of a single sequence to compute a representation of the 93 sequence [5]. LLMs are created using advanced ML techniques, especially deep neural networks, 94 trained on enormous amounts of text data from the Internet and other sources [4]. A few notable 95 LLMs include pretrained Bidirectional Encoder Representations from Transformers (BERT) [6], 96 Language Models for Dialog Applications (LaMDA) [7], Pathway Language Model (PaLM) [8], 97 Large Language Model Meta (LLaMA) [9], and Generative Pretrained Transformer (GPT)-3 and 98 later models [10-12]. 99 Recently, InstructGPT (GPT-3.5)—a GPT model employing 175 billion parameters with 100 supervised fine-tuning and reinforcement learning from human feedback [11]—and its dialogue- 101 optimized chatbot (ChatGPT) launched in November 2022 have significantly impacted NLP fields 102 [13]. By predicting the subsequent element of the texts, ChatGPT can comprehend user prompts 103 and generate human-like responses, expressed in ethical, sentimental, logical, and creative manner, 104 without any additional training (e.g., foundation model) [14]. Although GPT is a non-domain- 105 specific LLM, not exclusively intended to be used for medical or healthcare fields, recent 106 4 4 . Introduction 76 Thereafter, we removed questions with image data (n = 104) 131 and analyzed the remaining 290 questions without image data (Figure 1). 132 Using the ChatGPT API powered by GPT3.5, we initially tested its performance for the 133 original questions in Japanese language. Initially, we obtained a correct answer rate of 52.8% 134 (153/290) with an output error rate of 5.5% (16/290). Accordingly, we used updated prompts to 135 translate the original Japanese NMLE questions into English using ChatGPT before inputting them 136 publications have demonstrated that ChatGPT (GPT-3.5) possesses sufficient ability to pass the 107 United States Medical Licensing Examination [15, 16]. In contrast, another study reported 108 ChatGPT's inadequate performance on non-English-based Korean medical questions [17]. 109 Although the performance variation can be attributed to differences in languages, domestic 110 healthcare systems, diagnostic criteria, and treatment strategies, the relationship between these 111 differences and ChatGPT’s performance in answering medical questions remains unclear. 112 Furthermore, the performance of ChatGPT with the current GPT-4 model employing an estimated 113 10 trillion parameters [12] has not yet been evaluated on the latest Medical Licensing Examination, 114 which was originally written in non-English texts and held after the completion of GPT-4 model 115 training (August 2022) [18]. 116 Therefore, this study tested the performance of GPT (both GPT-3.5 and GPT-4 models) on 117 the 117th National Medical Licensing Examination (NMLE) (held in February 2023 in Japan), 118 which was originally conducted in the Japanese language. In particular, questions from the 119 previous year (116th NMLE exam held in February 2022) were used as a model and prompt 120 performance tuning set before using the latest questions (117th exam held in February 2023) as a 121 performance testing set to verify whether GPT can qualify for the actual minimal passing rate of 122 this examination. 123 Introduction 76 CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint publications have demonstrated that ChatGPT (GPT-3.5) possesses sufficient ability to pass the 107 United States Medical Licensing Examination [15, 16]. In contrast, another study reported 108 ChatGPT's inadequate performance on non-English-based Korean medical questions [17]. 109 Although the performance variation can be attributed to differences in languages, domestic 110 healthcare systems, diagnostic criteria, and treatment strategies, the relationship between these 111 differences and ChatGPT’s performance in answering medical questions remains unclear. 112 Furthermore, the performance of ChatGPT with the current GPT-4 model employing an estimated 113 10 trillion parameters [12] has not yet been evaluated on the latest Medical Licensing Examination, 114 which was originally written in non-English texts and held after the completion of GPT-4 model 115 training (August 2022) [18]. 116 Therefore, this study tested the performance of GPT (both GPT-3.5 and GPT-4 models) on 117 the 117th National Medical Licensing Examination (NMLE) (held in February 2023 in Japan), 118 which was originally conducted in the Japanese language. In particular, questions from the 119 previous year (116th NMLE exam held in February 2022) were used as a model and prompt 120 performance tuning set before using the latest questions (117th exam held in February 2023) as a 121 performance testing set to verify whether GPT can qualify for the actual minimal passing rate of 122 this examination. 123 124 125 Results 126 Improving performance through English translation and tuned prompts in 116th NMLE (2022) 127 Initially, we used the non-image-based questions from 116th NMLE in Japan to develop the optimal 128 input prompts for ChatGPT to maximize the correct answer rate. We extracted the question data 129 from the 116th NMLE containing 394 questions (originally 400 questions, but six were officially 130 removed from scoring evaluation). Results 126 Using the ChatGPT API powered by GPT3.5, we initially tested its performance for the 133 original questions in Japanese language. Initially, we obtained a correct answer rate of 52.8% 134 (153/290) with an output error rate of 5.5% (16/290). Accordingly, we used updated prompts to 135 translate the original Japanese NMLE questions into English using ChatGPT before inputting them 136 5 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint as questions. Although this marginally increased the correct answer rate to 56.2% (163/290), the 137 output errors increased to 14.8% (43/290; Figure 2). 138 To further improve the correct answer rate and reduce the errors, we tuned our prompts 139 for each question type (Basics of Medicine, Clinical Medicine, and Comprehension). In particular, 140 we provided sample outputs and directed the model to translate the questions into plain English 141 and create summaries before answering the questions (Figure 3). This tuned prompt improved the 142 correct answer rate to 63.1% (183/290) with a reduced output error rate of 7.6% (22/290). 143 Furthermore, we applied the above-tuned prompts to the GPT-4-based ChatGPT, which 144 demonstrated a correct answer rate of 82.8% (240/290) and a minimal error rate of 1.0% (3/290) 145 (Figure 2). 146 147 GPT-4-based ChatGPT performance on 117th (2023) NMLE with tuned prompt 148 Thereafter, we evaluated that the performance of the best model (GPT-4) with a tuned prompt for 149 the test set of 262 questions without image data from the 117th NMLE in Japan, held in February 150 4th and 5th, 2023, after the completion of GPT-4 model training in August 2022 (Figure 1). With a 151 tuned prompt, the best model achieved a correct answer rate of 78.6% (206/262) and an output 152 error rate of 0.8% (2/262) (Table 1). 153 The present results were compared with the actual minimal passing rate on the examination. Results 126 154 The current model with a tuned prompt scored 82.7% (129/156) for essential questions and 77.2% 155 (139/180) for basic and clinical questions, both of which qualified the minimum passing rates of 156 80.0% and 74.6%, respectively (Figure 2) [19]. Notably, we applied the GPT-4 model with tuned 157 prompts to the entire set of 395 questions (text-only) in the 117th NMLE, regardless of containing 158 image data (originally 400 questions, but five were officially removed from scoring evaluation). 159 This optimal model attained near-passing levels of 78.5% (157/200) for essential questions and 160 73.2% (216/295) for basic and clinical questions. 161 162 Exploratory analysis of incorrect ChatGPT responses and their associated explanations 163 To further enhance the performance of the model, we performed an exploratory analysis of 56 164 incorrect answers provided by the optimal GPT-4 model with tuned prompts for the 117th NMLE 165 questions. As listed in Table 2, the three primary factors contributing to the generation of incorrect 166 answers by the model included insufficient medical knowledge (33/56, 58.9%), Japan-specific 167 as questions. Although this marginally increased the correct answer rate to 56.2% (163/290), the 137 output errors increased to 14.8% (43/290; Figure 2). 138 (Figure 2). 146 147 GPT-4-based ChatGPT performance on 117th (2023) NMLE with tuned prompt 148 Thereafter, we evaluated that the performance of the best model (GPT-4) with a tuned prompt for 149 the test set of 262 questions without image data from the 117th NMLE in Japan, held in February 150 4th and 5th, 2023, after the completion of GPT-4 model training in August 2022 (Figure 1). With a 151 tuned prompt, the best model achieved a correct answer rate of 78.6% (206/262) and an output 152 error rate of 0.8% (2/262) (Table 1). 153 The present results were compared with the actual minimal passing rate on the examination. 154 Exploratory analysis of incorrect ChatGPT responses and their associated explanations 6 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint medical system information (17/56, 30.4%), and mathematical errors (4/56, 7.1%). Concerning the 168 insufficient medical knowledge, the areas of incorrect answers were not specific and spanned 169 across various medical fields. Notably, certain answers were outdated or critically incorrect in 170 current medical contexts (Figure 4). In terms of Japan-specific medical system, ChatGPT failed 171 to adequately answer questions related to Japanese medicolegal laws applicable in the medical and 172 healthcare field, guidance from the Ministry of Health, Labour, and Welfare (MHLW) in Japan, 173 and guidelines, especially those related to public health. Additionally, we noted several 174 mathematical errors such as in addition calculations (e.g., the explanation and addition formula 175 were correct, but the answer was wrong) and handling decimal points (because of translation errors 176 from the phrase “rounding to first decimal point” from Japanese). 177 178 179 Discussion 180 This study evaluated the performance of GPT on the Japanese Medical Licensing Examination. 181 The results indicate that 1) GPT-4 with a tuned prompt cleared the minimal passing rate on the 182 116th (2022) NMLE in Japan; 2) GPT-4 with tuned prompt qualified the minimum passing rate on 183 the latest 117th NMLE (2023); and 3) Inadequate medical knowledge, Japan-specific medical 184 system information, and mathematical errors were the primary factors associated with the incorrect 185 answers generated by the optimal model. Despite the absence of image data in the questions, this 186 study demonstrated the first attempt to use the best available ChatGPT model with tuned prompts 187 to achieve a minimum passing rate for the latest 117th NMLE in Japan. 188 This study provides several conclusions. First, GPT-4 with a tuned prompt cleared the 189 minimal passing rate on the 116th NMLE in Japan held in February 2022. Exploratory analysis of incorrect ChatGPT responses and their associated explanations Although GPT-3.5-based 190 ChatGPT achieved a correct answer rate of 52.8% for Japanese questions, it increased to 56.2% 191 after translating the questions into English. As GPT-3, the original GPT-3.5, was primarily trained 192 in English, it delivers a higher performance when responding to prompts in English compared to 193 other languages [10]. Similarly, a recent multilingual performance evaluation of GPT-4, an 194 improved version of GPT-3, confirmed that the best performance is more generally obtained with 195 English prompts [12]. After tuning our prompts to include a translation procedure into plain 196 English and modifying the output format based on the question type, the correct response rate 197 increased to 63 1% This finding is consistent with previous studies claiming that prompt 198 medical system information (17/56, 30.4%), and mathematical errors (4/56, 7.1%). Concerning the 168 insufficient medical knowledge, the areas of incorrect answers were not specific and spanned 169 across various medical fields. Notably, certain answers were outdated or critically incorrect in 170 current medical contexts (Figure 4). In terms of Japan-specific medical system, ChatGPT failed 171 to adequately answer questions related to Japanese medicolegal laws applicable in the medical and 172 healthcare field, guidance from the Ministry of Health, Labour, and Welfare (MHLW) in Japan, 173 and guidelines, especially those related to public health. Additionally, we noted several 174 mathematical errors such as in addition calculations (e.g., the explanation and addition formula 175 were correct, but the answer was wrong) and handling decimal points (because of translation errors 176 from the phrase “rounding to first decimal point” from Japanese). 177 medical system information (17/56, 30.4%), and mathematical errors (4/56, 7.1%). Concerning the 168 insufficient medical knowledge, the areas of incorrect answers were not specific and spanned 169 across various medical fields. Notably, certain answers were outdated or critically incorrect in 170 current medical contexts (Figure 4). In terms of Japan-specific medical system, ChatGPT failed 171 to adequately answer questions related to Japanese medicolegal laws applicable in the medical and 172 healthcare field, guidance from the Ministry of Health, Labour, and Welfare (MHLW) in Japan, 173 and guidelines, especially those related to public health. Exploratory analysis of incorrect ChatGPT responses and their associated explanations (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint engineering can improve model task performance [12, 20]. These improved correct response rates 199 can be attributed to English being the majority of the language in the training data, i.e., the Internet, 200 used by non-experts [21]. Although the error rate increased to 14.8% upon translating the Japanese 201 questions into English, it notably decreased to 7.6% after tuning the prompts by including the 202 format of the output. This result suggests that providing samples and standardizing the output 203 format can produce the desired output format and reduce the number of errors. Finally, upon 204 applying these optimized prompts to GPT-4-based ChatGPT, the correct response rate increased to 205 82.8% and the error rate plummeted to 1.0%. This significant improvement in performance can be 206 ascribed to the advanced architecture and training of GPT-4 [12]. 207 engineering can improve model task performance [12, 20]. These improved correct response rates 199 can be attributed to English being the majority of the language in the training data, i.e., the Internet, 200 used by non-experts [21]. Although the error rate increased to 14.8% upon translating the Japanese 201 questions into English, it notably decreased to 7.6% after tuning the prompts by including the 202 format of the output. This result suggests that providing samples and standardizing the output 203 format can produce the desired output format and reduce the number of errors. Finally, upon 204 applying these optimized prompts to GPT-4-based ChatGPT, the correct response rate increased to 205 82.8% and the error rate plummeted to 1.0%. This significant improvement in performance can be 206 ascribed to the advanced architecture and training of GPT-4 [12]. 207 Second, even in case of the latest 117th NMLE (2023), GPT-4 with tuned prompt qualified 208 the actual minimum passing rate. GPT-4 has passed various professional examinations in English, 209 including the practice bar exam with a score in the top 10% of examinees [12]. A previous study 210 reported that ChatGPT (GPT-3.5) failed to achieve the minimum passing rates [22]. However, this 211 study demonstrated that ChatGPT (GPT-4) can pass the 117th NMLE with the optimized prompt 212 tuning method proposed herein. Exploratory analysis of incorrect ChatGPT responses and their associated explanations Additionally, we noted several 174 mathematical errors such as in addition calculations (e.g., the explanation and addition formula 175 were correct, but the answer was wrong) and handling decimal points (because of translation errors 176 from the phrase “rounding to first decimal point” from Japanese). 177 178 This study evaluated the performance of GPT on the Japanese Medical Licensing Examination. 181 The results indicate that 1) GPT-4 with a tuned prompt cleared the minimal passing rate on the 182 116th (2022) NMLE in Japan; 2) GPT-4 with tuned prompt qualified the minimum passing rate on 183 the latest 117th NMLE (2023); and 3) Inadequate medical knowledge, Japan-specific medical 184 system information, and mathematical errors were the primary factors associated with the incorrect 185 answers generated by the optimal model. Despite the absence of image data in the questions, this 186 study demonstrated the first attempt to use the best available ChatGPT model with tuned prompts 187 to achieve a minimum passing rate for the latest 117th NMLE in Japan. 188 This study provides several conclusions. First, GPT-4 with a tuned prompt cleared the 189 minimal passing rate on the 116th NMLE in Japan held in February 2022. Although GPT-3.5-based 190 ChatGPT achieved a correct answer rate of 52.8% for Japanese questions, it increased to 56.2% 191 after translating the questions into English. As GPT-3, the original GPT-3.5, was primarily trained 192 in English, it delivers a higher performance when responding to prompts in English compared to 193 other languages [10]. Similarly, a recent multilingual performance evaluation of GPT-4, an 194 improved version of GPT-3, confirmed that the best performance is more generally obtained with 195 English prompts [12]. After tuning our prompts to include a translation procedure into plain 196 English and modifying the output format based on the question type, the correct response rate 197 increased to 63.1%. This finding is consistent with previous studies claiming that prompt 198 7 7 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Exploratory analysis of incorrect ChatGPT responses and their associated explanations As such, 233 calculation problems are reported as one of the areas where LLMs still exhibit relatively low 234 accuracy [24], indicating that calculation problems may be a relatively unsuitable field for current 235 ChatGPT. 236 systems considering the language differences. Furthermore, in certain cases related to 230 mathematical errors, the calculation formula in the explanation was correct, but the result and the 231 final answer output were incorrect. Moreover, an instruction of "approximating the decimal place" 232 was not properly comprehended by ChatGPT during the Japanese-to-English translation. As such, 233 calculation problems are reported as one of the areas where LLMs still exhibit relatively low 234 accuracy [24], indicating that calculation problems may be a relatively unsuitable field for current 235 ChatGPT. 236 As discussed, we express strong concerns regarding the use of the current ChatGPT for 237 medical purposes, as OpenAI has already indicated that the models should not be used for 238 providing triage, diagnosis, or treatment options for life-threating issues or severe medical 239 conditions [25]. Indeed, for use in medical settings, an approval must be obtained from regulatory 240 agencies, e.g., software as a medical device. Moreover, utilizing such technology is already 241 difficult with its several black-box aspects [12]. Various countries have released statements 242 regarding the applications of LLMs in medical fields [26, 27]. Although the versatility of these 243 models hinders the verification of their validity and they require enormous computational 244 resources and costs, we believe that the advanced medical foundation AI model [28] can replace 245 task-specific approach AI models and will appear not far off, with scientifically proven clinical 246 efficacy and safety in medical and healthcare fields. 247 The novelty of this study is that it is the first research to achieve a minimum passing rate 248 using 262 non-image questions in the latest 117th NMLE in Japan with the ChatGPT GPT-4 version 249 with the optimally tuned prompts. The limitations of this study were as follows. First, we only 250 used questions without image data to evaluate the performance of the best available model with 251 tuned prompts, although it might be fair to assess the ability of the model to pass the examination 252 using all questions, regardless of image data. Exploratory analysis of incorrect ChatGPT responses and their associated explanations The current results can be derived from the exquisite combination 213 of essential factors such as English translation and optimally tuned prompts for obtaining correct 214 answers through the best performance of the latest ChatGPT model. 215 Third, inadequate medical knowledge, information related to the medical and healthcare 216 system guidelines of Japan, and mathematical errors formed the three major factors of the incorrect 217 answers generated by the best available ChatGPT model with tuned prompts. Among the incorrect 218 answers associated under inadequate medical knowledge, no significant bias was observed for the 219 medical fields relevant to each question. Furthermore, even after providing incorrect answers, the 220 model output plausible but wrong medical explanations (so-called hallucinations in LLM outputs 221 [23]). Therefore, even if the model exhibits a performance level that surpasses the minimum score 222 for the NMLE, a broader range of specialized and up-to-date medical knowledge regarding 223 standard treatments should be inputted. In addition, output receivers should be equipped with 224 professional medical knowledge to assess the correctness of the output. For the Japan-specific 225 system, several incorrect answers were observed, especially in public health-related questions, 226 which are based on Japanese laws, guidelines, and unique systems. Although the GPT-4 powered 227 ChatGPT delivered improved performance in terms of output differences between the languages, 228 every country should perform their individual localization in terms of the applicable laws and 229 8 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint systems considering the language differences. Furthermore, in certain cases related to 230 mathematical errors, the calculation formula in the explanation was correct, but the result and the 231 final answer output were incorrect. Moreover, an instruction of "approximating the decimal place" 232 was not properly comprehended by ChatGPT during the Japanese-to-English translation. Materials and methods 271 Study overview 272 This study evaluated the performance of GPT models on the NMLE in Japan. We utilized both the 273 GPT-3.5 and GPT-4 models of ChatGPT (Open AI, Inc., San Francisco, CA, USA). Initially, the 274 questions from the 116th NMLE in Japan (February 2022) were used as a model and prompt tuning 275 set to optimize the performance of obtaining the correct answers and explanations. Subsequently, 276 we assessed the performance of the best ChatGPT model (GPT-4) with the tuned prompts for 277 answering the questions from the 117th NMLE in Japan (February 2023). 278 279 Input source 280 The questions and answers for the 116th NMLE in Japan were obtained from the official website 281 of the MHLW, Japan [29]. For the latest 117th NMLE, we manually performed optical character 282 recognition on the original question papers to create input data and extracted the official answers 283 from the MHLW website [19]. The examination comprised six blocks (A–F), with 75 questions in 284 blocks A, C, D, and E, and 50 questions in blocks B and F. Note that six questions in the 116th 285 NMLE and five in the 117th NMLE were excluded. In addition, all image-containing questions 286 were removed from both the prompt-tuning and the performance-testing datasets, because up till 287 early April 2023, only text-based questions could be used as input to the ChatGPT interface, 288 including the API. The number of image-containing questions was 104 in the 116th NMLE and 133 289 in the 117th NMLE. Thereafter, according to the Japanese NMLE scoring method, the remaining 290 questions without image data were classified into the categories of "Essential” and “Basic and 291 Clinical.” The 116th NMLE in Japan included 47 questions related to basics of medicine (essential), 292 24 questions of clinical medicine (essential), 14 questions on comprehension (essential), 65 293 questions regarding basics of medicine (general), 30 questions in basics of medicine (specifics), 294 31 questions of clinical medicine (general), 60 questions of clinical medicine (specifics), and 19 295 questions on comprehension. Exploratory analysis of incorrect ChatGPT responses and their associated explanations ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Exploratory analysis of incorrect ChatGPT responses and their associated explanations However, as revealed from the Results, we observed 253 a favorable model performance even upon using the entire question set in the 117th NMLE in Japan. 254 Second, the NMLE in Japan uniquely included strongly not-recommended "contraindication" 255 answer choices within the questions. The MHLW in Japan has set the minimum passing criteria 256 regarding selecting contraindication answer choices to be equal or less than three for the 116th 257 NMLE or two for the 117th NMLE. As the real number of contraindication answer choices were 258 not officially announced by the MHLW, we could not use them in the current performance 259 evaluation. 260 9 9 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint In conclusion, GPT-4 powered ChatGPT with optimally tuned prompts achieved a 261 minimum passing rate in the latest 117th NMLE in Japan. In addition, the model scored near- 262 passing levels for the entire test dataset of 395 questions, regardless of medical image data. The 263 upcoming GPT-4 version, which features enhanced image recognition capabilities, will easily 264 qualify the minimum passing rate and achieve top-tier scores, as reported in other English-based 265 examinations [12]. We again express strong concerns in terms of using of the current ChatGPT for 266 medical purposes so far. However, beyond its original design of answering examination questions 267 for humans, these AI models might have the potential be regarded as one of the best “sidekicks” 268 for solving problems and fulfilling the current needs in the medical and healthcare fields in the 269 near future. 270 10 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. Materials and methods 271 323 We further refined the prompts using the 116th NMLE questions to achieve higher rates of correct 324 answers and output in the desired format, because prompt tuning can improve the task accuracy 325 compared to training the entire model [12, 20]. The final optimized two-step prompts for the 326 English translation process and the process of answering the medical questions are illustrated in 327 Figure 3C, wherein each process comprised "system," "sample output," and "question input" 328 sections. We organized the output examples according to each medical question category (basics 329 of medicine, clinical medicine, and comprehension). In brief, ChatGPT was initially instructed to 330 translate the HTML-based Japanese questions into plain, direct, and improved English, while 331 NMLE in Japan for analyses. The entire set of 395 text-based questions, irrespective of image data, 302 from the 117th NMLE in Japan was considered for the exploratory analysis. 303 NMLE in Japan for analyses. The entire set of 395 text-based questions, irrespective of image data, 302 from the 117th NMLE in Japan was considered for the exploratory analysis. 303 Prompt engineering to maximize the correct answer rate 314 We used the 116th NMLE in Japan to generate the most suitable prompts for ChatGPT to answer 315 the 117th NMLE questions. Using the ChatGPT API, we first instructed ChatGPT to respond to the 316 original questions in Japanese language. We manually coded the Hyper Text Markup Language 317 (HTML) to represent the bold, italic, superscript, and subscript characters in the original text 318 (Figure 3A). Second, we instructed ChatGPT to translate the original Japanese NMLE questions 319 into English using its own capabilities before inputting them as questions (Figure 3B). In addition, 320 we compiled and analyzed the output errors. Thereafter, we provided prompts with restriction 321 sentences designed to prevent the reoccurrence of these errors, along with sample outputs 322 illustrating the desired output format. Finally, we inquired ChatGPT to improve the prompt itself. 323 We further refined the prompts using the 116th NMLE questions to achieve higher rates of correct 324 answers and output in the desired format, because prompt tuning can improve the task accuracy 325 compared to training the entire model [12, 20]. Materials and methods 271 CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. ch was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: iv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint NMLE in Japan for analyses. The entire set of 395 text-based questions, irrespective of image data, 302 from the 117th NMLE in Japan was considered for the exploratory analysis. 303 304 Generative Pretrained Transformer 305 The GPT, developed by OpenAI [14], is a type of AI model used for NLP tasks. Following the 306 research path from the original GPT, GPT-2, and GPT-3, OpenAI’s DL approach leverages 307 extensive amounts of data and intensive computation to create increasingly sophisticated and 308 capable language models [18]. ChatGPT has been fine-tuned from the initial GPT-3.5, and later, 309 GPT-4—a LLM trained in early 2022 to produce text [13, 30]. GPT-4 is OpenAI’s latest and most 310 advanced AI model that can solve difficult problems with greater accuracy [18]. In this study, we 311 used ChatGPT powered by both the GPT-3.5 and GPT-4 versions. 312 313 Prompt engineering to maximize the correct answer rate 314 We used the 116th NMLE in Japan to generate the most suitable prompts for ChatGPT to answer 315 the 117th NMLE questions. Using the ChatGPT API, we first instructed ChatGPT to respond to the 316 original questions in Japanese language. We manually coded the Hyper Text Markup Language 317 (HTML) to represent the bold, italic, superscript, and subscript characters in the original text 318 (Figure 3A). Second, we instructed ChatGPT to translate the original Japanese NMLE questions 319 into English using its own capabilities before inputting them as questions (Figure 3B). In addition, 320 we compiled and analyzed the output errors. Thereafter, we provided prompts with restriction 321 sentences designed to prevent the reoccurrence of these errors, along with sample outputs 322 illustrating the desired output format. Finally, we inquired ChatGPT to improve the prompt itself. Materials and methods 271 The 117th NMLE in Japan comprised 45 questions related to basics 296 of medicine (essential), 22 questions of clinical medicine (essential), 15 questions on 297 comprehension (essential), 61 questions from the basics of medicine (general), 27 questions on the 298 basics of medicine (specifics) 36 of clinical medicine (general) 46 questions related to clinical 299 p The questions and answers for the 116th NMLE in Japan were obtained from the official website 281 of the MHLW, Japan [29]. For the latest 117th NMLE, we manually performed optical character 282 recognition on the original question papers to create input data and extracted the official answers 283 from the MHLW website [19]. The examination comprised six blocks (A–F), with 75 questions in 284 blocks A, C, D, and E, and 50 questions in blocks B and F. Note that six questions in the 116th 285 NMLE and five in the 117th NMLE were excluded. In addition, all image-containing questions 286 were removed from both the prompt-tuning and the performance-testing datasets, because up till 287 early April 2023, only text-based questions could be used as input to the ChatGPT interface, 288 including the API. The number of image-containing questions was 104 in the 116th NMLE and 133 289 in the 117th NMLE. Thereafter, according to the Japanese NMLE scoring method, the remaining 290 questions without image data were classified into the categories of "Essential” and “Basic and 291 Clinical.” The 116th NMLE in Japan included 47 questions related to basics of medicine (essential), 292 24 questions of clinical medicine (essential), 14 questions on comprehension (essential), 65 293 questions regarding basics of medicine (general), 30 questions in basics of medicine (specifics), 294 31 questions of clinical medicine (general), 60 questions of clinical medicine (specifics), and 19 295 questions on comprehension. The 117th NMLE in Japan comprised 45 questions related to basics 296 of medicine (essential), 22 questions of clinical medicine (essential), 15 questions on 297 comprehension (essential), 61 questions from the basics of medicine (general), 27 questions on the 298 basics of medicine (specifics), 36 of clinical medicine (general), 46 questions related to clinical 299 medicine (specifics), and 10 questions regarding comprehension. Finally, we used 290 questions 300 (without image data) from the 116th NMLE and 262 questions (without image data) from the 117th 301 11 . Materials and methods 271 The final optimized two-step prompts for the 326 English translation process and the process of answering the medical questions are illustrated in 327 Figure 3C, wherein each process comprised "system," "sample output," and "question input" 328 sections. We organized the output examples according to each medical question category (basics 329 of medicine, clinical medicine, and comprehension). In brief, ChatGPT was initially instructed to 330 translate the HTML-based Japanese questions into plain, direct, and improved English, while 331 maintaining the original HTML codes without deleting or adding new text. In both processes, the 332 12 12 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint system of requirement and an exemplary output scenario were provided within the prompts. In the 333 question input section, the HTML-based Japanese questions were inputted for the English 334 translation process, and the English-translated questions were consequently inputted to the process 335 of answering the medical questions (Figure 3). To minimize output variability, all input prompts 336 were executed with the temperature parameter set to 0. 337 GPT-3.5-based analyses were performed using the ChatGPT API with custom Python code 338 on the Google Colaboratory interface. GPT-4-based analysis was conducted using ChatGPT 339 website console, with eight investigators (Y. T., T. N., K. A., T. E., R. M., S. K., H. K., and F. H.) 340 manually inputting prompts one by one and changing a thread each time. Specifically, they inputted 341 the questions, choices, and appropriate prompts into ChatGPT and summarized the output answers. 342 We used the GPT-3.5 version GPT3.5-turbo-0301 for the "Japanese,” “English,” and “English with 343 tuned prompt” analyses, and the GPT-4 model version released on March 14th 2023 for the 344 “English with tuned prompt” analysis. 345 346 Outcomes 347 The target outcome of this study is the correct answer rate. Materials and methods 271 CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. ch was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: iv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint analyzed the content of the incorrect answers along with their explanations to identify the areas in 364 which the application of the current ChatGPT for medicine may be relatively weak. 365 366 367 Acknowledgments 368 We express our gratitude to Yasuhiro Onogi and Yuichi Miyamae at MICIN, Inc. for their insightful 369 online discussions regarding this project. We thank Dr. Hozumi for dedicating his time to discuss 370 this topic with us. We also thank ChatGPT (GPT-4) and Enago English proofreading service for 371 English proofreading. 372 373 374 Data availability 375 The ChatGPT APIs used in this study are accessible via GitHub 376 (https://github.com/yudaitanaka1026/ChatGPT_NMLE_Japan). 377 378 379 Conflict of Interest 380 The authors declare no conflicts of interest relevant to this article. 381 382 383 Financial disclosure 384 None. 385 386 387 Author contributions 388 Conceptualization: Yudai Tanaka, Takuto Nakata, Ko Aiga, Hiroaki Kakizaki, and Akihiro 389 Nomura. 390 Data curation: Yudai Tanaka, Takuto Nakata, Ko Aiga, Takahide Etani, Ryota Muramatsu, Shun 391 Katagiri, Hiroyuki Kawai, Fumiya Higashino, and Masahiro Enomoto. 392 Formal analysis: Yudai Tanaka. 393 Methodology: Yudai Tanaka, Takuto Nakata, Hiroaki Kakizaki, and Akihiro Nomura. 394 analyzed the content of the incorrect answers along with their explanat 364 which the application of the current ChatGPT for medicine may be rel 365 366 367 Acknowledgments 368 We express our gratitude to Yasuhiro Onogi and Yuichi Miyamae at MIC 369 online discussions regarding this project. We thank Dr. Hozumi for de 370 this topic with us. We also thank ChatGPT (GPT-4) and Enago Engli 371 English proofreading. 372 373 374 Data availability 375 The ChatGPT APIs used in this study are accessible via GitHub 376 (https://github.com/yudaitanaka1026/ChatGPT_NMLE_Japan). 377 378 379 Conflict of Interest 380 The authors declare no conflicts of interest relevant to this article. Materials and methods 271 We manually compared ChatGPT’s 348 output answers with the official answers to determine the correctness of the output answers. 349 Accordingly, the correct answer rate was calculated as the number of correct answers divided by 350 the number of questions. We defined the output errors as incorrect answers. To evaluate the 351 potential performance for passing the 117th NMLE in Japan, we applied the minimum passing rates, 352 not the minimum passing scores, to evaluate the model performance because the image-containing 353 questions were excluded from the analyses. 354 355 Performance evaluation 356 In the primary performance evaluation, we assessed the correct answer rate for questions without 357 images in the 117th NMLE in Japan using the best ChatGPT model (GPT-4) with tuned prompt, 358 which was compared to the actual minimally passing rate on the examination. In the secondary 359 performance evaluation, we examined the correct answer rate for all questions in the 117th exam 360 using the optimal model and prompts. In addition, the medical reasonableness of the generated 361 explanations for each answer was assessed by two independent clinical physicians (M.N. and 362 ) d d bl h k d b h i d d li i l h i i (A ) h system of requirement and an exemplary output scenario were provided within the prompts. In the 333 question input section, the HTML-based Japanese questions were inputted for the English 334 translation process, and the English-translated questions were consequently inputted to the process 335 of answering the medical questions (Figure 3). To minimize output variability, all input prompts 336 were executed with the temperature parameter set to 0. 337 Performance evaluation 356 In the primary performance evaluation, we assessed the correct answer rate for questions without 357 images in the 117th NMLE in Japan using the best ChatGPT model (GPT-4) with tuned prompt, 358 which was compared to the actual minimally passing rate on the examination. In the secondary 359 performance evaluation, we examined the correct answer rate for all questions in the 117th exam 360 using the optimal model and prompts. In addition, the medical reasonableness of the generated 361 explanations for each answer was assessed by two independent clinical physicians (M.N. and 362 M.K.) and was double-checked by another independent clinical physician (A.N.). Furthermore, we 363 13 . thodology: Yudai Tanaka, Takuto Nakata, Hiroaki Kakizaki, and Akihiro Nomura. Materials and methods 271 (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Materials and methods 271 381 382 383 Financial disclosure 384 None. 385 386 387 Author contributions 388 Conceptualization: Yudai Tanaka, Takuto Nakata, Ko Aiga, Hiroaki K 389 Nomura. 390 Data curation: Yudai Tanaka, Takuto Nakata, Ko Aiga, Takahide Etan 391 Katagiri, Hiroyuki Kawai, Fumiya Higashino, and Masahiro Enomoto 392 Formal analysis: Yudai Tanaka. 393 Methodology: Yudai Tanaka, Takuto Nakata, Hiroaki Kakizaki, and A 394 analyzed the content of the incorrect answers along with their explanations to identify the areas in 364 which the application of the current ChatGPT for medicine may be relatively weak. 365 Data curation: Yudai Tanaka, Takuto Nakata, Ko Aiga, Takahide Etani, Ryota Muramatsu, Shun 391 Katagiri, Hiroyuki Kawai, Fumiya Higashino, and Masahiro Enomoto. 392 Data curation: Yudai Tanaka, Takuto Nakata, Ko Aiga, Takahide Etani, Ryota Muramatsu, Shun 391 Katagiri, Hiroyuki Kawai, Fumiya Higashino, and Masahiro Enomoto. 392 Methodology: Yudai Tanaka, Takuto Nakata, Hiroaki Kakizaki, and Akihiro Nomura. 394 14 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Project administration: Akihiro Nomura. 395 Supervision: Masayuki Takamura, Takashi Yoneda, and Hiroaki Kakizaki. 396 Validation: Masao Noda, Mitsuhiro Kometani, and Akihiro Nomura. 397 Visualization: Yudai Tanaka, Takuto Nakata, Ko Aiga, and Akihiro Nomura. 398 Writing – original draft: Yudai Tanaka, Takuto Nakata, Ko Aiga, and Akihiro Nomura. 399 Writing – review and editing: Yudai Tanaka, Takuto Nakata, Ko Aiga, Takahide Etani, Ryota 400 Muramatsu, Shun Katagiri, Hiroyuki Kawai, Fumiya Higashino, Masahiro Enomoto, Masao 401 Noda, Masayuki Takamura, Mitsuhiro Kometani, Takashi Yoneda, Hiroaki Kakizaki, and Akihiro 402 Nomura. 403 Supervision: Masayuki Takamura, Takashi Yoneda, and Hiroaki Kakizaki. 396 Validation: Masao Noda, Mitsuhiro Kometani, and Akihiro Nomura. 397 Visualization: Yudai Tanaka, Takuto Nakata, Ko Aiga, and Akihiro Nomur 398 15 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. References 404 References 404 1. Haug CJ, Drazen JM. Artificial Intelligence and Machine Learning in Clinical Medicine, 405 2023. N Engl J Med. 2023;388(13):1201-8. doi: 10.1056/NEJMra2302038. PubMed PMID: 406 36988595. 407 2. Nomura A, Noguchi M, Kometani M, Furukawa K, Yoneda T. Artificial Intelligence in 408 Current Diabetes Management and Prediction. Curr Diab Rep. 2021;21(12):61. Epub 20211213. 409 doi: 10.1007/s11892-021-01423-2. PubMed PMID: 34902070; PubMed Central PMCID: 410 PMCPMC8668843. 411 3. Aramaki E, Wakamiya S, Yada S, Nakamura Y. Natural Language Processing: from 412 Bedside to Everywhere. Yearb Med Inform. 2022;31(1):243-53. Epub 20220602. doi: 10.1055/s 413 0042-1742510. PubMed PMID: 35654422; PubMed Central PMCID: PMCPMC9719781. 414 4. Khurana D, Koli A, Khatter K, Singh S. Natural language processing: state of the art, 415 current trends and challenges. Multimed Tools Appl. 2023;82(3):3713-44. Epub 20220714. doi: 416 10.1007/s11042-022-13428-4. PubMed PMID: 35855771; PubMed Central PMCID: 417 PMCPMC9281254. 418 5. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all 419 you need. Advances in neural information processing systems. 2017;30. 420 6. Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional 421 Transformers for Language Understanding. Proceedings of the 2019 Conference of the North 422 American Chapter of the Association for Computational Linguistics: Human Language 423 Technologies. 2019;1:4171-86. doi: https://doi.org/10.18653/v1/N19-1423. 424 7. Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng H-T, et al. 425 LaMDA: Language Models for Dialog Applications2022 January 01, 2022:[arXiv:2201.08239 426 p.]. Available from: https://ui.adsabs.harvard.edu/abs/2022arXiv220108239T. 427 8. Chowdhery A, Narang S, Devlin J, Bosma M, Mishra G, Roberts A, et al. PaLM: Scaling 428 Language Modeling with Pathways2022 April 01, 2022:[arXiv:2204.02311 p.]. Available from: 429 https://ui.adsabs.harvard.edu/abs/2022arXiv220402311C. 430 9. Touvron H, Lavril T, Izacard G, Martinet X, Lachaux M-A, Lacroix T, et al. LLaMA: 431 Open and Efficient Foundation Language Models2023 February 01, 2023:[arXiv:2302.13971 p.] 432 A il bl f htt // i d b h d d / b /2023 Xi 230213971T 433 1. Haug CJ, Drazen JM. Artificial Intelligence and Machine Learning in Clinical Medicine, 405 2023. N Engl J Med. 2023;388(13):1201-8. doi: 10.1056/NEJMra2302038. PubMed PMID: 406 36988595. 407 2. Nomura A, Noguchi M, Kometani M, Furukawa K, Yoneda T. Artificial Intelligence in 408 Current Diabetes Management and Prediction. Curr Diab Rep. 2021;21(12):61. Epub 20211213. 409 doi: 10.1007/s11892-021-01423-2. PubMed PMID: 34902070; PubMed Central PMCID: 410 PMCPMC8668843. 411 3. Aramaki E, Wakamiya S, Yada S, Nakamura Y. References 404 Natural Language Processing: from 412 Bedside to Everywhere. Yearb Med Inform. 2022;31(1):243-53. Epub 20220602. doi: 10.1055/s- 413 0042-1742510. PubMed PMID: 35654422; PubMed Central PMCID: PMCPMC9719781. 414 4. Khurana D, Koli A, Khatter K, Singh S. Natural language processing: state of the art, 415 current trends and challenges. Multimed Tools Appl. 2023;82(3):3713-44. Epub 20220714. doi: 416 10.1007/s11042-022-13428-4. PubMed PMID: 35855771; PubMed Central PMCID: 417 PMCPMC9281254. 418 5. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all 419 you need. Advances in neural information processing systems. 2017;30. 420 6. Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional 421 Transformers for Language Understanding. Proceedings of the 2019 Conference of the North 422 American Chapter of the Association for Computational Linguistics: Huma 423 Technologies. 2019;1:4171-86. doi: https://doi.org/10.18653/v1/N19-1423. 424 7. Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng H-T, et al. 425 7. Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Chen 425 7. Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng H-T, et al. 425 LaMDA: Language Models for Dialog Applications2022 January 01, 2022:[arXiv:2201.08239 426 7. Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng H-T, et al. 425 LaMDA: Language Models for Dialog Applications2022 January 01, 2022:[arXiv:2201.08239 426 pp , , , , , g , LaMDA: Language Models for Dialog Applications2022 January 01, 2022:[arXiv:2201.08239 426 LaMDA: Language Models for Dialog Applications2022 January 01, 2022:[arXiv:2201.08239 426 p.]. Available from: https://ui.adsabs.harvard.edu/abs/2022arXiv220108239T. 427 p.]. Available from: https://ui.adsabs.harvard.edu/abs/2022arXiv220108239T. 7 8. Chowdhery A, Narang S, Devlin J, Bosma M, Mishra G, Roberts A, et al. PaLM: Scaling 428 Language Modeling with Pathways2022 April 01, 2022:[arXiv:2204.02311 p.]. Available from: 429 https://ui.adsabs.harvard.edu/abs/2022arXiv220402311C. 430 8. Chowdhery A, Narang S, Devlin J, Bosma M, Mishra G, Roberts A, et al. PaLM: Scaling 428 Language Modeling with Pathways2022 April 01, 2022:[arXiv:2204.02311 p.]. Available from: 429 https://ui.adsabs.harvard.edu/abs/2022arXiv220402311C. 430 Open and Efficient Foundation Language Models2023 February 01, 2023:[ 432 Available from: https://ui.adsabs.harvard.edu/abs/2023arXiv230213971T. 433 16 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. ch was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: iv preprint . References 404 CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint 10. Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, et al. Language models 434 are few-shot learners. Advances in neural information processing systems. 2020;33:1877-901. 435 11. Ouyang L, Wu J, Jiang X, Almeida D, Wainwright C, Mishkin P, et al. Training language 436 models to follow instructions with human feedback. Advances in Neural Information Processing 437 Systems. 2022;35:27730-44. 438 12. OpenAI. GPT-4 Technical Report2023 March 01, 2023:[arXiv:2303.08774 p.]. Available 439 from: https://ui.adsabs.harvard.edu/abs/2023arXiv230308774O. 440 12. OpenAI. GPT-4 Technical Report2023 March 01, 2023:[arXiv:2303.08774 p.]. Available 439 from: https://ui.adsabs.harvard.edu/abs/2023arXiv230308774O. 440 12. OpenAI. GPT-4 Technical Report2023 March 01, 2023:[arXiv:2303.08774 p.]. Available 439 from: https://ui.adsabs.harvard.edu/abs/2023arXiv230308774O. 440 13. OpenAI. Introducing ChatGPT 2022 [cited 2023 Apr 8]. Available from: 441 https://openai.com/blog/chatgpt. 442 13. OpenAI. Introducing ChatGPT 2022 [cited 2023 Apr 8]. Available from: 441 https://openai.com/blog/chatgpt. 442 14. Lee P, Bubeck S, Petro J. Benefits, Limits, and Risks of GPT-4 as an AI Chatbot for 443 Medicine. N Engl J Med. 2023;388(13):1233-9. doi: 10.1056/NEJMsr2214184. PubMed PMID: 444 36988602. 445 14. Lee P, Bubeck S, Petro J. Benefits, Limits, and Risks of GPT-4 as an AI Chatbot for 443 Medicine. N Engl J Med. 2023;388(13):1233-9. doi: 10.1056/NEJMsr2214184. PubMed PMID: 444 36988602. 445 15. Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepano C, et al. Performance 446 of ChatGPT on USMLE: Potential for AI-assisted medical education using large language 447 models. PLOS Digit Health. 2023;2(2):e0000198. Epub 20230209. doi: 448 15. Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepano C, et al. Performance 446 of ChatGPT on USMLE: Potential for AI-assisted medical education using large language 447 models. PLOS Digit Health. 2023;2(2):e0000198. Epub 20230209. doi: 448 10.1371/journal.pdig.0000198. PubMed PMID: 36812645; PubMed Central PMCID: 449 10.1371/journal.pdig.0000198. PubMed PMID: 36812645; PubMed Central PMCID: 449 PMCPMC9931230. 450 10.1371/journal.pdig.0000198. PubMed PMID: 36812645; PubMed Central PMCID: 449 PMCPMC9931230. 450 16. Gilson A, Safranek CW, Huang T, Socrates V, Chi L, Taylor RA, et al. How Does 451 ChatGPT Perform on the United States Medical Licensing Examination? The Implications of 452 Large Language Models for Medical Education and Knowledge Assessment. JMIR Med Educ. 453 2023;9:e45312. References 404 Epub 20230208. doi: 10.2196/45312. PubMed PMID: 36753318; PubMed 454 Central PMCID: PMCPMC9947764. 455 16. Gilson A, Safranek CW, Huang T, Socrates V, Chi L, Taylor RA, et al. How Does 451 ChatGPT Perform on the United States Medical Licensing Examination? The Implications of 452 Large Language Models for Medical Education and Knowledge Assessment. JMIR Med Educ. 453 2023;9:e45312. Epub 20230208. doi: 10.2196/45312. PubMed PMID: 36753318; PubMed 454 Central PMCID: PMCPMC9947764. 455 17. Huh S. Are ChatGPT's knowledge and interpretation ability comparable to those of 456 medical students in Korea for taking a parasitology examination?: a descriptive study. J Educ 457 Eval Health Prof. 2023;20:1. Epub 20230111. doi: 10.3352/jeehp.2023.20.1. PubMed PMID: 458 36627845; PubMed Central PMCID: PMCPMC9905868. 459 18. OpenAI. GPT-4 is OpenAI’s most advanced system, producing safer and more useful 460 reponses. 2023 [cited 2023 Apr 8]. Available from: https://openai.com/product/gpt-4. 461 18. OpenAI. GPT-4 is OpenAI’s most advanced system, producing safer and more useful 460 reponses. 2023 [cited 2023 Apr 8]. Available from: https://openai.com/product/gpt-4. 461 19. Ministry of Health Labour and Welfare in Japan. Announcement of the results of the 462 117th National Medical Licensing Examination in Japan. 2023 [cited 2023 Apr 10]. Available 463 from: https://www.mhlw.go.jp/general/sikaku/successlist/2023/siken01/about.html. 464 17 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint 20. Lester B, Al-Rfou R, Constant N. The power of scale for parameter-efficient prompt 465 tuning. arXiv preprint arXiv:210408691. 2021. 466 21. Web Technology Surveys. Usage statistics of content languages for websites. Available 467 from: https://w3techs.com/technologies/overview/content_language. 468 21. Web Technology Surveys. Usage statistics of content languages for websites. Available 467 from: https://w3techs.com/technologies/overview/content_language. 468 22. Kaneda Y, Tanimoto T, Ozaki A, Sato T, Takahashi K. Can ChatGPT Pass the 2023 469 Japanese National Medical Licensing Examination? Preprintsorg. 2023;2023030191. doi: 470 https://doi.org/10.20944/preprints202303.0191.v1. 471 22. Kaneda Y, Tanimoto T, Ozaki A, Sato T, Takahashi K. References 404 CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint xamples of prompts for English translation and answering medical questions. B: Simple “English prompts” used for Japanese-to-English translation and answering translated 514 questions. 515 B: Simple “English prompts” used for Japanese-to-English translation and answering translated 514 questions. 515 B: Simple “English prompts” used for Japanese-to-English translation and answering translated 514 questions. 515 Figure 1. Study overview. 496 g y Questions from the 116th NMLE in Japan were used as the prompt-tuning dataset and those from 497 117th NMLE were utilized as the performance-testing dataset after removing the image-based 498 questions. During the prompt tuning process, questions from the prompt-tuning dataset were 499 input into GPT-3.5-turbo and GPT-4, using simple prompts in both Japanese and English along 500 with tuned prompts in English. Subsequently, we evaluated the outputs from GPT-3.5-turbo and 501 GPT-4 with tuned prompts. After tuning the prompts, the ChatGPT (GPT-4) model optimized 502 with the tuned prompts was tested on the performance-testing dataset (117th NMLE). 503 504 References 404 Can ChatGPT Pass the 2023 469 Japanese National Medical Licensing Examination? Preprintsorg. 2023;2023030191. doi: 470 https://doi.org/10.20944/preprints202303.0191.v1. 471 23. Lee K, Firat O, Agarwal A, Fannjiang C, Sussillo D. Hallucinations in neural machine 472 translation. 2018. 473 24. Shakarian P, Koyyalamudi A, Ngu N, Mareedu L. An Independent Evaluation of 474 ChatGPT on Mathematical Word Problems (MWP)2023 February 01, 2023:[arXiv:2302.13814 475 24. Shakarian P, Koyyalamudi A, Ngu N, Mareedu L. An Independent Evaluation of 474 ChatGPT on Mathematical Word Problems (MWP)2023 February 01, 2023:[arXiv:2302.13814 475 p.]. Available from: https://ui.adsabs.harvard.edu/abs/2023arXiv230213814S. 476 p.]. Available from: https://ui.adsabs.harvard.edu/abs/2023arXiv230213814S. 476 25. OpenAI. Usage policies. 2023. Available from: https://openai.com/policies/usage- 477 policies. 478 26. Ordish J. Large Language Models and software as a medical device.: Medicines and 479 Healthcare products Regulatory Agency (MHRA); 2023. Available from: 480 Healthcare products Regulatory Agency (MHRA); 2023. Available from: 480 https://medregs.blog.gov.uk/2023/03/03/large-language-models-and-software-as-a-medical- 481 device/. 482 27. Ministry of Health Labour and Welfare in Japan. Software as a Medical Device 483 (Japanese). 2023. Available from: 484 https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000179749_00004.html. 485 28. Moor M, Banerjee O, Abad ZSH, Krumholz HM, Leskovec J, Topol EJ, et al. Foundation 486 models for generalist medical artificial intelligence. Nature. 2023;616(7956):259-65. Epub 487 20230412. doi: 10.1038/s41586-023-05881-4. PubMed PMID: 37045921. 488 29. Ministry of Health Labour and Welfare in Japan. Questions and answers of the 116th 489 National Medical Licensing Examination in Japan. 2022. Available from: 490 https://www.mhlw.go.jp/seisakunitsuite/bunya/kenkou_iryou/iryou/topics/tp220421-01.html. 491 30. OpenAI. What is ChatGPT? 2023. Available from: 492 https://help.openai.com/en/articles/6783457-what-is-chatgpt. 493 494 27. Ministry of Health Labour and Welfare in Japan. Software as a Medical Device 483 (Japanese). 2023. Available from: 484 27. Ministry of Health Labour and Welfare in Japan. Software as a Medical Device 483 (Japanese). 2023. Available from: 484 https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000179749_00004.html 485 https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000179749_00004.html. 485 28. Moor M, Banerjee O, Abad ZSH, Krumholz HM, Leskovec J, Topol EJ, et al. Foundation 486 models for generalist medical artificial intelligence. Nature. 2023;616(7956):259-65. Epub 487 20230412. doi: 10.1038/s41586-023-05881-4. PubMed PMID: 37045921. 488 29. Ministry of Health Labour and Welfare in Japan. Questions and answers of the 116th 489 National Medical Licensing Examination in Japan. 2022. Available from: 490 https://www.mhlw.go.jp/seisakunitsuite/bunya/kenkou_iryou/iryou/topics/tp220421-01.html. 491 30. OpenAI. What is ChatGPT? 2023. Available from: 492 https://www.mhlw.go.jp/seisakunitsuite/bunya/kenkou_iryou/iryou/topics/tp220421-01.html. 491 30 O AI Wh t i Ch tGPT? 2023 A il bl f 492 p g jp y _ y y p p 30. OpenAI. What is ChatGPT? 2023. Available from: 492 18 . Figure 2. Variations in the rate of correct answers across languages, pr 505 Translating the Japanese questions into English text improved the correct answer rate; however, 507 it increased the output error rate. Upon further tuning the prompts, the correct answer rate 508 improved and the output error decreased. Moreover, switching from the GPT-3.5 model to the 509 GPT-4 model enhanced the correct answer rate and almost eliminated errors. 510 it increased the output error rate. Upon further tuning the prompts, the correct answer rate 508 improved and the output error decreased. Moreover, switching from the GPT-3.5 model to the 509 GPT-4 model enhanced the correct answer rate and almost eliminated errors. 510 511 Figure 3. Examples of prompts for English translation and answering medical questions. 512 A: A simple “Japanese prompt” used for answering Japanese questions. 513 B: Simple “English prompts” used for Japanese-to-English translation and answering translated 514 questions. 515 C: Our optimized “English with tuned prompts”. 516 The final optimized two-step prompts comprised a "system," "sample output," and "question 517 input" sections. ChatGPT was initially instructed to translate HTML-based Japanese questions 518 into simple, direct, and improved English. In both processes, the system of requirement and an 519 exemplary output scenario were provided within the prompts. In the question input section, the 520 Japanese questions were inputted to the English translation process, and sequentially, the 521 English-translated questions were used to obtain the answers of the 117th NMLE questions. 522 523 Figure 4. Examples of potentially outdated or critically incorrect outputs from the model in 524 current medical contexts. 525 Figure legends 495 g g Figure 1. Study overview. 496 Questions from the 116th NMLE in Japan were used as the prompt-tuning dataset and those from 497 117th NMLE were utilized as the performance-testing dataset after removing the image-based 498 questions. During the prompt tuning process, questions from the prompt-tuning dataset were 499 input into GPT-3.5-turbo and GPT-4, using simple prompts in both Japanese and English along 500 with tuned prompts in English. Subsequently, we evaluated the outputs from GPT-3.5-turbo an 501 GPT-4 with tuned prompts. After tuning the prompts, the ChatGPT (GPT-4) model optimized 502 with the tuned prompts was tested on the performance-testing dataset (117th NMLE). 503 504 Figure 2. Variations in the rate of correct answers across languages, prompt tuning levels, 505 and GPT models. 506 Translating the Japanese questions into English text improved the correct answer rate; however 507 it increased the output error rate. Upon further tuning the prompts, the correct answer rate 508 improved and the output error decreased. Moreover, switching from the GPT-3.5 model to the 509 GPT-4 model enhanced the correct answer rate and almost eliminated errors. 510 511 Figure 3. Examples of prompts for English translation and answering medical questions. 512 A: A simple “Japanese prompt” used for answering Japanese questions. 513 B: Simple “English prompts” used for Japanese-to-English translation and answering translated 514 questions. 515 C: Our optimized “English with tuned prompts”. 516 The final optimized two-step prompts comprised a "system," "sample output," and "question 517 input" sections. ChatGPT was initially instructed to translate HTML-based Japanese questions 518 into simple, direct, and improved English. In both processes, the system of requirement and an 519 exemplary output scenario were provided within the prompts. In the question input section, the 520 Japanese questions were inputted to the English translation process, and sequentially, the 521 English-translated questions were used to obtain the answers of the 117th NMLE questions. 522 523 Figure 4. Examples of potentially outdated or critically incorrect outputs from the model i 524 current medical contexts. 525 Figure 4. Examples of potentially outdated or critically incorrect outputs from the model in 524 current medical contexts. 525 19 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint A: A question on the primary treatment for hyperventilation syndrome in the emergency 526 department. The suggestion of paper bag method for raising the carbon-dioxide concentration in 527 the blood has been commonly used in the past, but it is not always the first choice, as it can 528 worsen symptoms in certain patients with secondary hyperventilation, e.g., those with lung 529 diseases causing low blood oxygen levels. In such answers, it seems that outdated, traditional 530 information can prevail over the latest information, especially if it has been a standard practice 531 over a period and related information is widely available on the Internet. 532 B: A question on the initial outpatient treatment for a type-2 diabetes patient with poor control 533 and combined diabetic retinopathy and neuropathy. The long-term treatment goal for diabetes is 534 strict blood sugar control, but in this case, strict blood sugar control with sulfonylurea drugs 535 during the initial treatment may aggravate the risk of diabetic retinopathy, raising strong 536 concerns on ChatGPT's answer. 537 538 20 Tables Table 1. Performance of optimal GPT-4 model with tuned prompt for the 117th NMLE in Japan. Essential Basics and Clinical Basics of medicine (essential) Clinical medicine (essential) Comprehension (essential) Basics of medicine (general) Basics of medicine (specifics) Clinical medicine (general) Clinical medicine (specifics) Comprehension No. of questions without image data 45 22 15 61 27 36 46 10 No. of correct answers 36 19 12 47 25 22 37 8 No. of output errors 1 0 0 0 1 0 0 0 No. Figure 4. Examples of potentially outdated or critically incorrect outputs from the model in 524 current medical contexts. 525 of incorrect answers 8 3 3 14 1 14 9 2 Correct answer rate 80.0% 86.4% 80.0% 77.0% 92.6% 61.1% 80.4% 80.0% Output error rate 2.2% 0.0% 0.0% 0.0% 3.7% 0.0% 0.0% 0.0% Score weight x1 x3 x1 Total score (correct answer rate) 129/156 (82.7%) 139/180 (77.2%) Minimum passing rate 80.0% 74.6% nce of optimal GPT-4 model with tuned prompt for the 117th NMLE in Japan. 21 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Table 2. Summary of incorrect answers from the optimal model. Total incorrect answer N = 56 Insufficient medical knowledge 33 (58.9%) Breast surgery 1 Dermatology 2 Emergency medicine 2 Endocrinology 6 Gastroenterology 2 Immunology 1 Medical interview 1 Medical procedure 1 Nephrology 2 Neurology 1 Obstetrics and gynecology 2 Ophthalmology 1 Pediatrics 2 Physical examination 1 Psychiatry 1 Public health 1 Rehabilitation 1 Respiratory medicine 3 Rheumatology 1 Urology 1 Japan-specific medical system 17 (30.4%) Clinical research 1 Emergency 1 Psychiatry 1 Public health 14 Mathematical errors 4 (7.1%) Respiratory 1 Pediatrics 1 Cardiology 1 Medical interview 1 Others 2 (3.6%) Issue in English translation 1 Not providing an answer 1 Table 2. Summary of incorrect answers from the optimal model. Figure 3. A System Question input Question input B System Question input C "# system: Describe AI assistant's settings and so on You need an English translator, spell checker, and medical language expert who can translate your Japanese text into English. You want the translation to be improved and simplified to make it easier to understand for non-specialists at a high school level. Your request is to keep the meaning intact, but with a more literal translation. Your task is to provide multiple options for the answer, and use only HTML codes to provide the response. Restrictions: - The response is only one translation, containing only corrections and improvements to the Japanese text, not notes or anything else. - Do not remove any HTML code or add any new code to the response. - Your responses should only include translated English sentences with HTML codes. - You don't have to add the HTML code with options. - Set temperature = 0 =Sample= Japanese text: “Input a HTML-coded sample question in Japanese here” Sample Output: <p>What is commonly observed in patients with scleroderma kidney?</p> Aortic aneurysm\|sacral arthritis\|salmon pink rash\|nephrotic syndrome\|thrombotic microangiopathy [TMA]. # user: User's utterance Japanese sentences: “Input a HTML-coded original question in Japanese here” Tuned prompts for English translation "# system: Describe AI assistant's settings and so on You need an English translator, spell checker, and medical language expert who can translate your Japanese text into English. You want the translation to be improved and simplified to make it easier to understand for non-specialists at a high school level. Your request is to keep the meaning intact, but with a more literal translation. Your task is to provide multiple options for the answer, and use only HTML codes to provide the response. Restrictions: - The response is only one translation, containing only corrections and improvements to the Japanese text, not notes or anything else. - Do not remove any HTML code or add any new code to the response. - Your responses should only include translated English sentences with HTML codes. - You don't have to add the HTML code with options. - Set temperature = 0 =Sample= Japanese text: “Input a HTML-coded sample question in Japanese here” Sample Output: <p>What is commonly observed in patients with scleroderma kidney?</p> Aortic aneurysm\|sacral arthritis\|salmon pink rash\|nephrotic syndrome\|thrombotic microangiopathy [TMA]. Tuned prompts for answering medical questions System Sample output "# system: Describe AI assistant's settings and so on. As a doctor, provide a diagnosis, treatment, and prevention for any illness or disease based on a thorough examination of the patient's age, symptoms, and clinical course. Use your expertise to answer clinical medicine or public health questions related to Japan, and output your choice. Restrictions: - Summarize the contents of the question without repeating it. - Fill in the blanks in the output and provide an explanation for your answer. - For multiple-choice questions, choose the most probable option(s). - For questions without a specific number of selections, choose only one. Output: In summary, (summarise the question). The answer requires (number) of choices. The answer is (word) because (explain the rationale for your choice). arrow examination, and head MRI are not indicated for this patient's presentation. =SAMPLE= In summary, which two diseases exhibit low complementemia out of the options provided, which are Cellulitis, Bacterial pneumonia, Ischemic colitis, Acute glomerulonephritis, and Mixed cryoglobulinemia? The answer requires selecting (two) choices. The answer is (Acute glomerulonephritis) and (Mixed cryoglobulinemia) because these two diseases are known to be associated with low complementemia. Acute glomerulonephritis is an immune-mediated kidney disease that can lead to low levels of complement proteins. Mixed cryoglobulinemia is a type of vasculitis that is characterized by the presence of abnormal proteins in the blood, which can lead to low complement levels. Cellulitis, bacterial pneumonia, and ischemic colitis are not typically associated with low complementemia. # user: User's utterance Question: <p>Which physical examination is difficult to perform on a patient with impaired consciousness?</p> Tendon reflex\|neck stiffness\|light reflex\|finger-to-nose test\|knee raise test (lower limb dropping test)" Question input Tuned prompts for answering medical questions System Sample outpu "# system: Describe AI assistant's settings and so on. As a doctor, provide a diagnosis, treatment, and prevention for any illness or disease based on a thorough examination of the patient's age, symptoms, and clinical course. Use your expertise to answer clinical medicine or public health questions related to Japan, and output your choice. Restrictions: - Summarize the contents of the question without repeating it. - Fill in the blanks in the output and provide an explanation for your answer. - For multiple-choice questions, choose the most probable option(s). - For questions without a specific number of selections, choose only one. Output: In summary, (summarise the question). The answer requires (number) of choices. Tuned prompts for English translation “Input a HTML-coded original question in Japanese here” Table 2. Summary of incorrect answers from the optimal model. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint “Input a HTML-coded original question in Japanese here” Simple prompts for answering medical questions Question input System System Answer the following questions with reasons. Which of the following physical examinations is difficult to perform on a patient with impaired consciousness? Tendon reflex \| Nuchal rigidity \| Pupillary light reflex \| Finger-to- nose test \| Knee extension test (lower limb dropping test) ） Simple prompts for answering medical questions Question input Simple prompts for English translation System Translate the following Japanese into English. “Input a HTML-coded original question in Japanese here” Question input A B Figure 3. Table 2. Summary of incorrect answers from the optimal model. ummary of incorrect answers from the optimal model. Total incorrect answer N = 56 Insufficient medical knowledge 33 (58.9%) Breast surgery 1 Dermatology 2 Emergency medicine 2 Endocrinology 6 Gastroenterology 2 Immunology 1 Medical interview 1 Medical procedure 1 Nephrology 2 Neurology 1 Obstetrics and gynecology 2 Ophthalmology 1 Pediatrics 2 Physical examination 1 Psychiatry 1 Public health 1 Rehabilitation 1 Respiratory medicine 3 Rheumatology 1 Urology 1 Japan-specific medical system 17 (30.4%) Clinical research 1 Emergency 1 Psychiatry 1 Public health 14 Mathematical errors 4 (7.1%) Respiratory 1 Pediatrics 1 Cardiology 1 Medical interview 1 Others 2 (3.6%) Issue in English translation 1 Not providing an answer 1 22 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Figures Figure 1. Figures Figure 1. 23 Figure 1. 23 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint 24 24 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Figure 3. # user: User's utterance Japanese sentences: “Input a HTML-coded original question in Japanese here” Tuned prompts for English translation Tuned prompts for answering medical questions The answer is (word) because (explain the rationale for your choice). arrow examination, and head MRI are not indicated for this patient's presentation. =SAMPLE= In summary, which two diseases exhibit low complementemia out of the options provided, which are Cellulitis, Bacterial pneumonia, Ischemic colitis, Acute glomerulonephritis, and Mixed cryoglobulinemia? The answer requires selecting (two) choices. The answer is (Acute glomerulonephritis) and (Mixed cryoglobulinemia) because these two diseases are known to be associated with low complementemia. Acute glomerulonephritis is an immune-mediated kidney disease that can lead to low levels of complement proteins. Mixed cryoglobulinemia is a type of vasculitis that is characterized by the presence of abnormal proteins in the blood, which can lead to low complement levels. Cellulitis, bacterial pneumonia, and ischemic colitis are not typically associated with low complementemia. # user: User's utterance Question: <p>Which physical examination is difficult to perform on a patient with impaired consciousness?</p> Tendon reflex\|neck stiffness\|light reflex\|finger-to-nose test\|knee raise test (lower limb dropping test)" Question inpu Tuned prompts for answering medical questions Question input <p>Which physical examination is difficult to perform on a patient with impaired consciousness?</p> Tendon reflex\|neck stiffness\|light reflex\|finger-to-nose test\|knee raise test (lower limb dropping test)" 25 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 24, 2023. ; https://doi.org/10.1101/2023.04.17.23288603 doi: medRxiv preprint Figure 4. ure 4. 26
https://openalex.org/W2622496145	https://europepmc.org/articles/pmc6054011?pdf=render	English	null	Subsistence over symbolism: the role of transnational municipal networks on cities’ climate policy innovation and adoption	Mitigation and adaptation strategies for global change	2,017	cc-by	9,828	Mitig Adapt Strateg Glob Change (2018) 23:507–523 DOI 10.1007/s11027-017-9747-y ORIGINAL ARTICLE 1 Department of Environmental Systems Science, ETH Zurich (Swiss Federal Institute of Technology), Universitätstrasse 16, 8092 Zurich, Switzerland * Kaveh Rashidi kaveh.rashidi@usys.ethz.ch * Anthony Patt anthony.patt@usys.ethz.ch 1 Introduction Cities are the major source of greenhouse gas (GHG) emissions, with urban residents and activities accounting more than 70% of global anthropogenic GHG around the globe (World Bank 2010). Independent of efforts taken by national governments, numerous cities have started to launch their own mitigation policies (CDP 2015). Most of these activities aim at the reduction of GHG in the sectors of energy, transportation, urban land use, and waste (Bulkeley 2010; Xuemei 2007). Prior studies have explored several causes for cities’ adoption of climate policies. Some have focused on the socioeconomic and political status of cities as the major policy adoption drivers, while the others have mentioned that cities’ decisions are influenced by external factors such as the decisions of their national governments. y g One factor that may also play a role is the presence of city policy networks. While networks differ from each other, they generally have three defining characteristics. First, membership in the network is voluntary, meaning that cities are free to join or leave. Second, they are non-hierarchical, horizontal, and polycentric; therefore, they are considered to be self-governing. Third, the result of discussions—and consequently, decisions—are normally implemented by the member cities (Kern and Bulkeley 2009). Two networks, the Local Governments for Sustainability (ICLEI, from its previous name International Council for Local Climate Initiatives) and the more recent one known as C40, are the most noticeable transnational networks focusing on climate change mitigation programs, setting the grounds and facilitating the process for knowledge sharing. ICLEI is established in 1990 with the intention to assist local governments in transforming effectively towards a greener economy. It encompasses more than 1500 member cities around the world which represent more than 20% of the world population. ICLEI provides a platform for urban policy makers to exchange recent developments in the adoption of climate policies and the challenges they experi- enced in planning and implementation phases. ICLEI has offices around the globe and holds annual workshops and conferences in different member cities. C40 is the other large network. It now includes more than 80 cities, representing roughly 10% of the global population. Like ICLEI, the C40 network connects cities to address the issue of climate change through the development and implementation of programs and policies leading to a reduction of GHG emissions. Subsistence over symbolism: the role of transnational municipal networks on cities’ climate policy innovation and adoption Kaveh Rashidi1 & Anthony Patt1 Received: 22 November 2016 /Accepted: 2 May 2017 /Published online: 6 June 2017 # The Author(s) 2017. This article is an open access publication Abstract Urban areas account for the majority of global greenhouse gas emissions, and increasingly, it is city governments that are adopting and implementing climate mitigation policies. Many municipal governments have joined two different global city networks that aim to promote climate policy development at the urban scale, and there is qualitative evidence that such networks play an important role in motivating cities to adopt climate policies and helping them to implement them. Our study objective is to test this proposition quantitatively, making use of a global database on cities’ environmental policy adoption, and also taking into account a large number of other factors that could play a role in climate policy adoption. Controlling for these other factors, we find that network membership does make a significant difference in the number of different measures that city governments adopt. We also find that there are significant differences between the two different networks, suggesting that the nature of the services that such networks offer their members can play an important role. Our findings lead to the provision of a set of global mitigation strategies: First of all, joining the city networks can lead to a generation of global strategies which can result into climate mitigation benefits. However, cities are required to select the network which provides proper tailor made policies. Second, in the absence of concrete international commitments at the local level, city networks lay the ground for global governance and enable cities to adopt policies independently and proactively. Third, consideration of co-benefits of climate policies can optimize the develop- ment of global strategies. Keywords Climate mitigation policies . Networks . Policy diffusion . Cities . Climate chang Keywords Climate mitigation policies . Networks . Policy diffusion . Cities . Climate chang 1 Department of Environmental Systems Science, ETH Zurich (Swiss Federal Institute of Technology), Universitätstrasse 16, 8092 Zurich, Switzerland Mitig Adapt Strateg Glob Change (2018) 23:507–523 508 1 Introduction C40 organizes different conferences during the year to help cities learn from each other and provides advice and solutions to the cities. Both networks require their members to pay membership fees on an annual basis. The fees for C40, however, are substantially higher, and there is reason to believe that these higher fees are associated with more-carefully tailored policy analysis. This can be linked to the advisee’s willingness to cooperate and follow up for further advices when he or she has to pay for the service (Patt et al. 2006). There is evidence that city networks may play an important role in defining and developing climate-policy initiatives at the city level (Fünfgeld 2015). Studies have examined the rela- tionship between city networks and climate policy adoptions; however, these have mainly focused on the question of why cities join climate networks in the first place (Betsill and Bulkeley 2004; Buis 2009; Heinrichs et al. 2013; Kern and Bulkeley 2009; Pitt 2010). Only two studies have addressed the impact that networks have on the policies adopted (Krause 2012; Lee and Koski 2014), and for reasons that we explore in the following section, these leave major question still open, and the evidence remains weak as to whether network membership leads to an outcome of greater climate policy adoption. In this paper, we address this issue empirically. We go beyond prior studies, in that we take into account a wide range of other potential explanatory variables. We also differentiate between memberships in one or multiple networks. 2 Background Policy innovation is the adoption of policies that previously had not existed or are newly adopted by the government. Based on Mohr (1969), one can describe policy innovation as Bdirectly related to the motivation to innovate, inversely related to the strength of obstacles to innovation, and directly related to the resources available for overcoming such obstacles^ (p. 114). Several studies have focused on the questions of BWhat are the drivers of policy innovations in cities?^ and BWhy are some cities more innovative than the others?^ Walker (1969), in studying the diffusion of innovation and policy making among the American states, suggested that the likelihood of one state adopting a new policy is higher when others have already implemented it. In other words, cities tend to emulate policies that have already been tested elsewhere. Berry and Berry (1990) emphasized that the process of policy innovation in cities is influenced by not only internal factors such as socioeco- nomic developments, geographic characteristics, and political ideologies but also external factors that lead to the diffusion of ideas. Internal determinants of policy innovation have been shown to play an important role. Cities’ policy adoption can be linked to factors such as demographics, socioeconomic development, geography, and political ideology (Canon and Baum 1981; Gray 1973; Walker 1969). Several studies have drawn on this theory to examine the particular case of policies for climate adaptation and mitigation. Ryan (2015) listed three barriers to the adoption of climate policies by cities: lack of sufficient funding, adequate human resources, and information management; those cities that overcome these barriers are more likely to adopt more climate policies. The relationship between income and policy adoptions has been assessed by Collier (1997) and Mathy (2007); both observed positive correlations between income and the number of climate policies adopted. Another socioeconomic factor is the population of the cities. Betsill (2001) and Lee (2013) noted that more-populous cities have more resources to dedicate to the promotion of climate policies. Moreover, cities with more population are endowed with increased civic capacities and tend to push for more climate policies (Krause 2012; Lee and Koski 2014). The influence of citizens’ awareness of their governments is another factor that shapes the decisions of policy makers (Pierson 1993). One line of research has linked public awareness to the employment of women at various levels of occupations. Mitig Adapt Strateg Glob Change (2018) 23:507–523 509 2 Background These have to do with the institutional and governance context within which the city operates and the mechanism by which external factors operate can be classified as either vertical or horizontal diffusion. In vertical diffusion, policies are directed by the national governments to the local governments. These national decisions may thus shape the climate policies that are going to be adopted by municipal governments (Daley and Garand 2005). There is anecdotal evidence suggesting that cities in countries that had accepted the United Nations Framework Convention on Climate Change Kyoto Protocol earlier adopted more climate programs than those that joined later. National governments’ influence can also vary depending on the governing structure. In countries with a centralized (top-down) political system, the implementation of nationally legislated environmental and climate change policies seems to be more effective and quicker (Kimber 1995). The idea of horizontal diffusion suggests that the policies can be transferred from one city to another. Rogers (2010) defined such horizontal diffusion as Bthe process by which an innovation is communicated through certain channels over time among the member of a social system^ (p. 162). Berry and Berry (1990) emphasized that learning from the experience of the others and competition among cities are the primary drivers of horizontal diffusion. Hence, if one city implements a new policy and finds the policy to be successful and popular, there is a reasonable chance that the others that are informed of that innovation will adopt it as well. Two avenues for horizontal diffusion are attending international conferences and being a member of transnational networks. In fact, the two can be related, since transnational networks are a potentially important forum for local leaders to present their ideas, experiences, and achieve- ments, as well as to learn from others’ activities and expertise (Betsill and Bulkeley 2004; Meseguer 2005). Municipalities are now more connected than ever before, with connections made through networks, and provide opportunities for Bintermunicipal dialog,^ which results in global influences (Toly 2008). Local government associations can encourage professionalism for the member cities, especially those located in developing economies. Networks can play a role in Bagenda settings^ (Buis 2009). Busch (2015) believed that their functionalities are even more as they can be Bconsultant,^ Badvocate,^ Bcommitment brokers,^ and Bplatforms^ and thereby might result into different impacts on their members. Focusing on the Cities for Climate Protection (CCP) program, ICLEI (Betsill and Bulkeley 2004; Bulkeley et al. 2 Background Higher levels of female employment are an indicator of a society’s open mindedness and a government’s positive attitudes towards human rights and surrounding environ- ments. Torgler and Garcia-Valiñas (2007) found that women are generally more worried about climate change and related environmental problems. Geographic and local environmental factors also play a role, with a propensity towards natural hazards being a strong motivator of policy. Panayotou (1997) observed that governments react to disasters that they have experienced directly rather than disasters observed from afar. During recent years, the number of climate change-related hazards, such as hurricanes and coastal storms, has increased significantly, and when combined with projections of sea-level rise, one could expect that coastal cities would adopt more climate policies (Lee 2013; Nicholls et al. 2008) There are several studies that emphasize the importance of co-development benefits for climate policy (Bulkeley and Betsill 2005b; Krause 2011). Reduced fossil fuel consumption and air pollution are the most apparent local outcomes of many climate policies, with measureable improvements in public health and air visibility. Lowering fossil fuel consumption also reduces the dependency on importing energy, with benefits on balance of payments. One could expect that communities that are heavily dependent on the consumption of fossil fuels would be more likely to adopt certain climate policies, such as those that promote the installation of insulation systems in buildings, reducing energy consumption and the associated dependency on imported fossil energy sources. A large number of 510 Mitig Adapt Strateg Glob Change (2018) 23:507–523 cities suffer from air pollution, and governments have adopted various programs and initiatives to limit aerosol particulates (Betsill 2001; Lindseth 2004). Ryan (2015) found that for some cities, the co-benefits of climate policies are even more important as a motivating force than the mitigation of GHG emissions: co-benefits are simply more tangible, something that citizens can feel, compared to global benefits that come from reducing greenhouse gas emissions. These environmental factors can interact with policy makers’ ideological beliefs: the literature shows that governments that have implemented other environmental policies, addressing local concerns, tend to be more enthusiastic to adopt climate policies (Betsill 2001; Zahran et al. 2008a, b). In addition to internal factors leading to policy adoption, the literature also highlights the importance of external factors (Berry and Berry 1990). 2 Background 2003) concluded that the networks lay the groundwork for information and knowledge sharing. The networks also provide situations in which the member cities can compete with each other (Kern and Bulkeley 2009) and find the right partners to with which they could team up to attract external financial supports (Bulkeley et al. 2003). But networks can also be helpful in capacity building and rule setting (Andonova et al. 2009; Bulkeley et al. 2003), whereby the network operates as more than simply a venue for information exchange, but actually serves as in institutional locus of expertise and procedural guidance. They can provide benchmarking, recognition for cities, and some form of certification for the members that have achieved certain climate-policy goals (Kern and Bulkeley 2009). 511 Mitig Adapt Strateg Glob Change (2018) 23:507–523 There are increasing numbers of cities joining these climate mitigation city networks. Network studies have extensively addressed the questions of why cities join the networks (Betsill and Bulkeley 2004; Pitt 2010), what recognition benefits cities can expect (Buis 2009; Kern and Bulkeley 2009), and how memberships can provide visibility to leverage interna- tional funding (Betsill and Bulkeley 2004; Heinrichs et al. 2013). Moreover, the literature on cities’ climate policies and networks has mainly been qualitative in nature (Bontenbal 2013; Bulkeley 2005; de Jong et al. 2015; Schroeder et al. 2013; Toly 2008; Zeppel 2013), and studies that provide quantitative assessments are limited to a specific geographical boundary and do not provide generalized solutions (Busch 2016; Davies 2005; Kern and Bulkeley 2009; Krause 2012; Matisoff 2008; Pitt 2010). City networks, despite their increasing importance, are B… still relatively poorly understood with regards to their influence and impact on environmental outcome in local areas^ (Fünfgeld 2015) and Bnet effects of cities participation in transnational networks^ are not measured (Fünfgeld 2015). Two studies have gone the furthest in identifying the relationship between city network membership and climate policy adoption. Assessing the impact of ICLEI and the Mayors Climate Protection Agreement (MCPA) network membership in United States (US) cities, Krause (2012) applied ordinary least square (OLS) and propensity matching scores to answer the question of whether network membership had an influence on climate policy adoption. Krause found that it did, but left several issues unexamined. First, Krause studied only diffusion among American cities, raising the question of whether similar effects could be seen internationally, in particular in developing countries. Second, Krause studied a limited set of potential climate policies, leaving the question open of whether policies more tangentially related to mitigation would show the same effects. Third and related, Krause did not examine the question whether co-benefits to such policies could be an important driver. Lee and Koski (2014) reached a similar finding, doing so in an international context, but their methods raise some questions. They examined cities that were members of the C40 network and then looked at the relationship between additional membership in the ICLEI network and policy adoption. In this sense, their sample was already biased towards those cities that had already joined at least one network. Like Krause (2012), they also failed to take into account co-benefits as a potential driver. In light of these limitations, Fünfgeld (2015), writing a review paper of the field, highlighted a continued gap in the evidence for network membership effects. In our study, using a database of more than 127 cities, we address the following questions. First, can we observe a global impact of city network membership, in terms of the number climate policies that cities adopt? Second, and more nuanced, is there any difference between the city climate networks of C40 and ICLEI in terms of their impact on climate-policy adoptions? In studying this difference, we can differentiate between the horizontal transfers that take place when the network serves primarily as a forum for interaction, compared to when the network’s own institutions attempt to add additional value. 4 Dependent variable The policy count variable Y measures the total number of climate policies implemented by each city from 2010 to 2013. It encompasses policies from the following sectors: energy demand in buildings, finance, outdoor lighting, public procurement, transport, urban and land use, and waste and education. In total, a city can gain a maximum of 35 points for climate policies adopted. The dependent variable considers the algebraic sum of all the climate policies. Notably, only the policies that directly impact the GHG emissions have been selected. Table 1 lists these policies and the categories to which they belong. 3 Methods We have compiled data from various existing sources. The primary data sources are the reports submitted by cities to the CDP (formerly known as Carbon Disclosure Project) database (CDP 2015), which hosts a global database of cities, companies, and governments where environ- mental impact reports, actions on climate policies, and carbon footprints are reported to the CDP based on a self-reporting mechanism. The database of CDP is widely used in research 512 Mitig Adapt Strateg Glob Change (2018) 23:507–523 and policy making arenas. For variables related to a city’s dynamics, we use secondary data sources extracted from public sources and city municipality websites. We also introduce several explanatory variables controlling for environmental background of the cities to reduce the potential bias which may be caused by the voluntary nature of reporting the policy adoptions. Tables 1 and 2 present a full list of variables along with their sources. The dataset covers 127 cities around the world and includes 1651 data points in total, which is large enough to run statistical estimations. In order to assess the impact of city network membership on the adoption of climate policies, the following regression model is proposed Y ¼ β0 þ ∑ k j¼1 β jC j þ þαE þ δG þ γH þ ε ð1Þ ð1Þ In Eq. (1), Y denotes the total number of adopted policies by the city, β0 indicates the constant variable, Cj represents the control variables for the drivers of policy adoption while βj indicates the coefficients on each of those drivers. Two variables of E and G control for vertical diffusion of policies. The α reflects the coefficient of control variable E for government’s commitment towards climate change, δ is the coefficient for the governing structure G, γ is the coefficient for main variable H, ε is the error term, and k reflects the total number of variables in bundle of control variables. 5 Independent variables The independent variables Cj and H from Eq. 1 are listed in Table 2. In addition to the main variable counting network membership, we include a theory-driven set of control variables, also listed in Table 2. Going beyond previous studies, we include a set of variables capturing co-benefits of climate policy and a variable capturing the intrinsic propensity of a city to engage in environmental policy-making more generally. This latter factor makes use of the number of environmental policies in place other than those with a clear impact on climate mitigation or adaptation. Examples include the establishment of green spaces for recreation. Socioeconomic and geographic variables We have gathered GDP per capita from global databases and city municipalities for the income variable. We adjusted the values to 2009 US dollars, and we used the GDP deflator from the World Bank Database. We used logarithmic values of income to have a better distribution. We also issued logarithmic values of population 513 Mitig Adapt Strateg Glob Change (2018) 23:507–523 Table 1 Policies included in the dependent variables Category Policy Energy demand in buildings Transmission and distribution loss reduction Building codes and standards Smart grid implementation Low or zero carbon energy supply generation Combined heat and power Building performance rating and reporting Renewable onsite energy generation Financing mechanism for retrofit Clean energy procurement strategies Finance Carbon finance capacity building ESCO financing Carbon finance/markets Clean technology funds Outdoor lighting LED, CFL, and other luminaire technologies Solar powered lights Smart lighting Public procurement Encourage low carbon products Transport Improve fuel economy and reduce CO2 from motorized vehicles Improve the accessibility to public transit systems Infrastructure for non-motorized transport Improve bus transit times Improve the efficiency of freight systems and transport demand management Improve fuel economy and reduce CO2 from bus and/or light rail operations Urban land use Compact cities Limiting urban sprawl Urban agriculture Transit oriented development Eco-district development strategy Waste Waste to energy Methane recovery for reuse Improve the efficiency of waste collection Landfill gas capture Integrated waste management Wastewater to energy initiatives Education Climate change-focused curriculum Table 1 Policies included in the dependent variables for cities’ inhabitants. The hypothesis is that cities with larger populations will adopt more environmental and climate policies. Krause (2012) and Lee and Koski (2014) used a similar approach. The data have been gathered from the city databases and national statistics bureaus. for cities’ inhabitants. 5 Independent variables The hypothesis is that cities with larger populations will adopt more environmental and climate policies. Krause (2012) and Lee and Koski (2014) used a similar approach. The data have been gathered from the city databases and national statistics bureaus. National level variable values are extracted from the World Bank Database. The percentage of female employment with secondary education has been used as the proxy for educational level on climate change issues. Numerous studies have investigated the relationship between women and environmental concerns. For instance, Torgler and Garcia-Valiñas (2007) mention that in general, women demonstrate a greater level of concern towards environmental issues. Also, a larger level of women’s working age population correponds to Ba stronger preference for enivornemntal quality^ (Zhao et al. 2013). We would like to extend this further by examining relationship between females’ educational background and adoption of climate policies. The hypothesis is that a society which has more educated women at different employment levels should adopt more climate policies. A dichotomous variable for coastal location is used, taking the value of 1 if the city is coastal and 0 otherwise. Coastal areas are the first to be affected by climate change. As Nicholls et al. 5 Independent variables (2008) highlighted, coastal cities have either experienced the devastations Mitig Adapt Strateg Glob Change (2018) 23:507–523 514 Table 2 List of independent variables Variable Definition Source Socioeconomic and geographic Income Logarithmic values of GDP per capita for the cities City municipalities, National Bureau of Statistics Population Logarithmic values of population of cities City municipalities, National Bureau of Statistics Employment Percentage of female employment with secondary education World Bank Coastal location Binary variable, if city is coastal, it takes 1 otherwise 0 Authors’ desk-based research Co-development benefits Fossil fuel consumption Percentage of share of fossil fuel use to total energy consumption World Bank Air pollution Normalized values of air pollutions in cities World Bank Policy variables Environmental policies Normalized values of non-climate environmental policies CDP Target Binary variable, if city has GHG mitigation target, it takes 1 otherwise 0 Carbonn and C40 GHG (millions) Total amount of GHG in millions reported by cities ICLEI, C40, CDP, and cities municipalities Vertical diffusion National policies Years passed since the country accepted Kyoto protocol UNFCCC Governing structure Binary variable, if political system is top-down, it takes 1 otherwise (federal) 0 McGill (2016) caused by the climate change (in the form of coastal storms) or they are expecting it. In both cases, adaptation and mitigation policies would be required: one for the short-term prepared- ness in addressing local concerns in case of disaster (placing appropriate adaptation measures); the other is the policies with lasting impacts that could mitigate, eliminate, and avoid the occurrence of such disasters (applying climate mitigation policies) (Lee 2013; Nicholls et al. 2008). We, therefore, hypothesize that as the coastal cities are probably more exposed to the immediate impacts of climate change, they are more informed about the negative impacts and take more initiatives and policies to deal with it through not only implementing adaptation measures but also looking for ways to mitigate the likelihood of occurrence. Co-development benefits We have considered two variables to capture the co-development reasons that cities adopt climate policies. Fossil fuel use measures the share of fossil use against total energy consumption. We use the normalized values of cities’ air pollution from the World Bank Database. Policy variables Environmental policies unrelated to climate serves as a proxy for the local government’s political willingness to adopt climate policies. 5 Independent variables We include this variable to test the hypothesis that the higher number of policies adopted by a city to address environmental issues other Mitig Adapt Strateg Glob Change (2018) 23:507–523 515 than climate change (e.g., local air pollution), the more climate change mitigation policies are likely to be adopted as well. In other words, governments with greener attitudes adopt more climate policies (Betsill 2000; Bulkeley and Betsill 2005a). We extracted the values from the CDP database and normalized between 0 and 1. GHG target setting is measured as a binary variable to test the local government’s seriousness in tackling climate concerns. In the wake of climate change discussions, many cities have expressed their willingness to tackle GHG emissions by setting targets for reductions by 2020, 2025, or 2030. Target setting is an indicator of taking concrete steps in planning, designing, and implementing climate actions and in removing barriers such as a lack of human and financial resources (Holgate 2007). The other control variable is the logarithmic values of green- house gases per tons of CO2 emitted by cities. The assumption is that cities with a higher amount of GHG emissions have more incentive to adopt climate policies. The data has primarily been gathered from the CDP, Carbonn (a platform on which cities enter their GHG inventory data), and city municipalities. For the few cities without GHG accounting systems, national values—proportionally adjusted to the population—are used. The vertical diffusion element consists of two variables. First is the national policy variable which measures the years elapsed since the acceptance of Kyoto protocol (incorporated as E in Equation 1). This variable is in fact a policy variable that measures the influence of national governments on cities’ climate policy adoptions (commitments at the national level) as has been used similarly by Lee and Koski (2014). The list of countries with the years of acceptances has been extracted from the United Nations Framework Convention on Climate Change (UNFCCC) website. Similar to the work of (Lee 2013), we also introduce a variable to control for governing structure (indicated as G in Eq. 1) which measures the impact of governing system on adoption of climate polices at cities and sub-sovereign states. It is a binary variable and takes the value of 1 if the structure is top-down and 0 if it is otherwise, i.e., bottom-up (federal). 5 Independent variables This could matter, in the case of a governance system, whereby the central government has not only made international climate commitments but has also mandated city governments to adopt particular policies. These data have been collected from the list provided by McGill (2016). The main variable H, horizontal diffusion of policy through network membership, is a group of three dummy variables: one for membership in each of ICLEI and C40 and one interactive being membership in both networks. We have chosen these two networks based primarily on their global coverage, and the fact that they potentially contribute to a wide spectrum of climate policy adoption. Hence, we do not include data on membership in the Climate Alliance or the Covenant of Mayors, both of which are limited to Europe. We also do not examine membership in the Energy Cities network, which focuses on policies in the energy sector. With respect to the ICLEI network, we do not make use of data on cities’ membership in a particular ICLEI program, the Cities for Climate Protection (CCP). The CCP program is associated with cities adopting a particular set of climate policies, and membership is restricted to those cities that have done so. By contrast, participation in the wider ICLEI network may have an effect on cities’ policy adoption, but is not contingent on it. It is also worth noting that ICLEI has now rebranded the CCP and integrated into other agendas (van Staden 2017; Yi et al. 2017). 6 Results Figure 1 provides a first approximation of the effect of network members. It shows the mean number of climate policies adopted, differentiated by cities’ membership in climate networks. Cities belonging only to ICLEI adopt more policies on average than those belonging to no 516 Mitig Adapt Strateg Glob Change (2018) 23:507–523 networks, while cities belonging to C40, either instead of or in addition to ICLEI, adopt more policies than those belonging only to ICLEI. There is no statistical difference between those cities belonging only to C40 and those belonging to both networks, whereas the differences between other means are significant at the 95% confidence interval. While indicative of an overall trend, Fig. 1 does not reveal whether other factors, potentially correlating with network membership, could be explaining the observed differences in policy adoption. We used three regression models in order to do so, each of which could be considered appropriate for our data structure: ordinary least squared (OLS), Poisson, and negative binomial (NB). In all cases, the results from Poisson and NB models agree qualita- tively, while OLS results are in some contradictory with respect to the effects of control variables. However, OLS is somewhat problematic because the residuals are not normally distributed (p < .001). Moreover, overdispersion tests show that the null hypothesis of having no overdispersion is rejected, as p < .001, suggesting that of the three models, the NB provides the most efficient and robust estimates. We present all three models, but draw our qualitative conclusion from the NB results. Table 3 shows the results based on three estimation techniques. What these results allow us to see, as cannot been seen in Fig. 1, is the influence of the network membership variables when taking the control variables into account. In contrast to Fig. 1, it shows that ICLEI membership alone makes only a marginally significant impact on average numbers of policies adopted. Membership in the C40 or both networks makes a larger and clearly significant impact. On average, cities belonging to both networks adopt 1.69 more policies than those cities that are not members. This effect is roughly the same as is accounted for by a doubling of per capita income, close to half of the potential variance on the adoption of other environmental policies, and over half of the potential variance on air pollution. 6 Results Cities that are a member of both of the networks have adopted more climate policies than those that are members in either or none of the networks 0 3 6 9 12 15 None ICLEI C40 Both Climate policies adopted by cities Network membership Mitig Adapt Strateg Glob Change (2018) 23:507–523 517 Table 3 The results for the OLS methods and negative binomial methods (N = 127) Variables OLS Poisson NB Socioeconomic and geographic Income 3.253 * (1.235) 1.990 * (1.181) 2.015 * (1.218) Population −1.031 (0.661) 0.937 (1.084) 0.939 (1.103) Employment −0.075 * (0.022) 0.989 * (1.003) 0.987 *** (1.003) Coastal location 0.728 (0.607) 0.985 (1.074) 0.971 (1.092) Co-development benefits Fossil fuel consumption −0.045 * (0.024) 0.997 (1.002) 0.993 (1.003) Air pollution 5.142 * (2.975) 3.130 ** (1.495) 3.026 * (1.618) Policy variables Environmental policies 12.870 * (1.295) 3.397 * (1.137) 3.731 * (1.179) Target 1.310 (0.744) 1.221 (1.098) 1.214 * (1.117) GHG 0.000 * (0.000) 1.000 (1.000) 1.000 ** (1.000) Vertical diffusion National policies 0.018 (0.112) 0.984 (1.013) 0.983 (1.016) Governing structure 0.452 (0.799) 1.005 (1.092) 1.035 (1.116) Horizontal diffusion Networks (C40) 1.893 * (1.071) 1.440 * (1.127) 1.392 (1.159) Networks (ICLEI) 0.677 (0.786) 1.283 ** (1.113) 1.274 * (1.134) Networks (both) 3.594 * (1.041) 1.654 * (1.127) 1.691 *** (1.157) Intercept 1.959 (7.756) 0.826 (2.650) 0.812 (3.219) R2 0.704 0.697 0.697 Number in brackets indicate standard errors Statistical significance at 90% confidence level; 95% level; *99% level The results for the OLS methods and negative binomial methods (N = 127) by 0.993. Air pollution also has a positive significant relation with the city’s climate policy adoption. by 0.993. Air pollution also has a positive significant relation with the city’s climate policy adoption. Policy variables also show significant positive correlations with the adoption of climate policies. Cities that have implemented more environmental policies also adopt more climate policies, as do cities that have set GHG reduction targets. Cities with higher GHG emissions are more likely to adopt climate policies. Figure 2 illustrates graphically the results with respect to network membership. It plots predicted values for climate policy adoption based on the control variables, compared to observed values. What can be observed is that with few exceptions, cities that are members of both networks lie above the 45° line, indicating that observed policy adoption exceeds that which would be adopted based on other factors. Number in brackets indicate standard errors Statistical significance at 90% confidence level; 95% level; *99% level 6 Results Vertical diffusion was found to have no significant impact on the cities’ decisions in adopting climate policies. In terms of these other variables, income is significant at the 99% level, supporting the hypothesis that richer cities have a larger budget for climate change policies. Female employ- ment with higher education is also significant at 99%. No significant correlations were found for population or coastal locations. Both of the co-development benefits variables are statis- tically significant. Cities with more dependency on fossil fuel adopt more climate policies. A 1% increase into the dependency on fossil fuel consumption increases climate policies adopted 0 3 6 9 12 15 None ICLEI C40 Both Climate policies adopted by cities Network membership Fig. 1 The diagram shows the mean within cities policy adoption as a function of city networks. Error bars represent ±1.96 standard errors, the 95% confidence interval. 7 Discussion and conclusion 2 Plot of observed values of climate mitigation policies, compared to those predicted using the estimated model, and considering all factors other than network membership. The shaded circles represent the network membership of cities, and points above the 45° line represent the outperformers. As is shown in graph, the cities that belong to either or both of the networks are generally located above the line with other studies that show that policy advice is more frequently followed when the advisee has to pay for it (Patt et al. 2006). Moreover, networks with different profiles can play roles of a consultant, an advocate, a Bmatch maker^ or a commitment broker (Busch 2015), and one network can be strong in one or all of these features. This may lead to variations in the quality of services delivered, and potentially justify the differences in the likelihood of adopted policies by the member cities. Our results indicate that membership plays a role even when considering other important, and previously unexamined, driving factors. Most of the results with respect to the various control variables are consistent with the existing literature. The results suggest that the level of air pollution impacts the adoption of climate policies by cities. Cities are interested in the adoptions of climate policies that have public health and economic development co-benefits (Bulkeley and Betsill 2005a). Reducing air pollutants and increasing visibility are tangible co-benefits for several mitigation programs. The policy implication for development and research institutes is that when suggesting climate policies for cities, they should also place sufficient emphasis on the importance of co-benefits. The other co-benefits of climate policies are the reduction of fossil fuel consumption as the result of energy efficient activities. Apart from the few regions that produce their own fossil fuel, the rest are heavily dependent on the fossil fuel imported from somewhere else. Therefore, as was expected, the findings show that an increase in the share of fossil fuel consumption increases the adoption of climate policies. Among the socioeconomic factors, income and female’s employment influence the policy makers’ decisions in the adoption of the climate policies. The findings show that cities with higher incomes adopt more climate policies. Two possible interpretations can be made from this finding. 7 Discussion and conclusion The results show that joining international networks correlates with an increase in the adoption of climate policies. The results also show that cities in both networks adopt more climate policies than those in only one. Cities that are a member of the C40 network are more likely to adopt climate policies than those that are a member of ICLEI. One can speculate on several reasons, including the fact that C40, with fewer members, can offer a higher level of support to each member. C40 with 90 employees has to cope with 80 members (in theory above one person per city) while ICLEI with 200 employees works with 1500 member cities (less than 0.13 person per city). Another potential reason could be the higher cost of entry, consistent Mitig Adapt Strateg Glob Change (2018) 23:507–523 518 with other studies that show that policy advice is more frequently followed when the advisee has to pay for it (Patt et al. 2006). Moreover, networks with different profiles can play roles of a consultant, an advocate, a Bmatch maker^ or a commitment broker (Busch 2015), and one network can be strong in one or all of these features. This may lead to variations in the quality of services delivered, and potentially justify the differences in the likelihood of adopted policies by the member cities. Our results indicate that membership plays a role even when considering other important, and previously unexamined, driving factors. Most of the results with respect to the various control variables are consistent with the Fig. 2 Plot of observed values of climate mitigation policies, compared to those predicted using the estimated model, and considering all factors other than network membership. The shaded circles represent the network membership of cities, and points above the 45° line represent the outperformers. As is shown in graph, the cities that belong to either or both of the networks are generally located above the line Fig. 2 Plot of observed values of climate mitigation policies, compared to those predicted using the estimated model, and considering all factors other than network membership. The shaded circles represent the network membership of cities, and points above the 45° line represent the outperformers. As is shown in graph, the cities that belong to either or both of the networks are generally located above the line Fig. 7 Discussion and conclusion As opposed to cities in developing economies, cities with higher income have already dedicated resources for climate mitigation programs in their strategic development plans in consideration of the potential benefits and budget requirements. The results are consistent with previous literature that reports the importance of financial sources for the development of mitigation policies (Collier 1997; Mathy 2007) and the promotion of innovation in the cities (Mohr 1969), as well as the finding that successful implementation of policy innovations require larger Mitig Adapt Strateg Glob Change (2018) 23:507–523 519 administrative staffs with the knowledge of mitigation policies that can collaborate with contractors (Collier 1997; Mathy 2007). This is the case for wealthier cities where more individuals for mitigation programs can be allocated. Finally, we observe that higher rates of employment of females with higher education have a positive relationship with the adoption of climate policies. The positive correlation between higher education of inhabitants and the adoption of climate policies has already been found (Zahran et al. 2008a, b). Women are more concerned about environmental and climate change problems (Torgler and Garcia-Valiñas 2007) and therefore have different levels of involvement in occupations and local communities that help promote climate change concerns. administrative staffs with the knowledge of mitigation policies that can collaborate with contractors (Collier 1997; Mathy 2007). This is the case for wealthier cities where more individuals for mitigation programs can be allocated. Finally, we observe that higher rates of employment of females with higher education have a positive relationship with the adoption of climate policies. The positive correlation between higher education of inhabitants and the adoption of climate policies has already been found (Zahran et al. 2008a, b). Women are more concerned about environmental and climate change problems (Torgler and Garcia-Valiñas 2007) and therefore have different levels of involvement in occupations and local communities that help promote climate change concerns. p p g One previous study has found a negative correlation between GHG emissions and the adoption of climate policies (Zahran et al. 2008a, b). We find the opposite. The underlying reason could be that cities have become more concerned about climate change and their GHG footprint. In several cities, such as Shanghai and Tokyo, there exist city-level trading schemes that seek to adopt climate policies in response to their massive GHG emissions. 7 Discussion and conclusion We also find that cities that have already implemented environmental initiatives are also more willing to adopt climate policies. This is linked to the capacities that are built in these regions. Politically liberal and educated communities with a successful history of environmental activities react properly to the negative impacts of climate change and support the adoption of mitigation programs (Zahran et al. 2008a, b). Our results also show that cities that have set fixed targets for the reduction of GHG emissions (compared to the base year’s scenarios) adopt more climate polices. We have also found that vertical diffusion has no impact on the adoption of climate policies. The policy implication is that, for cities without concrete national commit- ment, the decisions regarding the adoption of climate policies can be made independently. For instance, in the absence of solid national commitment and law enforcement, various US cities have been actively engaging in climate change dialogs and adopting related policies. This can be linked to the fact that financial and fiscal autonomy can potentially bring about empower- ment to cities so they can adopt more efficient policies and allocate financial sources effectively (Lee 2013; UN HABITAT 2009). On the other hand, even in areas where commitments exist, cities can be more ambitious in adopting policies. This has been highlight- ed in the conference of parties (COP) 21 agreement (COP 21 2015) and conforms with current practical cases. For example, in Japan, with a persistent national commitment, some cities were able to reach the mitigation targets faster than what was instructed to them. The very good example is the cap-and-trade system initiated by the city of Tokyo (Nishida and Hua 2011) that has led to a significant amount of GHG emission reduction of 25% only in 5 years, much earlier than the initial plans (City of Tokyo 2016). One potential limitation bias in our results could stem from the voluntary nature of cities’ contribution to the CDP database. It is possible that cities less interested in enacting climate policies, and not joining a network, would not appear in our sample of cities. If such a selection bias exists, however, the result would be to decrease the observed effect of network member- ship. Hence, we can say that our results are, in this respect, conservative. It is possible that the effects of network membership are greater than those that we have observed. 7 Discussion and conclusion Another potential bias in such empirical studies is that resulting from unobserved variables. One can surmise, for example, that an unobserved variable accounts for network membership and climate policy adoption, meaning that the relationship observed between the two is not causal. We believe that we have gone beyond previous studies by including a wide range of such potential variables. In particular, we have included the variable for the adoption of environmental but non-climate policies, which we believe can be an indicator for cities’ overall commitment to 520 Mitig Adapt Strateg Glob Change (2018) 23:507–523 the environment. What we find is that including these variables makes the observed correlation between ICLEI membership and climate policy adoption appears to be insignificant, whereas without the inclusion of these variables, it had appeared to be significant. This stands in contrast to the results for C40 membership, membership in which persists as significant even with the additional variables added. This suggests two things. First, it suggests that the effect observed from C40 may not be as a result of a previously unobserved variable. Second, and more importantly, it suggests the existence of causality between C40 membership and policy adoption. It is important to note that we have not measured an adopted policies’ impact. In other words, the policies adopted through joining city networks might have larger mitigation impacts for some cities but have lesser impacts in others. Examining this issue, Pitt (2010) and Zhao et al. (2013) apply similar methods and arguments in their estimations of environmental policies and their impact on governments’ climate actions. We too would likely have been able to observe whether cities that adopt more policies actually make more progress in terms of emissions reduction or the diffusion of low carbon technologies. The fact is, however, that it is too soon to be able to measure such effects. The data for climate policy adoption only commence in 2012, and there are no data yet available to be able to estimate the effectiveness of these measures. This can be viewed as an important area for future research. City networks are increasingly receiving attention from cities around the world. In this paper, we have attempted to analyze the impact of membership in these two city networks on the adoption of climate policies at cities. 7 Discussion and conclusion Using a large global dataset of cities and controlling for economic, political, environmental, and other policy variables, we have found that joining city networks correlates with the adoption of more climate policies by cities. We also found that across the cities of our study, cities that are a member of C40 on average adopt more climate policies than those of ICLEI. Joining both of the networks increases the likelihood of more policies adopted; however, the impact is not significantly higher than joining either of the networks. Consistent with the previous literature, we have found that other socioeconomic, demographic, policy, and environmental variables also increase the policy adoption likelihood. Our results contribute to the provision of a set of global mitigation strategies through evaluation of the role of international city networks in shaping policy decisions. City networks can contribute to the generation of global strategies, which in turn can translate into climate mitigation benefits. Our results suggest that cities should decide to choose networks that provide sets of tailor-made policies that meet their actual requirements. Acknowledgements Funding for the research was provided by Climate-KIC and the European Research Council (grant number 313533). Additional support came from the South Pole Group. Helpful contributions from Patrick Bürgi, Martin Stadelmann, and Tobias Schmidt are acknowledged. All remaining errors remain those of the authors. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and repro- duction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. References Andonova L, Betsill M, Bulkeley H (2009) Transnational climate governance. Glob Environ Polit. Retrieved from http://www.mitpressjournals.org/doi/abs/10.1162/glep.2009.9.2.52 Andonova L, Betsill M, Bulkeley H (2009) Transnational climate governance. Glob Environ Polit. Retrieved from http://www.mitpressjournals.org/doi/abs/10.1162/glep.2009.9.2.52 Berry F, Berry W (1990) State lottery adoptions as policy innovations: an event history analysis. Am Polit Rev 84:395–415 Retrieved from http://journals.cambridge.org/abstract_S0003055400192561 Berry F, Berry W (1990) State lottery adoptions as policy innovations: an event history analysis. Am Polit Sci Rev 84:395–415 Retrieved from http://journals.cambridge.org/abstract_S0003055400192561 521 Mitig Adapt Strateg Glob Change (2018) 23:507–523 Betsill M (2000) Localizing global climate change: controlling greenhouse gas emissions in US cities. Retrieved from http://belfercenter.hks.harvard.edu/publication/2755/localizing global climate change.html Betsill M (2000) Localizing global climate change: controlling greenhouse gas emissions in US cities. Re from http://belfercenter.hks.harvard.edu/publication/2755/localizing_global_climate_change.html ( ) g g g g g g from http://belfercenter.hks.harvard.edu/publication/2755/localizing_global_climate_change.html Betsill M (2001) Mitigating climate change in US cities: opportunities and obstacles. Local Environ. Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13549830120091699 Betsill M (2001) Mitigating climate change in US cities: opportunities and obstacles. Local Environ. Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13549830120091699 Betsill M (2001) Mitigating climate change in US cities: opportunities and obstacles. Local Environ. Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13549830120091699 p Betsill MM, Bulkeley H (2004) Transnational networks and global environmental governance: the cities for climate protection program. Int Stud Q 48(2):471–493. doi:10.1111/j.0020-8833.2004.00310.x p Betsill MM, Bulkeley H (2004) Transnational networks and global environmental governance: the cities for climate protection program. Int Stud Q 48(2):471–493. doi:10.1111/j.0020-8833.2004.00310.x Betsill MM, Bulkeley H (2004) Transnational networks and global environmental governance: the cities for climate protection program. Int Stud Q 48(2):471–493. doi:10.1111/j.0020-8833.2004.00310.x Bontenbal MC (2013) Differences in learning practices and values in north-south city partnerships broader understanding of mutuality. Public Adm Dev 33(2):85–100. doi:10.1002/pad.1622 broader understanding of mutuality. Public Adm Dev 33(2):85 100. doi:10.1002/pad.1622 Buis H (2009) The role of local government associations in increasing the effectiveness of city-to-city cooper- ation. Habitat Int 33(2):190–194. doi:10.1016/j.habitatint.2008.10.017 g y p Buis H (2009) The role of local government associations in increasing the effectiveness of city-to-city cooper- ation. Habitat Int 33(2):190–194. doi:10.1016/j.habitatint.2008.10.017 Bulkeley H (2005) Reconfiguring environmental governance: towards a politics of scales and networks. Polit Geogr 24(8):875–902. doi:10.1016/j.polgeo.2005.07.002 Bulkeley H (2010) Cities and the governing of climate change. Annu Rev Environ Resour 35(November 2010): 229–253 Retrieved from http://www.annualreviews.org/eprint/CxNzxuwSsgGPrr4QvY88/full/10.1146 /annurev-environ-072809-101747?utm_source=&utm_medium=environ&utm_campaign=eprint Bulkeley H, Betsill M (2005a) Cities and climate change: urban sustainability and global environmental governance. Retrieved from https://books.google.ch/books?hl=en&lr=&id=1M_NEGvtg7YC&oi=fnd&pg=PR8&dq= Cities+and+climate+change:+Urban+sustainability+and+global+enviornmental+governance&ots=NxylcJ_tA8 &sig=tgrUIY9CK8_mgJkXpuyZkbPb0Zc Bulkeley H, Betsill M (2005b) Cities and climate change: urban sustainability and global environmental governance. Retrieved from https://books.google.ch/books?hl=en&lr=&id=1M_NEGvtg7YC&oi= fnd&pg=PR8&dq=Urban+sustainability+and+global+environmental+governance&ots=NxylcJ_tv0&sig=1 M-deHzEoyf9KZ77H34p25F6Qgw Bulkeley H, Dvies A, Evans B, Gibbs D, Kern K, Theobald K (2003) Environmental governance and transnational municipal networks in Europe. J Environ Policy Plan 5(3):235–254. doi:10.1080 /1523908032000154179 Busch H (2015) Linked for action? An analysis of transnational municipal climate networks in Germany. Int J Urban Sustain Dev 7(2):213–231. doi:10.1080/19463138.2015.1057144 Busch H (2016) Entangled cities—transnational municipal climate networks and urban governance. Retrieved from http://lup.lub.lu.se/record/78ae371f-bb48-4bcd-ad2d-920efb4fad7e Canon B, Baum L (1981) Patterns of adoption of tort law innovations: an application of diffusion theory to judicial doctrines. Am Polit Sci Rev 75(4):975–987 Retrieved from http://journals.cambridge.org/abstract_ S0003055400181237 CDP (2015) CDP cities 2015. Retrieved October 8, 2015, from https://www.cdp.net/en-US/Pages/events/2015 /cities/infographic.aspx g p p City of Tokyo (2016) Tokyo cap and trade\|environment of Tokyo. Retrieved from http://www.kankyo.metro. tokyo.jp/en/climate/cap_and_trade.html Collier U (1997) Local authorities and climate protection in the European union: putting subsidiarity into practice? Local Environ 2(1):39–57 Retrieved from http://www.tandfonline.com/doi/abs/10.1080 /13549839708725511 COP 21 (2015) COP 21. Retrieved June 1, 2016, from https://unfccc.int/resource/docs/2015/cop21/eng/l09r01.pdf Daley D, Garand J (2005) Horizontal diffusion, vertical diffusion, and internal pressure in state environmental policymaking, 1989-1998. Am Politics Res 33(5):615–644 Retrieved from http://apr.sagepub. com/content/33/5/615.short Davies AR (2005) Local action for climate change: transnational networks and the Irish experience. Local Environ 10(1):21–40. doi:10.1080/1354983042000309298 de Jong M, Joss S, Schraven D, Zhan C, Weijnen M (2015) Sustainable–smart–resilient–low carbon–eco– knowledge cities; making sense of a multitude of concepts promoting sustainable urbanization. J Clean Prod. doi:10.1016/j.jclepro.2015.02.004 Fünfgeld H (2015) Facilitating local climate change adaptation through transnational municipal networks. Current Opinion in Environmental Sustainability. Retrieved from http://www.sciencedirect. com/science/article/pii/S1877343514000839 Fünfgeld H (2015) Facilitating local climate change adaptation through transnational municipal networks. Current Opinion in Environmental Sustainability. Retrieved from http://www.sciencedirect. com/science/article/pii/S1877343514000839 Gray V (1973) Innovation in the states: a diffusion study. Am Polit Sci Rev 67(4):1174–1185 Retrieved from http://journals.cambridge.org/abstract_S0003055400147756 Gray V (1973) Innovation in the states: a diffusion study. Am Polit Sci Rev 67(4):1174–1185 Retrieved from http://journals.cambridge.org/abstract_S0003055400147756 p j g g _ Heinrichs D, Krellenberg K, Fragkias M (2013) Urban responses to climate change: theories and governance practice in cities of the Global South. Int J Urban Reg Res 37(6):1865–1878. doi:10.1111/1468-2427.12031 Heinrichs D, Krellenberg K, Fragkias M (2013) Urban responses to climate change: theories and governance practice in cities of the Global South. Int J Urban Reg Res 37(6):1865–1878. doi:10.1111/1468-2427.12031 Holgate C (2007) Factors and actors in climate change mitigation: a tale of two South African cities. Local Environ 12(5):471–484 Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13549830701656994 Holgate C (2007) Factors and actors in climate change mitigation: a tale of two South African cities. Local Environ 12(5):471–484 Retrieved from http://www.tandfonline.com/doi/abs/10.1080/13549830701656994 Kern K, Bulkeley H (2009) Cities, Europeanization and multi-level governance: governing climate change through transnational municipal networks. J Common Mark Stud 47(2):309–332. doi:10.1111/j.1468-5965.2009.00806.x Kern K, Bulkeley H (2009) Cities, Europeanization and multi-level governance: governing climate change through transnational municipal networks. J Common Mark Stud 47(2):309–332. doi:10.1111/j.1468-5965.2009.00806.x 522 Mitig Adapt Strateg Glob Change (2018) 23:507–523 Kimber CJM (1995) A comparison of environmental federalism in the United States and the European Union. Maryland Law Review, 54. Retrieved from http://heinonline.org/HOL/Page?handle=hein.journals/mllr54 &id=1670&div=46&collection=journals j Krause R (2011) Policy innovation, intergovernmental relations, and the adoption of climate protection initiatives by US cities. J Urban Aff 33(1):45–60 Retrieved from http://onlinelibrary.wiley.com/doi/10.1111/j.1467- 9906.2010.00510.x/pdf Krause RM (2012) An assessment of the impact that participation in local climate networks has on cities’ implementation of climate, energy, and transportation policies. Rev Policy Res 29(5):585–604. doi:10.1111 /j.1541-1338.2012.00582.x j Lee T (2013) Global cities and transnational climate change networks. Global Environ Politics 13(1):108–127. doi:10.1162/GLEP_a_00156 Lee T, Koski C (2014) Mitigating global warming in global cities: comparing participation and climate change policies of C40 cities. J Comp Policy Anal: Res Pract. Retrieved from http://www.tandfonline. com/doi/abs/10.1080/13876988.2014.910938 Lindseth G (2004) The Cities for Climate Protection Campaign (CCPC) and the framing of local climate policy. Local Environ 9(4):325–336. doi:10.1080/1354983042000246252 Mathy S (2007) Urban and rural policies and the climate change issue: the French experience of governance. Environ Sci 4(3):159–169 Retrieved from http://www.tandfonline.com/doi/abs/10.1080 /15693430701742701 Matisoff DC (2008) The adoption of state climate change policies and renewable portfolio standards: regional diffusion or internal determinants? Rev Policy Res 25(6):527–546. doi:10.1111/j.1541-1338.2008.00360.x McGill (2016) List of countries by system of government. Retrieved from http://cs.mcgill. ca/~rwest/wikispeedia/wpcd/wp/l/List_of_countries_by_system_of_government.htm Meseguer C (2005) Policy learning, policy diffusion, and the making of a new order. Ann Am Acad Pol 598(1): 67–82 Retrieved from http://ann.sagepub.com/content/598/1/67.short Mohr L (1969) Determinants of innovation in organizations. Am Pol Sci Rev 63(1):111–126 Retrieved from http://journals.cambridge.org/abstract_S0003055400261510 Nicholls R, Hanson S, Herweijer C, Patmore N (2008) Ranking port cities with high exposure and vulnerability to climate extremes. Retrieved from http://www.oecd-ilibrary.org/environment/ranking-port-cities-with-high- exposure-and-vulnerability-to-climate-extremes_011766488208?crawler=true&mimetype=application/pdf Nishida Y, Hua Y (2011) Motivating stakeholders to deliver change: Tokyo’s cap-and-trade program. Build Res Inf. Retrieved from http://www.tandfonline.com/doi/abs/10.1080/09613218.2011.596419 Panayotou T (1997) Demystifying the environmental Kuznets curve: turning a black box into a policy tool. Environ Dev Econ 2(4):465–484. doi:10.1017/S1355770X97000259 Panayotou T (1997) Demystifying the environmental Kuznets curve: turn Environ Dev Econ 2(4):465–484. doi:10.1017/S1355770X97000259 Patt AG, Bowles HR, Cash DW (2006) Mechanisms for enhancing the credibility of an adviser: prepayment aligned incentives. J Behav Decis Mak 19(4):347–359. doi:10.1002/bdm.532 Pierson P (1993) When effect becomes cause: policy feedback and political change. World Politics 45(4):595– 628 Retrieved from http://journals.cambridge.org/abstract_S0043887100008911 p j g g _ Pitt D (2010) The impact of internal and external characteristics on the adoption of climate mitigation policies by US municipalities. Environ Plann C Gov Policy 28:851–871 Retrieved from http://envplan. com/epc/fulltext/c28/c09175.pdf Rogers EM (2010) Diffusion of innovations, 4th Edition. Simon and Schuster. Retrieved from https://boo google.com/books?id=v1ii4QsB7jIC&pgis=1 Ryan D (2015) From commitment to action: a literature review on climate policy implementation at city level. Clim Chang 131(4):519–529 Retrieved from http://link.springer.com/article/10.1007/s10584-015-1402-6 Schroeder H, Burch S, Rayner S (2013) Novel multisector networks and entrepreneurship in urban climate governance. Environ Plann C GovPolicy 31(5):761–768. doi:10.1068/c3105ed Toly NJ (2008) Transnational municipal networks in climate politics: from global governance to global politics. Globalizations 5(3):341–356. doi:10.1080/14747730802252479 Torgler B, Garcia-Valiñas MA (2007) The determinants of individuals’ attitudes towards preventing environ- mental damage. Ecol Econ 63(2–3):536–552 Retrieved from http://www.sciencedirect. com/science/article/pii/S0921800906006021 p UN HABITAT (2009) Urban indicators guidelines - Bbetter information, better cities^: monitoring the habitat agenda and the millennium development goals-slum target, p 47. Retrieved from http://www.unhabitat.org S d M (2017) S i bl E T i i l l k I U T ( d) T d p UN HABITAT (2009) Urban indicators guidelines - Bbetter information, better cities^: monitoring the habitat agenda and the millennium development goals-slum target, p 47. Retrieved from http://www.unhabitat.org van Staden M (2017) Sustainable Energy Transition: local governments as key actors. In: Uyar T. (ed) Towards 100% renewable energy Springer Proceedings in Energy Springer Cham agenda and the millennium development goals-slum target, p 47. Retrieved from http://www.unhabitat.org van Staden M (2017) Sustainable Energy Transition: local governments as key actors. In: Uyar T. (ed) Towards 100% renewable energy. Springer Proceedings in Energy. Springer, Cham van Staden M (2017) Sustainable Energy Transition: local governments as key actors. In: Uyar T. (ed) Towards 100% renewable energy. Springer Proceedings in Energy. Springer, Cham Walker J (1969) The diffusion of innovations among the American states. A Retrieved from http://journals.cambridge.org/abstract_S0003055400000149 Walker J (1969) The diffusion of innovations among the American states. Am Pol Sci Rev 63(3):880– Retrieved from http://journals.cambridge.org/abstract_S0003055400000149 World Bank (2010) Cities and climate change: an urgent agenda. Retrieved May 19, 2015, from http://siteresou worldbank.org/INTUWM/Resources/340232-1205330656272/CitiesandClimateChange.pdf 523 Mitig Adapt Strateg Glob Change (2018) 23:507–523 Xuemei B (2007) Integrating global environmental concerns into urban management. J Ind Ecol 11(2):15–29 Retrieved from http://www.bonn-dialogues.com/file/get/9724.pdf Yi H, Krause RM, Feiock RC (2017) Back-pedaling or continuing quietly? Assessing the impact of ICLEI membership termination on cities’ sustainability actions. Environmental Politics 26(1):138–160. doi:10.1080 /09644016.2016.1244968 Zahran S, Brody S, Vedlitz A (2008a) Vulnerability and capacity: explaining local commitment to climate- change policy. Environ Plann C Gov Policy 26(3):544–562 Retrieved from http://envplan. com/epc/fulltext/c26/c2g.pdf p g p Zahran S, Grover H, Brody S, Vedlitz A (2008b) Risk, stress, and capacity explaining metropolitan commitment to climate protection. Urban Aff Rev 43(4):447–474 Retrieved from http://uar.sagepub.com/content/43 /4/447.short Zeppel H (2013) The ICLEI Cities for Climate Protection programme: local government networks in urban climate governance. Climate Chang Glob Policy Regimes. Retrieved from http://link.springer. com/chapter/10.1057/9781137006127_15 Zhao Y, Tang KK, Wang L (2013) Do renewable electricity policies promote renewable electricity generation? Evidence from panel data. Energy Policy 62:887–897. doi:10.1016/j.enpol.2013.07.072
https://openalex.org/W2514525755	https://inria.hal.science/hal-01637474/document	English	null	An Optimization Based Software Tool for Individual Automated Guideway Transit Systems	Lecture notes in computer science	2,016	cc-by	5,216	To cite this version: Ezzeddine Fatnassi. An Optimization Based Software Tool for Individual Automated Guideway Tran- sit Systems. 15th IFIP International Conference on Computer Information Systems and Industrial Management (CISIM), Sep 2016, Vilnius, Lithuania. pp.322-333, 10.1007/978-3-319-45378-1_29. hal-01637474 Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-01637474 https://inria.hal.science/hal-01637474v1 Submitted on 17 Nov 2017 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License An optimization based software tool for Individual automated Guideway Transit Systems Ezzeddine Fatnassi1 Institut Supérieur de Gestion de Tunis Université de Tunis 41, Rue de la Liberté - Bouchoucha - 2000 Bardo, Tunisie Abstract. Automated Guideway Transit is a form of public transporta- tion tools where fully driverless vehicles operate in order to move passen- gers between specic locations. Automated Guideway Transit includes a large variety of systems including mass transit systems and limited peo- ple automated guideway transit systems. In this paper, we focus on a specic limited people mover system called personal rapid transit. We develop in this paper an optimization based software which could man- age eciently the empty vehicles movements within the personal rapid transit system. Within this context, a branch and bound algorithm is proposed and its eciently is proved on a set of instances taken from the literature. Our algorithm is shown to get good quality results. Keywords: Automated guideway transit, Ondemand Transportation systems, Optimization, Branch And Bound 1 Details about Cplex could be found in https://www-01.ibm.com/software/ commerce/optimization/cplex-optimizer/ 2 Ezzeddine Fatnassi Ezzeddine Fatnassi 2 Typically, a software is one of the core component of any PRT system. It includes and supervises all the other PRT'components in order to deliver a high level of mobility for its users. Central control system for PRT manages the vehi- cles movements, the requested trips scheduling, the response of irregular opera- tions such as intrusion, accidents and so on. Consequently, implementing ecient softwares for PRT is of a high importance for such an intelligent transportation system. In this paper, we focus on implementing an optimization based software for PRT in order to eectively manage the empty vehicles management of PRT while reducing its total energy consumption. In fact for a on-demand transporta- tion service such as PRT, we could get up for a high number of empty vehicles moving to take passengers from dierent stations in the PRT'network which represents a high level of wasted transportation capacity. Therefore including optimization module within PRT'management software is of a high importance. In the literature, several operational optimization related studies to PRT were published such as dynamic routing [3], network design [19], simulation [5][11], optimized operational planning [6][9], energy minimization [18], total traveled distance [10][12][8], eet size[4] and so on. In this paper, we focus on the optimization problem related to PRT studied by Mrad and Hidri [18]. They proposed to study the routing electric vehicles with limited battery capacity in a static deterministic context. We propose a branch and bound resolution approach for that problem. We found that our branch and bound method outperforms the commercial solver Cplex' results1 for solving the PRT'optimization problem. An specic integration of our optimization approach within a PRT'software is also proposed. The remainder of this paper is organized as follows: Section 2 presents the formal problem denition. Section 3 develops our branch and bound optimization approach. Section 4 describes the computational results as well as the integration of our optimization approach within a PRT'software. Finally, Section 5 concludes the paper. 1 Introduction Personal Rapid Transit(PRT) is a relatively new mode of specic transporta- tion system. It falls under the automated guideway transit systems (AGT). In fact and as any classic AGT, PRT uses a set of automated vehicles which run on dedicated guideways. PRT is designed to move people in urban areas. PRT vehicles move people directly from origin station to destination station with a PRT'network of guideways without intermediate stops or transfers. The trans- portation service in PRT is done on-demand where one or a group of passengers ask for their transportation service. Typically, the main characteristics of PRT includes: i)Small electric driverless vehicles, ii) Direct origin-to-destination trans- portation service, iii) on demand transportation service and iv) Exclusive use of the PRT network by the PRT vehicles. In summary, the main characteristics of PRT includes the oer of a taxi- like transportation service by the use of driverless vehicles on a set of dedicated guideways. As the PRT is a complex, intelligent transportation service, it consists mainly on several advances components such as electric/electronic hardware, guideways, automated vehicles, stations, power sources and software[13]. 2 We also dene E′ = {(vi, vj); vi, vj ∈V ∗}. We also dene E′ = {(vi, vj); vi, vj ∈V ∗}. Note that G is a direct incomplete graph, because if the arc between node vi and node vj exists, the opposite arc does not exist. Our problem is a typical node routing problem that can be assimilated into the asymmetrical distance- constrained vehicle routing problem (ADCVRP). The main purpose in our prob- lem is to assign trips to vehicles with respect to the battery capacity of each vehicle in order to minimize the total consumption of electrical energy. An optimization based software tool for On-demand Transportation Systems An optimization based software tool for On-demand Transportation Systems 3 Let us dene a list of trips T that has cardinality \|T\| = n. Each trip i (= 1, . . . , n) will have its particular origin and arrival stations (OSi, ASi) and its particular departure and arrival times (OTi, ATi). We also suppose that the vehicle should return to the depot to charge its battery when necessary. The PRT problem is dened on an asymmetric graph G = {V , E}, where V = {v0, v1, v2, ˙,vn} is a set of nodes for which v0 denes the depot and v1, v2, ˙,vn denes the n dierent trips that the PRT system must cover. We dene V ∗= V /v0. Moreover, E = {(vi, vj); vi, vj ∈V } is a set of arcs that is dened as follows: If vi, vj ∈V ∗with ATi + Cost(ASi,OSj) ≤OTj, then the arc (i, j) exists and has cost cij, which represents the energy consumed from the arrival station of trip i (ASi) to the arrival station of trip j (ASj). Hence, each edge will have a combined cost that includes the cost of the movement from one trip to another and the cost of a trip itself. For each node i ∈V ∗we add an arc (0, i). The cost of this arc is c0i and represents the energy used to reach the arrival station of trip i from the depot. For each node i ∈V ∗we add an arc (i, 0). The cost of this arc is ci0 and represents the energy used to reach the depot from the arrival station of trip i. For each node i ∈V ∗we add an arc (i, 0). The cost of this arc is ci0 and represents the energy used to reach the depot from the arrival station of trip i. 2 Problem formulation We present in this section the problem formulation as presented in the work made by Mrad and Hidri[18]. The problem treated is based on the assumption that we have a predetermined list of trips to serve. Let us suppose that we have a set of PRT stations S, a depot D, and a PRT network that makes it possible to journey between any pair of stations. Let us also suppose that we have an unlimited number of vehicles that are initially located in the depot and have battery capacity B. Costi,j is a cost matrix that denes the cost of traveling from station i to station j. The cost of moving between station i and station j in the PRT network will be the length of the shortest path calculated using a FloydWarshall algorithm. 2.1 Flow-based mathematical formulation In this section, we adapt and present a ow-based mathematical formulation of our problem. This formulation was presented in [16] for the asymmetrical distance-constrained vehicle routing problem. We rst introduce the following integer variables: xij = {1 if node j is visited after node i, 0 otherwise, zij =    the shortest length traveled from the depot to customer j, as i is the predecessor of j 0 otherwise. zij =    the shortest length traveled from the depot to customer j, as i is the predecessor of j 0 otherwise. Let cij be the cost of the journey from node i to node j. Let δ+(i) is the set of edges that have vi as a root. Let δ−(i) is the set of edges that have vi as a sink. Ezzeddine Fatnassi PRT(1): Minimize ∑ (i,j)∈E cijxij (1) ∑ j∈δ+(i) xij = 1 ∀i ∈V ∗ (2) ∑ j∈δ−(i) xji = 1 ∀i ∈V ∗ (3) ∑ (i,j)∈E′ zij − ∑ (i,j)∈E′ zji − ∑ j∈V ∗ cijxij = 0 ∀i ∈V (4) zij ≤(B −cj0)xij ∀(i, j) ∈E′ (5) zij ≥(cij + c0i)xij ∀i ̸= 0, ∀(i, j) ∈E′ (6) z0i = c0ix0i ∀i ∈E′ (7) PRT(1): Minimize ∑ (i,j)∈E cijxij (1) ∑ j∈δ+(i) xij = 1 ∀i ∈V ∗ ∑ j∈δ−(i) xji = 1 ∀i ∈V ∗ ∑ (i,j)∈E′ zij − ∑ (i,j)∈E′ zji − ∑ j∈V ∗ cijxij = 0 ∀i ∈V (7) The objective (1) is to minimize the total charge used to make all the trips. Constraints (2) and (3) control the assignment of trips to routes. In fact, both constraints require that each node i ∈V ∗be visited just once. Constraint (4) ensures that the distance from node vi to any node vj by road is equal to the distance between the depot and node vj plus the distance from node vj to node vi. Constraint (5) ensures that the dierent routes are generated with respect to the battery capacity. In fact, it guarantees that the amount of charge the vehicle consumes to reach node vj from the depot is less than its battery capacity minus the charge needed for returning to the depot. 2.1 Flow-based mathematical formulation Moreover, in accordance with constraint (6), the total charge used to reach node vj from the depot is greater than or equal to the charge needed to follow the direct link between the depot and node vj. Finally, constraint (7) provides the initial value for z0i, which should be equal to the distance from the depot to node vi. 3 Branch-and-bound algorithm for static routing problem of personal rapid transit system For the general VRP, there exists a dierent method of solution that utilizes an exact algorithm. The reader is referred to [2] for a more extensive review of the exact method for the VRP. In this paper, we propose a branch-and-bound algorithm for solving the PRT problem. We present rst the general framework of our branch-and-bound procedure (B&B) [15][17][7]. The branch-and-bound algorithm is an exact method of nding optimal solutions for various distinctive optimization problems. This method was rst introduced by Land and Doig in 1960. Its main idea consists in enumerating all possible solutions in a search tree. The branch and bound rst solves a general global problem at the root of the search tree. Then, for all the other nodes the branch and bound will solve subproblems in which some constraints are relaxed and some variables are xed. An optimization based software tool for On-demand Transportation Systems An optimization based software tool for On-demand Transportation Systems 5 5 The branch and bound normally starts from a feasible initial solution given by an upper-bounding procedure. However, if we do not have a heuristic approach, the initial upper bound is set to innity. Once some constraints are relaxed, we need to dene a search strategy that consists of the main rules for choosing the next node to branch o and treat. There are two well-known strategies for branching: the depth-rst search (DFS), which consists in solving the most recently generated nodes rst; and the best-rst search, which consist in solving the most promising node rst [14]. Our B&B is presented in Algorithm 1. Based on Algorithm 1, there are several components that need to be dened such as the lower bounding procedure and the branching rules. In the next sections, we present details of the specic components of the B&B method. We should note that a solution S in Algorithm 1 represents a set of roads that cover each node in graph G exactly once. 3.1 Computation of lower bound for PRT problem The lower bound is one of the most important issues for the branch-and-bound algorithm. In fact, the branch and bound has to apply a lower-bounding proce- dure at each node of the search tree. The lower bound also has the advantage of giving a performance measure for heuristics. In the branch and bound, the lower bounds can be found by relaxing the mathematical formulation. There is a multitude of ways to relax these models. In our B&B, we decided to use the following linear model in order to generate the lower bounds. We will rst introduce the following integer variables xij = {1 if node of index j is visited after node of index i 0 Otherwise xij = {1 if node of index j is visited after node of index i 0 Otherwise xij = {1 if node of index j is visited after node of index i 0 Otherwise cij dene the cost of moving between nodes vi and vj in G. PRT: Min ∑ (i,j)∈E cijxij (8) ∑ j∈δ+(i) xij = 1∀i ∈V ∗ (9) ∑ j∈δ−(i) xji = 1∀i ∈V ∗ (10) xij ∈{0, 1}∀(i, j) ∈E (11) PRT: Min ∑ (i,j)∈E cijxij (8) (8) ∑ j∈δ+(i) xij = 1∀i ∈V ∗ (9) ∑ j∈δ−(i) xji = 1∀i ∈V ∗ (10) x ∈{0 1}∀(i j) ∈E (11) ∑ j∈δ+(i) xij = 1∀i ∈V ∗ ∑ j∈δ−(i) xji = 1∀i ∈V ∗ (10) xij ∈{0, 1}∀(i, j) ∈E (11) (11) In fact to generate the lower bounds, we decided to remove the battery con- straints from the mathematical model presented in [18] for solving the PRT prob- lem. The quality of this lower bound is not high, but with this lower-bounding scheme we obtain integer solutions. Solving the relaxed mathematical model, we obtained two types of cycles (routes): A cycle that respects the battery capacity of the PRT vehicles. A cycle that does not respect the battery capacity of the PRT vehicles which must be eliminated in future resolution of the relaxed linear program. A cycle that does not respect the battery capacity of the PRT vehicles which must be eliminated in future resolution of the relaxed linear program. 3.1 Computation of lower bound for PRT problem Ezzeddine Fatnassi 6 Algorithm 1 B&B 1: Compute Lower Bound of the problem 2: if Obtained Solution S Is Feasible then 3: Solution S is Optimal 4: Return Objective value of S 5: else 6: for all Infeasible Routes in Solution S do 7: Compute the ratio for each route 8: Choose the route with the largest ratio 9: for all Arcs in the chosen Route do 10: Relax the arc by putting its value to ∞ 11: if We get a feasible solution then 12: Update the Upper Bound 13: end if 14: Add Node to the tree 15: end for 16: end for 17: while ActualNode<MaxNodes do 18: Choose Subproblem with the smallest value of objective function 19: if There is more than one subproblem then 20: choose one randomly 21: end if 22: Calculate Lower Bound 23: for all Infeasible Routes in Solution S do 24: Compute the ratio for each route 25: Choose the route with the largest ratio 26: for all Arcs in the chosen Route do 27: Relax the arc by putting its value to ∞ 28: if We get a feasible solution then 29: Update the Upper Bound 30: end if 31: Add Node to the tree 32: end for 33: end for 34: if List of infeasible nodes is empty then 35: Return the upper Bound 36: end if 37: end while 38: end if 39: Return the upper Bound 3.2 Branching rules As mentioned in the previous section, solving of the lower bound can result in cycles that do not respect the battery capacity. Such a route is infeasible and thus should be eliminated (i.e., the cycle should not appear in any future relaxation of the problem). Accordingly, we take inspiration from previous work on solving either the asymmetric traveling salesman problem (ATSP) [14] or An optimization based software tool for On-demand Transportation Systems 7 7 the ADCVRP [1]. Normally, to eliminate an infeasible route, one arc should be removed. In our branch-and-bound method, we assign a large value to the selected arc that should be removed and then solve the relaxed program again. the ADCVRP [1]. Normally, to eliminate an infeasible route, one arc should be removed. In our branch-and-bound method, we assign a large value to the selected arc that should be removed and then solve the relaxed program again. Thus, in the future relaxations, the infeasible route will not reappear. The main disadvantage of this strategy is that the excluded edge may be present in the optimal solution, which can result in a failure of the method to nd the optimal solution. There are several ways to choose an arc to exclude from the infeasible route: There are several ways to choose an arc to exclude from the infeasible route: The rst way is to exclude the arc with the greatest cost. The second way is to exclude a random arc. A third way is to use the concept of the tolerance of an arc. This type of sensitivity operator will choose only one infeasible route to work on and branch from each time. Hence, for each infeasible route, the method will compute the ratio between the total electric charge used along the route and the number of arcs in that route. We then choose to work on the infeasible route with the largest ratio. For this route, we solve the dierent relaxation problems while eliminating one arc of the chosen route each time. We add to the tree the nodes that correspond to these dierent relaxations. In this paper, we choose to work with the third way as a branching rule for our branch-and-bound algorithm. 3.3 Example As in the previous iteration, we compute the ratio for each route: ρ(R2) = 43(Cost of the route R1) 5(number of arcs) = 8.6 (14) ρ(R3) = 33(Cost of the route R2) 4(number of arcs) = 8.25 (15) ρ(R2) = 43(Cost of the route R1) 5(number of arcs) = 8.6 (14) (14) ρ(R3) = 33(Cost of the route R2) 4(number of arcs) = 8.25 (15) (15) Hence, we decide to branch from route R1. We add additional node subprob- lems to the tree, where each corresponds to excluding one of the arcs (2, 4) (4, 6), and (6, 9) (see Figure 2).. The node counter is updated and increased by 3. The upper bound is not updated, since we again did not nd a feasible solution. This process is repeated until the maximum number of nodes is reached, the optimal solution is found, or the maximum running time is reached. 3.3 Example In this example, we explain the proposed B&B with the sensitivity operator. Sup- pose that we have 10 trips to serve and we have a battery capacity that permits the vehicle to run for 30 min. For simplicity, we assume that we have no upper bound. Therefore, the initial value of the upper bound is set to ∞. First, we solve the relaxed linear program. We obtain three routes R1 = (D, 0); (0, 3); (3, 7); (7, D), R2 = (D, 1); (1, 5); (5, 8); (8, 9); (9, D), and R3 = (D, 2); (2, 4); (4, 6); (6, D) of 28, 45, and 31 min, respectively. Therefore, we have two infeasible routes R2 and R3. For each of these two routes we calculate the associated ratio ρ as follows: ρ(R2) = 45(Cost of the route R1) 5(number of arcs) = 9 (12) ρ(R3) = 31(Cost of the route R2) 4( b f ) = 7.75 (13) ρ(R2) = 45(Cost of the route R1) 5(number of arcs) = 9 (12) (12) ρ(R3) = 31(Cost of the route R2) 4(number of arcs) = 7.75 (13) (13) The route R2 has the larger ratio (ρ(R2) = 9). Hence, we choose arcs from route R2 for branching. We add three nodes to the search tree, where each corresponds to solving the relaxed linear problem while excluding one arc (see Figure 1). The node counter is increased by 3. The upper bound is not updated, since we did not nd a feasible solution. We now have three subproblems on the tree, with costs 104, 107, and 105. We choose to branch from the node with the least cost (104). Solving the new ne Fatnassi 1 5 8 9 D 0 3 7 D 2 4 6 D 28 45 31 Infeasible Road 4 11 12 Infeasible Road 4 13 8-9 104 Fig. 1: First iteration of B&B 8 Ezzeddine Fatnassi Fig. 1: First iteration of B&B subproblem, we again have two infeasible routes R2 and R3 with costs 43 and 33. 4 Computational Results We have generated randomly 190 instances using the instances generator of Mrad and Hidri [18]. The size of the generated instances varies from 10 to 100 PRT'trips in steps of 5. For each trips'size, we generated 10 instances. Tests were made using a program coded in C++ and simulations are performed on a personal computer with Intel i5 2.3 GHZ processor and 6 GO of RAM(Windows 7). An optimization based software tool for On-demand Transportation Systems 9 9 2 4 6 9 D 0 3 7 D 1 5 8 D 28 43 33 Infeasible Road 4 13 12 Infeasible Road 4 11 104 8-9 1-5 104 105 5-8 107 2-4 4-6 6-9 105 108 107 Fig. 2: Second iteration of B&B Fig. 2: Second iteration of B&B We measure the performance of the tested algorithm by the GAP metric which is computed as follows: GAP = ((SOLB&B −LB) LB ) ∗100 (16) (16) SOLB&B represents the obtained solution for an instance using the B&B algorithm. The LB is the lower bound related to an instance of our problem. It is calculated based on the linear relaxation of the mathematical model given in [1]. All the mathematical models in this paper were coded using Cplex12.1. Results of our algorithm are proposed in Table 1. The B&B method founds an average GAP of 0.920 % in 356.717 seconds which could be described as good results. In fact and in many instances, the lower-bound technique used in our B&B can nd integer solutions that are very close to the optimal solution. Also one could note from table 1 that the average computation time for B&B increases at a reasonably low rate. We also compared the solutions obtained from B&B to those obtained by CPLEX for solving the valid mathematical model for the PRT problem [12]. Our B&B was able to nd better solutions than CPLEX in 30 instances, and the results were equal in 160 instances. We also performed more extensive statistical tests to evaluate the signicance of our results. In particular, we made a Wilcoxon matched-pairs signed-ranks test. This test yielded a P-value less than 0.0001 and showed that B&B is superior to Cplex. Therefore, we can state that B&B outperforms CPLEX. 4.1 Integration of the optimization module within a PRT software There is several ways to integrate optimization resolution techniques within in- telligent mobility softwares. In our PRT context and based on our problem, we propose a full integration procedure of our optimization technique which is presented in Algorithm 2. Algorithm 2 PRT software 1: Get the information related to the PRT network and the List of Trips 2: Construct the matrix Cost and Time. 3: Construct the graph G. 4: Use the B&B method to generate the optimal solution 5: The PRT control system uses the obtained data related to the dierent obtained roads from the B&B method to manage the PRT system. 4 Computational Results 10 Ezzeddine Fatnassi Table 1: Results of the B&B algorithm Table 1: Results of the B&B algorithm Number of trips Average Gap % Average Time (s) 10 0 0.217 15 0.565 76.056 20 0.980 151.140 25 0.980 151.140 30 0.837 302.896 35 1.893 480.634 40 0.337 309.723 45 0.867 362.849 50 0.781 481.219 55 0.976 541.620 60 0.387 363.099 65 0.638 439.567 70 0.898 500.468 75 2.237 423.149 80 0.988 484.330 85 1.112 492.855 90 0.557 364.977 95 1.051 305.209 100 1.398 546.469 Average 0.920 356.717 5 Conclusion In this paper, we proposed an optimization approach to minimizing the energy consumption for PRT. While focusing on a static problem related to PRT, a B&B resolution approach was proposed. This method adapted to the context of PRT was capable to solve various sized PRT test instances. In fact, our algorithm was shown to get a good quality results as it handles to nd an average gap of 0.920 in 356.717 seconds. As future works, we could implement new lower bounding procedures based on dierent type of relaxation. A branch and cut algorithm could also be devel- oped in order to speed up the performance of our algorithm. Also, one could note that the implementation of ecient new meta-heuristic and solution'representation seem to be very interesting in our context in order to enhance the solution qual- ity. New implemented meta-heuristic could also be implemented within our PRT software to solve large instances'size. Algorithm 2 PRT software 1: Get the information related to the PRT network and the List of Trip 1: Get the information related to the PRT netw 3: Construct the graph G. 4: Use the B&B method to generate the optimal solution 5: The PRT control system uses the obtained data related to the dierent obtained roads from the B&B method to manage the PRT system. Algorithm 2 includes several steps in order to manage eciently the PRT system. It starts by obtaining the related input such as the metric matrix related to the cost of moving between the dierent PRT stations and the information related to the static list of trips to serve. Next and based on these input data, the two matrix Cost and Time are constructed. Then, the optimization phase of our PRT software starts by building rst the graph G. The optimization phase then 11 n optimization based software tool for On-demand Tr An optimization based software tool for On-demand Transportation Systems ,and using the proposed B&B method, would solve the constructed problem using the obtained input data. The last step of our algorithm includes the transmission of the obtained roads representing the solution of the B&B method to the central PRT control system in order to route eciently the dierent vehicles available in the depot. References 1. Almoustafa, S., Hana, S., Mladenovi, N.: New exact method for large asymmetric distance-constrained vehicle routing problem. European Journal of Operational Research (2012) 2. Baldacci, R., Mingozzi, A., Roberti, R.: Recent exact algorithms for solving the vehicle routing problem under capacity and time window constraints. European Journal of Operational Research 218(1), 1 6 (2012), http://www.sciencedirect. com/science/article/pii/S0377221711006692 3. Chebbi, O., Chaouachi, J.: Modeling on-demand transit transportation system using an agent-based approach. In: Computer Information Systems and Industrial Management, pp. 316326. Springer (2015) 4. Chebbi, O., Chaouachi, J.: Optimal eet sizing of personal rapid transit system. In: Computer Information Systems and Industrial Management, pp. 327338. Springer (2015) 5. Chebbi, O., Chaouachi, J.: Proceedings of the Second International Afro-European Conference for Industrial Advancement AECIA 2015, chap. A Decentralized Man- agement Approach for On-Demand Transit Transportation System, pp. 175184. Springer International Publishing, Cham (2016), http://dx.doi.org/10.1007/ 978-3-319-29504-6_18 6. Chebbi, O., Chaouachi, J.: Reducing the wasted transportation capacity of per- sonal rapid transit systems: An integrated model and multi-objective optimiza- tion approach. Transportation Research Part E: Logistics and Transportation Re- view 89, 236 258 (2016), http://www.sciencedirect.com/science/article/ pii/S1366554515001647 Ezzeddine Fatnassi 12 7. Detienne, B., Sadykov, R., Tanaka, S.: The two-machine owshop total completion time problem: Branch-and-bound algorithms based on network-ow formulation. European Journal of Operational Research 252(3), 750 760 (2016), http://www. sciencedirect.com/science/article/pii/S0377221716300224 8. Fatnassi, E., Chaouachi, J.: Vns as an upper bound for an exact method to solve a class of on-demand transit transportation systems. Electronic Notes in Discrete Mathematics 47, 101108 (2015) 9. Fatnassi, E., Chaouachi, J., Klibi, W.: Planning and operating a shared goods and passengers on-demand rapid transit system for sustainable city-logistics. Trans- portation Research Part B: Methodological 81, 440460 (2015) 10. Fatnassi, E., Chebbi, O., Chaouachi, J.: Discrete honeybee mating optimiza- tion algorithm for the routing of battery-operated automated guidance electric vehicles in personal rapid transit systems. Swarm and Evolutionary Computa- tion 26, 35 49 (2016), http://www.sciencedirect.com/science/article/pii/ S2210650215000619 11. Fatnassi, E., Chebbi, O., Siala, J.C.: Two strategies for real time empty vehicle redistribution for the personal rapid transit system. In: Intelligent Transporta- tion Systems-(ITSC), 2013 16th International IEEE Conference on. pp. 18881893. IEEE (2013) 12. Fatnassi, E., Chebbi, O., Siala, J.C.: Comparison of two mathematical formulations for the oine routing of personal rapid transit system vehicles. In: The Interna- tional Conference on Methods and Models in Automation and Robotics (2014) 13. References Furman, B., Fabian, L., Ellis, S., Muller, P., Swenson, R.: Automated transit net- works (atn): A review of the state of the industry and prospects for the future. Tech. rep. (2014) 14. Germs, R., Goldengorin, B., Turkensteen, M.: Lower tolerance-based branch and bound algorithms for the {ATSP}. Computers & Operations Research 39(2), 291 298 (2012), http://www.sciencedirect.com/science/article/pii/ S0305054811000980 15. Kadri, A.A., Kacem, I., Labadi, K.: A branch-and-bound algorithm for solving the static rebalancing problem in bicycle-sharing systems. Computers & Industrial En- gineering 95, 41 52 (2016), http://www.sciencedirect.com/science/article/ pii/S0360835216300183 16. Kara, I.: Two indexed polonomyal size formulationsfor vehicle routing problems. Technical Report. Baskent University, Ankara/Turkey (2008) 17. Lee, W.C., Wang, J.Y., Lin, M.C.: A branch-and-bound algorithm for mini- mizing the total weighted completion time on parallel identical machines with two competing agents. Knowledge-Based Systems 105, 68 82 (2016), http: //www.sciencedirect.com/science/article/pii/S0950705116301010 18. Mrad, M., Hidri, L.: Optimal consumed electric energy while sequencing vehicle trips in a personal rapid transit transportation system. Computers & Industrial Engineering 79, 19 (2015) 19. Zheng, H., Peeta, S.: Network design for personal rapid transit under transit- oriented development. Transportation Research Part C: Emerging Technologies 55, 351362 (2015)
https://openalex.org/W4307735975	https://zenodo.org/record/7267171/files/37.%20-%20Priti%20Sivaramakrishanan%20%26%20Dr.%20Sybil%20-%20Jan-Feb%202021-12.pdf	English	null	GLOBAL CITIZENSHIP: ASPIRING FOR A WORLD OF PROSPERITY AND WELL BEING	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	6,124	Abstract: Abstract: The drive for globalisation has contributed to the emergence of global citizens who must work together and stand as one. Citizens who are a part of this globalisation process are not just citizens; rather, they are global citizens who have contributed to the world's various cultures coming together and seeming as one huge family. Despite the fact that the globalisation movement became more visible in the 1990s, the idea of global citizenship has existed in various forms since ancient times and may be found throughout history. With the advancement of technology and economic development there exist a paradoxical pattern that the world is diverse yet closely knit. One has to understand and behave in a socially just manner while dealing with a multi-diverse world or population. Justice is a basic requirement of life similar to the air we breathe, an essential element for living. Pre-service teachers play a pivotal role in shaping the future of students as they progress through the education system which becomes the bedrock for instilling values. It is therefore essential for pre-service teachers to be familiar with global citizenship and social justice. This study attempts to measure pre-service teachers perception towards global citizenship and social justice. The sample for the study consisted of 1054 pre service teachers of aided and unaided institutes. Both aided and unaided teacher training institutes play a major role in the formation of teachers fostering the value system in the lives of future generations. Hence this study considered it important to measure perception of pre service teachers towards global citizenship and social justice of aided and unaided institutes of Mumbai. Along with the responses administered from the tool, focused group interview was also conducted to validate the findings obtained through the quantitative responses. The findings of the study indicate that pre service teachers of aided institutes had a better perception of social justice and pre service teachers of unaided institutes had a better perception of global citizenship. Copyright © 2022 The Author(s): This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC BY-NC 4.0) which permits unrestricted use, distribution, and reproduction in any medium for non-commercial use provided the original author and source are credited. ISSN–2278-5655 ISSN–2278-5655 AMIERJ Aarhat Multidisciplinary International Education Research Journal Priti Sivaramakrishanan & Dr. Sybil Thomas Assistant Professor, St. Xavier’s Institute of Education ** Professor, Department of Education, University of Mumbai. * Assistant Professor, St. Xavier’s Institute of Education ** Professor, Department of Education, University of Mumbai Global Citizenship People abroad adopt the unique culture and traditions of India. The people of India are simultaneously acquiring new eating habits, traditions, and foreign languages. Whether it be ayurvedic procedures or ikega, SJIF Impact Factor 7.372 SJIF Impact Factor 7.372 Peer Reviewed Journal 265 ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Aarhat Multidisciplinary International Education Research Journal Original Research Article the Japanese culture has spread worldwide. The exchange of information throughout the globe is what has caused this. The globalization trend has led to this information exchange. The spread of ideas, knowledge, information, goods, and services has been aided by globalisation. Such knowledge, information, and idea exchange has aided in the global community's closer bonding as a single family. Globalization has thus helped to develop global citizens, who must come together and bond. This bonding as a single family is very much needed to create global citizens. This concept of boding together as one family has been expressed in the Sanskrit term ‘Vasudhaiva Kutumbakam’, has the meaning of "the world is one family". The above mentioned phrase is a reflection that the whole world has to live together in peace and harmony like a family. Human beings on earth are not just citizens of the state, region or county they belong to but are a part of the big world. The globalization movement has definitely helped the world. the Japanese culture has spread worldwide. The exchange of information throughout the globe is what has caused this. The globalization trend has led to this information exchange. The spread of ideas, knowledge, information, goods, and services has been aided by globalisation. Such knowledge, information, and idea exchange has aided in the global community's closer bonding as a single family. Globalization has thus helped to develop global citizens, who must come together and bond. This bonding as a single family is very much needed to create global citizens. This concept of boding together as one family has been expressed in the Sanskrit term ‘Vasudhaiva Kutumbakam’, has the meaning of "the world is one family". The above mentioned phrase is a reflection that the whole world has to live together in peace and harmony like a family. Human beings on earth are not just citizens of the state, region or county they belong to but are a part of the big world. Global Citizenship The globalization movement has definitely helped the world. In its literal sense, ‘Global’ means relating to the whole world and ‘Citizenship’ means belonging to a country. ‘Global citizenship’ therefore means one is a citizen of the world who adopts and understands communities across the world and is a part of the larger world community. According to UNESCO (2014) global citizenship refers to a sense of belongingness to a broader community and common humanity. According to Oxfam (2006) global citizen is one who is aware of the wider world and extends his needs to be exposed to and aware of the cultural and linguistic diversities across the globe. According to Oxfam (2006) global citizenship is one who displays the following char ding to Oxfam (2006) global citizenship is one who displays the following characteristic • is aware of the wider world and has a sense of their own role as a world citizen; • respects and values diversity; • has an understanding of how the world works economically, politically, socially, culturally, technologically and environmentally; • is outraged by social injustice; ticipates in and contributes to the community at a range of levels from local to global; • is willing to act to make the world a more sustainable place; • is willing to act to make the world a more sustainable place; • takes responsibility for their actions A citizen who tries to adopt the global citizenship values listed above has to behave in a just manner for himself and for others. A global citizen is aware of the different events and situations happening in the world and strives to make the world better through one’s own action and by fighting and striving for social justice. Social Justice Social Justice Each of these objectives is somehow related to the behaviours and conduct that people engage in on a daily basis. The seventeen goals are all connected to ideas of justice in some form or the other. For example goal 1 to ensure no poverty, goal 2 zero hunger goal 3 good health and well- being, goal 7 affordable and clean energy, goal 4 quality education. If these goals are to be achieved it is important that as a world everybody fights for what is unjust with an aim to provide resources that would help one to lead a dignified life. According to John Rawl social justice is based on the principle of equal liberty where each one has the privileges to enjoy the basic liberties of conscience, expression, association and democratic rights. The second principle is the principle of equality where the least advantaged members of society should be able to enjoy the different opportunities available in the society. According to Bradley’s theory, self- realization is essential and will enable individuals to act morally and become good in every situation which would help making the world morally better place. According to Rebore justice is a guide that organizes how people will continue their lives as members of a society. The pursuit of social justice, according to Singh (2011), is the equitable allocation of resources-both monetary and intangible-that are useful and valued. An analysis of the above definitions of justice clearly indicates that justice believes in equal distribution of resources, by providing each one with the desired opportunities through self realization. Social Justice Quote by Mahatma Gandhi clearly states that “Education should be so revolutionized as to answer the wants of the poorest villager, instead of answering those of an imperial exploiter.” Justice is a basic element of life similar to the air we breathe, which is very essential and helps the needs of the most wanted. With these thoughts in mind the sustainable development goals have listed 17 goals that need to be achieved by 2030. We are presently in year 2022 and still 8 mor SJIF Impact Factor 7.372 262 Peer Reviewed Journal ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Aarhat Multidisciplinary International Education Research Journal Original Research Article Original Research Article Original Research Article with the pandemic hitting us hard. Each of these objectives is somehow related to the behaviours and conduct that people engage in on a daily basis. The seventeen goals are all connected to ideas of justice in some form or the other. For example goal 1 to ensure no poverty, goal 2 zero hunger goal 3 good health and well- being, goal 7 affordable and clean energy, goal 4 quality education. If these goals are to be achieved it is important that as a world everybody fights for what is unjust with an aim to provide resources that would help one to lead a dignified life. According to John Rawl social justice is based on the principle of equal liberty where each one has the privileges to enjoy the basic liberties of conscience, expression, association and democratic rights. The second principle is the principle of equality where the least advantaged members of society should be able to enjoy the different opportunities available in the society. According to Bradley’s theory, self- realization is essential and will enable individuals to act morally and become good in every situation which would help making the world morally better place. According to Rebore justice is a guide that organizes how people will continue their lives as members of a society. The pursuit of social justice, according to Singh (2011), is the equitable allocation of resources-both monetary and intangible-that are useful and valued. An analysis of the above definitions of justice clearly indicates that justice believes in equal distribution of resources, by providing each one with the desired opportunities through self realization. with the pandemic hitting us hard. Global Citizenship and Social Justice Connect Global Citizenship and Social Justice Connect Global citizenship has been prevalent in various forms since ancient times, traces of which could be seen throughout history. During the Greco – Roman period (Dower, 2002) it was believed that human beings belonged to a more fundamental community and were bounded by duties and values to a larger moral community. Hindu philosophy always promoted that the world is one family in the famous quote ‘Vasudhaiva Kutumbakam’. According to Oxfam, (1997) a global citizen is one who is aware of the wider world and has a role to play in building a peaceful, sustainable and fairer planet. With the advancement of technology and economic development there exist a paradoxical pattern of the world being diverse yet closely connected to each individual in the world e.g. global markets where trades are effected across continents and countries yet there is a boundary of individual countries forming their own rules and laws to effect such trade. Music has been a core influence in breaking the country and continental barriers, music today is shared world over and there are various medium which promote international music to be available even in the remotest places of the world. Sports has been another major factor which is cheered and celebrated across the world be it the Olympics or any other global event which brings together a multi-diverse population to complete and exchange ideas as well as success. Diversity of thoughts and actions comes with its own challenges and conflicts. Resolving such global challenges and conflicts broods the concept of Social Justice which means One has to understand and behave in a socially just manner while dealing with a multi-diverse world or population. Justice is a basic requirement of life similar to the air we breathe, an essential element for living. According to Socrates justice is construed SJIF Impact Factor 7.372 263 Peer Reviewed Journal ISSN–2278-5655 Aarhat Multidisciplinary International Education Research Journal Original Research Article to mean the well-being of an individual by giving each person their due. Global citizenship promotes the well- being of each individual and social justice ensures a way to achieve this well-being. Global citizenship and Social justice complement each other in order to ensure peace and harmony in the world. Teachers play a pivotal role in shaping the future of students as they progress through the education system which becomes the bedrock to instilling values. Need for the study The National Education Policy 2020 recommends that global citizenship values be instilled among students of higher education and school education. If we are to create future children with a global mindset it is essential to know pre service teachers perceptions towards global citizenship. Creating future generations with a global frame or mind requires one to speak and voice out for justice for the well being of self and society. With the pandemic hitting us hard and having disturbed the well being and peace throughout the world it is very important that we instill among children to view and analyze everything with a global perspective. The promotion of global citizenship and social justice go hand in hand, therefore this study will enable us to ascertain the pre-service teachers' capacity to think globally and with a focus on social justice. Both aided and unaided teacher training institutes of Mumbai go a big way to shape and create aspiring teachers. Hence this study is very important to know about pre service teacher’s perceptions towards global citizenship and social justice. The results of this study will provide teacher education institutions of both aided and unaided institutes an insight into how pre-service teachers perceive themselves in relation to the world and others around them, as well as what needs to be done to support them in analyzing their actions for the benefit of the world's well- being. This research will also assist the many stakeholders in the education system in creating curricula that encourage social justice while fostering a sense of global citizenship. Global Citizenship and Social Justice Connect Thus it becomes essential for pre-service teachers to be familiarized with the linkages between global citizenship and social justice. Teachers play a pivotal role in shaping the future of students as they progress through the education system which becomes the bedrock to instilling values. Thus it becomes essential for pre-service teachers to be familiarized with the linkages between global citizenship and social justice. Hypothesis for the study 1. There is no significant difference in the perception of global citizenship of pre-service teachers on the basis of the type of Institution. 2. There is no significant difference in the perception of social justice of pre-service teachers on the basis of the type of Institution. Tools for the study Global Citizenship Scale Morais and Ogden (2010) Objectives of the Study 1. To study pre-service teachers’ perception towards global citizenship on the basis of ty 1. To study pre-service teachers’ perception towards global citizenship on the basis of type of institution. 2. To study pre-service teachers’ perception towards social justice on the basis of type of institution. H th i f th t d 1. To study pre-service teachers’ perception towards global citizenship on the basis of type of institution. 2. To study pre-service teachers’ perception towards social justice on the basis of type of institution. Hypothesis for the study Conclusion There is a significant difference in the global citizenship scores of pre service teachers on the basis of the type of institutions. The level of significance is at 0.05 level and the global citizenship mean scores of unaided institutions are more than aided institutions. 0.40% variance in global citizenship scores is associated with the type of institutions. Analysis of the study Hypothesis 1: There is no significant different in the perception of global citizenship of pre-service teachers on the basis of the type of Institution Table 1 Relevant statistics for ‘t’ test for global citizenship scores of pre service teachers on the basis of type of institutions Variable Group N Mean S.D. Df Tabulated Value ‘t’ ratio I.os. 100 W2 test Global Citizenship Aided 975 100.85 14.78 1054 At 0.01 level = 2.58 -2.31 do S manually at 0.05 level 0.40% Unaided 79 102.93 14.44 At 0.05 level 1.96 Interpretation of ‘t’ test The obtained value of ‘t’ ratio for the global citizenship scores of pre service teachers on the basis of type of institution is -2.31 which is more than the table value 1.96. Thus ‘t’ ratio is significant. Hence the null hypothesis is rejected. Table 1 Relevant statistics for ‘t’ test for global citizenship scores of pre service teachers on the basis of type of institutions Variable Group N Mean S.D. Df Tabulated Value ‘t’ ratio I.os. 100 W2 test Global Citizenship Aided 975 100.85 14.78 1054 At 0.01 level = 2.58 -2.31 do S manually at 0.05 level 0.40% Unaided 79 102.93 14.44 At 0.05 level 1.96 I t t ti f ‘t’ t t Interpretation of ‘t’ test The obtained value of ‘t’ ratio for the global citizenship scores of pre service teachers on the basis of type of institution is -2.31 which is more than the table value 1.96. Thus ‘t’ ratio is significant. Hence the null hypothesis is rejected. The obtained value of ‘t’ ratio for the global citizenship scores of pre service teachers on the basis of type of institution is -2.31 which is more than the table value 1.96. Thus ‘t’ ratio is significant. Hence the null hypothesis is rejected. Social Justice Scale Torres-Harding, S.R., Siers, B., & Olson, B. (2012) Pre service teacher’s perception towards social justice was measured with a readymade tool by Torres – Harding. Pre service teacher’s perception towards social justice was measured against four dimensions social justice attitude subscale, subjective norms around social justice, perceived behaviour control around social justice and intentions to engage in social justice. Global Citizenship Scale Morais and Ogden (2010) A readymade tool by Morais and Ogden (2010) was used to administer the perception of pre service teachers about global citizenship. The tool measured three dimensions of global citizenship social responsibility, global A readymade tool by Morais and Ogden (2010) was used to administer the perception of pre service teachers about global citizenship. The tool measured three dimensions of global citizenship social responsibility, global SJIF Impact Factor 7.372 264 Peer Reviewed Journal ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Aarhat Multidisciplinary International Education Research Journal competence and global civic engagement. competence and global civic engagement. Social Justice Scale Torres-Harding, S.R., Siers, B., & Olson, B. (2012) Discussion It is also possible that pre service teachers of unaided B.Ed institutions are more informed about their current global events (based on the tool) And can express their views about global issues using social media by expressing their views about international politics, global problems on a website, blog or chat room (based on the tool). It could also be that pre service teachers of unaided B.Ed institutions would me more interested in helping individuals and communities in solving global humanitarian problems, global environment, social or political problems (based on the tool). Pre service teachers of unaided institutions would also believe in purchasing products and brands that are grown by marginalized people and help them act as a global citizen (based on the tool). It may be because of these reasons that pre service teachers of unaided B.Ed institutions showed a better perception towards global citizenship than pre service teachers of aided B.Ed institutions (based on the tool). Hypothesis 2: There is no significant difference in the perception of social justice of pre-service teachers on the basis of the type of Institution Original Research Article Original Research Article be questioned and discussed which should be based on theory and logic. In the same study it was found that pre service teacher’s come from varied background and their knowledge of diversity and global issues was found to be limited (Reimer & McLean, 2005). In one more study it was found that the kind of teaching and learning strategies that are used to teach about global citizenship is very uncertain (Weber, et. Al, 2013). It another study it was also found that elementary teacher candidates felt it was important to teach about global issues but were less confident to teach about global aspects due to the lack of knowledge (An,2014). Probably the quality of programs promoting the ideas of global citizenship, pre service teachers knowledge regarding global issues and the learning experience used in teaching global citizenship would be a reason as to why there existed differences in the perception towards global citizenship scores of pre service teachers of aided and unaided colleges. Excerpts of the focus group interview did not reveal anything substantiate regarding the difference in the perception of global citizenship between pre service teachers of aided and unaided institutions. Discussion The mean global citizenship scores of unaided institutions are more than the mean scores of aided institutions. The pre service teachers of aided and unaided B.Ed colleges from whom the sample were collected, studied the two year B.Ed. syllabus have learnt concepts like globalization, global perspectives, global culture and global citizenship in the different courses. Global citizenship is still an emerging concept in higher education systems in India. Though the term global citizenship may be understood by pre service teachers and teacher educators, training pre service teachers about global citizenship would require coming up with programs that are objective and purposeful. The quality of programs organized to sensitize about global citizenship needs to 265 Peer Reviewed Journal ISSN–2278-5655 Aarhat Multidisciplinary International Education Research Journal Volume–XI, Issues– I Jan –Feb 2022 Journal Original Research Article be questioned and discussed which should be based on theory and logic. In the same study it was found that pre service teacher’s come from varied background and their knowledge of diversity and global issues was found to be limited (Reimer & McLean, 2005). In one more study it was found that the kind of teaching and learning strategies that are used to teach about global citizenship is very uncertain (Weber, et. Al, 2013). It another study it was also found that elementary teacher candidates felt it was important to teach about global issues but were less confident to teach about global aspects due to the lack of knowledge (An,2014). Probably the quality of programs promoting the ideas of global citizenship, pre service teachers knowledge regarding global issues and the learning experience used in teaching global citizenship would be a reason as to why there existed differences in the perception towards global citizenship scores of pre service teachers of aided and unaided colleges. Excerpts of the focus group interview did not reveal anything substantiate regarding the difference in the perception of global citizenship between pre service teachers of aided and unaided institutions. The perception towards global citizenship of pre service teachers of unaided institutions could be more because they could unconsciously adapt their behaviour and mannerisms, communication, strategies, when interacting with people from other cultures (based on the tool). Discussion The mean scores of aided pre service teachers are more than the mean scores of unaided pre service teachers. It may be possible that pre service teachers of aided institutions would be more proactive to talk to others about power, privilege and oppression (based on the tool). They would ensure that all individuals and groups especially the traditionally and marginalized have a chance to speak and be heard (based on the tool) they would allow individuals and groups to define and describe their problems and experiences (based on the tool). would allow individuals and groups to define and describe their problems and experiences (based on the tool). They would also be more keen to help individuals and groups to pursue and promote physical and emotional well – being of individuals and groups (based on the tool). They would help individuals and groups to pursue their chose goals in life (based on the tool). They would promote fair and equitable allocation of bargaining powers, obligations and resources in the society (based on the tool). Pre service teachers of aided institutions would support community organizations’ and institutions that would help individuals and groups to achieve their aims (based on the tool). It may also be due to the fact the pre service teachers of aided institutions would engage in acts of social justice (based on the tool). They would be confident to create a positive impact on other’s lives (based on the tool) They would possess an ability to work with individuals and groups in ways that are empowering (based on the tool). Apart from they being passionate to promote justice pre service teachers of aided institutions would also be capable to influence others to promote fairness and equality (based on the tool). They would ensure that people around them are engaged in activities through dialogue that address social justice issues (based on the tool). They would believe in motivating people to be supportive of efforts that promote social justice (based on the tool). And They would probably believe in their self to have a positive impact on the community (based on the tool). They would also be confident in their ability to talk to others about social injustices and the impact of social conditions on health and well-being (based on the tool). It was also seen that discourses in family, friends, education and travel mattered while conceptualizing social justice (Philpott, 2009). Discussion ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article I t t ti f ‘t ‘t t Aarhat Multidisciplinary International Education Research Journal Interpretation of ‘t-‘test The obtained value of ‘t’ ratio for the social justice scores of pre service teachers on the basis of type of institutions is 3.76 which is more than the table value. Thus ‘t’ ratio is significant. Hence the null hypothesis is rejected. Conclusion There is significant difference in the social justice scores of pre service teachers on the basis of type of institutions. The level of significance is 3.76 at 0.05 level and the mean scores of social justice scores of pre service teachers of unaided institutions are more than the pre service teachers of aided institutions. 1.23% variance in social justice scores is associated with the type of institutions. Discussion The perception towards global citizenship of pre service teachers of unaided institutions could be more because they could unconsciously adapt their behaviour and mannerisms, communication, strategies, when interacting with people from other cultures (based on the tool). It is also possible that pre service teachers of unaided B.Ed institutions are more informed about their current global events (based on the tool) And can express their views about global issues using social media by expressing their views about international politics, global problems on a website, blog or chat room (based on the tool). It could also be that pre service teachers of unaided B.Ed institutions would me more interested in helping individuals and communities in solving global humanitarian problems, global environment, social or political problems (based on the tool). Pre service teachers of unaided institutions would also believe in purchasing products and brands that are grown by marginalized people and help them act as a global citizen (based on the tool). It may be because of these reasons that pre service teachers of unaided B.Ed institutions showed a better perception towards global citizenship than pre service teachers of aided B.Ed institutions (based on the tool). H th i 2 Th i i ifi t diff i th ti f i l j ti f i t h Hypothesis 2: There is no significant difference in the perception of social justice of pre-service teachers on the basis of the type of Institution Table 2 Relevant statistics for ‘t’ test for SJ scores of pre service teachers on the basis of type of institutions Variable Group N Mean S.D. Df Tabulated Value ‘t’ ratio I.os. 100 test Social Justice Aided 975 143.25 17.32 1054 At 0.01 level = 2.58 3.76 S at 0.05 & 0.01 level 1.23% Unaided 79 138.98 19.09 At 0.05 level 1.96 SJIF Impact Factor 7.372 Peer Reviewed Journal 266 ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Interpretation of ‘t-‘test The obtained value of ‘t’ ratio for the social justice scores of pre service teachers on the basis of type of institutions is 3.76 which is more than the table value. Thus ‘t’ ratio is significant. Hence the null hypothesis is rejected. Gender cell at college conducted activities like talk show that promoted and supported the rights of both genders. It may be because of these reasons that pre service teachers of aided institutions have a better perception towards social justice than pre service teachers of unaided institutions. Participant K p Topics like discrimination, stereotyping, self esteem were taught and discussed in the graduation subject psychology and hence in the B.Ed course could relate, comprehend and discuss more about be more positive while being taught in the B.Ed course. p Topics like discrimination, stereotyping, self esteem were taught and discussed in the graduation subject psychology and hence in the B.Ed course could relate, comprehend and discuss more about be more positive while being taught in the B.Ed course. Participant N Issues on tolerance, brotherhood, secularism were discussed that helped us to think of equality and equity. Participant N Issues on tolerance, brotherhood, secularism were discussed that helped us to think of equality and equity. Participant L In the family often discussions about opposing caste system and encouraging inter caste relations was promoted. While travelling to Kashmir and South India I did experience diversity of the people through their language, culture and people which helped me to break stereotypes. There was also one particular incident where before the unification of Kashmir and India, people in Kashmir were very apprehensive about the travelers where I felt that justice for the Kashmiris’ was lacking through the people’s behvaiour. Participant L In the family often discussions about opposing caste system and encouraging inter caste relations was promoted. While travelling to Kashmir and South India I did experience diversity of the people through their language, culture and people which helped me to break stereotypes. There was also one particular incident where before the unification of Kashmir and India, people in Kashmir were very apprehensive about the travelers where I felt that justice for the Kashmiris’ was lacking through the people’s behvaiour. Most colleges now have a gender cell, where the institutions ensure that well-being of both male and female pre service teachers are taken care of. It may be possible that in the aided institutions the gender cell would be a functioning cell and programs might be conducted to ensure well-being of both genders. It may be because of this also that pre service teachers of aided institutions scored more in their perception and feelings towards justice. This is supported through the focused group interview of a respondent from aided institutions Participant K Gender cell at college conducted activities like talk show that promoted and supported the rights of both genders. Discussion This could also be a strong reason as to why the pre service teacher’s perception towards social justice is high due to personal discourses and relationships Pre service teachers of aided It was also seen that discourses in family, friends, education and travel mattered while conceptualizing social justice (Philpott, 2009). This could also be a strong reason as to why the pre service teacher’s perception towards social justice is high due to personal discourses and relationships. Pre service teachers of aided institutions would probably engage more in travelling and would be more alert to observe about things and obably engage more in travelling and would be more alert to observe about things and SJIF Impact Factor 7.372 SJIF Impact Factor 7.372 Peer Reviewed Journal 267 ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Aarhat Multidisciplinary International Education Research Journal Original Research Article issues that exist in the society. Probably in families of aided pre service teachers it would have been possible that more discourses or discussion about social justice would have occurred with the family. It may be possible that the subjects that had studied during their graduate or post graduation level would also have helped to comprehend and work in action towards justice. This is supported through the focused group interview of a respondent from aided institutions. Conclusion The study indicates that pre service teachers of aided institutes have better perception towards social justice and pre service teachers of unaided institutes have better perception towards global citizenship. From the study we can conclude that the perception of pre service teachers irrespective of aided or unaided institute needs to be built up and increased both towards global citizenship and social justice. With the four year B.Ed program SJIF Impact Factor 7.372 268 Peer Reviewed Journal ISSN–2278-5655 AMIERJ Volume–XI, Issues– I Jan –Feb 2022 Aarhat Multidisciplinary International Education Research Journal Original Research Article Aarhat Multidisciplinary International Education Research Journal Original Research Article to be integrated soon in colleges of Mumbai it is essential that at different levels of B.Ed education system programs are arranged to sensitize ideas and thoughts about global citizenship and social justice. Aided and unaided B.Ed colleges in Mumbai should aim at organizing seminars, webinars and workshops on themes evolving around social justice and global citizenship. This is very crucial if we are aiming for the future generations to be globally competent and act with the lens of justice. to be integrated soon in colleges of Mumbai it is essential that at different levels of B.Ed education system programs are arranged to sensitize ideas and thoughts about global citizenship and social justice. Aided and unaided B.Ed colleges in Mumbai should aim at organizing seminars, webinars and workshops on themes evolving around social justice and global citizenship. This is very crucial if we are aiming for the future generations to be globally competent and act with the lens of justice. g g y p j References References 1. UNESCO (2014). Global citizenship education: Preparing learners for the challenges of the 21st century http://unesdoc.unesco.org/images/0022/002277/227729e.pdf 2. Oxfam. Qualities of a Global Citizen. Retrieved April 6, 2011 from http://www.edu.gov.mb.ca/k12/cur/socstud/foundation_gr9/blms/9-2-4d.pdf 2. Oxfam. Qualities of a Global Citizen. Retrieved April 6, 2011 from http://www.edu.gov.mb.ca/k12/cur/socstud/foundation_gr9/blms/9-2-4d.p 3. Mahatma Gandhi quotes on Education \| Gandhi & Education (mkgandhi.org) https://unstats.un.org/sdgs/report/2022/The-Sustainable-Development-Goals-Report-2022.pdf 4. Rebore RW (2001). The Ethics of Educational Leadership. Prentice-Hall Inc, New Jersey 5. Dower, N. (2022). An introduction to global citizenship. In An Introduction to Global Citizenship. Edinburgh University Press 6. Sachi R: Vasudhaiva Kutumbakam - Original shloka in Maha Upanishad (rsachi.blogspot.com) 7. Oxfam, Op.Cit 6. Sachi R: Vasudhaiva Kutumbakam - Original shloka in Maha Upanishad (rsachi.blogspot.com) 7. Oxfam, Op.Cit 8. University of Mumbai. (201-2018). Syllabus for two years bachelor of education programme. Retrieved from B-Ed-2-Years-Syllabus.pdf (sscoe.edu.in) 9. Reimer, K., & McLean, L. R. (2009). Conceptual Clarity and Connections: Global Education and Teacher Candidates. Canadian Journal of Education, 32(4), 903-926. 9. Reimer, K., & McLean, L. R. (2009). Conceptual Clarity and Connections: Global Education and Teacher Candidates. Canadian Journal of Education, 32(4), 903-926. 10. Weber, N., Evans, M., MacDonald, A., & Ingram, L. A. (2013). Educating for the global dimension of citizenship in Canadian schools: A snapshot of teachers’ understandings and practices. Education Review, 3(2), 12-13 10. Weber, N., Evans, M., MacDonald, A., & Ingram, L. A. (2013). Educating for the global dimension of citizenship in Canadian schools: A snapshot of teachers’ understandings and practices. Education Review, 3(2), 12-13 11. An, S. (2014). Preparing elementary teachers as global citizenship educators. Journal of Education, 194(3), 25-38 11. An, S. (2014). Preparing elementary teachers as global citizenship educators. Journal of Education, 194(3), 25-38 12. Philpott, R. J. (2009). Exploring new teachers' understandings and practice of social justice education (Doctoral dissertation, Faculty of Education-Simon Fraser University). 12. Philpott, R. J. (2009). Exploring new teachers' understandings and practice of social justice education (Doctoral dissertation, Faculty of Education-Simon Fraser University). Peer Reviewed Journal SJIF Impact Factor 7.372 269
https://openalex.org/W4308103136	https://www.researchsquare.com/article/rs-1619477/latest.pdf	English	null	QSRR modeling of the chromatographic retention behavior of some quinolone and sulfonamide antibacterial agents using firefly algorithm coupled to support vector machine	BMC chemistry	2,022	cc-by	8,636	QSRR modeling of the chromatographic retention behavior for some quinolone and sulfonamide antibacterial agents using firefly algorithm coupled to support vector machine support vector machine Marwa A. Fouad (  marwa.fouad@pharma.cu.edu.eg ) Cairo University Enas H. Tolba National Organization for Drug Control and Research (NODCAR) Manal A. El-Shal National Organization for Drug Control and Research (NODCAR) Ahmed Serag Al-Azhar University Ahmed M. El Kerdawy Cairo University Research Article Keywords: Quinolones, Sulfonamides, Quantitative structure-retention relationship, Firefly algorithm, Support vector machine, y-randomization Posted Date: May 19th, 2022 DOI: https://doi.org/10.21203/rs.3.rs-1619477/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Research Article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/26 Page 1/26 Page 1/26 Page 1/26 Abstract Quinolone and sulfonamide are two classes of antibacterial agents that have an opulent history of medicinal chemistry features responsible for their improving bacterial spectrum, efficacy, pharmacokinetics, and adverse side effect profiles. The urgent need of their use and escalating rate of their resistance provokes the necessity for developing suitable analytical methods that speed up and facilitate their analysis. In this study, advanced firefly algorithm (FFA) coupled with support vector machine (SVM) were used to select the most significant descriptors and to construct two separate quantitative structure–retention relationship (QSRR) models using a series of 11 selected quinolones and 13 sulfonamide drugs, separately, in order to predict their retention factors in HPLC. Precisely, the effect of different pH range values and acetonitrile composition in the mobile phase on the retention behavior of quinolones and sulfonamides were studied, respectively. The obtained QSRR models showed high performance in both internal and external validation indicating their robustness and predictive ability. Y-randomization validation displayed that the obtained models are not obtained by chance. Besides, the obtained results shed the light on the molecular features that influence the retention behavior of these two classes under the current chromatographic conditions. 1. Introduction Such relationship provides a powerful alternative to the conventional trial-and-error approach with marked improvement in time and cost of experiments. In these mathematical models, a link between compounds’ chemical structures represented by their descriptors and the retention data in different chromatographic systems is built. The number of molecular descriptors that could be obtained for one analyte is enormous where some software could calculate up to 5000 descriptors per analyte [37]. Such massive increase in the dimensionality of the descriptors along with the possible incorporation of some nonempirical features could affect the performance of various QSRR models. Therefore, methods for feature selection are necessary to untangle this problem and decide which descriptors are important regarding the retention of compounds of interest. These methods ranged from classical type as forward and backward elimination to advanced nature inspired ones for example particle swarm optimization (PSO), genetic algorithm (GA) and its descendants (firefly, flower pollination and ant colony algorithms) [38–43]. Furthermore, different chemometric and artificial intelligence methods viz., partial least square (PLS), multiple linear regression (MLR), artificial neural networks (ANN) and support vector regression (SVR) proved to be effective in building reliable QSRR models owing to their ability in extracting maximal chemical information in addition to enhancing the speed and quality of analysis[44]. The application of QSRR models have been reported to different chemical families on reversed-phase liquid chromatography such as non- steroidal anti-inflammatory drugs [45], azole antifungal agents [46] and some pain killers drugs [47]. Support vector machine (SVM), a machine learning algorithm, was firstly published by Vapnik, Chervonenkis and co-workers [48]. The algorithm is based on finding a linear function that explains most of the variation of the response and at the same time links the nonlinear relationship between input and the target data [49]. Compared to conventional regression and neural network methods, SVM displays some advantages, including good generalization ability, global optimization and dimensional independence [50]. Thanks to its capability to model possible nonlinear relations between molecular descriptors and retention time, it has been incorporated in building powerful QSRR models [51, 52]. Previously our group developed two QSRR models aimed to provide some essence of the retention behavior for some β-lactams using multiple linear regression models combined with forward or firefly variable selection algorithms [44]. 1. Introduction Resistance to antibacterial agents is a major public health threat affecting humans worldwide mainly due to the uncontrolled use of such bioactive compounds, particularly in countries without standard treatment guidelines. Among those antibacterial agents, fluoroquinolones, a fluoro substituent series derived from nalidixic acid, showed escalating rate of resistance after domination over the therapeutic practice for a time particularly against gram-negative pathogens [1–3]. Such class of active compounds needs to be monitored carefully regarding their use and their abundance in the environment. Hence, from the analytical view, the urgent detection and analysis of these drugs become essential considering finding fast, simple, economical and accurate methods for their analysis. The literature survey revealed that quinolones could be determined thoroughly via high performance liquid chromatography in different matrices viz., biological fluids and tissues [4–10], milk and food of animal origin [11–16], marine products [17], honey [18], waste water [19–21] and in many pharmaceutical formulations [22–27]. Moreover, the relationship between the retention factors and lipophilicity of quinolones using RP- TLC has been assessed [28]. In addition, Wu et al [29] investigated the retention factors-activity relationship of some quinolones using micellar chromatography. On the other hand, sulfonamides are other synthetic antimicrobial agents, unfortunately with widespread resistance which made them infrequently utilized for medical interventions. However, the application of sulfonamides has been extended from their old capabilities as antimicrobial agents to another medical roles viz., anticancer, antiglaucoma, cyclooxygenase-2 and lipoxygenase inhibitors, anticonvulsant and hypoglycemic activities [30]. Regarding the analytical tools used in their detection, literature survey revealed that the determination of this class was also dominated by reversed phase liquid chromatography [31–34]. In context of their retention mechanisms, Cazenave-Gassiot et al discussed the correlation between sulfonamides retention factors and the proportion of modifier in the mobile phase using supercritical fluid Page 2/26 chromatography [35]. However, like quinolones, the separation behavior of this class on reversed phase liquid chromatography needs to be scrutinized. chromatography [35]. However, like quinolones, the separation behavior of this class on reversed phase liquid chromatography needs to be scrutinized. Among different models and theories applied to draw an image about the retention manners in reversed phase liquid chromatography, quantitative structure–retention relationship (QSRR) offers some useful insights not only to elucidate how different chemical drugs perform their retention upon analysis, but also to expect their retention chromatographic systems relatively well [36]. 2.1. Solvents, chemicals, sample preparation and instrumentation. The quinolones (Fig. S1) and sulfonamides (Fig. S2) under investigation were supplied by different pharmaceutical companies. Pure HPLC-grade acetonitrile, methanol and dimethylsulfoxide were supplied by Scarlau (Barcelona, Spain). Other chemicals used in this work including ortho-phosphoric acid, trifluoroacetic acid, sodium dihydrogen orthophosphate and sodium hydroxide were supplied by Honeywell Riedel-de Haën (Seelze, Germany). The instruments used in this study include Jenway 3510, Essex-UK, England pH meter equipped with a glass electrode and Agilent 1260 HPLC-UV series. Stock solutions (2 mg mL-1) of each drug were prepared using a suitable solvent either (methanol, dimethylsulfoxide, water, or acetonitrile). These solutions were stored at 4°C then diluted with the mobile phase to achieve sample concentrations ranging from (0.05–1 mg mL-1) before analysis. 1. Introduction Our scope in this report is to continue our work regarding QSRR modeling of other antibacterial agents (quinolones and sulfonamides), hopefully to highlight their reversed phase chromatographic retention mechanisms with respect to different ionization states and various percentage of organic modifiers for quinolones and sulfonamides, respectively. Owing to the complexity of the generated data, the use of advanced variable selection technique coupled to a machine learning approach seems imperative. Hence, firefly algorithm coupled to SVR has been employed to develop the QSRR models. Page 3/26 Page 3/26 Moreover, the obtained models have been assessed regarding their predictive ability with strict validation criteria, thus could be further employed to predict retention of potential degradation products and even metabolites of these compounds. 2.2. Chromatographic conditions The chromatographic elution of the quinolones was achieved using Inertsil® C18 column (250 mm x 4.6 mm, 5 µm) and detection was carried out at 275 nm. In a gradient mode, 5 mobile phases were prepared according to the plan of experiment and chromatographic system applied as programmed in Table 1, using acetonitrile and 28 mM sodium dihydrogen orthophosphate buffer prepared at different pHs 2.2, 3.5, 5.2, 6.5 and 8.2 using ortho-phosphoric acid or sodium hydroxide. However, the pH was measured again after mixing the buffer with acetonitrile and was found to be 3.2, 4.4, 5.9, 7.32 and 8.9, respectively. The system flow rate was adjusted at 1 ml min- 1. After each injection, the system was reconditioned by going back to the initial ratio and kept constant for 3 min. Data acquisition was performed on Agilent LC Chemstation software. The retention factors of eluted quinolones are listed in Table 2. Table (1) Gradient elution system used in quinolones’ separation Table (1) Gradient elution system used in quinolones’ separation Time (min) Acetonitrile % Buffer % 0 20 80 3 20 80 5 60 40 Page 4/26 Table (2) List of quinolones chromatographic retention factors (k) Page 4/26 Table (2) List of quinolones chromatographic retention factors (k) Table (2) List of quinolones chromatographic retention factors (k) Table (2) List of quinolones chromatographic retention factors (k) Page 4/26 Compound name pH 2.2 pH 3.5 pH 5.2 pH 6.5 pH 8.2 Gatifloxacin 1.580 1.603 1.576 1.566 1.635 Lomefloxacin 1.405 1.560 1.501 0.920 1.035 Moxifloxacin 1.558 1.606 1.592 1.749 1.840 Nadifloxacin 2.121 2.148 2.114 2.036 1.685 Norfloxacin 1.191 1.162 1.192 1.158 0.384 Ofloxacin 1.032 1.176 1.559 1.836 1.885 Ciprofloxacin 1.142 1.363 1.294 1.153 0.646 Gemifloxacin 1.557 1.576 1.578 1.584 1.633 Enrofloxacin 1.622 1.567 1.591 2.468 1.973 Danofloxacin 1.370 1.551 1.572 1.822 1.668 Sparfloxacin 1.560 1.576 1.567 2.486 1.987 Chromatographic separation of sulfonamides was achieved on a hypersil C18 column (150 mm x 4.6 mm, 5 µm) by applying isocratic elution based on a mobile phase consisting of acetonitrile and water acidified with trifluoroacetic acid (1 mL. L-1) in different ratios of 50:50, 45:55 or 30:70, v/v and at a flow rate of 0.8 ml min− 1. Ratio of 15:85, v/v was initially included but not considered for further assessment as many compounds were strongly retained onto the column. Analyses were operated at ambient temperature and detection was carried out at 270 nm. 2.2. Chromatographic conditions Data acquisition was performed on Agilent LC Chemstation software. A list of retention factors of eluted sulfonamides is shown in Table 3. Table (3) List of sulfonamides chromatographic retention factors (k) Page 5/26 Page 5/26 Compound name Acetonitrile% 50 45 30 Sulfacetamide Na 0.154 0.203 0.393 Sulfaguanidine 0.170 0.188 0.256 Sulfadiazine 0.174 0.228 0.443 Sulfaclozine 0.549 0.752 2.196 Sulfadimethoxine 0.419 0.567 1.433 Sulfadimidine 0.311 0.389 0.730 Sulfadoxine 0.395 0.524 1.276 Sulfathiazole 0.166 0.221 0.426 Sulfachloropyrazine Na 0.546 0.754 2.177 Sulfanilamide 0.154 0.194 0.295 Sulfamethoxazole 0.421 0.568 1.548 Sulfapyridine 0.306 0.359 0.597 Sulfaquinoxaline 0.519 0.716 2.221 2.3. QSRR modeling 2.3.1 Drawing Structures and molecular descriptors calculation The major microspecies of the studied quinolones at the pH of interest were estimated using MarvinSketch (6.0.3) [53] which resulted in 21 ions. The canonical smiles of these ions were imported into the Molecular Operating Environment (MOE, 2020.0901) software where they were converted into 3D structures and energy minimized using RMSD gradient of 0.05 kcal.mol− 1Å−1 with MMFF94x forcefield. The partial charges were automatically calculated. Finally, MOE molecular mechanical descriptors were computed for all the compounds resulting in a descriptor fund of 313 descriptors. The initial descriptor fund was reduced by removing zero values and constant descriptors. This ended to a descriptor fund of 293 descriptors. As for sulfonamides, PubChem database [54, 55] was used to introduce sulfonamides canonical SMILES into the MOE where they were converted to 3D structures and energy minimized using the same parameters previously mentioned for quinolones. Afterwards, MOE molecular mechanical descriptors were computed for all the compounds and a descriptor fund of 313 descriptors was obtained. The initial descriptor fund was reduced by removing zero values and constant descriptors, resulting in a fund of 112 descriptors in addition to acetonitrile % incorporated as a descriptor. Page 6/26 Page 6/26 Page 6/26 2.3.3 Descriptors selection and modeling Based on Durbin-Watson (DW) test, linearity of the datasets was tested via augmented partial residual plots (APARP) [56–58]. The test was performed using a custom script written in MATLAB (R2016 a)[59, 60]. The descriptors that continue to exist after the initial filtration were then used to build QSRR models. Firefly algorithm was implemented in MATLAB and applied for descriptor selection as an advanced nature- stimulated algorithm where the RMSECV of the SVR model was considered as a fitness function inside the algorithm for both datasets. The selected descriptors were finally incorporated in SVR final model building. Parameters of the algorithm was combinatorically optimized. These parameters were varied in intervals of specific increments, keeping in mind that in all optimization iterations, one parameter was always varied, while the others were saved constant. 2.3.2 Training set and test set generation The 21 quinolones major microspecies were divided into a calibration (training) set of 16 molecules and a test set of 5 molecules. Regarding sulfonamides, a total of 39 experimental retention factors resulted from three different ratios of mobile phase for the 13 compounds were used in building the QSRR model. The total experiments were split into a training set of 30 observations and an external validation test set of 9 observations. The selection of the calibration and the validation samples of quinolones and sulfonamides was based on keeping the distribution value for the retention factor maintained in both sets. 2.3.4 Model validation It was requisite to evaluate the applicability of the generated models based on model validation. In the present study, models were validated both internally as well as externally and any chance correlation was tested by the use of a y-scrambling technique: a method frequently used for this purpose. Internal validation was performed by using leave-one-out cross-validation (CVLOO) in quinolones model while by using leave-10%-out (CVL10%O) in sulfonamides. External validation was conducted by applying the model on external validation set of 5 microspecies of quinolones and 9 molecules of sulfonamides. The statistical quality of the models was judged by considering the root mean square errors (RMSE) of prediction and the validation correlation coefficients. External validation was conducted by applying the model on external validation set of 5 microspecies of quinolones and 9 molecules of sulfonamides. The statistical quality of the models was judged by considering the root mean square errors (RMSE) of prediction and the validation correlation coefficients. For Y-randomization, the output retention factors of the compounds were shuffled randomly, and the resulting dataset was examined by the FFA-SVM model using real (unscrambled) input descriptors to determine the correlation and predictive ability of the resulting models. The whole procedure was repeated 100 times for both datasets. Hotelling's T2 and William’s plot methods were used to determine the developed models applicability domains (AD) as previously described in our previous work [44]. 3. Results And Discussion 3.1 Optimization of the FFA and SVR parameters for the developed QSRR model of quinolones Firefly algorithm was used as a feature selection method to find the relevant descriptors in order to build reliable QSRR models. However, some parameters were needed to be optimized for proper descriptors’ selection. Initially, the selection of the fitness function used to evaluate the performance of the models was a critical step, thus, based on a previous study [44], the RMSECV was utilized as a fitness function computed by SVR model. Another critical parameter in the FFA is the absorption coefficient parameter “γ” which regulates the light intensity and thus controls the fireflies’ attractiveness, thus, this parameter has a powerful impact on the speed of the convergence and the behavior of the whole algorithm. Another valuable parameter is the “α” parameter that prevents sticking to the local optima through providing some sorts of random movements. Finally, the exploration phase of the FFA was controlled via the number of the fireflies used while the exploitation phase was controlled by the number of generations. The adjusted FFA parameters through combinatorial optimization are shown in Table 4. Table (4) Parameters of the firefly algorithm used for variable selection in QSRR modeling Parameter Quinolones Sulfonamides Number of fireflies 10 20 Generations 100 100 α 0.1 0.15 βο 1 1 γ 0.01 0.01 Concerning SVR, two parameters determine the quality of the model: Penalty error (C), a parameter that controls the trade-off between complexity of decision rule and frequency of error, and insensitive loss function (ɛ) which is a precision factor expressing the radius of the tube placed around the regression function f(x). Moreover, different types of kernels as basis function expansions were also assessed viz., polynomial, radial basis function (RBF) and sigmoid. Initially, the kernel function was examined through evaluating the performance of developed FFA-SVM models, the RBF was selected as best kernel function to model the nonlinearity of the generated data. The RBF kernel parameter regulates the amplitude of the Gaussian function and influence the generalization ability of SVM. Then, the parameters of C & ɛ were optimized. To optimize these parameters, their values was systematically varied in the training step through (CVLOO) and (CVL10%O) for quinolones and sulfonamides, respectively, with monitoring the RMSEcv of mode To obtain the optimal ɛ, the SVM with different ɛ values were trained, during this, we kept value of C as 1 initially, but after finding the optimal value of ɛ, the C value is further optimized. 3. Results And Discussion Page 7/26 Page 7/26 Page 7/26 3.2 QSRR modelling of quinolones in different ionization states For elucidating the chromatographic behavior of the studied quinolones, it is important to understand the relationship between the pH of the mobile phase and the ionization states of each compound, (Fig. S3). Some compounds behave ideally with respect to their ionization state i.e. moxifloxacin exists as a cation (polar) at acidic pHs (2.2 and 3.5) while exists as a neutral compound (hydrophobic) at pH (6.5 and 8.2), so this can rationalize the longer the retention factor of this compound in basic pH rather than the acidic one. Ciprofloxacin, lomefloxacin and norfloxacin exist in different dissociation forms in pHs (5.2 and 8.2) and this describes the fluctuation in their retention factors over these pHs. Nadifloxacin exists as a neutral compound at acidic pHs (2.2, 3.5 and 5.2) what describes the longer the retention factor at these lower pH values while its rapid elution and lower retention factor at basic pH 8.2 as it exists in anionic form. In contrast, ofloxacin and danofloxacin show different behavior, the cationic form of these analytes which appeared at acidic pHs (2.2 and 3.5) show lower retention factors while its anionic forms which are present at basic pHs (6.5 and 8.2) show higher retention factors and so the longer the time it is retained on the column. Additionally, gatifloxacin and gemifloxacin show stability in their retention factors although they can exist in different ionization state along the pH range (2.2–8.2). From these previous observations, the behavior of quinolone compounds cannot be predicted solely based on their ionization state and a more in-depth analysis is required that can predict their behavior successfully. It is noteworthy to mention that the microspecies of each compound could also be present in various ionic forms and in diverse percentages, thus it will be tricky to predict the retention behavior based on a single microspecies. To tackle this problem, we tried to choose the major microspecies as a representative for each molecule in the given pH taking into consideration not to choose the same microspecies at different pHs or different retention factors for the same ionization state. Considering this approach, we were able to derive a simple, interpretable QSRR model that can predict the retention factors of quinolones in their different ionic forms. 3.1 Optimization of the FFA and SVR parameters for the developed QSRR model of quinolones It was found that the best models were obtained using kernel type of (RBF), C = 1 and ɛ =0.01 for both datasets. The final developed FFA-SVM models were used to predict the retention factors of molecules in test set for quinolones and sulfonamides, respectively. Page 8/26 Page 8/26 3.2 QSRR modelling of quinolones in different ionization states The first step in QSRR model implemented for quinolones is to check the linearity of the data, augmented partial residual plots (APARP) along with Durbin-Watson (DW) test were used to check the residuals correlation [56–58]. The associated probability was found to be 0.045 (> 0.05) revealing the significance of the test and the nonlinearity of the data, therefore, nonlinear models as artificial neural network (ANN) and support vector regression (SVR) were tried for data modeling and the best results were obtained during using SVR. Five descriptors were chosen by the FFA and combined in building the SVM model. SMR is a 2D descriptor linked to molecular refractivity including implicit hydrogens [61]. This property is an atomic contribution model that assumes the correct protonation state. GCUT_SLOGP_1 is a 2D descriptor that uses atomic contribution to logP in place of partial charge. VSA is a 3D descriptor that is related to surface area, volume and shape of molecules, it represents van der Waals surface area [62]. Vsurf_EWmin 2 is a 3D descriptor and represents the 2nd lowest hydrophilic energy. Vsurf_IW6 is a 3D descriptor that represents the hydrophilic integy moment at (-4.0). Considering the selected descriptors, the model displays that quinolones retention is Page 9/26 Page 9/26 Page 9/26 based on their size and hydrophobic/hydrophilic nature which are the main elements that influence the retention in reversed phase liquid chromatography. based on their size and hydrophobic/hydrophilic nature which are the main elements that influence the retention in reversed phase liquid chromatography. Regarding the performance of the developed QSRR model, the agreement of the experimental and predicted retention factors shows the good predictive capability of the model as shown in Table 5. The nearness between the training set prediction and the cross-validation results point to the robustness of the resulted model and lack of any overfitting. As shown in Table 6, the results display the good prediction capability of the obtained model. The correlation between the experimental and predicted retention times for the training set, test set and CVLOO results are presented in (Fig. S4&S5). Spearman ranking correlation coefficient (ρ) was also calculated and found to be 0.976, 0.982, and 0.900 for the training set prediction (ρcal), CV LOO (ρLOO) and the external test set (ρpred), respectively, Table 6. Table (6): Quinolones and sulfonamides FFA-SVM model performance evaluation parameters 3.2 QSRR modelling of quinolones in different ionization states The closeness of ρ to “1” indicates a reasonable accuracy and excellent capability of the generated model to reproduce the experimental retention factor ranking (Fig. 1). Table (5): Experimental and predicted retention factors (k) of quinolones compounds in training set, cross- validation and test set Table (5): Experimental and predicted retention factors (k) of quinolones compounds in training set, cross- alidation and test set Table (5): Experimental and predicted retention factors (k) of quinolones compounds in training set, cross- validation and test set Page 10/26 Compound name Buffer pH Experimental k Training set prediction Residuals Cross- Validation CVLOO Residuals Lomefloxacin 6.5 0.920 1.175 0.255 1.247 0.327 Ciprofloxacin 6.5 1.153 1.164 0.011 1.262 0.109 Norfloxacin 3.5 1.162 1.255 0.093 1.342 0.18 Ofloxacin 3.5 1.176 1.261 0.085 1.385 0.209 Ciprofloxacin 3.5 1.363 1.330 -0.033 1.293 -0.07 Lomefloxacin 3.5 1.560 1.550 -0.01 1.401 -0.159 Gatifloxacin 6.5 1.566 1.555 -0.011 1.493 -0.073 Gemifloxacin 3.5 1.576 1.586 0.01 1.593 0.017 Gemifloxacin 6.5 1.584 1.575 -0.009 1.558 -0.026 Gatifloxacin 3.5 1.603 1.613 0.01 1.621 0.018 Moxifloxacin 3.5 1.606 1.616 0.01 1.638 0.032 Danofloxacin 8.2 1.668 1.659 -0.009 1.662 -0.006 Nadifloxacin 8.2 1.685 1.694 0.009 1.763 0.078 Moxifloxacin 6.5 1.749 1.739 -0.01 1.700 -0.049 Enrofloxacin 8.2 1.973 1.819 -0.154 1.773 -0.2 Nadifloxacin 3.5 2.148 1.828 -0.32 1.769 -0.379 Norfloxacin* 6.5 1.158 1.149 -0.009 Danofloxacin* 3.5 1.551 1.433 -0.118 Sparfloxacin* 3.5 1.576 1.563 -0.013 Enrofloxacin* 2.2 1.622 1.493 -0.129 Ofloxacin* 8.2 1.885 1.603 -0.282 () Test set compound Page 11/26 Page 11/26 Parameter Quinolones FFA-SVM Sulfonamides FFA-SVM R2 cal 0.931 0.900 R2 cal-adj 0.926 0.896 q2 LOO / q2 L10%O 0.808 0.812 R2 pred 0.879 0.820 RMSE cal 0.114 0.240 RMSECVLOO 0.163 0.328 RMSEpred 0.148 0.450 ρcal 0.976 0.988 ρLoo / ρL10%O 0.982 0.941 ρpred 0.900 0.883 3.3 QSRR modelling of sulfonamides using different organic modifiers QSRR modeling of sulfonamides was implemented with the aim of studying the associations between the retention factors of the examined compounds eluted upon using different percentages of acetonitrile in the mobile phase composition (Fig. S6): 50%, 45% and 30%, and their calculated constitutional, geometrical, physicochemical and electronical descriptors (independent variables),. Linearity of the data were also considered with the same procedures reported in quinolones with associated probability of 3.2− 17 (> 0.05) indicating the nonlinearity of the generated data. The FFA-SVR model was also utilized in this case resulting in two descriptors plus acetonitrile % in building the QSRR model. The selected features, Vsurf-D2 and vsurf- w2, are 3D descriptors that are related to the molecular hydrophobic and hydrophilic volume, respectively. Moreover, the influence of the third descriptor (acetonitrile % in the mobile phase) indicates that, in the resulted model, the analyte retention is based on its hydrophobic/hydrophilic nature which is a usual element that plays an important role in differential elution of analytes in reversed phase liquid chromatography. The results also display the good prediction capability of the obtained model as shown in Tables 6&7. The model training set, and test set correlation of the experimental and predicted retentions are presented in (Fig. S7), whereas the compounds’ experimental and predicted retentions in the CVLOO is presented in (Fig. S8) indicating the good correlation and the generalized ability of the developed QSRR sulfonamides model. Page 12/26 Page 12/26 Page 12/26 Spearman ranking correlation coefficient (ρ) was also calculated for the training set prediction (ρcal), CV L10%O (ρL10%O) and the external test set (ρpred) and was found to be 0.988, 0.941 and 0.883, respectively, (Fig. 2). The closeness of ρ to “1” indicates the capability of the generated model to reproduce the experimental retention factor ranking of the compounds under investigation in a reasonable accuracy. Spearman ranking correlation coefficient (ρ) was also calculated for the training set prediction (ρcal), CV L10%O (ρL10%O) and the external test set (ρpred) and was found to be 0.988, 0.941 and 0.883, respectively, (Fig. 2). The closeness of ρ to “1” indicates the capability of the generated model to reproduce the experimental retention factor ranking of the compounds under investigation in a reasonable accuracy. 3.4 Y scrambling and applicability domain of both models cRp2 values calculated for cross validation was also found to be 0.687and 0.791 (more than 0.5) for quinolones and sulfonamides QSRR models, respectively what denies that the obtained model is the result of a chance correlation. Statistical parameters of the scrambled models gathered around zero in a symmetrical way for both datasets (Fig. 3), representing that the scrambled models are of very low quality. cRp2 values calculated for cross validation was also found to be 0.687and 0.791 (more than 0.5) for quinolones and sulfonamides QSRR models, respectively what denies that the obtained model is the result of a chance correlation. The applicability domain of a QSPR is the structural, biological space or physicochemical knowledge or information on which the training set of the model has been developed, and for which it is applicable to make predictions for new compounds. In the William’s plot for the FFA-SVM models, the applicability domain is inside a squared area within ± 3 standard deviations and a leverage threshold h of 1.125 and 0.4 for quinolones and sulfonamides, respectively. Prediction is considered reliable only for those compounds that fall within this AD. Also, it can be seen that all compounds (training and test sets) are inside this area without any outlier (Fig. 4). Moreover, the residual plots for both classes show the differences between the predicted and the experimental retention factor (residuals) for the different compounds. The random dispersion of the residuals around the horizontal axis confirmed the predictability of the model, (Fig. S9&S10). 3.3 QSRR modelling of sulfonamides using different organic modifiers Table (7): Experimental and predicted retention factors (k) of sulfonamides compounds in training set, cross- validation and test set Table (7): Experimental and predicted retention factors (k) of sulfonamides compounds in training set, cross- validation and test set Page 13/26 Compound name Acetonitrile % Experimental k Training set prediction Residuals of Training set Cross- Validation CVL10%O Residuals of Cross- Validation Sulfacetamide Na 50% 0.154 0.153 -0.001 0.151 -0.002 Sulfacetamide Na 45% 0.203 0.194 -0.009 0.171 -0.032 Sulfacetamide Na 30% 0.393 0.385 -0.008 0.336 -0.057 Sulfaguanidine 50% 0.170 0.178 0.009 0.212 0.042 Sulfaguanidine 45% 0.188 0.181 -0.006 0.182 -0.006 Sulfaguanidine 30% 0.256 0.267 0.011 0.458 0.202 Sulfadiazine 50% 0.174 0.185 0.011 0.193 0.019 Sulfadiazine 45% 0.228 0.237 0.010 0.236 0.008 Sulfadiazine 30% 0.443 0.454 0.011 0.520 0.078 Sulfaclozine 50% 0.549 0.528 -0.021 0.460 -0.088 Sulfaclozine 45% 0.752 0.701 -0.051 0.657 -0.095 Sulfaclozine 30% 2.196 1.310 -0.886 1.082 -1.114 Sulfadimethoxine 50% 0.419 0.429 0.010 0.427 0.008 Sulfadimethoxine 45% 0.567 0.576 0.009 0.602 0.035 Sulfadimethoxine 30% 1.433 1.278 -0.155 1.006 -0.427 Sulfadimidine 50% 0.311 0.299 -0.012 0.289 -0.022 Sulfadimidine 45% 0.389 0.378 -0.010 0.352 -0.037 Compound name Acetonitrile % Experimental k Training set prediction Residuals of Training set Cross- Validation CVL10%O Residuals of Cross- Validation Sulfadimidine 30% 0.730 0.643 -0.087 0.555 -0.175 Sulfadoxine 50% 0.395 0.405 0.010 0.419 0.024 Sulfadoxine 45% 0.524 0.540 0.017 0.569 0.046 Sulfadoxine 30% 1.276 1.211 -0.065 0.962 -0.314 Sulfathiazole 50% 0.166 0.224 0.058 0.242 0.076 Sulfathiazole 45% 0.221 0.285 0.064 0.289 0.068 Sulfathiazole 30% 0 426 0 521 0 095 0 574 0 148 Compound name Page 14/26 Page 14/26 Sulfachloropyrazine Na 50% 0.546 0.568 0.022 0.623 0.078 Sulfachloropyrazine Na 45% 0.754 0.742 -0.012 0.698 -0.056 Sulfachloropyrazine Na 30% 2.177 1.233 -0.944 0.937 -1.240 Sulfanilamide 50% 0.154 0.165 0.011 0.231 0.078 Sulfanilamide 45% 0.194 0.183 -0.010 0.167 -0.026 Sulfanilamide 30% 0.295 0.307 0.012 0.409 0.114 Sulfacetamide Na 50% 0.154 0.153 -0.001 0.151 -0.002 Sulfacetamide Na 45% 0.203 0.194 -0.009 0.171 -0.032 Sulfacetamide Na 30% 0.393 0.385 -0.008 0.336 -0.057 Sulfaguanidine 50% 0.170 0.178 0.009 0.212 0.042 Sulfaguanidine 45% 0.188 0.181 -0.006 0.182 -0.006 Sulfaguanidine 30% 0.256 0.267 0.011 0.458 0.202 Compound name Acetonitrile % Experimental k Training set prediction Residuals of Training set Cross- Validation CVL10%O Residuals of Cross- Validation Sulfadiazine 50% 0.174 0.185 0.011 0.193 0.019 Sulfadiazine 45% 0.228 0.237 0.010 0.236 0.008 Sulfadiazine 30% 0.443 0.454 0.011 0.520 0.078 Sulfaclozine 50% 0.549 0.528 -0.021 0.460 -0.088 Sulfaclozine 45% 0.752 0.701 -0.051 0.657 -0.095 Sulfaclozine 30% 2.196 1.310 -0.886 1.082 -1.114 Sulfadimethoxine 50% 0.419 0.429 0.010 0.427 0.008 Sulfadimethoxine 45% 0.567 0.576 0.009 0.602 0.035 Sulfadimethoxine 30% 1.433 1.278 -0.155 1.006 -0.427 Sulfamethoxazole* 50% 0.421 0.248 Sulfamethoxazole* 45% 0.568 0.322 Sulfamethoxazole* 30% 1.548 0.644 Sulfapyridine* 50% 0.306 0.283 Sulfapyridine* 30% 0.597 0.443 Sulfaquinoxaline* 50% 0.519 0.530 Sulfaquinoxaline* 45% 0.716 0.719 Sulfaquinoxaline* 30% 2.221 1.279 -Test set compound () bl d l b l d f b h d l -Test set compound () 3.4 Y scrambling and applicability domain of both models Y-randomization or permutation test is another criterion used to validate our findings in this study especially with this low number of observations. In this test, it is suspected that the original QSRR model is significant if there is a solid link between the selected descriptors and the original response variables. Y- randomization was repeated for 100 times. If the statistical attributes of these randomized models are much lower than the original one, it can be decided that the model is sensible and had not been obtained by luck. The below equation was used to evaluate the quality of the obtained models from the 100 randomized matrices and to weigh against the original model quality. cRp 2 should be above 0.5 to ensure that the original model is not obtained by chance [63]. cR 2 p = R* R2 −R2 y √ Where (cRp 2) is the degree of variation in the values of the squared correlation coefficient average of the randomized models Ry 2 and the squared correlation coefficient of the original model R2. Where (cRp 2) is the degree of variation in the values of the squared correlation coefficient average of the randomized models Ry 2 and the squared correlation coefficient of the original model R2. Statistical parameters of the scrambled models gathered around zero in a symmetrical way for both datasets (Fig. 3), representing that the scrambled models are of very low quality. cRp2 values calculated for cross validation was also found to be 0.687and 0.791 (more than 0.5) for quinolones and sulfonamides QSRR models, respectively what denies that the obtained model is the result of a chance correlation. Where (cRp 2) is the degree of variation in the values of the squared correlation coefficient average of the randomized models Ry 2 and the squared correlation coefficient of the original model R2. Where (cRp 2) is the degree of variation in the values of the squared correlation coefficient average of the randomized models Ry 2 and the squared correlation coefficient of the original model R2. Statistical parameters of the scrambled models gathered around zero in a symmetrical way for both datasets (Fig. 3), representing that the scrambled models are of very low quality. 4. Conclusion Two QSRR models were built for prediction of the retention factors of quinolones and sulfonamides in HPLC system. The influence of the pH of the mobile phase on the retention factors and the ionization state of each Page 16/26 Page 16/26 Page 16/26 quinolone and the effect of acetonitrile composition in the mobile phase on the retention factors of sulfonamides were studied resulting in selection of 21 major microspecies of quinolones and 39 sulfonamides compounds. In both classes, significant descriptors that are related to the retention behavior in the chromatographic system were selected using the advanced FFA and then incorporated in building the QSRR models using SVM algorithm. In both, FFA-SVM models displayed that the analyte retention is quinolone and the effect of acetonitrile composition in the mobile phase on the retention factors of sulfonamides were studied resulting in selection of 21 major microspecies of quinolones and 39 sulfonamides compounds. In both classes, significant descriptors that are related to the retention behavior in the chromatographic system were selected using the advanced FFA and then incorporated in building the QSRR models using SVM algorithm. In both, FFA-SVM models displayed that the analyte retention is dependent on its hydrophobic/hydrophilic nature and/or its size. The two models showed high performance on both the training level and the validation level. In quinolones, the regression coefficients of the training set prediction (R2 cal), CV LOO (q2 LOO) and the external test set (R2 pred) were 0.931 (R2 adjusted = 0.926), 0.808 and 0.879, respectively, and with root mean square errors (RMSE) of 0.114, 0.163 and 0.148, respectively. In sulfonamides, the regression coefficients of the training set prediction (R2 cal), CV L10%O (q2 L10%O) and the external test set (R2 pred) were 0.900 (R2 adjusted = 0.896), 0.812 and 0.820, respectively, and with root mean square errors (RMSE) of 0.240, 0.450 and 0.328, respectively. In the Y-randomization validation test, the two models showed cRp 2 values of 0.687 and 0.791 for quinolones and sulfonamides, respectively indicating that both models are significant and not obtained by chance. 5.2. Consent for publication Not applicable The authors have no competing of interest The authors have no competing of interest 5.3. Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information files. 5.1. Ethics approval and consent to participate 5.1. Ethics approval and consent to participate 5.1. Ethics approval and consent to participate Not applicable 5.5. Funding No funding received 5.7. Acknowledgements Not applicable Marwa A. Fouad: Conceptualization, Writing - Review & Editing, Supervision Enas H. Tolba: Methodology, Investigation, Writing - Original Draft Manal A. El-Shal: Supervision Ahmed Serag: Methodology, Software, Writing - Original Draft and Ahmed M. El Kerdawy: Conceptualization, Methodology, Software, Writing - Review & Editing Marwa A. Fouad: Conceptualization, Writing - Review & Editing, Supervision Enas H. Tolba: Methodology, Investigation, Writing - Original Draft Manal A. El-Shal: Supervision Ahmed Serag: Methodology, Software, Writing - Original Draft and Ahmed M. El Kerdawy: Conceptualization, Methodology, Software, Writing - Review & Editing Page 17/26 Page 17/26 References 1. Doble A. Quinolones. xPharm: The Comprehensive Pharmacology Reference. 2007;:1–3. 2. Aldred KJ, Kerns RJ, Osheroff N. Mechanism of quinolone action and resistance. Biochemistry. 2014;53:1565–74. 3. Pham TDM, Ziora ZM, Blaskovich MAT. Quinolone antibiotics. MedChemComm. 2019;10:1719–39. M, Ziora ZM, Blaskovich MAT. Quinolone antibiotics. MedChemComm. 2019;1 3. Pham TDM, Ziora ZM, Blaskovich MAT. Quinolone antibiotics. MedChemCo 4. Gauhar S, Ali SA, Shoaib H, Shyum Naqvi SB, Muhammad IN. Development and Validation of a HPLC Method for Determination of Pefloxacin in Tablet and Human Plasma. Mashhad University of Medical Sciences. 2009;12:33–42. 5. Holtzapple CK, Buckley SA, Stanker LH. Determination of fluoroquinolones in serum using an on-line clean-up column coupled to high-performance immunoaffinity–reversed-phase liquid chromatography. Journal of Chromatography B: Biomedical Sciences and Applications. 2001;754:1–9. 6. Sultana N, Arayne MS, Shafi N, Naz A, Naz S, Shamshad H. A RP-HPLC Method For The Simultaneous Determination Of Diltiazem and Quinolones in Bulk, Formulations and Human Serum. Journal of the Chilean Chemical Society. 2009;54:358–62. 7. Haritova AM, Petrova DK, Stanilova SA. A Simple HPLC Method For Detection Of Fluoroquinolones in Serum Of Avian Species. Journal of Liquid Chromatography & Related Technologies. 2012;35:1130–9. 8. Nemutlu E, Kır S, Özyüncü Ö, Beksaç MS. Simultaneous Separation and Determination of Seven Quinolones Using HPLC: Analysis of Levofloxacin and Moxifloxacin in Plasma and Amniotic Fluid. Chromatographia. 2007;66:15–24. 9. Cavazos-Rocha N, Carmona-Alvarado I, Vera-Cabrera L, Waksman-de-Torres N, Salazar-Cavazos M de la L. HPLC Method for the Simultaneous Analysis of Fluoroquinolones and Oxazolidinones in Plasma. Journal of Chromatographic Science. 2014;52:1281–7. 9. Cavazos-Rocha N, Carmona-Alvarado I, Vera-Cabrera L, Waksman-de-Torres N, Salazar-Cavazos M de la L. HPLC Method for the Simultaneous Analysis of Fluoroquinolones and Oxazolidinones in Plasma. Journal of Chromatographic Science. 2014;52:1281–7. 10. Yu H, Mu H, Hu Y-M. Determination of fluoroquinolones, sulfonamides, and tetracyclines multiresidues simultaneously in porcine tissue by MSPD and HPLC–DAD. Journal of Pharmaceutical Analysis. 2012;2:76–81. 10. Yu H, Mu H, Hu Y-M. Determination of fluoroquinolones, sulfonamides, and tetracyclines multiresidues simultaneously in porcine tissue by MSPD and HPLC–DAD. Journal of Pharmaceutical Analysis. 2012;2:76–81. 11. Gili M, Marchis D, Stella P, Olivo F, Ostorero F, Franzoni M, et al. Multiresidue confirmatory method for determination of quinolones in milk by HPLC: method development and validation according to the criteria of Commission Decision 2002/657/EC. Italian Journal of Food Safety. 2013;2:9. 11. Gili M, Marchis D, Stella P, Olivo F, Ostorero F, Franzoni M, et al. References American Journal of Analytical Chemistry. 2011;02:200– 5. 19. Turiel E, Bordin G, Rodríguez AR. Determination of quinolones and fluoroquinolones in hospital sewage water by off-line and on-line solid-phase extraction procedures coupled to HPLC-UV. Journal of Separation Science. 2005;28:257–67. 20. Ibrahim HK, Abdel-Moety MM, Abdel-Gawad SA, Al-Ghobashy MA, Kawy MA. Validated electrochemical and chromatographic quantifications of some antibiotic residues in pharmaceutical industrial waste water. Environmental Science and Pollution Research. 2017;24:7023–34. 20. Ibrahim HK, Abdel-Moety MM, Abdel-Gawad SA, Al-Ghobashy MA, Kawy MA. Validated electrochemical and chromatographic quantifications of some antibiotic residues in pharmaceutical industrial waste water. Environmental Science and Pollution Research. 2017;24:7023–34. 21. Eva M. Golet, Alfredo C. Alder, Andreas Hartmann, Thomas A. Ternes and, Giger W. Trace Determination of Fluoroquinolone Antibacterial Agents in Urban Wastewater by Solid-Phase Extraction and Liquid Chromatography with Fluorescence Detection. Anal Chem. 2001;73:3632–8. 21. Eva M. Golet, Alfredo C. Alder, Andreas Hartmann, Thomas A. Ternes and, Giger W. Trace Determination of Fluoroquinolone Antibacterial Agents in Urban Wastewater by Solid-Phase Extraction and Liquid Chromatography with Fluorescence Detection. Anal Chem. 2001;73:3632–8. 22. Scherer R, Pereira J, Firme J, Lemos M, Lemos M. Determination of Ciprofloxacin in Pharmaceutical Formulations Using HPLC Method with UV Detection. Indian journal of pharmaceutical sciences. 2014;76:541–4. 22. Scherer R, Pereira J, Firme J, Lemos M, Lemos M. Determination of Ciprofloxacin in Pharmaceutical Formulations Using HPLC Method with UV Detection. Indian journal of pharmaceutical sciences. 2014;76:541–4. 23. Chamseddin C, Jira TH. Comparison of the chromatographic behavior of levofloxacin, ciprofloxacin and moxifloxacin on various HPLC phases. Die Pharmazie. 2011;66:244–8. 23. Chamseddin C, Jira TH. Comparison of the chromatographic behavior of levofloxacin, ciprofloxacin and moxifloxacin on various HPLC phases. Die Pharmazie. 2011;66:244–8. 24. Sversut RA, Serrou do Amaral M, César de Moraes Baroni A, Rodrigues PO, Rosa AM, Galana Gerlin MC, et al. Stability-indicating HPLC-DAD method for the simultaneous determination of fluoroquinolones and corticosteroids in ophthalmic formulations. Anal Methods. 2014;6:2125–33. 25. Puranik M, Bhawsar D, Rathi P, Yeole P. Simultaneous determination of ofloxacin and ornidazole in solid dosage form by RP-HPLC and HPTLC techniques. Indian Journal of Pharmaceutical Sciences. 25. Puranik M, Bhawsar D, Rathi P, Yeole P. Simultaneous determination of ofloxacin and ornidazole in solid dosage form by RP-HPLC and HPTLC techniques. Indian Journal of Pharmaceutical Sciences. 2010;72:513. 26. e Souza MJ, Bittencourt CF, Morsch LM. LC determination of enrofloxacin. Journal of pharmaceutical and biomedical analysis. 2002;28:1195–9. 26. e Souza MJ, Bittencourt CF, Morsch LM. References Multiresidue confirmatory method for determination of quinolones in milk by HPLC: method development and validation according to the criteria of Commission Decision 2002/657/EC. Italian Journal of Food Safety. 2013;2:9. 12. Christodoulou EA, Samanidou VF. Multiresidue HPLC analysis of ten quinolones in milk after solid phase extraction: Validation according to the European Union Decision 2002/657/EC. Journal of Separation Science. 2007;30:2421–9. 12. Christodoulou EA, Samanidou VF. Multiresidue HPLC analysis of ten quinolones in milk after solid phase extraction: Validation according to the European Union Decision 2002/657/EC. Journal of Separation Science. 2007;30:2421–9. 13. Stoilova N, Surleva A, Stoev G. Determination of quinolonones in food of animal origin by liquid chromatography coupled with fluorescence and mass spectrometric detection. Acta Chromatographica. 2014;26:599–614. 13. Stoilova N, Surleva A, Stoev G. Determination of quinolonones in food of animal origin by liquid chromatography coupled with fluorescence and mass spectrometric detection. Acta Chromatographica. 2014;26:599–614. Page 18/26 14. Sijun Zhao, Haiyang Jiang, Xuelian Li, Tiejun Mi, Cun Li and, Shen* J. Simultaneous Determination of Trace Levels of 10 Quinolones in Swine, Chicken, and Shrimp Muscle Tissues Using HPLC with Programmable Fluorescence Detection. :J Agric Food Chem . 2007;55:3829–34. 15. Chonan T, Fujimoto T, Inoue M, Tazawa T, Ogawa H. [Multiresidue determination of quinolones in animal and fishery products by HPLC]. Shokuhin eiseigaku zasshi Journal of the Food Hygienic Society of Japan. 2008;49:244–8. 16. Stoilova N, Petkova M. Developing and Validation Of Method For Detection of Quinolone Residues in Poultry Meat. Trakia Journal of Sciences. 2010;8:64–9. 16. Stoilova N, Petkova M. Developing and Validation Of Method For Detection of Quinolone Residues in Poultry Meat. Trakia Journal of Sciences. 2010;8:64–9. 17. Jang J, Lee K, Kwon K, Bae S, Kim HS. Simultaneous determination of thirteen quinolones in livestock and fishery products using ultra performance LC with electrospray ionization tandem mass spectrometry. Food Science and Biotechnology. 2013;22:1–9. 17. Jang J, Lee K, Kwon K, Bae S, Kim HS. Simultaneous determination of thirteen quinolones in livestock and fishery products using ultra performance LC with electrospray ionization tandem mass spectrometry. Food Science and Biotechnology. 2013;22:1–9. 18. Du W, Yao J, Li Y, Hashi Y. Rapid Determination of Residual Quinolones in Honey Samples by Fast HPLC with an On-Line Sample Pretreatment System. American Journal of Analytical Chemistry. 2011;02:200– 5. 18. Du W, Yao J, Li Y, Hashi Y. Rapid Determination of Residual Quinolones in Honey Samples by Fast HPLC with an On-Line Sample Pretreatment System. References LC determination of enrofloxacin. Journal of pharmaceutical and biomedical analysis. 2002;28:1195–9. 27. John P, Azeem W, Ashfaq M, Khan IU, Razzaq SN. Stability indicating RP-HPLC method for simultaneous determination of piroxicam and ofloxacin in binary combination. Pakistan journal of pharmaceutical sciences. 2015;28:1713–21. 27. John P, Azeem W, Ashfaq M, Khan IU, Razzaq SN. Stability indicating RP-HPLC method for simultaneous determination of piroxicam and ofloxacin in binary combination. Pakistan journal of pharmaceutical sciences. 2015;28:1713–21. Page 19/26 Page 19/26 28. Hubicka U, Żuromska-Witek B, Komsta Ł, Krzek J. Lipophilicity study of fifteen fluoroquinolones by reversed-phase thin-layer chromatography. Analytical Methods. 2015;7:3841–8. 29. Wu L-P, Chen Y, Wang S-R, Chen C, Ye L-M. Quantitative retention–activity relationship models for quinolones using biopartitioning micellar chromatography. Biomedical Chromatography. 2008;22:106– 14. 30. Parasca OM, Gheaţă F, Pânzariu A, Geangalău I, Profire L. Importance of sulfonamide moiety in current and future therapy. Revista medico-chirurgicala a Societatii de Medici si Naturalisti din Iasi. 117:558–64. 31. Ye S, Yao Z, Na G, Wang J, Ma D. Rapid simultaneous determination of 14 sulfonamides in wastewater by liquid chromatography tandem mass spectrometry. Journal of Separation Science. 2007;30:2360–9. 32. Malintan NT, Mohd MA. Determination of sulfonamides in selected Malaysian swine wastewater by high-performance liquid chromatography. Journal of Chromatography A. 2006;1127:154–60. 32. Malintan NT, Mohd MA. Determination of sulfonamides in selected Malaysian swine wastewater by high-performance liquid chromatography. Journal of Chromatography A. 2006;1127:154–60. 33. Saini B, Bansal G. Degradation Study on Sulfasalazine and a Validated HPLC-UV Method for its Stability Testing. Scientia pharmaceutica. 2014;82:295–306. 33. Saini B, Bansal G. Degradation Study on Sulfasalazine and a Validated HPLC-UV Method for its Stability Testing. Scientia pharmaceutica. 2014;82:295–306. 34. Ozkorucuklu SP, Sahin Y, Alsancak G. Determination of sulfamethoxazole in pharmaceutical formulations by flow injection system/HPLC with potentiometric detection using polypyrrole electrode. Journal of the Brazilian Chemical Society. 2011;22:2171–7. 34. Ozkorucuklu SP, Sahin Y, Alsancak G. Determination of sulfamethoxazole in pharmaceutical formulations by flow injection system/HPLC with potentiometric detection using polypyrrole electrode. Journal of the Brazilian Chemical Society. 2011;22:2171–7. 35. Cazenave-Gassiot A, Boughtflower R, Caldwell J, Coxhead R, Hitzel L, Lane S, et al. Prediction of retention for sulfonamides in supercritical fluid chromatography. Journal of Chromatography A. 2008;1189:254–65. 36. Kaliszan* R. QSRR: Quantitative Structure-(Chromatographic) Retention Relationships. 2007. https://doi.org/10.1021/CR068412Z. 37. Mauri A, Consonni V, Pavan M, Match RT-, 2006 undefined. Dragon software: An easy approach to molecular descriptor calculations. academia.edu. 38. Fister I, Yang X-S, Brest J, Fister D. References A Brief Review of Nature-Inspired Algorithms for Optimization. ELEKTROTEHNISKI VESTNIK. 2013;80:1–7. 39. Yang X-S. Flower Pollination Algorithm for Global Optimization. Springer, Berlin, Heidelberg; 2012. p. 240–9. 40. Yang X-S. Firefly Algorithms for Multimodal Optimization. Springer, Berlin, Heidelberg; 2009. 40. Yang X-S. Firefly Algorithms for Multimodal Optimization. Springer, Berlin, H 41. Qi Shen, Jian-Hui Jiang, Jing-chao Tao, Guo-li Shen and, Ru-Qin Yu. Modified Ant Colony Optimization Algorithm for Variable Selection in QSAR Modeling: QSAR Studies of Cyclooxygenase Inhibitors. J Chem Inf Model. 2005;45:1024–9. 42. Gonzalez M, Teran C, Saiz-Urra L, Teijeira M. Variable Selection Methods in QSAR: An Overview. Current Topics in Medicinal Chemistry. 2008;8:1606–27. 43. Shahlaei M. Descriptor Selection Methods in Quantitative Structure–Activity Relationship Studies: A Review Study. Chemical Reviews. 2013;113:8093–103. 44. Fouad MA, Tolba EH, El-shal MA, El AM. QSRR modeling for the chromatographic retention behavior of some ␤ -lactam antibiotics using forward and firefly variable selection algorithms coupled with multiple linear regression. 2018;1549:51–62. Page 20/26 45. Carlucci G, D’Archivio AA, Maggi MA, Mazzeo P, Ruggieri F. Investigation of retention behaviour of non- steroidal anti-inflammatory drugs in high-performance liquid chromatography by using quantitative structure–retention relationships. Analytica Chimica Acta. 2007;601:68–76. 46. Golubović J, Protić A, Zečević M, Otašević B, Mikić M, Živanović L. Quantitative structure–retention relationships of azole antifungal agents in reversed-phase high performance liquid chromatography. Talanta. 2012;100:329–37. 47. Ghasemi J, Saaidpour S. QSRR Prediction of the Chromatographic Retention Behavior of Painkiller Drugs. Journal of Chromatographic Science. 2009;47:156–63. 48. Vapnik N V. The Nature of Statistical Learning Theory. 2nd edition. New York: Springer science & business media; 2000. 49. Louis B, Agrawal VK, Khadikar P V. Prediction of intrinsic solubility of generic drugs using MLR, ANN and SVM analyses. European Journal of Medicinal Chemistry. 2010;45:4018–25. 50. Du H, Wang J, Yao X, Hu Z. Quantitative Structure-Retention Relationship Models for the Prediction of the Reversed-Phase HPLC Gradient Retention Based on the Heuristic Method and Support Vector Machine. Journal of Chromatographic Science. 2009;47:396–404. 51. Du H, Wang J, Yao X, Hu Z. Quantitative structure-retention relationship models for the prediction of the reversed-phase HPLC gradient retention based on the heuristic method and support vector machine. Journal of chromatographic science. 47:396–404. 51. Du H, Wang J, Yao X, Hu Z. Quantitative structure-retention relationship models for the prediction of the reversed-phase HPLC gradient retention based on the heuristic method and support vector machine. Journal of chromatographic science. 47:396–404. 52. References Li X, Luan F, Si H, Hu Z, Liu M. Prediction of retention times for a large set of pesticides or toxicants based on support vector machine and the heuristic method. Toxicology Letters. 2007;175:136–44. 52. Li X, Luan F, Si H, Hu Z, Liu M. Prediction of retention times for a large set of pesticides or toxicants based on support vector machine and the heuristic method. Toxicology Letters. 2007;175:136–44. 53. ChemAxon - Software Solutions and Services for Chemistry & Biology. 53. ChemAxon - Software Solutions and Services for Chemistry & Biology. 53. ChemAxon - Software Solutions and Services for Chemistry & Biology 54. Kim S, Thiessen PA, Bolton EE, Chen J, Fu G, Gindulyte A, et al. PubChem Substance and Compound databases. Nucleic Acids Research. 2016;44:D1202–13. 54. Kim S, Thiessen PA, Bolton EE, Chen J, Fu G, Gindulyte A, et al. PubChem Substance and Compound databases. Nucleic Acids Research. 2016;44:D1202–13. PubChem Project. https://pubchem.ncbi.nlm.nih.gov/. Accessed 20 Nov 2017. 56. Olivieri AC. Practical guidelines for reporting results in single- and multi-component analytical calibration: A tutorial. Analytica Chimica Acta. 2015;868:10–22. 57. Montgomery DC, Peck EA, Vining GG. Introduction to linear regression analysis. Wiley; 2012. 58. Centner, V.; De Noord, O.; Massart D. Detection of nonlinearity in multivariate calibration. Analytica Chimica Acta. 1998;376:153–68. 59. Martinez WL, Martinez AR, Solka JL. Exploratory data analysis with MATLAB. 2nd edition. CRC Press; 2011. 60. Menke W, Menke JE (Joshua E. Environmental data analysis with MatLab. 2nd edition. 2016. 61. Sharma MC, Kohli D V. Insight into the structural requirement of substituted quinazolinone biphenyl acylsulfonamides derivatives as Angiotensin II AT1 receptor antagonist: 2D and 3D QSAR approach. Journal of Saudi Chemical Society. 2014;18:35–45. 62. Guha R, Willighagen E. A Survey of Quantitative Descriptions of Molecular Structure. Current Topics in Medicinal Chemistry. 2020;12:1946–56. Page 21/26 Page 21/26 63. Gramatica P. Principles of QSAR models validation: internal and external. QSAR & Combinatorial Science. 2007;26:694–701. Figures Figure 1 FFA-SVM model experimental k ranking vs predicted k ranking in Quinolones training set prediction. FFA-SVM model experimental k ranking vs predicted k ranking in Quinolones training set prediction. Page 22/26 Figure 2 FFA-SVM model experimental k ranking vs predicted k ranking in Sulfonamides training set prediction FFA-SVM model experimental k ranking vs predicted k ranking in Sulfonamides training set prediction Page 23/26 gure 3 Figure 3 Y-randomization validation results for the FFA-SVM for (A) quinolones (B) sulfonamides modeling. Figure 3 Y-randomization validation results for the FFA-SVM for (A) quinolones (B) sulfonamides modeling. Y-randomization validation results for the FFA-SVM for (A) quinolones (B) sulfonamides modeling. Page 24/26 igure 4 Williams plot for FFA-SVM models of (A) quinolones with ±3 standard deviations, and a leverage thresh * of 1.125 as warning limits and (B) sulfonamides with ±3 standard deviations, and a leverage thresho f 0 4 as warning limits Circles represent training set cross validation prediction and diamonds represen Figure 4 Williams plot for FFA-SVM models of (A) quinolones with ±3 standard deviations, and a leverage threshold h* of 1.125 as warning limits and (B) sulfonamides with ±3 standard deviations, and a leverage threshold h* of 0.4 as warning limits. Circles represent training set cross-validation prediction and diamonds represent test set prediction. Page 25/26 Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. SupplementaryMaterials.docx quinolones.xlsx sulfonamides.xlsx Page 26/26
https://openalex.org/W1976555942	https://europepmc.org/articles/pmc4236167?pdf=render	English	null	Alouatta Trichromatic Color Vision: Cone Spectra and Physiological Responses Studied with Microspectrophotometry and Single Unit Retinal Electrophysiology	PloS one	2,014	cc-by	10,512	Abstract The howler monkeys (Alouatta sp.) are the only New World primates to exhibit routine trichromacy. Both males and females have three cone photopigments. However, in contrast to Old World monkeys, Alouatta has a locus control region upstream of each opsin gene on the X-chromosome and this might influence the retinal organization underlying its color vision. Post- mortem microspectrophotometry (MSP) was performed on the retinae of two male Alouatta to obtain rod and cone spectral sensitivities. The MSP data were consistent with only a single opsin being expressed in each cone and electrophysiological data were consistent with this primate expressing full trichromacy. To study the physiological organization of the retina underlying Alouatta trichromacy, we recorded from retinal ganglion cells of the same animals used for MSP measurements with a variety of achromatic and chromatic stimulus protocols. We found MC cells and PC cells in the Alouatta retina with similar properties to those previously found in the retina of other trichromatic primates. MC cells showed strong phasic responses to luminance changes and little response to chromatic pulses. PC cells showed strong tonic response to chromatic changes and small tonic response to luminance changes. Responses to other stimulus protocols (flicker photometry; changing the relative phase of red and green modulated lights; temporal modulation transfer functions) were also similar to those recorded in other trichromatic primates. MC cells also showed a pronounced frequency double response to chromatic modulation, and with luminance modulation response saturation accompanied by a phase advance between 10–20 Hz, characteristic of a contrast gain mechanism. This indicates a very similar retinal organization to Old- World monkeys. Cone-specific opsin expression in the presence of a locus control region for each opsin may call into question the hypothesis that this region exclusively controls opsin expression. Editor: Daniel Osorio, University of Sussex, United Kingdom Received May 13, 2014; Accepted October 27, 2014; Published November 18, 2014 Received May 13, 2014; Accepted October 27, 2014; Published November 18, 2014 Copyright: 2014 Silveira et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are included within the Supporting Information files. Alouatta Trichromatic Color Vision: Cone Spectra and Physiological Responses Studied with Microspectrophotometry and Single Unit Retinal Electrophysiology Luiz Carlos L. Silveira1,2, Ce´zar A. Saito1, Manoel da Silva Filho1, Jan Kremers3, Barry B. Lee5,6 1 Instituto de Cieˆncias Biolo´gicas, Universidade Federal do Para´, Bele´m, Para´, Brazil, 2 Nu´cleo de Medicina Tropical, Universidade Federal do Para´, Bele´m, Para´, Brazil, 3 Department of Ophthalmology, University Hospital Erlangen, Erlangen, Germany, 4 Division of Visual Science, Institute of Ophthalmology, University College London, London, England, United Kingdom, 5 State College of Optometry, State University of New York, New York, New York, United States of America, 6 Max Planck Institute for Biophysical Chemistry, Go¨ttingen, Germany Abstract Funding: This research was supported by the following grants: CNPq-PRONEX / FAPESPA / UFPA / FADESP #2268; CNPq-PRONEX / FAPESPA #316799/2009; FINEP IBN Net #1723. LCLS is a CNPq research fellow. JK was supported by the Hertie Foundation. The authors would like to thank Dr. Jose´ Augusto Pereira Carneiro Muniz of the Centro Nacional de Primatas (Ananindeua, Para´, Brazil) for provision of animals. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Email: luiz@ufpa.br * Email: luiz@ufpa.br ation: Silveira LCL, Saito CA, da Silva Filho M, Kremers J, Bowmaker JK, et al. (2014) Alouatta Trichromatic Color Vision: Cone Spe sponses Studied with Microspectrophotometry and Single Unit Retinal Electrophysiology. PLoS ONE 9(11): e113321. doi:10.1371/journ November 2014 \| Volume 9 \| Issue 11 \| e113321 aito CA, da Silva Filho M, Kremers J, Bowmaker JK, et al. (2014) Alouatta Trichromatic Color Vision: Cone Spectra and Physiologica Microspectrophotometry and Single Unit Retinal Electrophysiology. PLoS ONE 9(11): e113321. doi:10.1371/journal.pone.0113321 Introduction vision can be attained. This has been demonstrated using different behavioral methods [6], [7] and physiologically recordings of opponent cone responses from retinal ganglion cells and lateral geniculate nucleus (LGN) relay neurons [8–11]. In catarrhine primates (Old-World monkeys, apes, and humans), the genes for the middle- (M) and long-wavelength (L) sensitive opsins form a tandem array on the X chromosome and, together with the short-wavelength (S) cone opsin, coded on chromosome 7, form the basis for trichromacy [1–3]. In diurnal platyrrhine primates (New-World monkeys), usually only one longer wavelength opsin gene is found on the X chromosome but, depending of the species, two or more alleles are present [4], [5]. This means that all males are dichromats but if a female possesses two different alleles on her X chromosome pair, trichromatic color A major exception is the Alouatta. Electroretinographic studies in this platyrrhine species showed that in males and females both M- and L-opsins are present [12], and there is evidence in males and females for behavioral trichromacy [13]. The Alouatta L- and M-photopigments have spectral absorption maxima at about 560 and 530 nm, close to those of catarrhines [12], but there may be some inter-individual variability [14]. The genes coding these November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 1 Alouatta Color Vision opsins are present in a tandem array on the X chromosome, as in catarrhines, but with a different structure [4], [14], [15]. regimen and drinking water at libitum. Animal housing conditions, feeding regimen, and health were supervised by the CENP veterinarian staff. The animals were kept in environmental enriched cages as recommended by CENP primatologists. In catarrhines, a single locus control region (LCR) is situated upstream of the tandem gene array, and this is thought to control which gene is expressed [4], [15]. In the Alouatta, a LCR is present upstream of each opsin gene, suggesting that the opsin gene and upstream mechanisms are both duplicated [4], [15]. As a result, Alouatta trichromacy may have a different etiology compared to catarrhines [4], [15]. g y p g Each animal room measured 2.50 m x 4.00 m x 2.25 m, 10 m2 of total area. Each room pair housed 1 male and 5–6 female Alouatta. Houses and rooms were cleaned daily. They were washed with jets of water containing 10% hypochlorite. Introduction Animals were moved from one of the paired rooms to the other during cleaning. Clean water was provided for primates in 500 ml animal drinking bottles, two bottles per room. Bottles were refilled 3 times per day. Primate diet consisted of the following items: a) food pellets provided in plates, ad libitum; b) fruits, vegetables, once a day; cooked eggs, twice a week; diluted milk, twice a week. In addition, Alouatta were provided, daily with Embauba leaves which are part of the natural diet of these Amazonian primates. All animals housed in the CENP received continuous veterinary care which followed the following protocol: inspection in the first hour every day to observe each animal and to inspect the feeding plates and drinking bottles; thereafter regular inspection until the evening. There are also facilities for primate anesthesia and surgery. The facilities allow primate full veterinary assistance, including X-ray, ultrasound, and complete blood, urine, and stool clinical laboratory examination and testing. Each room is enriched with trunks, branches, and ropes to facilitate primate exercising and escape during disputes and force display. Platforms are available for resting and access for feeding. Anatomically, the retinae of diurnal platyrrhines are very similar to those of catarrhines [16–19]. Physiologically, early studies performed, for example, on the squirrel monkey, demonstrated the differences in color opponency in LGN between males and females [20]. Later, physiology performed in further platyrrhine species (Cebus sp., Saimiri sp., and Callithrix sp.) confirmed color opponency in the retinal and LGN cells of trichromatic females, and suggested a pattern that is broadly similar to that of catarrhines, with low achromatic and high chromatic contrast sensitivity of PC cells, and higher contrast sensitivity of MC cells; dichromatic male platyrrhines also showed a low contrast sensitivity of PC cells, indicating an adherence to a primate pattern [8], [9], [11], [21–26]. The retinal anatomy of the Alouatta shows primate character- istics in addition to some unique features. Franco and colleagues estimated the cone density for both retinas of one individual and found that they had an extremely high cone density of about 429,000 cones/mm2 and 357,000 cones/mm2, in the foveola [17]. This density is higher than in any other primate described so far, including humans, for which the highest reported individual value was 324,000 cones/mm2 [27]. Microspectrophotometry However, physiological characteristics of the Alouatta retina remain unknown. If the LCR controls which opsin is expressed, the dual LCRs in the Alouatta could provide a potential for mixed expression of two opsins in the same cone; this seems unlikely in face of the behavioral data, but the mechanisms for expression control in Alouatta is unknown. Physiologically, the characteristics of ganglion cells of the male Alouatta might be expected to resemble those of catarrhines. We here address these issues. Alouatta specimens are rarely available for physiological experi- ments. This brief note contains limited data from two male animals suggesting that pigment expression is indeed cone specific and that ganglion cells closely resemble the catarrhine pattern. Preliminary reports of these experiments have appeared elsewhere [29], [30]. Two eyes, one eye of each animal, were lightly fixed in 2% glutaraldehyde and subsequently examined. All procedures were carried out under dim red light. The eyes were enucleated, hemisected, and the anterior portion removed. One or two small pieces of retina, about 1 mm square, were taken from the fovea and parafoveal regions each, placed on a coverslip and mounted in saline containing 10% dextran, before being squashed with a second coverslip, which was sealed with wax. Although the tissue was in a relatively poor state of preservation, a small number of cone and rod outer segments were identified. g Microspectrophotometric recordings from these cells were performed in the conventional manner using a Liebman dual- beam microspectrophotometer [2], [6], [31]. Spectra were recorded at 2 nm intervals from 750 to 350 nm and from 351 to 749 nm on the return scan. The outward and return scans were averaged. A baseline spectrum was measured for each cell, with both beams in an unoccupied area close to the cell, and this was subtracted from the intracellular scan to derive the final spectrum. Two baseline scans were recorded for each cell and averaged. All cells were fully bleached with white light and post bleach spectra recorded. Records that either had very low absorbance or were clearly distorted were discarded. The wavelength of maximum absorbance (lmax) of both the absorbance spectra and difference spectra were determined by a standard computer program that best fits a visual pigment template to the right hand limb of the spectra [2], [6]. Introduction On the other hand, the retinal ganglion cell distribution in the Alouatta is similar to that observed in other diurnal platyrrhines and catarrhines [28]. On the day of experiment, the animal were sedated with an intramuscular injection of ketamine (ca 20 mg kg21) and then transferred to the Biological Sciences Institute of the Federal University of Para. Ethical statement All animal experiments were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23, revised 1996), and were approved by the Ethical Research Committee for Animal Experiments of the Institute of Biological Sciences, Federal University of Para´ (#MED004/2008). November 2014 \| Volume 9 \| Issue 11 \| e113321 Animals Two male Alouatta caraya, were obtained from the breeding colony of the Centro Nacional de Primatas – CENP (Ananindeua, Para´, Brazil) (Appendix S1). The animals were bred and kept in the CENP facilities until the day the experiment started. There, they were kept in housing conditions permitting social interaction with their conspecifics and submitted to an appropriate feeding A standardized computer program was used to estimate the wavelength of maximum absorbance (lmax) as fully detailed in Mollon and colleagues [6]. In summary, each of 20 relative absorbance values on the long-wave limb of the absorbance spectrum (corresponding to a 40 nm segment of the record and to November 2014 \| Volume 9 \| Issue 11 \| e113321 November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 2 Alouatta Color Vision [35]. Mean chromaticity was 595 nm, retinal illuminance levels was ca 2000 Td, but because of the small Alouatta eye the retinal flux per Troland is about 2 times that in humans. absorbance in the range approximately 45–90% of the maximum for the cell) was then referred to a standard template curve to obtain an estimate of the lmax: this operation amounts to finding the spectral location of the standard curve that gives the percentage absorbance value under consideration. The 20 individual estimates were then averaged to give the values given in the paper. The segment of the right hand limb of the spectrum was used because it includes the steepest part of the photopig- ment’s absorbance spectrum and so small changes in the wavelength correspond to large changes in absorbance, and secondly, the short-wave region in microspectrophotometric measurements is the most variable owing to factors including wavelength dependent scattering and the potential presence of photoproducts. For each cell, we recorded responses to four different stimulus series to extend cell classification and investigate the retinal mechanisms for Alouatta trichromacy. 1) Responses to 400 ms luminance and chromatic pulses were measured to assess the time course of responses; luminance pulses were incremental or decremental; chromatic pulses were redward or greenward. The cells’ spectral sensitivity and cone opponency were assessed by measuring responses to sinusoids in two different protocols. 2) In a modified HFP protocol, the relative modulation depths of a 638 nm red diode and a 554 nm green diode were varied while keeping mean chromaticity and mean luminance constants [36]. Electrophysiology 4) MC cells and PC cells modulation transfer functions (MTFs) were obtained by recording cell responses to sinusoidal stimuli at twelve temporal frequencies and multiple contrast levels. Lumi- nance modulation, with the red and green LEDs in phase, was employed for all cells. Responses of color opponent cells to chromatic modulation, with the LEDs out of phase, were also measured. For luminance modulation, luminance contrast was calculated as (Lmax 2 Lmin)/(Lmax+Lmin). For chromatic contrast, cone contrast was calculated in a similar manner, but using cone absorptions. Animals were initially anaesthetized with an intramuscular injection of ketamine (ca 20 mg kg21). Later, anesthesia was maintained by intravenous infusion of sufentanyl (0.5–4 mg kg21 h21). EEG and ECG were continuously monitored to ensure adequate depth of anesthesia and analgesia. Muscular paralysis was achieved by infusion of 5 mg kg21 h21 of gallamine triethiodide, and the animals were respired with O2 to which ca 1–2% CO2 had been added. End-tidal PCO2 was kept between 4 and 5% and body temperature maintained within the normal limits. The eye was prepared in a similar way to the eye of the macaque and Cebus, and recording of ganglion cell activity was performed as in those species [9], [33]. Briefly, the conjunctiva was sewn to a ring for stabilization and a cannula inserted behind the limbus. A tungsten-in-glass microelectrode was passed through the cannnula, and on touching the retinal surface the electrode was permitted to penetrate into the ganglion cell layer and the activity of a single neuron isolated. A two-dimensional gimbal system, rotating about the point of entry to the globe, permitted us to target different retinal regions. Following completion of record- ings, animals were killed with an overdose of barbiturate, Thionembutal (Abbott, Abbott Park, Illinois, USA), 35 mg/kg or higher. Death was assessed by cessation of ECG activity. For all sinusoidal modulation conditions, about 6 s of activity was averaged for each condition, and first and second harmonic amplitudes and phases were extracted. Animals Non-opponent cells show a null or response minimum at a particular 638/554 nm ratio while red-green opponent cells show a vigorous response at or close to that same ratio [9], [33], [36], [37]. Two to four (4.88–39 Hz) temporal frequencies were measured. Because the absorbance spectra of visual pigments become narrower towards shorter wavelengths, an empirical template has to be used to estimate the lmax. The template used in this study, and in all previous microspectrophotometric measurements from this lab recorded from both Old and New World monkeys, was based on the classical Dartnall nomogram expressed in units of lJ, an empirical relationship that can be used over the limited spectral range of primate rods and L and M cones. More recent empirical templates, such as those by Govardovskii and colleagues [32]), that can be used over the full spectral range, show, over the limited spectral range used here, little if any variation from the template employed in this study. 3) The cells’ spectral sensitivity and cone opponency were also investigated using a phase protocol [38], [39]. The relative phase of the luminance modulation of the 638 nm and 554 nm diodes was varied, with constant modulation amplitude. A phase of 6180 deg corresponds to chromatic modulation and of 0 deg to luminance modulation. For the phase protocols, MC cells were studied at 20 or 50% modulation contrast while PC cells were studied at 50 or 100% modulation contrast. Six temporal frequencies (1.22–39 Hz) were measured. November 2014 \| Volume 9 \| Issue 11 \| e113321 Microspectrophotometry Plots in (A), (B), and (D) show, respectively, rod, middle-wave sensitive cone, and long-wave sensitive cone pre-bleach (open symbols) and post-bleach (filled symbols) spectra. Histogram in (C) shows spectral distribution of lmax of individual rods and cones obtained from absorbance spectra. Figure 1. Absorbance spectra of visual pigments from Alouatta caraya. Plots in (A), (B), and (D) show, respectively, rod, middle-wave sensitive cone, and long-wave sensitive cone pre-bleach (open symbols) and post-bleach (filled symbols) spectra. Histogram in (C) shows spectral distribution of lmax of individual rods and cones obtained from absorbance spectra. doi:10.1371/journal.pone.0113321.g001 support for these classifications was provided by the quantitative analyses as described below. possibility of a limited degree of co-expression within each cone class. No published microspectrophotometric data from monkeys or humans has ever explicitly tested for co-expression. It would require very careful and somewhat difficult partial bleaching experiments due to the small spectral separation in these animals. We then measured cell responses to 400 ms step changes in luminance (increment or decrement) and color (redward or greenward pulses) at various contrast levels. Figure 2 shows the responses of an MC off-centre cell (A) and a +M-L PC off-centre cell (B). The MC cell showed strong phasic responses to luminance changes and little response to chromatic pulses. The PC cell showed strong tonic response to the green pulse, was strongly inhibited by the red pulse, and showed small tonic responses to luminance changes. These responses are very similar to those of macaque ganglion cells [33], [36], [37]. Microspectrophotometry Ten cone outer segments were analyzed and these fell into two clear populations, eight with lmax close to 530 nm (529.565.5 nm) and two with lmax around 557 nm (557.564.9 nm). All these outer segments were fully bleached after exposure to white light. No S cones were identified. Seven rod outer segments gave a mean lmax of 499.861.8 nm, but bleaching with white light revealed, in addition to the typical ‘retinal’ peak at about 370 nm, a stable photoproduct absorbing around 465 nm, presumably metarhodopsin III, which is some- times seen in lightly fixed tissue. The absorbance of the cone outer segments was low, 0.009 and 0.004 for the MWS and LWS cones, respectively, whereas the rods, which were better preserved, had a transverse absorbance of 0.029. A summary of the microspectro- photometric results is presented in Figure 1. These data are consistent with selective expression of the M and L opsins in different cones. Visual stimuli were presented using a Maxwellian view system with red and green diodes (LEDs) as light sources [8], [34]. The system could be rotated about the pupil in order to center the stimulus on a cell’s receptive field. The temporal waveforms for the LEDs were generated by a computer through 12-bit digital-to- analog converters. The LEDs were driven by a frequency- modulated pulse train which provided a highly linear relationship between driving voltage and light output. The emission spectra of the LEDs were measured with a spectroradiometer (Model pro- 703/PC, Photo Research, Burbank, CA, USA). The dominant wavelengths of the LEDs were 638 and 554 nm. The mean luminances of LEDs were set using heterochromatic flicker photometry (HFP) matches by an observer whose spectral luminosity function matched the 2 deg Judd’s spectral sensitivity Although the data are limited, the records fall into two clear spectral groups of L and M cones, but as with all primate microspectrophotometric data, we cannot explicitly exclude the November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 3 Alouatta Color Vision Figure 1. Absorbance spectra of visual pigments from Alouatta caraya. Plots in (A), (B), and (D) show, respectively, rod, middle-wave sensitive cone, and long-wave sensitive cone pre-bleach (open symbols) and post-bleach (filled symbols) spectra. Histogram in (C) shows spectral distribution of lmax of individual rods and cones obtained from absorbance spectra. doi:10.1371/journal.pone.0113321.g001 Figure 1. Absorbance spectra of visual pigments from Alouatta caraya. November 2014 \| Volume 9 \| Issue 11 \| e113321 Electrophysiology We recorded from 34 ganglion cells from the retinae of the two animals (Dataset S1). Cell receptive fields were plotted on a tangent screen and were located within 10 deg of the fovea. Initial cell classification was performed with flashed spots of different colors and gave similar results to those previously obtained from recordings from the retinae of macaques and Cebus [9], [33], [37], [40]. Cells that were classified as PC cells at this stage could be clearly identified by their vigorous response, either excitatory or inhibitory, to green and red lights. KC cells with excitatory S-cone input strongly responded to short wavelength light. Cells that were classified as MC cells could be identified by their good response to low achromatic contrast. Using these criteria, we identified 6 PC cells, 1 S-cone On cell, and 17 MC cells. The remaining 10 cells of our sample remained unclassified mostly because these cells were not recorded for the necessary time for clear classification. Further Figures 3 and 4 shows Alouatta MC and PC cell responses to heterochromatically modulating stimuli. As described in the Methods section, an amplitude protocol based on HFP (Figure 3) and a phase protocol (Figure 4) were used to characterize the spectral sensitivity and color-opponency of ganglion cells. Alouatta MC cells’ responses to the HFP protocol displayed a minimum when the contrast ratio of the counterphase-modulated red to green lights was changed. A ratio of unity would suggest a spectral sensitivity similar to the human photopic luminosity function (Vl). As an example, Figure 3A shows responses of an November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 4 Alouatta Color Vision ll to 9.8 Hz stimuli; response amplitude and phase are against the relative modulation of the red and green LEDs. presence of a substantial second-harmonic component with a amplitude that peaked near a contrast ratio of unity (Figure 3 2. Ganglion cell responses to luminance and color step changes. Responses to 50% contrast, 400 ms step changes in luminan ent or decrement) and color (redward or greenward pulses) of an MC off-centre cell (A) and a +M-L PC off-centre cell (B). The MC cell show esponses to luminance changes and little responses to chromatic pulses. The PC cell showed strong tonic responses to the greenward puls ngly inhibited by the redward pulse, and showed small tonic responses to luminance changes. Electrophysiology Histograms are averages of 20 sweeps; 4 m 371/journal.pone.0113321.g002 ouatta Co o s o e 2. Ganglion cell responses to luminance and color step changes. Responses to 50% contrast, 400 ms step changes in luminanc ment or decrement) and color (redward or greenward pulses) of an MC off-centre cell (A) and a +M-L PC off-centre cell (B). The MC cell showe responses to luminance changes and little responses to chromatic pulses. The PC cell showed strong tonic responses to the greenward puls rongly inhibited by the redward pulse, and showed small tonic responses to luminance changes. Histograms are averages of 20 sweeps; 4 ms .1371/journal.pone.0113321.g002 Figure 2. Ganglion cell responses to luminance and color step changes. Responses to 50% contrast, 400 ms step changes in luminance (increment or decrement) and color (redward or greenward pulses) of an MC off-centre cell (A) and a +M-L PC off-centre cell (B). The MC cell showed phasic responses to luminance changes and little responses to chromatic pulses. The PC cell showed strong tonic responses to the greenward pulse, was strongly inhibited by the redward pulse, and showed small tonic responses to luminance changes. Histograms are averages of 20 sweeps; 4 ms/ bin. doi:10.1371/journal.pone.0113321.g002 doi:10.1371/journal.pone.0113321.g002 MC cell to 9.8 Hz stimuli; response amplitude and phase are plotted against the relative modulation of the red and green LEDs. There is a minimal response of the MC cell first harmonic (filled diamonds) amplitude (upper panel), coinciding with a sudden shift in phase of about 180 deg (middle panel), at a relative amplitude of the red and green LEDs of about 2. Alouatta MC cells also showed a pronounced frequency-doubled response, expressed by the presence of a substantial second-harmonic component with an amplitude that peaked near a contrast ratio of unity (Figure 3A, upper and middle panels, empty squares) whilst phase remained stable. The histograms below the plots illustrate actual responses to two cycles of modulation (Figure 3A, lower panel). Arrows indicate the R/G values for the histograms. The frequency-doubled response is clear in the central histogram. November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 5 Alouatta Color Vision Figure 3. Ganglion cell responses to heterochromatic flicker photometry stimuli. (A) MC and (B) PC cell responses to heterochromatic stimuli. HFP protocol: The relative modulation depths of counterphase modulated 554 and 638 nm LEDs were manipulated. Electrophysiology In the HFP protocol, Alouatta MC cell responses displayed a minimum, coinciding with a sudden shift in phase, when the modulation ratio of the red and green lights was slightly larger than unity. The amplitudes of the second harmonic component peaked around a contrast ratio of unity, where it was substantially larger than the amplitude of the first harmonic component, whilst its phase remained stable. The resultant frequency doubled response can be observed in the corresponding histogram. Alouatta PC cells showed vigorous response and no phase shift for nearly all stimulus conditions. doi:10.1371/journal.pone.0113321.g003 response shaping (Figure 3B, upper and middle panels, empty squares). Again, the histograms set below the plots illustrate actual responses to two cycles of modulation (Figure 3B, lower panel). On average, response minima of MC cells from the two animals tested was close to a luminance ratio between the red and green diodes of one (mean: 1.08, s.d.: 0.26, n = 16) but the two animals differed significantly in mean null values (means 1.25 (n = 9), 0.86 (n = 7), p,0.02, t test). It is established that HFP luminance nulls of humans is related to relative L/M cone numerosity in the retina [41], [42], but the mechanisms controlling relative numerosity are poorly understood; with the different genetic foundation in the Alouatta this problem is compounded. In conclusion, the responses of Alouatta MC cells and PC cells to the HFP protocol were similar to those of their counterparts found in the retinas of macaques and trichromatic Cebus [9], [33], [36], [37]. In the phase protocol, rather than changing the relative amplitude of modulated lights, their relative phase is varied. This protocol can determine relative weights and temporal properties of cone mechanisms in ganglion cell responses [39]. At 10 Hz and above, macaque MC cells respond to the phase protocol with a first-harmonic maximum when the green and red lights are in phase; their responses reach a minimum when the green and red stimuli are modulated in counterphase. At lower temporal frequencies, the phase of minimum response moves away from zero as in human psychophysical results [38], [39], [43]. At higher frequencies, a similar pattern was found in Alouatta MC cells, as shown in Figure 4A. Response amplitude and phase have been plotted as a function of the relative phase of the two LEDs. Electrophysiology MC and PC cells were stimulated at 9.8 Hz and 1.22 Hz, respectively. Stimulus size was, 4 deg mean retinal illuminance was 2000 Td. Each data point was obtained from the Fourier analysis on averaged responses in episodes each lasting 6 s. Top and middle panels: response amplitude and phase as a function of the red/ green amplitude ratio, respectively. Filled diamonds and empty squares represent the parameters for response first and second harmonics, respectively. Bottom panels: the histograms illustrate actual averaged responses to two modulation cycles, the arrows point at the employed red/ green ratios for the histograms, and the red and green curves represent the temporal luminance modulation of the red and green lights for each stimulus condition. In the HFP protocol, Alouatta MC cell responses displayed a minimum, coinciding with a sudden shift in phase, when the modulation ratio of the red and green lights was slightly larger than unity. The amplitudes of the second harmonic component peaked around a contrast ratio of unity, where it was substantially larger than the amplitude of the first harmonic component, whilst its phase remained stable. The resultant frequency doubled response can be observed in the corresponding histogram. Alouatta PC cells showed vigorous response and no phase shift for nearly all stimulus conditions. doi:10.1371/journal.pone.0113321.g003 Figure 3. Ganglion cell responses to heterochromatic flicker photometry stimuli. (A) MC and (B) PC cell responses to heterochromatic stimuli. HFP protocol: The relative modulation depths of counterphase modulated 554 and 638 nm LEDs were manipulated. MC and PC cells were stimulated at 9.8 Hz and 1.22 Hz, respectively. Stimulus size was, 4 deg mean retinal illuminance was 2000 Td. Each data point was obtained from the Fourier analysis on averaged responses in episodes each lasting 6 s. Top and middle panels: response amplitude and phase as a function of the red/ green amplitude ratio, respectively. Filled diamonds and empty squares represent the parameters for response first and second harmonics, respectively. Bottom panels: the histograms illustrate actual averaged responses to two modulation cycles, the arrows point at the employed red/ green ratios for the histograms, and the red and green curves represent the temporal luminance modulation of the red and green lights for each stimulus condition. November 2014 \| Volume 9 \| Issue 11 \| e113321 Electrophysiology MC and PC cells were stimulated at 9.8 Hz and 1.22 Hz, respectively. Stimulus size was 4 deg, mean retinal illuminance was 2000 Td. Response amplitudes (top panels) and phases (middle panels) are shown as a function of the phase difference between the luminance modulation in the red and green LEDs. Filled diamonds and empty squares represent the parameters for response first and second harmonics, respectively, extracted by Fourier analysis. Bottom panels: the histograms illustrate actual responses to two cycles of modulation, the arrows indicate the red/green phase difference for the histograms, and the red and green curves represent how the phase of the red and green lights changed for each stimulus condition. Luminance modulation corresponds to a relative phase of 0 deg, chromatic modulation to a relative phase of 6180 deg. MC cells responded to the phase protocol with a maximal response first harmonic amplitude when the green and red lights were modulated in phase and their response phase changed continuously with the phase difference between the modulation in the two LEDs. PC cells displayed a minimal response when the green and red lights were modulated in phase and their response phases changed abruptly in the region of minimal response. doi:10.1371/journal.pone.0113321.g004 Alouatta, a highly significant difference (p,0.001, t test). This would suggest a species difference. is a response minimum near +/2180 deg. The response phase changes gradually with a more rapid transition near +/2180 deg. The second-harmonic response is seen to be very vigorous at this phase. In the macaque [39], second-harmonic response compo- nents were sometimes seen but were seldom large enough to disturb analysis of first-harmonic components. In the sample of Alouatta cells, frequency-doubled components were very promi- nent and made analysis of first-harmonic components noisy, especially at lower temporal frequencies. To validate this observation, we compared the ratio of luminance 1st harmonic to chromatic 2nd harmonic responses for samples of cells from the two species, using very similar stimuli (50% modulation contrast, 4.88 Hz, same LED peak wavelengths and mean luminance; macaque data were taken from previous work [34]. Ratios were 2.65 (s.d. 1.42, n = 13) for macaque and 0.758 (sd. 0.31, n = 13) for is a response minimum near +/2180 deg. The response phase changes gradually with a more rapid transition near +/2180 deg. The second-harmonic response is seen to be very vigorous at this phase. Electrophysiology There The frequency-doubled response to chromatic modulation of MC cells in the Alouatta was very obvious as in the example in Figure 3. As in the macaque [36], it was not present at higher temporal frequencies (39 Hz), with some cell-to-cell variability. This is taken up below. Alouatta PC cells showed vigorous responses to nearly all heterochromatic stimuli; the first harmonic component to 1.22 Hz stimuli (Figure 3B filled diamonds) displayed a broad peak at ratios around unity (upper panel). The phases were similar for all ratios (middle panel). The second harmonic-response follows the first harmonic amplitude but is always smaller and thus is related to November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 6 Alouatta Color Vision Figure 4. Ganglion cell response to stimulus phase changes. (A) MC and (B) PC cell responses to heterochromatic stimuli. Phase protocol. The relative phases of the 554 and 638 nm LEDs were modulated with fixed modulation depths. MC and PC cells were stimulated at 9.8 Hz and 1.22 Hz, respectively. Stimulus size was 4 deg, mean retinal illuminance was 2000 Td. Response amplitudes (top panels) and phases (middle panels) are shown as a function of the phase difference between the luminance modulation in the red and green LEDs. Filled diamonds and empty squares represent the parameters for response first and second harmonics, respectively, extracted by Fourier analysis. Bottom panels: the histograms illustrate actual responses to two cycles of modulation, the arrows indicate the red/green phase difference for the histograms, and the red and green curves represent how the phase of the red and green lights changed for each stimulus condition. Luminance modulation corresponds to a relative phase of 0 deg, chromatic modulation to a relative phase of 6180 deg. MC cells responded to the phase protocol with a maximal response first harmonic amplitude when the green and red lights were modulated in phase and their response phase changed continuously with the phase difference between the modulation in the two LEDs. PC cells displayed a minimal response when the green and red lights were modulated in phase and their response phases changed abruptly in the region of minimal response. doi:10.1371/journal.pone.0113321.g004 Figure 4. Ganglion cell response to stimulus phase changes. (A) MC and (B) PC cell responses to heterochromatic stimuli. Phase protocol. The relative phases of the 554 and 638 nm LEDs were modulated with fixed modulation depths. November 2014 \| Volume 9 \| Issue 11 \| e113321 Electrophysiology In the macaque [39], second-harmonic response compo- nents were sometimes seen but were seldom large enough to disturb analysis of first-harmonic components. In the sample of Alouatta cells, frequency-doubled components were very promi- nent and made analysis of first-harmonic components noisy, especially at lower temporal frequencies. To validate this observation, we compared the ratio of luminance 1st harmonic to chromatic 2nd harmonic responses for samples of cells from the two species, using very similar stimuli (50% modulation contrast, 4.88 Hz, same LED peak wavelengths and mean luminance; macaque data were taken from previous work [34]. Ratios were 2.65 (s.d. 1.42, n = 13) for macaque and 0.758 (sd. 0.31, n = 13) for PC cells responded to the phase protocol with a minimum when the green and red lights were in phase; their responses increased with increasing phase difference between the green and red stimuli and were maximal when the two modulated in counterphase (Figure 4B, upper panel). In addition, the cell response showed an abrupt change in phase in the region of minimum response (Figure 4B, middle panel). These results suggest responses of Alouatta MC and PC cells were similar to those of the corresponding ganglion cells found in the retinas of macaques [39] and trichromatic Cebus [9], although differences in detail are likely to be present. There appeared to be more variability in the opponent weighting for PC cells than in similar macaque data, but our cell sample was not large. November 2014 \| Volume 9 \| Issue 11 \| e113321 November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 7 Alouatta Color Vision To explore further the comparison between Alouatta ganglion cell responses on the one hand and those of ganglion cells in macaques and in trichromatic female platyrrhines on the other, we employed sinusoidal luminance and chromatic modulation at different temporal frequencies and contrasts. Figure 5A–B illus- trates response amplitudes and phases of Alouatta MC (Figure 5A) and PC cells (Figure 5B) as a function of contrast; luminance modulation was used for the MC cell and chromatic modulation for the PC cell. Data representing response amplitude as a function of contrast were fitted with Naka-Rushton functions [44]. Electrophysiology MC cell vigorously responded to low levels of luminance contrast at frequencies of 5–40 Hz, but responses rapidly saturated accom- panied by response phase advancement, especially at intermediate temporal frequencies (9.8 Hz), as seen in the phase plot. PC cells were most sensitive to chromatic contrast (Figure 5B), but responses did not exhibit saturation or phase advance when contrast was increased. Figure 5C illustrates the averaged tempo- ral MTFs that were obtained from Alouatta MC cells (n = 7) and PC cells (n = 4). For both cell classes, we obtained luminance temporal MTFs and for PC cells we also obtained red-green temporal MTFs. The MTFs show the contrast gain as a function of temporal frequency, contrast gain being the initial slope (at zero contrast) of the Naka-Rushton functions fitted to the amplitude- versus-contrast data, cone contrast being used for the red-green temporal MTFs. MC cells were much more sensitive than PC cells to temporal luminance modulation across the temporal frequency range studied and responded vigorously to high temporal frequencies. On the other hand, PC cells were most sensitive to red-green contrast, especially at low and intermediate temporal frequencies. and responded to the HFP protocol with an amplitude peak near equal luminance. In addition, two specific features found in macaque ganglion cells were also encountered. MC cells responded to equal-luminance red-green modulation with twice the stimulation frequency [36]; this effect was marked in Alouatta MC cells. Also, MC cells showed response saturation and phase advancement with increasing contrast whereas PC cells did not. These latter features of MC cells are characteristic of a contrast gain control mechanisms [45], [46], and found in MC cells but not in PC cells of catarrhines [47], [48] and platyrrhines [8], [9]. Many properties are shared by the MC and PC cells of Alouatta, macaques [33], [37], and trichromatic female Callithrix [8] and Cebus [9]. The MC and PC cells of dichromatic male or female Callithrix [8] and Cebus [9] also exhibit the similar responses to luminance-modulated stimuli as in their trichromatic conspecifics. In conclusion, although a relatively small number of cells were measured, they displayed response properties that closely match those of other trichromatic species and individuals strongly suggesting that the processing in the retina of Alouatta is very similar. Electrophysiology Alouatta PC cells comprised the same subclasses found in other trichromats: some were excited by red light and inhibited by green light, while others were excited by green light and inhibited by red light. These responses reflected the excitatory or inhibitory inputs they received from M cones (530 nm absorption peak) or L cones (558 nm absorption peak). No study has yet morphologically identified Alouatta bipolar cells, but we suggest from the responses obtained from PC cells that single-cone midget bipolar cells similar to those that have been found in catarrhines [49] and platyrrhines [50], [51] will be present in the Alouatta retina. As complete trichromacy seems to have evolved relatively recently in Alouatta (with a different origin compared to catarrhines [4], [15]), it is likely that the receptoral mechanisms were able to take over pre-existent post-receptoral retinal wiring, already utilized in females of the species, and to use them to send trichromatic signals to higher visual centers. Physiological substrate for Alouatta trichromacy This work provides direct physiological data on the functional organization of the Alouatta retina for comparison with data in other species. The comparison may shed light on basic principles of primate retinal organization and on possible specific adaptations to trichromatic vision in different primate species. We therefore conclude that the functional organization of the retina is very similar in all anthropoid primates and was probably present before the split between catarrhines and platyrrhines 40 million years ago (MYA). There may well be differences in detail, so that the frequency-doubled MC response to chromatic modulation was very obvious in Alouatta, compared to earlier recordings in trichromatic Callithrix and Cebus. Another feature of trichromatic color vision is cone specificity in the surrounds of PC- cells of macaques [52], [53]. This is present to some degree in the Callithrix [54] but the situation in Alouatta is unknown. We stress that quantitative differences between retinae of different primate species are likely. We have found microspectrophotometric signatures of two populations of middle-to-long wavelength sensitive cones bearing separate expressions of photopigments with absorption peaks close to 530 nm and 558 nm absorption peaks. The presence of these two cone types is the receptoral prerequisite for a red-green color opponent pathway in the primate retina [1], [2], [5]. Cone-specific expression of M and L opsins is not unexpected from ERG and behavioral data [12], but it remains unclear how this occurs when each opsin gene has its own LCR. In addition, we have found post-receptoral red-green color opponent mechanisms with very similar properties of those previously found in macaques [33], [37], trichromatic female Callithrix [8] and trichromatic female Cebus [9]. Although the cell sample was limited due to the small number of animals available, the general features found in the macaque were apparent. Discussion Physiological substrate for Alouatta trichromacy Contrast sensitivity of Alouatta retinal ganglion cells Contrast sensitivity of Alouatta retinal ganglion cells Temporal MTFs (contrast gain as a function of temporal frequency) obtained from Alouatta MC and PC cells followed the same general pattern of temporal MTFs obtained from macaques [34] and Cebus [9]. The low luminance contrast gain of PC cells observed in trichromatic primates have been considered to be largely due to mutual cancellation of the opponent cone signals [33], [34], [37]. PC cells of dichromatic and trichromatic Cebus monkeys showed similar low gain [9]. PC cells of dichromatic platyrrhines might be described as color blind versions of cone- opponent PC cells of trichromatic primates. The functional significance played by such a numerous, luminance contrast insensitive cell class remains uncertain. In any event, PC cells in male Alouatta appear to have properties as in the macaque. November 2014 \| Volume 9 \| Issue 11 \| e113321 Physiological properties of Alouatta retinal ganglion cells Physiological properties of Alouatta retinal ganglion cells Alouatta MC and PC ganglion cells exhibit similarities with MC and PC ganglion cells of catarrhine primates. To list these similarities: MC cells were sensitive to luminance contrast, exhibited phasic responses to luminance steps and responded to the phase protocol and HFP protocols as in the macaque. PC cells were less sensitive to luminance contrast, exhibited tonic responses to chromatic pulses, responded to the phase protocol with a minimum response when the green and red lights were in phase November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org 8 Alouatta Color Vision Figure 5. Ganglion cell response to stimulus contrast. (A–B) Response amplitude (left panels) and phase (right panels) as a function of stimulus luminance contrast for Alouatta MC cell (A) and PC cell (B). The results obtained with three temporal frequencies are given: 1.2 Hz (filled squares), 9.8 Hz (filled triangles), and 39 Hz (filled diamonds) for the MC cell; 1.2 Hz (filled squares), 9.8 Hz (filled triangles), and 19.4 Hz (filled diamonds) for the PC cell. Data representing response amplitude as a function of contrast have been fitted with Naka-Rushton functions. MC cell vigorously responded to low levels of luminance contrast, but responses rapidly saturated accompanied by advancement in response phase, especially at intermediate and high temporal frequencies. PC cells were quite insensitive to low levels of luminance contrast, but the responses did not exhibit saturation or phase advancement. (C) Temporal modulation transfer functions (temporal MTFs) for Alouatta MC cells (average of 7 cells) stimulated with luminance sinusoids (filled squares) and a PC cell stimulated with luminance (empty diamonds) and averaged responses of 4 cells to red-green (filled diamonds) sinusoids. Contrast gain was defined as the initial slope of the Naka-Rushton functions fitted to the amplitude versus contrast data such as those illustrated in the left panels (A–B). Michelson contrast and cone contrast were used for the luminance and red-green chromatic temporal MTFs, respectively. MC cells were much more sensitive than PC cells to temporal luminance modulation at all the temporal frequencies range and vigorously responded to very high temporal frequencies. On the other hand, PC cells were very sensitive to red-green contrast, especially at low and intermediate temporal frequencies. doi:10.1371/journal.pone.0113321.g005 Figure 5. Ganglion cell response to stimulus contrast. Physiological properties of Alouatta retinal ganglion cells (A–B) Response amplitude (left panels) and phase (right panels) as a function of stimulus luminance contrast for Alouatta MC cell (A) and PC cell (B). The results obtained with three temporal frequencies are given: 1.2 Hz (filled squares), 9.8 Hz (filled triangles), and 39 Hz (filled diamonds) for the MC cell; 1.2 Hz (filled squares), 9.8 Hz (filled triangles), and 19.4 Hz (filled diamonds) for the PC cell. Data representing response amplitude as a function of contrast have been fitted with Naka-Rushton functions. MC cell vigorously responded to low levels of luminance contrast, but responses rapidly saturated accompanied by advancement in response phase, especially at intermediate and high temporal frequencies. PC cells were quite insensitive to low levels of luminance contrast, but the responses did not exhibit saturation or phase advancement. (C) Temporal modulation transfer functions (temporal MTFs) for Alouatta MC cells (average of 7 cells) stimulated with luminance sinusoids (filled squares) and a PC cell stimulated with luminance (empty diamonds) and averaged responses of 4 cells to red-green (filled diamonds) sinusoids. Contrast gain was defined as the initial slope of the Naka-Rushton functions fitted to the amplitude versus contrast data such as those illustrated in the left panels (A–B). Michelson contrast and cone contrast were used for the luminance and red-green chromatic temporal MTFs, respectively. MC cells were much more sensitive than PC cells to temporal luminance modulation at all the temporal frequencies range and vigorously responded to very high temporal frequencies. On the other hand, PC cells were very sensitive to red-green contrast, especially at low and intermediate temporal frequencies. doi:10.1371/journal.pone.0113321.g005 References Yeh T, Lee BB, Kremers J, Cowing JA, Hunt DM, et al. (1995a) Visual responses in the lateral geniculate of dichromatic and trichromatic marmosets (Callithrix jacchus). J Neurosci 15: 7892–7904. 30. Saito CA, Silveira LCL, da Silva Filho M, Lee BB, Kremers J (2005) Physiology of howler monkey retinal ganglion cells: temporal achromatic contrast sensitivity. ARVO Abstracts, Inv Ophthalmol Vis Sci 46: E-Abstract 2240– B1009. 9. Lee BB, Silveira LCL, Yamada ES, Hunt DM, Kremers J, et al. (2000) Visual responses of ganglion cells of a New World primate, the capuchin monkey, Cebus apella. J Physiol (Lond) 528: 573–590. 31. Liebman PA, Entine G (1964) Sensitive low-light level microspectrophotometer detection of photosensitive pigments of retinal cones. J Opt Soc Am 54: 1451– 1459. 10. Hashemi-Nezhad M, Blessing E, Dreher B, Martin P (2008) Segregation of short-wavelength sensitive (‘‘blue’’) cone signals among neurons in the lateral geniculate nucleus and striate cortex of marmosets. Vision Res 48: 2604–2614. 32. Govardovskii VI, Fyhrquist N, Reuter T, Kuzmin DG, Donner K (2000) In search of the visual pigment template. Vis Neurosci 17: 509–528. 11. Martin PR, Blessing E, Buzas P, Szmajda B, Forte J (2011). Transmission of colour and acuity signals by parvocellular cells in marmoset monkeys. J Physiol (Lond) 589: 2795–2812. 33. Lee BB, Martin PR, Valberg A (1989a) Sensitivity of macaque retinal ganglion cells to chromatic and luminance flicker. J Physiol (Lond) 414: 223–243. 34. Lee BB, Pokorny J, Smith VC, Martin PR, Valberg A (1990) Luminance and chromatic modulation sensitivity of macaque ganglion cells and human observers. J Opt Soc Am A 7: 2223–2236. 12. Jacobs GH, Neitz M, Deegan JF, Neitz J (1996a) Trichromatic colour vision in New World monkeys. Nature 382: 156–158. y 13. Arau´jo AC, Didonet JJ, Arau´jo CS, Saletti PG, Borges TR, et al. (2008) Color vision in the black howler monkey (Alouatta caraya). Vis Neurosci 25: 243–248. 35. Judd DB (1951) Colorimetry and artificial daylight. In: Technical Committee No. 7, Report of Secretariat United States Commission, International Commission on Illumination, Twelfth Session. Stockholm, Sweden. 1–60. 14. Matsushita Y, Oota H, Welker BJ, Pavelka MS, Kawamura S (2014) Color vision variation as evidenced by hybrid L/M opsin genes in wild populations of trichromatic Alouatta New World monkeys. Int J Primatol 35: 71–87. 36. Lee BB, Martin PR, Valberg A (1988) The physiological basis of heterochro- matic flicker photometry demonstrated in the ganglion cells of the macaque retina. References 1. Nathans J, Thomas D, Hogness DS (1986) Molecular genetics of human color vision: The genes encoding blue, green, and red pigments. Science 232: 193– 202. 1. Nathans J, Thomas D, Hogness DS (1986) Molecular genetics of human color vision: The genes encoding blue, green, and red pigments. Science 232: 193– 202. 25. Solomon SG, White AJ, Martin PR (1999) Temporal contrast sensitivity in the lateral geniculate nucleus of a New World monkey, the marmoset Callithrix jacchus. J Physiol (Lond) 517: 907–917. 2. Bowmaker JK, Astell S, Hunt DM, Mollon JD (1991) Photosensitive and photostable pigments in the retinae of Old World monkeys. J Exp Biol 156: 1– 19. 26. Usrey WM, Reid RC (2000) Visual physiology of the lateral geniculate nucleus in two species of NewWorld monkey: Saimiri sciureus and Aotus trivirgatus. J Physiol (Lond) 523: 755–769. 3. Ibbotson RE, Hunt DM, Bowmaker JK, Mollon JD (1992) Sequence divergence and copy number of the middle- and long-wave photopigment genes in Old World monkeys. Proc R Soc Lond B 247: 145–154. 27. Curcio CA, Sloan KR, Kalina RE, Hendrickson AE (1990) Human photoreceptor topography. J Comp Neurol 292: 497–523. 28. de Athaide LM, Muniz JAPC, Gomes BD, Finlay BL, Silveira LCL (2009) Distribution and topography of the ganglion cells and displaced amacrine cells in the howler monkey retina (Alouatta caraya). In: Nascimento SMC, Almeida VNM, Castelo Branco M, editors. Abstracts Book of the 20th Symposium of the International Colour Vision Society (ICVS). Braga, Portugal: Universidade do Minho, Centro de Fı´sica. 85. 4. Hunt DM, Dulai KS, Cowing JA, Julliot C, Mollon JD, et al. (1998) Molecular evolution of trichromacy in primates. Vision Res 38: 3299–3306. evolution of trichromacy in primates. Vision Res 38: 3299–3306. 5. Jacobs GH (2007) New World monkeys and color. Int J Primatol 28: 729–759. 6. Mollon JD, Bowmaker JK, Jacobs GH (1984) Variations of colour vision in a New World primate can be explained by polymorphism of retinal photopig- ments. Proc R Soc Lond B 222: 373–399. 29. Saito CA, da Silva Filho M, Lee BB, Bowmaker JK, Kremers J, et al. (2004) Alouatta trichromatic color vision – single-unit recording from retinal ganglion cells and microspectrophotometry. ARVO Abstracts, Inv Ophthalmol Vis Sci 45: E-Abstract 4276–B737. 7. Goulart PRK, Bonci DMO, Galva˜o OF, Silveira LCL, Ventura DF (2013) Color discrimination in the tufted capuchin monkey, Sapajus spp. PLOS ONE 8(4): e62255O, 1–9. ( ) 8. References J Physiol (Lond) 404: 323–347. 15. Dulai KS, von Dornum M, Mollon JD, Hunt DM (1999). The evolution of trichromatic color vision by opsin gene duplication in New World and Old World primates. Genome Res 9: 629–638. y 37. Lee BB, Martin PR, Valberg A (1989b) Amplitude and phase of responses of macaque retinal ganglion cells to flickering stimuli. J Physiol (Lond) 414: 245– 263. 16. Silveira LCL, Picanc¸o-Diniz CW, Sampaio LFS, Oswaldo-Cruz E (1989) Retinal ganglion cell distribution in the Cebus monkey: a comparison with the cortical magnification factors. Vision Res 29: 1471–1483. 38. Lindsey DT, Pokorny J, Smith VC (1986) Phase dependent sensitivity to heterochromatic flicker. J Opt Soc Am A 3: 921–927. 17. Franco ECS, Finlay BL, Silveira LCL, Yamada ES, Crowley JC (2000) Conservation of absolute foveal area in New World primates: a constraint on eye size and conformation. Brain Behav Evol 56: 276–286. 39. Smith VC, Lee BB, Pokorny J, Martin PR, Valberg A (1992) Responses of macaque ganglion cells to the relative phase of heterochromatically modulated lights. J Physiol (Lond) 458: 191–221. 18. Silveira LCL (2004) Comparative study of the primate retina. In: Kaas JH, Collins CE, editors. The Primate Visual System. Boca Raton, Florida, USA: CRC Press. 29–51. 40. Silveira LCL, Lee BB, Yamada ES, Kremers J, Hunt DM, et al. (1999) Ganglion cells of a short wavelength sensitive cone pathway in New World monkeys: morphology and physiology. Vis Neurosci 16: 333–343. 19. Silveira LCL, Gru¨nert U, Kremers J, Lee BB, Martin PR (2005) Comparative anatomy and physiology of the primate retina. In: Kremers J, editor. The primate visual system: A comparative approach. Chichester, England, UK: John Wiley & Sons. 127–160. 41. Brainard DH, Roorda A, Yamauchi Y, Calderone JB, Metha AB, et al. (2000) Functional consequences of the relative numbers of L and M cones. J Opt Soc Am A 17: 607–614. 42. Kremers J, Scholl HPN, Knau H, Berendschot TTJM, Usui T, et al. (2000) L/ M-cone ratios in human trichromats assessed by psychophysics, electroretinog- raphy and retinal densitometry. J Opt Soc Am A 17: 517–526. y 20. Jacobs GH, De Valois RL (1965) Chromatic opponent cells in squirrel monkey lateral geniculate nucleus. Nature 206: 487–489. raphy and retinal densitometry. J Opt Soc Am A 17: 517–526. g 21. Kremers J, Weiss S (1997). Receptive field dimensions of lateral geniculate cells in the common marmoset (Callithrix jacchus). Vision Res 37: 2171–2181. Evolution of primate trichromacy generally related to foraging behavior. Stephen Polyak was probably the first to explicitly point out that detecting orange and red fruits against the green foliage would benefit from trichromacy [58]. There have been several studies measuring fruits and leaves surface reflectance and quantifying primate foraging behavior in natural conditions which were then used to provide quantitative data for theories of trichromacy evolution [59–64]. These ideas focused in the evolutionary advantage of acquiring a better distinction of reddish fruits against green foliage [65], [66], yellowish dappled with red young nutritive leaves against mature green leaves [62], and reddish falling leaves against healthy green leaves [67]. These possibilities are not mutually exclusive. It should be observed that both catarrhines and Alouatta distinguish themselves from the majority of platyrrhines by having a diet mainly based on fruits and on leaves [67], [68]. For the Alouatta, there have been reports from field studies of their foraging behavior associated with quantitative analysis in the color domain Among the possible scenarios for the evolution of trichromatic vision in primates [9], [55], a common view is that polymorphic color vision arose in prosimians and passed to anthropoids before the split between catarrhines and platyrrhines. Full trichromacy then arose independently at least twice, in catarrhines and in the Alouatta [4], [56], [57]. The absorption peaks of the two LWS cone opsins found in the two groups are usually similar (Jacobs et al. [12] and the present study), possibly indicating an optimization for red-green color opponency in regular trichro- mats. Another possibility, originally proposed by Mollon [55], and supported by others based on molecular data [4], [14], [15], is that trichromatic vision in platyrrhines and catarrhines evolved separately. However, the close similarities in retinal anatomy and physiology between the two groups, further underlined here, make, in our view, parallel evolution unlikely. There are several hypotheses about the evolutionary pressures that might have driven primates to trichromacy and they are November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 9 Alouatta Color Vision of surface reflectance underlying the importance of selecting both the appropriate fruits [60], [61] and leaves [64] for their diet. Checklist S1 The ARRIVE Guidelines Checklist. (DOC) Checklist S1 The ARRIVE Guidelines Checklist. (DOC) Checklist S1 The ARRIVE Guidelines Checklist. (DOC) Appendix S1 Alouatta housing conditions including feeding regimens and environmental enrichment. (DOC) Conceived and designed the experiments: LCLS JK JKB BBL. Performed the experiments: LCLS CAS MSF JK JKB BBL. Analyzed the data: LCLS CAS JK JKB BBL. Contributed reagents/materials/analysis tools: LCLS CAS MSF JK JKB BBL. Contributed to the writing of the manuscript: LCLS CAS MSF JK JKB BBL. Author Contributions Appendix S1 Alouatta housing conditions including feeding regimens and environmental enrichment. (DOC) References Silveira LCL, Lee BB, Yamada ES, Kremers J, Hunt DM (1998) Post-receptoral mechanisms of colour vision in New World primates. Vision Res 38: 3329–3337. 62. Lucas PW, Darvell BW, Lee PKD, Yuen TDB, Choong MF (1998) Colour cues for leaf food selection by long-tailed macaques (Macaca fascicularis) with a new suggestion for the evolution of trichromatic colour vision. Folia Primatologica 69: 139–154. 51. Chan TL, Martin PR, Clunas N, Gru¨nert U (2001) Bipolar cell diversity in the primate retina: morphologic and immunocytochemical analysis of a New World monkey, the marmoset Callithrix jacchus. J Comp Neurol 437: 219–239. 52. Martin PR, Lee BB, White AJR, Solomon SG, Ru¨ttiger L (2001) Chromatic sensitivity of ganglion cells in the peripheral primate retina. Nature 410: 933– 936. 63. Lucas PW, Dominy NJ, Riba-Hernandez P, Stoner KE, Yamashita N, et al. (2003) Evolution and function of routine trichromatic vision in primates. Evolution 57: 2636–2643. 53. Solomon SG, Lee BB, White AJ, Ru¨ttiger L, Martin PR (2005) Chromatic organization of ganglion cell receptive fields in the peripheral retina. J Neurosci 25: 4527–4539. 64. Stoner KE, Riba-Hernandez P, Lucas PW (2005) Comparative use of color vision for frugivory by sympatric species of platyrrhines. Am J Primatology 67: 399–409. 54. Szmajda BA, Gru¨nert U, Martin PR (2005) Mosaic properties of midget and parasol ganglion cells in the marmoset retina. Vis Neurosci 22: 395–404. 54. Szmajda BA, Gru¨nert U, Martin PR (2005) Mosaic properties 65. Snodderly DM (1979) Visual discriminations encountered in food foraging by a neotropical primate: implications for the evolution of color vision. In: Burtt EHJ, editor. The behavioral significance of color. New York, New York, USA: Garland STMP Press. 238–279. 55. Mollon JD (1991) Uses and evolutionary origins of primate color vision. In: Cronly-Dillon JR, Gregory RL, editors. Evolution of the eye and visual system. Houndmills, Basingstoke, Hampshire, England, UK: MacMillan Press. 306–319. 66. Mollon JD, Jordan G (1988/1989) Eine evolutiona¨re Interpretation des menschlichen Farbensehens. Die Farbe 35/36: 139–170. g p g 56. Kainz PM, Neitz J, Neitz M (1998) Recent evolution of uniform trichromacy in a New World monkey. Vision Res 38: 3315–3320. 67. Kremers J, Silveira LCL, Yamada ES, Lee BB (1999) The ecology and evolution of primate colour vision. In: Gegenfurtner KR, Sharpe LT, editors. Color vision: From genes to perception. Cambridge, England, UK: Cambridge University Press. 123–142. 57. Bowmaker JK (2008) Evolution of vertebrates visual pigments. Vision Res 48: 2022–2041. 58. References 43. Smith VC, Pokorny J (1975). Spectral sensitivity of the foveal cone photopigments between 400 and 500 nm. Vision Res 15: 161–171. 43. Smith VC, Pokorny J (1975). Spectral sensitivity of the in the common marmoset (Callithrix jacchus). Vision Res 37: 217 photopigments between 400 and 500 nm. Vision Res 15: 161–17 22. Kremers J, Weiss S, Zrenner E (1997) Temporal properties of marmoset lateral geniculate cells. Vision Res 37: 2649–2660. 44. Naka KI, Rushton WAH (1966) S-potentials from colour units in the retina of fish (Cyprinidae). J Physiol (Lond) 185: 536–555. 23. White AJR, Wilder HD, Goodchild AK, Sefton AJ, Martin PR (1998) Segregation of receptive field properties in the lateral geniculate nucleus of a New-World Monkey, the marmoset Callithrix jacchus. J Neurophysiol 80: 2063– 2076. 45. Shapley RM, Victor JD (1978) The effect of contrast on the transfer characteristics of cat retinal ganglion cells. J Physiol (Lond) 285: 275–298. . Shapley RM, Victor JD (1979) Nonlinear spatial summation and 46. Shapley RM, Victor JD (1979) Nonlinear spatial summation and the contrast gain control of cat retinal ganglion cells. J Physiol (Lond) 290: 141–161. 24. White AJR, Solomon SG, Martin PR (2001) Spatial properties of koniocellular cells in the lateral geniculate nucleus of the marmoset Callithrix jacchus. J Physiol (Lond) 533: 519–535. 47. Benardete EA, Kaplan E, Knight BW (1992) Contrast gain control in the primate retina: P cells are not X like, some M cells are. Vis Neurosci 8, 483–486. PLOS ONE \| www.plosone.org November 2014 \| Volume 9 \| Issue 11 \| e113321 10 November 2014 \| Volume 9 \| Issue 11 \| e113321 Alouatta Color Vision 48. Yeh T, Lee BB, Kremers J (1995b) The temporal response of ganglion cells of the macaque retina to conespecific modulation. J Opt Soc Am A 12, 456–464. 60. Regan BC, Julliot C, Simmen B, Vie´not F, Charles-Dominique P, et al. (1998) Frugivory and colour vision in Alouatta seniculus, a trichromatic platyrrhine monkey. Vision Res 38: 3321–3327. 49. Boycott BB, Wa¨ssle H (1991) Morphological classification of bipolar cells of the primate retina. Eur J Neurosci 3: 1069–1088. y 61. Regan BC, Julliot C, Simmen B, Vie´not F, Charles-Dominique P, et al. (2001) Fruits, foliage and the evolution of primate colour vision. Phil Trans R Soc Lond B 356: 229–283. p J 50. 68. Fleagle JG (1988) Primate adaptation and evolution. San Diego, California, U.S.A.: Academic Press. 486 p. References Polyak S (1957) The vertebrate visual system. Chicago, Illinois, USA: Chicago University Press. 1390 p. 68. Fleagle JG (1988) Primate adaptation and evolution. San Diego, California, U.S.A.: Academic Press. 486 p. 59. Osorio D, Vorobyev M (1996) Colour vision as an adaptation to frugivory in primates. Proc R Soc Lond B 263: 593–539. November 2014 \| Volume 9 \| Issue 11 \| e113321 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 11
https://openalex.org/W2059083012	https://research.rug.nl/files/2333781/2014PLoSOneFuermaier.pdf	English	null	Good Vibrations – Effects of Whole Body Vibration on Attention in Healthy Individuals and Individuals with ADHD	PloS one	2,014	cc-by	9,574	University of Groningen University of Groningen University of Groningen Good vibrations Fuermaier, Anselm B.M.; Tucha, Lara; Koerts, Janneke; van Heuvelen, Marieke J. G.; van der Zee, Eddy A.; Lange, Klaus W.; Tucha, Oliver Fuermaier, Anselm B.M.; Tucha, Lara; Koerts, Janneke; van Heuvelen, Marieke J. G.; van der Zee, Eddy A.; Lange, Klaus W.; Tucha, Oliver DOI: 10.1371/journal.pone.0090747 DOI: 10.1371/journal.pone.0090747 IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version P blisher's PDF also kno n as Version of record IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Document Version Publisher's PDF, also known as Version of record Document Version Publisher's PDF, also known as Version of record Publication date: 2014 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Fuermaier, A. B. M., Tucha, L., Koerts, J., van Heuvelen, M. J. G., van der Zee, E. A., Lange, K. W., & Tucha, O. (2014). Good vibrations: Effects of whole body vibration on attention in healthy individuals and individuals with ADHD. PLoS ONE, 9(2), Article e90747. https://doi.org/10.1371/journal.pone.0090747 Citation for published version (APA): Fuermaier, A. B. M., Tucha, L., Koerts, J., van Heuvelen, M. J. G., van der Zee, E. A., Lange, K. W., & Tucha, O. (2014). Good vibrations: Effects of whole body vibration on attention in healthy individuals and individuals with ADHD. PLoS ONE, 9(2), Article e90747. https://doi.org/10.1371/journal.pone.0090747 Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Citation for published version (APA): Fuermaier, A. B. M., Tucha, L., Koerts, J., van Heuvelen, M. J. G., van der Zee, E. A., Lange, K. W., & Tucha, O. (2014). Good vibrations: Effects of whole body vibration on attention in healthy individuals and individuals with ADHD. PLoS ONE, 9(2), Article e90747. https://doi.org/10.1371/journal.pone.0090747 Copyright strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of th or copyright holder(s), unless the work is under an open content license (like Creative Commons). The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne- amendment. Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 24-10-2024 Abstract Funding: The authors have no support or funding to report. Funding: The authors have no support or funding to report. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: klaus.lange@ur.de Competing Interests: The authors have declared that no competing interests exist. * E-mail: klaus.lange@ur.de Good Vibrations – Effects of Whole Body Vibration on Attention in Healthy Individuals and Individuals with ADHD Anselm B. M. Fuermaier1, Lara Tucha1, Janneke Koerts1, Marieke J. G. van Heuvelen2, Eddy A. van der Zee3, Klaus W. Lange4*, Oliver Tucha1 1 Department of Clinical and Developmental Neuropsychology, University of Groningen, Groningen, the Netherlands, 2 Center for Human Movement Sciences, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands, 3 Center of Behaviour and Neuroscience, Department of Molecular Neurobiology, University of Groningen, Groningen, the Netherlands, 4 Department of Experimental Psychology, University of Regensburg, Regensburg, Germany Abstract Objectives: Most of the current treatment strategies of ADHD are associated with a number of disadvantages which strengthen the need for alternative or additional approaches for the treatment of ADHD. In this respect, Whole Body Vibration (WBV) might be interesting as it was found to have beneficial effects on a variety of physiological measures. The present study explored the effects of WBV on attention of healthy individuals and adults diagnosed with ADHD. Methods: Eighty-three healthy individuals and seventeen adults diagnosed with ADHD participated in the study. WBV treatment was applied passively, while participants were sitting on a chair which was mounted on a vibrating platform. A repeated measure design was employed in order to explore potential effects of WBV treatment on attention within subjects. Attention (i.e. inhibitory control) was measured with a color-word interference paradigm. Results: A period of two minutes of WBV treatment had significant beneficial effects of small to medium size on attention of both healthy individuals and adults with ADHD. The effect of WBV treatment on attention did not differ significantly between groups. Conclusions: WBV was demonstrated to improve cognitive performance of healthy individuals as well as of individuals with ADHD. WBV treatment is relatively inexpensive and easy to apply and might therefore be of potential relevance for clinical use. The application of WBV treatment as a cognitive enhancement strategy and as a potential treatment of cognitive impairments is discussed. Citation: Fuermaier ABM, Tucha L, Koerts J, van Heuvelen MJG, van der Zee EA, et al. (2014) Good Vibrations – Effects of Whole Body Vibration on Attention in Healthy Individuals and Individuals with ADHD. PLoS ONE 9(2): e90747. doi:10.1371/journal.pone.0090747 Editor: Yu-Feng Zang, Hangzhou Normal University, China Received November 26, 2013; Accepted February 6, 2014; Published February 28, 2014 Copyright: 2014 Fuermaier et al. This is an open-access article distributed under the terms of the Creative Commons Att unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. opyright: 2014 Fuermaier et al. This is an open-access article distributed under the terms of the Creative Commons Attribution restricted use, distribution, and reproduction in any medium, provided the original author and source are credited. er et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits and reproduction in any medium, provided the original author and source are credited. Citation: Fuermaier ABM, Tucha L, Koerts J, van Heuvelen MJG, van der Zee EA, et al. (2014) Good Vibrations – Effects of Whole Body Vibration on Attention in Healthy Individuals and Individuals with ADHD. PLoS ONE 9(2): e90747. doi:10.1371/journal.pone.0090747 Participants Participation in the study was announced to undergraduate students of the University of Groningen, the Netherlands. The description of the study objectives emphasized that the aim of the study was to examine cognition of typically developed students as well as of students with problems of attention. Therefore, a considerable number of students diagnosed with ADHD indicated their interest to participate in the study. In total, 83 healthy individuals and 17 individuals with ADHD took part in the experiment. All participants were undergraduate students from the University of Groningen, the Netherlands. Participation was voluntary and no monetary compensation was offered. Healthy individuals (43 female, 40 male) had a mean age of 22.5 years (SD = 3.7 years), ranging from 18 to 31 years. The self- reported body height of individuals ranged from 163 cm to 205 cm (M = 179 cm, SD = 10 cm), with a body weight ranging from 50 kg to 105 kg across individuals (M = 72 kg; SD = 12 kg). Furthermore, the healthy individuals indicated to spend on average 3.8 hours per week on active exercise (SD = 3.6 hours), which ranged from no active exercise at all (0 hours) to 20 hours per week across individuals. None of the healthy individuals reported any history of neurological or psychiatric diseases and none were currently treated with medication known to affect the central nervous system. Furthermore, none of the healthy individuals indicated to suffer from red-green color deficiency which was a prerequisite to perform the Stroop Color-Word Interference task (see below). In addition, all healthy individuals successfully completed two plates of the Ishihara Color Test [38] in order to ensure unimpaired red-green color perception [39]. in order to ensure unimpaired red-green color perception [39]. Individuals with ADHD (9 female, 8 male) had a mean age of 24.2 years (SD = 1.9 years), ranging from 21 to 28 years. According to self-reports, the body height ranged from 159 cm to 190 cm (M = 176 cm, SD = 8 cm) with a body weight ranging from 45 kg to 101 kg (M = 75 kg; SD = 17 kg) across individuals with ADHD. Furthermore, individuals with ADHD reported to spend in average 3.2 hours per week on active exercise (SD = 5.1 hours), ranging from no exercise at all (0 hours) to 20 hours per week across individuals. Methods pp WBV can be described as a training method which exposures the whole body of an individual to low frequency environmental vibration. In contrast to the use of WBV in combination with dynamic exercises in the fields of sports and fitness, a passive application of WBV is realized by holding static poses (i.e. standing or sitting) during the time of exposure to vibration. Beneficial effects of WBV were found on various physiological measures, including balance, mobility, posture control, oxygen uptake, heart rate, blood pressure, blood flow and muscle strength [23–27]. Moreover, recent animal research provided evidence for a potential value of WBV in improving cognition, as indicated by improved maze learning and an enhanced neuronal activity in mice following the application of WBV treatment [28,29]. The effect of WBV on cognition in humans, however, was not examined in detail and therefore remains widely unsolved. Several studies applied WBV on healthy individuals and examined whether cognitive functioning was different when participants performed cognitive tasks while they received WBV. While some studies did not find any effect of WBV on cognition (i.e. on short- term memory and reasoning skills) [30,31], others revealed impaired short-term memory, long-term memory and arithmetic reasoning skills while WBV was applied [32–34]. A field study to predict the effects of WBV on professional drivers exposed participants to different levels of WBV by testing them in a van which was driving either on asphalt or cobblestone or when the van was halted. Participants had to perform a visual cognitive task and showed a decreased performance while being exposed to WBV compared to a condition of no WBV (van halted) [35]. In contrast to these detrimental effects of WBV on cognition, WBV has also been suggested as an alerting system for improving vigilance [36]. It was argued that particular bands of vibration frequencies cause individuals to increase muscle tension which might facilitate alertness [36]. A study conducted by Mueller and colleagues [37] showed positive effects of passive vibration (proprioceptive stimulation) of the forearm on both neuropsycho- logical performance (as assessed with a computerized attention task) and neurophysiological measures (as assessed with event- related potentials) in patients with traumatic brain injury (TBI) as well as healthy individuals. In this study, proprioceptive stimula- tion was found to improve task performance of a collapsed group of patients with TBI and healthy individuals. Introduction To address this need for further treatment options, the present study explores whether Whole Body Vibration (WBV) can be a beneficial approach to the treatment of adult ADHD. Methods Moreover, patients with TBI showed longer ERP latencies (P300) than healthy individuals. These P300 latencies were shortened by vibratory stimuli in patients with TBI, but not in healthy individuals. This study, therefore, demonstrated that passive vibration may have the potential to improve pathological cognitive processes in patients with neurological conditions and may thus be of relevance for the treatment of cognitive dysfunctions and the field of cognitive rehabilitation. So far, the majority of studies applied WBV at the same time as the cognitive tasks were performed whereas prolonged effects of WBV on subsequent cognitive functioning have not been examined. This is surprising (1) as it appears plausible that vibratory stimuli can cause distraction when they are performed during cognitive testing and (2) when considering that the prolonged effects of WBV are of particular clinical relevance. Thus, the aim of the present study was to explore the effects of WBV on attention in a group of healthy individuals and in a clinical sample of individuals with cognitive dysfunctions, i.e. adults with ADHD. The potential value of WBV as a cognitive Introduction behavioral therapy (CBT) and stimulant drug treatment (e.g. using methylphenidate). Both intervention strategies were shown to have beneficial effects on adults with ADHD, although the first-line choice of treatment strategies is often a stimulant drug therapy [15–18]. However, pharmacological treatment of ADHD is associated with several disadvantages for the individuals. For example, it was found that a pharmacological treatment may lead to unsatisfactory effects in a number of patients [17,19]. In particular, clinical side effects can occur together with stimulant medication, such as insomnia, dry mouth, decreased appetite, and headache [20]. Furthermore, it was demonstrated that methyl- phenidate improves cognitive functions of adults with ADHD, but does not normalize their level of performance [11,21]. Conse- quently, there is a need for alternative or additional treatment options which (1) can be given in addition to the effective treatments (e.g. no interaction with stimulant drug treatment), (2) are time and cost effective, (3) have no detrimental side effects and (4) are effective on symptoms that are most detrimental to the Attention deficit hyperactivity disorder (ADHD) is associated with a reduced quality of life [1]. Children and adults with ADHD frequently experience problems in social interaction, which can lead to social rejection, social isolation and discrimination [2–5]. Moreover, ADHD is associated with lower academic achievement which was shown to be directly linked to ADHD symptoms and cognitive deficits [6]. With regard to cognition of adults with ADHD, neuropsychological research revealed problems in atten- tion and executive dysfunctions as core features of ADHD, including problems in working memory, inhibitory control, distractibility, vigilance, set-shifting, and task planning [7–13]. Accordingly, subjective experiences of inattention and executive dysfunctioning were shown in a considerable proportion of adults with ADHD by using self-report instruments [14]. Conventional behavioral and pharmacological treatment strat- egies of adults with ADHD include, beside others, cognitive February 2014 \| Volume 9 \| Issue 2 \| e90747 1 February 2014 \| Volume 9 \| Issue 2 \| e90747 PLOS ONE \| www.plosone.org Effects of WBV on Cognition in ADHD Effects of WBV on Cognition in ADHD enhancement strategy and its application for the treatment of cognitive dysfunctions are discussed. enhancement strategy and its application for the treatment of cognitive dysfunctions are discussed. patients’ functioning, such as problems related to attention [13,22]. Self-report Measures of ADHD Symptoms Figure 1. Drawing of the vibrating platform with the mounted wooden platform and chair. Accelerations were measured at location A, B, C and D to determine the actual vibration frequency and amplitude. doi:10.1371/journal.pone.0090747.g001 Two scales of the self-report ADHD Rating Scale (ARS, DSM- IV) were applied, one for current symptoms and one for retrospective symptoms in childhood. The ADHD Rating Scale is a self-report instrument for ADHD symptoms made up of the 18 DSM-IV criteria for ADHD [41,43]. In the Dutch version of the ADHD Rating Scale, five DSM-IV criteria containing double statements were rephrased into two statements, so that the total number of items was 23. In the analysis of the number of fulfilled symptoms, the scores based on the 23 items were recalculated to the original 18 DSM-IV criteria. The items were rated on a 4- point scale (0 = rarely or never, 1 = sometimes, 2 = often, 3 = very often) based on participant’s behavior in the past six months (scale for current symptoms) or based on the participant’s behavior in childhood up to twelve years in age (scale for retrospective symptoms). A symptom was considered to be present, if the answer given to the item was often or very often (score 2 or 3). There were two dimensions, one for inattentive symptoms (nine criteria) and one for hyperactive-impulsive symptoms (nine criteria). To qualify for a clinical diagnosis of adult ADHD, individuals must have met six or more of the nine DSM-IV criteria of inattention and/or hyperactivity-impulsivity in childhood, and at least five of the nine DSM-IV criteria of inattention and/or hyperactivity-impulsivity in adulthood [43]. Furthermore, two self-report scales were applied in order to quantify ADHD symptomatology currently and retrospectively. Childhood ADHD symptoms were self-rated with the short version of the Wender Utah Rating Scale (WURS). The WURS contains 25 items which are rated on a five-point Likert scale [44,45]. Severity of current ADHD symptoms was self-rated with the ADHD self-report scale [44]. This scale consists of 18 items which correspond to the diagnostic criteria of DSM-IV and which are rated on a four-point Likert scale [44,46]. A sum score was calculated for each rating scale. Measurement of Attention To explore the effects of passive WBV on attention, the Stroop Color-Word Interference task (Stroop, 1935) was performed. The Stroop Participants The actual vertical displace- ments (frequency and amplitude) were therefore measured on different locations of the chair (see A, B, C and D in Figure 1) based on acceleration data without a person sitting on the chair. The measured displacements (frequency/amplitude) were 30 Hz/ 0.44 mm (location A), 30 Hz/0.44 mm (location B), 30 Hz/ 0.66 mm (location C) and 30 Hz/0.50 mm (location D). The modified Vibe 300 was located in a quiet laboratory at the Department of Clinical and Developmental Neuropsychology of the University of Groningen, the Netherlands. (M = 30.8; SD = 8.6) [41–44]. In the diagnostic assessment of the 17 individuals with ADHD, 5 individuals met DSM-IV criteria for ADHD – predominantly inattentive type (ADHD-I), 1 patient met criteria for ADHD – hyperactive-impulsive type (ADHD-H) and 11 individuals met criteria for ADHD – combined type (ADHD- C). Moreover, one of the individuals with ADHD reported to suffer from comorbid disorders (anxiety and depression) and was therefore continuously taking antidepressant medication (including the time of the assessment) as prescribed by the attending physician (150 mg per day). Furthermore, four individuals with ADHD were currently treated with methylphenidate and received individually tailored and clinically appropriate doses with a mean dose of 40.5 mg per day, with individual doses ranging from 10 mg to 72 mg. Pharmacological treatment was not discontinued for this study so that the individuals were on their usual prescribed medication at the time of their participation. Healthy individuals and individuals with ADHD did not differ significantly in either of the demographic variables, including age (t(98) = 1.778, p = .079), gender (x2 (1) = 0.127; p = .721), body height (t (98) = 1.072, p = .286), body weight (t (98) = 0.825, p = .411) and amount of exercise (t (98) = 0.555, p = .580). Participants None of the individuals with ADHD indicated to suffer from red-green color deficiency which was again ensured by successful completion of two plates of the Ishihara Color Test [38,39]. Moreover, all individuals in the ADHD group reported that they had previously been diagnosed with ADHD by a qualified psychologist or physician. The diagnosis of ADHD was confirmed by a clinical psychiatric interview with an experienced psychologist belonging to the research group. The diagnostic interview involved a retrospective diagnosis of an ADHD in childhood and currently as devised by Barkley and Murphey [40]. In addition, individuals in the ADHD group qualified for inclusion on the basis of four self-report measures for ADHD symptoms. All patients completed the self- report ADHD Rating Scale (ARS, DSM-IV) for retrospective symptoms in childhood (patients fulfilled in average 6.1 criteria for inattention and 6.1 criteria for hyperactivity/impulsivity) and the self-report ADHD Rating Scale (ARS, DSM-IV) for current symptoms (patients fulfilled in average 6.8 criteria for inattention and 5.9 criteria for hyperactivity/impulsivity). Moreover, individ- uals with ADHD completed the short form of the Wender Utah Rating Scale for childhood ADHD symptoms (WURS) (M = 40.2; SD = 11.0) and the ADHD self-report scale for current symptoms Thus, the aim of the present study was to explore the effects of WBV on attention in a group of healthy individuals and in a clinical sample of individuals with cognitive dysfunctions, i.e. adults with ADHD. The potential value of WBV as a cognitive February 2014 \| Volume 9 \| Issue 2 \| e90747 2 PLOS ONE \| www.plosone.org Effects of WBV on Cognition in ADHD (Figure 1). Both the wooden platform and the chair were attached to the Vibe 300 with bolts from underneath the Vibe 300 in order to keep deviations from the vibrating frequency and amplitude on a minimum. The manufacturer settings of 30 Hz vibration frequency and 4 mm vibration amplitude were applied as a previous study showed this setup to be comfortable for the participant as well as efficient [47]. However, the construction of a wooden platform and chair mounted on the vibrating device was assumed to cause some deviations from the manufacturer settings with regard to the vibration frequency and amplitude. Materials Self-report Measures of ADHD Symptoms Procedure At the beginning of the experiment, all participants completed a questionnaire asking for descriptive information including age, sex, body height, body weight, time spent on average per week on exercise as well as red-green color deficiency. Furthermore, participants were asked for any history of psychiatric or neurological diseases as well as pharmacological treatment. For those students who indicated to be diagnosed with ADHD, a psychiatric interview was conducted followed by the completion of four self-report scales of ADHD symptoms. Subsequently, all participants conducted four practice trials of the Stroop Color-Word Interference task. For these practice trials, four parallel versions of the task were used, separated by a resting period of three minutes between each time the task was performed. The four practice trials were applied in order to minimize practice effects during the experimental trials. Furthermore, explorative analyses were carried out in order to examine whether the effects of WBV were determined by demographic variables, including age, gender, body height, body weight and amount of active exercise. This was done by correlating ipsative scores of healthy individuals (indicating the effect of WBV on cognition) with their demographic variables (Spearmen rank correlation). According to Cohen [48], negligible effects (r,0.1), small effects (0.1#r,0.3), medium effects (0.3#r, 0.5) and large effects (r$0.5) were distinguished [48]. Finally, gender effects of WBV treatment were analyzed by comparing female and male individuals (Mann-Whitney-U-Test). Prior to the first experimental trial, the participant was requested to take seat on the chair mounted on the Vibe 300. The participant was instructed to sit in upright position throughout the whole experiment, the arms on the rest and both feet on the wooden platform. The participant was further instructed to keep body movements to a minimum. The experiment consisted of eight experimental trials. Each trial started with a two-minute period of experimental treatment, either a period of vibration (vibration condition) or a resting period of no vibration (resting condition). Immediately after the experimental treatment, the Color-Word Interference Test and the Color Block Test of the Stroop Color-Word Interference task were performed. The Color- Word Interference Test always preceded the Color Block Test. Subsequently, a resting period of three minutes followed prior to the beginning of the subsequent experimental trial. The eight experimental trials comprised four trials of vibration and four trials of resting. The sequence of the eight experimental trials followed an ABBA-design in order to balance rank order effects (i.e. Results Effects of WBV Treatment on Attention Data analysis revealed that WBV treatment significantly improved attention performance (M6SD) in both healthy individuals (resting condition: 3.46260.419; vibration condition: 3.28960.358; Z = 4.835, p#.001) and individuals with ADHD (resting condition: 3.63560.373; vibration condition: 3.38160.419; Z = 3.243, p = .001). While the difference in healthy individuals was of small size (d = 0.44), a medium effect (d = 0.64) was found in individuals with ADHD (Figure 3). A separate exploration of the effects of WBV treatment among different subtypes of ADHD showed beneficial effects in all three groups, including patients of the inattentive subtype (n = 5; resting condition: 3.68760.438; vibration condition: 3.50460.419), the patient of the hyperactive-impulsive subtype (n = 1; resting condition: 3.055; vibration condition: 2.965) and patients of the combined subtype (n = 11; resting condition: 3.66360.331; vibration condition: 3.36360.398). Procedure practice effects) when comparing the two treatment conditions. A schematic drawing of the procedure of eight experimental trials is presented in Figure 2. The duration of the assessment was about 90 minutes. Whole Body Vibration Figure 1. Drawing of the vibrating platform with the mounted wooden platform and chair. Accelerations were measured at location A, B, C and D to determine the actual vibration frequency and amplitude. doi:10.1371/journal.pone.0090747.g001 Passive WBV was applied by using a vibrating platform (Vibe 300, Tonic Vibe, Nantes, France). A wooden platform (0.5 m60.9 m) was mounted on the Vibe 300 to enlarge the platform. On top of the platform, a wooden chair was mounted PLOS ONE \| www.plosone.org February 2014 \| Volume 9 \| Issue 2 \| e90747 3 Effects of WBV on Cognition in ADHD Effects of WBV on Cognition in ADHD Color-Word Interference task measure an aspect of attention, i.e. inhibitory control [9]. In the present study, two conditions of the Stroop Color-Word Interference task were applied, the Color Block Test and the Color-Word Interference Test. In the Color Block Test, 20 squares each printed in one of four possible colors (yellow, blue, green or red), were presented on a card. The participant’s task was to name the colors of the squares as fast as possible. In the Color- Word Interference Test, a list of 52 color names was presented (yellow, blue, green or red). Each word was printed in one of four possible colors (yellow, blue, green or red), however, the color ink of the words was different from the color names (e.g. the word green was printed in blue). The participant was requested to name the ink colors of the words as quickly as possible, thereby ignoring the written color name. In both tests, the time to complete the task was measured. An interference quotient was calculated by dividing the time needed for the Color-Word Interference Test by the time needed for the Color Block Test. The interference quotient represented a measure of attention (i.e. inhibitory control), with a larger quotient indicating more problems in inhibitory control. 12 parallel versions (for four practice trials and eight experimental trials) were designed in order to apply the tests in a repeated measure design. All participants signed an informed consent prior to the study and were debriefed after the assessment. All participants were informed that participation was voluntary and that they had the right to refuse or stop participating in the study at any time. Statistical Analysis Since the cognitive task were performed repeatedly (eight times) under two different treatment conditions (vibration or resting), mean scores were calculated across the four trials for each treatment condition, separately for healthy individuals and individuals with ADHD. Nonparametric statistical tests (Wilcoxon signed-rank test for dependent samples) were applied in order to compare cognitive performance (mean interference quotient) between the vibration condition and the resting condition, separately for healthy individuals and individuals with ADHD. A significance level of a = .05 was applied for all comparisons. Moreover, effect sizes (Cohen’s d) were calculated for all comparisons. Following Cohen’s guidelines for interpreting effect sizes (Cohen, 1988), negligible effects (d,0.20), small effects (0.20#d,0.50), medium effects (0.50#d,0.80) and large effects (d$0.80) were distinguished [48]. In addition, the effect of WBV on cognition was compared between healthy individuals and individuals with ADHD. For this reason, ipsative scores were calculated for each individual by subtracting the mean interference quotient of vibration conditions per participant from the mean interference quotient of resting conditions per participant. Ipsative scores, which indicated the effect of WBV on cognition, were then compared between healthy individuals and individuals with ADHD by applying a nonpara- metric statistical test for independent samples (Mann-Whitney-U- Test). Discussion The present study showed in a large sample of healthy young individuals that WBV treatment has a beneficial effect on subsequent cognitive functioning. Already a short interval of two minutes WBV treatment resulted in an acute beneficial effect on attention (i.e. inhibitory control) as measured with the Stroop Color- Word Interference task. Even though evidence of long-term beneficial effects of WBV treatment on cognition is still lacking, the present effects of WBV on attention are promising as the treatment period was rather short (two minutes), emphasizing the significance of the effect. Furthermore, it was assumed that there is only little room for exercise related improvements of cognition in early adulthood [49]. Because the sample of the present study represents a homogenous selection of young, physically healthy and mentally high functioning individuals, WBV treatment appears to have the potential to be an effective cognitive enhancer. This is of high relevance as better cognitive functioning is associated with higher academic achievement and socioeconomic status as well as facilitation of social integration [50] and is further assumed to provide a buffer against cognitive impairment (cognitive reserve) as a consequence of the effects of brain pathology or aging [51]. Demographic changes of the society with an increasing number of older people create considerable concerns (e.g. rise of financial burdens in the health sector) and underline the societal interest in maintaining and improving physical health and cognitive abilities of people. If WBV will be proven to have long-term beneficial effects on cognition, WBV treatment might be a suitable approach to contribute in achieving this goal, in particular, because it is inexpensive, non-invasive, easy to apply and safe (especially when applied passively). For example, studies on the musculoskeletal system demonstrated that WBV can be applied safely on children, older individuals and individuals with neurological and psychiatric conditions including individuals with movement disorders [52– 57]. Ethics Statement The study was approved by the Ethical Committee Psychology (ECP) affiliated to the University of Groningen, the Netherlands. A comparison of ipsative scores between healthy individuals (n = 83; 0.17360.293) and the entire group of individuals with PLOS ONE \| www.plosone.org February 2014 \| Volume 9 \| Issue 2 \| e90747 February 2014 \| Volume 9 \| Issue 2 \| e90747 4 Effects of WBV on Cognition in ADHD Figure 2. Schematic drawing of the experimental procedure consisting of eight experimental trials. V = Vibration condition: WBV treatment for two minutes; R = Resting condition: No WBV treatment for two minutes; B = Break: Breaks of three minutes (resting); Stroop test = Stroop Color-Word Interference task (Color-Word Interference Test, Color Block Test); For analysis, mean scores were calculated across the four experimental trials for each treatment condition (vibration and resting). doi:10.1371/journal.pone.0090747.g002 Figure 2. Schematic drawing of the experimental procedure consisting of eight experimental trials. V = Vibration condition: WBV treatment for two minutes; R = Resting condition: No WBV treatment for two minutes; B = Break: Breaks of three minutes (resting); Stroop test = Stroop Color-Word Interference task (Color-Word Interference Test, Color Block Test); For analysis, mean scores were calculated across the four experimental trials for each treatment condition (vibration and resting). doi:10.1371/journal.pone.0090747.g002 ADHD (n = 17; 0.25460.218) revealed a non-significant differ- ence of small size (Z = 1.542, p = 0.123, d = 0.31). February 2014 \| Volume 9 \| Issue 2 \| e90747 Limitations and Future Directions Furthermore, the positive effects of WBV on cognition as shown in the present study on a sample of individuals with ADHD support findings reported by Mueller and colleagues [37], who found positive effects of proprioceptive stimulation (vibration) on cognition and event-related potentials in patients with TBI and thereby suggested that vibratory stimuli may be of potential relevance in a supplementary meaning for the treatment of pathologic cognitive functions. Second, improvements of attention were measured directly after WBV treatment has been applied (acute effects). To establish WBV as a treatment approach, data showing longer lasting effects of cognitive enhancement are required. Therefore, sustained effects of WBV on cognition needs to be explored and should be subject of future research. Longitudinal studies should also take effects of vibratory stimuli on other domains than cognition into consideration, as it was reported that chronic exposure to WBV (e.g. in the occupational setting) is associated with alterations of mood, including fatigue, depression and anxiety [62]. Third, the present study focused on a single aspect of cognition, i.e. inhibitory control. However, it remains unsolved whether WBV has a selective effect on only particular aspects of cognition or whether a broad range of cognitive functions can benefit from WBV treatment. Hence, future studies should examine which functions are most sensitive to WBV. p g g If cognition enhancing effects of WBV treatment can be replicated in controlled trials, there are several advantages related to a clinical application of WBV treatment emphasizing its value as a novel strategy in the treatment of cognitive dysfunctions of individuals with ADHD. First, WBV does not interfere with other interventions (such as pharmacological treatment) and can thus be applied in addition to conventional interventions. Second, WBV is easy to apply and is (in comparison to other methods) cost effective. A WBV device can be set up in private or institutional settings (e.g. at home, in school or at work) and short periods of WBV treatment of only a few minutes per day might be sufficient to cause clinically relevant effects. In addition, a high compliance rate is expected as this approach requires only little individual effort (e.g. in form of active exercise). Third, no detrimental side effects are currently known by the application of passive WBV. Fourth, and most importantly, WBV was found to have positive acute effects on attention (i.e. Limitations and Future Directions The present study must be viewed in the context of some limitations. First, with regard to the conclusions aiming at clinical application of WBV treatment, it has to be taken into consider- ation that the cognition-enhancing effects of WBV treatment on ADHD were observed in a rather small sample (n = 17) of high functioning individuals diagnosed with ADHD. All individuals with ADHD were college students, implying a relatively high level of cognitive functioning. Furthermore, at the time of the assessment, four individuals with ADHD reported to be treated with stimulant medication which compromises the conclusions drawn upon the effects of WBV treatment on cognition. For this reason, the generalization of the observed effects to the population of adults with ADHD and individuals with other neuropsycho- logical disorders might be questioned. However, assuming that there is only little room for improvement in high functioning individuals (ceiling effects), significant effects of small to medium size revealed in two intellectually high functioning groups of people (healthy college students and college students with ADHD) underline the potential of WBV for the treatment of neuropsy- chological dysfunctions. A replication of the present findings in a large group of healthy individuals and individuals with neuropsy- chological dysfunctions appears necessary. Future studies address- ing effects of WBV treatment in healthy individuals would benefit from the application of screening instruments for psychiatric disorders (e.g. for depression or ADHD) in order to obtain more homogeneous groups and to exclude individuals with high scores on respective symptom scales suggesting a psychiatric diagnosis. g Moreover, it has to be emphasized that the present study revealed positive effects of WBV treatment on performance in subsequent cognitive performance and differed in this respect from the majority of previous research in which participants were exposed to WBV at the same time as they performed cognitive tasks [30–35]. These studies reported either no effects or revealed even negative effects of WBV on cognition, which is not surprising considering the distraction that vibratory stimuli may have induced to the simultaneous execution of cognitive tasks. Correlations between WBV Treatment Effect and Demographic Variables An exploration of the effects of WBV treatment among different subtypes of ADHD indicated the largest benefit in patients of the combined subtype, even though small sample sizes hamper a reliable comparison of different subtypes of ADHD and therefore requires replication in larger samples. Furthermore, and more importantly in this context, it has to be emphasized that the effects of WBV treatment on subsequent attention performance as shown in the present study are promising but do not yet justify a recommendation for clinical use. Future studies are required examining long-term effects of WBV treatment on cognition by implementing a variety of control conditions (e.g. vibration of specific body parts only or presenta- tion of auditory stimuli) in order to explore the specific mechanism of WBV treatment which causes cognitive enhancement. (d = 0.44). The finding that adults with ADHD benefit more from WBV treatment than healthy individuals cannot be explained by demographic variables since group did not differ in these variables. From a clinical perspective, this finding is very meaningful considering that problems related to attention are common neuropsychological deficits associated with ADHD which has been shown to adversely affect individuals’ occupational function- ing and quality of life [1,11,58,59]. Since improved cognitive functions may support social integration and occupational functioning and may also lead to an increased quality of life [1,11,58,59], the present findings consequently stress the potential value of WBV in the treatment of cognitive impairments of individuals with ADHD. An exploration of the effects of WBV treatment among different subtypes of ADHD indicated the largest benefit in patients of the combined subtype, even though small sample sizes hamper a reliable comparison of different subtypes of ADHD and therefore requires replication in larger samples. Furthermore, and more importantly in this context, it has to be emphasized that the effects of WBV treatment on subsequent attention performance as shown in the present study are promising but do not yet justify a recommendation for clinical use. Future studies are required examining long-term effects of WBV treatment on cognition by implementing a variety of control conditions (e.g. vibration of specific body parts only or presenta- tion of auditory stimuli) in order to explore the specific mechanism of WBV treatment which causes cognitive enhancement. Correlations between WBV Treatment Effect and Demographic Variables Explorative analyses of the association between the effect of WBV treatment and demographic variables in healthy individuals showed that WBV effects were only non-significantly related in small to negligible size to any of the demographic variables including age (r = 2.023; p = .836), body height (r = 2.178; p = .107), body weight (r = 2.091; p = .414) and amount of exercise (r = 2.098; p = .380). Female and male individuals did also not differ significantly concerning the effect of WBV treatment as also indicated by a negligible effect size, Z = 2 1.258; p = .209, d = 0.19. Figure 3. Mean interference quotient for performance of the Stroop Color-Word Interference task following resting and vibration, for both healthy individuals and individuals with ADHD (M6SE). doi:10.1371/journal.pone.0090747.g003 In accordance with the studies showing that patients with neurological conditions can benefit from WBV treatment (i.e. in physiological measures, but also cognition) [37], the present study revealed that WBV treatment improves cognitive functioning (i.e. inhibitory control as a measure of attention) of adults diagnosed with ADHD. The effects of WBV treatment on attention in adults with ADHD were of medium size (d = 0.64) and were thus somewhat larger as the effects observed in healthy individuals Figure 3. Mean interference quotient for performance of the Stroop Color-Word Interference task following resting and vibration, for both healthy individuals and individuals with ADHD (M6SE). doi 10 1371/journal pone 0090747 g003 doi:10.1371/journal.pone.0090747.g003 February 2014 \| Volume 9 \| Issue 2 \| e90747 PLOS ONE \| www.plosone.org February 2014 \| Volume 9 \| Issue 2 \| e90747 5 Effects of WBV on Cognition in ADHD (d = 0.44). The finding that adults with ADHD benefit more from WBV treatment than healthy individuals cannot be explained by demographic variables since group did not differ in these variables. From a clinical perspective, this finding is very meaningful considering that problems related to attention are common neuropsychological deficits associated with ADHD which has been shown to adversely affect individuals’ occupational function- ing and quality of life [1,11,58,59]. Since improved cognitive functions may support social integration and occupational functioning and may also lead to an increased quality of life [1,11,58,59], the present findings consequently stress the potential value of WBV in the treatment of cognitive impairments of individuals with ADHD. February 2014 \| Volume 9 \| Issue 2 \| e90747 References Tannock R (1998) Attention deficit hyperactivity disorder: Advances in cognitive, neurobiological, and genetic research. Journal of Child Psychology and Psychiatry 39: 65–99. 23. Bogaerts A, Verschueren S, Delecluse C, Claessens AL, Boonen S (2007) Effects of whole body vibration training on postural control in older individuals: A 1 year randomized controlled trial. Gait Posture 26: 309–316. 9. Boonstra AM, Oosterlaan J, Sergeant JA, Buitelaar JK (2005) Executive functioning in adult ADHD: A meta-analytic review. Psychol Med 35: 1097– 1108. 24. Cardinale M, Bosco C (2003) The use of vibration as an exercise intervention. Exerc Sport Sci Rev 31: 621–624. 10. Fuermaier ABM, Tucha L, Koerts J, Aschenbrenner S, Westermann C, et al. (2013) Complex prospective memory in adults with attention deficit hyperac- tivity disorder. PloS ONE 8(3): e58338. 25. Cochrane DI, Sartor F, Winwood K, Stannard SR, Narici MV, et al. (2008) A comparison of the physiologic effects of acute whole-body vibration exercise in young and older people. Arch Phys Med Rehabil 89: 815–821. 11. Tucha L, Tucha O, Sontag TA, Stasik D, Laufkoetter R, et al. (2011) Differential effects of methylphenidate on problem solving in adults with ADHD. Journal of Attention Disorders 15: 161–173. 26. Lam FMH, Lau RWK, Chung RCK, Pang MYC (2012) The effect of whole body vibration on balance, mobility and falls in older adults: A systematic review and meta-analysis. Maturitas 72: 206–213. 12. Tucha O, Mecklinger L, Laufko¨tter R, Kaunzinger I, Paul GM, et al. (2005) Clustering and switching on verbal and figural fluency functions in adults with attention deficit hyperactivity disorder. Cognitive Neuropsychiatry 10: 231–248. 27. Stewart JM, Karman C, Montgomery LD, McLeod KJ (2005) Plantar vibration improves leg fluid flow in perimenopausal women. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 288: 623–629. 13. Tucha L, Tucha O, Laufkoetter R, Walitza S, Klein HE, et al. (2008) Neuropsychological assessment of attention in adults with different subtypes of attention-deficit/hyperactivity disorder. J Neural Transm 115: 269–278. 28. Keijser N, Piersma D, Postema F, Venema BJ, Luiten PGM, et al. (2011) Improved cognitive performance as a result of whole body stimulation in mice and men. The 9th Dutch Endo-Neuro-Psycho (ENP) Meeting: 125. 14. Biederman J, Petty CR, Fried R, Black S, Faneuil A, et al. (2008) Discordance between psychometric testing and questionnaire-based definitions of executive function deficits in individuals with ADHD. Journal of Attention Disorders 12: 92–102. 29. References 16. Wigal SB (2009) Efficacy and safety limitations of attention-deficit hyperactivity disorder pharmacotherapy in children and adults. Cns Drugs 23: 13–21. 1. Agarwal R, Goldenberg M, Perry R, IsHak WW (2012) The quality of life of adults with attention deficit hyperactivity disorder: A systematic review. Innovations in clinical neuroscience 9: 10–21. 17. Wilens TE, Spencer TJ, Biederman J (2002) A review of the pharmacotherapy of adults with attention-deficit/hyperactivity disorder. Journal of attention disorders 5: 189–202. 2. Pescosolido BA, Fettes DL, Martin JK, Monahan J, McLeod JD (2007) Perceived dangerousness of children with mental health problems and support for coerced treatment. Psychiatric Services 58: 619–625. 18. Sonuga-Barke EJS, Brandeis D, Cortese S, Daley D, Ferrin M, et al. (2013) Nonpharmacological interventions for ADHD: Systematic review and meta- analyses of randomized controlled trials of dietary and psychological treatments. Am J Psychiatry 170: 275–289. y 3. Canu WH, Carlson CL (2007) Rejection sensitivity and social outcomes of young adult men with ADHD. Journal of Attention Disorders 10: 261–275. 4. Mueller AK, Fuermaier ABM, Koerts J, Tucha L (2012) Stigma in attention deficit hyperactivity disorder. ADHD Attention Deficit and Hyperactivity Disorders 4: 101–114. 19. Wigal T, Swanson JM, Regino R, Lerner MA, Soliman I, et al. (1999) Stimulant medications for the treatment of ADHD: Efficacy and limitations. Ment Retard Dev Disabil Res Rev 5: 215–224. 5. Lange KW, Reichl S, Lange KM, Tucha L, Tucha O (2010) The history of attention deficit hyperactivity disorder. ADHD Attention Deficit and Hyperac- tivity Disorders 2: 241–255. 20. Adler LA, Spencer T, McGough JJ, Jiang H, Muniz R (2009) Long-term effectiveness and safety of dexmethylphenidate extended-release capsules in adult ADHD. Journal of Attention Disorders 12: 449–459. 6. Daley D, Birchwood J (2010) ADHD and academic performance: Why does ADHD impact on academic performance and what can be done to support ADHD children in the classroom? Child Care Health and Development 36: 455–464. 21. Tucha O, Mecklinger L, Laufkoetter R, Klein HE, Walitza S, et al. (2006) Methylphenidate-induced improvements of various measures of attention in adults with attention deficit hyperactivity disorder. J Neural Transm 113: 1575– 1592. 7. Barkley RA (1997) Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory of ADHD. Psychol Bull 121: 65–94. 22. Schoechlin C, Engel RR (2005) Neuropsychological performance in adult attention-deficit hyperactivity disorder: Meta-analysis of empirical data. Archives of Clinical Neuropsychology 20: 727–744. 8. Limitations and Future Directions inhibitory control), an aspect of cognition which is crucial for everyday life functioning and which has been found to be impaired in both children and adults with ADHD [9,10,13,60,61]. Finally, and fourth, so far only little is known about the mechanisms underlying the WBV induced cognitive enhance- ment, and the present study does not contribute to the understanding of these mechanisms. The present study compared the effects of a two-minute period of WBV treatment with a resting condition in which no vibratory stimuli were applied. The implementation of a variety of control conditions would be necessary in order to explore which elements of WBV treatment are responsible for the cognition enhancing effect. For example, no information was obtained (1) whether it might be efficient enough to stimulate particular body parts (e.g. the hands) of individuals to induce cognition enhancing effects, (2) which settings of vibratory stimuli induce most beneficial effects (i.e. frequency and ampli- tude) and (3) to which extent the produced noise of the vibrating device contributes to the present results. It was hypothesized that WBV changes muscle length, leading to stimulation of muscle spindles and a reflex response of the muscle [24]. This may enhance muscle activity, resulting in increased oxygen uptake and heart rate. Furthermore, other mechanoreceptors (such as free nerve endings) might be stimulated PLOS ONE \| www.plosone.org February 2014 \| Volume 9 \| Issue 2 \| e90747 6 Effects of WBV on Cognition in ADHD which might induce sensory stimulation in cortical brain areas. Despite these findings, the underlying mechanism, determining factors and the neurobiological basis of the effects of WBV on cognition in humans remain however widely unresolved and should be addressed in future research. A series of animal studies (on different mouse strains) has been performed [28,29,63,64] revealing that WBV treatment enhanced cognition (e.g. spatial memory) in young and aged mice compared to non-vibrated control mice. Examinations of brains indicated that WBV activates/increases various mechanisms, including (I) the activity of the cholinergic system in the forebrain, (II) the transportation of glucose across the blood-brain barrier, (III) expression of immediate early genes (which make neurons more responsive), (IV) production of proteins necessary for neuronal plasticity and (V) the neurogenesis (i.e. generating new neurons in the adult brain). Limitations and Future Directions Furthermore, following WBV, (VI) an increased concen- tration of tyrosine hydroxylase has been found, an enzyme responsible for catalyzing the synthesis of a precursor of the neurotransmitter dopamine. Dopamine has been shown to affect movement, motivation and cognition [65–67] and is also assumed to be involved in the pathophysiology of ADHD [68,69]. considerations on brain arousal and stochastic resonance as well as on empirical evidence suggested that auditory stimuli (background white noise) can be beneficial to cognitive performance [70,71]. In these studies, it was argued that whereas auditive noise is detrimental to cognition of healthy individuals, a moderate amount of auditive noise might improve cognitive functions in individuals with hypodopaminergic states, such as individuals with ADHD. With regard to the results of the present study, the beneficial effect of WBV treatment on cognition in adults with ADHD is in line with the prediction of this model. However, as a positive effect of WBV treatment on cognition was also shown in healthy individuals, the model of white noise affecting cognition is not supported and does therefore not appear to be the primary mechanism of the cognitive enhancement found in the present study. However, the present findings in combination with the results of So¨derlund and colleagues might indicate that sensory stimulation, independent whether it is auditive noise or mechanic vibration, has the potential to improve cognition in individuals suffering from ADHD. Author Contributions In addition, it can be speculated that the beneficial effects on cognition are not only the result of the vibratory stimuli, but also of the noise which is produced by the vibrating device. Unfortu- nately, a condition controlling for noise was not implemented in the present study design. A recent model based on theoretical Conceived and designed the experiments: ABMF LT JK MJGvH EAvdZ KWL OT. Performed the experiments: ABMF LT. Analyzed the data: ABMF LT JK OT. Contributed reagents/materials/analysis tools: MJGvH EAvdZ OT. Wrote the paper: ABMF LT JK OT. Effects of WBV on Cognition in ADHD Ishihara S (1917) Tests for color-blindness. Tokyo: Hongo Haruk 59. Boonstra AM, Kooij JJS, Oosterlaan J, Sergeant JA, Buitelaar JK (2010) To act or not to act, that’s the problem: Primarily inhibition difficulties in adult ADHD. Neuropsychology 24: 209–221. 39. Weale RA (1953) Ishihara tables - comparison of the 9th and 10th editions. Br Med J 1: 1148–1148. 40. Barkley RA, Murphy KR (1998) Attention-deficit hyperactivity disorder: A clinical workbook. New York: Guilford. p y gy 60. Tucha O, Walitza S, Mecklinger L, Sontag T-A, Kuebber S, et al. (2006) Attentional functioning in children with ADHD - predominantly hyperactive- impulsive type and children with ADHD - combined type. J Neural Transm 113: 1943–1953. 41. DuPaul GJ, Power TJ, Anastopoulos AD, Reid R (1998) ADHD Rating Scale- IV. Checklists, norms and clinical interpretation. New York: The Guilford Press. 42. Kooij JJS, Boonstra AM, Swinkels SHN, Bekker EM, de Noord I, et al. (2008) Reliability, validity, and utility of instruments for self-report and informant report concerning symptoms of ADHD in adult patients. Journal of Attention Disorders 11: 445–458. 61. Fuermaier ABM, Tucha L, Koerts J, Aschenbrenner S, Weisbrod M, et al. (2013) Source discrimination in adults with attention deficit hyperactivity disorder. PLoS ONE 8(5): e65134. ( ) 62. Abbate C, Micali E, Giorgianni C, Munao F, Brecciaroli R, et al. (2004) Affective correlates of occupational exposure to whole-body vibration - A case- control study. Psychother Psychosom 73: 375–379. 43. Kooij JJS, Buitelaar JK, Van den Oord EJ, Furer JW, Rijnders CAT, et al. (2005) Internal and external validity of attention-deficit hyperactivity disorder in a population-based sample of adults. Psychol Med 35: 817–827. 63. Timmer M, Van der Zee EA, Riedel G (2006) Whole body vibration and behavior: Investigation of the role of various neurotransmitter systems. Federation of European Neuroscience Societies Abstract 3: 089.31. 44. Ro¨sler M, Retz-Junginger P, Retz W, Stieglitz R (2008) Homburger ADHS- Skalen fu¨r Erwachsene. Untersuchungsverfahren zur syndromalen und kategor- ialen Diagnostik der Aufmerksamkeitsdefizit-/Hyperaktivita¨tssto¨rung (ADHS) im Erwachsenenalter. Go¨ttingen: Hogrefe. 64. Lahr MMH, Postema F, Venema BJ, Luiten PGM, Riedel G, et al. (2009) Whole body stimulation functions as a cognitive enhancer in young and old mice. 8th Dutch Endo-Neuro-Psycho Meeting Abstract 170. 45. Ward MF, Wender PH, Reimherr FW (1993) The Wender Utah Rating-Scale - an aid in the retrospective diagnosis of childhood attention-deficit hyperactivity disorder. Am J Psychiatry 150: 885–890. 65. Effects of WBV on Cognition in ADHD 31. Ljungberg JK, Neely G (2007) Stress, subjective experience and cognitive performance during exposure to noise and vibration. J Environ Psychol 27: 44– 54. 53. Kawanabe K, Kawashima A, Sashimoto I, Takeda T, Sato Y, et al. (2007) Effect of whole-body vibration exercise and muscle strengthening, balance, and walking exercises on walking ability in the elderly. Keio J Med 56: 1–10. g g y y 54. Santos-Filho SD, Cameron MH, Bernardo-Filho M (2012) Benefits of whole- body vibration with an oscillating platform for people with multiple sclerosis: A systematic review. Multiple sclerosis international. 2012: 274728. 32. Sandover J, Champion DF (1984) Some effects of a combined noise and vibration environment on a mental arithmetic task. J Sound Vibrat 95: 203–212. 33. Sherwood N, Griffin MJ (1990) Effects of whole-body vibration on short-term- memory. Aviation Space and Environmental Medicine 61: 1092–1097. 55. Stark C, Nikopoulou-Smyrni P, Stabrey A, Semler O, Schoenau E (2010) Effect of a new physiotherapy concept on bone mineral density, muscle force and gross motor function in children with bilateral cerebral palsy. J Musculoskelet Neuron Interact 10: 151–158. 34. Sherwood N, Griffin MJ (1992) Evidence of impaired learning 34. Sherwood N, Griffin MJ (1992) Evidence of impaired learning during whole- body vibration. Journal of Sound and Vibration 152: 219–225. body vibration. Journal of Sound and Vibration 152: 219–225. 35. Costa N, Arezes PM, Melo RB (2012) Effects of vibration exposure on professional drivers: A field test for quantifying visual and cognitive performance. Work - A Journal of Prevention Assessment & Rehabilitation 41: 3039–3042. 56. Sitja Rabert M, Rigau Comas D, Fort Vanmeerhaeghe A, Santoyo Medina C, Roque i Figuls M, et al. (2012) Whole-body vibration training for patients with neurodegenerative disease. The Cochrane database of systematic reviews 2: CD009097. 36. Poulton EC (1978) Increased vigilance with vertical vibration at 5-Hz - an alerting mechanism. Appl Ergon 9: 73–76. 57. King LK, Almeida QJ, Ahonen H. (2009) Short-term effects of vibration therapy on motor impairments in Parkinson’s disease. Neurorehabilitation 25: 297–306. 37. Mueller SV, von Schweder AJ, Frank B, Dengler R, Munte TF, et al. (2002) The effects of proprioceptive stimulation on cognitive processes in patients after traumatic brain injury. Arch Phys Med Rehabil 83: 115–121. 58. Barkley RA, Murphy KR (2010) Impairment in occupational functioning and adult ADHD: The predictive utility of executive function (EF) ratings versus EF tests. Archives of Clinical Neuropsychology 25: 157–173. 38. References Van der Zee EA, Riedel G, Rutgers EH, De Vries C, Postema F, et al. (2010) Enhanced neuronal activity in selective brain regions of mice induced by whole body stimulation. Federation of European Neuroscience Societies Abstract 5: 024.49. 15. Vidal-Estrada R, Bosch-Munso R, Nogueira-Morais M, Casas-Brugue M, Ramos-Quiroga JA (2012) Psychological treatment of attention deficit hyperactivity disorder in adults: A systematic review. Actas Esp Psiquiatr 40: 147–154. 30. Ljungberg J, Neely G, Lundstrom R (2004) Cognitive performance and subjective experience during combined exposures to whole-body vibration and noise. Int Arch Occup Environ Health 77: 217–221. 7 PLOS ONE \| www.plosone.org February 2014 \| Volume 9 \| Issue 2 \| e90747 February 2014 \| Volume 9 \| Issue 2 \| e90747 Effects of WBV on Cognition in ADHD Effects of WBV on Cognition in ADHD Effects of WBV on Cognition in ADHD Arias-Carrion O, Poeppel E (2007) Dopamine, learning, and reward-seeking behavior. Acta Neurobiol Exp 67: 481–488. 66. Arnsten AFT, Wang MJ, Paspalas CD (2012) Neuromodulation of thought: Flexibilities and vulnerabilities in prefrontal cortical network synapses. Neuron 76: 223–239. 46. American Psychiatric Association, editor. (1994) Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Press. 47. Regterschot R, van Heuvelen MJG, Zeinstra EB, Fuermaier ABM, Tucha L (submitted) Whole body vibration improves cognition in healthy young adults. 67. Schultz W (2007) Multiple dopamine functions at different time courses. Annu Rev Neurosci 30: 259–288. 48. Cohen J (1988) Statistical power analysis for the behavioral sciences. New Jersey: Hillsdale. 68. Tripp G, Wickens JR (2009) Neurobiology of ADHD. Neuropharmacology 57: 579–589. 49. Hillman CH, Erickson KI, Kramer AF (2008) Be smart, exercise your heart: Exercise effects on brain and cognition. Nature Reviews Neuroscience 9: 58–65. 69. Durston S, Konrad K (2007) Integrating genetic, psychopharmacological and neuroimaging studies: A converging methods approach to understanding the neurobiology of ADHD. Developmental Review 27: 374–395. 50. Heckman JJ (2011) The economics of inequality: The value of early childhood education. American Educator 35: 31–35. 70. So¨derlund GBW, Sikstro¨m S, Loftesnes JM, Sonuga-Barke EJ (2010) The effects of background white noise on memory performance in inattentive school children. Behavioral and Brain Functions 6. 51. Stern Y (2009) Cognitive reserve. Neuropsychologia 47: 2015–2028. 52. del Pozo-Cruz B, Adsuar JC, Parraca JA, del Pozo-Cruz J, Olivares PR, et al. (2012) Using whole-body vibration training in patients affected with common neurological diseases: A systematic literature review. Journal of Alternative and Complementary Medicine 18: 29–41. 71. So¨derlund G, Sikstro¨m S, Smart A (2007) Listen to the noise: Noise is beneficial for cognitive performance in ADHD. Journal of Child Psychology and Psychiatry 48: 840–847. February 2014 \| Volume 9 \| Issue 2 \| e90747 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 8
https://openalex.org/W2026255607	https://hal.science/hal-01765110/document	English	null	Quantitative Analysis of Lavender (&lt;i&gt;Lavandula angustifolia&lt;/i&gt;) Essential Oil Using Multiblock Data from Infrared Spectroscopy	American journal of analytical chemistry	2,014	cc-by	9,603	Quantitative Analysis of Lavender (Lavandula angustifolia) Essential Oil Using Multiblock Data from Infrared Spectroscopy Nathalie Dupuy, Vincent Gaydou, Jacky Kister Nathalie Dupuy, Vincent Gaydou, Jacky Kister To cite this version: Nathalie Dupuy, Vincent Gaydou, Jacky Kister. Quantitative Analysis of Lavender (Lavandula an- gustifolia) Essential Oil Using Multiblock Data from Infrared Spectroscopy. American Journal of Analytical Chemistry, 2014, 05 (10), pp.633-645. 10.4236/ajac.2014.510071. hal-01765110 How to cite this paper: Dupuy, N., Gaydou, V. and Kister, J. (2014) Quantitative Analysis of Lavender (Lavandula angustifo- lia) Essential Oil Using Multiblock Data from Infrared Spectroscopy. American Journal of Analytical Chemistry, 5, 633-645. http://dx.doi.org/10.4236/ajac.2014.510071 Quantitative Analysis of Lavender (Lavandula angustifolia) Essential Oil Using Multiblock Data from Infrared Spectroscopy Nathalie Dupuy, Vincent Gaydou, Jacky Kister Nathalie Dupuy, Vincent Gaydou, Jacky Kister Aix Marseille University, LISA, METICA, EA4672, F-13397 Marseille 20, France Email: nathalie.dupuy@univ-amu.fr Received 30 April 2014; revised 14 June 2014; accepted 26 June 2014 Copyright © 2014 by authors and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licenses/by/4.0/ Copyright © 2014 by authors and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licenses/by/4.0/ HAL Id: hal-01765110 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. American Journal of Analytical Chemistry, 2014, 5, 633-645 Published Online July 2014 in SciRes. http://www.scirp.org/journal/ajac http://dx.doi.org/10.4236/ajac.2014.510071 American Journal of Analytical Chemistry, 2014, 5, 633-645 Published Online July 2014 in SciRes. http://www.scirp.org/journal/ajac http://dx.doi.org/10.4236/ajac.2014.510071 Abstract Near-infrared and mid-infrared spectroscopies were currently used to analyze natural compounds. During the last ten years various multiblocks methods were developed such as Concatenated PLS, Hierarchical-PLS (H-PLS), and MultiBlock-PLS (MB-PLS). These three algorithms were used to analyze 55 lavender (Lavandula angustifolia) essential oil samples. The results obtained were compared to the ones obtained respectively in NIR and MIR ranges. The accuracies of the models depend on the spectroscopic technique, pretreatment and the PLS methods. The results showed that the choice of the factor numbers used to build the multiblock models was the most important parameter for the H-PLS and MB-PLS methods. Keywords Keywords 2.2. Gas Chromatography The composition of the 55 lavender essential oil samples, was determined by internal normalization using an Agilent 7890A GC System, equipped with a capillary column SUPELCOWAX 10 of 30 m (DI: 25 µm, phase thickness: 0.25 µm), a FID detector, dihydrogen as gas vector, and following the analysis protocol established by the French National Association for Normalization (AFNOR) [13]. The mean composition was done in the Table 1. 2.1. Essential Oil Samples The lavender (Lavandula angustifolia) essential oils investigated in this work were provided by the Office Na- tional Interprofessionnel des Plantes à Parfum, Aromatiques et Médicinales (ONIPPAM) and they belong to the label “AOC Huiles essentielles de Lavande de Haute-Provence” (Origin Protected Designation of “Haute-Pro- vence”). The oils were obtained by vapor phase distillation of the flowery part recently cut, of the L. angustifolia species, spontaneous growing or cultivated in south of France. The samples were left in two groups, including a group of calibration (42 samples) and a group of validation (13 samples). The samples with the minimum and maximum contents for each compound investigated were added in calibration set. The other samples were ran- domly selected between prediction and calibration set. 1. Introduction angustifolia es- sential oils are used in perfumery, cosmetic, for antimicrobial activity, and anti-colic properties [11]. These oils are characterized by high level of linalool and linalyl acetate, moderate levels of lavandulyl acetate, terpen-4-ol and lavandulol. The amount of 1,8-cineole and camphor often vary between very low to moderate [12]. In this work, five compounds of lavender’s essential oil were quantified by each chemometric method and these meth- ods were compared. Four chemometric methods have been used to build the regression models: PLS method was used independently on NIR and MIR data, and three multiblock methods (concatenated, H-PLS and MB- PLS methods) were applied on NIR and MIR data in a simultaneous way. 1. Introduction Vibrational spectroscopies such as NIR and MIR, when associated to multivariate analysis, have been proved to be a powerful tool in various product analyses like gasoline samples [1], diesels [2], fuel [3] [4], olive oils [5] [6] or lavandin essential oils [7]. These analytical spectroscopic methods, besides being shorter in time than the usual ones (ASTM methods), present good accuracy and precision; are non-destructive; and can be used for quality control. During the last decade, a lot of progress appeared in the analytic world. The time necessary to obtain analytic data decreases, so for one manufactured product, multiple measurements are done (NIR, MIR spectro- scopies, liquid or gas chromatography, sensorial analysis). The data are considered as independents; the NIR or MIR data could be used to quantify compounds by using regression methods as Partial Least Square (PLS) re- N. Dupuy et al. gression. Each group of variables, or each matrix, is usually called a block and is measured on the same obser- vations (in rows). It is possible to find complementary information using two different analytic methods. In this case the analyst could use multiples predictors blocks and multiblocks responses blocks. In the literature several multiblocks methods were described [8]. The first method called concatenated method consists in concatenating the descriptor block into the same matrix and then applying Partial Least Square (PLS) regression. Sometimes this method works well, but the individual blocks must be scaled to obtain interpretable results. The second method consists in considering each block independently at the beginning. Principal Component Analysis (PCA) could be applied on each block and then the scores obtained in each block are collected together to form a super matrix. The PLS regression is then applied on the super matrix. This method presented by Tenenhaus and Vinzi [9] is called Hierarchical-PLS (H-PLS). The third method is proposed by Wangen and Kowalski [10], the PLS regression is applied on each block and then the scores obtained in each block are collected together to form a super matrix. The PLS regression is then applied on the super matrix. This whole process is called MultiBlock- PLS (MB-PLS). The objective of this study is to realize the determination of the five main compounds which compose the lavender (Lavandula angustifolia) essential oil, using multiblock chemometric methods. This plant, native from the Mediterranean Basin, is widely cultivated for essential oil production. Pure L. 2.3. FT-NIR FT-NIR spectra were recorded with a Nicolet Antaris spectrometer interfaced to a personal computer. Samples were filled into glass tubes of 2 mm. All spectra were computed at 8 cm−1 resolution between 4000 and 10,000 cm−1, thanks to the software result integration 2.1 Thermo Nicolet. Co-addition of symmetrical interferograms on 100 scans was performed for each spectrum. A reference spectrum and sample spectrum was recorded with glass tubes of 2 mm. 3. Chemometric Methods 3. Chemometric Methods 2.4. FT-MIR FT-MIR spectra were recorded with a Thermo Nicolet AVATAR 370 spectrometer interfaced to a personal computer. Samples were deposed on an Attenuated Total Reflectance (ATR) accessory. All spectra were com- 634 Table 1. Products identified by GC. Retention time (s) Mean (%) 3-octanone 314 0.88 Limonene 385 0.54 1,8-cineole 391 0.32 (Z)-β-ocimene 405 3.08 (E)-β-ocimene 428 1.92 Linalool 572 28.96 Camphor 662 0.29 Borneol 722 0.91 Lavandulol 751 3.56 Terpinen-4-ol 789 0.49 α terpineol 799 0.35 Linalyl acetate 1042 37.03 Lavandulyl acetate 1144 4.12 E-β-caryophyllene 1545 3.73 puted at 2 cm−1 resolution between 700 and 4000 cm−1. Co-addition of symmetrical interferograms on 100 scans was performed for each spectrum. A reference spectrum was recorded before each sample spectrum. puted at 2 cm−1 resolution between 700 and 4000 cm−1. Co-addition of symmetrical interferograms on 100 scans was performed for each spectrum. A reference spectrum was recorded before each sample spectrum. 3.1. Partial Component Analysis (PCA) Principal component analysis [14] involves a mathematical procedure that transforms a number of possibly cor- related variables into a smaller number of uncorrelated variables called principal components. The first principal component accounts for as much of the variability in the data as possible, and each succeeding component ac- counts for as much of the remaining variability as possible. Data sets with many variables can be simplified through variable reduction and thereby be more easily interpreted. 2.5. Software The chemometric applications were performed by the UNSCRAMBLER software Version 9.2 from CAMO (Computer Aided Modelling, Trondheim, Norway) and by the MATLAB software Version 7.4 from The Math Works Inc. (Natick, Units States). 3.2. Partial Least Square (PLS) Regression 3.2. Partial Least Square (PLS) Regression PLS is a supervised analysis that is based on the relation between the signal intensity and the characteristics of the sample [15]. Interference and overlapping of the spectral information may be overcome by using powerful multicomponent analysis such as PLS regression. PLS allows a sophisticated statistical approach using a spectral region rather than unique and isolated analytical bands [16] [17]. The algorithm is based on the ability to ma- thematically correlate spectral data to a property matrix of interest while simultaneously accounting for all other significant spectral factors that perturb the spectrum. It is thus a multivariate regression method that uses a se- lected spectral region and is based on the use of latent variables. To construct a model, the first step is to per- form a calibration. This involves collecting a calibration set of reference samples which should contain all chemical and physical variations to be expected in the unknown samples, which will be predicted later. The purpose of this calibration is to establish a multiple linear regression between the NIR spectra data or MIR spec- tra and the various parameters of the sample set [volatile compounds (water in the majority), lipid rates, or va- 635 rietal origins]. Cross-validation was applied in regression to fix the required number of latent variables for mod- el construction. So, the optimal number of latent variables is determined on the basis of prediction of samples kept out from the individual model. The second step is to validate the model using a prediction set (different from the calibration one), i.e. to compare the values obtained by the model to the values obtained by the refer- ence method. The evaluation of the calibration performance is estimated by computing the standard error of calibration (RMSEC) after comparing the real modification with the computed one for each component. The formula for the standard error of calibration is [18]: ( ) 2 1 RMSEC 1 N i i i C C N p = ′ − = −− ∑ (1) (1) where Ci is the known value, i C′ is the calculated value, N is the number of samples and p is the number of in- dependent variables in the regression. 3.2. Partial Least Square (PLS) Regression The standard error of prediction (RMSEP) gives an estimation of the prediction performance during the step of validation of the calibration equation: ( ) 2 1 RMSEP N i i i C C M = ′ − = ∑ (2) (2) where Ci is the known value, i C′ is the value calculated by the calibration equation, and M is the number of samples in the prediction set. where Ci is the known value, i C′ is the value calculated by the calibration equation, and M is the number of samples in the prediction set. The predictive ability of the model should also be expressed by the bias and the square of correlation coeffi- cient (R2) also called determination coefficient, usually called Q2 in prediction. The regression coefficients are the numerical coefficients which express the link between the predictor variations and the response variations. The bias is systematic difference between predicted and measured values. The bias is computed as average value of the residuals. The residual is the measure of the variation which is not taken into account by the model. The residual for a given sample and a given variable is computed as the difference between observed value and fitted (projected or predicted) value of the variable on the sample. For this study, the full cross validation is chosen to validate all models. Data arrangement [19] for combined-PLS: Data arrangement [19] for combined-PLS: Data arrangement [19] for combined-PLS: g X-block: virgin olive oil spectra data have been arranged in two ways by taking NIR and MIR absorbances as columns, yielding the X-matrix (the X-matrix data constituted the independent set of variables). Y-block: the Y-block data were the set of dependent variables. For combining NIR and MIR spectra, the data were normalized at the unit vector in order to give the same importance for the two spectral regions. 3.3. Multiblock Methods Nowadays, it is possible to arrange, for the same sets of samples, several blocks of analytical variables. These various data sets or these various blocks are obtained by means of various analytical methods, for example data stemming from the spectroscopy NIR and from the spectroscopy MIR. Every block contains information relative to the variance of samples and it was intended to use multiblock methods because. These methods are able to treat all the blocks in a simultaneous way and three multiblock methods were used for this study. Data arrangement for S-PLS models: Data arrangement for S-PLS models: g The third way was proposed by Wangen and Kowalski [10] Partial Least Square (PLS) regression was applied on each block and then the scores obtained in each block were collected together to form a super matrix. The PLS regression was then applied on the super matrix. This method was called S-PLS. The main difference be- tween this algorithm and the algorithm of H-PLS is that for the S-PLS, the scores of every block of data are cal- culated by means of a PLS associated with quantitative information. Then the blocks of scores are concatenated to form the “super block” matrix of and finally a new PLS is realized on this super block matrix. The method of the S-PLS was worked out to treat relations of variance (and of covariance) between several blocks of data. This method is able of taking into account the variances of K blocks of analytical variables and of putting them in connection with several blocks of variables to be explained (quantitative information). The algorithm of the me- thod S-PLS is also clearly presented in the publication of Westerhuis et al. [25]. y p p X-block: PLS scores of the lavender oil NIR and MIR data have been calculated separately and concatenated. Y-block: the Y-block data were the set dependent variable. b oc : S sco es o t e ave de o N a d data ave bee ca cu ated sepa ate y a d co cate ated. Y-block: the Y-block data were the set dependent variable. 4. Results and Discussion Compound identification of the lavender samples were done using retention indices and co-elution with authen- tic samples of the five compounds investigated. The percentages were determined by the method of area norma- lization and without the application of response factor corrections according to standard methods [25]. The va- riables to be explain (quantitative content data) were obtained by internal normalization. In Table 2, the mean and range of the main five compounds, i.e. linalool, lavandulol, linalyl acetate, lavan- dulyl acetate and ß-caryophyllene are given with both calibration and validation sets according to the targeted compounds. As shown in this table, linalool and linalyl acetate are the main compounds (28.8% and 37.2% re- spectively for the calibration set). The three other compounds present a range around 5%, in the two sets. The spectroscopic variables from NIR and MIR were used for building chemometrics models using multiblock me- thods, concatenated block, hierarchical PLS method and MB-PLS method. The efficiently of NIR and MIR ranges were first studied, then the 2 spectral ranges were concatenated, the H-PLS and S-PLS methods were checked. For each of following studies same groups of samples in calibration and in prediction were preserved. 3.5. H-PLS Models This method presented by Wold et al. [20] was called H-PLS. Data arrangement: X-block: PCA of the lavender oil NIR and MIR data have been calculated separately. PCA was performed on the calibration set and the same model was used for the decomposition of the prediction one. The scores of each PCA were extracted in order to build another data matrix named TT. Y-block: the Y-block data were the set dependent variable. 636 N. Dupuy et al. A PLS cycle [21] is done between TT and each predictor from which a super weight and an updated super score TT are obtained normalized to unit length [22]. These cycles are repeated until TT converged [23]. Evalua- tion of error for this method is estimated as a classical PLS. The algorithm of the method H-PLS is very clearly presented in the publication of Westerhuis et al. [24]. A PLS cycle [21] is done between TT and each predictor from which a super weight and an updated super score TT are obtained normalized to unit length [22]. These cycles are repeated until TT converged [23]. Evalua- tion of error for this method is estimated as a classical PLS. The algorithm of the method H-PLS is very clearly presented in the publication of Westerhuis et al. [24]. 3.7. Data Pretreatment Data analysis was carried out using the full spectra. Mean centering was used to improve the smaller spectral differences removing the common information from the spectra. Absorbance normalized value was also em- ployed. None of the other mathematical treatments (multiplicative scatter correction, second derivative [26], etc.) or wavelength ranges tested improved the prediction accuracy of models. During the data processing, the Stan- dard Normal Variate (SNV) correction pretreatment [27] could be used. The SNV pretreatment is a row-oriented transformation that removes scatter effects from spectra by centering and scaling each individual spectrum. To perform the variable matrices, some pretreatments (or preprocessings) were done. This step is very important to study matrix of spectra. The pretreatments could be combined in order to optimize the models. In this work, in- fluences of pretreatments were compared. 4.1. Near Infrared Studies Several models of regression were then elaborated with samples of calibration to predict contents in linalool, la- vandulol, linalyl acetate, lavandulyl acetate and ß-caryophyllene in the [4500 - 5000 cm−1] and [6000 to 7200 cm−1] spectral ranges. Various pretreatments were tested for each targeted compounds. The pretreatments which allowed obtaining the best regression models were baseline correction followed by SNV for the 5 compounds. The best calibration model characteristics appear in left part of Table 3. The number of factors used for each ofthese models of regression varies from 2 to 5 and R2 obtained for linalool (0.99), linalyl acetate (0.99) and la- 637 N. Dupuy et al. N. Dupuy et al. Table 2. Content of the five main compounds in lavender essential oil samples investigated. Table 2. Content of the five main compounds in lavender essential oil samples investigated. Linalool (%) Lavandulol (%) Linalyl acetate (%) Lavandulyl acetate (%) ß-caryophyllene (%) Calibration set N = 42 Min-Max 23.4 - 33.8 1.75 - 5.31 28.5 - 41.5 3.05 - 4.90 0.49 - 5.48 Mean 28.8 3.38 37.2 4.12 3.70 Deviation standard 2.14 0.79 2.56 0.42 1.11 Prediction set N = 13 Min-Max 24.8 - 32.8 1.90 - 5.07 32.9 - 40.0 3.43 - 4.84 1.28 - 5.24 Mean 29.4 3.91 36.8 4.12 3.78 Deviation standard 1.92 0.66 1.96 0.36 1.18 Table 3. Characteristics of PLS models regression (NIR, MIR and concatenated data). 4.1. Near Infrared Studies Compounds FN R2 RMSEC (%) REC (%) Q2 RMSEP (%) REC (%) Near infrared Linalool 5 0.99 0.199 0.69 0.96 0.399 1.4 Lavandulol 3 0.92 0.254 7.5 0.71 0.321 9.5 Linalyl acetate 5 0.99 0.247 0.67 0.76 1.58 4.2 Lavandulyl acetate 2 0.76 0.202 4.9 0.54 0.331 8.0 ß-caryophyllene 4 0.89 0.161 4.4 0.53 0.353 9.5 Mid infrared Linalool 8 0.99 0.071 0.25 0.97 0.269 0.91 Lavandulol 15 0.99 0.013 0.38 0.85 0.254 7.5 Linalyl acetate 7 0.98 0.352 0.95 0.64 1.12 3.0 Lavandulyl acetate 9 0.99 0.010 0.24 0.90 0.144 3.5 ß-caryophyllene 13 0.99 0.011 0.30 0.78 0.261 7.0 Concatenated matrix Linalool 10 0.99 0.048 0.17 0.98 0.277 0.94 Lavandulol 8 0.99 0.045 1.3 0.70 0.323 9.6 Linalyl acetate 10 0.99 0.061 0.16 0.74 1.09 2.9 Lavandulyl acetate 13 0.99 0.012 0.29 0.79 0.243 5.9 ß-caryophyllene 9 0.99 0.025 0.68 0.69 0.327 8.8 FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative error of calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. 2 Linalool (%) Lavandulol (%) Linalyl acetate (%) Lavandulyl acetate (%) ß-caryophyllene (%) FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative erro calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. vandulol (0.92) are close to 1. On the other hand, for lavandulyl acetate and ß-caryophyllene, calculated R2 (0.76 and 0.89 respectively) are low. vandulol (0.92) are close to 1. On the other hand, for lavandulyl acetate and ß-caryophyllene, calculated R2 (0.76 and 0.89 respectively) are low. The calculated RMSEC is included between 0.161% and 0.254%. These values are relatively close to some of the others and they express the error of prediction realized by the models to predict the compound contents. When the RMSEC is expressed with respect to the average compound content, we calculate the REC which are relative errors and allow more easily comparison models between them. So REC obtained for linalool (0.69%) and linalyl acetate (0.67%) are particularly close to zero (with regard to three other REC) and explain a good quality of prediction of these compounds contents in calibration samples. Then, the five models of regression are validated by means of the validation sample’s set. N. Dupuy et al. 4.1. Near Infrared Studies Q2 obtained are respectively lower than R2 and the RMSEP is respectively higher to RMSEC. The calculated REP translates a good prediction of the linalool content of the validation samples. The REP of the other models presented low prediction qualities in particular for the lavan- 638 N. Dupuy et al. dulol and ß-caryophyllene because the REC obtained for these two compounds are around 10%. We can observe that the validation of the regression model of the lavandulyl acetate is bad because the REP is equal to 4.2% while the REC obtained during the calibration of this model was 0.67%. N. Dupuy et al. dulol and ß-caryophyllene because the REC obtained for these two compounds are around 10%. We can observe that the validation of the regression model of the lavandulyl acetate is bad because the REP is equal to 4.2% while the REC obtained during the calibration of this model was 0.67%. dulol and ß-caryophyllene because the REC obtained for these two compounds are around 10%. We can observe that the validation of the regression model of the lavandulyl acetate is bad because the REP is equal to 4.2% while the REC obtained during the calibration of this model was 0.67%. 4.3. Multiblock Method: NIR + MIR Studies The multiblock method allowed organizing chemical information from NIR and MIR data. Three multiblock methods were compared: the concatenated matrix (CONC) method, the hierarchical PLS (H-PLS) method and the MultiBlock-PLS (MB-PLS) method. All the regression models were elaborated from NIR and MIR data of the calibration sample set. The pretreatments used were baseline correction followed by a SNV on NIR data and no pretreatment on MIR data. The spectral ranges previously determined, were kept for application of multib- lock methods. 4.2. Medium Infrared Studies With calibration samples set, 5 regression models were then elaborated in the 700 and 1900 cm−1 spectral range to predict contents in linalool, lavandulol, linalyl acetate, lavandulyl acetate and ß-caryophyllene of the calibra- tion set samples. The various pretreatments were tested for each compounds, however the best regression mod- els were obtained without making any pretreatment. The best characteristics of regression models obtained are given in Table 3 which shows that, without any pretreatment, it was possible to elaborate good calibration mod- els. R2 obtained are all very close to one. The RMSEC are very close to zero in comparison with those obtained with the NIR data. The models are then validated and the characteristics of the validation of each regression model are given in Table 2. Q2 obtained are respectively lower than R2 in particular for linalyl acetate regression model (0.64 and 0.98 respectively for this compound). The RMSEP are higher to the RMSEC (in particular for the linalyl acetate) and the REP obtained for linalool regression model is particularly successful (REP = 0.91%). Other regression models present relatively small REP (3% and 3.5% respectively for linalyl and lavandulyl ace- tates; and 7% and 7.5% respectively for lavandulol and ß-caryophyllene). The pretreatments of baseline correc- tion and SNV used with the NIR data allowed to improve regression models, while the MIR data allowed to ob- tain good qualities of regression models without any pretreatment. On the other hand, factors number used with the MIR data is higher to those used by the NIR data and the REP obtained by means of the MIR data are small- er than those obtained with the NIR data about is the targeted compound. In conclusion, MIR spectroscopy is the most suitable spectroscopic domain. However, it is possible that simultaneous use of these two spectroscopic domains allows elaborating even more precise regression models and it is in this optic that the multiblocks me- thods were applied. 4.3. Multiblock Method: NIR + MIR Studies 4.3.1. Study on Concatenated Matrix 9 3 2 8 7 3 . 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 a r, P C ): ( 1 3 L a v a nd u ly l a c et at e, 1) B 0 = 4 .6 8 29 3 7 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 850 700 1200 1050 1550 1400 6000 4500 6950 Regression coefficients 1750 linalool Linalyl acetate Lavandulyl acetate lavandulol caryophyllene Wavenumber (cm -1) linalool Figure 1. First vector of regression of 5 targeted compounds (CONC). set up with the calibration samples set. Previously, 42 samples participated in the calibration. 10 of these sam- ples were chosen to constitute the “test set” of H-PLS. Then, by means of 32 remaining calibration samples, 31 regression models of H-PLS were elaborated with a PCA vector score number varying between 1 to 31. The great matrices scores thus obtained present dimensions going from 2 to 62 variables because the number of vec- tors scores calculated for every block of data is identical. 31 regression models of H-PLS were then validated by means of the “test set”. Figure 2 concerns only linalool content and shows on vertical axis, the RMSEP’s inten- sity of the test set, in function of 2 variables. The first variable (in-depth axis) is the number of PCA score vector calculated for each block (NIR and MIR block) during the first stage of H-PLS. The second variable (horizontal axis) is the number of PLS score vector calculated during the second stage of H-PLS. 4.3.1. Study on Concatenated Matrix The two calibration matrixes sample set were normalized after the pretreatment then the data were concatenated. The calibration model characteristics are presented according to the targeted compound, in Table 3. The regres- sion models elaborated for each compound presented good accuracy. Indeed R² are very close to 1 and the RMSEC are close to 0. The calculated REC are thus low. The regression models were then validated with the validation sample set. Q2 are smaller than R2 (for the same compound) in particular for ß-caryophyllene of which R2 is 0.998 while Q2 is 0.69. The obtained RMSEP are significantly higher in relation to RMSEC. The REP obtained for linalyl acetate is particularly interesting because it is lower than the obtained REP (for the same compound) by using only MIR data (3.0% with MIR data and 2.9% with the concatenated matrix method). The bias of validation of each model is relatively low in relation to the mean content of corresponding com- pound. The Figure 1 presents the first vectors of regression of the five targeted compounds. These regression vectors are expressed according to NIR and MIR spectroscopic variables. The regression coefficients stemming from MIR data are on the left of the dotted line and those stemming from NIR data are positioned to the right of this line. The intensity of the coefficients stemming from MIR data is higher than those outcomes from NIR data. This observation confirms the relevance of the MIR data with regard to the NIR data. 4.3.2. Study on H-PLS y The method of hierarchical PLS (H-PLS) was organized by following the algorithm described by Westerhuis [24]. To estimate the optimal factor number used for the first stage of H-PLS (the PCA stage), a “test set” was 639 N. Dupuy et al. Figure 1. First vector of regression of 5 targeted compounds (CONC). set up with the calibration samples set. Previously, 42 samples participated in the calibration. 10 of these sam- ples were chosen to constitute the “test set” of H-PLS. Then, by means of 32 remaining calibration samples, 31 regression models of H-PLS were elaborated with a PCA vector score number varying between 1 to 31. The great matrices scores thus obtained present dimensions going from 2 to 62 variables because the number of vec- tors scores calculated for every block of data is identical. 4.3.1. Study on Concatenated Matrix 31 regression models of H-PLS were then validated by means of the “test set”. Figure 2 concerns only linalool content and shows on vertical axis, the RMSEP’s inten- sity of the test set, in function of 2 variables. The first variable (in-depth axis) is the number of PCA score vector calculated for each block (NIR and MIR block) during the first stage of H-PLS. The second variable (horizontal axis) is the number of PLS score vector calculated during the second stage of H-PLS. This Figure 2 has the shape of a quarter of bowl. When the number of vectors scores is low, the test set’s RMSEP values are high and they exceed the scale of the RMSEP. H i i f ’ RMSEP i b d Th di f hi i i i d i h 7 0 0 . 1 4 6 7 8 6 . 9 3 2 8 7 3 . 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 -v a r, P C ): ( 0 6 L in a lo l, 1 ) B0 = 23 .5 4 73 0 6 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 7 0 0 . 1 4 6 7 8 6 . 9 3 2 8 7 3 . 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4.3.1. Study on Concatenated Matrix 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 Y -v a r, P C ): ( 1 2 L in a ly l a ce ta te ,1 ) B 0 = -7 .1 0 69 5 9 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 7 0 0 . 1 4 6 7 8 6 . 9 3 2 8 7 3 . 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 a r, P C ): ( 1 3 L a v a nd u ly l a c et at e, 1) B 0 = 4 .6 8 29 3 7 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 850 700 1200 1050 1550 1400 6000 4500 6950 Regression coefficients 1750 linalool Linalyl acetate Lavandulyl acetate lavandulol caryophyllene Wavenumber (cm -1) Figure 1. First vector of regression of 5 targeted compounds (CONC). 7 0 0 . 1 4 6 7 8 6 . 9 3 2 8 7 3 . 4.3.1. Study on Concatenated Matrix 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 -v a r, P C ): ( 0 6 L in a lo l, 1 ) B0 = 23 .5 4 73 0 6 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 7 0 0 . 1 4 6 7 8 6 . 9 3 2 8 7 3 . 7 1 9 9 6 0 . 5 0 5 1 0 4 7 . 2 9 2 1 1 3 4 . 0 7 8 1 2 2 0 . 8 6 5 1 3 0 7 . 6 5 1 1 3 9 4 . 4 3 8 1 4 8 1 . 2 2 4 1 5 6 8 . 0 1 1 1 6 5 4 . 7 9 7 1 7 4 1 . 5 8 4 1 8 2 8 . 3 7 0 4 5 5 8 . 8 9 6 4 9 0 6 . 0 2 0 6 2 4 8 . 2 3 2 6 5 9 5 . 3 5 6 6 9 4 2 . 4 8 0 Y -v a r, P C ): ( 1 2 L in a ly l a ce ta te ,1 ) B 0 = -7 .1 0 69 5 9 X - V a r ia b le s R e g r e s s io n C oe f ic ie nt s (B ) 7 0 0 . 1 4 6 7 8 6 . 4.3.1. Study on Concatenated Matrix This Figure 2 has the shape of a quarter of bowl. When the number of vectors scores is low, the test set’s RMSEP values are high and they exceed the scale of the RMSEP. However a minimum of test set’s RMSEP is observed. The coordinates of this minimum are registered in the white box on Figure 2. For each compound the number of PCA scores included in the model varied and the best results are obtained for 21 scores for the linalool, 12 scores for the lavandulol, 9 for linalyl acetate, 8 for lavan- dulyl acetate and 7 for ß-caryophyllene. Then, new regressions models were elaborated by means of 42 calibra- tion samples of calibration set. NIR and MIR data were used in the algorithm of the H-PLS after the variable pretreatments (baseline correction + SNV for NIR data) and by using only the selected spectroscopic variables. For each compound, the optimal number of vector scores calculated by blocks is given in Table 4. These numbers were optimized by means of the test set (in this way, the dimension of the H-PLS great matrix scores, is equal to the double of the number of vector scores calculated by block, because two blocks of data are used in this study). The characteristics of the regression models obtained for each targeted compounds are given in Table 4 (in H-PLS part). R2 calculated are close to 1 with the exception of lavandulyl acetate (0.76) and of ß-caryophyllene (0.52). Obtained RMSEC are small with regard to the average contents of targeted compounds (with the exception of ß-caryophyllene). The H-PLS model is then validated with the validation sample set. Characteristics of validation are given in Table 3. Q2 concerning the regression model of linalool (0.89) is the closest to one. Q2 for other regression models between 0.8 and 0.3, indicate a relatively bad validation. The ob- 640 N. Dupuy et al. Figure 2. Test set’s RMSEP in function of the number of score calculated for each block (first stage of H-PLS) and in function of regression number vector (second stage of H-PLS) for linalool analysis. Figure 2. Test set’s RMSEP in function of the number of score calculated for each block (first stage of H-PLS) and in function of regression number vector (second stage of H-PLS) for linalool analysis. Figure 2. 4.3.1. Study on Concatenated Matrix y The multiblock PLS (S-PLS) method was organized by following the algorithm described by Westerhuis [24]. As for H-PLS study, the optimal PLS score number calculated, to make the super score matrix, was estimated by means of a “test set” by 10 samples. The 31 supers scores matrixes were built by means of 32 remaining samples of calibration (outside the test set). The dimensions of these matrices are included between 2 and 62. 31 models of regression S-PLS are then validated by means of the test set. However a minimum of test set’s RMSEP is ob- served. The number of PLS scores calculated for each block was optimized to 9 for the linalool, 3 for the lavan- dulol, 10 for linalyl acetate, 3 for lavandulyl acetate and for ß-caryophyllene. Then, models of regression are elaborated by means of 42 samples of calibration set. NIR and MIR data were used in the algorithm of the S- PLS after the pretreatment. For each compound, the number of vector scores calculated by block is given in Ta- ble 5. Five regression models were elaborated and the characteristics of these models are given in Table 5. Calcu- lated R2 are close to one for linalool and linalyl acetate. Concerning lavandulol and lavandulyl acetate, R2 are respectively equal to 0.73 and 0.70. It indicates that the regression models concerning these two products are of less efficient than the regression models concerning linalool and linalyl acetate. R2 obtained for ß-caryophyllene, is around 0.6 showing a bad regression model. The obtained RMSEC are not small any more with regard to the average content of the targeted compounds. So, the obtained REC are high, in particular for ß-caryophyllene of which the REC is 18%. The S-PLS regression models were validated by validation sample set and the validation characteristics are given in Table 5. Q2 obtained are sharply lower than corresponding R2, respectively for each of targeted compounds. The RMSEP are then more brought up than the RMSEC and the obtained REP are thus of less good quality than the REC. For example for the lavandulol, the REC is equal to 12% while the REP is 16%. The model of regression of linalool presented satisfactory characteristics of validation since the REP is 1.95%. The validation characteristics of linalyl and lavandulyl acetate models are less good because the calcu- lated REP for these models is 3.45% and 6.86% respectively. 4.3.1. Study on Concatenated Matrix Test set’s RMSEP in function of the number of score calculated for each block (first stage of H-PLS and in function of regression number vector (second stage of H-PLS) for linalool analysis. Table 4. Characteristics of H-PLS regression models. Compounds FN R2 RMSEC (%) REC (%) Q2 RMSEP (%) REC (%) Linalool 4 0.94 0.496 1.7 0.89 0.536 1.8 Lavandulol 16 0.99 0.073 2.2 0.67 0.495 14 Linalyl acetate 14 0.97 0.416 1.1 0.78 1.01 3.0 Lavandulyl acetate 8 0.76 0.197 4.8 0.32 0.313 7.6 ß-caryophyllene 12 0.52 0.719 19 0.39 0.913 24 FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative error of calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative error of calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. tained REP are also high, particularly for the lavandulol (14.7%) and ß-caryophyllene (24.7%). For the linalool, the first regression vector calculated using the H-PLS method includes 42 variables. These variables correspond to PCA vector scores, normalized and concatenated. These vectors form the super score matrix and it is from this matrix that the regression vectors are calculated. Figure 3 gives a graphic representation of the H-PLS first regression vector of linalool. Variables stemming from vectors score PCA of MIR data is to the left of the dotted line and those outcomes of vectors score PCA of NIR data is positioned to the right. This vector of regression allows observing the weight of vectors scores PCA stemming from MIR and NIR spectroscopic domains. The vectors scores 5 and 13 are the most important concerning the block of MIR data. This report is surprising be- cause it is the first vectors calculated scores which explain the maximum of variance. It thus means that the use- ful information is scattered in the data and it is not significant towards the general variance of the data. The first vector scores calculated with the block of NIR data is the one which has most importance for this spectral do- main. 641 N. Dupuy et al. Figure 3. First vector of regression of linalool (H-PLS). 4.3.3. 4.3.1. Study on Concatenated Matrix Study on MultiBlock-PLS The multiblock PLS (S-PLS) method was organized by following the algorithm described by Westerhuis [24]. As for H-PLS study, the optimal PLS score number calculated, to make the super score matrix, was estimated by means of a “test set” by 10 samples. The 31 supers scores matrixes were built by means of 32 remaining samples of calibration (outside the test set). The dimensions of these matrices are included between 2 and 62. 31 models of regression S-PLS are then validated by means of the test set. However a minimum of test set’s RMSEP is ob- served. The number of PLS scores calculated for each block was optimized to 9 for the linalool, 3 for the lavan- dulol, 10 for linalyl acetate, 3 for lavandulyl acetate and for ß-caryophyllene. Then, models of regression are elaborated by means of 42 samples of calibration set. NIR and MIR data were used in the algorithm of the S- PLS after the pretreatment. For each compound, the number of vector scores calculated by block is given in Ta- ble 5. Five regression models were elaborated and the characteristics of these models are given in Table 5. Calcu- 2 2 Figure 3. First vector of regression of linalool (H-PLS). 4.3.3. Study on MultiBlock-PLS y The multiblock PLS (S-PLS) method was organized by following the algorithm described by Westerhuis [24]. As for H-PLS study, the optimal PLS score number calculated, to make the super score matrix, was estimated by means of a “test set” by 10 samples. The 31 supers scores matrixes were built by means of 32 remaining samples of calibration (outside the test set). The dimensions of these matrices are included between 2 and 62. 31 models of regression S-PLS are then validated by means of the test set. However a minimum of test set’s RMSEP is ob- served. The number of PLS scores calculated for each block was optimized to 9 for the linalool, 3 for the lavan- dulol, 10 for linalyl acetate, 3 for lavandulyl acetate and for ß-caryophyllene. Then, models of regression are elaborated by means of 42 samples of calibration set. NIR and MIR data were used in the algorithm of the S- PLS after the pretreatment. For each compound, the number of vector scores calculated by block is given in Ta- ble 5. 4.3.1. Study on Concatenated Matrix This regression vector allows observing the weight of PLS vector scores stemming from spectroscopic domains. We can notice that it is the first calculated PLS vector scores which have most importance concerning the block of MIR data. Whereas calculated PLS vector scores with the block of NIR data present low intensities in regression coefficient except for vector scores 2 and 9 for the NIR block (vectors scores 11 and 18 on Figure 4). The vector score 18 explain a small proportion of the va- riance (lower than 1%) but the corresponding regression coefficient has its vector score is brought up. It means that a very low variance influenced strongly the regression. 4.3.1. Study on Concatenated Matrix The ß-caryophyllene regression model is bad with a REP higher than 25%. g Figure 4 gives a graphic representation of the first S-PLS regression vector concerning linalool. Regression 642 N. Dupuy et al. N. Dupuy et al. Figure 4. First vector of regression of linalool (S-PLS). Table 5. Characteristics of S-PLS regression models. Compounds FN R2 RMSEC (%) REC (%) Q2 RMSEP (%) REC (%) Linalool 5 0.099 0.104 0.36 0.87 0.57 2.0 Lavandulol 5 0.73 0.408 12 0.45 0.54 16 Linalyl acetate 4 0.99 0.254 0.68 0.67 1.28 3.5 Lavandulyl acetate 13 0.70 0.217 5.3 0.46 0.282 6.9 ß-caryophyllene 6 0.60 0.661 18 0.52 0.931 21 FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative error of calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. FN: suggested number of factor used, R2: coefficient of regression, RMSEC: root mean square error of calibration, REC: relative erro calibration, Q2: coefficient of determination, RMSEP: root mean square error of prediction, REP: relative error of prediction. coefficients stemming from vectors score PLS of MIR data are to the left of the dotted line and those stemming from vectors score PLS of NIR data is positioned to the right of this dotted line. The percentages given in brack- ets indicate the proportion of variance explained. This regression vector allows observing the weight of PLS vector scores stemming from spectroscopic domains. We can notice that it is the first calculated PLS vector scores which have most importance concerning the block of MIR data. Whereas calculated PLS vector scores with the block of NIR data present low intensities in regression coefficient except for vector scores 2 and 9 for the NIR block (vectors scores 11 and 18 on Figure 4). The vector score 18 explain a small proportion of the va- riance (lower than 1%) but the corresponding regression coefficient has its vector score is brought up. It means that a very low variance influenced strongly the regression. coefficients stemming from vectors score PLS of MIR data are to the left of the dotted line and those stemming from vectors score PLS of NIR data is positioned to the right of this dotted line. The percentages given in brack- ets indicate the proportion of variance explained. 5. Conclusion This study on the determination of main compounds of lavender essential oil samples illustrates capacities and limits of the multiblock methods. Although numerous compounds characterized this essential oil, the range con- 643 tent of these last ones is relatively low. From this little variation in composition, chemometric methods allowed elaborating reliable methods of quantification for the five principal compounds contained in these essential oils: linalool, lavandulol, linalyl acetate, lavandulyl acetate and ß-caryophyllene. The affected accuracy is included between 1% for linalool and 8% for ß-caryophyllene. For this study, MIR data were the most adapted data to the implementation of the regression models. The method of concatenated table gave interesting results but other multiblock methods did not show particular capacity. References [1] Oliveira, F., Brandao, C., Ramalho, H., da Costa, L., Suarez, P. and Rubim, J. (2007) Adulteration of Diesel/Biodiesel Blends by Vegetable Oil as Determined by Fourier Transform (FT) near Infrared Spectrometry and FT-Raman Spec- troscopy. Analytica Chimica Acta, 594, 194-199. http://dx.doi.org/10.1016/j.aca.2007.01.045 [1] Oliveira, F., Brandao, C., Ramalho, H., da Costa, L., Suarez, P. and Rubim, J. (2007) Adulteration of Diesel/Biodiesel Blends by Vegetable Oil as Determined by Fourier Transform (FT) near Infrared Spectrometry and FT-Raman Spec- troscopy. Analytica Chimica Acta, 594, 194-199. http://dx.doi.org/10.1016/j.aca.2007.01.045 [2] Santos, V., Oliveira, F.C.C.., Lima, D.G., Petry, A.C., Garcia, E., Suarez, P.A.Z. and Rubim, J.C. (2005) A Compara- tive Study of Diesel Analysis by FTIR, FTNIR and FT-Raman Spectroscopy Using PLS and Artificial Neural Network Analysis. Analytica Chimica Acta, 547, 188-192. http://dx.doi.org/10.1016/j.aca.2005.05.042 [2] Santos, V., Oliveira, F.C.C.., Lima, D.G., Petry, A.C., Garcia, E., Suarez, P.A.Z. and Rubim, J.C. (2005) A Compara- tive Study of Diesel Analysis by FTIR, FTNIR and FT-Raman Spectroscopy Using PLS and Artificial Neural Network Analysis. Analytica Chimica Acta, 547, 188-192. http://dx.doi.org/10.1016/j.aca.2005.05.042 [3] Cooper, J., Larkin, C., Christopher, M., Schmitigal, J., Morris, R. and Abdelkader, M. (2011) Rapid Analysis of Jet Fuel Using a Handheld Near-Infrared (NIR) Analyzer. Applied Spectroscopy, 65, 187-192. http://dx.doi.org/10.1366/10-06076 [4] Silva, A., Pontes, L., Pimentel, M. and Pontes, M. (2012) Detection of Adulteration in Hydrated Ethyl Alcohol Fuel Using Infrared Spectroscopy and Supervised Pattern Recognition Methods. Talanta, 93, 129-134. http://dx.doi.org/10.1016/j.talanta.2012.01.060 [5] Casale, M., Casolino, C., Oliveri, P. and Forina, M. (2010) Characterisation of Table Olive Cultivar by NIR Spectros- copy. Food Chemistry, 118, 163-168. http://dx.doi.org/10.1016/j.foodchem.2009.04.091 [6] Galtier, O., Dupuy, N., Le Dréau, Y., Ollivier, D., Kister, J. and Artaud, J. (2008) Lipid Compositions and French Registered Designations of Origins of Virgin Olive Oils Predicted by Chemometric Analysis of Mid-Infrared Spectra. Applied Spectroscopy, 62, 583-590. http://dx.doi.org/10.1366/000370208784344479 [7] Bombarda, I., Dupuy, N., Le Van Da, J. and Gaydou, E. (2008) Comparative Chemometric Analyses of Geographic Origins and Compositions of Lavandin var. Grosso Essential Oils by Mid Infrared Spectroscopy and Gas Chromato- graphy. Analytica Chimica Acta, 613, 31-39. http://dx.doi.org/10.1016/j.aca.2008.02.038 [8] Gerlach, R., Kowalski, B. and Wold, S. (1979) Partial Least Squares Path Modeling with Latent Variables. Analytica Chimica Acta, 3, 417-421. http://dx.doi.org/10.1016/S0003-2670(01)85039-X [9] Bastien, P., Esposito Vinzi, V. and Tenenhaus, M. (2005) PLS Generalised Linear Regression. Computational Statis- tics Data Analysis, 48, 17-46. http://dx.doi.org/10.1016/j.csda.2004.02.005 [10] Wangen, L. and Kowalsky, B.A. (1988) Multiblock Partial Least Squares Algorithm for Investigating Complex Chemical Systems. Acknowledgements The authors are grateful to the Office National Interprofessionnel des Plantes à Parfum, Aromatiques et Médici- nales (ONIPPAM) for providing authentic lavander essential oil samples. References Journal of Chemometrics, 3, 3-20. http://dx.doi.org/10.1002/cem.1180030104 [11] Lis-Balchin, M. and Hart, S. (1999) Studies on the Mode of Action of the Essential Oil of Lavender (La tifolia P. Miller). Phytotherapy Research, 13, 540-546. h //d d i /10 1002/(SICI)1099 1573(199909)13 6 540 AID PTR523 3 0 CO 2 I [11] Lis-Balchin, M. and Hart, S. (1999) Studies on the Mode of Action of the Essential Oil of Lavender (Lavandula angus- tifolia P. Miller). Phytotherapy Research, 13, 540-546. http://dx.doi.org/10.1002/(SICI)1099-1573(199909)13:6<540::AID-PTR523>3.0.CO;2-I f ) y py http://dx.doi.org/10.1002/(SICI)1099-1573(199909)13:6<540::AID-PTR523>3.0.CO;2-I [12] McGimpsey, J.A. and Porter, N.G. (1999) A Growers Guide for Commercial Production. New Zealand Institute for Crop and Research Limited, Christchurch. [13] Norme, N.F. (2000) Huiles essentielles—Analyse par chromatographie en phase gazeuse sur colonne capillaire— Méthode générale. AFNOR, Ed., Paris. [14] Wold, S., Esbensen, K. and Geladi, P. (1987) Principal Component Analysis. Chemometrics Intelligent Laboratory System, 2, 37-52. http://dx.doi.org/10.1016/0169-7439(87)80084-9 [15] Fuller, M.P. and Griffiths, P.R. (1978) Diffuse Reflectance Measurements by Infrared Fourier Transform Spectrometry. Analytical Chemistry, 50, 1906-1910. http://dx.doi.org/10.1021/ac50035a045 [16] Martens, H. and Naes, T. (1989) Multivariate Calibration. John Wiley and Sons, Chichester. [17] Liang, Y. and Kvalheim, O. (1996) Heuristic Evolving Latent Projections: Resolving Two-Way Multicomponent Data. Chemometrics and Intelligent Laboratory System, 32, 1-10. [18] Dupuy, N., Duponchel, L., Amram, B., Huvenne, J.P. and Legrand, P. (1994) Quantitative Analysis of Latex in Paper 644 N. Dupuy et al. Coating by ATR-FTIR Spectroscopy. Journal of Chemometrics, 8, 333-347. [19] Karoui, R., Mouazen, A., Ramon, H., Schoonheydt, R. and De Baerdemaeker, J. (2006) Feasibility Study of Discrimi- nating the Manufacturing Process and Sampling Zone in Ripened Soft Cheeses Using Attenuated Total Reflectance MIR and Fiber Optic Diffuse Reflectance VIS-NIR Spectroscopy. Food Research International, 39, 588-597. http://dx.doi.org/10.1016/j.foodres.2005.12.002 [20] Wold, S., Kettaneh, N. and Tjessem, K. (1996) Hierarchical Multiblock PLS and PC Models for Easier Model Inter- pretation and as an Alternative to Variable Selection. Journal of Chemometrics, 10, 463-482. http://dx.doi.org/10.1002/(SICI)1099-128X(199609)10:5/6<463::AID-CEM445>3.0.CO;2-L [21] MacGregor, J. and Kourti, T. (1995) Process Analysis, Monitoring and Diagnosis, Using Multiva Methods. Chemometrics and Intelligent Laboratory System, 28, 3-21. http://dx.doi.org/10.1016/0169-74 [22] Qin, J., Vallaand, S. and Piovoso, M. (2001) On Unifying Multiblock Analysis with Application to Decentralized Process Monitoring. Journal of Chemometrics, 15, 715-742. http://dx.doi.org/10.1002/cem.667 [23] Jaworski, A., Wikiel, K. and Wikiel, H. (2005) Application of Multiblock and Hierarchical PCA and PLS Models for Analysis of AC Voltammetric Data. Electroanalysis, 17, 1477-1485. http://dx.doi.org/10.1002/elan.200503290 [24] Westerhuis, J., Kourti, T. and MacGregor, J. References (1998) Analysis of Multiblock and Hierarchical PCA and PLS Models. Journal of Chemometrics, 12, 301-317. [24] Westerhuis, J., Kourti, T. and MacGregor, J. (1998) Analysis of Multiblock and Hierarchical PCA and PLS Models. Journal of Chemometrics, 12, 301-317. http://dx.doi.org/10.1002/(SICI)1099-128X(199809/10)12:5<301::AID-CEM515>3.0.CO;2-S [25] Westerhuis, J. and Coenegracht, P. (1997) Multivariate Modelling of the Pharmaceutical Two-Step Process of Wet Granulation and Tableting with Multiblock PLS. Journal of Chemometrics, 11, 379-382. http://dx.doi.org/10.1002/(SICI)1099-128X(199709/10)11:5<379::AID-CEM482>3.0.CO;2-8 [26] Savitzky, A. and Golay, M. (1964) Smoothing and Differentiation of Data by Simplified Least Squares Procedures. Analytical Chemistry, 36, 1627-1639. http://dx.doi.org/10.1021/ac60214a047 [27] Zhang, M., Sheng, G., Mu, Y., Li, Y., Yu, H., Harada, H. and Li, Y. (2009) Rapid and Accurate Determination of VFAs and Ethanol in the Effluent of an Anaerobic H2-Producing Bioreactor Using Near-Infrared Spectroscopy. Water Research, 43, 1823-1830. http://dx.doi.org/10.1016/j.watres.2009.01.018 Scientific Research Publishing (SCIRP) is one of the largest Open Access journal publishers. It is currently publishing more than 200 open access, online, peer-reviewed journals covering a wide range of academic disciplines. SCIRP serves the worldwide academic communities and contributes to the progress and application of science with its publication. Other selected journals from SCIRP are listed as below. Submit your manuscript to us via either submit@scirp.org or Online Submission Portal. Other selected journals from SCIRP are listed as below. Submit your manuscript to us via eith submit@scirp.org or Online Submission Portal. Other selected journals from SCIRP are listed as below. Submit your manuscript to us via either submit@scirp.org or Online Submission Portal.
https://openalex.org/W2610160269	https://openpsychologyjournal.com/VOLUME/10/PAGE/55/PDF/	English	null	The More I Can Choose, The More I Am Disappointed: The “Illusion of Control” in Children’s Decision-Making	The Open psychology journal	2,017	cc-by	4,246	Conclusion: Children have experienced a sense of agency for their choice, thus leading to an illusion of control for the decision process and to the so-called “wishful thinking”. Keywords: Sense of agency, Illusion of control, Disappointment, Decision making, Child development. Method: 107 children (age range 7-10 years) rated their emotions before and after discovering the outcome, in two experimental conditions: choice condition, where the child could decide which of the two remaining tickets to choose in order to win some candies, and no choice condition, where the child could not decide as only one ticket was left. Background: Decision making is a complex psychological process driven by emotions. Among the most unpleasant ones are the situations when the obtained outcome is not the one expected. This emotional experience is influenced by sense of agency, i.e. the feeling that we voluntarily control our actions and, through them, events in the world. Negative counterfactual emotions as disappointment have been marginally analyzed in children’s decision-making, and the study of children’s sense of agency could help to understand them. Obj ti Objective: uring childhood the valence of disappointment in decision making in relation to the possibility of choosing or not. To evaluate during childhood the valence of disappointment in decision making in relation to the possibility of choo Results: The self-attribution of a positive emotional state was significantly higher in the choice condition than in the no choice condition, so the possibility to pick up the ticket made children happier in general, by promoting an “illusion of control”, which is absent in the no choice condition. Then, after discovering the bad outcome, the emotions collapse, settling at substantially similar values. The More I Can Choose, The More I Am Disappointed: The “Illusion of Control” in Children’s Decision-Making 1Dipartimento di Scienze Umane e Sociali, Università degli Studi di Bergamo, Bergamo, Italy 2Unità di Ricerca sulla Teoria della Mente, Dipartimento di Psicologia, Università Cattolica del Sacro Cuore, Milano, It l 3Donders Institute for Brain, Mind and Behavior, Radboud University, Nijmegen, The Netherlands 3Donders Institute for Brain, Mind and Behavior, Radboud University, Nijmegen, The Netherlands Revised: February 01, 2017 Accepted: March 06, 2017 Received: August 11, 2016 Send Orders for Reprints to reprints@benthamscience.ae Send Orders for Reprints to reprints@benthamscience.ae The Open Psychology Journal, 2017, 10, 55-60 55 1. INTRODUCTION What makes the distinction between the two emotions possible is the differential agency attribution, i.e. personal agency for regret and external agency for disappointment [6]. Sense of agency is a key psychological component of human behavior, as it refers to the feeling that we voluntarily control our actions and, through them, events in the world [7]. So, sense of agency is relevant for decision making, as it allows us to establish a link between our intentions and our actions, and between our actions and their final outcomes. This second link, between actions and outcomes, constitutes an important signal that informs the sense of agency, because if outcomes match our expectations, our sense of agency is reinforced (the feeling that “I did that”), whereas a mismatch impoverishes our sense of agency [8, 9]. However, people often tend to overestimate their sense of agency, thus falling in the bias of the so-called “illusion of control” [10], i.e. the tendency to believe that they can control or at least influence outcomes that they undoubtedly have no influence over. This unrealistic perceived control of events that cannot be under our control is evident in decision making, for example when people prefer to pick their lottery numbers than have them randomly allocated, and are willing to even pay for such an opportunity, or when rolling dice in craps, as they tend to throw harder for high numbers and softer for low numbers. This illusion is so pervasive that it lays at the heart of superstitious thinking also in the adult world. As regards the development of the sense of agency, researches have shown that young children present a reduced awareness about the possibilities to determine outcomes through actions, and that this altered sense of agency drives them easily towards the illusion of control [11, 12]. The valence of the outcome plays a role as well, as young children tend to consider an outcome as intended when it satisfies their desire [13], and it has been also shown that the valence of the outcome drives children’s judgements and decisions in bargaining games - see for example [14 - 16]. So, as regards counterfactual emotions such as regret and disappointment, it may be that children do not truly experience them, but rather a negative feeling related to the frustration of not obtaining a desired outcome. 1. INTRODUCTION Decision making is a complex psychological process, involving both deliberative and emotional components. For a long time, decision making has been conceived as a primarily cognitive process, requiring to estimate which among * Address correspondence to this author at the Università degli Studi di Bergamo, Dipartimento di Scienze Umane e Sociali Address: Piazzale S. Agostino, 2 - 24129 Bergamo, Italy, Tel: 00390352052934; E-mail: ilaria.castelli@unibg.it 1874-3501/17 2017 Bentham Open 56 The Open Psychology Journal, 2017, Volume 10 56 Castelli et al. various options would have led to the best outcome, i.e. the maximization of one’s own profit. This conception has been supported by another one, deeply rooted in our western culture, i.e. the negatively interfering role of emotions (or “passions”) on human behavior. Such a conception has been strongly revised over the past decades, that have showed an increase of researches about the role of emotions in decision-making [1, 2]. various options would have led to the best outcome, i.e. the maximization of one’s own profit. This conception has been supported by another one, deeply rooted in our western culture, i.e. the negatively interfering role of emotions (or “passions”) on human behavior. Such a conception has been strongly revised over the past decades, that have showed an increase of researches about the role of emotions in decision-making [1, 2]. It is possible to distinguish between two main ways in which emotions are involved in decision-making. The first one regards expected emotions, i.e. the predictions about the emotional consequences of a certain outcome after the decision. The second one regards immediate emotions, i.e. the feelings that people experience during the decision- making process. Nowadays, in the psychological scientific literature there is a remarkable agreement on the role of emotions as powerful drivers of human decision-making across all its phases, i.e. before making the decision, during the decisional process and at the end, once the results of the decision are discovered. Among the most unpleasant emotional experiences in decision making we can find the situations when the obtained outcome is not the one we expected. In this case, we experience emotions such as regret and disappointment that typically occur when we compare the actual outcome with other possible outcome/s that could have occurred had we only decided differently [3]. Regret and disappointment are close to each other, because their hedonic valence is the same (i.e. negative), but with a different intensity [4, 5]. 2.2. Procedure A simple decision making task was created for the purposes of the present study. The scenario was derived from [21], adjusted for children. The experimenter explained that the children would play an amusement-park lottery, where he/she can win some candies (good outcome = 10 candies; bad outcome = 1 candy). The task included two conditions. In the choice condition, the child could decide which of two remaining tickets to choose in order to win some candies. In the no choice condition, the child could not decide which ticket to choose as another child had decided first, and so they would receive the remaining ticket. In both contitions the outcome was always bad (one candy), and the child was asked to rate his/her emotions on a thermometer ranging from 1 (negative/bad feelings) to 7 (positive/good feelings) with smiles depicted on it. Adequate understanding of the emotion thermometer was assessed before the administration of the task. The child was asked to rate his/her emotions across the entire decision process, i.e. at three levels: The child was asked to rate his/her emotions across the entire decision process, i.e. at three levels: level 1: before making the decision (in the choice condition) and before being given the remaining ticket (in the no choice condition). level 1: before making the decision (in the choice condition) and before being given the remaining ticket (in the no choice condition). level 2: when having the ticket in his/her hands, but before discovering the outcome. level 3: after discovering the outcome. Tasks were submitted individually in one session in a quiet room at school. Each child played the two tasks, i.e. the choice task and the no choice task as a within subjects design. The order of the two task conditions was counterbalanced across subjects. 1. INTRODUCTION A crucial point here is the understanding of agency as a necessary condition to experience such type of emotions: [17] showed that children that experience regret made a different decision the day after in the same task, but this remained true independent of who was responsible for the choice, i.e. the child, a dice, a dice launched by another person. However, if children themselves decided, i.e. they did not use a dice nor let another person use it, the intensity of the emotion was higher. This result is interesting, because it puts the light again on the role of the sense of agency in the emotions involved in decision making. As regards disappointment, to the best of our knowledge, so far only one study has explored it: [18] showed that around seven/eight years of age children understand the different situational antecedents of regret and disappointment, but they are still not able to adopt adequate strategies in order to manage anticipatory regret and anticipatory disappointment until nine/ten years. We think that the study of children’s sense of agency and of the possible presence of the bias of the “illusion of control” could contribute to the understanding of counterfactual emotions in decision making, especially of disappointment, that has been marginally studied in developmental decision making so far. The emotion of disappointment, in fact, should not be influenced by the understanding of one's own responsibility as in the case of regret, so it should simply depend on the bad outcome of the decision irrespective of whether the subject had the opportunity to choose or not. So, the aim of this study is to evaluate the valence of disappointment in relation to the The Open Psychology Journal, 2017, Volume 10 57 The Open Psychology Journal, 2017, Volume 10 57 Disappointment in Children’s Decision-Making possibility of choosing or not, and the different ratings of this emotion at various moments of the decision-making process, i.e. before making a choice, after making the choice but not knowing the outcome, and after knowing the (bad) outcome. possibility of choosing or not, and the different ratings of this emotion at various moments of the decision-making process, i.e. before making a choice, after making the choice but not knowing the outcome, and after knowing the (bad) outcome. 2.1. Participants The participants were 107 children attending the 2 nd, 3 rd, 4 th, 5 th years of a primary school in the North of Italy. They were neither referred to social services, nor reported for learning and socio-relational difficulties. Parents’ informed written consent was obtained for each participant, and directors of the schools granted permission to carry on the research. The research was conducted according to APA ethical standards, and was approved by the local ethics committee. The study was conformed to the ethical principles of the 1964 Helsinki Declaration. In order to create a homogeneous group under the cognitive and the linguistic profile, the Coloured Progressive Matrices – CPM [19] and the Peabody Picture Vocabulary Test – PPVT [20] were respectively used. On the basis of this assessment, nine children (3 strangers and 6 Italians) were excluded from the sample since their performance to the CPM resulted at the 25 th percentile. Two age-groups were created: young (2 nd and 3 rd classes, N = 54, 19 males, 35 females) with a mean age of 7 years and 7 months (sd = 6.95 months), and old (4 th and 5 th classes, N = 44, 21 males, 23 females) with a mean age of 9 years and 9 months (sd = 8.23 months). 3. RESULTS Since preliminary analyses showed no statistically significant difference between the level 1 and level 2 in both experimental conditions (p >.05), an average score between the self-attribution of positive emotional state at level 1 and level 2 was computed, both for the choice condition and for that of no choice condition. From a descriptive overview (Table 1), the means of the emotions before discovering the bad outcome in the choice condition are higher than the means of the emotions before discovering the bad outcome in the no choice condition in both age groups. Paired samples T-tests confirm this trend: the self-attribution of a positive emotional state is significantly higher in the choice condition than in the no choice condition (young: paired samples t=7.26, df =48, p <.001; old: t=6.93, df=48, p <.001). This may suggest that the possibility to pick up the ticket could make children happier in general, by promoting an “illusion of control”, which is absent in the no choice condition. Then, after discovering the bad outcome, the emotions 58 The Open Psychology Journal, 2017, Volume 10 Castelli et al. 58 collapse, settling at substantially similar values, independently of the experimental condition for both age groups. Table 1. Descriptive statistics of the explored variables. N Min Max Mean Stand. Dev. CC 1- emotion before outcome 98 3,00 7,00 5,99 ,84 NCC 2 - emotion before outcome 98 2,50 7,00 4,92 ,98 CC 1 - emotion after outcome 98 1 7 2,19 1,21 NCC 2 - emotion after outcome 98 1 7 2,09 1,23 1 Choice Condition2 No Choice Condition collapse, settling at substantially similar values, independently of the experimental condition for both age groups. Table 1. Descriptive statistics of the explored variables. N Min Max Mean Stand. Dev. CC 1- emotion before outcome 98 3,00 7,00 5,99 ,84 NCC 2 - emotion before outcome 98 2,50 7,00 4,92 ,98 CC 1 - emotion after outcome 98 1 7 2,19 1,21 NCC 2 - emotion after outcome 98 1 7 2,09 1,23 1 Choice Condition2 No Choice Condition In order to better understand if children experience different intensity of emotions – namely disappointment – between the choice and the no choice conditions, we calculated the relative difference between the self-attribution of a positive emotional state before and after discovering the bad outcome for the choice condition and for the no choice condition separately. DISCUSSION AND CONCLUSION The understanding of complex emotions involved in decision-making is a challenging achievement during childhood. With regard to the understanding of negative emotions such as disappointment, our results show that children between seven/nine years of age experience different intensity of this emotion based to the possibility of choosing or not. Our findings are in line with those reported by [18] about disappointment, but we cannot claim anything about the way children use this emotion, as we did not directly evaluate the anticipation of such emotion and the use of such anticipation on subsequent decision-making as did [18]. Instead, we tried to examine the intensity of the emotional experience across the decision-making process, i.e. immediate emotions or the feelings that people experience during the decision-making process, following the distinction proposed by [1]. Monitoring children’s ratings of emotions at various points of the decisional process has offered a relevant result, which can be explained by referring to the function of agency, analogous to what happens in adults [6]. Notably, the difference in the ratings of the emotions before and after knowing the outcome is higher in the choice condition than in the no choice condition, and this can be likely explained by referring to the antecedent condition of the decision itself. Children have presumably experienced a sense of agency for their choice, thus leading to an illusion of control for the decision process and to so-called “wishful thinking”. This likely raised the level of the emotions until the moment of the discovery of the bad outcome: the higher the emotions before knowing it, the higher the difference between “before” and “after” the decision. However, we did not find an age effect, as found by [12], and this could be probably attributed to the different research paradigms used in the two studies. The illusion of control seems to be quite pervasive of human behavior and hard to be eliminated also in adults, as demonstrated by [22]: their participants were exposed to undesired outcomes that occurred independently of their behavior, and they were warned that the outcomes might have alternative causes, other than the participants’ actions, a strategy that has been shown to reduce positive illusions. However, when participants received this information in an experiment in which the outcomes were undesired, their illusion was enhanced rather than reduced. This study has also some limitations. 3. RESULTS These two differences should be the measure of the intensity of the effect that the bad outcome has on the initial emotion in the two different conditions, i.e. in the choice condition and in the no choice condition. We ran a General Linear Model for repeated measures comparing the indexes of the intensity of emotions, i.e. the choice difference and the no choice difference, as dependent variables, and age and gender as between-subject factors. We found a significant main effect of the experimental condition (F(1, 94) = 88.88, p = .001, ηρ 2 = .486, θ = 1), showing that the initial condition, i.e. having or not having the possibility to choose, drives the intensity of the effect that the bad outcome has on the initial emotion. In the choice condition, in fact, the initial ratings of emotions are higher than in the no choice condition, so the final difference between the ratings of emotions at two points of the decisional process is higher than the final difference in the no choice condition. This evidence seems to support the idea that children fall into the bias of the “illusion of control”. In fact, the sense of agency may have played a role in this experience: the possibility to choose the ticket seems to sustain the emotional rates at the beginning of the decision making process, thus making children happier compared to the situation with no possibility to choose. INFORMED CONSENT Parents’ informed written consent was obtained for each participant before starting the research. Parents’ informed written consent was obtained for each participant before starting the research. The directors of the schools granted permission to carry out the research. The research was conducted according to APA ethical standards, and was approved by the local ethics committee. The study was conformed to the ethical principles of the 1964 Helsinki Declaration. FUNDING This study received no funding. ETHICAL APPROVAL All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. CONFLICT OF INTEREST The author confirms that this article content has no conflict of interest. The author confirms that this article content has no conflict of interest. ACKNOWLEDGEMENTS We would like to thank Laura Sanvito for helping in collecting data. A special thanks to the children, parents and teachers who made this research possible. DISCUSSION AND CONCLUSION In order to be sure that all children – especially younger ones – understood the request to rate the intensity of their emotion at various moments of the decisional process, we used the thermometer of emotions, without asking to label them. So, we found that the participants completed the emotional state scale The Open Psychology Journal, 2017, Volume 10 59 Disappointment in Children’s Decision-Making 59 differently when they had the possibility to choose their ticket vs. when they had not, thus showing that the possibility to choose influences the subsequent emotional experience. However, we do not have the report about the precise labelled emotion (disappointment), that could also have been mixed with other ones (such as anger, or sadness). Future research could consider to use a post-manipulation measure that potentially captures, via self report, what the participants are exactly feeling, in order to better understand the development of decision-making abilities and of the emotions involved in these complex processes. REFERENCES [1] Loewenstein G, Lerner JS. Davidson RJ, Scherer KR, Goldsmith HH. The role of affect in decision making. In: Handbook of Affective Science. Oxford University Press 2003; pp. 619-42. [1] Loewenstein G, Lerner JS. Davidson RJ, Scherer KR, Goldsmith HH. The role of affect in decision making. In: Handbook of Affective Science. Oxford University Press 2003; pp. 619-42. [2] Lerner JS, Li Y, Valdesolo P, Kassam KS. Emotion and decision making. Annu Rev Psychol 2015; 66: 799-823. [http://dx.doi.org/10.1146/annurev-psych-010213-115043] [PMID: 25251484] [2] Lerner JS, Li Y, Valdesolo P, Kassam KS. Emotion and decision making. Annu Rev Psychol 2015; 66: 799-823. [http://dx.doi.org/10.1146/annurev-psych-010213-115043] [PMID: 25251484] [3] Coricelli G, Dolan RJ, Sirigu A. Brain, emotion and decision making: the paradigmatic example of regret. Trends Cogn Sci (Regul Ed) 2007; 11(6): 258-65. [http://dx.doi.org/10.1016/j.tics.2007.04.003] [PMID: 17475537] [3] Coricelli G, Dolan RJ, Sirigu A. Brain, emotion and decision making: the paradigmatic example of regret. Trends Cogn Sci (Regul Ed) 2007; 11(6): 258-65. [http://dx.doi.org/10.1016/j.tics.2007.04.003] [PMID: 17475537] [4] Zeelenberg M, Van Dijk WW, Manstead AS, et al. On bad decisions and disconfirmed expectancies: The psychology of regret and disappointment. Cogn Emotion 2000; 14: 521-41. [http://dx.doi.org/10.1080/026999300402781] [4] Zeelenberg M, Van Dijk WW, Manstead AS, et al. On bad decisions and disconfirmed expectancies: The psychology of regret and disappointment. Cogn Emotion 2000; 14: 521-41. [http://dx.doi.org/10.1080/026999300402781] [5] Van Dijk E, Zeelenberg M. Investigating the Appraisal Patterns of Regret and Disappointment. Motiv Emot 2002; 26: 321-31. [http://dx.doi.org/10.1023/A:1022823221146] [5] Van Dijk E, Zeelenberg M. Investigating the Appraisal Patterns of Regret and Disappointment. Motiv Emot 2002; 26: 321-31. [http://dx.doi.org/10.1023/A:1022823221146] [6] Giorgetta C, Grecucci A, Bonini N, et al. Waves of regret: a meg study of emotion and decision-making. Neuropsychologia 2013; 51(1): 38-51. [6] Giorgetta C, Grecucci A, Bonini N, et al. Waves of regret: a meg study of emotion and decision-making. Neuropsychologia 2013; 51(1): 38-51. [http://dx.doi.org/10.1016/j.neuropsychologia.2012.10.015] [PMID: 23137945] [7] Haggard P, Chambon V. Sense of agency. Curr Biol 2012; 22(10): R390-2. [http://dx.doi.org/10.1016/j.cub.2012.02.040] [PMID: 22625851] [7] Haggard P, Chambon V. Sense of agency. Curr Biol 2012; 22(10): R390-2. [http://dx.doi.org/10.1016/j.cub.2012.02.040] [PMID: 22625851] [8] Blakemore SJ, Wolpert DM, Frith CD. Abnormalities in the awareness of action. Trends Cogn Sci (Regul Ed) 2002; 6(6): 237-42. [http://dx.doi.org/10.1016/S1364-6613(02)01907-1] [PMID: 12039604] [9] Wegner DM, Wheatley T. Apparent mental causation. Sources of the experience of will. Am Psychol 1999; 54(7): 480-92. [http://dx.doi.org/10.1037/0003-066X.54.7.480] [PMID: 10424155] [10] Langer EJ. The illusion of control. J Pers Soc Psychol 1975; 32: 311-28. REFERENCES [http://dx.doi.org/10.1037/0022-3514.32.2.311] [11] Weisz JR, Yeates KO, Robertson D, et al. Perceived contingency of skill and chance events – A developmental analysis. Dev Psychol 1982; 60 The Open Psychology Journal, 2017, Volume 10 Castelli et al. 18: 898-905. [http://dx.doi.org/10.1037/0012-1649.18.6.898] 18: 898-905. [http://dx.doi.org/10.1037/0012-1649.18.6.898] [12] van Elk M, Rutjens BT, van der Pligt J. The development of the illusion of control and sense of agency in 7- to-12-year old children and adults. Cognition 2015; 145: 1-12. [http://dx.doi.org/10.1016/j.cognition.2015.08.004] [PMID: 26298422] [13] Astington JW. The paradox of intention: Assessing children’s metarepresentational understanding. In: Malle BF, Moses LJ, Baldwin DA, Eds. Intentions and intentionality: Foundations of social cognition. MIT Press 2001; pp. 85-103. [14] Castelli I, Massaro D, Sanfey AG, et al. Fairness and intentionality in children’s decision-making. Int Rev Econ 2010; 57: 269-88. [http://dx.doi.org/10.1007/s12232-010-0101-x] [15] Castelli I, Massaro D, Sanfey AG, et al. “What is fair for you?” judgments and decisions about fairness and theory of mind. Eur J Dev Psychol 2013; 11: 49-62. [http://dx.doi.org/10.1080/17405629.2013.806264] [16] Castelli I, Massaro D, Bicchieri C, Chavez A, Marchetti A. Fairness norms and theory of mind in an ultimatum game: judgments, offers, and decisions in school-aged children. PLoS One 2014; 9(8): e105024. [http://dx.doi.org/10.1371/journal.pone.0105024] [PMID: 25118863] [17] OConnor E, McCormack T, Beck SR, Feeney A. Regret and adaptive decision making in young children. J Exp Child Psychol 2015; 135: 86-92. [17] OConnor E, McCormack T, Beck SR, Feeney A. Regret and adaptive decision making in young children. J Exp Child Psychol 2015; 135: 86-92. [http://dx.doi.org/10.1016/j.jecp.2015.03.003] [PMID: 25843700] 86-92. [http://dx.doi.org/10.1016/j.jecp.2015.03.003] [PMID: 25843700] [http://dx.doi.org/10.1016/j.jecp.2015.03.003] [PMID: 25843700] [18] Guttentag R, Ferrell J. Children’s understanding of anticipatory regret and disappointment. Cogn Emotion 2008; 22: 815-32. [http://dx.doi.org/10.1080/02699930701541542] rogressive matrices: Séries A, Ab, B Clamart Seine: Editions scientiﬁques et psychotechniques 1947. [19] Raven JC. Progressive matrices: Séries A, Ab, B Clamart Seine: Editions scientiﬁques et psychotechniques 1947. [20] Dunn L, Dunn L. Peabody Picture Vocabulary Test-Revised Manual. Pines, MN: American Guidance Service 1981. [21] Van Dijk E, Zeelenberg M. On the psychology of ‘if only’: Regret and the comparison between factual and counterfactual outcomes. Organ Behav Hum 2005; 97: 152-60. [http://dx.doi.org/10.1016/j.obhdp.2005.04.001] [22] Matute H, Blanco F. Reducing the illusion of control when an action is followed by an undesired outcome. Psychon Bull Rev 2014; 21(4): 1087-93. [http://dx.doi.org/10.3758/s13423-014-0584-7] [PMID: 24448764] © 2017 Castelli et al. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: (https://creativecommons.org/licenses/by/4.0/legalcode). This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
https://openalex.org/W4361897180	https://figshare.com/ndownloader/files/39859095	English	null	Supplementary Tables 1 and 2 and Supplementary Figures 1 through 6 from Alternative Polyadenylation in Triple-Negative Breast Tumors Allows NRAS and c-JUN to Bypass PUMILIO Posttranscriptional Regulation	null	2,023	cc-by	6,817	Shortened 3'UTRs APA site gene ID lost PUM sites log fold changeP-value diff. expr gene symbol p.1301.6 1301 1 3.34604 1.20E-05 COL11A1 p.81611.1 81611 1 1.84336 0.000678749 ANP32E p.10097.1 10097 1 1.76786 3.37E-05 ACTR2 p.51026.1 51026 1 1.74004 9.13E-06 GOLT1B p.51659.1 51659 1 1.59379 6.14E-05 GINS2 p.79572.1 79572 1 1.59155 0.00035399 ATP13A3 p.5062.1 5062 2 1.49865 8.96E-05 PAK2 p.4893.2 4893 1 1.48177 0.002016624 NRAS p.8507.1 8507 1 1.37817 0.007869441 ENC1 p.10276.2 10276 2 1.37096 0.00099952 NET1 p.9877.1 9877 2 1.33103 4.95E-06 ZC3H11A p.80155.1 80155 1 1.31275 4.48E-07 NAA15 p.55706.1 55706 1 1.31015 0.000572818 TMEM48 p.92140.2 92140 3 1.30391 4.98E-06 MTDH p.25932.1 25932 2 1.30381 0.00644907 CLIC4 p.528.3 528 1 1.29909 0.000144482 ATP6V1C1 p.84955.2 84955 2 1.27068 4.94E-05 NUDCD1 p.8543.1 8543 1 1.26135 0.019855937 LMO4 p.55284.1 55284 1 1.24239 6.14E-07 UBE2W p.117178.3 117178 1 1.23431 9.53E-05 SSX2IP p.51018.4 51018 1 1.22593 2.35E-06 RRP15 p.57532.6 57532 3 1.22504 1.72E-05 NUFIP2 p.112858.2 112858 1 1.18547 7.93E-05 TP53RK p.8669.2 8669 1 1.1763 5.05E-06 EIF3J p.10890.4 10890 1 1.09651 9.14E-08 RAB10 p.3267.1 3267 1 1.09645 4.22E-06 AGFG1 p.6745.2 6745 1 1.0902 0.000747828 SSR1 p.79677.2 79677 2 1.08962 3.66E-05 SMC6 p.23527.2 23527 2 1.07493 0.000213667 ACAP2 p.5203.1 5203 1 1.07094 0.00070456 PFDN4 p.79888.2 79888 1 1.05371 0.001510546 LPCAT1 p.26046.2 26046 1 1.05281 0.000231483 LTN1 p.153339.1 153339 1 1.04776 0.000311249 TMEM167A p.55110.1 55110 2 1.03488 0.001739127 MAGOHB p.5469.1 5469 2 1.02688 0.050409219 MED1 p.54800.1 54800 2 1.0235 0.002063174 KLHL24 p.11320.3 11320 4 1.00629 0.028935489 MGAT4A p.54149.1 54149 1 0.997606 0.006795977 C21orf91 p.55971.1 55971 1 0.989322 0.001160397 BAIAP2L1 p.22797.2 22797 1 0.986222 0.007900227 TFEC p.7994.2 7994 1 0.966617 0.02085234 MYST3 p.10059.2 10059 1 0.957199 0.000454561 DNM1L p.25917.2 25917 1 0.93798 0.000124526 THUMPD3 p.2786.2 2786 1 0.932943 0.03126166 GNG4 Shortened 3'UTRs APA site gene ID lost PUM sites log fold changeP-value diff. Shortened 3'UTRs APA site gene ID lost PUM sites log fold changeP-value diff. expr gene symbol p.1301.6 1301 1 3.34604 1.20E-05 COL11A1 p.81611.1 81611 1 1.84336 0.000678749 ANP32E p.10097.1 10097 1 1.76786 3.37E-05 ACTR2 p.51026.1 51026 1 1.74004 9.13E-06 GOLT1B p.51659.1 51659 1 1.59379 6.14E-05 GINS2 p.79572.1 79572 1 1.59155 0.00035399 ATP13A3 p.5062.1 5062 2 1.49865 8.96E-05 PAK2 p.4893.2 4893 1 1.48177 0.002016624 NRAS p.8507.1 8507 1 1.37817 0.007869441 ENC1 p.10276.2 10276 2 1.37096 0.00099952 NET1 p.9877.1 9877 2 1.33103 4.95E-06 ZC3H11A p.80155.1 80155 1 1.31275 4.48E-07 NAA15 p.55706.1 55706 1 1.31015 0.000572818 TMEM48 p.92140.2 92140 3 1.30391 4.98E-06 MTDH p.25932.1 25932 2 1.30381 0.00644907 CLIC4 p.528.3 528 1 1.29909 0.000144482 ATP6V1C1 p.84955.2 84955 2 1.27068 4.94E-05 NUDCD1 p.8543.1 8543 1 1.26135 0.019855937 LMO4 p.55284.1 55284 1 1.24239 6.14E-07 UBE2W p.117178.3 117178 1 1.23431 9.53E-05 SSX2IP p.51018.4 51018 1 1.22593 2.35E-06 RRP15 p.57532.6 57532 3 1.22504 1.72E-05 NUFIP2 p.112858.2 112858 1 1.18547 7.93E-05 TP53RK p.8669.2 8669 1 1.1763 5.05E-06 EIF3J p.10890.4 10890 1 1.09651 9.14E-08 RAB10 p.3267.1 3267 1 1.09645 4.22E-06 AGFG1 p.6745.2 6745 1 1.0902 0.000747828 SSR1 p.79677.2 79677 2 1.08962 3.66E-05 SMC6 p.23527.2 23527 2 1.07493 0.000213667 ACAP2 p.5203.1 5203 1 1.07094 0.00070456 PFDN4 p.79888.2 79888 1 1.05371 0.001510546 LPCAT1 p.26046.2 26046 1 1.05281 0.000231483 LTN1 p.153339.1 153339 1 1.04776 0.000311249 TMEM167A p.55110.1 55110 2 1.03488 0.001739127 MAGOHB p.5469.1 5469 2 1.02688 0.050409219 MED1 p.54800.1 54800 2 1.0235 0.002063174 KLHL24 p.11320.3 11320 4 1.00629 0.028935489 MGAT4A p.54149.1 54149 1 0.997606 0.006795977 C21orf91 p.55971.1 55971 1 0.989322 0.001160397 BAIAP2L1 p.22797.2 22797 1 0.986222 0.007900227 TFEC p.7994.2 7994 1 0.966617 0.02085234 MYST3 p.10059.2 10059 1 0.957199 0.000454561 DNM1L p.25917.2 25917 1 0.93798 0.000124526 THUMPD3 p.2786.2 2786 1 0.932943 0.03126166 GNG4 expr gene symbol p.1301.6 1301 1 3.34604 1.20E-05 COL11A1 p.81611.1 81611 1 1.84336 0.000678749 ANP32E p.10097.1 10097 1 1.76786 3.37E-05 ACTR2 p.51026.1 51026 1 1.74004 9.13E-06 GOLT1B p.51659.1 51659 1 1.59379 6.14E-05 GINS2 p.79572.1 79572 1 1.59155 0.00035399 ATP13A3 p.5062.1 5062 2 1.49865 8.96E-05 PAK2 p.4893.2 4893 1 1.48177 0.002016624 NRAS p.8507.1 8507 1 1.37817 0.007869441 ENC1 p.10276.2 10276 2 1.37096 0.00099952 NET1 p.9877.1 9877 2 1.33103 4.95E-06 ZC3H11A p.80155.1 80155 1 1.31275 4.48E-07 NAA15 p.55706.1 55706 1 1.31015 0.000572818 TMEM48 p.92140.2 92140 3 1.30391 4.98E-06 MTDH p.25932.1 25932 2 1.30381 0.00644907 CLIC4 p.528.3 528 1 1.29909 0.000144482 ATP6V1C1 p.84955.2 84955 2 1.27068 4.94E-05 NUDCD1 p.8543.1 8543 1 1.26135 0.019855937 LMO4 p.55284.1 55284 1 1.24239 6.14E-07 UBE2W p.117178.3 117178 1 1.23431 9.53E-05 SSX2IP p.51018.4 51018 1 1.22593 2.35E-06 RRP15 p.57532.6 57532 3 1.22504 1.72E-05 NUFIP2 p.112858.2 112858 1 1.18547 7.93E-05 TP53RK p.8669.2 8669 1 1.1763 5.05E-06 EIF3J p.10890.4 10890 1 1.09651 9.14E-08 RAB10 p.3267.1 3267 1 1.09645 4.22E-06 AGFG1 p.6745.2 6745 1 1.0902 0.000747828 SSR1 p.79677.2 79677 2 1.08962 3.66E-05 SMC6 p.23527.2 23527 2 1.07493 0.000213667 ACAP2 p.5203.1 5203 1 1.07094 0.00070456 PFDN4 p.79888.2 79888 1 1.05371 0.001510546 LPCAT1 p.26046.2 26046 1 1.05281 0.000231483 LTN1 p.153339.1 153339 1 1.04776 0.000311249 TMEM167A p.55110.1 55110 2 1.03488 0.001739127 MAGOHB p.5469.1 5469 2 1.02688 0.050409219 MED1 p.54800.1 54800 2 1.0235 0.002063174 KLHL24 p.11320.3 11320 4 1.00629 0.028935489 MGAT4A p.54149.1 54149 1 0.997606 0.006795977 C21orf91 p.55971.1 55971 1 0.989322 0.001160397 BAIAP2L1 p.22797.2 22797 1 0.986222 0.007900227 TFEC p.7994.2 7994 1 0.966617 0.02085234 MYST3 p.10059.2 10059 1 0.957199 0.000454561 DNM1L p.25917.2 25917 1 0.93798 0.000124526 THUMPD3 p.2786.2 2786 1 0.932943 0.03126166 GNG4 p.133619.2 133619 2 0.932475 6.05E-05 PRRC1 p.56852.2 56852 2 0.928459 0.0137244 RAD18 p.6731.7 6731 1 0.917953 2.07E-05 SRP72 p.57720.3 57720 1 0.911116 5.37E-05 GPR107 p.8634.2 8634 1 0.891051 0.001914598 RTCD1 p.25836.1 25836 4 0.876987 0.005399843 NIPBL p.135228.3 135228 2 0.875933 0.001249262 CD109 p.157680.4 157680 2 0.872486 0.005159905 VPS13B p.9097.2 9097 1 0.871557 0.0013773 USP14 p.2885.2 2885 2 0.865555 0.000138914 GRB2 p.8030.1 8030 1 0.863186 0.011890665 CCDC6 p.125228.3 125228 1 0.858339 0.011574755 C18orf19 p.7175.1 7175 1 0.857776 0.000151936 TPR p.57089.1 57089 1 0.848759 1.12E-06 ENTPD7 p.90806.2 90806 1 0.837183 0.001403134 ANGEL2 p.55207.1 55207 1 0.820868 0.022784284 ARL8B p.10203.1 10203 2 0.807331 0.003429625 CALCRL p.55153.2 55153 3 0.80623 0.002703631 SDAD1 p.7227.1 7227 3 0.80539 0.049064692 TRPS1 p.283209.1 283209 1 0.802976 0.013351419 PGM2L1 p.8065.2 8065 1 0.802513 0.000151257 CUL5 p.79828.1 79828 1 0.797501 0.000505914 METTL8 p.56943.1 56943 1 0.797174 0.000932962 ENY2 p.84901.2 84901 1 0.793092 2.86E-06 NFATC2IP p.255520.1 255520 1 0.792406 0.006672674 ELMOD2 p.5087.1 5087 1 0.790327 0.081207269 PBX1 p.84259.1 84259 1 0.788728 0.00425738 DCUN1D5 p.286148.5 286148 1 0.779111 0.003963532 DPY19L4 p.134430.1 134430 4 0.777661 0.000135153 WDR36 p.1737.3 1737 1 0.776573 0.004454858 DLAT p.56474.1 56474 1 0.765375 0.003952036 CTPS2 p.6198.1 6198 1 0.765057 0.020871033 RPS6KB1 p.116150.1 116150 1 0.750093 0.0322726 NUS1 p.5048.2 5048 3 0.746829 0.002203512 PAFAH1B1 p.9969.5 9969 1 0.742792 0.011596419 MED13 p.56888.2 56888 2 0.741086 0.000265759 KCMF1 p.9126.2 9126 1 0.738587 0.002221341 SMC3 p.51122.3 51122 1 0.732167 0.026314018 COMMD2 p.6272.2 6272 2 0.722032 0.019363657 SORT1 p.9166.1 9166 1 0.71989 0.022713572 EBAG9 p.9868.1 9868 1 0.715714 0.003029771 TOMM70A p.93663.2 93663 1 0.714849 0.018461196 ARHGAP18 p.808.1 808 1 0.70889 0.00084125 CALM3 p.2932.1 2932 4 0.706221 0.00148557 GSK3B p.5873.1 5873 1 0.687671 0.009466502 RAB27A p.55775.2 55775 1 0.684457 0.000295737 TDP1 p.55342.1 55342 1 0.6826 0.000318993 STRBP p.3093.1 3093 2 0.661935 0.000702017 UBE2K p.6924.1 6924 1 0.661442 0.000129949 TCEB3 p.6249.4 6249 2 0.659843 0.01775643 CLIP1 p.94104.2 94104 1 0.658328 0.000385359 GCFC1 p.64225.1 64225 2 0.65562 0.04834206 ATL2 p.55109.3 55109 2 0.644681 0.00450987 AGGF1 p.1362.1 1362 1 0.644267 0.164735867 CPD p.127018.1 127018 1 0.640176 0.013850442 LYPLAL1 p.54726.2 54726 2 0.634034 0.000739965 OTUD4 p.7803.3 7803 4 0.632354 0.024478859 PTP4A1 p.4287.2 4287 4 0.625629 0.008551638 ATXN3 p.4026.4 4026 1 0.623232 0.037022049 LPP p.9140.4 9140 1 0.622513 0.007886147 ATG12 p.25978.1 25978 1 0.61774 0.039418085 CHMP2B p.58528.1 58528 1 0.614936 0.096105373 RRAGD p.22936.1 22936 2 0.610072 0.049572333 ELL2 p.79939.1 79939 2 0.609574 0.006803917 SLC35E1 p.8545.1 8545 1 0.601315 0.016649492 CGGBP1 p.23161.1 23161 1 0.596947 0.000534007 SNX13 p.11052.2 11052 2 0.57751 0.001072231 CPSF6 p.163590.2 163590 1 0.575751 0.000151844 TOR1AIP2 p.11163.1 11163 2 0.574426 0.06813305 NUDT4 p.4205.2 4205 2 0.571992 0.058540682 MEF2A p.9958.6 9958 1 0.571921 0.018085617 USP15 p.9373.2 9373 1 0.557663 0.039796087 PLAA p.1432.1 1432 2 0.546392 0.060952202 MAPK14 p.143384.2 143384 1 0.541097 0.018544243 C10orf46 p.8774.1 8774 1 0.537813 0.009392055 NAPG p.57456.1 57456 1 0.531792 0.005923904 KIAA1143 p.10892.1 10892 4 0.520433 0.084999374 MALT1 p.5586.1 5586 1 0.518789 0.039169085 PKN2 p.23704.1 23704 1 0.516073 0.201070101 KCNE4 p.518.1 518 1 0.514817 0.007245646 ATP5G3 p.57168.1 57168 1 0.514457 0.00304206 ASPHD2 p.9262.2 9262 1 0.514405 0.044452468 STK17B p.26984.2 26984 1 0.493885 0.023470657 SEC22A p.1629.1 1629 1 0.492596 0.016998642 DBT p.55471.1 55471 1 0.490821 0.025843519 C2orf56 p.55320.1 55320 1 0.483879 0.003715179 MIS18BP1 p.5527.5 5527 1 0.477724 0.035496227 PPP2R5C p.163.2 163 1 0.472397 0.118516768 AP2B1 p.54994.2 54994 1 0.470611 0.015759 C20orf11 p.51762.2 51762 1 0.454929 0.019597276 RAB8B p.9306.3 9306 1 0.451307 0.08547743 SOCS6 p.377.1 377 1 0.449607 0.02246586 ARF3 p.1540.1 1540 4 0.444419 0.104850773 CYLD p.11169.2 11169 1 0.44378 4.21E-05 WDHD1 p.23142.1 23142 1 0.430607 0.044486472 DCUN1D4 p.10100.3 10100 3 0.426983 0.106741319 TSPAN2 p.9180.4 9180 1 0.42312 0.300312912 OSMR p.286205.2 286205 2 0.409166 0.000401905 SCAI p.11127.2 11127 1 0.398149 0.044541766 KIF3A p.4780.2 4780 1 0.390668 0.035514343 NFE2L2 p.285282.2 285282 1 0.390133 0.068636111 RABL3 p.1027.2 1027 1 0.38651 0.345813534 CDKN1B p.23057.2 23057 1 0.377734 0.128271244 NMNAT2 p.7182.1 7182 4 0.377357 0.001919509 NR2C2 p.55973.2 55973 1 0.37611 0.003439249 BCAP29 p.23621.1 23621 1 0.375918 0.088379782 BACE1 p.488.3 488 2 0.364977 0.036378751 ATP2A2 p.285672.1 285672 1 0.36385 0.035363477 SREK1IP1 p.961.1 961 2 0.361648 0.360347354 CD47 p.81550.2 81550 1 0.360271 0.01056503 TDRD3 p.4097.1 4097 1 0.351628 0.062194883 MAFG p.148808.1 148808 2 0.338065 0.279902064 MFSD4 p.91775.2 91775 1 0.337272 0.024718201 FAM55C p.5786.1 5786 1 0.334142 0.046893298 PTPRA p.55031.1 55031 1 0.334125 0.057210628 USP47 p.23362.4 23362 2 0.333083 0.503671369 PSD3 p.57539.1 57539 1 0.332467 0.076105229 WDR35 p.57148.2 57148 2 0.332243 0.009685689 RALGAPB p.89796.2 89796 1 0.331881 0.284561598 NAV1 p.361.1 361 1 0.320445 0.326873028 AQP4 p.55796.2 55796 5 0.317619 0.006151172 MBNL3 p.4763.4 4763 3 0.313912 0.084647177 NF1 p.4238.1 4238 1 0.311914 0.12584843 MFAP3 p.55186.2 55186 1 0.305336 0.177787916 SLC25A36 p.9962.1 9962 1 0.303373 0.084565323 SLC23A2 p.92126.2 92126 1 0.301246 0.056845437 DSEL p.64093.1 64093 1 0.295977 0.121331237 SMOC1 p.93.1 93 4 0.290566 0.170491525 ACVR2B p.2137.1 2137 1 0.287783 0.073052301 EXTL3 p.10413.1 10413 1 0.286149 0.449788783 YAP1 p.5737.1 5737 2 0.28555 0.351363717 PTGFR p.4848.2 4848 1 0.28483 0.196843522 CNOT2 p.10725.1 10725 2 0.277295 0.28728263 NFAT5 p.137492.2 137492 1 0.276748 0.265356893 VPS37A p.128637.1 128637 2 0.271548 0.054531759 TBC1D20 p.79685.1 79685 1 0.27136 0.087746336 SAP30L p.8763.1 8763 1 0.270601 0.544178404 CD164 p.29974.1 29974 3 0.268273 0.031311624 A1CF p.51552.1 51552 2 0.268008 0.13921524 RAB14 p.3572.1 3572 1 0.264442 0.420538133 IL6ST p.9139.2 9139 1 0.259734 0.064771407 CBFA2T2 p.208.1 208 1 0.257889 0.139680389 AKT2 p.23064.1 23064 1 0.254912 0.049366953 SETX p.161742.1 161742 5 0.250255 0.281846114 SPRED1 p.64398.3 64398 1 0.249777 0.386893254 MPP5 p.5578.2 5578 3 0.243364 0.04786041 PRKCA p.6400.3 6400 1 0.242434 0.240415741 SEL1L p.91404.1 91404 3 0.23832 0.348952623 SESTD1 p.83548.3 83548 1 0.237782 0.129399801 COG3 p.54545.1 54545 2 0.235296 0.284736973 MTMR12 p.23493.1 23493 1 0.227029 0.61623904 HEY2 p.6641.2 6641 1 0.224681 0.295617098 SNTB1 p.80014.1 80014 1 0.22367 0.244246628 WWC2 p.1374.1 1374 1 0.217355 0.04804195 CPT1A p.5728.2 5728 1 0.210932 0.357658116 PTEN p.26092.2 26092 1 0.206119 0.446531071 TOR1AIP1 p.3709.1 3709 1 0.205776 0.432292013 ITPR2 p.387338.1 387338 1 0.205606 0.253652804 NSUN4 p.8492.2 8492 2 0.200164 0.101984163 PRSS12 p.84333.2 84333 1 0.195877 0.297644452 PCGF5 p.6452.2 6452 1 0.193542 0.17595779 SH3BP2 p.26268.1 26268 1 0.191478 0.231516911 FBXO9 p.81573.4 81573 2 0.188175 0.137506831 ANKRD13C p.9469.1 9469 1 0.184347 0.102089212 CHST3 p.84236.1 84236 1 0.178655 0.29395844 RHBDD1 p.115.1 115 1 0.178495 0.332623239 ADCY9 p.23001.1 23001 1 0.177003 0.43282014 WDFY3 p.7799.4 7799 1 0.174549 0.077378442 PRDM2 p.152137.1 152137 2 0.173612 0.491083372 CCDC50 p.9218.1 9218 2 0.170494 0.556554949 VAPA p.3423.5 3423 2 0.166722 0.312045127 IDS p.6886.1 6886 2 0.164429 0.136548178 TAL1 p.10657.1 10657 1 0.152746 0.318756281 KHDRBS1 p.143684.1 143684 2 0.152616 0.396998309 FAM76B p.56243.4 56243 2 0.152163 0.414429553 KIAA1217 p.1983.6 1983 1 0.151745 0.493572749 EIF5 p.7520.2 7520 1 0.149639 0.309652006 XRCC5 p.148932.1 148932 1 0.149625 0.266023362 MOBKL2C p.56990.1 56990 1 0.141538 0.589484246 CDC42SE2 p.84056.1 84056 2 0.138149 0.176426541 KATNAL1 p.23390.1 23390 1 0.136651 0.315837341 ZDHHC17 p.2800.1 2800 1 0.129072 0.380388077 GOLGA1 p.56889.1 56889 1 0.128404 0.684977517 TM9SF3 p.27230.2 27230 1 0.125736 0.45835779 SERP1 p.83876.1 83876 1 0.125323 0.132301439 MRO p.51343.1 51343 1 0.119375 0.50695366 FZR1 p.57496.3 57496 5 0.119347 0.120758719 MKL2 p.79817.2 79817 1 0.114455 0.494455151 MOBKL2B p.801.3 801 2 0.109818 0.553505488 CALM1 p.51100.1 51100 2 0.108578 0.698189614 SH3GLB1 p.133746.1 133746 2 0.106794 0.474647079 JMY p.4204.1 4204 1 0.105201 0.454633022 MECP2 p.154810.1 154810 2 0.100534 0.325051574 AMOTL1 p.11276.2 11276 1 0.0988962 0.581457768 SYNRG p.26065.3 26065 3 0.0979143 0.585140705 LSM14A p.79776.1 79776 1 0.0975801 0.167849972 ZFHX4 p.4093.1 4093 2 0.0962148 0.330471158 SMAD9 p.102.2 102 3 0.0960521 0.763043747 ADAM10 p.55719.3 55719 1 0.0946039 0.547054847 FAM178A p.2674.2 2674 1 0.085502 0.883468476 GFRA1 p.143098.2 143098 2 0.0838597 0.567212111 MPP7 p.7057.2 7057 4 0.08262 0.788089149 THBS1 p.2889.1 2889 1 0.0762866 0.672692496 RAPGEF1 p.9236.2 9236 1 0.0577848 0.80076936 CCPG1 p.1106.3 1106 1 0.0564997 0.672185612 CHD2 p.2729.4 2729 1 0.0407808 0.88883392 GCLC p.83999.1 83999 1 0.0403445 0.73255589 KREMEN1 p.4660.2 4660 1 0.0372442 0.544487442 PPP1R12B p.7204.3 7204 1 0.0286435 0.878192293 TRIO p.4208.1 4208 2 0.0280783 0.851422555 MEF2C p.10026.5 10026 2 0.0178547 0.94038073 PIGK p.10424.4 10424 2 0.0134776 0.93881968 PGRMC2 p.3340.3 3340 1 0.0121369 0.9089681 NDST1 p.50808.2 50808 1 0.0118442 0.950998769 AK3 p.51621.1 51621 2 0.0094747 0.956808803 KLF13 p.54855.2 54855 1 0.00896938 0.977763612 FAM46C p.81488.1 81488 1 0.00768531 0.963160408 GRINL1A p.56605.4 56605 1 0.00643096 0.974446673 ERO1LB p.10735.1 10735 1 0.00393394 0.98576474 STAG2 p.4851.1 4851 1 -0.00162789 0.989875918 NOTCH1 p.5494.1 5494 4 -0.0122894 0.946107026 PPM1A p.8897.2 8897 1 -0.0139983 0.906803551 MTMR3 p.160518.2 160518 2 -0.0152684 0.7962659 DENND5B p.894.1 894 1 -0.0197799 0.913461919 CCND2 p.7020.2 7020 2 -0.0274538 0.956229834 TFAP2A p.79819.6 79819 1 -0.0322384 0.737824555 WDR78 p.9839.2 9839 2 -0.0383425 0.827712284 ZEB2 p.5337.3 5337 1 -0.0410481 0.766926242 PLD1 p.8313.1 8313 1 -0.0445589 0.619988739 AXIN2 p.22822.2 22822 6 -0.0457664 0.90306617 PHLDA1 p.8649.2 8649 2 -0.0532616 0.796255436 LAMTOR3 p.83464.1 83464 1 -0.0568737 0.719683748 APH1B p.84668.1 84668 1 -0.0607184 0.639382996 FAM126A p.23140.3 23140 1 -0.0662972 0.469960771 ZZEF1 p.140885.2 140885 1 -0.0689954 0.743861141 SIRPA p.153396.1 153396 1 -0.0710312 0.67834599 TMEM161B p.7058.5 7058 1 -0.0856479 0.815942131 THBS2 p.1902.1 1902 1 -0.104382 0.52560646 LPAR1 p.130574.2 130574 4 -0.107264 0.365770639 LYPD6 p.79745.3 79745 2 -0.107655 0.616144501 CLIP4 p.89795.1 89795 1 -0.131101 0.119900809 NAV3 p.54756.1 54756 2 -0.131135 0.271271129 IL17RD p.54665.3 54665 1 -0.135868 0.392906769 RSBN1 p.10015.2 10015 1 -0.144086 0.493182855 PDCD6IP p.51201.1 51201 1 -0.145307 0.430826575 ZDHHC2 p.66008.4 66008 1 -0.150806 0.524273993 TRAK2 p.55715.3 55715 1 -0.163888 0.13452526 DOK4 p.23405.2 23405 2 -0.172374 0.486693047 DICER1 p.23387.3 23387 2 -0.174896 0.392031886 SIK3 p.538.1 538 1 -0.178314 0.402682107 ATP7A p.11099.1 11099 1 -0.185514 0.013349693 PTPN21 p.11334.2 11334 1 -0.201129 0.228502379 TUSC2 p.1602.1 1602 2 -0.202125 0.460408697 DACH1 p.29123.7 29123 1 -0.202347 0.410008021 ANKRD11 p.7556.1 7556 1 -0.20608 0.204396309 ZNF10 p.2060.3 2060 2 -0.211391 0.244547165 EPS15 p.29068.1 29068 2 -0.227284 0.296079005 ZBTB44 p.3915.3 3915 1 -0.265352 0.353620201 LAMC1 p.10217.3 10217 2 -0.274235 0.130316239 CTDSPL p.6500.6 6500 1 -0.280407 0.151985502 SKP1 p.153222.3 153222 3 -0.298181 0.093405869 C5orf41 p.6777.1 6777 1 -0.316894 0.030150255 STAT5B p.83700.2 83700 2 -0.318731 8.57E-05 JAM3 p.5412.4 5412 1 -0.326029 0.13943951 UBL3 p.23768.1 23768 1 -0.359359 0.000294524 FLRT2 p.5937.1 5937 1 -0.366876 0.110424796 RBMS1 p.2119.2 2119 1 -0.37498 0.111021787 ETV5 p.6867.1 6867 1 -0.380351 0.306685828 TACC1 p.26053.2 26053 2 -0.382417 0.031395058 AUTS2 p.57616.1 57616 1 -0.406895 0.024206049 TSHZ3 p.79875.3 79875 1 -0.412747 0.106648899 THSD4 p.6926.1 6926 3 -0.429427 0.17521498 TBX3 p.92.1 92 1 -0.475598 0.011835869 ACVR2A p.10929.1 10929 1 -0.509401 0.013936609 SRSF8 p.1500.2 1500 1 -0.521875 0.040376858 CTNND1 p.7905.2 7905 2 -0.546689 0.074385262 REEP5 p.6558.1 6558 2 -0.563669 0.186800463 SLC12A2 p.89797.1 89797 1 -0.565451 0.003070541 NAV2 p.55450.1 55450 1 -0.709479 0.210456846 CAMK2N1 p.4211.1 4211 1 -0.777265 8.24E-06 MEIS1 p.80310.1 80310 1 -0.80233 4.40E-06 PDGFD p.116496.2 116496 1 -0.884717 0.034323472 FAM129A p.36.1 36 1 -0.918384 0.047647336 ACADSB p.23452.1 23452 1 -0.976362 1.14E-07 ANGPTL2 p.10350.3 10350 1 -1.20567 6.29E-16 ABCA9 p.3725.2 3725 2 -1.21448 0.000139363 JUN p.120.1 120 1 -1.40224 0.000412452 ADD3 p.7021.1 7021 2 -1.46108 0.022040325 TFAP2B p.1846.3 1846 1 -1.55448 0.005487796 DUSP4 p.3479.1 3479 1 -1.6844 2.18E-09 IGF1 p.8614.2 8614 1 -2.77791 8.24E-11 STC2 engthened 3'UTRs APA site gene ID Gain PUM sites log fold change P-value diff. Shortened 3'UTRs APA site gene ID lost PUM sites log fold changeP-value diff. expr gene symbol p.1301.6 1301 1 3.34604 1.20E-05 COL11A1 p.81611.1 81611 1 1.84336 0.000678749 ANP32E p.10097.1 10097 1 1.76786 3.37E-05 ACTR2 p.51026.1 51026 1 1.74004 9.13E-06 GOLT1B p.51659.1 51659 1 1.59379 6.14E-05 GINS2 p.79572.1 79572 1 1.59155 0.00035399 ATP13A3 p.5062.1 5062 2 1.49865 8.96E-05 PAK2 p.4893.2 4893 1 1.48177 0.002016624 NRAS p.8507.1 8507 1 1.37817 0.007869441 ENC1 p.10276.2 10276 2 1.37096 0.00099952 NET1 p.9877.1 9877 2 1.33103 4.95E-06 ZC3H11A p.80155.1 80155 1 1.31275 4.48E-07 NAA15 p.55706.1 55706 1 1.31015 0.000572818 TMEM48 p.92140.2 92140 3 1.30391 4.98E-06 MTDH p.25932.1 25932 2 1.30381 0.00644907 CLIC4 p.528.3 528 1 1.29909 0.000144482 ATP6V1C1 p.84955.2 84955 2 1.27068 4.94E-05 NUDCD1 p.8543.1 8543 1 1.26135 0.019855937 LMO4 p.55284.1 55284 1 1.24239 6.14E-07 UBE2W p.117178.3 117178 1 1.23431 9.53E-05 SSX2IP p.51018.4 51018 1 1.22593 2.35E-06 RRP15 p.57532.6 57532 3 1.22504 1.72E-05 NUFIP2 p.112858.2 112858 1 1.18547 7.93E-05 TP53RK p.8669.2 8669 1 1.1763 5.05E-06 EIF3J p.10890.4 10890 1 1.09651 9.14E-08 RAB10 p.3267.1 3267 1 1.09645 4.22E-06 AGFG1 p.6745.2 6745 1 1.0902 0.000747828 SSR1 p.79677.2 79677 2 1.08962 3.66E-05 SMC6 p.23527.2 23527 2 1.07493 0.000213667 ACAP2 p.5203.1 5203 1 1.07094 0.00070456 PFDN4 p.79888.2 79888 1 1.05371 0.001510546 LPCAT1 p.26046.2 26046 1 1.05281 0.000231483 LTN1 p.153339.1 153339 1 1.04776 0.000311249 TMEM167A p.55110.1 55110 2 1.03488 0.001739127 MAGOHB p.5469.1 5469 2 1.02688 0.050409219 MED1 p.54800.1 54800 2 1.0235 0.002063174 KLHL24 p.11320.3 11320 4 1.00629 0.028935489 MGAT4A p.54149.1 54149 1 0.997606 0.006795977 C21orf91 p.55971.1 55971 1 0.989322 0.001160397 BAIAP2L1 p.22797.2 22797 1 0.986222 0.007900227 TFEC p.7994.2 7994 1 0.966617 0.02085234 MYST3 p.10059.2 10059 1 0.957199 0.000454561 DNM1L p.25917.2 25917 1 0.93798 0.000124526 THUMPD3 p.2786.2 2786 1 0.932943 0.03126166 GNG4 expr. Miles_Supp Fig 5 Supp Table 1: APA shortened transcripts in Basal-like and Triple-negative (A) List of shortened mRNAs from Basal-like breast tumors. (B) List of shortened mRNAs from Triple-negative breast tumors. Supp Fig 1: 3’UTR length variations in an independent cohort of triple-negative breast tumors (A) Mean and standard deviation of the E2F4 transcript from normal breast tissue and triple- negative breast tumors. (B) RT-PCR results from triple-negative breast tumors of the E2F4 3’UTR (negative control). Expression of the E2F4 coding sequence set to 1. (C) RT-PCR results from normal breast tissue mapping the length of the E2F4 3’UTR. (D) RT-PCR of the FOXO1 mRNA and 3’UTR from triple-negative breast tumors. (E) RT-PCR of the FOXO1 mRNA and 3’UTR from normal breast tissue. (F) RT-PCR of the PTEN mRNA and 3’UTR from triple-negative breast tumors. (G) RT-PCR of the PTEN mRNA and 3’UTR from normal breast tissue. (H) Percentage of triple-negative tumors displaying APA of E2F4, FOXO1 and PTEN. Shortened 3'UTRs APA site gene ID lost PUM sites log fold changeP-value diff. expr gene symbol p.1301.6 1301 1 3.34604 1.20E-05 COL11A1 p.81611.1 81611 1 1.84336 0.000678749 ANP32E p.10097.1 10097 1 1.76786 3.37E-05 ACTR2 p.51026.1 51026 1 1.74004 9.13E-06 GOLT1B p.51659.1 51659 1 1.59379 6.14E-05 GINS2 p.79572.1 79572 1 1.59155 0.00035399 ATP13A3 p.5062.1 5062 2 1.49865 8.96E-05 PAK2 p.4893.2 4893 1 1.48177 0.002016624 NRAS p.8507.1 8507 1 1.37817 0.007869441 ENC1 p.10276.2 10276 2 1.37096 0.00099952 NET1 p.9877.1 9877 2 1.33103 4.95E-06 ZC3H11A p.80155.1 80155 1 1.31275 4.48E-07 NAA15 p.55706.1 55706 1 1.31015 0.000572818 TMEM48 p.92140.2 92140 3 1.30391 4.98E-06 MTDH p.25932.1 25932 2 1.30381 0.00644907 CLIC4 p.528.3 528 1 1.29909 0.000144482 ATP6V1C1 p.84955.2 84955 2 1.27068 4.94E-05 NUDCD1 p.8543.1 8543 1 1.26135 0.019855937 LMO4 p.55284.1 55284 1 1.24239 6.14E-07 UBE2W p.117178.3 117178 1 1.23431 9.53E-05 SSX2IP p.51018.4 51018 1 1.22593 2.35E-06 RRP15 p.57532.6 57532 3 1.22504 1.72E-05 NUFIP2 p.112858.2 112858 1 1.18547 7.93E-05 TP53RK p.8669.2 8669 1 1.1763 5.05E-06 EIF3J p.10890.4 10890 1 1.09651 9.14E-08 RAB10 p.3267.1 3267 1 1.09645 4.22E-06 AGFG1 p.6745.2 6745 1 1.0902 0.000747828 SSR1 p.79677.2 79677 2 1.08962 3.66E-05 SMC6 p.23527.2 23527 2 1.07493 0.000213667 ACAP2 p.5203.1 5203 1 1.07094 0.00070456 PFDN4 p.79888.2 79888 1 1.05371 0.001510546 LPCAT1 p.26046.2 26046 1 1.05281 0.000231483 LTN1 p.153339.1 153339 1 1.04776 0.000311249 TMEM167A p.55110.1 55110 2 1.03488 0.001739127 MAGOHB p.5469.1 5469 2 1.02688 0.050409219 MED1 p.54800.1 54800 2 1.0235 0.002063174 KLHL24 p.11320.3 11320 4 1.00629 0.028935489 MGAT4A p.54149.1 54149 1 0.997606 0.006795977 C21orf91 p.55971.1 55971 1 0.989322 0.001160397 BAIAP2L1 p.22797.2 22797 1 0.986222 0.007900227 TFEC p.7994.2 7994 1 0.966617 0.02085234 MYST3 p.10059.2 10059 1 0.957199 0.000454561 DNM1L p.25917.2 25917 1 0.93798 0.000124526 THUMPD3 p.2786.2 2786 1 0.932943 0.03126166 GNG4 gene symbol p.2308.2 2308 1 -1.6295 1.64E-08 FOXO1 p.6309.2 6309 1 -1.62248 2.12E-06 SC5DL p.9252.2 9252 1 -1.54632 0.00011932 RPS6KA5 p.3667.2 3667 8 -1.4955 0.001657433 IRS1 p.57125.1 57125 1 -1.30928 0.000306387 PLXDC1 p.56898.1 56898 1 -1.2938 8.58E-05 BDH2 p.6925.1 6925 3 -1.25683 0.000600007 TCF4 p.3176.2 3176 2 -1.19904 0.00027889 HNMT p.81563.5 81563 1 -1.16602 0.001680965 C1orf21 p.55603.1 55603 1 -1.12472 0.007161292 FAM46A p.11170.1 11170 1 -1.09491 0.006247135 FAM107A p.81539.1 81539 3 -0.828634 0.019664202 SLC38A1 p.10491.3 10491 1 -0.790634 0.000170561 CRTAP p.51312.2 51312 1 -0.782017 0.058080082 SLC25A37 p.26469.2 26469 1 -0.70474 0.011680346 PTPN18 p.131566.7 131566 1 -0.690306 0.00437571 DCBLD2 p.9218.6 9218 2 -0.671798 0.001230882 VAPA p.79365.1 79365 1 -0.668554 0.112433371 BHLHE41 p.573.2 573 1 -0.631085 0.025300848 BAG1 p.6622.1 6622 1 -0.600955 4.63E-06 SNCA p.64778.2 64778 1 -0.578246 0.038028609 FNDC3B p.200958.1 200958 2 -0.553899 0.135360905 MUC20 p.64388.1 64388 1 -0.534599 0.002645254 GREM2 p.171425.3 171425 1 -0.516825 0.015380678 CLYBL p.23038.1 23038 2 -0.509143 0.000334 WDTC1 p.175.1 175 2 -0.429467 0.154322571 AGA p.84897.2 84897 1 -0.406707 0.018423573 TBRG1 p.8874.2 8874 1 -0.399612 0.043162955 ARHGEF7 p.55333.3 55333 1 -0.396708 0.114101202 SYNJ2BP p.54491.1 54491 1 -0.289698 0.053320848 FAM105A p.23607.2 23607 2 -0.225057 0.1281634 CD2AP p.51249.1 51249 1 -0.167554 0.508842086 TMEM69 p.80306.1 80306 1 -0.145682 0.567196653 MED28 p.79752.2 79752 1 -0.144008 0.677482834 ZFAND1 p.9562.1 9562 1 0.00641579 0.970841107 MINPP1 p.115350.1 115350 2 0.0147064 0.899185368 FCRL1 p.1977.1 1977 3 0.0440604 0.654658669 EIF4E p.6672.5 6672 1 0.0604781 0.78030873 SP100 p.6651.1 6651 1 0.0629671 0.755235526 SON p.65108.1 65108 1 0.0936532 0.697927499 MARCKSL1 p.149371.1 149371 1 0.180538 0.358534392 EXOC8 p.6738.1 6738 1 0.211184 0.241726643 TROVE2 p.4883.1 4883 1 0.226194 0.201443 NPR3 p.90268.2 90268 1 0.29415 0.063667699 FAM105B p.3225.2 3225 1 0.306857 0.395120885 HOXC9 p.3842.3 3842 1 0.307704 0.107106004 TNPO1 p.23213.2 23213 1 1.05363 0.020403615 SULF1 p.259266.1 259266 1 1.25827 4.23E-05 ASPM p.51514.1 51514 1 1.87144 1.63E-06 DTL p.3225.2 3225 1 0.306857 0.395120885 HOXC9 p.3842.3 3842 1 0.307704 0.107106004 TNPO1 p.23213.2 23213 1 1.05363 0.020403615 SULF1 p.259266.1 259266 1 1.25827 4.23E-05 ASPM p.51514.1 51514 1 1.87144 1.63E-06 DTL Miles Supplementary Table 2: Primers and siRNAs Primers Miles Supplementary Table 2: Primers and siRNAs Primers 3’RACE RT-PCR Primer Sequence P7-t25-vn oligo-dT primer CAAGCAGAAGACGGCATACGAGATTTTTTTTTTTTTTTTTTTTTTTTTVN CSTF2 forward ATCCTTGCCTGCGAATGTCC CSTF2 reverse GGGTGGTCCTCGGCTCTC CSTF3 forward TGAAGTGGATAGAAAACCAGAATACCC CSTF3 reverse TGGAAACAGATAGGAGGAGGGAGA CPSF2 forward CTTTGGAACCCTTGCCACCT CPSF2 reverse TGCGGACTGCTACTTGATTGTTG TMEM119 proximal forward CCCTGGCAACATTGTGAGACC TMEM119 distal forward TCTCCCCCATCCCTCCATCT FAM100B proximal forward CAGGAGTTTTTCAGGCAAGTTTTTC FAM100B distal forward AGTGGAGAGCCTGCCTTTGG P7 forward CAAGCAGAAGACGGCATACGAGA NRAS distal forward GGATGTCCGTGGAAGTTGTAA NRAS proximal forward TTTGTTTCCCCTAGCTTACTGA JUN distal forward GGTGCTTATTCTCAAAGCAGGA JUN proximal forward GCTTCTGTACTCCTTAAGAACACA actin forward distal CAGCCAGGGCTTACCTGT Tumor RT-PCR Primers Sequence FOXA1 TUMOR RT FOR CTCCTACTGGACGCTGCAC FOXA1 TUMOR RT REV GCAAGTAGCAGCCGTTCTC SMG1 TUMOR RT FOR GACGTGCTTAATGAAAGTAGCAA SMG1 TUMOR RT REV AAGGCCAAGACAGCAAGC CCND1 TUMOR RT FOR AGATCATCCGCAAACACG CCND1 TUMOR RT REV TGAACTTCACATCTGTGGCA STAT5B UTR 2 TUMOR FOR CATGTGGCTGTTTGAATAAGAGA STAT5B UTR 2 TUMOR REV GACGAAGACTCACTGGAGCA STAT5B CDS TUMOR FOR GCTAAAGCTGTTGATGGATACG STAT5B CDS TUMOR REV GATGCGTTCACAAACTCAGG PTEN UTR2 TUMOR FOR GTCATGCATGCAGATGGAA PTEN UTR2 TUMOR REV CTGCCAAATACTACAGTTAAAGCC PTEN CDS TUMOR FOR ACACGACGGGAAGACAAGTT PTEN CDS TUMOR REV CTTTGATATCACCACACACAGG FOXO1 UTR1 TUMOR FOR AGACTTGGCAGCAAAGACATT FOXO1 UTR1 TUMOR REV GCAGCACATAACCTGCACA FOXO1 UTR2 TUMOR FOR TGAATCCTTGGTAGCTCTCTGA FOXO1 UTR1 TUMOR REV TTGCTTTCCAGACAGACCAG RHOF UTR2 TUMOR FOR CGTCAGCTAAACCTGAGCTTC RHOF UTR2 TUMOR REV GGGAGGGTCTCAGAAGGC RHOF UTR1 TUMOR FOR CTGCGGCTAGACTCTTGGA RHOF UTR1 TUMOR REV GGGAGTTTGTGGTCAGAGC IRS1 UTR1 TUMOR FOR GCACTCTGTAGGTTAGTCCATCA IRS1 UTR1 TUMOR REV CCGCCAAGGTCTTAATTCAC IRS1 UTR2 TUMOR FOR GCCATGGGTCCTTTCTGA IRS1 UTR2 TUMOR REV ACAGACTTACAGAACTACGGAAGG c-JUN RT-PCR for tumor CDNA1 aacgtgacagatgagcagga c-JUN RT-PCR rev tumor CDNA1 gtgttctggctgtgcagttc c-JUN RT-PCR for tumor CDNA2 gaaggaggagcctcagacag c-JUN RT-PCR rev tumor CDNA2 ctgggactccatgtcgatg c-JUN RT-PCR for tumor 3utr1 aagttgcgacggagagaaaa c-JUN RT-PCR rev tumor 3utr1 acgcaacccagtccaactt c-JUN RT-PCR for tumor 3utr2 tcaataaacagcttcatgccttt c-JUN RT-PCR rev tumor 3utr2 acaacactgggcaggatacc c-JUN RT-PCR for tumor 3utr2-1 ggatagtgcgatgtttcagga c-JUN RT-PCR rev tumor 3utr2-1 gacttctcagtgggctgtcc NRAS RT-PCR FOR TUMOR1 CDS 1 tggtgaaacctgtttgttgg NRAS RT-PCR REV TUMOR1 CDS 1 tcgcctgtcctcatgtattg NRAS RT-PCR FOR TUMOR1 CDS 2 acaaaacaagcccacgaact NRAS RT-PCR REV TUMOR1 CDS 2 aacaccctgtctggtcttgg NRAS RT-PCR FOR TUMOR1 3UTR1 caccccaggtttttcatctg NRAS RT-PCR REV TUMOR1 3'UTR1 gaatttcacgtttgcggttt NRAS RT-PCR FOR TUMOR1 3UTR1-1 tttgagagcagatgccatttt NRAS RT-PCR REV TUMOR1 3'UTR1-1 ctgggaaagccccaaaac NRAS RT-PCR FOR TUMOR1 3UTR2 ctctacagcttcccccacag NRAS RT-PCR REV TUMOR1 3'UTR2 ggtccattagcaccatgctt NRAS RT-PCR FOR TUMOR1 3UTR2-1 agcctgggtgacaagagaga NRAS RT-PCR REV TUMOR1 3'UTR2-1 cctttgtgccaaaattcaca E2F4 RT-PCR FOR TUMORS CDS1 tcacagaggacgtgcagaac E2F4 RT-PCR REV TUMORS CDS1 agggtatctccagcaaagca E2F4 RT-PCR FOR TUMORS CDS1-1 ggacccaacccttctacctc E2F4 RT-PCR REV TUMORS CDS1-1 tgacagctctttggggagtt E2F4 RT-PCR FOR TUMORS 3UTR1 accgaggagctgccattac E2F4 RT-PCR REV TUMORS 3UTR1 ctggcagaagcaagacactg E2F4 RT-PCR FOR TUMORS 3UTR1-1 cacttctgctgggcctttag E2F4 RT-PCR REV TUMORS 3UTR1-1 ccccagagacctctgttcct CSTF2 TUMOR Forward TAGCTCCTGGAGGAGGAATG CSTF2 TUMOR Reverse CGTTCCATGGACACTGGTC CPSF2 TUMOR Forward GCAGCTTCTGACAAATGTCCT CPSF2 TUMOR Reverse CCTGCTGTGTCCACTGCTAT E2F1 TUMOR Forward TGCAGAGCAGATGGTTATGG E2F1 TUMOR Reverse GAAAGTTCTCCGAAGAGTCCA E2F4 TUMOR Forward GAGATACCCTCTTGGCCATC E2F4 TUMOR Reverse CCCATTGAGACCCTCTGG CFIM25 TUMOR FOR ATACTGGGTCGTCAGGATGG CFIM25 TUMOR REV TCTCCACCAGTTACCAATGC KRT7 For tumor AGGTGGATGCCCTGAATG KRT7 rev tumor GCTCTGTCAACTCCGTCTCA KRT19 for tumor TTGTCCTGCAGATCGACAAT KRT 19 rev tumor GTCTCAAACTTGGTTCGGAAG TIMP1 for tumor AGTGTCTGCGGATACTTCCA TIMP1 rev tumor TTCCAGCAATGAGAAACTCCT PUM1 tumor forward ACTTGCTCTGGAGGGAAGAG PUM1 tumor reverse TGGATTCATCTGTTTCCCAA PUM2 tumor forward TGGAAGTGGTAACTCTTTGGG PUM2 tumor reverse TGCCAACTGATGAACCAAAT RT-PCR Primers Sequence cycd1 Forward CCGTCCATGCGGAAGATC cycd1 Reverse ATGGCCAGCGGGAAGAC bim Forward TGGCAAAGCAACCTTCTGATG bim Reverse GCAGGCTGCAATTGTCTACCT trail Forward GCTCTGGGCCGCAAAAT trail Reverse TGCAAGTTGCTCAGGAATGAA DR5 Forward GGGCCACAGGGACACCTT DR5 Reverse GCATCTCGCCCGGTTTT siRNA sense PUM2-136881 GCUUUCUCACCAUUCAAUGU PUM2-136882 GCAUGGUAGAAUAUGUAUU PUM2-136883 CGAGUAAGUAGACAAGAAU PUM1-138317 GCAACUAAUUCAGCUAAUC PUM1-138318 GCUAAUCAACAGACCACCC PUM1-138319 CCAGACACUCCCUAUUUUA CSTF77-145771 GCAUUCUCAUUCGAGAGGC CSTF77-145772 GCUGUCCGAAGAGUUUAUC CSTF77-145773 GCACUCAUGGAAUAUUACU PABPN1-J-011803-05 GGAACGGCCUGGAGUCUGA PABPN1-J-011803-06 AGUCAACCGUGUUACCAUA PABPN1-J-011803-07 GAGUCCACCUCCAGGCAAU PABPN1-J-011803-08 UAUCAAAGCUCGAGUCAGG PUM1-138317 GCAACUAAUUCAGCUAAUC PUM1-138318 GCUAAUCAACAGACCACCC PUM1-138319 CCAGACACUCCCUAUUUUA CSTF77-145771 GCAUUCUCAUUCGAGAGGC CSTF77-145772 GCUGUCCGAAGAGUUUAUC CSTF77-145773 GCACUCAUGGAAUAUUACU PABPN1-J-011803-05 GGAACGGCCUGGAGUCUGA PABPN1-J-011803-06 AGUCAACCGUGUUACCAUA PABPN1-J-011803-07 GAGUCCACCUCCAGGCAAU PABPN1-J-011803-08 UAUCAAAGCUCGAGUCAGG CDS UTR2 4 0 0.2 0.4 0.6 0.8 1 1.2 CDS UTR 2 Relative expression Tumor mRNA length: PTEN Breast tissue mRNA length: PTEN SHORT LONG Tumor Tumor mRNA length: FOXO1 0 0.2 0.4 0.6 0.8 1 1.2 1 2 3 4 5 SHORT LONG Breast tissue mRNA length: FOXO1 Relative expression Normal breast samples C CDS1 CDS2 UTR1 UTR1-1 Tumor Breast Tissue 2F4 3’UTR length in triple-negative reast tumors (Control substrate) 36/36 8/8 3’UTR position 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative expression E2F4 cDNA Length in triple- negative breast tumors 3’UTR position 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative E2F4 length E2F4 cDNA length in normal breast tissue 3’UTR position B E H G E 2 F 4 F O X O 1 P T E N 0 5 1 0 1 5 E 2 F 4 , F O X O 1 A N D P T E N G e n e s % o f tu m o rs d is p la y in g A P A A C 0 0.4 0.8 1.2 CDS1 CDS2 UTR1 UTR1-1 Relative E2F4 length Tumor Breast Tissue E2F4 3’UTR length in triple-negative breast tumors (Control substrate) 36/36 8/8 3’UTR position 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative expression E2F4 cDNA Length in triple- negative breast tumors 3’UTR position 0 0.5 1 1.5 CDS1 Relative E2F4 length E2F4 cD B C R1-1 gative rate) 36/36 8/8 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative expression E2F4 cDNA Length in triple- negative breast tumors 3’UTR position 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative E2F4 length E2F4 cDNA length in normal breast tissue 3’UTR position B B 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative expression E2F4 cDNA Length in triple- negative breast tumors 3’UTR position B C A 0 0.4 0.8 1.2 CDS1 CDS2 UTR1 UTR1-1 Relative E2F4 length Tumor Breast Tissue E2F4 3’UTR length in triple-negative breast tumors (Control substrate) 36/36 8/8 3’UTR position A C 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 Relative E2F4 length E2F4 cDNA length in normal breast tissue 3’UTR position E2F4 cDNA length in normal breast tissue SHORT LONG ngth: FOXO1 0 0.2 0.4 0.6 0.8 1 1.2 1 2 3 4 5 SHORT LONG Breast tissue mRNA length: FOXO1 Relative expression l b l E H 5 1 0 1 5 E 2 F 4 , F O X O 1 A N D P T E N m o rs d is p la y in g A P A D E E H 0 0.2 0.4 0.6 0.8 1 1.2 CDS UTR 2 Relative expression Breast tissue mRNA length: PTEN mRNA position Normal breast samples Miles_Supp Fig 1 G E 2 F 4 F O X O 1 P T E N 0 G e n e s % o f tu m 0 0.5 1 1.5 CDS UTR2 Relative expression 4 Tumor mRNA length: PTEN Tumor mRNA position F F F G Miles_Supp Fig 1 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Relative expression NRAS shortened 3'UTRs triple-negative 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Average shortening of the NRAS 3'UTR in triple-negative breast tumors 0 0.2 0.4 0.6 0.8 1 1.2 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Average NRAS length normal breast tissue Relative Expression Relative Expression 3’UTR position 3’UTR position 3’UTR position C D F Miles_Supp Fig 2 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Relative expression NRAS shortened 3'UTRs triple-negative 3’UTR position C 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 relative RNA level to CDS NRAS full length in triple -negative tumors 3’UTR position A NRAS full length in triple -negative tumors NRAS shortened 3'UTRs triple-negative NRAS full length in triple -negative tumors C A 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Average shortening of the NRAS 3'UTR in triple-negative breast tumors Relative Expression 3 UTR position 3’UTR position D 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Average length of full-length 3'UTR in triple-negative breast tumors Relative Expression 3 UTR position 3’UTR position B D D B 0 0.5 1 1.5 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Relative NRAS length NRAS cDNA length normal breast 3 UTR position 3’UTR position E 0 0.2 0.4 0.6 0.8 1 1.2 CDS1 CDS2 UTR1 UTR1-1 UTR2 UTR2-1 Average NRAS length normal breast tissue Relative Expression 3’UTR position F Mil S Fi 2 F E Miles_Supp Fig 2 UTR2-1 t tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression JUN 3'UTR shortening in triple negative breast tumors UTR2-1 ple 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Average length of JUN shortening in triple- negative breast tumors UTR2-1 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative JUN length Average JUN length normal breast tissue Relative expression C D F 3’UTR position 3’UTR position 3’UTR position Miles_Supp Fig 3 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 JUN full length in triple-negative breast tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression JUN 3'UTR shortening in triple negative breast tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Average Full length 3'UTRs of JUN in triple negative breast tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Average length of JUN shortening in triple- negative breast tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 JUN cDNA length normal breast tissue 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative JUN length Average JUN length normal breast tissue Relative expression C D F 3’UTR position 3’UTR position 3’UTR position 3’UTR position 3’UTR position 3’UTR position Miles_Supp Fig 3 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression JUN full length in triple-negative breast tumors A 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression JUN 3'UTR shortening in triple negative breast tumors C C C A JUN 3'UTR shortening in triple negative breast tumors JUN full length in triple-negative breast tumors 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression Average Full length 3'UTRs of JUN in triple negative breast tumors B D 3’UTR position 3’UTR position 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Average length of JUN shortening in triple- negative breast tumors Relative expression D 3’UTR position 3’UTR position B B D 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative JUN length Average JUN length normal breast tissue F 3 UTR position 3’UTR position Miles_Supp Fig 3 0 0.5 1 1.5 CDS UTR1 UTR2 UTR2-1 Relative expression JUN cDNA length normal breast tissue E 3 UTR position 3’UTR position F F E 0 5 1 5 0hr 2hr 6hr 0 0.5 1 1.5 0hr 2hr 6hr Scr PUM1 PUM2 time after Actinomycin D FOXO1 Full Length 3’UTR FOXO1 Extended 3’UTR shRNA MDA-MB-231 cells WB:TUB WB:PUM1 Scr 1 2 1 2 PUM1 shRNA PUM2 WB:PUM2 A 0 5 1 0hr 2hr 6hr 0 0.5 1 0hr 2hr 6hr Scr PUM1 PUM2 PTEN Long 3’UTR PTEN Short 3’UTR time after Actinomycin D shRNA E MCF7 cells WB:TUB WB:PUM1 Scr 1-5 1-9 PUM1 shRNA MCF7 cells WB:TUB WB:PUM2 Scr 2-1 2-4 PUM2 shRNA D 0 6 12 18 Long Short PRE mut Scr PUM1 PUM2 NRAS Luciferase construct shRNA Relative Luciferase MBA-MD-231 cells: NRAS NRAS Luciferase construct Relative Luciferase MCF7 cells: NRAS shRNA 0 5 10 15 Long short PRE mut Scr PUM1 PUM2 F s_Supp Fig 4 MDA-MB-231 cells WB:TUB WB:PUM1 Scr 1 2 1 2 PUM1 shRNA PUM2 WB:PUM2 A MCF7 cells WB:TUB WB:PUM1 Scr 1-5 1-9 PUM1 shRNA MCF7 cells WB:TUB WB:PUM2 Scr 2-1 2-4 PUM2 shRNA D MCF7 cells WB:TUB WB:PUM1 Scr 1-5 1-9 PUM1 shRNA MCF7 cells WB:TUB WB:PUM2 Scr 2-1 2-4 PUM2 shRNA D D MDA-MB-231 cells WB:TUB WB:PUM1 Scr 1 2 1 2 PUM1 shRNA PUM2 WB:PUM2 A A A 0.5 1 1.5 0.5 1 1.5 Scr PUM1 PUM2 tive Luciferase mRNA FOXO1 Full Length 3’UTR FOXO1 Extended 3’UTR shRNA B E 0 6 12 18 Long Short PRE mut Relative Luciferase MBA-MD-231 cells: NRAS 0 0.5 1 1.5 0hr 2hr 6hr 0 0.5 1 1.5 0hr 2hr 6hr Scr PUM1 PUM2 Relative Luciferase mRNA time after Actinomycin D FOXO1 Full Length 3’UTR FOXO1 Extended 3’UTR shRNA B E Relative Luciferase E 0 6 12 18 Long Short PRE mut Relative Luciferase MBA-MD-231 cells: NRAS E B B time after Actinomycin D 0 5 0hr 2hr 6hr 0 0.5 1 0hr 2hr 6hr Scr PUM1 PUM2 PTEN Long 3’UTR PTEN Short 3’UTR time after Actinomycin D shRNA construct shRNA NRAS Luciferase construct Relative Luciferase MCF7 cells: NRAS shRNA 0 5 10 15 Long short PRE mut Scr PUM1 PUM2 F s Supp Fig 4 construct shRNA Relative Luciferase MCF7 cells: NRAS shRNA 0 5 10 15 Long short PRE mut Scr PUM1 PUM2 F F C NRAS Luciferase construct time after Actinomycin D A 0 0.4 0.8 1.2 0hr 2hr 6hr Relative Luciferase mRNA 0 0.4 0.8 1.2 0hr 2hr 6hr Scr PUM1 PUM2 NRAS Long 3’UTR NRAS Short 3’UTR time after Actinomycin D shRNA B 0 0.5 1 0hr 2hr 6hr time after Actinomycin D Relative Luciferase mRNA 0 0.5 1 0hr 2hr 6hr Scr PUM1 PUM2 c-JUN Long 3’UTR c-JUN Short 3’UTR shRNA A 0 0.4 0.8 1.2 0hr 2hr 6hr Relative Luciferase mRNA 0 0.4 0.8 1.2 0hr 2hr 6hr Scr PUM1 PUM2 NRAS Long 3’UTR NRAS Short 3’UTR time after Actinomycin D shRNA A A A 0 0.4 0.8 1.2 0hr 2hr 6hr Relative Luciferase mRNA 0 0.4 0.8 1.2 0hr 2hr 6hr Scr PUM1 PUM2 NRAS Long 3’UTR NRAS Short 3’UTR time after Actinomycin D shRNA A 0 0.4 0.8 1.2 0hr 2hr 6hr Relative Luciferase mRNA NRAS Long 3’UTR 0 0.4 0.8 .2 0hr 2hr 6hr Scr PUM1 PUM2 NRAS Short 3’UTR shRNA time after Actinomycin D B 0 0.5 1 0hr 2hr 6hr Relative Luciferase mRNA 0 0.5 1 0hr 2hr 6hr Scr PUM1 PUM2 c-JUN Long 3’UTR c-JUN Short 3’UTR shRNA B 0 0.5 1 0hr 2hr 6hr time after Actinomycin D Relative Luciferase mRNA 0 0.5 1 0hr 2hr 6hr Scr PUM1 PUM2 c-JUN Long 3’UTR c-JUN Short 3’UTR shRNA B time after Actinomycin D Miles_Supp Fig 5 -100 -80 -60 -40 -20 0 20 FAM110 NRAS JUN TMEM119 SCR PUM1 PUM2 PABPN1 MCF7 cells shRNA Relative Distal/proximal RT-PCR FAM110 NRAS JUN TMEM119 SCR PUM1 PUM2 PABPN1 shRNA MCF10A cells MCF10A cells B M1 Scr 1 2 PUM1 A MCF10A cells WB:TUB WB:PUM2 Scr 1 2 PUM2 shRNA B D WB:TUB WB:PABPN1 siRNA MCF10A cells WB:TUB WB:PABPN1 siRNA MCF7 cells -100 -80 -60 -40 -20 0 20 FAM110 NRAS JUN TMEM119 SCR PUM1 PUM2 PABPN1 shRNA Relative Distal/proximal RT-PCR FAM110 NRAS JUN TMEM119 SCR PUM1 PUM2 PABPN1 shRNA B -100 -80 -60 -40 -20 0 20 FAM110 NRAS JUN TMEM119 SCR PUM1 PUM2 PABPN1 shRNA Relative Distal/proximal RT-PCR B B A MCF7 cells MCF10A cells MCF10A cells 1 Scr 1 2 PUM1 MCF10A cells WB:TUB WB:PUM2 Scr 1 2 PUM2 shRNA D WB:TUB WB:PABPN1 siRNA MCF10A cells WB:TUB WB:PABPN1 siRNA MCF7 cells MCF10A cells WB:TUB WB:PUM1 Scr 1 2 PUM1 shRNA MCF10A cells WB:TUB WB:PUM2 Scr 1 2 PUM2 shRNA C D WB:TUB WB:PABPN1 siRNA MCF10A cells WB:T WB:P siRNA C D E Miles_Supp Fig 6 Miles_Supp Fig 6 Supp Table 2: List of primers and siRNAs used in this study Supp Fig 1: 3’UTR length variations in an independent cohort of triple-negative breast tumors Supp Fig 2: Alternative poly-adenylation of the NRAS in Triple-negative breast tumors (A) RT-PCR results from triple-negative breast tumors which retain a full length NRAS 3’UTR. Expression of the NRAS coding sequence set to 1, only tumors which have more than 50% of UTR region 2 in this collection. (B) Mean and standard deviation of NRAS expression from all of the tumors included in A. (C) RT-PCR results from triple-negative breast tumors which have shortened the NRAS 3’UTR. Expression of the NRAS coding sequence set to 1, only tumors which have less than 50% of UTR region 2 in this collection. (D) Mean and standard deviation of NRAS expression from all of the tumors included in C. (E) RT-PCR results from normal breast tissue mapping the length of the NRAS 3’UTR. (F) Mean and standard deviation of NRAS expression from the normal breast tissue included in E. Supp Fig 4: APA of FOXO1, PTEN and NRAS change their sensitivity to PUM activity. g 5: APA of FOXO1, PTEN and NRAS change their sensitivity to PUM activity (A) mRNA stability assays of the NRAS short and long 3’UTR isoforms from MDA-MB-231 cells depleted of PUM1, PUM2 or Scrambled controls sequences. (B) mRNA stability assays of the c- JUN short and long 3’UTR isoforms from MDA-MB-231 cells depleted of PUM1, PUM2 or Scrambled controls sequences. g 3: Alternative poly-adenylation of the c-JUN in Triple-negative breast tumo (A) RT-PCR results from triple-negative breast tumors which retain a full length c-JUN 3’UTR. Expression of the c-JUN coding sequence set to 1, only tumors which have more than 50% of UTR region 2 in this collection. (B) Mean and standard deviation of c-JUN expression from all of the tumors included in A. (C) RT-PCR results from triple-negative breast tumors which have shortened the c-JUN 3’UTR. Expression of the c-JUN coding sequence set to 1, only tumors which have less than 50% of UTR region 2 in this collection. (D) Mean and standard deviation of c-JUN expression from all of the tumors included in C. (E) RT-PCR results from normal breast tissue mapping the length of the c-JUN 3’UTR. (F) Mean and standard deviation of c-JUN expression from the normal breast tissue included in E. (A) Western blots of PUM1, PUM2, and TUB from MDA-MB-231 cells infected with two independent shRNAs targeting PUM1, PUM2 or Scrambled control sequences. (B) mRNA stability assays of the FOXO1 short and long 3’UTR isoforms from MDA-MB-231 cells depleted of PUM1, PUM2 or Scrambled controls sequences. (C) mRNA stability assays of the PTEN short and long 3’UTR isoforms from MDA-MB-231 cells depleted of PUM1, PUM2 or Scrambled controls sequences. (D) Western blots of PUM1, PUM2, and TUB from MCF7 cells infected with two independent shRNAs targeting PUM1, PUM2 or Scrambled control sequences. (E) Luciferase expression levels from MDA-MB-231 cells depleted of PUM1, PUM2 and Scrambled sequences and then transfected with Luciferase constructs containing either the short, long or PRE-mutant 3’UTR isoforms of NRAS. Long isoform in Scramble depleted cells set to 1. (p=0.01). (F) Luciferase expression levels from MCF7 cells depleted of PUM1, PUM2 and Scrambled sequences and then transfected with Luciferase constructs containing either the short, long or PRE-mutant 3’UTR isoforms of NRAS. Long isoform in Scramble depleted cells set to 1. (p=0.01). Supp Figure 6: Pumilio does not affect alternative poly-adenylation status of candidate substrates. (A) 3’RACE RT-PCR of FAM110, NRAS, c-JUN and TMEM119 from MCF10a cells depleted of PUM1, PUM2, PAPBN1 or Scrambled control sequences by shRNA. Distal APA site divided by proximal site from Scrambled treated cells set to 0. (B) 3’RACE RT-PCR of FAM110, NRAS, c- JUN and TMEM119 from MCF7 cells depleted of PUM1, PUM2, PAPBN1 or Scrambled control sequences by shRNA. Distal APA site divided by proximal site from Scrambled treated cells set to 0. (C) Western blots of PUM1, PUM2, PABPN1 and TUB from MCF10a cells treated with shRNAs or siRNAs targeting PUM, PABPN1 or Scrambled control sequences. (D) Western blots of PABPN1 and TUB from MCF7 cells treated with siRNAs targeting PABPN1 or Scrambled control sequences. (E) RT-qPCR results from RPE cells transfected an empty control vector (pCDNA) or one containing CSTF3.
https://openalex.org/W4226100021	http://wrap.warwick.ac.uk/168447/1/WRAP-peculiar-double-consciousness-TWAIL-22.pdf	English	null	The Peculiar Double-Consciousness of TWAIL	Social Science Research Network	2,022	cc-by	11,785	Manuscript version: Published Version The version presented in WRAP is the published version (Version of Record). The version presented in WRAP is the published version (Version of Record). Persistent WRAP URL: http://wrap.warwick.ac.uk/168447 How to cite: How to cite: The repository item page linked to above, will contain details on accessing citation guidance from the publisher. Manuscript version: Published Version Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes this work of researchers of the University of Warwick available open access under the following conditions. This article is made available under the Creative Commons Attribution 4.0 International license (CC BY 4.0) and may be reused according to the conditions of the license. For more details see: http://creativecommons.org/licenses/by/4.0/. For more information, please contact the WRAP Team at: wrap@warwick.ac.uk warwick.ac.uk/lib-publications Indonesian Journal of International Law Indonesian Journal of International Law Volume 19 Number 2 Third World Approaches to International Law Article 3 The Peculiar Double-Consciousness of TWAIL The Peculiar Double-Consciousness of TWAIL Mohsen Al Attar Warwick University, United Kingdom Follow this and additional works at: https://scholarhub.ui.ac.id/ijil Part of the International Law Commons Recommended Citation Recommended Citation Al Attar, Mohsen (2022) "The Peculiar Double-Consciousness of TWAIL," Indonesian Journal of International Law: Vol. 19: No. 2, Article 3. DOI: 10.17304/ijil.vol19.2.3 Available at: https://scholarhub.ui.ac.id/ijil/vol19/iss2/3 This Article is brought to you for free and open access by the Faculty of Law at UI Scholars Hub. It has been accepted for inclusion in Indonesian Journal of International Law by an authorized editor of UI Scholars Hub. Indonesian Journal of International Law (2022), Vol. 19 No. 2, pp. 239-262 https://doi.org/10.17304/ijil.vol19.2.3 THE PECULIAR DOUBLE-CONSCIOUSNESS OF TWAIL Mohsen Al-Attar Warwick University Correspondence: mohsen.al-attar@warwick.ac.uk 1 Mohsen al Attar, “TWAIL: A Paradox within a Paradox,” International Community Law Review 22 (2020): 163. 2 Antony Anghie and Bhupinder S. Chimni, “Third World Approaches to International Law and Individual Responsibility in Internal Conflicts,” Chinese Journal of International Law 2, no. 1 (2003): 77–103. Abstract The Eurocentrism colouring much mainstream scholarship is shunned by TWAIL, which has centred on international legal scholarship regarding the views of people historically marginalised by imperialism. In a single generation, TWAIL interventions have shifted the academia’s perceptions of international law. This forces scholars to account for the partialities of the purportedly universal international legal regime. However, TWAIL embraces iconoclasm and critique, and is also denoted by a formalist streak. Many of its scholars are committed to the regime, towing a counter-intuitive and inconsistent orthodox line. This study used WEB Du Bois’ concept of double-consciousness to explain the tension gripping TWAIL scholars. The formerly colonised are caught in a loop, aspiring to belong to the academy and the world. However, they recognise the discord between their emancipatory hope for international law and its predatory reality. Du Bois beseeches the colonised to fuse their clashing worldviews to produce a radical consciousness that advances human freedom. Keywords : double-consciousness, epistemology, TWAIL, Third World imaginaries, W. E. B. du Bois Submitted : 14 March 2022 \| Revised : 9 June 2022 \| Accepted : 21 June 2022 Submitted : 14 March 2022 \| Revised : 9 June 2022 \| Accepted : 21 June 2022 Copyright © 2022 – Mohsen Al-Attar Published by Lembaga Pengkajian Hukum Internasional I. INTRODUCTION TWAIL scholars seek to subvert international law, proffering counter- narratives and critical analyses that question or collapse the regime’s foundations. The scholars also rescue the regime by professing an optimism in international law that supports the hagiographical accounts common to orthodox scholarship. Previous studies described this dynamic as an irreconcilable paradox.1 TWAIL scholars are not naïve to the contradictions because they embrace half-conformity. They are sensitive to the risks of abandoning international law to positivists, naturalists, and formalists.2 The scholars have also enjoyed success in their critique-cum-rescue stance. Furthermore, Copyright © 2022 – Mohsen Al-Attar Published by Lembaga Pengkajian Hukum Internasional Mohsen Al-Attar TWAIL has prompted a re-think of international law by chastening orthodox. The contradictions and inconsistencies produced by these competing loyalties compel people to accept a cognitive pretzel to advance Third World liberation through international law. Reflections on TWAIL’s character and purpose have grown since the emergence of the paradox argument. Subsequently, double-consciousness is a more useful metonym when deliberating the TWAIL scholars’ attitude toward international law. Du Bois explained in the Soul of Black Folks that the oppressed are cursed and blessed by a double consciousness.3 They engage the world through their oppressors’ eyes because of colonialism’s powerful pedagogical force. The oppressed even internalise certain colonial precepts and values in their struggle for emancipation.4 The colonised have their experiences, lives, and ontologies. These are grounded in oppression and resistance, as well as in the history, culture, and episteme of Third World peoples. The oppressed relate to the oppressor’s normative contexts but they do not abandon their dreams of freedom, even when they appear idiosyncratic to others. Du Bois does not treat a double-consciousness as a deficiency despite the tension it generates. Conversely, it nourishes Third World imaginaries, stimulating a richer understanding of the struggle for re-humanisation. Double- consciousness allows the colonised to capitalise on their distinct worldview to advance a radical and unitary consciousness. W. E. B. Du Bois’ device helps explore the concurrent commitments prevailing in the TWAIL universe.5 TWAIL scholars belong and wish to belong to the international community and its associated order. Additionally, they want to enjoy freedom guided by their norms and in pursuit of their aspirations.6 This double-consciousness is evident across the gamut of TWAIL scholarship. Its interlocutors are committed to the regime as constituted, but also to an ideal, as they imagined. 3 W. E. B. Du Bois and Jonathan Scott Holloway, The Souls of Black Folk, First Yale University Press edi tion (New Haven: Yale University Press, 2015), 5. 4 Reiland Rabaka, Du Bois: A Critical Introduction, Key Contemporary Thinkers (Cambridge, UK; Med ford, MA, USA: Polity Press, 2021), 60. 5 Du Bois and Holloway, The Souls of Black Folk, 5. 6 James Thuo Gathii, “The Promise of International Law: A Third World View,” American University International Law Review 36, no. 3 (2021): 102. 7 “Critical Pedagogy Symposium: Critical Thinking and Teaching as Common Sense – Random Reflec tions,” Opinio Juris (blog), Antony Anghie, accessed 31 August 2020, http://opiniojuris.org/2020/08/31/ critical-pedagogy-symposium-critical-thinking-and-teaching-as-common-sense-random-reflections/. 4 Reiland Rabaka, Du Bois: A Critical Introduction, Key Contemporary Thinkers (Cambridge, UK; Me ford, MA, USA: Polity Press, 2021), 60. 3 W. E. B. Du Bois and Jonathan Scott Holloway, The Souls of Black Folk, First Yale University Press edi tion (New Haven: Yale University Press, 2015), 5. 4 Reiland Rabaka, Du Bois: A Critical Introduction, Key Contemporary Thinkers (Cambridge, UK; Med ford, MA, USA: Polity Press, 2021), 60. 5 Du Bois and Holloway, The Souls of Black Folk, 5. 6 James Thuo Gathii “The Promise of International Law: A Third World View ” American University I. INTRODUCTION This might not have been evident in an essay by Antony Anghie for the international law blog Opinio Juris. Although Antony Anghie professes commitment to “the classical positivist approach”, he urges his students to familiarise themselves with its disadvantages for Third World states.7 Until a better frame comes along, Third World law students must 240 The Peculiar Double-consciousness of TWAIL master the doctrine to produce “better outcomes”.8 While rejecting the racial hierarchy that generates an illiberal dialectic of superiority and inferiority, they cherish the equality tenet entrenched in the liberal promise. In what is perhaps In the most quoted passage of The Souls of Black Folk, Du Bois stated: “It is a peculiar sensation, this double-consciousness, this sense of always looking at one’s self through the eyes of others, of measuring one’s soul by the tape of a world that looks on in amused contempt and pity.”9 Similar to other TWAIL scholars, Anghie are familiar with the international law brutalities for Third World people. They are conscious that they are looking at themselves through the eyes of others. However, it remains the instrumentality of choice, guiding their thinking and being as scholars, pedagogues, and people. The next section describes the terms of the order in international law. These terms produce disciplinary boundaries, but their epistemological character stretches beyond simple study frames. Moreover, this study examined TWAIL scholarship, dividing the account into its early days as a counter-narrative and its blossoming into an epistemic exploration. The final section discusses Du Bois’ double-consciousness. This could become TWAIL’s greatest asset, as proclaimed by the radical sociologist. TWAIL considers scholarship a way of thinking, resisting, and being, has contributed to the re-humanisation of Third World people in their eyes and those of their former and current oppressors. Therefore, it lays the foundation for the radical unitary consciousness recommended by Du Bois for the anti-colonial. y f 10 Doreen Massey, “Negotiating Disciplinary Boundaries,” Current Sociology 47, no. 4 (999): 6. 11 Ibid, 6. 8 Ibid. 9 Du Bois and Holloway, The Souls of Black Folk, 5. 10 Doreen Massey, “Negotiating Disciplinary Boundaries,” Current Sociology 47, no. 4 (999): 6. 11 Ibid, 6. 8 Ibid. 9 Du Bois and Holloway, The Souls of Black Folk, 5. 9 Du Bois and Holloway, The Souls of Black Folk, 5 II. DISCIPLINARY OR EPISTEMOLOGICAL BOUNDAR IES? THE INTERNATIONAL LEGAL FRAME Academic disciplines operate within constructed and policed boundaries, and most are dominated by a few theoretical frameworks and conceptual ideas.10 Scholars invest resources into researching, developing, and verifying what they consider plausibility in their fields. Plausibility gradually morph into truths and commitments perceived as the natural order of things. The academics’ commitments outline their fields, producing disciplinary boundaries without which the study could become incoherent or illogic. Disciplinary boundaries require the fusion of internal and external historiographical approaches.11 Scholars adopt internal parameters to unify the methods and theories necessary to grow coherent knowledge. This results 241 Mohsen Al-Attar in a scholarly community, with each parameter contributing to expanding the conceptualised debates and knowledge pool. Furthermore, sensitivity to the external is imperative. Scholars monitor the production of knowledge beyond their boundaries to account for the evolutionary and trans-disciplinary scientific development. Engaging with external parameters ensures scholars nurture the discipline for context and impact beyond itself. Subsequently, all boundaries enable scholars to maintain the relevance of their discipline. Immanuel Wallerstein stated that people define a discipline by what lies beyond it.12 This implies the difficulties surrounding, containing, and constraining the boundaries. The difficulties circumscribe scientific development within an imagined locale, population, period, and epistemic tradition. Scholars inoculate disciplinary knowledge against entropy by probing supportive axioms and mining new ones. Establishing disciplinary boundaries is recognisable and prudent, but it comes at a cost, specifically narrowing and erasing perspectives. Disciplinary boundaries exclude the viewpoints of those that do not fit the frame, do not conform to it, or reject received logic. Since frames claim the truth that generates commitments, a more troubling outcome is the diminishing of the truths and commitments of others. In this situation, disciplinary boundaries reproduce relations of influence and power.13 They protect a centre ground, pushing everything else to the periphery, sometimes beyond the field. For instance, applying critical and radical labels exorcises non-conforming scholars to the periphery, reinforcing a disciplinary hierarchy. Although some conventional scholars recognise heterodox scholarship as study, they treat it as inferior, fanciful, or eccentric, while others dismiss it altogether.14 Since academia is neither democratic nor representative, an obvious consequence most research fields suffer from severe forms of misrecognition (status inequality) and misrepresentation (participatory inequality).15 Non- European scholars were long prevented from commenting on international legal matters (misrepresentation). 12 Ibid. 13 Ibid.,12. 14 Brad R. Roth, “Governmental Illegitimacy and Neocolonialism: Response to Review by James Thuo Gathii,” Michigan Law Review 98, no. 6 (2000): 2056–65, doi: https://doi.org/10.2307/1290273. 15 Nancy Fraser, “Reframing Justice in a Globalising World,” New Left Review 36 (2005): 69. Brad R. Roth, Governmental Illegitimacy and Neocolonialism: Response to Review by James Gathii,” Michigan Law Review 98, no. 6 (2000): 2056–65, doi: https://doi.org/10.2307/1290273. 15 Nancy Fraser, “Reframing Justice in a Globalising World,” New Left Review 36 (2005): 69. II. DISCIPLINARY OR EPISTEMOLOGICAL BOUNDAR IES? THE INTERNATIONAL LEGAL FRAME When they were permitted into the academy, they could not draw upon non-European epistemologies when articulating their views (misrecognition). They had to hang forego their identities when discussing international legality. Consequently, orthodox scholars nurtured this perception, representing their studies as universal and themselves as neutral. Universality and neutrality manifest as virulent discrimination against the non-European jurist. They dissolve the import of culture in the conception of 242 The Peculiar Double-consciousness of TWAIL behaviour and its regulation. Moreover, discrimination in studies and science bolster the agent-subject dialectic that has coloured relationships between the metropolis and colonies since the European imperial era. Academics pursue studies because of passion, belief, expectations, and personal gain. These motives are not disinterested because the motivational context for producing scholarship is complex. Study could easily advance knowledge within a self-interested context. In the infamous torture memos, John Yoo rationalised his way around an absolute prohibition on torture.16 He produced the legal argument after his government committed the heinous acts. The self-proclaimed neutrality of science and scholarship exacerbates power differentials, as reflected in the TWAIL corpus. As oppression, Study is a morbid heading but fits in various fields, including anthropology, geography, history, and medicine. Since Europe universalised itself, its jurists have commanded a monopoly over the development of international law. In elsewhere, it is argued “From Vattel to Westlake to Yoo, centuries of international legal academics produce scholarship in service of Euro-American imperial ambition and might”17, and also “Their arguments structure the regime of international law in self- serving ways”.18 It is impossible to divorce contemporary order aspects, such as economic, geopolitical, and legal, from Europe’s imperial-colonial period. Imperial powers boasted that their civilisational project was devised to enlighten others. However, this gift of light was a fair trade-off when they could fill their coffers with the resources of others. These same jurists grounded the brutalities scarring international law with many examples. They deployed malleable notions of morality and civilisation in the early stages of international law’s development to rationalise slavery, colonisation, and genocide. Moreover, the jurists veered to legalistic and liberal arguments in subsequent times to support analogous schemes.19 The transatlantic slave trade, the colonisation of the Indian subcontinent, the genocide of the Herero and the Nama, and the counter-independence wars Europe waged in the 20th century were treated as extra-legal events. International law did not prohibit these practices, meaning the Europeans committed no legal wrong. 16 Clare Keefe Coleman, “Teaching the Torture Memos: Making Decisions under Conditions of Uncer tainty,” Journal of Legal Education 62 (2012): 81. 17 Mohsen al Attar, “Must International Legal Pedagogy Remain Eurocentric?,” Asian Journal of Interna tional Law 11, no. 1 (January 2021): 180, https://doi.org/10.1017/S2044251321000138. 18 Ibid, 180. 19 Ibid, 180. 20 Franziska Boehme, “Reactive Remembrance: The Political Struggle over Apologies and Reparations between Germany and Namibia for the Herero Genocide,” Journal of Human Rights 19, no. 2 (14 March 2020): 238–55, https://doi.org/10.1080/14754835.2020.1727729. 21 Brenna Bhandar, “Theft in Broad Daylight: Racism and Neoliberal Legality,” Law and Critique 32, no. 3 (November 2021): 288, doi: https://doi.org/10.1007/s10978-021-09308-9. 22 Babatunde Fagbayibo, “Some Thoughts on Centering Pan-African Epistemic in the Teaching of Public International Law in African Universities,” International Community Law Review 21, no. 2 (2019): 170–89, doi: https://doi.org/10.1163/18719732-12341397. 23 Mohsen al Attar, “Must International Legal Pedagogy Remain Eurocentric?,” Asian Journal of Interna tional Law 11, no. 1 (January 2021): 180, doi: https://doi.org/10.1017/S2044251321000138, 197. 24 Fagbayibo, ‘Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities’, 172. 25 Ibid. 20 Franziska Boehme, “Reactive Remembrance: The Political Struggle over Apologies and Reparations between Germany and Namibia for the Herero Genocide,” Journal of Human Rights 19, no. 2 (14 March 2020): 238–55, https://doi.org/10.1080/14754835.2020.1727729. 23 Mohsen al Attar, “Must International Legal Pedagogy Remain Eurocentric?,” Asian Journal of Intern tional Law 11, no. 1 (January 2021): 180, doi: https://doi.org/10.1017/S2044251321000138, 197. 21 Brenna Bhandar, “Theft in Broad Daylight: Racism and Neoliberal Legality,” Law and Critique 32, n 3 (November 2021): 288, doi: https://doi.org/10.1007/s10978-021-09308-9. 24 Fagbayibo, ‘Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities’, 172. 25 Ibid 22 Babatunde Fagbayibo, “Some Thoughts on Centering Pan-African Epistemic in the Teaching of Pub International Law in African Universities,” International Community Law Review 21, no. 2 (2019): 170–8 doi: https://doi.org/10.1163/18719732-12341397. 24 Fagbayibo, ‘Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities’, 172. 25 Ibid. II. DISCIPLINARY OR EPISTEMOLOGICAL BOUNDAR IES? THE INTERNATIONAL LEGAL FRAME Tony Blair, the former British Prime Minister, expressed regret only for the enslavement of millions, confident in the extra- 243 Mohsen Al-Attar legality of the transatlantic slave trade. European international law did not prohibit slavery, colonisation, genocide, or plunder. It means that jurists or the law bore no responsibility for the savagery committed or the surviving legacies. Similar to Blair’s viewpoint, Germany insisted on its authority to decide how it valued the Herero and the Nama, the peoples it ethnically cleansed.20 Germany, the genocide perpetrator, also controls how Namibia uses the compensation. To borrow Brenna Bhandar’s astute quip, international law remains perpetually innocent.21 Until recently, this hagiographical account of international was the only take on the offer. In the Third World, international law students read Vitoria and Vattel, Grotius and Goethe, Westlake and Wesley, Lauterpacht and Locke.22 The absence of the involvement non-Europeans in the regime’s constitution was inconsequential. Hagiography is useful when beneficiaries of the past seek to justify the present. Therefore, misrecognition and misrepresentation do not undermine the legitimacy of the scientific universalist claims. Many leading textbooks and scholarship on international law reproduce this subjective record. According to the Teaching and Researching International Law in Asia report, law faculties in Asia depend on the same textbooks prevailing in European universities.23 In the few instances where they rely on a local text, it is a translation rather than original scholarship. Similarly, Babatunde Fagbayibo stated that teaching public international law in Africa remains unresponsive to the imperative of decolonization.”24 “The curriculum in many universities across the continent remains steeped in Eurocentric cannons and does little to disrupt hegemonic assumptions that place European thinkers at the heart of the development of international law.”25 Also, only recently have Third World jurists compiled state practice across the Asian and African continents, consistently treating European subjectivity as human objectivity. As a student and scholar, I recall being perplexed by the Eurocentrism 244 The Peculiar Double-consciousness of TWAIL of international law. Why did we not study non-European histories and viewpoints if the regime was international? Reading history made me aware of the lengthy civilizational records in regions beyond Europe. How did these people conduct relations with others? What was the dynamic between co- existing civilizations? Furthermore, because of my anti-colonial disposition, I questioned how European international law responded to historic and ongoing brutalities committed against non-European peoples. Alas, the textbooks on offer were silent. 26 Mohsen al Attar and Vernon Tava, “TWAIL Pedagogy - Legal Education for Emancipation,” Palestine Yearbook of International Law 15 (2009): 7–40. 27 George Galindo, “Splitting TWAIL,” Windsor Yearbook of Access to Justice 33, no. 3 (2016): 37–56; Karin Mickelson, “Taking Stock of TWAIL Histories,” International Community Law Review 10, no. 4 (2008): 355–62. 28 Obiora Chinedu Okafor, “Critical Third World Approaches to International Law (TWAIL): Theory, Methodology, or Both?,” International Community Law Review 10, no. 4 (2008): 371–78. 29 Karin Mickelson, “Taking Stock of TWAIL Histories,” International Community Law Review 10, no. 4 (2008); Karin Mickelson, “Rhetoric and Rage: Third World Voices in International Legal Discourse,” Wisconsin International Law Journal 16 (1997): 353. 30 Philip James Drake, Beyond the Bounds of Formalism: Social Justice and Legal Education (UK: Hud dersfield, 2020). II. DISCIPLINARY OR EPISTEMOLOGICAL BOUNDAR IES? THE INTERNATIONAL LEGAL FRAME European scholarship was meant to suffice, which often felt like asking students to grade their homework. It thus delighted me to learn of the scholarly movement now known as TWAIL, and I was grateful for the vistas it revealed.26 Obiora Chinedu Okafor, “Critical Third World Approaches to International Law (TWAIL): Theory, ethodology, or Both?,” International Community Law Review 10, no. 4 (2008): 371–78. George Galindo, “Splitting TWAIL,” Windsor Yearbook of Access to Justice 33, no. 3 (2016): 37–56; rin Mickelson, “Taking Stock of TWAIL Histories,” International Community Law Review 10, no. 4 008): 355–62. 26 Mohsen al Attar and Vernon Tava, “TWAIL Pedagogy - Legal Education for Emancipation,” Palesti Yearbook of International Law 15 (2009): 7–40. III. GAZING (FROM) BEYOND THE BOUNDARIES TWAIL has been around for a single generation, depending on the interlocutor’s preferred approach toward periodisation.27 TWAIL was a cri de guerre before it was a theory and developed its methods to become chic.28 Led by an insurgent and irreverent community of scholars, TWAIL pursued varied forms of subversion to liberate international legal scholarship from its Eurocentric straitjacket and use the academic platform to advance the emancipation of Third World peoples.29 For the legal academy, the birth of TWAIL was transcendental, bringing down the curtain on white supremacist mythmaking about jus gentium. Law is an academic discipline operating within the ambit of conservatism, which is understandable considering its dual role in academia. It prepares the next cadre of legal practitioners.30 To achieve this goal, law schools engage with the regulatory regime as a component in the state system governed, ordered, and coerced by the law. Any law school that cannot familiarise students with law’s operational matrix does a great disservice, graduating individuals ill- equipped to grapple with legal disputes. Consequently, jurists and publicists 245 Mohsen Al-Attar are often faithful formalists and universalists.31 They believe in the rules, even when they propose to use their expertise to bend or skirt those rules. In this case, due process, procedures, and bureaucracy are beloved bedfellows, a tendency that holds whether the scholar is progressive or reactionary. The universalising ethos of the framework besots even the most committed human rights advocate, snubbing the inevitable relativism upon which human dignity leans.32 TWAIL scholars adopt a different outlook, where the law is the highest normative and coercive instrument. To only account for the law’s regulatory purpose is an equally perverse dereliction of duty. Students fail to grasp the power dynamics that shape the regime that preserves. This creates the struggle between law as natural order that reifies the status quo and a contingent proposition that stimulates critical inquiry. According to TWAIL, once the law is stripped of its holier-than-thou halo, people are left with a system of command and structure most comfortable when drawing swords to excuse the past and defend the status quo. The geopolitical prejudice behind this impetus is glaring and galling. At the civilisational level, European jurists have committed to codifying norms and rules of behaviour. They ignore the influence of their morality and material interests upon their preferences. 31 Ibid. 32 Kajit John Bagu, “Ideological Refuge v Jurisprudence of Insurgency: Cultural, Relativism and Univer salism in the Human Rights Discourse,” Warwick Student Law Review 1 (2011): 1–18. 33 Mohsen al Attar, “Subverting Eurocentric Epistemology: The Value of Nonsense When Designing Coun terfactuals,” Contingency in International Law: On The Possibilities of Different Legal Histories, ed. Ingo Venzke and Kevin Heller (Oxford University Press, 2021); James Thuo Gathii, “International Law and Eurocentricity,” European Journal of International Law 9, no. 1 (1 January 1998): 184–211, doi: https:// doi.org/10.1093/ejil/9.1.184. 34 Antony Anghie, Imperialism, Sovereignty, and the Making of International Law, Cambridge Studies in International and Comparative Law (Cambridge University Press, 2004). III. GAZING (FROM) BEYOND THE BOUNDARIES Orthodox scholars had proven more responsive to the TWAIL critique only when Third World scholars revealed and rebelled against the Eurocentric presumptions upon which they fashioned the regime.33 What is the TWAIL critique? It is layered, varied, and variable. The two aspects foundational to the movement are the counter-narrative and epistemic renewal. Each method and its implications for understanding international law are captured in two examples. In Imperialism, Sovereignty, and the Making of International Law, Antony Anghie collapsed the moral foundations that European international law boasted of.34 Rather than notions of universality and equality, the regime’s architects were guided by a dynamic of difference and of dehumanisation. Much of the regime was structured with a hierarchical view of civilisation, producing the codification of European normative preferences and the denigration of those 246 The Peculiar Double-consciousness of TWAIL of others. Moreover, Anghie examined the current geopolitical relations, highlighting how demonisation informs modern legality. In his seminal book, Anghie used the Euro-American response to the bombing of the World Trade Centre and their willingness to rewrite swathes of international legal practice. He aimed to bolster his claim about the persistence of this predatory dynamic (Yoo). Jus ad bellum and jus in bello are ornamental by classifying the other side as sub-human. Ten years later, James Gathii published War, Commerce, and International Law35 Gathii centred the powerful Euro-American interests that manipulated war and international legality in the name of mammon. Echoing Susan Marks, Gathii showed that immiseration is good business and international law is a powerful ally when imperial states wish to confiscate the wealth of others.36 Similar to Anghie, Gathii relied upon historical examples to show the amenability of international law to promoting First World economic interests. He also examined his claim through a contemporary lens, focusing on the American invasion of Iraq in 2003 and the law reforms implemented by American agents. Many changes to oil and gas, intellectual property, and financial laws advanced the interests of American companies contracted for the reconstruction of Iraq post-invasion. These same agents espouse views years after Iraq’s descent into chaos. According to Paul Bremer, the agents believe that the Iraqi people are infinitely better off currently from under Saddam, even after all the problems. They have the men and women of the American armed forces and elsewhere to thank for it.37 This shows that imperialism is an endless battleground. 37 CBS News, “Paul Bremer, Who Helped Run Iraq Occupation, Is Now a Ski Instructor,” News, CBS News, accessed 7 March 2022, https://www.cbsnews.com/news/paul-bremer-who-helped-run-iraq-occupa tion-is-now-a-ski-instructor/. 36 Susan Marks, “Human Rights and the Bottom Billion,” European Human Rights Law Review, no (2009): 37–49. 39 Michelle Burgis-Kasthala, “Scholarship as Dialogue? TWAIL and the Politics of Methodology,” Journ of International Criminal Justice 14, no. 4 (2016): 921–38, doi: https://doi.org/10.1093/jicj/mqw044. 35 James Thuo Gathii, War, Commerce, and International Law (Oxford University Press, 2010). 38 Raúl Alberto Mora, “Counter-Narrative,” Key Concepts in Intercultural Dialogue 36, 2014. 39 Mi h ll B i K h l “S h l hi Di l ? TWAIL d h P li i f M h d l ” J 35 James Thuo Gathii, War, Commerce, and International Law (Oxford University Press, 2010). 36 Susan Marks, “Human Rights and the Bottom Billion,” European Human Rights Law Review, no. 1 (2009): 37–49. 37 CBS News, “Paul Bremer, Who Helped Run Iraq Occupation, Is Now a Ski Instructor,” News, CBS News, accessed 7 March 2022, https://www.cbsnews.com/news/paul-bremer-who-helped-run-iraq-occupa tion-is-now-a-ski-instructor/. 38 Raúl Alberto Mora, “Counter-Narrative,” Key Concepts in Intercultural Dialogue 36, 2014. 39 Michelle Burgis-Kasthala, “Scholarship as Dialogue? TWAIL and the Politics of Methodology,” Journal of International Criminal Justice 14, no. 4 (2016): 921–38, doi: https://doi.org/10.1093/jicj/mqw044. 44 Bhupinder S. Chimni, “Asian Civilizations and International Law: Some Reflections,” Asian Jour nal of International Law 1, no. 1 (2011): 39–42; Rowland J. V. Cole, “Africa’s Approach to Internation al Law: Aspects of the Political and Economic Denominators,” African Yearbook of International Law Online / Annuaire Africain de Droit International Online 18, no. 1 (2010): 287–310, doi: https://doi. org/10.1163/22116176-01801011. 42 Hiroshi Fukurai, “Original Nation Approaches to “Inter- National” Law (ONAIL): Decoupling of the Nation and the State and the Search for New Legal Orders,” Indiana Journal of Global Legal Studies 26, no. 1 (2019): 199–262; Ogba Adejoh Sylvester and Okpanachi Idoko Anthony, “Decolonization in Africa and Pan-Africanism,” Yönetim Bilimleri Dergisi 12, no. 23 (2014); Amy Maguire, “Contemporary Anti- Colonial Self-Determination Claims and the Decolonisation of International Law,” Griffith Law Review 22, no. 1 (2013): 238–68, doi: https://doi.org/10.1080/10383441.2013.10854774. 40 Jan Jansen, “Politics of Remembrance, Colonialism and the Algerian War of Independence in France’, in A European Memory? Contested Histories and Politics of Remembrance, by Malgorzata Pakier and Bo Stråth,” Studies in Contemporary European History (New York: Berghahn Books, 2010), 287. ( ) , p g 43 S.S.M.W. Seneviratne and K.A.A.N. Thilakarathna, “Recognition of New States: An Analysis of Se lected Case Studies from a Third World Approach to International Law (TWAIL)” (Colombo, 2020).l 41 Sundhya Pahuja, Decolonising International Law: Development, Economic Growth and the Politics of Universality (Cambridge University Press, 2011). 40 Jan Jansen, “Politics of Remembrance, Colonialism and the Algerian War of Independence in France’, in A European Memory? Contested Histories and Politics of Remembrance, by Malgorzata Pakier and Bo Stråth,” Studies in Contemporary European History (New York: Berghahn Books, 2010), 287. 41 Sundhya Pahuja, Decolonising International Law: Development, Economic Growth and the Politics of Universality (Cambridge University Press, 2011). 42 Hiroshi Fukurai, “Original Nation Approaches to “Inter- National” Law (ONAIL): Decoupling of the Nation and the State and the Search for New Legal Orders,” Indiana Journal of Global Legal Studies 26, no. 1 (2019): 199–262; Ogba Adejoh Sylvester and Okpanachi Idoko Anthony, “Decolonization in Africa and Pan-Africanism,” Yönetim Bilimleri Dergisi 12, no. 23 (2014); Amy Maguire, “Contemporary Anti- Colonial Self-Determination Claims and the Decolonisation of International Law,” Griffith Law Review 22, no. 1 (2013): 238–68, doi: https://doi.org/10.1080/10383441.2013.10854774. 43 S.S.M.W. Seneviratne and K.A.A.N. Thilakarathna, “Recognition of New States: An Analysis of Se lected Case Studies from a Third World Approach to International Law (TWAIL)” (Colombo, 2020). 44 Bhupinder S. Chimni, “Asian Civilizations and International Law: Some Reflections,” Asian Jour nal of International Law 1, no. 1 (2011): 39–42; Rowland J. V. Cole, “Africa’s Approach to Internation al Law: Aspects of the Political and Economic Denominators,” African Yearbook of International Law Online / Annuaire Africain de Droit International Online 18, no. 1 (2010): 287–310, doi: https://doi. org/10.1163/22116176-01801011. ( ) , p g 43 S.S.M.W. Seneviratne and K.A.A.N. Thilakarathna, “Recognition of New States: An Analysis of Se lected Case Studies from a Third World Approach to International Law (TWAIL)” (Colombo, 2020). 44 Bhupinder S. Chimni, “Asian Civilizations and International Law: Some Reflections,” Asian Jour nal of International Law 1, no. 1 (2011): 39–42; Rowland J. V. Cole, “Africa’s Approach to Internation III. GAZING (FROM) BEYOND THE BOUNDARIES TWAIL verifies the reflections on international law that originate outside the Eurosphere, accounting for Third World stories when telling the narrative.44 Its databases 248 The Peculiar Double-consciousness of TWAIL comprise hundreds of articles that crash against international legal orthodoxy.45 Although a domineering Eurocentric epistemology constrained most, a few broke free. They probed the implications of non-European thought for the future of international legality. This leads to epistemic renewal, a second method vital to the TWAIL critique.46 Across the centuries, Third World peoples opposed the world- making ambitions of European international law. TWAIL is a praxis of re- remembering and re-humanisation, allowing Third World peoples to honour their dead while forging space for their living. This was a natural outgrowth of the counter-narrative method. Mora asserts that “the effect of a counter- narrative is to empower and give agency to those communities.”47 “Choosing their own words and telling their own stories” allows marginalised communities to provide alternative perspectives, helping to create complex narratives that present their realities.48 These complex realities include distinct worldviews, aspirations, and epistemologies, which revival is also part of the TWAIL method. Recent TWAIL scholars excavated Third World resistance to the epistemic violence that circumscribes and potentially improves international law. This leads to epistemic renewal, a second method vital to the TWAIL critique.46 Across the centuries, Third World peoples opposed the world- making ambitions of European international law. TWAIL is a praxis of re- remembering and re-humanisation, allowing Third World peoples to honour their dead while forging space for their living. This was a natural outgrowth of the counter-narrative method. Mora asserts that “the effect of a counter- narrative is to empower and give agency to those communities.”47 “Choosing their own words and telling their own stories” allows marginalised communities to provide alternative perspectives, helping to create complex narratives that present their realities.48 These complex realities include distinct worldviews, aspirations, and epistemologies, which revival is also part of the TWAIL method. Recent TWAIL scholars excavated Third World resistance to the epistemic violence that circumscribes and potentially improves international law. Examples are spreading, and two texts explain the impetus toward epistemic renewal similar to the counter-narrative. Fagbayibo is leading the charge on a Pan-African episteme.49 It aims to reform legal pedagogy and account for African epistemologies when teaching international law across Africa and beyond. 45 James Thuo Gathii, “TWAIL: A Brief History of Its Origins, Its Decentralized Network, and a Tentative Bibliography,” Trade, Law and Development 3, no. 1 (2011): 26–64. 46 Pooja Parmar, “TWAIL: An Epistemological Inquiry,” International Community Law Review 10, no. 4 (2008): 363–70. 47 Raúl Alberto Mora, “Counter-Narrative,” Key Concepts in Intercultural Dialogue 36, 2014. 48 Ibid. 49 Babatunde Fagbayibo, ‘Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities’, International Community Law Review 21, no. 2 (2019): 170–89, doi: https://doi.org/10.1163/18719732-12341397. 50 Ibid, 185. III. GAZING (FROM) BEYOND THE BOUNDARIES Anghie and Gathii practiced the counter-narrative method favoured in TWAIL scholarship. Raúl Alberto Mora defines counter-narratives as “narratives that arise from the vantage point of those historically marginalized.”38 Orthodox scholars neither told nor reflected on international law from the context of the colonised, the enslaved, or the massacred. These viewpoints exist beyond their disciplinary boundaries, hence do not elicit scholarly curiosity, at least not within the centres of power.39 According to Anghie, the millions massacred by Europe are central to the story of international law. He mentioned the names of the Arawak and the Taino, of 247 Mohsen Al-Attar the Third World civilians, annihilated by American, English, French, and German weapons of mass destruction. Anghie also mentioned the hundreds of thousands of Algerians whose nameless memories fill La Seine, floating within sight of the pulpit at Notre Dame.40 Anghie centred European international law’s complicity in the demonic suffering Europe inflicted upon Third World people. Similarly, Gathii expanded on the predatory economic interests that informed this depraved behaviour. The violence and legal interventions, such as Bremer’s Iraqi law reform package, were intentional. International law was a crafty instrument when facilitating programs of dispossession. Until the lions have their historians, the hunt will always be told from the hunter’s perspective. Reflecting Chinua Achebe’s aphorism, TWAIL established a platform where Third World peoples are not disappeared into the netherworld of extra-legality. Subsequently, its scholars adopted the counter-narrative to subvert European international law-making’s hitherto parochial and dehumanising ethos. The interlocutors of international law’s history spoke about and to Third World people born into law, as stated by Pahuja.41 Although they could comply with or contest their representations, European states and publicists devised the order. Newly independent states were to pledge fealty to the same frame that denied their rationality, normativity, or humanity, even during formal decolonisation.42 This demand for orthodoxy epitomised the regime’s continued commitment to Eurocentrism. Therefore, Third World states must cohabitate with imperial predators and prostrate before them, a hazing ceremony recast as the doctrine of recognition to gain acceptance.43 Anghie, Gathii, and other TWAIL scholars used the counter-narrative method to expose the predation within European international law. 49 Babatunde Fagbayibo, ‘Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities’, International Community Law Review 21, no. 2 (2019): 170–89, doi: https://doi.org/10.1163/18719732-12341397. 50 Ibid, 185. 5 Ibid, 187. 52 Hiroshi Fukurai, “Fourth World Approaches to International Law (FWAIL) and Asia’s Indigenous Strug gles and Quests for Recognition under International Law,” Asian Journal of Law and Society 5, no. 1 (May 2018): 221–31, doi: https://doi.org/10.1017/als.2018.10. 53 Ibid 224 , 55 Mohsen al Attar, “Subverting Eurocentric Epistemology: The Value of Nonsense When Designing Coun terfactuals,” Contingency in International Law: On The Possibilities of Different Legal Historie, 185–88. 56 James T Gathii, “Writing Race and Identity in a Global Context: What CRT and TWAIL Can Learn From Each Other,” UCLA Law Review, no. 67 (2020). III. GAZING (FROM) BEYOND THE BOUNDARIES He proposes a “critical integrative approach” that builds upon the Afrocentricity movement “to widen the knowledge base of students by dispelling the myths of ‘universalism.”50 Scholars should thus expose students to the historical contributions Africa made to international law, while adopting a trans-disciplinary approach toward its teaching. A trans- disciplinary approach coheres better with African epistemologies, where the isolation of fields is inconsistent with their preferred pedagogical approach. Fagbayibo also proposed understanding international law materials such as plays, art exhibitions, music, novels, poems, and films. This helps convey the social imaginaries that inspire African peoples and overcome global order 249 Mohsen Al-Attar Mohsen Al-Attar 51 Ibid, 187. 52 Hiroshi Fukurai, “Fourth World Approaches to International Law (FWAIL) and Asia’s Indigenous Strug gles and Quests for Recognition under International Law,” Asian Journal of Law and Society 5, no. 1 (May 2018): 221–31, doi: https://doi.org/10.1017/als.2018.10. 53 Ibid, 224. 54 Ibid, 227. 55 Mohsen al Attar, “Subverting Eurocentric Epistemology: The Value of Nonsense When Designing Coun terfactuals,” Contingency in International Law: On The Possibilities of Different Legal Historie, 185–88. 56 James T Gathii, “Writing Race and Identity in a Global Context: What CRT and TWAIL Can Learn From Each Other,” UCLA Law Review, no. 67 (2020). asymmetry.51 Hiroshi Fukurai stated the need for Fourth World Approaches to International Law52 while celebrating TWAIL for explaining “the predatory role of international law against the interests and aspirations of the Third World countries.”53 Praise notwithstanding, he also chastises TWAIL for its state-centrism and for dismissing the concerns of communities of indigenous peoples corralled within the state system, termed the Fourth World. TWAIL was promoted to adopt more inclusive conceptions of nation and sovereignty to support the cause of sub-nation communities. It was also recommended bifurcating nation and state to recognize that a collective and united identity manifests without the legal construction of the state.54 Additionally, Fukurai proposed revisiting standing to facilitate the participation of indigenous peoples in debates about international law. The environmental crisis was used to highlight its contribution toward protecting ecological biodiversity. Malcolm Shaw and Jan Klabbers showed commendable inclusivity in their textbooks, acknowledging critical approaches and making them part of the narrative of international law. However, European positivism and naturalism could not narrate the stories of the Third World’s relationship to international law,55 forcing TWAIL to make this banality obvious. Orthodox publicists follow the great-men-of-history model of scholarship, centring a white Christian European state as the main protagonist. In contrast, TWAIL tells the story of international law from the victims’ perspective, while Fagbayibo and Fukurai proposed epistemic renewal. Non-Europeans are erased from the international legal narrative, followed by their worldviews being censored. Fagbayibo and Fukurai were driven by cultural recognition and epistemological equivalency to question the aetiological commitments of international law and TWAIL. In a single generation, TWAIL has made it uncomfortable for the white supremacist version of international law to roam unhindered.56 TWAIL no longer enjoys the freedom of the academy, and is confronted by people every time it cleanses itself of the blood of ancestors, contemporaries, and unborn. Because of TWAIL, teaching international law without European histories of predation and Third World records of resistance and without contemplating 250 The Peculiar Double-consciousness of TWAIL non-European epistemologies is to propagandise rather than educate.57 non-European epistemologies is to propagandise rather than educate.57 The penultimate section states that both strands of TWAIL scholarship exist in tandem but are in conflict. Although TWAIL scholarship is not paradoxical, the scholars are trapped in a double consciousness. 61 Karin Mickelson, “Taking Stock of TWAIL Histories,” International Community Law Review 10, no. 4 (2008): 356. “In the process of carrying out my research, I found that none of the existing literature seemed to provide an adequate account of the approach taken by developing countries to international environmen 57 Balakrishnan Rajagopal, “Counter-Hegemonic International Law: Rethinking Human Rights and Devel opment as a Third World Strategy,” Third World Quarterly 27, no. 5 (July 2006): 767–83, doi: https://doi. org/10.1080/01436590600780078. 58 Darryl C. Thomas, “The Black Radical Tradition - Theory and Practice: Black Studies and the Scholarship of Cedric Robinson,” Race & Class 47, no. 2 (October 2005): 1–22, doi: https://doi. org/10.1177/0306396805058077. 59 Alyson L. Mahar, Virginie Cobigo, and Heather Stuart, “Conceptualizing Belonging,” Disability and Rehabilitation 35, no. 12 (June 2013): 1026–32, doi: https://doi.org/10.3109/09638288.2012.717584. 60 Ibid. 61 Karin Mickelson, “Taking Stock of TWAIL Histories,” International Community Law Review 10, no. 4 (2008): 356. “In the process of carrying out my research, I found that none of the existing literature seemed to provide an adequate account of the approach taken by developing countries to international environmen Alyson L. Mahar, Virginie Cobigo, and Heather Stuart, “Conceptualizing Belonging,” Disability and habilitation 35, no. 12 (June 2013): 1026–32, doi: https://doi.org/10.3109/09638288.2012.717584. Ibid 57 Balakrishnan Rajagopal, “Counter-Hegemonic International Law: Rethinking Human Rights and Dev opment as a Third World Strategy,” Third World Quarterly 27, no. 5 (July 2006): 767–83, doi: https://d org/10.1080/01436590600780078. Darryl C. Thomas, “The Black Radical Tradition - Theory and Practice: Black Studies and the holarship of Cedric Robinson,” Race & Class 47, no. 2 (October 2005): 1–22, doi: https://doi. g/10.1177/0306396805058077. IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL TWAIL is borne of a collective anti-colonial consciousness entangled in historical struggles for belonging and being.58 Scholars wish to belong to the academy of publicists, as seen in their participation in traditional activities. They also challenge many of the regime’s foundations, insisting it is inconsistent, parochial, and predatory. Although belonging and being are not mutually exclusive, they are not coaeval. Belonging involves feeling needed, important, integral, valued, respected, or in harmony with the group or system.59 As social creatures, each of these sentiments is critical to human self-perception and engagement with the world. These sentiments gain meaning by manifesting within a communal setting because connectedness and perceived cohesion with a group are key determinants in experiences of belonging.60 Subsequently, the behaviour of TWAIL scholars should not be surprising. They are international academic lawyers by trade and wish to connect with their cohort similar to other professionals. Acceptance and affirmation are also important since the academics’ profiles and reputations are contingent on standing before peers. Being is a more complex proposition. While survival standards are objective, the experience of being or existing is a personal affair. Moreover, since humans cannot be isolated, being is intertwined with culture and surroundings. Individuals experience culture differently, making it difficult to identify the variables upon which being rests. Humans understand the exclusion of their viewpoints meant a denial of being, precipitating the formation of TWAIL.61 251 Mohsen Al-Attar Mohsen Al-Attar The TWAIL network is grounded in the united recognition that humans need democratisation of international legal scholarship in two senses. (i) First, they need to contextualise international law’s privileging of European and North American voices. This requires providing institutional and imaginative opportunities for participation from the third world. (ii) Second, they need to criticise mainstream international law’s politics and scholarship for ensuring it helps reproduce structures that marginalise and dominate third-world people.62 As much as TWAIL scholars wished to belong to the academy, they acknowledged their existence beyond it and wished to fashion their space of flourishing. Struggles for belonging and being are omnipresent in TWAIL, as is the tension between them. They predominantly hail from regions with potent anti-colonial histories and are characterised by the same tensions identified by Fanon two generations ago. Moreover, struggles for belonging and being are trained legal scholars, most of whom studied former colonial powers’ legal systems and their institutions. , 63 Mohsen al Attar, “Reframing the “Universality” of International Law in a Globalizing World’,” McGill Law Journal 59, no. 1 (2013): 95, doi: https://doi.org/10.7202/1018986ar. 64 Jayan Nayar, “Thinking Being-Otherwise: Returning the Anti-Colonial to Philosophy,” Beyond Law and Development : Resistance, Empowerment and Social Injustice, ed. Abdul Paliwala (NY: Routledge, 2022). tal law.” 62 Ibid, 357–58. 63 Mohsen al Attar, “Reframing the “Universality” of International Law in a Globalizing World’,” McGill Law Journal 59, no. 1 (2013): 95, doi: https://doi.org/10.7202/1018986ar. 64 Jayan Nayar, “Thinking Being-Otherwise: Returning the Anti-Colonial to Philosophy,” Beyond Law and Development : Resistance, Empowerment and Social Injustice, ed. Abdul Paliwala (NY: Routledge, 2022). IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL Wang Tieya stated that Third World countries are adamantly opposed to the imperialistic, colonialist, oppressive, and exploitative principles and rules of traditional international law. However, they do not reject international law63 due to geopolitical, economic, and infrastructural reasons. This study focused on existential factors. Jayan Nayar commented on the non-Europeans’ denial of being during colonization, followed by the imposition of European subjectivity. According to Jayan Nayr, colonisation is the closure of the meaning of being within a bounded totality. It involves the transmogrification of already-being - through the philosophical negation and material subjugation of the encountered other-ed. These include non-being, or without a name, meaning, cosmologies, pasts - into a subject-being of colonial or ordering. It is a process that subjectifies the encountered ‘Other-ed’ and makes the other a subject of a totalised order of governmentality.64 This means that TWAIL scholars must embrace the bounded totality of international law to belong. However, this requires accepting a historical record rife with their dehumanisation. In Nayar’s language, TWAIL must submit to a colonial form of order, including their subjugation as non-beings. For Third World scholars, the cost of belonging is the surrender of the self. “It is a peculiar sensation, this double-consciousness,” declared Du Bois, 252 The Peculiar Double-consciousness of TWAIL “this sense of always looking at one’s self through the eyes of others, of measuring one’s soul by the tape of a world that looks on in amused contempt and pity.”65 Third World peoples have struggled for self-consciousness because of colonialism. They perpetually measure themselves against their existence, recast as underdevelopment, seeking to exorcise their culture and belonging from their being. According to Du Bois, Third World peoples undertake this task in a Eurocentric world where the most pervasive ideas about colonized peoples are contemptuous and degrading.66 The narrative routinely elevates misconceptions, misunderstandings, and stereotypes to a scientific level, adumbrating white superiority, and black inferiority. Du Bois used the racism of the dominant epistemology to explain the duality gripping the colonised. He stated that one ever feels his two-ness—an American, a Negro, two souls, thoughts, unreconciled strivings, warring ideals in one dark body, whose dogged strength keeps it from being torn asunder.67 As the racial hierarchy habituates white peoples to ideas about their superiority, black peoples are also conditioned and brutalised by their purported inferiority. , 69 Mohsen al Attar, “Subverting Racism in / through International Law Scholarship,” Opinio Juris (blog), accessed 3 March 2021, http://opiniojuris.org/2021/03/03/subverting-racism-in-international-law-scholar ship/; Anna-Spain Bradley, “International Law’s Racism Problem,” Opinio Juris (blog), accessed 4 Septem ber 2019, http://opiniojuris.org/2019/09/04/international-laws-racism-problem/; Adelle Blackett, “Follow the Drinking Gourd: Our Road to Teaching Critical Race Theory and Slavery and the Law, Contempla tively, at McGill,” McGill Law Journal 62, no. 4 (2017): 1251, doi: https://doi.org/10.7202/1043165ar. 65 Du Bois and Holloway, The Souls of Black Folk, 5. 66 Ibid, 8-9. 67 Ibid, 5. 68 Ibid, 9. 69 Mohsen al Attar, “Subverting Racism in / through International Law Scholarship,” Opinio Juris (blog), accessed 3 March 2021, http://opiniojuris.org/2021/03/03/subverting-racism-in-international-law-scholar ship/; Anna-Spain Bradley, “International Law’s Racism Problem,” Opinio Juris (blog), accessed 4 Septem ber 2019, http://opiniojuris.org/2019/09/04/international-laws-racism-problem/; Adelle Blackett, “Follow the Drinking Gourd: Our Road to Teaching Critical Race Theory and Slavery and the Law, Contempla tively, at McGill,” McGill Law Journal 62, no. 4 (2017): 1251, doi: https://doi.org/10.7202/1043165ar. 65 Du Bois and Holloway, The Souls of Black Folk, 5. 66 Ibid 8 9 IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL Colonised peoples clash with themselves, where they may inhabit a white supremacist world but have their reality. This includes cultural values and epistemological tendencies intertwined with their identity. The colonised are often translated to conflict with the world, a dyad explained using double-consciousness as an invaluable metonym. Therefore, handicapped people should not be asked to race with the world but allowed to give all their time and thought to their social problems.68 Despite a potent sense of being, colonised peoples have internalised colonial ideas and values that belittle their identities. They measure their existence against colonial ideals predicated on their dehumanisation, producing the chaotic internal struggle lamented by Du Bois. Du Bois’ double-consciousness captures the duality that grips TWAIL. According to prominent TWAIL scholars, international law is Eurocentric, parochial, predatory, divisive, and racist.69 These scholars remain committed to Eurocentric international law irrespective of the deficiencies. Previous scholarship described this dynamic as a paradox: Similar with most TWAIL scholars, I suppose I am more sanguine about 253 Mohsen Al-Attar TWAIL than international law. TWAIL provides a compelling critical apparatus that borrows and rejects analytical methods, celebrates and condemns normative pursuits, and submits to and subverts tenets of international law. Is it a theory? Yes, but a theory that challenges what legal theory is meant to be. Those looking to TWAIL for deliverance and disappointment commit self-harm because it cannot defend and forsake international law. Like the criticised international legal regime, I am forced to conclude that, at this stage, TWAIL has little more than paradoxes to offer. However, oh, what succulent paradoxes they are.70 Since making that argument, perceptions of TWAIL have developed. Rather than spiraling into a paradox, scholars are caught in a pendulum of their double-consciousness. According to Pahuja, TWAIL was born into a colonial, Christian, liberal, and capitalist international lawyers academy. Its scholars remain wedded to the construct, incapable of eschewing the compulsion to belong despite the counter-narratives and critiques. Counter-narratives empower and give agency.71 By selecting their words and telling their stories, TWAIL scholars have chipped at the international legal construct, forcing widespread acknowledgment of their realities.72 It was not a matter of belonging alone but belonging as they are. This means that TWAIL is a praxis of re-remembering and re-humanisation. Similarly, Du Bois advocated entreating the colonised to convert their double-consciousness into a radical and unitary consciousness grounded in anti-colonial history, culture, and struggle. 70 Mohsen al Attar, “TWAIL: A Paradox within a Paradox,” International Community Law Review 22 (2020): 195–96. 71 Raúl Alberto Mora, “Counter-Narrative,” Key Concepts in Intercultural Dialogue 36, 2014. 72 Ibid. 73 Du Bois and Holloway, The Souls of Black Folk, 152. IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL Such a double life with double thoughts, duties, and social classes must result in double words and ideals. This tempts the mind to pretense, revolt, hypocrisy, or radicalism.73 For Du Bois, the aim was to counter dehumanisation and make people’s consciousness whole. He intended his racial metaphor of double-consciousness to support the advance of the colonised toward a unitary self. The goal was to counter epistemic violence and double-consciousness with epistemological equivalency and radical consciousness. The history of the American Negro is the history of this strife, the longing to attain self-conscious manhood, to merge the double self into a better and truer self. In this merging, the Negro wishes neither of the older selves to be lost. He would not Africanise America, for America has too much to teach the world and Africa. The Negro soul would not be bleached n a flood of white 254 The Peculiar Double-consciousness of TWAIL Americanism, for his blood has a message for the world. He simply wishes to make it possible for a man to be both a Negro and an American.74 Belonging and being need not elicit an existential crisis, though they presuppose a radical and unitary consciousness. TWAIL echoes Du Bois’ ambition, with many scholars envisioning a regime of international law. In this case, the cultures and spiritualities guiding Third World peoples occupy centre stage in shared legality. Christopher Weeramantry wrote about this in the International Court of Justice (ICJ) case concerning the Gabčíkovo-Nagymaros Project.75 He stated that since the main forms of civilisation have been integrated into the structure and the Statute of the Court, the Court as being charged with the duty to refer to the wisdom of the world’s several civilisations. There is no stouter demand for epistemological equivalency than in this ruling. In contrast to the earlier goal of folding Third World states into the international legal regime as conceptualised by Europe, many TWAIL scholars are exploring rebuilding international law to account for Third World epistemologies. These epistemologies are produced in disparate continents and published in wide-ranging places. Therefore, the TWAIL scholarship is much better understood as part of a large, protracted, ongoing struggle against international law. Its scholars were taught, and they sought to re-make to ensure that it could become more international than it had received.76 TWAIL advances the possibility of radical international legality. 74 Ibid, 5. 75 Gabčikovo-Nagymaros Project (Hungary/Slovakia), Judgment, ICJ Reports 1997, p. 7, Separate Opinion of Vice-President Weeramantry, 88. 76 James Thuo Gathii, “The Promise of International Law: A Third World View,” American University International Law Review 36, no. 3 (2021): 408. 77 Fagbayibo, “Some Thoughts on Centring Pan-African Epistemic,” 185. 78 The Graduate Institute of International and Development Studies was the latest to advertise for such a post (January 2022). 76 James Thuo Gathii, “The Promise of International Law: A Third World View,” American Univers International Law Review 36, no. 3 (2021): 408. g y g g p 78 The Graduate Institute of International and Development Studies was the latest to advertise for such post (January 2022). 77 Fagbayibo, “Some Thoughts on Centring Pan-African Epistemic,” 185. Ibid, 5. 75 Gabčikovo-Nagymaros Project (Hungary/Slovakia), Judgment, ICJ Reports 1997, p. 7, Separate Opini of Vice-President Weeramantry, 88. IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL Such as legality urges experimentation with unique social imaginaries, transforming perceptions of international law. Orthodox scholars now recognise the incompleteness of their representations. To account for the subjectivity of their purported universality, some have referred to theirs as European approaches to international law. Others, such as Shaw and Klabbers, adapted their textbooks to account for critical approaches.77 There are also job notices that express a desire for academics familiar with Global South perspectives.78 This study also discussed the impact of this process on TWAIL scholars and their observations about international law. Anti-colonial nationalist international lawyers such as Anand, Bedjaoui, and Elias sought independence from colonialism. However, they also pursued a reconstruction of world order 255 Mohsen Al-Attar along egalitarian lines. Bedjaou observed that his generation of scholars did not locate international law in colonialism but in imperialism, arguing that colonialism is a stage in pursuing hegemony.79 This means that Bedjaoui and his cohort sought national independence, even when only formal, nominal, and fragile.80 They viewed decolonisation through material rather than an epistemic prism. Despite later attempts to read more into their aims, Bedjaoui stated that they wanted to integrate themselves into the international juridical order and progressively correct international commerce.81 According to Adom Getachew, Bedjaoui’s nationalism was sensible and persuasive. However, it is anachronistic currently, having been eclipsed by an irreversible momentum toward global economic integration. Once technology compressed time and space, political economy was caught up to render nationalism redundant, even when useful in obscuring class cleavages.82 Current TWAIL scholars are not enamoured with the spent promise of nationalism, establishing what Gathii termed a subaltern epistemic location from where they remake the world. These third-world scholars did not seek to fit conventional tropes of the discipline.83 They do not feel bound by post-enlightenment social knowledge, its structures of thought, and related constructions of political subjectivity, to which international law is central.84 While the scholars remain constrained by the problematic knowledge structures, the international legal regime forces them to operate within and seek to “[de- centre] the Westphalian myth” and take international law “beyond the usual pathways.85This is in line with Du Bois, which underscored the importance of the maturation of Third World scholarship for belonging and being. In the examination of Du Bois, Rebaka celebrated the colorful prose, poetry, and music making up The Souls of Black Folk: “Du Bois writes from different perspectives and speaks in different voices. 79 Mark Toufayan, Emmanuelle Jouannet, and Hélène Ruiz Fabri, eds., Droit International et Nouvelles Approches Sur Le Tiers-Monde: Entre Répétition et Renouveau (Société de législation comparée, 2013), 81. (Author’s translation) 80 Ibid, 83. 81 Ibid, 84. 82 Harry W Arthurs, “Law and Learning in an Era of Globalization,” German Law Journal 10, no. 07 (2009): 629. 83 James Thuo Gathii, “The Promise of International Law: A Third World View,” American University International Law Review 36, no. 3 (2021): 407. 84 Ibid, 409–10. 85 Ibid, 416. 86 Reiland Rabaka, Du Bois: A Critical Introduction, Key Contemporary Thinkers (Cambridge, UK; Med ford, MA, USA: Polity Press, 2021), 46. Ibid, 416. 86 Reiland Rabaka, Du Bois: A Critical Introduction, Key Contemporary Thinkers (Cambridge, UK; Med ford, MA, USA: Polity Press, 2021), 46. 79 Mark Toufayan, Emmanuelle Jouannet, and Hélène Ruiz Fabri, eds., Droit International et Nouvelles Approches Sur Le Tiers-Monde: Entre Répétition et Renouveau (Société de législation comparée, 2013), 81. (Author’s translation) 79 Mark Toufayan, Emmanuelle Jouannet, and Hélène Ruiz Fabri, eds., Droit International et Nouvelles Approches Sur Le Tiers-Monde: Entre Répétition et Renouveau (Société de législation comparée, 2013), 81. (Author’s translation) 80 Ibid, 83. 81 Ibid, 84. 82 Harry W Arthurs, “Law and Learning in an Era of Globalization,” German Law Journal 10, no. 07 y f 88 Michael Fakhri, “Third World Sovereignty, Indigenous Sovereignty, and Food Sovereignty: Living with Sovereignty despite the Map,” Transnational Legal Theory 9, no. 3–4 (2018): 218–53; Vusi Gu mede, “Leadership for Africa’s Development: Revisiting Indigenous African Leadership and Setting the Agenda for Political Leadership,” Journal of Black Studies 48, no. 1 (2017): 74–90, doi: https://doi. org/10.1177/0021934716678392. IV. THE DOUBLE-CONSCIOUSNESS OF TWAIL This is because race and racism are multidimensional, polyvocal, ever-changing, and require varied criticism and conceptual counter-attacks.86 These devices are critical to Du Bois’ self- actualization. He feels no compulsion to bleach his Negro soul in a flood of 256 The Peculiar Double-consciousness of TWAIL white Americanism because he knows that Negro blood has a message for the world.87 Similar characteristics are noticeable among TWAIL scholars and in the scholarship. Critical scholarship is a way of thinking, resisting, subverting, and being. Their scholarship is based on the lived experiences of Third World peoples, accounting for the multiplicity of contexts confronted by this disparate group.88 The people are critical in their conclusions and starting points. They eschew truth claims but expose the fiction of international law, inviting engagement with power, systems, structures, ideologies, and epistemologies that preserve the status quo. Furthermore, the people speak about international law and its possibilities in ways unimaginable by the orthodoxy. This means that TWAIL scholarship has upended the publicist community, supplanting certainty with doubt, truth with contingency. Its avant-garde quality has challenged and changed the order terms that international legal scholars once took for granted. For TWAIL, scholarship provokes its interlocutors to engage with the ideals of “non-dominant, non-elite, and subaltern individuals and groupings.”89 The scholars did not pursue this out of a misguided and Eurocentric desire to save the other. Conversely, they are stricken by a longing to attain self-conscious manhood (sic), to merge double self into a better and truer self, as Du Bois observed.90 TWAIL scholars seek to create conditions for a better and truer international law. y f 91 Cedric J. Robinson, Black Marxism: The Making of the Black Radical Tradition (Chapel Hill, N. University of North Carolina Press, 2000), 316. 89 James Thuo Gathii, “The Promise of International Law: A Third World View,” American Univers International Law Review 36, no. 3 (2021): 410. 87 Du Bois and Holloway, The Souls of Black Folk, 5. 87 Du Bois and Holloway, The Souls of Black Folk, 5. 88 Michael Fakhri, “Third World Sovereignty, Indigenous Sovereignty, and Food Sovereignty: Living with Sovereignty despite the Map,” Transnational Legal Theory 9, no. 3–4 (2018): 218–53; Vusi Gu mede, “Leadership for Africa’s Development: Revisiting Indigenous African Leadership and Setting the Agenda for Political Leadership,” Journal of Black Studies 48, no. 1 (2017): 74–90, doi: https://doi. org/10.1177/0021934716678392. 89 James Thuo Gathii, “The Promise of International Law: A Third World View,” American University International Law Review 36, no. 3 (2021): 410. 90 Du Bois and Holloway, The Souls of Black Folk, 5. 91 Cedric J. Robinson, Black Marxism: The Making of the Black Radical Tradition (Chapel Hill, N.C: University of North Carolina Press, 2000), 316. ( ) 90 Du Bois and Holloway, The Souls of Black Folk, 5. 92 Georges Abi-Saab, “The Third World Intellectual in Praxis: Confrontation, Participation, or Operation behind Enemy Lines?,” Third World Quarterly 37, no. 11 (November 2016): 1957–71, doi: https://doi.or g/10.1080/01436597.2016.1212653; Kwadwo Appiagyei-Atua, “Ethical Dimensions of Third-World Ap proaches to International Law (TWAIL): A Critical Review,” African Journal of Legal Studies 8, no. 3–4 (29 April 2015): 209–35, doi: https://doi.org/10.1163/17087384-12342063; Mohsen Al Attar and Rosalie Miller, “Towards an Emancipatory International Law: The Bolivarian Reconstruction,” Third World Quar terly 31, no. 3 (April 2010): 347–63, doi: https://doi.org/10.1080/01436597.2010.488469; Michael Fakhri, “Third World Sovereignty, Indigenous Sovereignty, and Food Sovereignty: Living with Sovereignty de spite the Map,” Transnational Legal Theory 9, no. 3–4 (2018). V. CONCLUSION In a provocative passage from a seminal book, Robinson forewarned that “if we are to survive, we must take nothing dead and selecting wisely from among the dying.91 Much of European international law is on life support, which is inevitable because the regime would collapse due to its contradictions. TWAIL is a movement that nourishes rather than dictates. Its genesis was political and powerful, centring the colonial and anti-colonial international law. However, it occupies an uncertain space, not revolution, reform, or renewal. According to Du Bois, it could not be any other way for the anti- 257 Mohsen Al-Attar colonial. TWAIL’s double-consciousness results from attempts to ethnically cleanse Third World peoples of themselves. It is a form of embryonic critique and inchoate because its scholars are wrenched in opposing directions, wedded and widowed. They were bound to embrace some characteristics of the dominant order regardless of the virulent resistance. Embrace and submission are not interchangeable, and TWAIL’s iconoclasm is cathartic. Some of its scholars pursue internal renovations to the frame. Furthermore, TWAIL’s more revolutionary interlocutors are inspired by avant- garde defiance, stirred by the eclectic and visionary attitudes of Third World peoples in brutal modernity.92 Its radical consciousness is vital to fashioning an international law suitable for the modern era. The scholars pursue this not in opposition, contempt, or condemnation of other peoples. Conversely, they do this to honor the ideals Third World peoples wish for international law, a hagiography upon which TWAIL could build. ACKNOWLEDGMENT This article is the author’s forthcoming manuscript on TWAIL, to be published by Oxford University Press in 2023 under the title A Guerrilla at the Hague: TWAIL and the Future of International Law. 258 The Peculiar Double-consciousness of TWAIL Journal Articles Abi-Saab, Georges. “The Third World Intellectual in Praxis: Confrontation, Participation, or Operation behind Enemy Lines?.” Third World Quarterly 37, no. 11 (November 2016): 1957–71. doi: https://doi.org/10.1080/01436597.2016 .1212653. Anghie, Antony and Bhupinder S. Chimni. “Third World Approaches to International Law and Individual Responsibility in Internal Conflicts.” Chinese Journal of International Law 2, no. 1 (2003): 77–103. Appiagyei-Atua, Kwadwo. “Ethical Dimensions of Third-World Approaches to International Law (TWAIL): A Critical Review.” African Journal of Legal Studies 8, no. 3–4 (29 April 2015): 209–35. doi: https://doi.org/10.1163/17087384- 12342063. Arthurs, Harry W. “Law and Learning in an Era of Globalization.” German Law Journal 10, no. 07 (2009): 629. Attar, Mohsen al. “Must International Legal Pedagogy Remain Eurocentric?.” Asian Journal of International Law 11, no. 1 (January 2021): 180. doi: https://doi. org/10.1017/S2044251321000138. Attar, Mohsen al. “Subverting Eurocentric Epistemology: The Value of Nonsense When Designing Counterfactual.” Contingency in International Law: On The Possibilities of Different Legal Histories ed. Venzke, Ingo and Kevin Heller. (Oxford University Press, 2021) Attar, Mohsen al. “Reframing the “Universality” of International Law in a Globalizing World’.” McGill Law Journal 59, no. 1 (2013): 95. doi: https://doi. org/10.7202/1018986ar. Attar, Mohsen al. “Subverting Racism in / through International Law Scholarship.” Opinio Juris (blog). Accessed 3 March 2021. http://opiniojuris.org/2021/03/03/ subverting-racism-in-international-law-scholarship/. Attar, Mohsen Al and Rosalie Miller. “Towards an Emancipatory International Law: The Bolivarian Reconstruction.” Third World Quarterly 31, no. 3 (April 2010): 347–63. doi: https://doi.org/10.1080/01436597.2010.488469. Attar, Mohsen al. “TWAIL: A Paradox within a Paradox.” International Community Law Review 22 (2020): 16. Bhandar, Brenna. “Theft in Broad Daylight: Racism and Neoliberal Legality.” Law and Critique 32, no. 3 (November 2021): 288. doi: https://doi.org/10.1007/ s10978-021-09308-9. Blackett, Adelle. “Follow the Drinking Gourd: Our Road to Teaching Critical Race Theory and Slavery and the Law, Contemplatively, at McGill.” McGill Law Journal 62, no. 4 (2017): 1251. doi: https://doi.org/10.7202/1043165ar. Bradley, Anna-Spain. “International Law’s Racism Problem.” Opinio Juris (blog). Accessed 4 September 2019. http://opiniojuris.org/2019/09/04/international- laws-racism-problem/. Boehme, Franziska. “Reactive Remembrance: The Political Struggle over Apologies and Reparations between Germany and Namibia for the Herero Genocide.” Journal of Human Rights 19, no. 2 (14 March 2020): 238–55. doi: https://doi.org /10.1080/14754835.2020.1727729. 259 Mohsen Al-Attar Burgis-Kasthala, Michelle. “Scholarship as Dialogue? TWAIL and the Politics of Methodology.” Journal of International Criminal Justice 14, no. 4 (2016): 921– 38. doi: https://doi.org/10.1093/jicj/mqw044.l Chimni, Bhupinder S. “Asian Civilizations and International Law: Some Reflections.” Asian Journal of International Law 1, no. 1 (2011): 39–42. Cole, Rowland J. V. Journal Articles “Africa’s Approach to International Law: Aspects of the Political and Economic Denominators.” African Yearbook of International Law Online / Annuaire Africain de Droit International Online 18, no. 1 (2010): 287–310. doi: https://doi.org/10.1163/22116176-01801011. Coleman, Clare Keefe. “Teaching the Torture Memos: Making Decisions under Conditions of Uncertainty.” Journal of Legal Education 62 (2012): 81. Drake, Philip James. “Beyond the Bounds of Formalism: Social Justice and Legal Education.” UK, Huddersfield, 2020. Fagbayibo, Babatunde. “Some Thoughts on Centring Pan-African Epistemic in the Teaching of Public International Law in African Universities.” International Community Law Review 21, no. 2 (2019): 170–89. doi: https://doi. org/10.1163/18719732-12341397. Fakhri, Michael. “Third World Sovereignty, Indigenous Sovereignty, and Food Sovereignty: Living with Sovereignty despite the Map.” Transnational Legal Theory 9, no. 3–4 (2018): 218–53. Fraser, Nancy. “Reframing Justice in a Globalising World.” New Left Review 36 (2005): 69. Fukurai, Hiroshi. “Fourth World Approaches to International Law (FWAIL) and Asia’s Indigenous Struggles and Quests for Recognition under International Law.” Asian Journal of Law and Society 5, no. 1 (May 2018): 221–31. doi: https://doi.org/10.1017/als.2018.10. Fukurai, Hiroshi. “Original Nation Approaches to “Inter- National” Law (ONAIL): Decoupling of the Nation and the State and the Search for New Legal Orders.” Indiana Journal of Global Legal Studies 26, no. 1 (2019): 199–262. Galindo, George. “Splitting TWAIL.” Windsor Yearbook of Access to Justice 33, no. 3 (2016): 37–56; Gathii, James Thuo. “The Promise of International Law: A Third World View.” American University International Law Review 36, no. 3 (2021): 102. Gathii, James Thuo. “TWAIL: A Brief History of Its Origins, Its Decentralized Network, and a Tentative Bibliography.” Trade, Law and Development 3, no. 1 (2011): 26–64. Gathii, James T. “Writing Race and Identity in a Global Context: What CRT and TWAIL Can Learn From Each Other.” UCLA Law Review, no. 67 (2020). Gumede, Vusi. “Leadership for Africa’s Development: Revisiting Indigenous African Leadership and Setting the Agenda for Political Leadership.” Journal of Black Studies 48, no. 1 (2017): 74–90. doi: https://doi.org/10.1177/0021934716678392. John, Bagu, Kajit. “Ideological Refuge v Jurisprudence of Insurgency: Cultural, Relativism and Universalism in the Human Rights Discourse.” Warwick Student Law Review 1 (2011): 1–18. Maguire, Amy. “Contemporary Anti-Colonial Self-Determination Claims and the Decolonisation of International Law.” Griffith Law Review 22, no. 1 (2013): 238– 68. doi: https://doi.org/10.1080/10383441.2013.10854774. 260 The Peculiar Double-consciousness of TWAIL Mahar, Alyson L. Virginie Cobigo, and Heather Stuart. “Conceptualizing Belonging.” Disability and Rehabilitation 35, no. 12 (June 2013): 1026–32. doi: https://doi.or g/10.3109/09638288.2012.717584. Marks, Susan. Journal Articles “Human Rights and the Bottom Billion.” European Human Rights Law Review, no. 1 (2009): 37–49. Massey, Doreen. “Negotiating Disciplinary Boundaries.” Current Sociology 47, no. 4 (1996): 6. Mickelson. “Taking Stock of TWAIL Histories’; Karin Mickelson, ‘Rhetoric and Rage: Third World Voices in International Legal Discourse.” Wis. Int’l L. J., no. 16 (1997): 353. Mora, Raúl Alberto. “Counter-Narrative.” Key Concepts in Intercultural Dialogue 36, 2014. Nayar, Jayan. “Thinking Being-Otherwise: Returning the Anti-Colonial to Philosophy.” Beyond Law and Development : Resistance, Empowerment and Social Injustice, ed. Abdul Paliwala. NY: Routledge, 2022. Okafor, Obiora Chinedu. “Critical Third World Approaches to International Law (TWAIL) Parmar, Pooja. “TWAIL: An Epistemological Inquiry.” International Community Law Review 10, no. 4 (2008): 363–70. Rajagopal, Balakrishnan. “Counter-Hegemonic International Law: Rethinking Human Rights and Development as a Third World Strategy.” Third World Quarterly 27, no. 5 (July 2006): 767–83. doi: https://doi.org/10.1080/01436590600780078. Roth, Brad R. “Governmental Illegitimacy and Neocolonialism: Response to Review by James Thuo Gathii.” Michigan Law Review 98, no. 6 (2000): 2056–65. doi: https://doi.org/10.2307/1290273. Seneviratne, S.S.M.W. and K.A.A.N. Thilakarathna. “Recognition of New States: An Analysis of Selected Case Studies from a Third World Approach to International Law (TWAIL).” Colombo, 2020. Sylvester, Ogba Adejoh and Okpanachi Idoko Anthony. “Decolonization in Africa and Pan-Africanism.” Yönetim Bilimleri Dergisi 12, no. 23 (2014) Theory, Methodology, or Both?.” International Community Law Review 10, no. 4 (2008): 371–78. Thomas, Darryl C. “The Black Radical Tradition - Theory and Practice: Black Studies and the Scholarship of Cedric Robinson.” Race & Class 47, no. 2 (October 2005): 1–22. doi: https://doi.org/10.1177/0306396805058077. Internet Anghie, Antony. ‘Critical Pedagogy Symposium: Critical Thinking and Teaching as Common Sense – Random Reflections’ Opinio Juris (blog). Accessed 31 August 2020. http://opiniojuris.org/2020/08/31/critical-pedagogy-symposium-critical- thinking-and-teaching-as-common-sense-random-reflections/. Attar, Mohsen al. “Subverting Racism in / through International Law Scholarship.” Opinio Juris (blog). Accessed 3 March 2021. http://opiniojuris.org/2021/03/03/ subverting-racism-in-international-law-scholarship/. Book and Book Chapters Anghie, Antony. Imperialism, Sovereignty, and the Making of International Law. Cambridge Studies in International and Comparative Law. Cambridge University Press, 2004. Du Bois, W. E. B. and Jonathan Scott Holloway, The Souls of Black Folk. First Yale University Press edition (New Haven [Connecticut]: Yale University Press, 2015. Gathii, James Thuo. War, Commerce, and International Law. Oxford University Press, 2010. Mark Toufayan, Emmanuelle Jouannet, and Hélène Ruiz Fabri, eds., Droit International et Nouvelles Approches Sur Le Tiers-Monde: Entre Répétition et 261 Mohsen Al-Attar Renouveau (Société de législation comparée, 2013), 81. (Author’s translation) Pahuja, Sundhya. Decolonising International Law: Development, Economic Growth and the Politics of Universality. Cambridge University Press, 2011. abaka, Reiland. Du Bois: A Critical Introduction. Key Contemporary Thinkers, Cambridge, UK; Medford, MA, USA: Polity Press, 2021. Robinson, Cedric J. Black Marxism: The Making of the Black Radical Tradition. Chapel Hill, N.C: University of North Carolina Press, 2000. Jansen, Jan. “Politics of Remembrance, Colonialism and the Algerian War of Independence in France’, in A European Memory? Contested Histories and Politics of Remembrance, by Malgorzata Pakier and Bo Stråth.” Studies in Contemporary European History. New York: Berghahn Books, 2010. Others CBS News. “Paul Bremer, Who Helped Run Iraq Occupation, Is Now a Ski Instructor.” News, CBS News. Accessed 7 March 2022. https://www.cbsnews.com/news/ paul-bremer-who-helped-run-iraq-occupation-is-now-a-ski-instructor/. Gabčikovo-Nagymaros Project (HungarylSlovakia). Judgment, ICJ Reports 1997. p. 7, Separate Opinion of Vice-President Weeramantry, p.88. The Graduate Institute of International and Development Studies was the latest to advertise for such a post (January 2022). 262
https://openalex.org/W3194844109	https://www.researchsquare.com/article/rs-198381/latest.pdf	English	null	Hybrid (Dynamical-Statistical) Models to Improve Seasonal Drought Prediction: Application to Regional Meteorological Droughts in China	Research Square (Research Square)	2,021	cc-by	13,101	Hybrid (Dynamical-Statistical) Models to Improve Seasonal Drought Prediction: Application to Regional Meteorological Droughts in China Zhiyong Wu (  zywu@hhu.edu.cn ) Hohai University College of Hydrology and Water Resources https://orcid.org/0000-0001-7186-3776 Hao Yin Hohai University College of Hydrology and Water Resources Hai He Hohai University College of Hydrology and Water Resources Yuan Li Hohai University College of Hydrology and Water Resources Research Article Keywords: Seasonal drought prediction, Hybrid models, Machine learning, China Posted Date: February 23rd, 2021 DOI: https://doi.org/10.21203/rs.3.rs-198381/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Hybrid (dynamical-statistical) models to improve seasonal drought 1 prediction: application to regional meteorological droughts in China 2 Zhiyong Wu1,2 · Hao Yin1 · Hai He1 · Yuan Li1 3 1College of Hydrology and Water Resources, Hohai University, Nanjing 210098, China 4 2Yangtze Institute for Conservation and Development, Hohai University, Nanjing 210098, China 5 Correspondence to: Zhiyong Wu (wzyhhu@gmail.com) 6 Hybrid (dynamical-statistical) models to improve seasonal dro 1 prediction: application to regional meteorological droughts in Ch 2 Zhiyong Wu1,2 · Hao Yin1 · Hai He1 · Yuan Li1 3 1College of Hydrology and Water Resources, Hohai University, Nanjing 210098, China 4 2Yangtze Institute for Conservation and Development, Hohai University, Nanjing 210098, China 5 Correspondence to: Zhiyong Wu (wzyhhu@gmail.com) 6 Abstract 7 Accurate drought prediction is important for drought resistance and water resources management. However, the seasonal 8 drought prediction is of low accuracy for both dynamical and statistical models. In this study, we combined dynamical models 9 and machine learning to construct hybrid (dynamical-statistical) models. We used the random forest approach to identify 10 representative regions based on geopotential height, sea-level pressure, and 2-m temperature. The least absolute shrinkage and 11 selection operator (Lasso and an artificial neural network (ANN) were used to construct the statistical models, with atmospheric 12 variables as predictors and 3-month Standardized Precipitation Index (SPI3) as the predictand. The atmospheric variables 13 forecasted by the European Centre for Medium-Range Weather Forecasts (ECMWF) SEAS5 model were processed as 14 predictors to force the statistical models. The resulting hybrid models, constructed using dynamical models and machine 15 learning, were named as dynamic-Lasso (‘D-Lasso’) and dynamic-ANN (‘D-ANN’) separately. The results suggested that 16 prediction skills were improved by the hybrid models; compared to the best available dynamical model (UK Met Office), D- 17 ANN extends the forecast horizons by 6, 21, and 4 days in northern, eastern, and southern China, respectively. In spring and 18 summer, the correlation skills were also improved. The effective prediction of the atmospheric anomalies over the eastern and 19 southern Tibetan Plateau and the Northwest Pacific region was identified as the main contributor to successful seasonal drought 20 prediction. Overall, the hybrid models were able to predict drought processes effectively, and D-ANN outperformed the D- 21 Lasso in drought onset and persistence phases. 22 1 Introduction 24 Droughts are natural disasters that occur under all climate regimes (Liu et al. 2016). Global chronically drought-prone areas 25 have increased significantly, from 16.19% between 1902 and 1949 to 41.09% between 1950 and 2008 (Wang et al. 2014). 26 Globally, an estimated 55 million people are affected by droughts every year, and as many as 700 million people are at risk of 27 1 being displaced as a result of drought by 2030 (World Health Organization 2019). According to the Food and Agriculture 28 Organization (FAO) of the United States, developing countries suffered losses of $29 billion in agriculture due to droughts 29 between 2005 and 2015 (Conforti et al. 2018). Therefore, effective drought prediction is important to improve drought 30 resistance and water resources management, and reduce drought losses. 31 being displaced as a result of drought by 2030 (World Health Organization 2019). According to the Food and Agriculture 28 Organization (FAO) of the United States, developing countries suffered losses of $29 billion in agriculture due to droughts 29 between 2005 and 2015 (Conforti et al. 2018). Therefore, effective drought prediction is important to improve drought 30 resistance and water resources management, and reduce drought losses. 31 With the increase of our understanding of the climate system and computational capabilities, predicting droughts using general 32 circulation models (GCMs) has become a fundamental approach (Hao et al. 2018). Seasonal drought indices, such as the 33 Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI), can be calculated 34 from GCM precipitation and temperature outputs (Lang et al. 2020; Ma et al. 2015; Ujeneza and Abiodun 2015). The advantage 35 of GCMs is that they are based on physical principles and can capture the nonlinear changes in the ocean, atmosphere, and on 36 land. In recent years, significant advances have been made in drought prediction, with the widespread use of post-processing 37 and multi-model ensembles (Li et al. 2020; Mo and Lyon 2015). Nevertheless, GCMs are limited by chaotic properties, initial 38 value errors, and model biases, which lead to uncertainties in the forecasted precipitation with a lead time of more than two 39 weeks (Li et al. 2017; Merryfield et al. 2020). Statistical methods usually first identify a range of predictors from historical 40 hydroclimatic observations (reanalysis) which are then input into statistical models to predict drought. 1 Introduction 24 2019), and this approach has also been used to improve drought prediction (Agana and Homaifar 2017; 63 Khan et al. 2020; Rezaeianzadeh and Tabari 2012; Zhang et al. 2019). However, drought mechanisms vary between different 64 regions and across temporal-spatial scales, meaning that further research is needed to determine whether ML is effective for 65 seasonal drought prediction. 66 has helped improve climate predictions significantly as well as assist in interpreting the underlying mechanisms (Pan et al. 62 2020; Yoo-Geun et al. 2019), and this approach has also been used to improve drought prediction (Agana and Homaifar 2017; 63 Khan et al. 2020; Rezaeianzadeh and Tabari 2012; Zhang et al. 2019). However, drought mechanisms vary between different 64 regions and across temporal-spatial scales, meaning that further research is needed to determine whether ML is effective for 65 seasonal drought prediction. 66 China has long been affected by drought due to its geographical location and climatic regime (Ayantobo et al. 2017), and in 67 recent years, extreme droughts have become increasingly frequent and severe in eastern and southern China (Chen et al. 2018; 68 Jin et al. 2013; Ma et al. 2020). For example, the great drought in southwestern China from September 2009 to March 2010 69 affected more than 69 million people and 6.6 million hectares of crops, with economic losses of more than 40 billion yuan 70 (China Meteorological Administration 2011). In 2013, drought in southern China affected more than 80 million people and 71 damaged 7.9 million hectares of crops (China Meteorological Administration 2014). While progress has been made in drought 72 prediction in China using dynamical models, statistical methods, and hybrid models (Xu et al. 2018a; Xu et al. 2018b; Zhang 73 et al. 2019; Zhu et al. 2020), there have been few attempts to employ atmospheric circulation datasets. Therefore, the 74 applicability of ML for seasonal drought prediction in different regions of China remains unknown. 75 In this study, we combine dynamical models and ML to construct hybrid models. We use a random forest (RF) approach to 76 identify typical regions from atmospheric circulation variables. The least absolute shrinkage and selection operator (Lasso) 77 and an artificial neural network (ANN) are used to extract statistical relationships between atmospheric variables in 78 representative regions and meteorological drought. The atmospheric variables forecasted by the dynamical models are then 79 processed to generate drought-prediction models. 1 Introduction 24 Statistical methods are 41 less computationally demanding and usually provide a baseline level of skill (Kirtman et al. 2013). However, unclear physical 42 mechanisms and unstable statistical relationships in climate change are always the problems of statistical methods (Hao et al. 43 2018). To combine the advantages of dynamical and statistical models, many researchers have attempted to construct hybrid 44 (dynamical-statistical) models. 45 Hybrid models mainly involve two steps, first, calibration to correct the bias (and ensemble spread) of GCMs forecasts and, 46 second, merging forecasts from multiple sources (Madadgar et al. 2016; Ribeiro and Pires 2016; Yan et al. 2017). Previous 47 efforts have focused on combining forecasts from multiple dynamical and statistical models, which has achieved excellent 48 results (Hao et al. 2016; Madadgar et al. 2016; Wang et al. 2012). Nevertheless, some variables, such as geopotential height 49 and temperature, yield higher skill scores with long lead times and are often used to analyze atmospheric circulation anomalies 50 during droughts but are rarely used for prediction. For seasonal drought, large-scale circulation is an important mechanism, 51 and the role of atmospheric anomalies varies between droughts (Lhotka et al. 2020; Li et al. 2018b; Okumura et al. 2017; 52 Zhuang et al. 2020). Thus, some studies have aimed to extract information from atmospheric circulation patterns to predict 53 drought events. For example, Costa-Cabral et al. (2016) were able to project meteorological drought in California using the 54 North Pacific High (NPC) sea-level pressure anomaly; in northeastern Europe, more than 65% of drought events were detected 55 one month in advance by Lavaysse et al. (2018); and Richardson et al. (2020) explored the potential benefits of using weather 56 pattern predictions to improve sub-seasonal forecasts of meteorological drought. In these studies, the selection of an 57 appropriate method to obtain key information from atmospheric variables was critical. 58 With the explosive growth of observation (and reanalysis) data and the considerable increases in computing capabilities, 59 climate prediction with machine learning (ML) has become an active area of research (Reichstein et al. 2019). The most 60 significant advantage of ML is its ability to mine information from ‘big data’ for future prediction. In the past few years, ML 61 2 has helped improve climate predictions significantly as well as assist in interpreting the underlying mechanisms (Pan et al. 62 2020; Yoo-Geun et al. Three of the nine drought regions of China (Fig. 1), namely Northern China (North), Eastern China (East), and Southern China 86 (South), were selected as the study regions, covering approximately 0.69, 0.91, and 0.37 million km2, respectively. The East 87 Asian Monsoon climate greatly influences these regions, and precipitation shows large inter- and intra-annual variability (Cai 88 et al. 2017; Liu and Huang 2019; Zhang 2015). 89 1 Introduction 24 The remainder of this paper is structured as follows: Section 2 introduces 80 the study area and data; the ML models and predictor construction methods are described in Sect. 3; Section 4 presents the 81 prediction results; we will discuss the study limitations and directions for future work in Sect. 5; and the conclusions are 82 presented in Sect. 6. 83 2.2.1 Observations 94 Observed daily precipitation amounts from the Climate Data Centre (CDC) of the National Meteorological Information Centre, 95 China Meteorological Administration (CMA), were used, providing 0.5° × 0.5° gridded data for the period 1979–2019 96 (http://data.cma.cn/data/detail/dataCode/SURF_CLI_CHN_PRE_DAY_GRID_0.5.html). These data were initially used to 97 calculate area-averaged precipitation over the three selected regions. 98 Observed daily precipitation amounts from the Climate Data Centre (CDC) of the National Meteorological Information Centre, 95 China Meteorological Administration (CMA), were used, providing 0.5° × 0.5° gridded data for the period 1979–2019 96 (http://data.cma.cn/data/detail/dataCode/SURF_CLI_CHN_PRE_DAY_GRID_0.5.html). These data were initially used to 97 calculate area-averaged precipitation over the three selected regions. 98 The following atmospheric circulation information was obtained from the European Centre for Medium-Range Weather 99 Forecasts (ECMWF) ERA5 reanalysis dataset (https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5) at a 2° × 100 2° resolution for the area 30–180° E, 10° S–90° N: geopotential heights at 200, 500, and 850 hPa (GH200, GH500, and 101 GH850); 2-m temperature (T2M); and sea-level pressure (SLP). 102 2.1 Study areas 85 Three of the nine drought regions of China (Fig. 1), namely Northern China (North), Eastern China (East), and Southern China 86 (South), were selected as the study regions, covering approximately 0.69, 0.91, and 0.37 million km2, respectively. The East 87 Asian Monsoon climate greatly influences these regions, and precipitation shows large inter- and intra-annual variability (Cai 88 et al. 2017; Liu and Huang 2019; Zhang 2015). 89 3 3 Fig. 1 The location of the drought regions of China and the coverage of atmospheric variables (30°-180°E, 10°S-90°N). The gra shaded areas are the study regions. 1-3 represent Northern, Eastern, and Southern China, respectively Fig. 1 The location of the drought regions of China and the coverage of atmospheric variables (30°-180°E, 10°S-90°N). The g haded areas are the study regions. 1-3 represent Northern, Eastern, and Southern China, respectively Fig. 1 The location of the drought regions of China and the coverage of atmospheric variables (30°-180°E, 10°S-90°N). The gray 91 shaded areas are the study regions. 1-3 represent Northern, Eastern, and Southern China, respectively 92 3.1 Daily-updated Standardized Precipitation Index 112 As a seasonal drought index, the SPI3 was calculated for the three study regions using area-averaged precipitation data. 113 Traditionally, the SPI3 varies on a monthly scale, reflecting cumulative precipitation over the past three months; however, a 114 monthly timescale does not well reflect the evolution of drought onset, persistence, and relief, presenting limitations for 115 meeting the requirements for real-time monitoring and prediction for effective management. Therefore, we followed the World 116 Meteorological Organization (WMO) recommendation to adjust the sliding time window of the SPI3 to 90 days (Svoboda et 117 al. 2012). Thus, we obtained information on the daily variation of meteorological droughts while the mathematical treatment 118 remained unchanged. Full calculation details are given by Liu et al. (2018). 119 2.2.2 GCM hindcast data 103 We focused on seasonal drought prediction for the next 90 days, corresponding to the ‘seasonal scale’ discussed throughout 104 the paper. For this, data for the atmospheric variables (in 90-day blocks) were downloaded from the ECMWF SEAS5 hindcast 105 4 4 dataset for the period 1993–2016 (https://cds.climate.copernicus.eu/cdsapp#!/dataset/seasonal-original-single-levels), 106 including GH200, GH500, GH850, T2M, and SLP. For comparison, we selected hindcast data from the ECMWF, UK Met 107 Office (UKMO), and Météo France (https://cds.climate.copernicus.eu/cdsapp#!/dataset/seasonal-original-single-levels) 108 hindcast precipitation (Table 1). 109 dataset for the period 1993–2016 (https://cds.climate.copernicus.eu/cdsapp#!/dataset/seasonal-original-single-levels), 106 including GH200, GH500, GH850, T2M, and SLP. For comparison, we selected hindcast data from the ECMWF, UK Met 107 Office (UKMO), and Météo France (https://cds.climate.copernicus.eu/cdsapp#!/dataset/seasonal-original-single-levels) 108 hindcast precipitation (Table 1). 109 Table 1 Data used in this study 110 Data type Variable name Pressure level (hPa) Abbreviation Data source Atmospheric variables 2-m temperature surface T2M Reanalysis: ERA5 reanalysis Hindcast: ECMWF SEAS5 Sea-level pressure surface SLP Geopotential height 200/500/850 GH200/GH500/GH850 Precipitation Total precipitation surface TP Observation: CMA Hindcast: ECMWF/UKMO/Météo France Table 1 Data used in this study 110 Data type Variable name Pressure level (hPa) Abbreviation Data source Atmospheric variables 2-m temperature surface T2M Reanalysis: ERA5 reanalysis Hindcast: ECMWF SEAS5 Sea-level pressure surface SLP Geopotential height 200/500/850 GH200/GH500/GH850 Precipitation Total precipitation surface TP Observation: CMA Hindcast: ECMWF/UKMO/Météo France 3.3.1 Construction of ML models 132 We constructed ML models for each of the three drought regions using 32 years of data from 1980 to 2011 for training and 8 133 years of data from 2012 to 2019 for validation. The models adopted the daily five-layer SA90 dataset as the predictor and the 134 SPI3 as the predictand. The model structure is shown in Fig. 2. First, we flattened the five-layer SA90 into a column of 55161 135 data points (the input neurons), extracted typical data through the RF, and then produced the output SPI3 data by forcing the 136 Lasso and ANN. The ML models (RF, Lasso, and ANN) are described in detail in Supplement 1–3. 137 138 Fig. 2 ML model structure, where ANN contains a single layer and 16 nodes. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP 139 are the 90-day standardized anomalies of GH200, GH500, GH850, SLP and T2M, respectively. Lasso is a linear model, while ANN 140 is linear 141 Fig. 2 ML model structure, where ANN contains a single layer and 16 nodes. SA9 139 Fig. 2 ML model structure, where ANN contains a single layer and 16 nodes. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP 139 are the 90-day standardized anomalies of GH200, GH500, GH850, SLP and T2M, respectively. Lasso is a linear model, while ANN 140 is linear 141 3.2 Predictor construction 120 Hart and Grumm (2001) first used standardized anomalies to delineate weather-scale events, and this approach has been used 121 in an increasing number of studies to identify atmospheric and oceanic anomalies (Duan et al. 2014; Liu et al. 2018; Liu et al. 122 2017). Here, we calculated 90-day standardized anomalies (SA90) using the following equation: 123 SA90 = 𝑋90−𝜇90 𝜎90 , (1) 124 SA90 = 𝑋90−𝜇90 𝜎90 , 124 (1) where 𝑋90 represents the mean values of atmospheric variables over the past 90 days; and 𝜇90 and 𝜎90 are the mean and 125 standard deviation of the 90-day period, respectively. The climatic background period selected for this study was 1981–2010. 126 For example, 𝑋90 for 1 April 2000 represented the 90-day average for the period from 3 January 2000 to 1 April 2000; and 𝜇90 127 and 𝜎90 are the 90-day average and standard deviation for the period from 3 January 1981 to 1 April 2010. The ERA5 reanalysis 128 5 5 dataset for the 40-year period between 1980 and 2019 was processed into SA90, from which we obtained the SA90 for all of 129 the considered atmospheric variables (i.e. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP). 130 3.3.3 GCM hindcast precipitation 151 The SPI3 calculated from the ECMWF, UKMO, and Météo France hindcast precipitation datasets were compared with the 152 hybrid models. For prediction, we use observed precipitation for the past (90-1-N) days, GCM outputs for the prospective N 153 days, and then calculated the Nth-day SPI3. 154 3.3.2 Calculation of the prospective Nth-day SPI3 142 The ML models constructed in Sect. 3.3.1 incorporate the contemporaneous statistical relationships between the predictors and 143 the SPI3 and, therefore cannot predict the future. During the training and validation periods, the models were forced with 144 ERA5 data. For prediction, we first calculated SA90 using ERA5 reanalysis data for the past (90-1-N) days, ECMWF outputs 145 6 for the prospective N days, and then forced the ML models to predict the Nth-day SPI3, as shown in Fig. 3. Therefore, the 146 drought prediction models we construct were hybrid (dynamical–statistical) models, which we name dynamic-Lasso (‘D- 147 Lasso’) and dynamic-ANN (‘D-ANN’), respectively. 148 for the prospective N days, and then forced the ML models to predict the Nth-day SPI3, as shown in Fig. 3. Therefore, the 146 drought prediction models we construct were hybrid (dynamical–statistical) models, which we name dynamic-Lasso (‘D- 147 Lasso’) and dynamic-ANN (‘D-ANN’), respectively. 148 149 Fig. 3 Schematic representation of the calculation process for the prospective Nth-day SPI3 150 Fig. 3 Schematic representation of the calculation process for the prospective Nth- 150 chematic representation of the calculation process for the prospective Nth-day SPI3 3.3.4 Evaluation metrics 155 Root-mean-square error (RMSE) was adopted as the loss function to describe the error between the model output and the 156 observed data, and for the inverse calculation of the network residuals, as follows: 157 Root-mean-square error (RMSE) was adopted as the loss function to describe the error between the model output and the 156 observed data, and for the inverse calculation of the network residuals, as follows: 157 RMSE = √ ∑ (𝑦𝑖−𝑥𝑖)2 𝑛 𝑖=1 𝑛 , (2) 158 RMSE = √ ∑ (𝑦𝑖−𝑥𝑖)2 𝑛 𝑖=1 𝑛 , (2) 158 RMSE = √ ∑ (𝑦𝑖−𝑥𝑖)2 𝑛 𝑖=1 𝑛 , 158 (2) where 𝑥𝑖 represents the observed value, 𝑦𝑖 represents the predictand, and n represents the sample size. 159 where 𝑥𝑖 represents the observed value, 𝑦𝑖 represents the predictand, and n represents the sample size. 159 The correlation coefficient (Corr) was used as the performance evaluation function for ML to measure the correlation between 160 the model output and the expected values: 161 The correlation coefficient (Corr) was used as the performance evaluation function for ML to measure the correlation between 160 the model output and the expected values: 161 Corr = 1 − ∑ (𝑥𝑖−𝑥̅)(𝑦𝑖−𝑦̅) 𝑛 𝑖=1 √∑ (𝑥𝑖−𝑥̅)2 𝑛 𝑖=1 √∑ (𝑦𝑖−𝑦̅)2 𝑛 𝑖=1 , (3) 162 Corr = 1 − ∑ (𝑥𝑖−𝑥̅)(𝑦𝑖−𝑦̅) 𝑛 𝑖=1 √∑ (𝑥𝑖−𝑥̅)2 𝑛 𝑖=1 √∑ (𝑦𝑖−𝑦̅)2 𝑛 𝑖=1 , 162 (3) where 𝑥𝑖 is the observed value, 𝑦𝑖 is the predictand, 𝑥̅ = 1 𝑛∑ 𝑥𝑖 𝑛 𝑖=1 (the sample mean), and analogously for 𝑦̅; and n represents 163 the sample size. 164 where 𝑥𝑖 is the observed value, 𝑦𝑖 is the predictand, 𝑥̅ = 1 𝑛∑ 𝑥𝑖 𝑛 𝑖=1 (the sample mean), and analogously for 𝑦̅; and n represents 163 the sample size. 164 7 A drought event is defined as a number of consecutive days (> 60 days) with daily updated SPI3 values of < -0.5. Based on 167 this approach, the droughts identified between 1979 and 2019 in the study regions are shown in Table 2. 3.3.4 Evaluation metrics 155 168 Table 2 Drought events identified in the three study regions between 1979 and 2019 69 Regions North East South Dates (Start– End) 1981/05/03–1981/08/16 1984/01/16–1984/05/23 1986/03/23–1986/08/09 1989/09/03–1989/11/11 1991/09/09–1992/01/25 1992/06/10–1992/09/22 1995/03/07–1995/06/19 1995/12/31–1996/05/02 1997/06/11–1997/11/14 1998/11/02–1999/04/10 1999/07/14–1999/10/31 2000/03/23–2000/06/26 2001/04/12–2001/08/18 2002/08/04–2002/12/05 2008/12/23–2009/02/24 2010/12/16–2011/02/27 2011/05/26–2011/08/25 2013/03/22–2013/05/25 2014/07/15–2014/09/27 1981/06/18–1981/10/06 1984/01/12–1984/04/07 1985/05/31–1985/09/23 1986/02/20–1986/06/22 1986/08/18–1986/11/14 1988/06/23–1988/09/04 1991/10/04–1992/01/04 1992/07/26–1993/01/13 1995/12/28–1996/03/25 1998/10/26–1999/04/23 2000/04/07–2000/06/22 2001/04/05–2001/06/19 2001/06/21–2001/08/30 2001/09/01–2001/11/04 2003/09/25–2003/12/10 2004/03/07–2004/05/14 2004/10/11–2005/01/08 2007/05/06–2007/07/11 2007/11/21–2008/01/27 2009/09/27–2009/12/02 2011/03/07–2011/08/04 2013/03/14–2013/05/15 2013/07/28–2013/10/09 2019/10/10–2019/12/21 1984/01/11–1984/04/17 1985/05/20–1985/09/05 1986/09/10–1986/11/17 1988/06/07–1988/08/31 1989/07/06–1990/01/10 1991/04/03–1991/08/15 1991/10/18–1991/12/26 1992/09/12–1993/01/13 1996/01/04–1996/03/27 1996/11/14–1997/01/23 1998/09/21–1998/12/01 1999/01/25–1999/05/25 2002/03/09–2002/06/17 2003/03/25–2003/06/13 2003/07/03–2004/02/04 2004/05/22–2004/08/03 2004/10/09–2005/02/05 2005/11/12–2006/02/26 2007/11/19–2008/01/30 2009/01/31–2009/04/19 2009/10/01–2009/12/01 2011/03/11–2011/10/24 2012/09/20–2012/11/21 2018/04/05–2018/07/06 4.2 Simulated SPI3 in validation periods 170 4.2 Simulated SPI3 in validation periods 170 4.2 Simulated SPI3 in validation periods 170 4.2 Simulated SPI3 in validation periods 170 According to Sect. 3, we constructed ML models based on data for the period 1980–2011. To test the robustness of the models, 171 we forced the Lasso and ANN models for the period between 2012 and 2019 (Fig. 4). Both of these models’ simulations were 172 consistent with the observed SPI3 trends in all three regions, although ANN performed better overall, especially when 173 simulating extreme values. This indicates that the ANN has a better information-extraction capability and better captures 174 nonlinear atmospheric processes. In addition, the models performed much better for the east and south regions than the north 175 study region, which may reflect the differing drought mechanisms in these regions. 176 8 8 Fig. 4 Temporal trends in the observed and simulated SPI3 during the validation period (1 January 2012 to 31 December 2019). The black curve is the observed SPI3, and green and red curves are the Lasso and ANN simulations, respectively. The grey shading shows the identified drought events in Sect. 4.1 Fig. 4 Temporal trends in the observed and simulated SPI3 during the validation p 178 Fig. 4 Temporal trends in the observed and simulated SPI3 during the validation period (1 January 2012 to 31 December 2019). The 178 black curve is the observed SPI3, and green and red curves are the Lasso and ANN simulations, respectively. The grey shading 179 shows the identified drought events in Sect. 4.1 180 4.3 Seasonal drought prediction 181 Between 1993 and 2016, ECMWF hindcasts data for the first day of each month, as do D-Lasso and D-ANN. Therefore, we 182 generated five sets of seasonal drought prediction products (SPI3) based on the Météo France, ECMWF, UKMO, D-Lasso, 183 and D-ANN datasets. 184 Between 1993 and 2016, ECMWF hindcasts data for the first day of each month, as do D-Lasso and D-ANN. Therefore, we 182 generated five sets of seasonal drought prediction products (SPI3) based on the Météo France, ECMWF, UKMO, D-Lasso, 183 and D-ANN datasets. 184 Fig. 5 shows the correlation results of the SPI3 between 1993 and 2016. The strength of the correlations between all five 185 models decreases as the lead time increases. At three weeks or less, the correlation between the ECMWF and UKMO outputs 186 Fig. 5 shows the correlation results of the SPI3 between 1993 and 2016. The strength of the correlations between all five 185 models decreases as the lead time increases. At three weeks or less, the correlation between the ECMWF and UKMO outputs 186 9 is stronger than that of the hybrid models, indicating that dynamic models are more accurate in simulating precipitation with 187 short lead times, which is consistent with other analyses (Lang et al. 2014; Li et al. 2017). However, when the lead time exceeds 188 20 days, the hybrid models always outperformed the dynamic models; in particular, D-ANN showed the best performance for 189 lead times above 30 days, which indicates that nonlinear hybrid models further improved drought predictions. 190 is stronger than that of the hybrid models, indicating that dynamic models are more accurate in simulating precipitation with 187 short lead times, which is consistent with other analyses (Lang et al. 2014; Li et al. 2017). However, when the lead time exceeds 188 20 days, the hybrid models always outperformed the dynamic models; in particular, D-ANN showed the best performance for 189 lead times above 30 days, which indicates that nonlinear hybrid models further improved drought predictions. 190 Data correlations also differed between the study regions. For example, outputs from both the dynamical and hybrid models 191 in the north region had lower correlation skills than the other two models. Among the three dynamical models, the UKMO 192 model was best for seasonal drought prediction. 4.3 Seasonal drought prediction 181 Following Buizza and Leutbecher (2015), we determined the forecast lead 193 times at which the correlation skills of the models were no longer higher than 0.5, defined as the forecast skill horizon. Based 194 on this, D-ANN extended the forecast skill horizon by 6, 21, and 4 days in the north, east, and south regions, respectively, 195 relative to the UKMO model. In particular, the correlation of D-ANN was higher than 0.5 with a 76-day lead time in the east 196 region, while the correlations of the Météo France, ECMWF, and UKMO data were 0.18, 0.29, and 0.34, respectively. 197 g p y 198 Fig. 5 Correlation skills between five models predicting SPI3 at different lead times. Black, purple, blue, green, and red curves 199 represent Météo France, ECMWF, UKMO, D-Lasso, and D-ANN outputs, respectively. The grey dashed lines indicate a 200 correlation of 0.5 201 98 Fig 5 Correlation skills between five models predicting SPI3 at different lead times Black purple blue green and red curves 99 Fig. 5 Correlation skills between five models predicting SPI3 at different lead times. Black, purple, blue, green, and red curves 199 represent Météo France, ECMWF, UKMO, D-Lasso, and D-ANN outputs, respectively. The grey dashed lines indicate a 200 correlation of 0.5 201 10 4.5 Interpretation of prediction mechanisms 214 Fig. 7(a–c) shows that D-ANN can effectively predict seasonal droughts across the different regions. The color maps (Fig. 215 7(d–e)) quantify the contribution of each grid predictor to the prediction, where positive and negative indicate that the predictor 216 contributes to positive and negative to SPI3, respectively. 217 The representative regions identified by the RF for GH850 and SLP are mainly located in the east and south of the Tibetan 218 Plateau and the Northwest Pacific. In addition, the representative regions in the north are sparser than in the east and south, 219 which means the RF could not effectively extract key information from the atmospheric variables. Importantly, the East Asian 220 Summer/Winter Monsoons are fundamental factors affecting droughts and floods in China (Ding et al. 2008; Wen et al. 2000; 221 Yihui and Chan 2005; Zhou and Wu 2010). Interestingly, the representative regions were almost exclusively located in those 222 areas where the CDC calculates the East Asian Summer/Winter Monsoons Indices (http://cmdp.ncc– 223 cma.net/Monitoring/monsoon.htm). Further details of these indices are provided by Shi and Zhu (1996) and Zhang et al. (2003). 224 The representative regions identified by the RF for GH850 and SLP are mainly located in the east and south of the Tibetan 218 Plateau and the Northwest Pacific. In addition, the representative regions in the north are sparser than in the east and south, 219 which means the RF could not effectively extract key information from the atmospheric variables. Importantly, the East Asian 220 Summer/Winter Monsoons are fundamental factors affecting droughts and floods in China (Ding et al. 2008; Wen et al. 2000; 221 Yihui and Chan 2005; Zhou and Wu 2010). Interestingly, the representative regions were almost exclusively located in those 222 areas where the CDC calculates the East Asian Summer/Winter Monsoons Indices (http://cmdp.ncc– 223 cma.net/Monitoring/monsoon.htm). Further details of these indices are provided by Shi and Zhu (1996) and Zhang et al. (2003). 224 12 g ) p y ( ) g ( Fig. 4.4 Prediction skills across all months 202 Figure 6 shows the SPI3 correlation skills for the UKMO and D-ANN models for all target months. Both underperform 203 between September and December. To show the differences between the two models, the skills based on D-ANN were 204 subtracted from the UKMO outputs (Fig. 6(e–i)). Compared to the UKMO outputs, D-ANN performed poorly from October 205 to January but was more skillful across the spring and summer months. For lead times exceeding 30 days, D-ANN has a clear 206 advantage. However, the correlation skills varied considerably among the regions, with higher skills in the east and south 207 regions compared to the north. 208 Figure 6 shows the SPI3 correlation skills for the UKMO and D-ANN models for all target months. Both underperform 203 between September and December. To show the differences between the two models, the skills based on D-ANN were 204 subtracted from the UKMO outputs (Fig. 6(e–i)). Compared to the UKMO outputs, D-ANN performed poorly from October 205 to January but was more skillful across the spring and summer months. For lead times exceeding 30 days, D-ANN has a clear 206 advantage. However, the correlation skills varied considerably among the regions, with higher skills in the east and south 207 regions compared to the north. 208 regions compared to the north. 08 09 Fig. 6 Comparison of the correlation skills of UKMO and D-ANN models in different target months in the study regions: a–c are 10 Corr(UKMO), d–f are Corr(D-ANN), and e–i are Corr(D-ANN) – Corr(UKMO). Shaded areas represent Corr > 0.5. Horizontal 11 coordinates indicate different lead times (pentads); the first pentad includes the 1–5-day predictions, the second includes 6–10-day 12 predictions, and so on. Vertical coordinates are target months 13 209 Fig. 6 Comparison of the correlation skills of UKMO and D-ANN models in different target months in the study regions: a–c are 210 Corr(UKMO), d–f are Corr(D-ANN), and e–i are Corr(D-ANN) – Corr(UKMO). Shaded areas represent Corr > 0.5. Horizontal 211 coordinates indicate different lead times (pentads); the first pentad includes the 1–5-day predictions, the second includes 6–10-day 212 predictions, and so on. Vertical coordinates are target months 213 11 an Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter nsoon Index Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter 230 Monsoon Index 231 4.5 Interpretation of prediction mechanisms 214 7 Interpretations of the physical mechanisms of the D-ANN model: a–c, three droughts in the north, east, and south regions (black and red curves represent observations and D-ANN predictions, respectively, and blue dots are initial prediction moments); d–i, colors represent the contribution of the D-ANN model extracted from each grid, with positive values in red and negative values in blue; d–f, the two black boxes (10°–20° N, 100°–150° E and 25°–35° N, 100°–150° E) show the calculation areas of East Fig. 7 Interpretations of the physical mechanisms of the D-ANN model: a–c, three droughts in the north, east, and south regions 226 (black and red curves represent observations and D-ANN predictions, respectively, and blue dots are initial prediction moments); 227 d–i, colors represent the contribution of the D-ANN model extracted from each grid, with positive values in red and negative 228 values in blue; d–f, the two black boxes (10°–20° N, 100°–150° E and 25°–35° N, 100°–150° E) show the calculation areas of East 229 12 Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter 230 Monsoon Index 231 Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter 230 Monsoon Index 231 5.1 Network structure 248 All of our results are based on the single-layer, 16-node ANN model described in Sect. 3. In this discussion, we focus on two 249 important parameters in the ANN model, namely the number of layers and nodes. 250 All of our results are based on the single-layer, 16-node ANN model described in Sect. 3. In this discussion, we focus on two 249 important parameters in the ANN model, namely the number of layers and nodes. 250 4.6 Predicting drought process 232 We divided individual drought periods into three phases of onset, persistence, and relief. For example, for the drought from 233 15 July 2014 to 27 September 2014, the onset, persistence, and relief periods correspond to 15 July–09 August, 10 August–3 234 September, and 4 September–27 September, respectively. Fig. 8 shows the analysis of all droughts in the east study region 235 from 1993 to 2016, and results for the north and south regions are included in Supplement 4. 236 The UKMO and D-ANN models showed different prediction abilities for all drought phases. Importantly, effective prediction 237 of the onset phases of droughts is crucial for drought resistance. The UKMO model performed better for lead times < 45 days, 238 while D-ANN performed better with lead times > 45 days. For the drought persistence phase, D-ANN always produced better 239 predictions, with correlation values of 0.59 with a 60-day lead time, indicating that this model is better at predicting drought 240 intensity. However, D-ANN predictions during the relief phases of droughts were poorer with lead times > 45 days, which 241 likely reflects the requirement for large precipitation inputs during this phase. 242 43 Fig. 8 Predictions of drought phases with different lead times (days) in the east study region of China: UKMO and D-ANN outputs 44 are shown in blue and red, respectively; a–e, drought onset phase; f–j, drought persistence phase; k–o, drought relief phase. Corr 45 and RMSE represent correlation coefficient and root-mean-square error, respectively 46 Fig. 8 Predictions of drought phases with different lead times (days) in the east study region of China: UKMO and D-ANN outputs 244 are shown in blue and red, respectively; a–e, drought onset phase; f–j, drought persistence phase; k–o, drought relief phase. Corr 245 and RMSE represent correlation coefficient and root-mean-square error, respectively 246 13 5.1.1 Hidden layer nodes: more or less? 251 If hidden layer nodes are too few, it is difficult for a network to learn and acquire information-processing capabilities. 252 Conversely, too many nodes in a hidden layer result in over-fitting (Zou et al. 2009). We trained the model by modifying the 253 hidden nodes in the network while maintaining the number of layers at 1, the results of which are shown in Table 3. We found 254 that when fewer than 16 nodes are used, model performance was compromised, and when more than 16 nodes are included, 255 the model provides greater complexity but does not out-perform the validation dataset. 256 Table 3 Performance of the ANN model with different numbers of nodes Regions Metrics Nodes 4 8 16 32 64 North Corr Training 0.84 0.95 0.97 0.95 0.98 Validation 0.69 0.70 0.74 0.70 0.70 RMSE Training 0.58 0.33 0.25 0.31 0.22 Validation 0.67 0.65 0.64 0.66 0.66 East Corr Training 0.83 0.95 0.96 0.97 0.93 Validation 0.82 0.83 0.86 0.85 0.82 RMSE Training 0.60 0.34 0.31 0.24 0.37 Validation 0.61 0.61 0.55 0.58 0.58 South Corr Training 0.94 0.94 0.96 0.95 0.93 Validation 0.76 0.77 0.78 0.78 0.77 RMSE Training 0.36 0.35 0.28 0.32 0.39 Validation 0.68 0.69 0.65 0.66 0.64 5.1.2 Hidden layers: deep or shallow? 258 Regions Metrics Nodes 4 8 16 32 64 North Corr Training 0.84 0.95 0.97 0.95 0.98 Validation 0.69 0.70 0.74 0.70 0.70 RMSE Training 0.58 0.33 0.25 0.31 0.22 Validation 0.67 0.65 0.64 0.66 0.66 East Corr Training 0.83 0.95 0.96 0.97 0.93 Validation 0.82 0.83 0.86 0.85 0.82 RMSE Training 0.60 0.34 0.31 0.24 0.37 Validation 0.61 0.61 0.55 0.58 0.58 South Corr Training 0.94 0.94 0.96 0.95 0.93 Validation 0.76 0.77 0.78 0.78 0.77 RMSE Training 0.36 0.35 0.28 0.32 0.39 Validation 0.68 0.69 0.65 0.66 0.64 5.1.2 Hidden layers: deep or shallow? 258 The neural network learns from the data through the nodes in each layer, and the stacking of multiple layers allows the structure 259 of complex applications to be learned (Pan 2019). However, it remains questionable a deeper network (i.e. with more layers) 260 produced a stronger model (Tan 2020). To test this, keeping the number of nodes at 16, we varied the number of layers, as 261 shown in Table 4. Based on this, we found that network performance did not improve as the number of layers was increased. 262 From the analyses in Sec. 5.1.1 and Sec. 5.1.2, we found that the relationship between contemporaneous atmospheric 263 circulation parameters and regional meteorological drought may not be overly complex, meaning that a single-layer, 16-node 264 ANN model was able to effectively represent nonlinear relationships. 265 Table 4 Performance of ANN models with different numbers of layers 266 Regions Metrics Layers 1 2 3 The neural network learns from the data through the nodes in each layer, and the stacking of multiple layers allows the structure 259 of complex applications to be learned (Pan 2019). However, it remains questionable a deeper network (i.e. with more layers) 260 produced a stronger model (Tan 2020). To test this, keeping the number of nodes at 16, we varied the number of layers, as 261 shown in Table 4. Based on this, we found that network performance did not improve as the number of layers was increased. 262 From the analyses in Sec. 5.1.1 and Sec. 5.1.2, we found that the relationship between contemporaneous atmospheric 263 circulation parameters and regional meteorological drought may not be overly complex, meaning that a single-layer, 16-node 264 ANN model was able to effectively represent nonlinear relationships. 5.2.1 Anomaly correlation coefficients (ACC) of forecasted atmospheric variables 268 5.2.1 Anomaly correlation coefficients (ACC) of forecasted atmospheric variables 268 5.2.1 Anomaly correlation coefficients (ACC) of forecasted atmospheric variables 268 We used the forecasted atmospheric circulation dataset from the ECMWF as a predictor to force the hybrid models. Therefore, 269 the models contained two kinds of error, first from the ML models and, second, from the GCM. This section focuses on the 270 impact of the second error source. Fig. 9 shows that the ACC of all variables decreased with increasing lead time, which 271 indicates that the error from the ECMWF dataset introduced a lot of uncertainty to the prediction. Indeed, forecasting skills 272 are almost entirely lost for lead times above 50 days. 273 274 Fig. 9 Boxplots of ACC for all atmospheric variables (30°–180° E, 10°S–90° N) with increasing lead times. The top to bottom five 275 lines of a box plot represent the minimum, first quartile, median, third quartile, and maximum values of data, respectively 276 Fig. 9 Boxplots of ACC for all atmospheric variables (30°–180° E, 10°S–90° N) w 275 276 Fig. 9 Boxplots of ACC for all atmospheric variables (30°–180° E, 10°S–90° N) with increasing lead times. The top to bottom five 275 lines of a box plot represent the minimum, first quartile, median, third quartile, and maximum values of data, respectively 276 5.2.2 Perfect drought prediction 277 5.1.2 Hidden layers: deep or shallow? 258 265 14 North Corr Training 0.97 0.97 0.98 Validation 0.74 0.70 0.73 RMSE Training 0.25 0.30 0.19 Validation 0.64 0.65 0.63 East Corr Training 0.96 0.95 0.98 Validation 0.86 0.77 0.82 RMSE Training 0.31 0.39 0.31 Validation 0.55 0.67 0.69 South Corr Training 0.96 0.81 0.98 Validation 0.78 0.65 0.75 RMSE Training 0.28 0.62 0.24 Validation 0.65 0.78 0.72 5.2 Impact of forecast error from the ECMWF dataset 267 North Corr Training 0.97 0.97 0.98 Validation 0.74 0.70 0.73 RMSE Training 0.25 0.30 0.19 Validation 0.64 0.65 0.63 East Corr Training 0.96 0.95 0.98 Validation 0.86 0.77 0.82 RMSE Training 0.31 0.39 0.31 Validation 0.55 0.67 0.69 South Corr Training 0.96 0.81 0.98 Validation 0.78 0.65 0.75 RMSE Training 0.28 0.62 0.24 Validation 0.65 0.78 0.72 We replaced all of the atmospheric data for the prospective 90 days with reanalysis data to re-force the hybrid models, which 278 we named ‘Perfect-D-Lasso’ and ‘Perfect-D-ANN’ (Fig. 10). The correlation skills of the Perfect hybrid models were almost 279 we named ‘Perfect-D-Lasso’ and ‘Perfect-D-ANN’ (Fig. 10). The correlation 279 5.2.2 Perfect drought prediction 277 ced all of the atmospheric data for the prospective 90 days with reanalysis data to re-force the hybrid models, which d ‘Perfect-D-Lasso’ and ‘Perfect-D-ANN’ (Fig. 10). The correlation skills of the Perfect hybrid models were almost 15 always above 0.8. Moreover, the performance of the models differed between the study regions, with the Perfect-D-ANN 280 model performing best in the north region. Methods for post-processing GCM precipitation outputs have also been included 281 in the dynamical-statistical prediction methods (Schepen et al. 2016), and so it is necessary to correct the forecasted 282 atmospheric variables accordingly. 283 model performing best in the north region. Methods for post-processing GCM precipitation outputs have also been included 281 in the dynamical-statistical prediction methods (Schepen et al. 2016), and so it is necessary to correct the forecasted 282 atmospheric variables accordingly. 283 284 Fig. 10 Correlation skills of four models predicting the SPI3 with different lead times in the different study regions. The solid 285 green curves represent D-Lasso model predictions; the solid red curves represent D-ANN model predictions; the dashed green 286 curves represent Perfect-D-Lasso model predictions; and the dashed red curves represent the Perfect-D-ANN model predictions. 287 The dashed grey lines indicate a correlation of 0.5 288 84 Fig. 10 Correlation skills of four models predicting the SPI3 with different lead times in the different study regions. The solid 285 green curves represent D-Lasso model predictions; the solid red curves represent D-ANN model predictions; the dashed green 286 curves represent Perfect-D-Lasso model predictions; and the dashed red curves represent the Perfect-D-ANN model predictions. 287 The dashed grey lines indicate a correlation of 0.5 288 Acknowledgments 316 This work is supported by National Natural Science Foundation of China (grants No. 51779071); National Key Research and 317 Development Program of China (Grants No. 2017YFC1502403 and 2018YFC0407701). 318 6 Conclusions 289 51779071); National Key Resea 317 1980–2011 and validated for the period 2012–2019, demonstrating good model robustness; (4) hybrid models (D-Lasso and 296 D-ANN) were constructed by combining the ECMWF and ML models; (5) month-by-month drought prediction was performed 297 for the period between 1993 and 2016. Compared to the dynamical models (Météo France, ECMWF, and UKMO), D-ANN 298 not only extended the forecast skill horizons in all regions but offers higher predictive abilities in spring and summer; (6) we 299 quantified the contribution of representative regions extracted using a RF approach, finding that these areas mainly correspond 300 to those areas where the CDC calculates the East Asian Summer/Winter Monsoon indices; and (7) the D-ANN model is more 301 skillful at predicting the onset and persistence phases of drought but performs comparatively poorly during the relief phase. 302 Important limitations of this study are that the selected predictors fail to incorporate the precursors of drought, such as sea 303 surface temperature, sea ice, and plateau snow cover (Li et al. 2018a; Schubert et al. 2016; Wang et al. 2017), and second, the 304 atmospheric variable outputs from the ECMWF model could not be adequately post-processed. 305 Declarations 306 Competing interests The authors declare no competing interests. 307 308 Availability of data and material All the datasets used in this paper are publicly accessible. 309 310 Authors' contributions ZW, with the help of HY and HH, designed the research. ZW and HH obtain 311 it. ZW conducted case analysis and wrote the original draft. HY, HH, and YL jointly reviewed t 312 authors finally read and approved the paper. 313 314 Consent for publication Not applicable 315 Availability of data and material All the datasets used in this paper are publicly accessible. 309 Authors' contributions ZW, with the help of HY and HH, designed the research. ZW and HH obtained the funding supporting 311 it. ZW conducted case analysis and wrote the original draft. HY, HH, and YL jointly reviewed the entire manuscript. All 312 authors finally read and approved the paper. 313 Agana N, Homaifar A (2017) A Deep Learning Based Approach for Long-Term Drought Prediction. doi:10.1109/SECON.2017.7925314 320 Ayantobo OO, Li Y, Song S, Yao N (2017) Spatial comparability of drought characteristics and related return periods in mainland China 321 over 1961–2013 Journal of Hydrology 550:549-567 doi:https://doi.org/10.1016/j.jhydrol.2017.05.019 322 Buizza R, Leutbecher M (2015) The forecast skill horizon Quarterly Journal of the Royal Meteorological Society 141:3366-3382 323 Cai R, Tan H, Kontoyiannis H (2017) Robust surface warming in offshore China Seas and its relationship to the East Asian Monsoon wind 324 field and ocean forcing on interdecadal time scales Journal of Climate 30:8987-9005 325 6 Conclusions 289 Seasonal drought prediction is important for drought resistance and water resources management. Here, we constructed 290 seasonal drought prediction models with atmospheric variables (GH200, GH500, GH850, T2M, and SLP) as predictors and 291 the SPI3 as the predictand. The resulting models were applied and evaluated for the prediction of meteorological droughts in 292 north, east, and south regions of China. The main conclusions can be summarized as follows: (1) the daily-updated SPI3 was 293 used to identify droughts for the last 40 years in the three drought regions; (2) the five-layer atmospheric variables were 294 processed as model predictors; (3) the Lasso and ANN models were trained using the ERA5 reanalysis dataset for the period 295 16 1980–2011 and validated for the period 2012–2019, demonstrating good model robustness; (4) hybrid models (D-La 296 D-ANN) were constructed by combining the ECMWF and ML models; (5) month-by-month drought prediction was pe 297 for the period between 1993 and 2016. Compared to the dynamical models (Météo France, ECMWF, and UKMO), 298 not only extended the forecast skill horizons in all regions but offers higher predictive abilities in spring and summer 299 quantified the contribution of representative regions extracted using a RF approach, finding that these areas mainly cor 300 to those areas where the CDC calculates the East Asian Summer/Winter Monsoon indices; and (7) the D-ANN model 301 skillful at predicting the onset and persistence phases of drought but performs comparatively poorly during the relief p 302 Important limitations of this study are that the selected predictors fail to incorporate the precursors of drought, suc 303 surface temperature, sea ice, and plateau snow cover (Li et al. 2018a; Schubert et al. 2016; Wang et al. 2017), and sec 304 atmospheric variable outputs from the ECMWF model could not be adequately post-processed. 305 Declarations 306 Competing interests The authors declare no competing interests. 307 308 Availability of data and material All the datasets used in this paper are publicly accessible. 309 310 Authors' contributions ZW, with the help of HY and HH, designed the research. ZW and HH obtained the funding sup 311 it. ZW conducted case analysis and wrote the original draft. HY, HH, and YL jointly reviewed the entire manuscr 312 authors finally read and approved the paper. 313 314 Consent for publication Not applicable 315 Acknowledgments 316 This work is supported by National Natural Science Foundation of China (grants No. Buizza R, Leutbecher M (2015) The forecast skill horizon Quarterly Journal of the Royal Meteorological Society 141:3366 3382 323 Cai R, Tan H, Kontoyiannis H (2017) Robust surface warming in offshore China Seas and its relationship to the East Asian Monsoon wind 324 field and ocean forcing on interdecadal time scales Journal of Climate 30:8987-9005 325 Agana N, Homaifar A (2017) A Deep Learning Based Approach for Long-Term Drought Prediction. doi:10.1109/SECON.2017.7925314 320 Ayantobo OO, Li Y, Song S, Yao N (2017) Spatial comparability of drought characteristics and related return periods in mainland China 321 over 1961–2013 Journal of Hydrology 550:549-567 doi:https://doi.org/10.1016/j.jhydrol.2017.05.019 322 Buizza R, Leutbecher M (2015) The forecast skill horizon Quarterly Journal of the Royal Meteorological Society 141:3366-3382 323 Cai R Tan H Kontoyiannis H (2017) Robust surface warming in offshore China Seas and its relationship to the East Asian Monsoon wind 324 References 319 Part I: Observed evidences International Journal of Climatology: A Journal of the Royal Meteorological Society 337 28:1139-1161 338 Duan W, He B, Takara K, Luo P, Nover D, Yamashiki Y, Huang W (2014) Anomalous atmospheric events leading to Kyushu's flash floods, 339 July 11–14, 2012 Natural Hazards 73:1255-1267 340 Hao Z, Hao F, Xia Y, Singh VP, Hong Y, Shen X, Ouyang W (2016) A statistical m 341 NLDAS-2 Journal of Applied Meteorology and Climatology 55:1049-1061 342 Hao Z, Hao F, Xia Y, Singh VP, Hong Y, Shen X, Ouyang W (2016) A statistical method for categorical drought prediction based on 341 NLDAS-2 Journal of Applied Meteorology and Climatology 55:1049-1061 342 Hao Z, Singh V, Xia Y (2018) Seasonal Drought Prediction: Advances, Challenges 343 doi:10.1002/2016rg000549 344 Hao Z, Singh V, Xia Y (2018) Seasonal Drought Prediction: Advances, Challenges, and Future Prospects Reviews of Geophysics 56 343 doi:10.1002/2016rg000549 344 Hart RE, Grumm RH (2001) Using Normalized Climatological Anomalies to Rank Synoptic-Scale Events Objectively Monthly Weather 345 Review 129:2426-2442 346 Jin D, Guan Z, Tang W (2013) The Extreme Drought Event during Winter–Spring of 2011 in East China: Combined Influences of 347 Teleconnection in Midhigh Latitudes and Thermal Forcing in Maritime Continent Region Journal of Climate 26:8210-8222 348 doi:10.1175/JCLI-D-12-00652.1 349 Khan N, Sachindra DA, Shahid S, Ahmed K, Shiru MS, Nawaz N (2020) Prediction of droughts over Pakistan using machine learning 350 algorithms Advances in Water Resources 139:103562 351 Kirtman B, Anderson D, Brunet G, Kang I-S, Scaife AA, Smith D (2013) Prediction from weeks to decades. In: Climate science for serving 352 society. Springer, pp 205-235 353 Luo L, Ye A, Duan Q (2020) Do CFSv2 Seasonal Forecasts Help Improve the Forecast of Me China? Water 12:2010 Lang Y, Luo L, Ye A, Duan Q (2020) Do CFSv2 Seasonal Forecasts Help Improve the 354 China? Water 12:2010 355 Lang Y, Luo L, Ye A, Duan Q (2020) Do CFSv2 Seasonal Forecasts Help Improve the Forecast of Meteorological Drought over Mainland 354 China? Water 12:2010 355 Lang Y et al. References 319 (2014) Evaluating Skill of Seasonal Precipitation and Temperature Predictions of NCEP CFSv2 Forecasts over 17 356 Hydroclimatic Regions in China Journal of Hydrometeorology 15:1546-1559 357 Lavaysse C, Vogt J, Toreti A, Carrera ML, Pappenberger F (2018) On the use of weather regimes to forecast meteorological drought over 358 Europe Nat Hazards Earth Syst Sci 18:3297-3309 doi:10.5194/nhess-18-3297-2018 359 Lhotka O, Trnka M, Kyselý J, Markonis Y, Balek J, Možný M (2020) Atmospheric circulation as a factor contributing to increasing drought 360 severity in Central Europe Journal of Geophysical Research: Atmospheres 125:e2019JD032269 361 6 , , y ý , , , ý ( ) p g g g severity in Central Europe Journal of Geophysical Research: Atmospheres 125:e2019JD032269 361 Li H, Chen H, Wang H, Sun J, Ma J (2018a) Can Barents Sea ice decline in spring enhance summer hot drought events over northeastern 362 China? Journal of Climate 31:4705-4725 363 China? References 319 Journal of Climate 31:4705-4725 363 Li Q, Yang S, Wu T, Liu X (2017) Subseasonal Dynamical Prediction of East Asian Co 364 Li Q, Yang S, Wu T, Liu X (2017) Subseasonal Dynamical Prediction of East Asian Cold Surges Weather & Forecasting 32 364 Li X, Li D, Li X, Chen L (2018b) Prolonged seasonal drought events over northern China and their possible causes International Journal of 365 Climatology 38:4802-4817 366 Li Q, Yang S, Wu T, Liu X (2017) Subseasonal Dynamical Prediction of East Asian Cold Surges Weather & Forecasting 32 364 Li X, Li D, Li X, Chen L (2018b) Prolonged seasonal drought events over northern China and their possible causes International Journal of 365 Climatology 38:4802-4817 366 Li X, Li D, Li X, Chen L (2018b) Prolonged seasonal drought events over northern Chi 365 Climatology 38:4802-4817 366 gy Li Y, Wu Z, He H, Wang QJ, Xu H, Lu G (2020) Post-processing sub-seasonal precipitation forecasts at various spatiotemporal scales across 367 China during boreal summer monsoon Journal of Hydrology:125742 doi:https://doi.org/10.1016/j.jhydrol.2020.125742 368 Li Y, Wu Z, He H, Wang QJ, Xu H, Lu G (2020) Post-processing sub-seasonal precipitation forecasts at various spatiotemporal scales across 367 China during boreal summer monsoon Journal of Hydrology:125742 doi:https://doi.org/10.1016/j.jhydrol.2020.125742 368 Li Y, Wu Z, He H, Wang QJ, Xu H, Lu G (2020) Post-processing sub-seasonal precipita 367 China during boreal summer monsoon Journal of Hydrology:125742 doi:http 368 Liu Y, Huang R (2019) Linkages between the south and east asian monsoon water vapor transport during boreal summer Journal of Climate 369 32:4509-4524 370 Liu Y, Huang R (2019) Linkages between the south and east asian monsoon water vapor transport during boreal summer Journal of Climate 369 32:4509-4524 370 Liu Z, Lu G, He H, He J (2018) A conceptual prediction model for seasonal drought processes using atmospheric and oceanic standardized 371 anomalies: application to regional drought processes in China Hydrology and Earth System Sciences 22:529-546 372 Liu Z, Lu G, He H, He J (2018) A conceptual prediction model for seasonal drought processes using atmospheric and oceanic standardized 371 anomalies: application to regional drought processes in China Hydrology and Earth System Sciences 22:529-546 372 , , , ( ) p p g p g p anomalies: application to regional drought processes in China Hydrology and Earth System Sciences 22:529-546 372 pp g g p y gy Liu Z, Lu G, He H, Wu Z, He J (2017) Anomalous features of water vapor transpor 373 Southwest China Water 9:244 374 Liu Z, Lu G, He H, Wu Z, He J (2017) Anomalous features of water vapor transport during severe summer and early fall droughts in 373 Southwest China Water 9:244 374 Liu Z, Lu G, He H, Wu Z, He J (2017) Anomalous features of water vapor transport during severe summer and early fall droughts in 373 Southwest China Water 9:244 374 Liu Z, Törnros T, Menzel L (2016) A probabilistic prediction network for hydrolog 375 assessment Water Resources Research 52:6243-6262 376 Ma F, Yuan X, ye A (2015) Seasonal Drought Predictability and Forecast Skill over China Journal of Geophysical Research: Atmospheres 377 120:n/a-n/a doi:10.1002/2015JD023185 378 Ma F, Yuan X, ye A (2015) Seasonal Drought Predictability and Forecast Skill over Ch 377 120:n/a-n/a doi:10.1002/2015JD023185 378 References 319 17 Chen S, Zhang L, Liu X, Guo M, She D (2018) The Use of SPEI and TVDI to Assess Temporal-Spatial Variations in Drought Conditions 326 in the Middle and Lower Reaches of the Yangtze River Basin, China Advances in Meteorology 2018:9362041 327 doi:10.1155/2018/9362041 328 Chen S, Zhang L, Liu X, Guo M, She D (2018) The Use of SPEI and TVDI to Assess Temporal-Spatial Variations in Drought Conditions 326 in the Middle and Lower Reaches of the Yangtze River Basin, China Advances in Meteorology 2018:9362041 327 Chen S, Zhang L, Liu X, Guo M, She D (2018) The Use of SPEI and TVDI to Assess 326 China Meteorological Administration (2011) China Meteorological Disaster Yearbook (CMDY) China Meteorological Disaster Yearbook 329 vol 978-7-5029-5446-8. Meteorological Press, Beijing, China 330 China Meteorological Administration (2014) China Meteorological Disaster Yearb 331 vol 978-7-5029-5556-4. Meteorological Press, Beijing, China 332 Ahmed S, Markova G (2018) Impact of disasters and crises on agriculture and food security, 2 Conforti P, Ahmed S, Markova G (2018) Impact of disasters and crises on a 333 Conforti P, Ahmed S, Markova G (2018) Impact of disasters and crises on agriculture and food security, 2017 333 Costa-Cabral M, Rath JS, Mills WB, Roy SB, Bromirski PD, Milesi C (2016) Projecting and forecasting winter precipitation extremes and 334 meteorological drought in California using the North Pacific high sea level pressure anomaly Journal of Climate 29:5009-5026 335 Ding Y, Wang Z, Sun Y (2008) Inter‐decadal variation of the summer precipitation in East China and its association with decreasing Asian 336 summer monsoon. Part I: Observed evidences International Journal of Climatology: A Journal of the Royal Meteorological Society 337 28 1139 1161 338 Costa-Cabral M, Rath JS, Mills WB, Roy SB, Bromirski PD, Milesi C (2016) Projecting and forecasting winter precipitation extremes and 334 meteorological drought in California using the North Pacific high sea level pressure anomaly Journal of Climate 29:5009-5026 335 Ding Y, Wang Z, Sun Y (2008) Inter‐decadal variation of the summer precipitation in East China and its association with decreasing Asian 336 summer monsoon. 6ee9bcb0725b. Ujeneza EL, Abiodun BJ (2015) Drought regimes in Southern Africa and how well GCMs simulate them Climate Dynamics 44:1595-1609 410 Wang C, Yang K, Li Y, Wu D, Bo Y (2017) Impacts of spatiotemporal anomalies of Tibetan Plateau snow cover on summer precipitation 411 in eastern China Journal of Climate 30:885-903 412 Ujeneza EL, Abiodun BJ (2015) Drought regimes in Southern Africa and how well GCMs sim j , ( ) g g y Wang C, Yang K, Li Y, Wu D, Bo Y (2017) Impacts of spatiotemporal anomalies of Tibetan Plateau snow cover on summer precipitation 411 in eastern China Journal of Climate 30:885-903 412 Wang Q et al. (2014) Temporal-spatial characteristics of severe drought events and their impact on agriculture on a global scale Quaternary 413 International 349:10-21 414 Wang Q et al. (2014) Temporal-spatial characteristics of severe drought events and their impact on agriculture on a global scale Quaternary 413 International 349:10-21 414 Wang QJ, Schepen A, Robertson DE (2012) Merging seasonal rainfall forecasts from multiple statistical models through Bayesian model 415 averaging Journal of Climate 25:5524-5537 416 Wang QJ, Schepen A, Robertson DE (2012) Merging seasonal rainfall forecasts from multiple statistical models through Bayesian model 415 averaging Journal of Climate 25:5524-5537 416 Wen C, Graf HF, Ronghui H (2000) The interannual variability of East Asian Winter Monsoon and its relation to the summer monsoon 417 Advances in Atmospheric Sciences 17:48-60 doi:10.1007/s00376-000-0042-5 418 Wen C, Graf HF, Ronghui H (2000) The interannual variability of East Asian Winter Monsoon and its relation to the summer monsoon 417 Advances in Atmospheric Sciences 17:48-60 doi:10.1007/s00376-000-0042-5 418 World Health Organization (2019) Drought. 120:n/a-n/a doi:10.1002/2015JD023185 120:n/a-n/a doi:10.1002/2015JD023185 378 Ma S, Zhu C, Liu J (2020) Combined Impacts of Warm Central Equatorial Pacific Sea Surface Temperatures and Anthropogenic Warming 379 on the 2019 Severe Drought in East China Advances in Atmospheric Sciences 37:1149-1163 doi:10.1007/s00376-020-0077-8 380 Ma S, Zhu C, Liu J (2020) Combined Impacts of Warm Central Equatorial Pacific Sea Surface Temperatures and Anthropogenic Warming 379 on the 2019 Severe Drought in East China Advances in Atmospheric Sciences 37:1149-1163 doi:10.1007/s00376-020-0077-8 380 18 Madadgar S, Aghakouchak A, Shukla S, Wood AW, Cheng L, Hsu K-L, Svoboda M (2016) A hybrid statistical-dynamical framework for 381 meteorological drought prediction: Application to the southwestern United States Water Resources Research 52:5095-5110 382 Merryfield WJ et al. 6ee9bcb0725b. https://www.who.int/health-topics/drought 419 g ( ) g p p g Xu L, Chen N, Zhang X (2018a) A comparison of large-scale climate signals and the North American Multi-Model Ensemble (NMME) for 420 drought prediction in China Journal of Hydrology 557:378-390 doi:https://doi.org/10.1016/j.jhydrol.2017.12.044 421 g ( ) g p p g Xu L, Chen N, Zhang X (2018a) A comparison of large-scale climate signals and the North American Multi-Model Ensemble (NMME) for 420 drought prediction in China Journal of Hydrology 557:378-390 doi:https://doi.org/10.1016/j.jhydrol.2017.12.044 421 Xu L, Chen N, Zhang X (2018a) A comparison of large-scale climate signals and the N 420 drought prediction in China Journal of Hydrology 557:378-390 doi:https://do 421 on in China Journal of Hydrology 557:378-390 doi:https://doi.org/10.1016/j.jhydrol.2017.12.04 Xu L, Chen N, Zhang X, Chen Z (2018b) An evaluation of statistical, NMME and hybrid models for drought prediction in China Journal of 422 Hydrology 566:235-249 423 Xu L, Chen N, Zhang X, Chen Z (2018b) An evaluation of statistical, NMME and hybrid models for drought prediction in China Journal of 422 Hydrology 566:235-249 423 y gy Yan H, Moradkhani H, Zarekarizi M (2017) A probabilistic drought forecasting framework: A combined dynamical and statistical approach 424 Journal of Hydrology 548:291-304 doi:https://doi.org/10.1016/j.jhydrol.2017.03.004 425 Yan H, Moradkhani H, Zarekarizi M (2017) A probabilistic drought forecasting framework: A combined dynamical and statistical approach 424 Journal of Hydrology 548:291-304 doi:https://doi.org/10.1016/j.jhydrol.2017.03.004 425 Yihui D, Chan JCL (2005) The East Asian summer monsoon: an overview Meteorology and Atmospheric Physics 89:117-142 426 doi:10.1007/s00703-005-0125-z 427 Yihui D, Chan JCL (2005) The East Asian summer monsoon: an overview Meteorology and Atmospheric Physics 89:117-142 426 doi:10.1007/s00703-005-0125-z 427 Yoo-Geun H, Jeong-Hwan K, Jing-Jia L (2019) Deep learning for multi-year ENSO forecasts Nature 573 428 Zhang Q, Tao S, Chen L (2003) Interannual variation of East Asian Summer Monsoon index and East Asian atmospheric circulation Acta 429 Meteorologica Sinica:559-568 430 Zhang Q, Tao S, Chen L (2003) Interannual variation of East Asian Summer Monsoon index and East Asian atmospheric circulation Acta 429 Meteorologica Sinica:559-568 430 120:n/a-n/a doi:10.1002/2015JD023185 (2020) Current and emerging developments in subseasonal to decadal prediction Bulletin of the American 383 Meteorological Society 101:E869-E896 384 Mo KC, Lyon B (2015) Global Meteorological Drought Prediction Using the North American Multi-Model Ensemble Journal of 385 Hydrometeorology 16:1409-1424 doi:10.1175/JHM-D-14-0192.1 386 Mo KC, Lyon B (2015) Global Meteorological Drought Prediction Using the No 385 Hydrometeorology 16:1409-1424 doi:10.1175/JHM-D-14-0192.1 386 Okumura YM, DiNezio P, Deser C (2017) Evolving impacts of multiyear La Niña events on atmospheric circulation and US drought 387 Geophysical Research Letters 44:11-614 388 Okumura YM, DiNezio P, Deser C (2017) Evolving impacts of multiyear La Niña 387 Geophysical Research Letters 44:11-614 388 Okumura YM, DiNezio P, Deser C (2017) Evolving impacts of multiyear La Niña events on atmospheric circulation and US drought 387 Geophysical Research Letters 44:11-614 388 p y Pan B (2019) Advancing Precipitation Prediction Using a Composite of Models and Data 9 Pan B, Anderson GJ, Goncalves A, Lucas DD, Bonfils CJW, Lee J (2020) Improving seasonal forecast using probabilistic deep learning 390 arXiv preprint arXiv:201014610 391 Pan B, Anderson GJ, Goncalves A, Lucas DD, Bonfils CJW, Lee J (2020) Improving seasonal forecast using probabilistic deep learning 390 arXiv preprint arXiv:201014610 391 p p Reichstein M, Camps-Valls G, Stevens B, Jung M, Denzler J, Carvalhais N, Prabhat M (2019) Deep learning and process understanding for 392 data-driven Earth system science Nature 566:195 doi:10.1038/s41586-019-0912-1 393 p p Reichstein M, Camps-Valls G, Stevens B, Jung M, Denzler J, Carvalhais N, Prabhat M (2019) Deep learning and process understanding for 392 data-driven Earth system science Nature 566:195 doi:10.1038/s41586-019-0912-1 393 Rezaeianzadeh M, Tabari H (2012) MLP-based drought forecasting in different climatic regions Theoretical and Applied Climatology 109 394 doi:10.1007/s00704-012-0592-3 395 Rezaeianzadeh M, Tabari H (2012) MLP-based drought forecasting in different climatic regions Theoretical and Applied Climatology 109 394 doi:10.1007/s00704-012-0592-3 395 396 doi:10.1007/s00704-012-0592-3 395 Ribeiro AFS, Pires CAL (2016) Seasonal drought predictability in Portugal using statistical–dynamical techniques Physics and Chemistry 396 of the Earth, Parts A/B/C 94:155-166 397 Ribeiro AFS, Pires CAL (2016) Seasonal drought predictability in Portugal using stati 396 of the Earth, Parts A/B/C 94:155-166 397 Ribeiro AFS, Pires CAL (2016) Seasonal drought predictability in Portugal using statistical–dynamical techniques Physics and Chemistry 396 of the Earth, Parts A/B/C 94:155-166 397 Richardson D, Fowler H, Kilsby C, Neal R, Dankers R (2020) Improving sub-seasonal forecast skill of meteorological drought: a weather 398 pattern approach Natural Hazards and Earth System Sciences 20:107-124 doi:10.5194/nhess-20-107-2020 399 Richardson D, Fowler H, Kilsby C, Neal R, Dankers R (2020) Improving sub-seasonal forec pattern approach Natural Hazards and Earth System Sciences 20:107-124 doi:10.51 Richardson D, Fowler H, Kilsby C, Neal R, Dankers R (2020) Improving sub-seasonal forecast skill of meteorological drought: a weather 398 pattern approach Natural Hazards and Earth System Sciences 20:107-124 doi:10.5194/nhess-20-107-2020 399 400 Schepen A, Wang QJ, Robertson DE (2016) Application to Post-processing of Meteorological Seasonal Forecasting. 120:n/a-n/a doi:10.1002/2015JD023185 In: Duan Q, 400 Pappenberger F, Thielen J, Wood A, Cloke HL, Schaake JC (eds) Handbook of Hydrometeorological Ensemble Forecasting. 401 Springer Berlin Heidelberg Berlin Heidelberg pp 1-29 doi:10 1007/978-3-642-40457-3 18-1 402 Schepen A, Wang QJ, Robertson DE (2016) Application to Post processing of Meteorological Seasonal Forecasting. In: Duan Q, 400 Pappenberger F, Thielen J, Wood A, Cloke HL, Schaake JC (eds) Handbook of Hydrometeorological Ensemble Forecasting. 401 Springer Berlin Heidelberg, Berlin, Heidelberg, pp 1-29. doi:10.1007/978-3-642-40457-3_18-1 402 Springer Berlin Heidelberg, Berlin, Heidelberg, pp 1-29. doi:10.1007/978-3-6 402 p g g g pp Schubert SD et al. (2016) Global meteorological drought: a synthesis of current understanding with a focus on SST drivers of precipitation 403 deficits Journal of Climate 29:3989-4019 404 p g g g pp Schubert SD et al. (2016) Global meteorological drought: a synthesis of current understanding with a focus on SST drivers of precipitation 403 deficits Journal of Climate 29:3989-4019 404 Shi N, Zhu Q (1996) An abrupt change in the intensity of the East Asian summer monsoon index and its relationship with temperature and 405 precipitation over East China International Journal of Climatology: A Journal of the Royal Meteorological Society 16:757-764 406 407 Shi N, Zhu Q (1996) An abrupt change in the intensity of the East Asian summer monsoon index and its relationship with temperature and 405 precipitation over East China International Journal of Climatology: A Journal of the Royal Meteorological Society 16:757-764 406 Svoboda M, Hayes M, Wood D (2012) Standardized precipitation index user guide Worl 407 Svoboda M, Hayes M, Wood D (2012) Standardized precipitation index user guide World Meteorological Organization Geneva, Switzerland 407 Tan T (2020) Are Deeper Networks Better? —A Case Study. https://towardsdatascience.com/are-deeper-networks-better-a-case-study- 408 Svoboda M, Hayes M, Wood D (2012) Standardized precipitation index user guide World Meteorological Organization Geneva, Switzerland 407 Tan T (2020) Are Deeper Networks Better? —A Case Study. https://towardsdatascience.com/are-deeper-networks-better-a-case-study- 408 6ee9bcb0725b. 409 Zhu Q, Luo Y, Zhou D, Xu Y-P, Wang G, Tian Y (2020) Drought prediction using in situ and remote sensing products with SVM over the 437 Xiang River Basin, China Natural Hazards doi:10.1007/s11069-020-04394-x 438 Zhu Q, Luo Y, Zhou D, Xu Y-P, Wang G, Tian Y (2020) Drought prediction using in situ and remote sensing products with SVM over the 437 Xiang River Basin, China Natural Hazards doi:10.1007/s11069-020-04394-x 438 Zhuang Y, Fu R, Wang H (2020) Large‐Scale Atmospheric Circulation Patterns Associated with US Great Plains Warm Season Droughts 439 Revealed by Self‐Organizing Maps Journal of Geophysical Research Atmospheres 125 doi:10.1029/2019JD031460 440 Zou W, Li Y, Tang A (2009) Effects of the number of hidden nodes used in a structured-based neural network on the reliability of image 441 classification Neural Computing and Applications 18:249-260 doi:10.1007/s00521-008-0177-3 442 443 Zhuang Y, Fu R, Wang H (2020) Large‐Scale Atmospheric Circulation Patterns Associated with US Great Plains Warm Season Droughts 439 Revealed by Self‐Organizing Maps Journal of Geophysical Research Atmospheres 125 doi:10.1029/2019JD031460 440 y g g p p y p Zou W, Li Y, Tang A (2009) Effects of the number of hidden nodes used in a structured-based neural network on the reliability of image 441 classification Neural Computing and Applications 18:249-260 doi:10.1007/s00521-008-0177-3 442 Meteorologica Sinica:559-568 g Zhang R (2015) Changes in East Asian summer monsoon and summer rainfall over eastern China during recent decades Science Bulletin 431 60:1222-1224 doi:10.1007/s11434-015-0824-x 432 g Zhang R (2015) Changes in East Asian summer monsoon and summer rainfall over eastern China during recent decades Science Bulletin 431 60:1222-1224 doi:10.1007/s11434-015-0824-x 432 Zhang R (2015) Changes in East Asian summer monsoon and summer rainfall over ea 431 60:1222-1224 doi:10.1007/s11434-015-0824-x 432 Zhang R, Chen Z-Y, Xu L-J, Ou C-Q (2019) Meteorological drought forecasting based on a statistical model with machine learning 433 techniques in Shaanxi province, China Science of The Total Environment 665:338-346 434 Zhang R, Chen Z-Y, Xu L-J, Ou C-Q (2019) Meteorological drought forecasting based on a statistical model with machine learning 433 techniques in Shaanxi province, China Science of The Total Environment 665:338-346 434 Zhou LT, Wu R (2010) Respective impacts of the East Asian winter monsoon and ENSO on winter rainfall in China Journal of Geophysical 435 Research: Atmospheres 115 436 Zhou LT, Wu R (2010) Respective impacts of the East Asian winter monsoon and ENSO on winter rainfall in China Journal of Geophysical 435 Research: Atmospheres 115 436 19 20 Figures Figure 1 Figure 1 The location of the drought regions of China and the coverage of atmospheric variables (30°-180°E, 10°S- 90°N). The gray shaded areas are the study regions. 1-3 represent Northern, Eastern, and Southern China, respectively Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. The location of the drought regions of China and the coverage of atmospheric variables (30°-180°E, 10°S- 90°N). The gray shaded areas are the study regions. 1-3 represent Northern, Eastern, and Southern China, respectively Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figure 2 ML model structure, where ANN contains a single layer and 16 nodes. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP are the 90-day standardized anomalies of GH200, GH500, GH850, SLP and T2M, respectively. Lasso is a linear model, while ANN is linear Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figure 2 ML model structure, where ANN contains a single layer and 16 nodes. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP are the 90-day standardized anomalies of GH200, GH500, GH850, SLP and T2M, respectively. Lasso is a linear model, while ANN is linear Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever igure 2 Figure 2 Figure 2 ML model structure, where ANN contains a single layer and 16 nodes. SA90GH200, SA90GH500, SA90GH850, SA90T2M, and SA90SLP are the 90-day standardized anomalies of GH200, GH500, GH850, SLP and T2M, respectively. Lasso is a linear model, while ANN is linear Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figure 3 Figure 3 Schematic representation of the calculation process for the prospective Nth-day SPI3 Schematic representation of the calculation process for the prospective Nth-day SPI3 Schematic representation of the calculation process for the prospective Nth day SPI3 Figure 4 Temporal trends in the observed and simulated SPI3 during the validation period (1 January 2012 to 31 December 2019). The black curve is the observed SPI3, and green and red curves are the Lasso and AN simulations, respectively. The grey shading shows the identi¦ed drought events in Sect. 4.1 Figure 4 Temporal trends in the observed and simulated SPI3 during the validation period (1 January 2012 to 31 December 2019). The black curve is the observed SPI3, and green and red curves are the Lasso and ANN simulations, respectively. The grey shading shows the identi¦ed drought events in Sect. 4.1 gure 5 orrelation skills between ¦ve models predicting SPI3 at different lead times. Black, purple, blue, green d red curves represent Météo France, ECMWF, UKMO, D-Lasso, and D-ANN outputs, respectively. The ey dashed lines indicate a correlation of 0.5 Figure 5 Correlation skills between ¦ve models predicting SPI3 at different lead times. Black, purple, blue, green, and red curves represent Météo France, ECMWF, UKMO, D-Lasso, and D-ANN outputs, respectively. The grey dashed lines indicate a correlation of 0.5 Figure 6 Comparison of the correlation skills of UKMO and D-ANN models in different target months in the study regions: a–c are Corr(UKMO), d–f are Corr(D-ANN), and e–i are Corr(D-ANN) – Corr(UKMO). Shaded are represent Corr > 0.5. Horizontal coordinates indicate different lead times (pentads); the ¦rst pentad includes the 1–5-day predictions, the second includes 6–10-day predictions, and so on. Vertical coordinates are target months Figure 6 Comparison of the correlation skills of UKMO and D-ANN models in different target months in the study regions: a–c are Corr(UKMO), d–f are Corr(D-ANN), and e–i are Corr(D-ANN) – Corr(UKMO). Shaded areas represent Corr > 0.5. Horizontal coordinates indicate different lead times (pentads); the ¦rst pentad includes the 1–5-day predictions, the second includes 6–10-day predictions, and so on. Vertical coordinates are target months Figure 7 Interpretations of the physical mechanisms of the D-ANN model: a–c, three droughts in the north, east, and south regions (black and red curves represent observations and D-ANN predictions, respectively, and blue dots are initial prediction moments); d–i, colors represent the contribution of the D-ANN model extracted from each grid, with positive values in red and negative values in blue; d–f, the two black boxes (10°–20° N, 100°–150° E and 25°–35° N, 100°–150° E) show the calculation areas of East Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter Monsoon Index Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Interpretations of the physical mechanisms of the D-ANN model: a–c, three droughts in the north, east, and south regions (black and red curves represent observations and D-ANN predictions, respectively, and blue dots are initial prediction moments); d–i, colors represent the contribution of the D-ANN model extracted from each grid, with positive values in red and negative values in blue; d–f, the two black boxes (10°–20° N, 100°–150° E and 25°–35° N, 100°–150° E) show the calculation areas of East Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter Monsoon Index Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figure 7 Interpretations of the physical mechanisms of the D-ANN model: a–c, three droughts in the north, east, and south regions (black and red curves represent observations and D-ANN predictions, respectively, and blue dots are initial prediction moments); d–i, colors represent the contribution of the D-ANN model extracted from each grid, with positive values in red and negative values in blue; d–f, the two black boxes (10°–20° N, 100°–150° E and 25°–35° N, 100°–150° E) show the calculation areas of East Asian Summer Monsoon Index; g–i, the black box (10°–50° N, 110°–160° E) shows the calculation area of East Asian Winter Monsoon Index Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figure 8 Figure 8 Predictions of drought phases with different lead times (days) in the east study region of China: UKMO and D-ANN outputs are shown in blue and red, respectively; a–e, drought onset phase; f–j, drought persistence phase; k–o, drought relief phase. Corr and RMSE represent correlation coe¨cient and root- mean-square error, respectively Predictions of drought phases with different lead times (days) in the east study region of China: UKMO and D-ANN outputs are shown in blue and red, respectively; a–e, drought onset phase; f–j, drought persistence phase; k–o, drought relief phase. Corr and RMSE represent correlation coe¨cient and root- mean-square error, respectively mean square error, respectively Figure 9 Boxplots of ACC for all atmospheric variables (30°–180° E, 10°S–90° N) with increasing lead times. The top to bottom ¦ve lines of a box plot represent the minimum, ¦rst quartile, median, third quartile, and maximum values of data, respectively Boxplots of ACC for all atmospheric variables (30°–180° E, 10°S–90° N) with increasing lead times. The top to bottom ¦ve lines of a box plot represent the minimum, ¦rst quartile, median, third quartile, and maximum values of data, respectively p p ( , ) g top to bottom ¦ve lines of a box plot represent the minimum, ¦rst quartile, median, third quartile, and maximum values of data, respectively Figure 10 Correlation skills of four models predicting the SPI3 with different lead times in the different study regions. The solid green curves represent D-Lasso model predictions; the solid red curves represent D-A d l di ti th d h d t P f t D L d l di ti d th d h p y Figure 10 C l ti kill f f d l di ti th SPI3 ith diff t l d ti i th diff t t d Figure 10 Correlation skills of four models predicting the SPI3 with different lead times in the different study regions. The solid green curves represent D-Lasso model predictions; the solid red curves represent D-ANN model predictions; the dashed green curves represent Perfect-D-Lasso model predictions; and the dashed red curves represent the Perfect-D-ANN model predictions. The dashed grey lines indicate a correlation of 0.5 red curves represent the Perfect-D-ANN model predictions. The dashed grey lines indicate a correlation of 0.5 Supplementary Files This is a list of supplementary ¦les associated with this preprint. Click to download. SupplementaryMaterial.docx
https://openalex.org/W2948557996	https://hal.inrae.fr/hal-02886415/document	English	null	Adventitious Virus Detection in Cells by High-Throughput Sequencing of Newly Synthesized RNAs: Unambiguous Differentiation of Cell Infection from Carryover of Viral Nucleic Acids	MSphere	2,019	cc-by	8,598	RESEARCH ARTICLE Applied and Environmental Science RESEARCH ARTICLE Applied and Environmental Science Adventitious Virus Detection in Cells by High-Throughput Sequencing of Newly Synthesized RNAs: Unambiguous Differentiation of Cell Infection from Carryover of Viral Nucleic Acids Justine Cheval,a Erika Muth,a Gaëlle Gonzalez,b Muriel Coulpier,b Pascale Beurdeley,a Stéphane Cr aPathoQuest, Paris, France bUMR1161 Virologie, ANSES, INRA, Ecole Nationale Vétérinaire d’Alfort, Université Paris-12 Est, Maisons-Alfort, Paris, France cEcole Nationale Vétérinaire d’Alfort, Université Paris-Est, Maisons-Alfort, Paris, France dPathogen Discovery Laboratory, Biology of Infection Unit, Institut Pasteur, Paris, France ABSTRACT The use of high-throughput sequencing (HTS) to identify viruses in bio- logicals differs from current molecular approaches, since its use enables an unbiased approach to detection without the need to design speciﬁc primers to preamplify tar- get sequences. Its broad range of detection and analytical sensitivity make it an im- portant tool to ensure that biologicals are free from adventitious viruses. Similar to other molecular methods, however, identiﬁcation of viral sequences in cells by HTS does not prove viral infection, since this could reﬂect carryover of inert viral se- quences from reagents or other sources or the presence of transcriptionally inactive cellular sequences. Due to the broad range of detection associated with HTS, the above can potentially be perceived as a drawback for the testing of pharmaceutical biological products using this method. In order to avoid the identiﬁcation of in- ert viral sequences, we present a methodology based on metabolic RNA labeling and sequencing, which enables the speciﬁc identiﬁcation of newly synthesized viral RNAs in infected cells, resulting in the ability to unambiguously distinguish active infection by DNA or RNA viruses from inert nucleic acids. In the present study, we report the ability to differentiate Vero cells acutely infected by a single-stranded positive-sense RNA virus (tick-borne encephalitis virus) from cells which have been in contact with nonreplicating virus particles. Additionally, we also found a laboratory contamination by the squirrel monkey retrovirus of our Vero cell line, which was derived from an Old World (African green) monkey, a type of contamination which until now has been identiﬁed only in cells derived from primates from the New World. Citation Cheval J, Muth E, Gonzalez G, Coulpier M, Beurdeley P, Cruveiller S, Eloit M. 2019. Adventitious virus detection in cells by high- throughput sequencing of newly synthesized RNAs: unambiguous differentiation of cell infection from carryover of viral nucleic acids. mSphere 4:e00298-19. https://doi.org/10.1128/ mSphere.00298-19. Editor Matthew B. Frieman, University of Maryland, College Park Copyright © 2019 Cheval et al. This is an open- access article distributed under the terms of the Creative Commons Attribution 4.0 International license. Address correspondence to Marc Eloit, marc.eloit@pathoquest.com. Adventitious Virus Detection in Cells by High-Throughput Sequencing of Newly Synthesized RNAs: Unambiguous Differentiation of Cell Infection from Carryover of Viral Nucleic Acids Received 29 April 2019 Accepted 13 May 2019 Published 5 June 2019 Citation Cheval J, Muth E, Gonzalez G, Coulpier M, Beurdeley P, Cruveiller S, Eloit M. 2019. Adventitious virus detection in cells by high- throughput sequencing of newly synthesized RNAs: unambiguous differentiation of cell infection from carryover of viral nucleic acids. mSphere 4:e00298-19. https://doi.org/10.1128/ mSphere.00298-19. IMPORTANCE The use of high-throughput sequencing (HTS) to identify viral con- tamination of biological products is extremely sensitive and provides a broad range of detection. Nevertheless, viral sequences identiﬁed can also be inert. Ex- amples include contamination resulting from reagents or the presence of inacti- vated viruses in animal-derived components of the cell culture medium. We therefore developed a method that relies on the sequencing of newly synthe- sized RNAs, an unequivocal sign of the presence of a transcriptionally active vi- rus. This improvement in the speciﬁcity of viral testing increases the acceptabil- ity of HTS as a standard test for cells used in manufacturing biologicals and in biotherapies. Address correspondence to Marc Eloit, marc.eloit@pathoquest.com. Received 29 April 2019 Accepted 13 May 2019 Published 5 June 2019 KEYWORDS 4-thiouridine, RNA-seq, Vero, biologicals, high-throughput sequencing, metagenomics, squirrel monkey retrovirus, tick-borne encephalitis virus, virus safety msphere.asm.org 1 May/June 2019 Volume 4 Issue 3 e00298-19 Cheval et al. Cheval et al. V iral contamination is a topic of major concern for biological products. This includes both the risk of contamination of good manufacturing practice (GMP) facilities and the ﬁnal drug product. Virus testing of raw materials, cells, and virus seeds is key for the safety of the drug product. This is particularly critical for live vaccines, gene therapy viral vectors, and cell therapy drug products, since their production does not include downstream adventitious virus elimination steps. As a result, the safety of these products relies heavily on viral testing during the production process. All previously reported contaminations of products based on cell cultures were due to unexpected animal viruses that remained unidentiﬁed despite viral testing of raw materials or production cells. The details of these contaminations have been reviewed by others (1). In fact, classical viral testing is limited, since many viruses do not grow in cell lines used for in vitro tests or in rodents or eggs used for in vivo tests (2). Adventitious Virus Detection in Cells by High-Throughput Sequencing of Newly Synthesized RNAs: Unambiguous Differentiation of Cell Infection from Carryover of Viral Nucleic Acids While targeted PCR represents a potential alternative approach, it targets only a small range of known viruses and it lacks the ability to identify the ever-increasing number of newly discov- ered viruses or unknown viruses. V With the above in mind, high-throughput sequencing (HTS)-based viral testing methods are particularly appealing due to their unbiased and wide viral detection range which includes unknown viruses. This is particularly important, as the catalog of animal viruses is still incomplete, an intrinsic limit to any targeted testing method. HTS is a nucleic acid-based technology that makes no a priori assumptions regarding viruses to be detected, since it identiﬁes all viral nucleic acid sequences found in a sample. Similar to other molecular methods like PCR, the identiﬁcation of viral sequences in cells by HTS does not prove viral infection, since many other sources of inert sequences exist. These inert sequences can be exogenous sequences, an example being bovine virus from fetal bovine serum (FBS) even after it has been properly inactivated by gamma irradiation. These sequences can be found associated with cells, because inactivated viruses can still bind to cells. Other sources of inert sequences are present in reagents and can contaminate a sample during the nucleic acid extraction procedure or subsequent steps (3–5). While the discovery of inert viruses is not an issue for targeted PCR, because this method is limited to a few viruses, it represents a major problem when using HTS, as this method has a much broader range of viral detection, a main advantage of the technology. As a result, HTS can potentially identify inert viral sequences leading to unnecessary, lengthy, and resource-consuming investigations to rule out or conﬁrm cellular infection. In order to distinguish the identiﬁcation of inert sequences versus active cell infection, we decided to focus HTS solely on real markers of cell infection for the present study, speciﬁcally the synthesis of RNAs that encode viral proteins and/or regulate the expression of other genes. We have recently demonstrated that sequenc- ing positive-sense or negative-sense cellular viral RNAs that differ from genomic nucleic acids can selectively pinpoint productive or latent viral cell infection (6). In the present paper, we present a streamlined RNA sequencing (RNA-Seq) pipeline based on meta- bolic RNA labeling which enables speciﬁc identiﬁcation of RNAs synthesized in cells and provides unambiguous evidence of active cellular infection versus inert nucleic acids. Adventitious Virus Detection in Cells by High-Throughput Sequencing of Newly Synthesized RNAs: Unambiguous Differentiation of Cell Infection from Carryover of Viral Nucleic Acids We report the results corresponding to Vero cells acutely infected by a positive-sense single-stranded RNA virus (tick-borne encephalitis virus [TBEV]) and the identiﬁcation of the unexpected laboratory persistent contamination of the cells by squirrel monkey retrovirus (SMRV). May/June 2019 Volume 4 Issue 3 e00298-19 RESULTS Identiﬁcation of adventitious viruses by agnostic RNA-Seq in Vero cells. Vero cells were ﬁrst put in contact with a high dose of tick-borne encephalitis virus (TBEV) at 4°C (day 0 [D0]). At this temperature, only passive virus binding to cell receptors occurs, and virus entry, an active event, is blocked. Therefore, this experimental setting mimics the carryover of a nonreplicating virus. RNAs were extracted and sequenced as a marker of DNA or RNA virus infection. The results of the agnostic analysis and those msphere.asm.org 2 May/June 2019 Volume 4 Issue 3 e00298-19 Virus Detection by HTS of Newly Synthesized RNAs TABLE 1 Number of reads on TBEV and SMRV genomes and horizontal coverage of the genomea Virus and parameter Value for parameter at the following time and treatment Day 0, no 4sU Day 1, no 4sU Day 1, 4sU alkylation Day 1, 4sU, no alkylation TBEV No. of readsb 160,085 (0.27) 6,408,291 (0.35) 5,240,211 (0.36) 5,722,338 (0.32) Horizontal coverage (%)c 100 100 100 100 SMRV No. of reads 1,807,960 (0.39) 1,601,090 (0.40) 1,816,788 (0.48) 2,479,933 (0.37) Horizontal coverage (%) 100 100 100 100 aReads were mapped on the genomes of TBEV and SMRV found by the agnostic procedure (see Table S1 in the supplemental material). bThe number of reads shows the number of reads of the virus on the genome. The values in parentheses are the percentages of negative-sense/total number of reads. cHorizontal coverage of the genome is shown as a percentage of the genome. TABLE 1 Number of reads on TBEV and SMRV genomes and horizontal coverage of the genomea of the mapping of the reads against the two main viral hits found by the agnostic analysis (TBEV and squirrel monkey retrovirus [SMRV]) are shown, respectively, in Table S1 in the supplemental material and in Table 1. The main viral species detected at D0 was, as expected, TBEV, but also, unexpectedly, SMRV (Table 1). More than 160,000 TBEV reads out of a total of around 150 million raw reads (Table S1) were identiﬁed, covering the whole genome. Vero cells were then shifted to 37°C to enable virus entry and then incubated for 1 day prior to harvest. The number of reads strongly increased with between 5.2 million and 6.4 million TBEV reads recorded. RESULTS Additionally, between 1.6 and 1.8 million reads mapping to SMRV-H (an SMRV isolated from a human lymphoid cell line [7]) were also identiﬁed independent of the day of harvest. Taking these results together, this means that the SMRV transcripts were expressed by the cells without any relationship to experimental infection by TBEV. We also identiﬁed a number of other hits (Table S1). The main additional hit was baboon endogenous virus, a known endogenous virus of Vero cells (8). We used the human genome as a reference (GRCh37/hg19) for depleting host sequences, which explains why this baboon endogenous virus was detected. A few hundred reads mapping to endogenous human retroviruses were also recorded. In our experience, this ﬁnding is frequent in primate/human cell lines. We also found a few bovine viral diarrhea virus (BVDV) reads typically associated with the use of gamma-irradiated bovine serum. In addition, we identiﬁed a few reads (50) targeted to different herpesviruses which we considered background noise. Differentiation of cell infection versus carryover of inert sequences. Since our primary objective was to mimic challenging conditions for differentiation between cell infection from carryover while testing the ability of HTS to detect early infection of cells, we compared results of cells put in contact with high doses of TBEV blocked for virus replication at 4°C with those of cells infected with the same dose of virus 24 h postinfection. The former mimicked cells with inactivated virus or free nucleic acids, and the latter mimicked cells infected just before being banked. Since TBEV is a positive-sense single-stranded RNA (ssRNA) virus, the negative-sense RNA was used as a marker of virus replication. The three conditions tested at D1 resulted in 0.32% to 0.36% of the reads being negative sense compared to 0.27% at D0, a very small but highly signiﬁcant difference (P 0.0001 by chi-square test) (6). This type of comparative analysis is not relevant for the chronic infection of cells by SMRV, a retrovirus for which transcription uses a DNA provirus as the matrix and leads mainly to positive- but also to negative-sense RNAs (9). We then examined the TBEV rate of T¡C conversion and their distribution along the whole viral genomes following metabolic labeling by 4-thiouridine (4sU) of newly synthesized RNAs (Table 2 and Fig. 1 and 2). May/June 2019 Volume 4 Issue 3 e00298-19 RESULTS At D1 and in the absence of metabolic labeling, the T¡C rate was very low (0.16%) and similar to those of T¡A or T¡A (0.05 to 0.13%), resulting in a calculated background conversion index of 1.78. Similar results were obtained at D0, indicating good reproducibility of the background level of conversion. These background rates of T conversion were distributed along the whole May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 3 Cheval et al. TABLE 2 Conversion rate of T nucleotides and conversion index TABLE 2 Conversion rate of T nucleotides and conversion index Sample and parameter Value for parameter for sample at the following time and treatmenta: Day 0, no 4sU Day 1, no 4sU Day 1, 4sU alkylation Day 1, 4sU, no alkylation TBEV T¡C rate (%) 0.15 0.16/0.16 0.87/0.82 0.16/0.14 T¡A rate (%) 0.05 0.05/0.05 0.10/0.10 0.05/0.05 T¡G (%) 0.11 0.13/0.13 0.18/0.18 0.12/0.12 Conversion index 2.12 1.78/1.76 6.09/5.83 1.81/1.60 SMRV T¡C rate (%) 0.15 0.15/0.11 3.49/3.45 0.15/0.11 T¡A rate (%) 0.05 0.05/0.05 0.10/0.10 0.05/0.05 T¡G rate (%) 0.07 0.08/0.08 0.11/0.12 0.07/0.08 Conversion index 2.41 2.42/1.80 32.95/31.67 2.46/1.80 Cellular transcripts (internal control) T¡C rate (%) 0.10 0.06 1.83 0.05 T¡A rate (%) 0.03 0.02 0.04 0.01 T¡G rate (%) 0.06 0.05 0.08 0.03 Conversion index 2.12 1.60 29.67 2.34 aIn each TBEV and SMRV cell, the ﬁrst value (before the slash) and the second value (after the slash) correspond to the conversion rate or index based on the external consensus sequence derived from the D0 reads (“TBEV REFERENCE” and “SMRV REFERENCE”) or from internal consensus sequences directly derived from the reads of the experimental condition (“INTERNAL REFERENCE”), respectively. aIn each TBEV and SMRV cell, the ﬁrst value (before the slash) and the second value (after the slash) correspond to the conversion rate or index based on the external consensus sequence derived from the D0 reads (“TBEV REFERENCE” and “SMRV REFERENCE”) or from internal consensus sequences directly derived from the reads of the experimental condition (“INTERNAL REFERENCE”), respectively. genomes (Fig. 1). In clear contrast, the T¡C conversion rate for labeled and alkylated RNAs at D1 was much higher (0.87%), resulting in a conversion index of 6.09, a 3.4-fold increase compared to the background at D1. The conversion index at D1 for the labeled and alkylated SMRV-infected cells was 32.95, 13.6-fold over the background level. These speciﬁc T¡C conversion rates were also distributed along the whole viral genomes (Fig. msphere.asm.org 4 RESULTS 1), supporting the use of the mean conversion rates as a robust indicator of efﬁcacy of labeling. Comparisons between metabolically labeled and nonlabeled RNAs would necessi- tate two conditions of culture. As a result, we also compared the TBEV and SMRV conversion indexes obtained at D1 for the 4sU-labeled culture, with and without RNA alkylation. This necessitates only one condition of culture, followed by RNA extraction and alkylation, or no treatment. The low level of conversions in cells with 4sU-labeled, nonalkylated RNA did not impair the detection by BLAST analysis of potential viral hits (Table S1). As shown in Table 2, the conversion index of the 4sU-labeled, nonalkylated RNAs remained low and close to that of the nonlabeled condition (1.81 and 2.46 for TBEV and SMRV, respectively, increasing 3.4- and 13.4-fold, respectively, in the alkylated condition). This suggests that nonalkylated RNAs extracted from the same cell culture can be used to calculate background conversion rates. We have also taken advantage of the fact that even in the best cases, the rate of T¡C conversion was low (below 5% of T positions) and therefore should not modify the consensus sequence, which represents the most frequently found base for each nucleotide position. For SMRV and TBEV, we recalculated all conversion rates and their distribution along each genome (Fig. 1 and Table 2). As expected, this did not signiﬁcantly modify the results obtained when the nonlabeled RNA sequence (D0 reference) was used. The only conversion indexes above the background level were those of 4sU-labeled and alkylated RNAs of TBEV- and SMRV-infected cells at D1. The conversion indexes at D1 were 3.3- and 17.6-fold over the background of nonlabeled cells for TBEV and SMRV, respectively. Therefore, our results show that after 4Su labeling of cells, RNA-Seq was able to speciﬁcally identify newly synthesized viral RNAs with a high signal-to-background noise ratio in a one-shot test, without the need for an external reference sequence. Characterization of expression of SMRV sequences in Vero cells. With the reads mapping to SMRV-H, we derived a consensus full-length sequence, deposited in GenBank (10) as proviral DNA under the name SMRV-Vero (GenBank accession number May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 4 Virus Detection by HTS of Newly Synthesized RNAs FIG 1 Distribution of conversion rates along the genomes of SMRV and TBEV. RESULTS Reverse transcription-PCRs (RT-PCRs) targeting the gag and env gene together with the primer binding site conﬁrmed the presence of SMRV expression in the Vero cell line. Sanger sequencing conﬁrmed the sequence of the PBS at position nt 467 (SMRV-H position). RT-PCRs were negative for the same lot of cells that was banked 3 passages after receiving the cells from ATCC and also for a vial of another lot of Vero cells (lot 70005907) tested directly without any cultivation step. This demonstrated that contamination occurred during the passage of cells in the cell culture laboratory. RT-PCR investigations. Reverse transcription-PCRs (RT-PCRs) targeting the gag and env gene together with the primer binding site conﬁrmed the presence of SMRV expression in the Vero cell line. Sanger sequencing conﬁrmed the sequence of the PBS at position nt 467 (SMRV-H position). RT-PCRs were negative for the same lot of cells that was banked 3 passages after receiving the cells from ATCC and also for a vial of another lot of Vero cells (lot 70005907) tested directly without any cultivation step. This demonstrated that contamination occurred during the passage of cells in the cell culture laboratory. RESULTS Conversion rates were calculated in reference to the reference sequence derived from day 0 (“TBEV REFERENCE” and “SMRV REFERENCE”) or from an internal consensus sequence based on the reads from each experimental condition (“INTERNAL REFFERENCE”). For the D0 time point, the two modes of calculation are therefore equivalent. (T to C, G, or A, in blue, orange, or green lines, respectively). Abbreviations: D0 and DO, day 0; 4Su, 4-thiouridine; alkyl, alkylation. FIG 1 Distribution of conversion rates along the genomes of SMRV and TBEV. Conversion rates were calculated in reference to the reference sequence derived from day 0 (“TBEV REFERENCE” and “SMRV REFERENCE”) or from an internal consensus sequence based on the reads from each experimental condition (“INTERNAL REFFERENCE”). For the D0 time point, the two modes of calculation are therefore equivalent. (T to C, G, or A, in blue, orange, or green lines, respectively). Abbreviations: D0 and DO, day 0; 4Su, 4-thiouridine; alkyl, alkylation. MK561030). The full DNA genome has a length of 8,787 nucleotides (nt) and shows 99% identity with SMRV-H. Compared to the original SMRV isolated from squirrel monkey lungs, SMRV-Vero, like SMRV-H, shows an insertion of one (SMRV-Vero) or two (SMRV-H) nucleotide G in the primer binding site (PBS) (at nt 467 in SMRV-H). The four open reading frames (ORFs) described for SMRV-H are also present as shown in Table 3. Amino acid identity ranges from 98% to 99% for all proteins except for Gag (93%). The Gag precursor protein is cleaved into p10, p16, p19, and p35 proteins. Most of the variability (Fig. S1) was concentrated in the p35 protein, the major capsid protein (11). MK561030). The full DNA genome has a length of 8,787 nucleotides (nt) and shows 99% identity with SMRV-H. Compared to the original SMRV isolated from squirrel monkey lungs, SMRV-Vero, like SMRV-H, shows an insertion of one (SMRV-Vero) or two (SMRV-H) nucleotide G in the primer binding site (PBS) (at nt 467 in SMRV-H). The four open reading frames (ORFs) described for SMRV-H are also present as shown in Table 3. Amino acid identity ranges from 98% to 99% for all proteins except for Gag (93%). The Gag precursor protein is cleaved into p10, p16, p19, and p35 proteins. Most of the variability (Fig. S1) was concentrated in the p35 protein, the major capsid protein (11). RT-PCR investigations. DISCUSSION The efﬁcacy of HTS for the testing of biologicals has been demonstrated by the conﬁrmation of viral contamination of a marketed live vaccine despite no ﬁndings by conventional testing (12). The potential expanded use of HTS for testing of biologicals, however, would need to address the concern of ﬁnding viral hits that do not present safety concerns. Our objective was to design HTS strategies able to add speciﬁcity of detection of living viruses in cells in addition to the broad range of viral detection and May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 5 Cheval et al. FIG 2 Conversion indexes expressed as the ratio of the T¡C conversion rate to the average of the (T¡A, T¡G) conversion rates. Conversion rates for viruses were calculated according to the two types of reference sequence as in Fig. 1. Conversion rates for cellular RNA controls are calculated according the CONTROL REFERENCE sequence. Abbreviations: N, no; Y, yes. FIG 2 Conversion indexes expressed as the ratio of the T¡C conversion rate to the average of the (T¡A, T¡G) conversion rates. Conversion rates for viruses were calculated according to the two types of reference sequence as in Fig. 1. Conversion rates for cellular RNA controls are calculated according the CONTROL REFERENCE sequence. Abbreviations: N, no; Y, yes. analytical sensitivity that this technology provides versus current testing methodolo- gies. We have previously demonstrated that our protocol of deep sequencing of RNAs is sensitive enough to detect a single human herpesvirus type 4 (HHV-4)-infected cell in a background of 105 virus-free cells and also has the ability to detect cells infected at very low multiplicity of infection (MOI) by BVDV 2 days after infection (6). Coupled with strand analysis as biomarkers of virus expression during replication and even latency (13, 14), this protocol increases speciﬁcity compared to sequencing all types of cellular nucleic acids (DNA and RNA) (6). Nevertheless, this method is not devoid of limits. For example, since antigenomes are present in small amount for single-stranded positive-sense RNA [ss()RNA] viruses, very deep sequencing is required. For some single-stranded negative-sense RNA [ss()RNA] viruses, full-length antigenomes are also present in mature virions, which results in the requirement for an additional, virus species-speciﬁc bioinformatic analysis in order to identify subgenomic transcripts associated with cell infection. May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 6 DISCUSSION of aa) in: Amino acid identity (%) SMRV-H SMRV-V Gag 740 742 93 Protease 90 90 98 Pol 798 798 99 Env 575 575 98 May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 6 msphere.asm.org 6 Virus Detection by HTS of Newly Synthesized RNAs bution of viral hits between puriﬁed RNAs and crude RNAs, which would be difﬁcult to standardize and validate. For the present study, we have taken advantage of a recently described thiol (SH)-linked alkylation for the metabolic sequencing of RNA (SLAM seq), which enables 4sU detection in RNA species after metabolic RNA labeling (16). g After performing metabolic RNA labeling for 6 h, we prepared total RNA followed by thiolalkylation or no treatment and then randomly deep sequenced the libraries. We based our interpretation on the frequency of the T¡C conversion compared to the background of T¡G and T¡A. Using the ss()RNA virus TBEV as a model, we demonstrated the ability to differentiate “old” RNAs of nonreplicating viruses bound at the cell surface from RNAs associated with live viruses. This is an obvious advantage compared to the detection of whole nucleic acids, since these are not speciﬁc to live viruses. This method also overcome the difﬁculty of detecting the small amount of antigenomic RNA strand as a marker of ss()virus expression (6). Another advantage of our technique is that only RNA is labeled. This is an interesting feature since RNAs are per se evidence of viral transcription for DNA viruses and retroviruses (which are transcribed from a DNA provirus). Even some transcripts incorporated in inactivated virions of some enveloped double-stranded DNA (dsDNA) viruses, such as herpesvi- ruses (17, 18), poxvirus (19), or mimivirus (20), would not be labeled. p A good demonstration of the advantages of our nontargeted approach was that we were able to detect the expression of a strain of SMRV very close to SMRV-H in the Vero ATCC CCL81 cell line used for our TBEV infection studies. We have named this strain SMRV-Vero. When tested a posteriori by RT-PCR, it appeared that the Vero cells from ATCC were SMRV free and that the contamination arose after receiving the cells from ATCC in the laboratory in which the cells were cultivated, from a currently unknown origin. DISCUSSION This was unexpected, as SMRV, a type D betaretrovirus, is known to infect only primates from the New World (21, 22), and Vero cells are derived from African green monkeys, an Old World primate. Until now, only different simian endogenous retrovi- ruses (SERVs) have been identiﬁed in Vero cells (8, 12, 23, 24). This includes one type close to baboon endogenous virus (BAEV) gammaretrovirus (25) and another close to exogenous simian retrovirus (SRV) (26). These sequences, which include complete proviruses (27), show a large diversity in the cell origin, but most harbor inactivating deletions or frameshift mutations (ATCC CCL81, JCRBO111, or Vero 76) (24, 27). The SMRV-Vero strain that we identiﬁed does not present any obvious loss-of-function mutations, insertions, or deletions. Compared to the SMRV strain isolated from mon- keys in canine cells, the G insertion in the PBS of SMRV-Vero is reminiscent of the GG insertion found in SMRV-H, a clone persistently infecting human lymphoid cell lines and should therefore be compatible with priming of reverse transcription by tRNALys1,2. Mutations with reference to SMRV-H were identiﬁed in the p35 capsid protein. It is currently not known whether SMRV-Vero produces an infectious virus. More character- ization of this virus is currently ongoing. Regarding the testing of cells (e.g., cell banks, drug products in cell therapy, “control cells” tested during vaccine production, cells used for bioassay of raw materials), routine usage of our technique requires only the addition of 4sU to the cell medium a few hours before cell harvest. In a ﬁrst modality of usage, agnostic HTS analysis is completed using the sequences obtained with an index sample (nonalkylated RNA from the 4sU-labeled cells or nonlabeled cells). The use of nonalkylated RNA from the 4sU-labeled cells is manage- able, since alkylation can be performed as a second step, allowing comparison of sequences derived from libraries with the same contents, and because such compari- son does not necessitate a parallel, nonlabeled culture. A consensus reference se- quence has to be derived from 4sU-labeled, nonalkylated RNAs. Conversion rates and conversion ratios then have to be calculated by mapping the reads derived from the 4sU-labeled and alkylated RNAs onto this reference sequence. Our results suggest that if nonlabeled cells are used as a reference, an increase of at least threefold between the conversion indexes could be used to distinguish newly synthesized RNAs. DISCUSSION Cell-free viral RNAs can also be present in standard virus stocks frozen at 80°C or in liquid nitrogen just after harvest and therefore can introduce background noise during validation without any relevance to real samples. analytical sensitivity that this technology provides versus current testing methodolo- gies. We have previously demonstrated that our protocol of deep sequencing of RNAs is sensitive enough to detect a single human herpesvirus type 4 (HHV-4)-infected cell in a background of 105 virus-free cells and also has the ability to detect cells infected at very low multiplicity of infection (MOI) by BVDV 2 days after infection (6). Coupled with strand analysis as biomarkers of virus expression during replication and even latency (13, 14), this protocol increases speciﬁcity compared to sequencing all types of cellular nucleic acids (DNA and RNA) (6). Nevertheless, this method is not devoid of limits. For example, since antigenomes are present in small amount for single-stranded positive-sense RNA [ss()RNA] viruses, very deep sequencing is required. For some single-stranded negative-sense RNA [ss()RNA] viruses, full-length antigenomes are also present in mature virions, which results in the requirement for an additional, virus species-speciﬁc bioinformatic analysis in order to identify subgenomic transcripts associated with cell infection. Cell-free viral RNAs can also be present in standard virus stocks frozen at 80°C or in liquid nitrogen just after harvest and therefore can introduce background noise during validation without any relevance to real samples. As a result of the above, we decided to evaluate an ultimate evolution of the concept of targeting RNAs speciﬁcally for viral infection, the identiﬁcation of newly synthesized RNAs, since this could distinguish viral expression from carryover, inde- pendently of the type of virus. A classical method for such analysis would be to label nascent RNAs with modiﬁed uridine analogs (e.g., 4-thiouridine [4sU], 5-etyniluridine [EU], and 5=-bromouridine [BrU]), followed by puriﬁcation of labeled de novo RNAs (see the review in reference 15). To draw any conclusion regarding carryover of inert viruses versus cell infection, these techniques would have necessitated comparing the distri- TABLE 3 Comparison of SMRV-H and SMRV-Vero proteins Protein Length of protein (no. May/June 2019 Volume 4 Issue 3 e00298-19 MATERIALS AND METHODS Cells and viruses. A vial of Vero cells (ATCC-CCL-81; lot 62488537; ATCC, Molsheim, France) was cultivated in minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS) after receiving the cells and frozen at passage 3 (the ﬁrst passage of the vial of cells received from ATCC was designated passage 1). A vial was then defrosted in a biosafety level 3 (BSL-3) laboratory to be used as cell substrate for virus cultivation, and the cells were grown again in MEM supplemented with 10% FBS. Cells were used at passage 18. A second vial of Vero cells (lot 70005907) was obtained from the same source and used directly for PCR testing. Tick-borne encephalitis virus (TBEV), a member of the family Flaviviridae, possesses a single-stranded RNA with positive polarity [ssRNA()] genome. The Hypr strain (28) was kindly supplied by Sarah Moutailler, ANSES, Maisons-Alfort, France. TBEV infection of Vero cells. Vero cells were plated at 400,000 cells/well in three wells of a six-well culture plate (9.5 cm2 [cell growth area]) in order to reach 106 cells/well after 24 h. Cells were then infected with the Hypr strain TBEV at a multiplicity of infection of 1 and incubated for 1 h on ice with agitation. The medium was removed in one well just after incubation, and the cells were lysed with 1 ml of TRIzol and stored at 80°C until RNA extraction (condition 1, day 0 with no 4sU [“D0 no 4sU”]). For the other two wells (conditions 2, 3, and 4, see next paragraph), the medium was removed and replaced by MEM plus 10% FBS and incubated overnight at 37°C. y p g 4sU labeling and RNA extraction. The addition of 4-thiouridine (4sU) into the cell culture medium enables 4sU nucleotides to be incorporated into newly synthesized RNA. The reverse transcription of 4sU displays a certain percentage of misincorporation resulting in a T¡C transition in the cDNA that can be identiﬁed by sequencing (16). Media containing 800 M 4sU was prepared by adding 8 l of 100 mM 4sU in 992 l of MEM. The day after viral infection (day 1 [D1]), the medium was removed and replaced by medium without 4sU in one well (condition 2, day 1 with no 4sU [“D1 - no 4sU”]) or 4sU-containing medium (800 M) for the other well (conditions 3 and 4, day 1 with 4sU [“D1 with 4sU”]). DISCUSSION We found that the same threshold could be used in our study when RNA from 4sU-labeled, msphere.asm.org 7 Cheval et al. nonalkylated RNAs was used as a reference. This is despite reports of a signiﬁcant T¡C conversion rate in 4sU-labeled nonalkylated RNAs (16). In a second and privileged modality of usage, the protocol uses only one set of sequences from the tested cell sample. Speciﬁcally, 4sU-labeled and alkylated RNAs would be sequenced and analyzed by the agnostic pipeline. The performance of the agnostic pipeline is not signiﬁcantly modiﬁed (Table 1), as only a minority of T positions are converted to C. In case of any viral hits, the targeted pipeline will use the same set of sequences to ﬁrst deﬁne a consensus sequence, itself marginally impacted by the low rate of T¡C conversions, and then to calculate the conversion rate and index. In this setting, the conversion index will be used to determine whether the viral hit corresponds to an RNA that has been recently transcribed. Our results suggest that a conversion index above 3 would provide a good margin of safety above the back- ground. Proper optimization of 4sU labeling conditions and threshold determination with the cells to be tested could improve the quality of the results and are mandatory in the case of GMP routine testing. The protocol needs to compute statistics and the frequency of T conversions and is not useful when a few reads are evidenced (for example, see BVDV in Table S1 in the supplemental material). In practice, such a low number of reads can hardly be mistaken with a viral infection, as shown when using worst-case models (6). In conclusion, using transcriptome analysis by HTS coupled with RNA strand analysis and identiﬁcation of newly synthesized RNAs after metabolic labeling enables the ability to broadly and speciﬁcally test cells for active infection versus inert viral sequence contamination, thus representing technical progress for the viral safety testing of biological products. May/June 2019 Volume 4 Issue 3 e00298-19 MATERIALS AND METHODS The bacterial rRNA database was initially down- loaded from the EMBL-EBI ENA rRNA database (ftp://ftp.ebi.ac.uk/pub/databases/ena/rRNA/release) and reworked using in-house sequence cleaning and clustering processes. Remaining reads were considered sequences of interest and were assembled into contigs with CLC Assembly Cell solution (Qiagen Bioinformatics). Resulting contigs and nonassembled reads (singletons) were aligned using BLAST alignment (32) on viral and comprehensive databases. Contigs and singletons were ﬁrst aligned on a viral nucleotide database. Hits with an E value below 103 were aligned on a comprehensive nucleotide database. If the best hit was still a viral taxonomy, hits were reported. The nucleotide viral and comprehensive databases were downloaded during November 2017 from the EMBL-EBI nucleotide sequence database (https://www.ebi.ac.uk/). The viral database is restricted to viral sequences of the EMBL-EBI nucleotide sequence database, including subgenomic sequences. Proprietary software was developed to remove duplicate and low-conﬁdence sequences (e.g., too short, multiple taxonomies, low-quality associated keywords). Contigs without any viral nucleotide hits were similarly aligned successively on viral and comprehensive protein databases to check for more-distant viral hits. Protein viral and comprehensive databases were downloaded during November 2017 from the Uniref100 database (https://www.uniprot.org). Though the Uniref100 database is already nonredundant, we per- formed a taxonomic cleaning process to generate the ﬁnal databases. The taxonomic assignation reported the best hit results with contigs not assigned after these two rounds of alignment being classiﬁed as unknown or nonviral species. The data from condition 1 (“DO no 4sU”) enabled the identiﬁcation of contigs covering the whole genome of TBEV (n 1; length 10850 bp) and SMRV (n 4; cumulated length 8639 bp). For SMRV, a new assembly of the reads with Megahit v1.1.3 (33) gave a unique contig covering the whole genome. The resulting sequences, obtained at D0, were labeled “TBEV REFERENCE” and “SMRV REFERENCE.” Estimation of T¡C conversion ratio. In order to detect viral sequences with a very high number of T¡C conversions, the set of quality ﬁltered reads was mapped back to TBEV/SMRV REFERENCE with minimap2 (34). The pileup module of the htsbox software (https://github.com/lh3/htsbox) was then used to detect all mismatches (with a base quality at least equal to 30) at every position of the TBEV or SMRV REFERENCE. The global variation proﬁle was then analyzed by a proprietary script (PathoQuest, Paris, France) to deﬁne each nucleotide conversion rate. MATERIALS AND METHODS Six hours later, the medium was removed and replaced by fresh medium without 4sU in the condition “D1 - no 4sU” or by fresh 4sU-containing medium (800 M) in the two “D1 with 4sU” conditions. Three hours later, the medium was removed from the three wells, and the cells were lysed with 1 ml of TRIzol and stored at 80°C until RNA extraction. RNA extraction was performed in the dark using a chloroform-isoamyl alcohol mix (24:1) (catalog no. 25666; Sigma-Aldrich, St. Louis, MO, USA) followed by isopropanol-ethanol precipitation. During extraction, reducing agent was used to maintain the 4sU-treated sample under reducing conditions. Alkylation was performed using the SLAMseq kinetic kit with anabolic kinetics module (catalog no. 061; Lexogen, Vienna, Austria) for the “D1 with 4sU” condition. Total extracted RNA was mixed with iodoacetamide, which modiﬁes the 4-thiol group of 4sU-containing nucleotides via the addition of a carboxyamidomethyl group, leading to condition 3, “D1 with 4sUalkylation.” This alkylation ampliﬁes the frequency of T¡C misincorporations during reverse transcription. Condition 4 was labeled “D1 with 4sU no alkylation.” The RNA was then puriﬁed using ethanol precipitation before library preparation. Library preparation and sequencing. The SMARTer Stranded Total RNA-Seq kit (Pico Input Mam- malian) (ClonTech, Mountain View, USA) was used for direct construction of libraries starting with 10 ng msphere.asm.org 8 May/June 2019 Volume 4 Issue 3 e00298-19 Virus Detection by HTS of Newly Synthesized RNAs of RNA. The workﬂow used with this kit incorporates a proprietary technology that depletes ribosomal cDNA using probes speciﬁc to mammalian rRNA and some mitochondrial RNA. Sequencing was per- formed on the NextSeq system (Illumina, San Diego, CA, USA) using the NextSeq 500/550 high output kit v2 (catalog no. FC-404-2002; Illumina). Single-read sequencing was utilized with a read length of 150 nucleotides, resulting in the generation of approximatively 125 million reads per sample. Agnostic bioinformatic analysis. The raw data reads were ﬁltered to select high-quality and relevant reads. Raw data were sorted to suppress or cut duplicates, low-quality reads, and homopolymers (proprietary software; PathoQuest, Paris, France). Sequences introduced during the preparation of Illumina libraries (adapters, primers) were removed with Skewer (29). Reads aligned with BWA (30) to (i) the human genome (reference GRCh37/hg19 [31]) used as a proxy to mimic primate (Vero cells) host background or to (ii) bacterial rRNA were discarded. May/June 2019 Volume 4 Issue 3 e00298-19 MATERIALS AND METHODS The proportion of converted nucleotides was compared to the total number of aligned nucleotides. For example, the T¡C conversion rate was calculated using the following formula: T →C rate No. of C nucleotides identified when a T was expected Total no. of expected T T →C rate No. of C nucleotides identified when a T was expected Total no. of expected T T →C rate No. of C nucleotides identified when a T was expected Total no. of expected T T →C rate No. of C nucleotides identified when a T was expected Total no. of expected T The conversion rates for each time point were normalized with the following conversion index: conversion index T →C rate MeanT →A, T →G rates conversion index T C rate MeanT →A, T →G rates MeanT →A, T →G rates Additionally, for viral hits, conversion rates were also calculated using as reference the consensus sequence derived from the reads from each experimental condition. This consensus sequence was named “INTERNAL REFERENCE.” Hence, for the D0 time point, the “TBEV REFERENCE” and “SMRV REFERENCE” were equivalent to “INTERNAL REFERENCE” sequences. As a quality control for the labeling, we checked the mean conversion index of a set of exons using nonlabeled cells as a reference. Exons were used from the following human genes (35) (RefSeq accession number): C1orf43 (NM_015449), CHMP2A (NM_014453), EMC7 (NM_020154), and GPI (NM_000175). We used these human exons in order to identify their equivalent in the Chlorocebus sabaeus genome, from where the Vero cells are derived. The complete assembly of Chlorocebus sabaeus (accession number GCF_000409795.2) was retrieved from NCBI assembly database (https://www.ncbi.nlm.nih.gov/ assembly/). Selected human exons were mapped onto C. sabaeus assembly using minimap2 (34), and the resulting bam ﬁle was converted to a bed ﬁle using the bamtobed module from the BEDTools utility (36). Only hits with mapping quality higher than 30 were retained (41 exons), and the corresponding sequences were extracted from C. sabaeus assembly using the getfasta module from the BEDTools utility and indexed for analyses as the “CONTROL REFERENCE” sequence to calculate conversion rates and indexes. Labeling was considered satisfactory if the conversion index was superior to 10. Stranded analysis. A targeted and stranded analysis was performed on the identiﬁed TBEV reads. This analysis was based on a more stringent mapping alignment of ﬁltered reads with the alignment providing a detailed horizontal genome coverage and depth proﬁle. Local alignments were performed with BWA (30). Since the sample libraries were prepared using the SMARTer Stranded RNA-Seq kit, the RNA strand information was also retained. REFERENCES 1. Klug B, Robertson JS, Condit RC, Seligman SJ, Laderoute MP, Sheets R, Williamson A-L, Gurwith M, Kochhar S, Chapman L, Carbery B, Mac LM, Chen RT, Brighton Collaboration Viral Vaccine Vector Safety Working Group. 2016. Adventitious agents and live viral vectored vaccines: con- siderations for archiving samples of biological materials for retrospective analysis. Vaccine 34:6617–6625. https://doi.org/10.1016/j.vaccine.2016 .02.015. viral anti-sense transcription. Virol J 14:9. https://doi.org/10.1186/s12985 -016-0667-3. 10. Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW. 2016. GenBank. Nucleic Acids Res 44:D67–D72. https://doi.org/10.1093/nar/gkv1276. Acids Res 44:D67–D72. https://doi.org/10.1093/nar/gkv1276. 11. Devare SG, Stephenson JR. 1979. Primate retroviruses: intracistronic mapping of type D viral gag gene by use of nonconditional replication mutants. J Virol 29:1035–1043. 2. Mallet L, Gisonni-Lex L. 2014. Need for new technologies for detection of adventitious agents in vaccines and other biological products. PDA J Pharm Sci Technol 68:556–562. https://doi.org/10.5731/pdajpst.2014 .01012. 2. Mallet L, Gisonni-Lex L. 2014. Need for new technologies for detection of adventitious agents in vaccines and other biological products. PDA J Pharm Sci Technol 68:556–562. https://doi.org/10.5731/pdajpst.2014 .01012. 12. Victoria JG, Wang C, Jones MS, Jaing C, McLoughlin K, Gardner S, Delwart EL. 2010. Viral nucleic acids in live-attenuated vaccines: detection of minority variants and an adventitious virus. J Virol 84:6033–6040. https://doi.org/10.1128/JVI.02690-09. 13. Zhang X, Ma X, Jing S, Zhang H, Zhang Y. 2018. Non-coding RNAs and retroviruses. Retrovirology 15:20. https://doi.org/10.1186/s12977-018 -0403-8. 3. Naccache SN, Greninger AL, Lee D, Coffey LL, Phan T, Rein-Weston A, Aronsohn A, Hackett J, Jr, Delwart EL, Chiu CY. 2013. The perils of pathogen discovery: origin of a novel parvovirus-like hybrid genome traced to nucleic acid extraction spin columns. J Virol 87:11966–11977. https://doi.org/10.1128/JVI.02323-13. 3. Naccache SN, Greninger AL, Lee D, Coffey LL, Phan T, Rein-Weston A, Aronsohn A, Hackett J, Jr, Delwart EL, Chiu CY. 2013. The perils of pathogen discovery: origin of a novel parvovirus-like hybrid genome traced to nucleic acid extraction spin columns. J Virol 87:11966–11977. https://doi.org/10.1128/JVI.02323-13. 14. Knipe DM, Raja P, Lee J. 2017. Viral gene products actively promote latent infection by epigenetic silencing mechanisms. Curr Opin Virol 23:68–74. https://doi.org/10.1016/j.coviro.2017.03.010. 4. Smuts H, Kew M, Khan A, Korsman S. 2014. Novel hybrid parvovirus-like virus, NIH-CQV/PHV, contaminants in silica column-based nucleic acid extraction kits. J Virol 88:1398. https://doi.org/10.1128/JVI.03206-13. 4. Smuts H, Kew M, Khan A, Korsman S. 2014. Novel hybrid parvovirus-like virus, NIH-CQV/PHV, contaminants in silica column-based nucleic acid extraction kits. J Virol 88:1398. https://doi.org/10.1128/JVI.03206-13. 15. Tani H, Akimitsu N. REFERENCES 2012. Genome-wide technology for determining RNA stability in mammalian cells: historical perspective and recent advan- tages based on modiﬁed nucleotide labeling. RNA Biol 9:1233–1238. https://doi.org/10.4161/rna.22036. 5. Ngoi CN, Siqueira J, Li L, Deng X, Mugo P, Graham SM, Price MA, Sanders EJ, Delwart E. 2017. Corrigendum: the plasma virome of febrile adult Kenyans shows frequent parvovirus B19 infections and a novel arbovirus (Kadipiro virus). J Gen Virol 98:517. https://doi.org/10.1099/jgv.0.000762. 5. Ngoi CN, Siqueira J, Li L, Deng X, Mugo P, Graham SM, Price MA, Sanders EJ, Delwart E. 2017. Corrigendum: the plasma virome of febrile adult Kenyans shows frequent parvovirus B19 infections and a novel arbovirus (Kadipiro virus). J Gen Virol 98:517. https://doi.org/10.1099/jgv.0.000762. 16. Herzog VA, Reichholf B, Neumann T, Rescheneder P, Bhat P, Burkard TR, Wlotzka W, von Haeseler A, Zuber J, Ameres SL. 2017. Thiol-linked alkylation of RNA to assess expression dynamics. Nat Methods 14: 1198–1204. https://doi.org/10.1038/nmeth.4435. 6. Brussel A, Brack K, Muth E, Zirwes R, Cheval J, Hebert C, Charpin J-M, Marinaci A, Flan B, Ruppach H, Beurdeley P, Eloit M. 2019. Use of a new RNA next generation sequencing approach for the speciﬁc detection of virus infection in cells. Biologicals 59:29–36. https://doi.org/10.1016/j .biologicals.2019.03.008. 6. Brussel A, Brack K, Muth E, Zirwes R, Cheval J, Hebert C, Charpin J-M, Marinaci A, Flan B, Ruppach H, Beurdeley P, Eloit M. 2019. Use of a new RNA next generation sequencing approach for the speciﬁc detection of virus infection in cells. Biologicals 59:29–36. https://doi.org/10.1016/j .biologicals.2019.03.008. 17. Terhune SS, Schröer J, Shenk T. 2004. RNAs are packaged into human cytomegalovirus virions in proportion to their intracellular con- centration. J Virol 78:10390–10398. https://doi.org/10.1128/JVI.78.19 .10390-10398.2004. 7. Oda T, Ikeda S, Watanabe S, Hatsushika M, Akiyama K, Mitsunobu F. 1988. Molecular cloning, complete nucleotide sequence, and gene struc- ture of the provirus genome of a retrovirus produced in a human lymphoblastoid cell line. Virology 167:468–476. https://doi.org/10.1016/ 0042-6822(88)90109-2. 18. Sciortino MT, Suzuki M, Taddeo B, Roizman B. 2001. RNAs extracted from herpes simplex virus 1 virions: apparent selectivity of viral but not cellular RNAs packaged in virions. J Virol 75:8105–8116. https://doi.org/ 10.1128/JVI.75.17.8105-8116.2001. 8. Ma H, Ma Y, Ma W, Williams DK, Galvin TA, Khan AS. 2011. Chemical induction of endogenous retrovirus particles from the Vero cell line of African green monkeys. J Virol 85:6579–6588. https://doi.org/10.1128/ JVI.00147-11. 19. Grossegesse M, Doellinger J, Haldemann B, Schaade L, Nitsche A. 2017. T →C rate No. of C nucleotides identified when a T was expected Total no. of expected T As a result, a mapping alignment analysis was able to provide May/June 2019 Volume 4 Issue 3 e00298-19 msphere.asm.org 9 Cheval et al. information on the mother strand (the DNA or RNA template sequence from which the RNA was transcribed) of each read (sense or reverse relative to the mother strand) as described previously (6). information on the mother strand (the DNA or RNA template sequence from which the RNA was transcribed) of each read (sense or reverse relative to the mother strand) as described previously (6). SMRV RT-PCR. Primers used for RT-PCR included SMRV-gag-4F (ACCGTGTTTTTGGTCCTGAG), SMRV- gag-4R (GGGCACTGCTGTAGGAACAT), SMRV-PBS-F (CTGCGGGACAGAGCAAGT), SMRV-PBS-R (TCCCATGA TTGGGTCTTACC), SMRV-env-4F (GTACAGCAGGACTCGGGGTA), and SMRV-env-4R (CGTCTTCGTTCGAGGT CTTC). cDNA was tested using 100 ng of RNA to be tested in a ﬁnal volume of 10 l with the SuperScript III First Strand Synthesis System kit (catalog no. 18080-051; Thermo Fisher, Waltham, USA) and then diluted to 2 ng/l. From the RNA samples, 2 l was used for each tube, and PCR was conducted with the PrimeSTAR GXL DNA polymerase kit (catalog no. R050A, Ozyme, Montigny-le-Bretonneux, France). Thirty cycles were performed, and each cycle consisted of 10 s at 98°C, 15 s at 60°C, and 1 min at 68°C. Data availability. Access to the proprietary software tools used in this study will be granted to researchers upon request. ACKNOWLEDGMENTS We thank Marielle Cochet for her help in the SMRV investigations. The sponsor of the study was PathoQuest. Employees of PathoQuest (E.M., J.C., P.B., and S.C.) participated in writing the study report and in the decision to submit a manuscript for publication. E.M., J.C., P.B., and S.C. are employees of PathoQuest. E.M., J.C., and M.E. hold shares in PathoQuest. E.M., J.C., P.B., and S.C. are employees of PathoQuest. E.M., J.C., and M.E. hold shares in PathoQuest. SUPPLEMENTAL MATERIAL Supplemental material for this article may be found at https://doi.org/10.1128/ mSphere.00298-19. Supplemental material for this article may be found at https://doi.org/10.1128/ mSphere.00298-19. FIG S1, PDF ﬁle, 0.1 MB. TABLE S1, PDF ﬁle, 0.1 MB. E.M., J.C., P.B., and S.C. are employees of PathoQuest. E.M., J.C., and M.E. hold shares in PathoQuest. REFERENCES A next-generation sequencing approach uncovers viral tran- scripts incorporated in poxvirus virions. Viruses 9:E296. https://doi .org/10.3390/v9100296. 20. Raoult D, Audic S, Robert C, Abergel C, Renesto P, Ogata H, La Scola B, Suzan 9. Manghera M, Magnusson A, Douville RN. 2017. The sense behind retro- May/June 2019 Volume 4 Issue 3 e00298-19 Virus Detection by HTS of Newly Synthesized RNAs M, Claverie J-M. 2004. The 1.2-megabase genome sequence of Mimivirus. Science 306:1344–1350. https://doi.org/10.1126/science.1101485. M, Claverie J-M. 2004. The 1.2-megabase genome sequence of Mimivirus. Science 306:1344–1350. https://doi.org/10.1126/science.1101485. 28. Wallner G, Mandl CW, Ecker M, Holzmann H, Stiasny K, Kunz C, Heinz FX. 1996. Characterization and complete genome sequences of high- and low- virulence variants of tick-borne encephalitis virus. J Gen Virol 77:1035–1042. https://doi.org/10.1099/0022-1317-77-5-1035. 21. Fine D, Schochetman G. 1978. Type D primate retroviruses: a review. Cancer Res 38:3123–3139. 22. Colcher D, Heberling RL, Kalter SS, Schlom J. 1977. Squirrel monkey retrovirus: an endogenous virus of a New World primate. J Virol 23: 294–301. 29. Jiang H, Lei R, Ding S-W, Zhu S. 2014. Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads. BMC Bioin- formatics 15:182. https://doi.org/10.1186/1471-2105-15-182. 23. Onions D, Côté C, Love B, Toms B, Koduri S, Armstrong A, Chang A, Kolman J. 2011. Ensuring the safety of vaccine cell substrates by mas- sively parallel sequencing of the transcriptome. Vaccine 29:7117–7121. https://doi.org/10.1016/j.vaccine.2011.05.071. 30. Li H, Durbin R. 2009. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25:1754–1760. https://doi .org/10.1093/bioinformatics/btp324. 31. Kent WJ, Sugnet CW, Furey TS, Roskin KM, Pringle TH, Zahler AM, Haussler D. 2002. The human genome browser at UCSC. Genome Res 12:996–1006. https://doi.org/10.1101/gr.229102. 24. Osada N, Kohara A, Yamaji T, Hirayama N, Kasai F, Sekizuka T, Kuroda M, Hanada K. 2014. The genome landscape of the African green monkey kidney-derived Vero cell line. DNA Res 21:673–683. https://doi.org/10 .1093/dnares/dsu029. 32. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215:403–410. https://doi.org/10.1016/ S0022-2836(05)80360-2. 25. van der Kuyl AC, Dekker JT, Goudsmit J. 1995. Distribution of baboon endogenous virus among species of African monkeys suggests multiple ancient cross-species transmissions in shared habitats. J Virol 69: 7877–7887. 33. Li D, Liu C-M, Luo R, Sadakane K, Lam T-W. 2015. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31:1674–1676. https://doi.org/ 10.1093/bioinformatics/btv033. 26. van der Kuyl AC, Mang R, Dekker JT, Goudsmit J. 1997. Complete nucleotide sequence of simian endogenous type D retrovirus with intact genome organization: evidence for ancestry to simian retrovirus and baboon endogenous virus. J Virol 71:3666–3676. 34. Li H. 2018. Minimap2: pairwise alignment for nucleotide sequences. Bioinfor- matics 34:3094–3100. https://doi.org/10.1093/bioinformatics/bty191. 35. Eisenberg E, Levanon EY. 2013. May/June 2019 Volume 4 Issue 3 e00298-19 Human housekeeping genes, revisited. Trends Genet 29:569–574. https://doi.org/10.1016/j.tig.2013.05.010. 27. Sakuma C, Sekizuka T, Kuroda M, Kasai F, Saito K, Ikeda M, Yamaji T, Osada N, Hanada K. 2018. Novel endogenous simian retroviral inte- grations in Vero cells: implications for quality control of a human vaccine cell substrate. Sci Rep 8:644. https://doi.org/10.1038/s41598 -017-18934-2. 36. Quinlan AR, Hall IM. 2010. BEDTools: a ﬂexible suite of utilities for comparing genomic features. Bioinformatics 26:841–842. https://doi .org/10.1093/bioinformatics/btq033. msphere.asm.org 11 msphere.asm.org 11
https://openalex.org/W3136974198	https://www.frontiersin.org/articles/10.3389/feart.2021.646772/pdf	English	null	The Genesis of the Askartor Be-Mo Deposit in the North Xinjiang, Northwest China: Evidence From Geology, Geochemistry, U-Pb, and Re-Os Geochronology	Frontiers in earth science	2,021	cc-by	13,481	Edited by: Chang-Zhi Wu, Chang’an University, China Reviewed by: Fuquan Yang, Chinese Academy of Geologi-cal Sciences (CAGS), China Rui Wang, China University of Geosciences, China Correspondence: Tang Yong tangyong@vip.gyig.ac.cn Zhang Hui zhanghui@vip.gyig.ac.cn Edited by: Chang-Zhi Wu, Chang’an University, China Reviewed by: Fuquan Yang, Chinese Academy of Geologi-cal Sciences (CAGS), China Rui Wang, China University of Geosciences, China Correspondence: Tang Yong tangyong@vip.gyig.ac.cn Zhang Hui zhanghui@vip.gyig.ac.cn Edited by: Chang-Zhi Wu, Chang’an University, China Reviewed by: Fuquan Yang, Chinese Academy of Geologi-cal Sciences (CAGS), China Rui Wang, China University of Geosciences, China Correspondence: Tang Yong tangyong@vip.gyig.ac.cn Zhang Hui zhanghui@vip.gyig.ac.cn Specialty section: This article was submitted to Economic Geology, a section of the journal Frontiers in Earth Science Specialty section: This article was submitted to Economic Geology, a section of the journal Frontiers in Earth Science Received: 28 December 2020 Accepted: 03 February 2021 Published: 16 March 2021 ORIGINAL RESEARCH published: 16 March 2021 doi: 10.3389/feart.2021.646772 The Genesis of the Askartor Be-Mo Deposit in the North Xinjiang, Northwest China: Evidence From Geology, Geochemistry, U-Pb, and Re-Os Geochronology Tang Yong, Zhang Hui* and Lv Zheng-Hang Key Laboratory for High Temperature and High-Pressure Study of the Earth’s Interior, Institute of Geochemistry, Chinese Academy of Sciences, Guiyang, China The Askartor Be-Mo deposit is located in the southeastern area of the Chinese Altay orogenic belt in Xinjiang, NW China. Zircon U-Pb data show that there are two periods of magmatic activities in the Askartor Be-Mo ore district, namely, the Devonian granodiorite (386.8 ± 2.6 Ma) and biotite granite (385.4 ± 4.4 Ma), and the Triassic two-mica granite (247.5 ± 2.2 Ma) and muscovite granite (231.4 ± 2.0 Ma). The zircon U-Pb age of pegmatoid orebody is 220.6 ± 1.6 Ma which coincides with the molybdenite Re-Os isochron age of 228.7 ± 7.1 Ma. The two-mica and muscovite granites belong to the high- K Calc-alkaline series with peraluminous features, and are characterized by high SiO2 (71.92–75.41 wt%), and Al2O3 (13.43–15.98 wt%), and low TiO2 (0.01–0.25 wt%), Fe2O3 (0.11–1.14 wt%) and CaO (0.07–0.76 wt%). The highly fractionated element ratios of Y/Ho, Zr/Hf and Nb/Ta, and the rare earth element tetrad effect occur in the muscovite granite, indicating the ﬂuid exsolution occurs at the late stage of magma evolution, and the muscovite granite experienced the strong self-metasomatism. Rayleigh fractional calculations show that the Askartor Be-Mo deposit is the product of multistage fractional crystallization of initial Be-enriched magma. INTRODUCTION Because of its unique physical and chemical properties, beryllium (Be) is widely used in many important industrial products used in the aerospace, computer, defense, medical, nuclear, and telecommunications industries. The two most important Be ores are bertrandite [Be2Si2O7(OH)2] and beryl [Be3Al2Si6O18]. At present, 90% of the source mineral of the world’s beryllium produce comes from bertrandite, but industrial beryl is critical for the production of the very high purity beryllium metal needed for some strategic applications (Foley et al., 2017). At present, two distinct classes of deposits currently account for most beryllium ores (Barton and Young, 2002; London and Evensen, 2002): they are 1) volcanic and carbonated-hosted deposits that contain the mineral bertrandite, and 2) pegmatite-type deposits that have an abundance of the mineral beryl. Received: 28 December 2020 Accepted: 03 February 2021 Published: 16 March 2021 Keywords: Askartor Be deposit, U-Pb age, tetrad effect, melt-ﬂuid interaction, NW China GEOLOGICAL BACKGROUND et al., 2011; Lv et al., 2012) and internal evolution (Zhang et al., 2008; Lv et al., 2012; Stepanov and Hermann, 2013; Yin et al., 2013) of pegmatites in this area have been studied systematically. However, the mechanism of extraordinary enrichment of beryllium in pegmatites is not clear. Based on the solubility of beryl and the partition coefﬁcient of Be, Evensen and London (2002) considered that extended fractionation beyond 90% total solidiﬁcation by at least a three-step process is required to achieve beryl saturation in granitic magma systems, but there is a lack of factual evidence. Citation: Yong T, Hui Z and Zheng-Hang L (2021) The Genesis of the Askartor Be- Mo Deposit in the North Xinjiang, Northwest China: Evidence From Geology, Geochemistry, U-Pb, and Re-Os Geochronology. Front. Earth Sci. 9:646772. doi: 10.3389/feart.2021.646772 g y y The Chinese Altay orogen, located in Northwest China, is the most important beryllium ore belt in China. Its deposit is mainly pegmatite type. The geological characteristics (Zou and Li, 2006; Qin et al., 2013; Yang et al., 2018; Zhang et al., 2019), geochronology (Wang et al., 2007; Chen, 2011; Ren March 2021 \| Volume 9 \| Article 646772 1 Frontiers in Earth Science \| www.frontiersin.org Yong et al. Yong et al. Genesis of Askartor Be-Mo Deposit FIGURE 1 \| Regional geological map of Xinjiang Altay modiﬁed after Windley et al. (2002). FIGURE 1 \| Regional geological map of Xinjiang Altay modiﬁed after Windley et al. (2002). Regional Geology High precision U-Pb dating results of LA- ICPMS, SHRIMP, and SIMS shows that most of the granitoids were emplaced during the Early to Middle Paleozoic time, with minor ones in the Permian period (Zou et al., 1988; Chen and Jahn, 2002; Wang et al., 2006; Yuan et al., 2007; Sun et al., 2008). The Early Middle Paleozoic I and S-type granitoid, which is emplaced from 507 to 360 Ma, display arc-related geochemical characters (Yuan et al., 2007; Wang et al., 2009; Cai et al., 2011a, Cai et al., 2011b). They are usually deformed and partly metamorphosed to become foliated (Wang et al., 2009). In contrast, the A or I-type granitoids emplaced at ca. 290–250 Ma are mostly undeformed and possess within-plate geochemical characteristics (Tong et al., 2014). The Askartor Be-Mo deposit is composed of primary ore and placer. The primary ore consists of pegmatoid-type beryllium ore body in the upper part and granitite-type beryllium ore body in the lower part, and the two parts show a gradual transitional relationship. Detailed surface geological mapping, underground mining workings and drill holes demonstrate that the following six zones are successively encountered from bottom to top (Figure 3): (1) medium-to ﬁne-grain muscovite granite. These rocks are mainly composed of quartz (30–35 vol%), plagioclase (35 vol%), K-feldspar (±20 vol%), and muscovite (5–10 vol%). The Chinese Altay orogen is an important rare metal metallogenic belt in China, and its deposit is mainly pegmatite. Approximately 100,000 pegmatite dykes exposed in the Chinese Altay (Zou and Li, 2006). These pegmatite dykes formed during the Devonian synorogenic to Jurassic post- orogenic to anorogenic stage, and mostly concentrated in Triassic (e.g., Ren et al., 2011; Lv et al., 2021 and references therein). The Askartor Be-Mo deposit is the only rare metal deposit related to granite. The Chinese Altay orogen is an important rare metal metallogenic belt in China, and its deposit is mainly pegmatite. Approximately 100,000 pegmatite dykes exposed in the Chinese Altay (Zou and Li, 2006). These pegmatite dykes formed during the Devonian synorogenic to Jurassic post- orogenic to anorogenic stage, and mostly concentrated in Triassic (e.g., Ren et al., 2011; Lv et al., 2021 and references therein). The Askartor Be-Mo deposit is the only rare metal deposit related to granite. (2) beryl-bearing ﬁne-grain albitized muscovite granite. They are grey-white, and have a ﬁne-grain granitic structure with a massive texture. Regional Geology Genesis of Askartor Be-Mo Deposit 11,200 tons of ore, which occupy only 2% of the total reserves, have been produced until 1996. FIGURE 2 \| Simpliﬁed geological map of the Askartor Be-Mo deposit district modiﬁed after Wang et al. (2015). p The Upper Ordovician Habahe subgroup sporadically crops out in the mining area, and is composed of biotite-quartz schist and biotite gneiss. Magmatic rocks develop well and make up 95% of the whole mining area, including granodiorites, biotite granites, two mica granites, and medium-ﬁne grain muscovite granites (Figure 2). Granodiorites are the most voluminous of these intrusive rocks. These rocks are grey-black, ﬁne to medium- grained, with a massive texture. They are composed of quartz (20–25 vol%), plagioclase (50 vol%), K-feldspar (±10 vol%), and biotite (10–15 vol%). Accessory minerals include zircon, apatite, titanite, and magnetite. Biotite granites are the ore-hosting rock, with an outcrop area of about 0.9 km2. These biotite granites are grey-white, and have an unequal-particle hypidiomorphic structure with a massive texture. Petrographic observation shows that the mineral compositions mainly involve quartz (30–35 vol%), plagioclase (±40 vol%), K-feldspar (±15 vol%), biotite (5–10 vol%), and muscovite (±2 vol%). Biotite granites are intruded by two mica granites, and the boundary between those two intrusions is sharp and clear. The surface outcrop of the two mica granites is up to 5 km2. They are grey-white, and have a medium-to coarse grain granitic structure with a massive texture. The two mica granites are mainly composed of quartz (35 vol%), plagioclase (∼35 vol%), K-feldspar (20 vol%), biotite (±5 vol%), and muscovite (±5 vol%). Muscovite granites occur in the upper portion of the two mica granites, and the transition between both granite types is gradational. The Muscovite granites have a ﬁne-to medium-grain granitic structure. These rocks are mainly composed of quartz (30–35 vol%), plagioclase (35 vol%), K-feldspar (±20 vol%), and muscovite (5–10 vol%). FIGURE 2 \| Simpliﬁed geological map of the Askartor Be-Mo deposit district modiﬁed after Wang et al. (2015). cratons (e.g., Xiao et al., 2008, Xiao et al., 2009, Xiao et al., 2015; Cai et al., 2016; Xiao et al., 2018). The granitoids occupy over ∼40% of the Chinese Altay orogen (Zou et al., 1988). Regional Geology The Altay orogen, which is the southernmost part of the Central Asian Orogenic Belt (CAOB), is situated between the south Siberian Craton to the north and the Junggar Block to the south. According to Windley et al. (2002), the Altay orogen in China is composed of the following six terranes, and details of the terranes can be found in Windley et al. (2002). The rare metal deposits are mainly distributed in Terrane 3 and Terrane 4 (Figure 1). The Askartor deposit is the only granite-related Be-Mo deposit in the Chinese Altay orogen. The Askartor granite pluton is a composite rock body with obvious vertical zoning, from bottom to top, the rock types change from two-mica granite, through muscovite granite, to pegmatite. In addition, there are granodiorite and biotite granite in the ore district. Zou and Li (2006) presented the geological features, and ore body morphology. The ages of different rock types have been presented by Wang et al. (2015), Ding et al. (2016), Zhang et al. (2017). However, the processes that might have contributed to the genesis of this deposit have not been noticed. Therefore, in this study, we have carried out a systemic study of the geology, geochronology, and geochemistry, in order to establish the mineralization model of the Askartor Be-Mo deposit, and identify the key factors restricting the enrichment of Be. Recent studies have showed that the Chinese Altay orogen blet is a subduction-related accretionary orogen blet. The tectonic setting of the Chinese Altay varied from a passive continental margin to active continental margin during the Middle Cambrian, and then, it underwent a prolonged subduction process from Ordovician to Carboniferous, characterized by massive magma activities, and high-temperature metamorphism (e.g., Sun et al., 2008; Jiang et al., 2010; Cai et al., 2010; Cai et al., 2011a; Cai et al., 2012a; Cai et al., 2011b). During the Permian period, the Chinese Altai converged consecutively with the East and West Junggar arcs because of the closure of the Paleo-Asian Ocean (e.g., Cai et al., 2012b; Li et al., 2015; Cai et al., 2016; Broussolle et al., 2018). Since the Triassic period, the Chinese Altai has entered into a post- orogenic stage After the amalgamation of the Siberia and Tarim March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 2 Yong et al. Regional Geology These rocks are mainly composed of quartz (30–35 vol%), albite (±40 vol%), K-feldspar (±20 vol%), and muscovite (±10 vol%). Beryl sporadically occurs in these rocks. (3) banded muscovite-quartz-albite zone (Figures 4A,B). This zone is usually 4∼5 m wide, 0.1∼2 m long and spans 10 m at its widest place. Meanwhile, it has a strong Be-mineralization and a gradual transition relationship with ﬁne-grained albitization muscovite granitite zone. Albite, microcline, quartz, muscovite, and beryl formed the zone’s clear banded structure. Geology of the ore District g The Arskartor Be deposit is located in the southeastern of the Central Altay, and about 80 km north of the Qinghe County (Figure 1). This deposit has mined since 1954, and a total of March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 3 Yong et al. Yong et al. Genesis of Askartor Be-Mo Deposit FIGURE 3 \| Cross sections of the Askartor No.1 Be-Mo deposit. FIGURE 3 \| Cross sections of the Askartor No.1 Be-Mo deposit. FIGURE 3 \| Cross sections of the Askartor No.1 Be-Mo deposit. (4) Massive quartz zone (Figure 4D). This zone mainly consists of quartz, the colors of which are ivory, rosy, and transparency with dense block structure and the widest place of the core quartz reaches 32 m. Beryl only appears along the marginal areas, reﬂecting a weak Be- mineralization. There is pyrite, blende, bismuthinite, nesting of gypsum and barite vein in this zone, besides quartz, and the molybdenite appears in the inside of quartz vein’s edge with little content. 0.1–5 m wide, is the primary Be-mineralization zone. The main mineral compositions are muscovite (20–30 vol%), quartz (10–20 vol%), beryl (3–5 vol%), and microcline (2–20 vol%). The molybdenite presents star-dotted lamellar and laminated shapes, coexisting with pyrite, bismuthinite, blende, spessartite, apatite, spodumene, and niobite. (6) (6) muscovite-microcline-albite zone: This zone, which is 2–10 m thick with weak Be-mineralization, is in the top side of the deposit, contacting the biotite granitite wall- rock directly. The main mineral compositions are quartz (6) muscovite-microcline-albite zone: This zone, which is 2–10 m thick with weak Be-mineralization, is in the top side of the deposit, contacting the biotite granitite wall- rock directly. The main mineral compositions are quartz (6) muscovite-microcline-albite zone: This zone, which is 2–10 m thick with weak Be-mineralization, is in the top side of the deposit, contacting the biotite granitite wall- rock directly. The main mineral compositions are quartz (5) Beryl-bearing muscovite-quartz zone (Figure 4C). This zone, showing circular distribution around the massive quartz with March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 4 Yong et al. Genesis of Askartor Be-Mo Deposit FIGURE 4 \| Examples of beryl in different zones of the No.1 orebody. FIGURE 4 \| Examples of beryl in different zones of the No.1 orebody. Sciences (SKLODG, CAS). Major elements were analyzed by standard X-ray ﬂuorescence (XRF). Geology of the ore District Samples were prepared as glass discs, formed by mixing 0.70 g of rock powder (dried at 110°C) with 7.0 g of lithium tetraborate for 15 min at 1,100°C in 95%P-5%Au crucibles. Analyses were performed on a PANalytical Axios PW4400 X-ray ﬂuorescence spectrometer. The analytical precision, as determined on the Chinese National Standards GSR-1 and GSR-3, was better than 5%. Loss on ignition (LOI) was obtained using 1 g power heated to 1,100°C for 1 h. (30 vol%), microcline (39 vol%), plagioclase (20 vol%), and muscovite (10 vol%), and the accessory minerals are spessartite, pyrite and molybdenite. Samples Rock samples were examined by optical microscopy, and unaltered or least altered samples were selected for geochemical analysis. These samples were cleaned with deionized water, crushed, and powered with an agate mill. Five samples were selected for zircon U-Pb dating in this study. These samples are granodiorite (sample name: AKG01, Location: N47°17′32″, E90°17′38″), biotite granite (sample name: AKG02, location:N47°18′21″,E90°17′39″), two-mica granite (sample name: AKG04, N47°17′33″, E90°17′34″), ﬁne-grain muscovite (sample name: AKG05, location: N47°18′19″, E90°17′43″), and the banded pegmatite (sample name: AKG06, N47°18′20″, E90°17′45″), respectively. Seven molybdenite samples were collected from ﬁne-grain muscovite granite and pegmatite. Trace elements were analyzed with a Finnigan MAT ELEMENT magnetic sector ICP-MS. 50 mg of powdered granite sample were dissolved in a high-pressure PolyTetraFluoroEthylene (PTFE) bomb for 48 h using HF + HNO3. Rh was used as an internal standard to monitor signal drift during counting. The USGS and Chinese National Standards BCR-1, BHVO-1, and GSR-3 were chosen for calibrating element concentrations. The discrepancy between the triplicates is less than 5% for all the elements (Qi et al., 2000). Zircon U-Pb Dating The zircon U-Pb dating samples were crashed, and grinded, and then separated by heavy liquid and magnetic separation methods. Zircons were picked by hand under a binocular microscope and mounted in epoxy resin. Before U-Pb dating analysis, The Major and Trace Elements Analysis Major and trace elements were determined at the State Key Laboratory of Ore Deposit Geochemistry, Chinese Academy of Major and Trace Elements Analysis Major and Trace Elements Analysis j y Major and trace elements were determined at the State Key Laboratory of Ore Deposit Geochemistry, Chinese Academy of Major and trace elements were determined at the State Key Laboratory of Ore Deposit Geochemistry, Chinese Academy of March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 5 Yong et al. Genesis of Askartor Be-Mo Deposit FIGURE 5 \| Representative cathodoluminescence (CL) images of zircons from the different intrusions with analytical spots of the Askartor deposit. (A) granodiorite; (B) biotite granite; (C) two-mica granite; (D) muscovite granite; (E) pegmatite. FIGURE 5 \| Representative cathodoluminescence (CL) images of zircons from the different intrusions with analytical spots of the Askartor deposit. (A) granodiorite; (B) biotite granite; (C) two-mica granite; (D) muscovite granite; (E) pegmatite. FIGURE 5 \| Representative cathodoluminescence (CL) images of zircons from the different intrusions with analytical spots of the Askartor deposit. (A) granodiorite; (B) biotite granite; (C) two-mica granite; (D) muscovite granite; (E) pegmatite. analyses were done at SKLODG, CAS. The analysis procedures are similar to that provided by Du et al. (2004). The ISOPLOT program (Ludwig, 2003) was used to calculate the Re-Os isochron age. The decay constant used in this calculation was λ187Re 1.666 × 10–11 year−1 (Smoliar et al., 1996). Uncertainty in Re-Os model ages was 1.02% uncertainty. The uncertainty includes the 187Re decay constant and uncertainty in Re and Os concentrations. Cathodoluminescence (CL) images were used to assess the zircon at the Beijing zircon dating navigation technology limited company, and CL images performed at JSM6510 scanning electron microscope produced by JEOL Corporation (Japan). Zircon U-Pb dating was done by an Elan DRC-e ICP-MS coupled with a GeoLasPro 193 nm Laser-Ablation System at SKLODG, CAS. The conditions were summarized as follows: the laser frequency was 8 Hz with an energy of 34–40 mJ; the beam diameter was 30 μm; and the acquisition was 30 s for background and 60 s for signal. Helium was used as a carrier gas to transport the ablated materials from the laser-ablation cell to the ICP-MS torch. The external standard is Zircon 91,500, and the internal standard to calculate U, Th, and Pb concentrations of unknowns was NIST610. Glass was used as an internal standard to Zircon U-Pb Ages g Granodiorite (AKG01): the zircon grains from granodiorite are mostly light-yellow brown, transparent, and short prismatic. Their sizes are between 100 and −300 μm. Almost all grains show clear oscillatory zoning (Figure 5A). The contents of Th and U are respectively 8–253 ppm and 172–326 ppm, and the Th/ U ratios are between 0.41 and −0.85 (Table 1), indicating that these zircons are of magmatic origin. Twenty spots were selected for U-Pb analysis and these points were tightly grouped, yielding a 206Pb/238U age of 386.8 ± 2.6 Ma (MSWD 0.54) (Figure 6A), representing the crystallization age of the pluton. Raw data were processed using the ICPMSDataCal program (Liu et al., 2008). Uncertainties of individual analyses and weighted mean ages are reported within 1σ errors; Common Pb correction used the 204Pb methods of Andersen (2002). The weighted mean U-Pb ages and Concordia plots were processed using ISOPLOT software (Ludwig 2003). Molybdenite Re-Os Dating Seven molybdenite samples were collected from ﬁne-grain granite and pegmatite. Molybdenite with a purity of >99% were carefully handpicked under a binocular microscope. Re-Os isotope Biotite granite (AKG02): the feature of zircon grains from biotite granite is similar to that in granodiorite (Figure 5B). The contents of Th and U are 73∼453 ppm and 102∼646 ppm, March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 6 Genesis of Askartor Be-Mo Deposit Yong et al. TABLE 1 \| LA-ICPMS U-Pb data of zircons from the Askaetor Be deposit. Molybdenite Re-Os Dating Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG01.01 114 189 0.60 0.05164 0.00273 0.43061 0.02071 0.06106 0.00112 364 15 382 7 AKG01.02 116 233 0.50 0.04802 0.00216 0.40546 0.01694 0.06058 0.00096 346 12 379 6 AKG01.03 123 241 0.51 0.04695 0.00319 0.39674 0.02423 0.06084 0.00132 339 18 381 8 AKG01.04 120 222 0.54 0.05362 0.00220 0.46003 0.01773 0.06159 0.00098 384 12 385 6 AKG01.05 146 212 0.69 0.05318 0.00242 0.45116 0.02016 0.06043 0.00094 378 14 378 6 AKG01.06 160 277 0.58 0.05267 0.00247 0.45924 0.02003 0.06214 0.00102 384 14 389 6 AKG01.07 153 287 0.53 0.04878 0.00217 0.42531 0.01754 0.06251 0.00099 360 12 391 6 AKG01.08 130 222 0.58 0.05370 0.00250 0.46034 0.02064 0.06139 0.00105 384 14 384 6 AKG01.09 210 263 0.80 0.05074 0.00258 0.42444 0.02039 0.06044 0.00116 359 15 378 7 AKG01.10 112 185 0.60 0.04680 0.00249 0.41514 0.02213 0.06297 0.00120 353 16 394 7 AKG01.11 176 219 0.80 0.05420 0.00233 0.46224 0.01759 0.06187 0.00097 386 12 387 6 AKG01.12 149 326 0.46 0.04624 0.00185 0.39560 0.01460 0.06182 0.00090 338 11 387 5 AKG01.13 129 215 0.60 0.05190 0.00443 0.45338 0.03755 0.06280 0.00142 380 26 393 9 AKG01.14 112 219 0.51 0.05318 0.00301 0.45134 0.02534 0.06144 0.00135 378 18 384 8 AKG01.15 108 172 0.62 0.04750 0.00261 0.41188 0.02123 0.06300 0.00129 350 15 394 8 AKG01.16 127 308 0.41 0.04889 0.00225 0.42556 0.01818 0.06292 0.00106 360 13 393 6 AKG01.17 185 281 0.66 0.04605 0.00232 0.39918 0.01924 0.06253 0.00123 341 14 391 7 AKG01.18 124 188 0.66 0.05419 0.00273 0.47174 0.02214 0.06299 0.00116 392 15 394 7 AKG01.19 102 206 0.50 0.05198 0.00273 0.44181 0.02155 0.06250 0.00121 372 15 391 7 AKG01.20 157 249 0.63 0.05019 0.00234 0.43714 0.01953 0.06235 0.00104 368 14 390 6 AKG01.21 91 196 0.46 0.05867 0.00322 0.50699 0.02966 0.06161 0.00122 416 20 385 7 AKG01.22 184 282 0.65 0.05268 0.00212 0.45656 0.01764 0.06186 0.00095 382 12 387 6 AKG01.23 232 311 0.75 0.05240 0.00225 0.45382 0.01851 0.06205 0.00106 380 13 388 6 AKG01.24 253 297 0.85 0.05368 0.00227 0.46268 0.01870 0.06178 0.00105 386 13 386 6 Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG01.01 73 102 0.72 0.06418 0.00470 0.53331 0.03401 0.06208 0.00151 434 23 388 9 AKG01.02 182 337 0.54 0.05139 0.00192 0.44326 0.01562 0.06212 0.00089 373 11 388 5 AKG01.03 155 178 0.87 0.05423 0.00264 0.47109 0.02247 0.06244 0.00106 392 16 390 6 AKG01.04 109 242 0.45 0.05391 0.00258 0.45821 0.02113 0.06130 0.00090 383 15 384 5 AKG01.05 301 339 0.89 0.05167 0.00198 0.44134 0.01561 0.06155 0.00086 371 11 385 5 AKG01.06 339 363 0.93 0.05158 0.00183 0.44679 0.01488 0.06199 0.00077 375 10 388 5 AKG01.07 85 163 0.52 0.05297 0.00276 0.45214 0.02318 0.06173 0.00114 379 16 386 7 AKG01.08 82 181 0.45 0.06687 0.00316 0.57185 0.02548 0.06135 0.00113 459 16 384 7 AKG01.09 81 156 0.52 0.05182 0.00308 0.43671 0.02144 0.06210 0.00109 368 15 388 7 AKG01.10 122 207 0.59 0.05923 0.00266 0.51857 0.02309 0.06226 0.00100 424 15 389 6 AKG01.11 123 140 0.88 0.05480 0.00266 0.47020 0.02110 0.06266 0.00105 391 15 392 6 AKG01.12 453 497 0.91 0.05195 0.00153 0.45040 0.01274 0.06154 0.00078 378 9 385 5 AKG01.13 218 320 0.68 0.05196 0.00201 0.45766 0.01591 0.06303 0.00083 383 11 394 5 AKG01.14 221 420 0.53 0.04953 0.00187 0.42675 0.01517 0.06117 0.00075 361 11 383 5 AKG01.15 174 211 0.83 0.05082 0.00227 0.44076 0.01884 0.06214 0.00093 371 13 389 6 AKG01.16 273 329 0.83 0.05443 0.00199 0.47273 0.01595 0.06174 0.00080 393 11 386 5 AKG01.17 93 161 0.58 0.05384 0.00261 0.45893 0.01995 0.06235 0.00112 384 14 390 7 AKG01.18 208 328 0.63 0.05005 0.00333 0.42699 0.02773 0.06047 0.00134 361 20 379 8 AKG01.19 149 191 0.78 0.05056 0.00220 0.42830 0.01684 0.06143 0.00095 362 12 384 6 AKG01.20 251 646 0.39 0.05066 0.00148 0.44541 0.01285 0.06221 0.00074 374 9 389 4 AKG01.21 137 269 0.51 0.05317 0.00245 0.46259 0.02204 0.06240 0.00104 386 15 390 6 AKG01.22 97 182 0.53 0.05282 0.00245 0.44924 0.02006 0.06178 0.00103 377 14 386 6 AKG01.23 120 304 0.40 0.05901 0.00223 0.44341 0.01598 0.05430 0.00074 373 11 341 5 AKG01.24 343 435 0.79 0.05139 0.00176 0.41835 0.01375 0.05813 0.00073 355 10 364 4 Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG04.01 1,592 11,768 0.14 0.04965 0.00161 0.28534 0.00956 0.03956 0.00066 255 8 250 4 AKG04.02 1,680 12,157 0.14 0.05605 0.00187 0.32628 0.01686 0.03903 0.00123 287 13 247 8 AKG04.03 1,160 8,796 0.13 0.05658 0.00171 0.31940 0.00928 0.03910 0.00047 281 7 247 3 AKG04.04 60,118 15,458 3.89 0.06415 0.00178 0.36682 0.01357 0.03885 0.00082 317 10 246 5 AKG04.05 1,486 10,054 0.15 0.05197 0.00164 0.29778 0.01214 0.03901 0.00089 265 9 247 6 (Continued on following page) TABLE 1 \| LA-ICPMS U-Pb data of zircons from the Askaetor Be deposit. Molybdenite Re-Os Dating March 2021 \| Volume 9 \| Article 646772 7 Frontiers in Earth Science \| www.frontiersin.org Genesis of Askartor Be-Mo Deposit Yong et al. TABLE 1 \| (Continued) LA-ICPMS U-Pb data of zircons from the Askaetor Be deposit. Frontiers in Earth Science \| www.frontiersin.org March 2021 \| Volume 9 \| Article 646772 Molybdenite Re-Os Dating Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG04.06 772 9,875 0.08 0.05248 0.00161 0.30379 0.00996 0.03984 0.00054 269 8 252 3 AKG04.07 911 8,517 0.11 0.04810 0.00175 0.28231 0.01325 0.03942 0.00093 252 10 249 6 AKG04.08 2066 15,936 0.13 0.04504 0.00167 0.25770 0.01127 0.03992 0.00154 233 9 252 10 AKG04.09 1,507 10,554 0.14 0.05904 0.00163 0.34457 0.01288 0.03914 0.00092 301 10 248 6 AKG04.10 1,040 9,661 0.11 0.05247 0.00162 0.29513 0.00845 0.03839 0.00062 263 7 243 4 AKG04.11 617 9,999 0.06 0.04926 0.00144 0.27990 0.00824 0.03841 0.00068 251 7 243 4 AKG04.12 772 11,048 0.07 0.05071 0.00159 0.29490 0.00942 0.03907 0.00074 262 7 247 5 AKG04.13 1,561 8,696 0.18 0.05884 0.00191 0.36347 0.01064 0.04147 0.00058 315 8 262 4 AKG04.14 1,187 11,746 0.10 0.05119 0.00172 0.29746 0.01073 0.03867 0.00084 264 8 245 5 AKG04.15 482 10,083 0.05 0.05817 0.00180 0.33289 0.01242 0.03852 0.00089 292 9 244 6 AKG04.16 455 7,250 0.06 0.04612 0.00146 0.26626 0.00815 0.03939 0.00054 240 7 249 3 AKG04.17 1,080 7,121 0.15 0.05212 0.00227 0.29562 0.01312 0.03980 0.00085 263 10 252 5 Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG05.01 466 8,360 0.06 0.05887 0.00396 0.27643 0.01825 0.03406 0.00046 248 15 216 3 AKG05.02 1,539 9,060 0.17 0.07022 0.00370 0.36196 0.01832 0.03739 0.00055 314 14 237 3 AKG05.03 1,385 32,709 0.04 0.05839 0.00128 0.30657 0.00648 0.03672 0.00050 272 5 232 3 AKG05.04 2,175 9,735 0.22 0.06268 0.00175 0.33484 0.00922 0.03708 0.00051 293 7 235 3 AKG05.05 1,244 7,222 0.17 0.06124 0.00281 0.30387 0.01322 0.03599 0.00053 269 10 228 3 AKG05.06 4,544 24,671 0.18 0.07573 0.00706 0.36702 0.03374 0.03515 0.00053 317 25 223 3 AKG05.07 1,453 6,478 0.22 0.05318 0.00213 0.28493 0.00935 0.03679 0.00068 255 7 233 4 AKG05.08 515 10,428 0.05 0.04876 0.00215 0.25751 0.01019 0.03616 0.00056 233 8 229 3 AKG05.09 1,048 6,806 0.15 0.05874 0.00285 0.28896 0.01335 0.03568 0.00052 258 11 226 3 AKG05.10 12,226 6,492 1.88 0.14124 0.00673 0.84036 0.04060 0.04005 0.00079 619 22 253 5 AKG05.11 4,420 8,378 0.53 0.09897 0.00833 0.54416 0.04485 0.03988 0.00069 441 29 252 4 AKG05.12 2,713 9,038 0.30 0.06395 0.00301 0.34327 0.01609 0.03606 0.00060 300 12 228 4 AKG05.13 1,671 18,199 0.09 0.05063 0.00322 0.23443 0.01453 0.03358 0.00047 214 12 213 3 AKG05.14 971 6,915 0.14 0.05554 0.00366 0.27971 0.01792 0.03652 0.00055 250 14 231 3 AKG05.15 1,130 25,526 0.04 0.05239 0.00243 0.28043 0.01268 0.03597 0.00060 251 10 228 4 AKG05.16 858 7,029 0.12 0.05235 0.00356 0.26304 0.01704 0.03645 0.00074 237 14 231 5 AKG05.17 2,947 5,148 0.57 0.07749 0.00518 0.44492 0.02590 0.03994 0.00093 374 18 252 6 AKG05–18 1,217 7,419 0.16 0.07066 0.00401 0.35344 0.01911 0.03628 0.00062 307 14 230 4 Sample Concentration (ppm) Th/U Isotopic ratio 207Pb/235U 1σ 206Pb/238U 1σ Ages (Ma) Th U 207Pb/206Pb 1σ 207Pb/235U 1σ 206Pb/238U 1σ AKG06.01 3,936 6,450 0.61 0.0778 0.0030 0.3940 0.0135 0.0353 0.0008 337 10 224 5 AKG06.02 3,533 6,229 0.57 0.0623 0.0024 0.3125 0.0117 0.0344 0.0007 276 9 218 4 AKG06.03 1,124 5,935 0.19 0.0456 0.0015 0.2249 0.0070 0.0339 0.0006 206 6 215 3 AKG06.04 2,150 4,780 0.45 0.0495 0.0019 0.2630 0.0098 0.0364 0.0006 237 8 230 4 AKG06.05 952 4,751 0.20 0.0474 0.0019 0.2396 0.0091 0.0348 0.0006 218 7 220 4 AKG06.06 1,603 4,578 0.35 0.0494 0.0019 0.2522 0.0093 0.0350 0.0006 228 8 222 3 AKG06.07 2,383 5,664 0.42 0.0443 0.0016 0.2227 0.0078 0.0342 0.0006 204 7 217 4 AKG06.08 796 6,841 0.12 0.0434 0.0016 0.2183 0.0076 0.0343 0.0006 201 6 218 4 AKG06.09 1867 4,985 0.37 0.0557 0.0019 0.2807 0.0096 0.0342 0.0006 251 8 217 4 AKG06.10 811 6,042 0.13 0.0391 0.0015 0.2009 0.0071 0.0352 0.0006 186 6 223 4 AKG06.11 1,594 7,267 0.22 0.0419 0.0016 0.2116 0.0075 0.0345 0.0006 195 6 219 4 AKG06.12 291 5,837 0.05 0.0432 0.0014 0.2148 0.0066 0.0341 0.0005 198 6 216 3 AKG06.13 2075 7,608 0.27 0.0395 0.0016 0.1959 0.0081 0.0340 0.0006 182 7 215 4 AKG06.14 890 5,567 0.16 0.0400 0.0015 0.2030 0.0068 0.0351 0.0007 188 6 223 4 AKG06.15 576 4,896 0.12 0.0353 0.0012 0.1814 0.0059 0.0353 0.0005 169 5 224 3 AKG06.16 372 4,870 0.08 0.0354 0.0013 0.1807 0.0063 0.0354 0.0005 169 5 224 3 AKG06.17 2,609 5,378 0.49 0.0387 0.0016 0.1992 0.0082 0.0355 0.0007 184 7 225 4 AKG06.18 1,122 5,172 0.22 0.0404 0.0014 0.2084 0.0071 0.0355 0.0005 192 6 225 3 AKG06.19 1743 6,841 0.25 0.0375 0.0012 0.1916 0.0059 0.0349 0.0005 178 5 221 3 AKG06.20 1953 4,277 0.46 0.0458 0.0026 0.2348 0.0131 0.0353 0.0008 214 11 224 5 AKG06.21 2,160 4,034 0.54 0.0474 0.0017 0.2434 0.0090 0.0348 0.0006 221 7 221 4 AKG0622 538 4,539 0.12 0.0436 0.0015 0.2278 0.0076 0.0351 0.0005 208 6 223 3 AKG0623 459 4,607 0.10 0.0449 0.0020 0.2320 0.0099 0.0350 0.0007 212 8 222 4 AKG0624 3,936 6,450 0.61 0.0778 0.0030 0.3940 0.0135 0.0353 0.0008 337 10 224 5 Frontiers in Earth Science \| www.frontiersin.org March 2021 \| Volume 9 \| Article 646772 8 8 Genesis of Askartor Be-Mo Deposit Yong et al. Molybdenite Re-Os Dating Yong et al. FIGURE 6 \| Zircon U-Pb age of the different intrusions from the Askartor Be-Mo deposit. (A) granodiorite; (B) biotite granite; (C) two-mica granite; (D) muscovite granite; (E) pegmatite, and (F) Re-Os isochron age of molybdenites. FIGURE 6 \| Zircon U-Pb age of the different intrusions from the Askartor Be-Mo deposit. (A) granodiorite; (B) biotite granite; (C) two-mica granite; (D) muscovite granite; (E) pegmatite, and (F) Re-Os isochron age of molybdenites. melting. The zircons mostly have low Th/U ratios of 0.1–0.2. Seventeen analytical points were conducted on the zircons with legible oscillatory zones, and sixteen of them yielded consistent and concordant 206Pb/238U ages with a weighted mean age of 247.5 ± 2.2 Ma (MSWD 0.45) (Figure 6C), which represents the crystallization time of the muscovite granite. The rest one analytical point yielded an older 206Pb/238U age of 262 ± 3.6 Ma (Table 1, AKG04.16), which may reﬂect the incorporation of common lead after crystallization. respectively, and the Th/U ratios ranged from 0.39 to 0.93. Twenty-three yielded concordant 206Pb/238U ages with a weighted mean age of 385.8 ± 4.4 Ma (MSWD 0.42, n 23) (Figure 6B), representing the crystallization age of this rock. One point showed a younger age of 341 ± 5 Ma (Table 1, AKG02.23), which may be affected by minor Pb loss. Two-mica granite (AKG04): The zircon grains separated from two-mica granite are mostly brown and opaque, and varying in size of 80–220 μm, with euhedral habit featuring tetragonal dipyramid ± prism. They show weak and intense luminescences in CL photos, with reservation of primary oscillatory zone in some zircons (Figure 5C). The CL features indicate that the zircons have suffered from low degree of recrystallization induced by metamictimation due to high contents of U (7,121∼15936 ppm) and Th (mostly fall in range of 455∼2066 ppm) and ﬂuid alteration after crystallization from Fine-grain muscovite granite (AKG05): Zircons from ﬁne- grain muscovite granite generally exhibit euhedral tetragonal- dipyramid or stubby tetragonal prism shapes and range from 80 to 180 μm in size. Most of them are opaque and brown to dark brown in color, with weak oscillatory zoning structure (Figure 5D). Th and U contents of the zircon grain are 466–4,544 ppm and 5,148–32,709 ppm, respectively, with Th/U March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 9 Yong et al. Yong et al. Molybdenite Re-Os Dating Genesis of Askartor Be-Mo Deposit TABLE 2 \| The whole-rock compositions of different intrusions from Arskartor Be deposit. Molybdenite Re-Os Dating Rock type Two-mica granite Muscovite granite Sample AKG-12–04K-A AKG-12–03A-A AKG-12–03B-D AKG-12–03C AKG-12–03D AKG-12–03E AKG-12–04D-A AKG-12–04E-A AKG-12–04F AKG-12–04Q-D AKG-12–04C CaO 0.19 0.26 0.25 0.16 0.25 0.22 0.07 0.33 0.15 0.09 0.09 Na2O 4.04 4.36 3.77 4.49 4.53 4.76 5.51 5.41 5.96 6.03 4.52 K2O 4.47 4.14 4.80 3.87 3.77 3.96 3.90 2.92 4.39 3.07 3.38 P2O5 0.11 0.09 0.11 0.11 0.11 0.14 0.11 0.19 0.15 0.13 0.10 LOI 0.55 0.74 1.00 0.91 1.00 0.84 1.04 0.69 0.49 0.89 0.75 Total 99.08 99.77 99.46 99.42 99.47 99.51 98.80 99.26 99.31 99.69 99.72 A/CNK 1.24 1.22 1.16 1.23 1.21 1.20 1.14 1.12 1.08 1.13 1.37 ppm Li 70.8 66.2 45.5 56.5 45.5 37.5 30.4 23.5 23.7 29.7 36.3 Be 10.0 10.5 9.7 10.3 11.0 8.0 200.6 311.9 300.8 355.9 175.2 Rb 359.0 323.0 349.0 344.0 317.0 431.0 448.0 587.0 454.0 318.0 385.0 Sr 34.6 6.8 7.7 5.0 15.0 8.3 14.6 21.0 21.1 12.4 5.9 Y 2.8 2.3 2.9 4.4 3.0 3.9 5.3 5.8 5.6 5.6 2.5 Zr 15.1 18.2 15.5 33.2 14.6 23.0 26.7 23.0 20.8 21.4 22.5 Nb 17.5 14.4 8.6 13.9 15.5 21.4 98.7 96.7 110.0 105.0 80.6 Ba 6.6 7.2 5.2 3.1 4.0 13.6 15.4 32.3 30.6 19.5 5.9 La 1.35 1.82 1.89 1.79 1.05 1.18 0.41 0.68 1.04 0.90 0.58 Ce 2.93 3.54 3.74 3.36 2.26 2.20 0.79 1.56 2.05 1.51 1.16 Pr 0.35 0.43 0.49 0.42 0.24 0.27 0.10 0.19 0.21 0.11 0.13 Nd 1.20 1.45 1.58 1.41 0.70 0.84 0.32 0.68 0.66 0.40 0.41 Sm 0.32 0.32 0.38 0.37 0.21 0.23 0.08 0.16 0.20 0.10 0.10 Eu 0.05 0.04 0.05 0.03 0.04 0.03 0.01 0.02 0.03 0.02 0.02 Gd 0.25 0.26 0.26 0.31 0.18 0.20 0.09 0.18 0.15 0.12 0.10 Tb 0.06 0.05 0.05 0.08 0.04 0.05 0.04 0.06 0.05 0.06 0.03 Dy 0.35 0.30 0.37 0.58 0.29 0.39 0.40 0.53 0.44 0.53 0.21 Ho 0.07 0.07 0.07 0.12 0.07 0.08 0.12 0.14 0.09 0.12 0.05 Er 0.22 0.21 0.26 0.45 0.26 0.32 0.35 0.39 0.38 0.47 0.22 Tm 0.04 0.04 0.04 0.09 0.06 0.07 0.09 0.12 0.08 0.08 0.06 Yb 0.27 0.30 0.33 0.66 0.46 0.59 0.77 0.88 0.70 0.81 0.59 Lu 0.04 0.04 0.05 0.11 0.07 0.09 0.11 0.11 0.11 0.11 0.10 Hf 0.8 0.9 0.7 1.7 0.8 1.5 4.3 4.1 3.5 4.0 1.6 Ta 3.2 2.2 1.3 2.0 2.4 3.4 61.4 116.0 98.2 78.8 51.2 Th 2.8 2.7 2.1 2.9 1.5 1.6 3.4 3.9 4.8 5.4 0.8 U 2.1 8.0 1.5 3.2 1.4 2.0 18.4 16.6 10.8 5.8 2.3 Nb/Ta 5.4 6.6 6.8 6.9 6.5 6.2 1.6 0.8 1.1 1.3 1.6 Zr/Hf 18.5 19.3 21.8 19.2 17.4 15.1 6.2 5.6 5.9 5.3 14.2 Y/Ho 38.6 34.7 40.0 35.8 42.9 48.9 45.4 42.6 60.3 45.5 54.2 Sr/Eu 666.7 168.9 144.4 167.9 419.4 261.3 1,440.6 900.2 841.5 563.1 370.7 REE 7.50 8.87 9.56 9.76 5.93 6.52 3.66 5.70 6.20 5.33 3.75 δEu 0.56 0.43 0.52 0.27 0.56 0.44 0.37 0.42 0.43 0.61 0.48 δCe 1.03 0.96 0.94 0.93 1.08 0.94 0.94 1.04 1.06 1.16 1.00 TE3.4 0.38 0.28 0.32 0.33 0.31 0.31 0.30 0.38 0.31 0.18 0.28 ce \| www.frontiersin.org March 2021 \| Volume 11 (Continued) The whole-rock compositions of different intrusions from Arskartor Be deposit. Molybdenite Re-Os Dating Rock type Two-mica granite Muscovite granite Sample AKG-12–04K-A AKG-12–03A-A AKG-12–03B-D AKG-12–03C AKG-12–03D AKG-12–03E AKG-12–04D-A AKG-12–04E-A AKG-12–04F AKG-12–04Q-D AKG-12–04C wt% SiO2 74.77 73.87 73.37 73.10 73.05 74.25 75.06 74.58 74.21 73.70 74.40 TiO2 0.07 0.09 0.24 0.19 0.20 0.16 0.09 0.05 0.15 0.25 0.04 Al2O3 14.47 14.38 13.43 14.86 14.04 14.16 14.06 14.16 14.40 13.61 14.71 TFe2O3 0.76 0.83 0.92 1.11 1.14 0.97 0.71 0.49 0.90 0.97 0.35 MnO 0.03 0.04 0.04 0.05 0.04 0.05 0.02 0.02 0.05 0.02 0.02 MgO 0.17 0.20 2.37 0.27 0.29 0.22 0.17 0.12 0.19 0.04 0.05 CaO 0.36 0.53 0.60 0.74 0.76 0.42 0.48 0.58 0.56 0.60 0.10 Na2O 4.28 4.21 3.47 3.72 3.06 3.41 4.06 4.58 3.64 3.31 4.85 K2O 3.89 4.11 3.99 4.39 4.12 4.01 4.13 3.83 4.46 4.12 3.81 P2O5 0.17 0.19 0.16 0.18 0.19 0.18 0.17 0.14 0.18 0.18 0.07 LOI 0.87 0.94 1.32 0.54 1.23 1.63 0.72 0.66 0.97 1.09 1.12 Total 99.84 99.39 99.91 99.15 98.12 99.45 99.67 99.21 99.69 97.89 99.53 A/CNK 1.22 1.17 1.21 1.22 1.29 1.32 1.17 1.11 1.22 1.24 1.20 ppm Li 107.0 148.0 148.0 183.0 211.0 236.0 100.0 75.1 151.0 97.8 59.7 Be 12.6 12.8 19.5 12.6 21.7 15.3 9.1 9.0 12.7 13.8 13.9 Rb 345.0 401.0 386.0 343.0 345.0 350.0 309.0 271.0 364.0 369.0 338.0 Sr 21.7 32.8 46.4 52.8 47.8 43.2 19.2 14.7 24.1 25.7 18.3 Y 6.8 5.9 8.0 8.6 9.1 5.6 4.0 3.4 5.8 3.2 2.4 Zr 59.8 67.4 83.9 106.0 119.0 92.3 54.0 33.3 83.8 60.0 16.7 Nb 19.7 20.3 17.6 18.9 19.9 21.2 14.4 12.1 17.2 13.1 16.0 Ba 58.4 88.7 135.0 131.0 104.0 115.0 26.2 15.2 50.7 77.3 16.4 La 4.86 11.60 13.70 23.50 19.90 12.90 8.57 5.76 10.90 8.07 0.67 Ce 15.00 26.20 28.80 50.50 41.90 34.00 19.30 11.00 23.10 21.40 1.30 Pr 1.35 3.08 3.52 6.23 5.27 3.39 2.22 1.49 2.97 2.10 0.16 Nd 5.31 11.10 13.40 23.10 19.60 12.60 8.08 5.43 11.30 7.59 0.50 Sm 1.09 1.89 2.28 3.89 3.61 2.14 1.43 1.04 2.05 1.34 0.16 Eu 0.13 0.25 0.32 0.41 0.36 0.25 0.16 0.11 0.20 0.16 0.02 Gd 0.89 1.19 1.55 2.35 2.20 1.41 0.98 0.58 1.30 0.77 0.14 Tb 0.17 0.19 0.23 0.35 0.33 0.21 0.14 0.10 0.21 0.13 0.03 Dy 1.04 0.95 1.20 1.53 1.54 1.00 0.64 0.52 0.98 0.55 0.24 Ho 0.22 0.18 0.24 0.28 0.28 0.18 0.12 0.10 0.17 0.09 0.05 Er 0.69 0.54 0.82 0.82 0.85 0.54 0.37 0.33 0.56 0.30 0.20 Tm 0.12 0.09 0.15 0.12 0.12 0.08 0.05 0.05 0.08 0.05 0.04 Yb 0.78 0.67 1.23 0.80 0.83 0.60 0.38 0.35 0.54 0.30 0.35 Lu 0.13 0.10 0.19 0.11 0.12 0.10 0.05 0.05 0.08 0.04 0.05 Hf 2.3 2.3 2.6 3.0 3.6 2.9 1.9 1.3 2.8 1.9 1.0 Ta 3.3 2.1 2.6 2.6 3.1 2.1 1.7 1.5 3.0 2.6 2.1 Th 8.0 11.1 12.2 21.1 19.8 13.6 8.5 4.8 15.3 9.6 1.4 U 1.8 4.6 1.9 40.1 11.3 2.1 4.9 0.9 13.3 1.5 1.5 Nb/Ta 6.0 9.8 6.8 7.2 6.3 9.9 8.2 8.0 5.7 5.1 7.7 Zr/Hf 26.3 30.0 32.0 34.9 32.8 31.4 28.9 26.0 30.0 31.1 16.4 Y/Ho 30.5 32.7 33.4 31.2 32.6 30.2 34.1 34.5 33.6 34.4 48.6 Sr/Eu 172.8 131.7 143.8 129.2 131.4 170.8 123.7 128.1 121.5 158.0 877.8 REE 31.77 58.03 67.63 113.98 96.91 69.40 42.48 26.92 54.44 42.89 3.91 δEu 0.39 0.50 0.52 0.41 0.39 0.44 0.40 0.45 0.37 0.49 0.42 δCe 1.40 1.05 0.99 1.00 0.98 1.23 1.06 0.90 0.97 1.25 0.96 TE3.4 1.03 1.04 1.02 1.02 1.00 0.99 0.98 1.03 1.03 1.06 1.16 Rock type Muscovite granite Beryl-bearing muscovite granite sample AKG-12–04G AKG-12–04J AKG-12–04M AKG-12–04N AKY-12–04L AKG-12–04O AKG-12–04B AKG-12–04H AKG-12–04R-A AKG-12–04S AKG-12–04I wt% SiO2 74.33 74.85 75.41 74.87 74.86 74.04 72.64 75.05 71.92 74.13 75.04 TiO2 0.06 0.05 0.04 0.03 0.04 0.03 0.01 0.01 0.01 0.01 0.03 Al2O3 14.70 14.81 13.78 14.61 14.51 15.00 15.34 14.15 15.98 15.19 15.41 TFe2O3 0.51 0.42 0.27 0.33 0.34 0.41 0.17 0.43 0.22 0.11 0.32 MnO 0.02 0.02 0.01 0.02 0.02 0.08 0.02 0.04 0.01 0.04 0.08 MgO 0.09 0.04 0.03 0.02 0.04 0.02 0.00 0.04 0.03 0.00 0.00 (Continued on following page) March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 10 Yong et al. March 2021 \| Volume 9 \| Article 646772 Molybdenite Re-Os Dating Yong et al. Genesis of Askartor Be-Mo Deposit TABLE 2 \| (Continued) The whole-rock compositions of different intrusions from Arskartor Be deposit. Molybdenite Re-Os Dating Two-mica granite Muscovite granite AKG-12–04K-A AKG-12–03A-A AKG-12–03B-D AKG-12–03C AKG-12–03D AKG-12–03E AKG-12–04D-A AKG-12–04E-A AKG-12–04F AKG-12–04Q-D AKG-12–04C 0.19 0.26 0.25 0.16 0.25 0.22 0.07 0.33 0.15 0.09 0.09 4.04 4.36 3.77 4.49 4.53 4.76 5.51 5.41 5.96 6.03 4.52 4.47 4.14 4.80 3.87 3.77 3.96 3.90 2.92 4.39 3.07 3.38 0.11 0.09 0.11 0.11 0.11 0.14 0.11 0.19 0.15 0.13 0.10 0.55 0.74 1.00 0.91 1.00 0.84 1.04 0.69 0.49 0.89 0.75 99.08 99.77 99.46 99.42 99.47 99.51 98.80 99.26 99.31 99.69 99.72 1.24 1.22 1.16 1.23 1.21 1.20 1.14 1.12 1.08 1.13 1.37 ppm 70.8 66.2 45.5 56.5 45.5 37.5 30.4 23.5 23.7 29.7 36.3 10.0 10.5 9.7 10.3 11.0 8.0 200.6 311.9 300.8 355.9 175.2 359.0 323.0 349.0 344.0 317.0 431.0 448.0 587.0 454.0 318.0 385.0 34.6 6.8 7.7 5.0 15.0 8.3 14.6 21.0 21.1 12.4 5.9 2.8 2.3 2.9 4.4 3.0 3.9 5.3 5.8 5.6 5.6 2.5 15.1 18.2 15.5 33.2 14.6 23.0 26.7 23.0 20.8 21.4 22.5 17.5 14.4 8.6 13.9 15.5 21.4 98.7 96.7 110.0 105.0 80.6 6.6 7.2 5.2 3.1 4.0 13.6 15.4 32.3 30.6 19.5 5.9 1.35 1.82 1.89 1.79 1.05 1.18 0.41 0.68 1.04 0.90 0.58 2.93 3.54 3.74 3.36 2.26 2.20 0.79 1.56 2.05 1.51 1.16 0.35 0.43 0.49 0.42 0.24 0.27 0.10 0.19 0.21 0.11 0.13 1.20 1.45 1.58 1.41 0.70 0.84 0.32 0.68 0.66 0.40 0.41 0.32 0.32 0.38 0.37 0.21 0.23 0.08 0.16 0.20 0.10 0.10 0.05 0.04 0.05 0.03 0.04 0.03 0.01 0.02 0.03 0.02 0.02 0.25 0.26 0.26 0.31 0.18 0.20 0.09 0.18 0.15 0.12 0.10 0.06 0.05 0.05 0.08 0.04 0.05 0.04 0.06 0.05 0.06 0.03 0.35 0.30 0.37 0.58 0.29 0.39 0.40 0.53 0.44 0.53 0.21 0.07 0.07 0.07 0.12 0.07 0.08 0.12 0.14 0.09 0.12 0.05 0.22 0.21 0.26 0.45 0.26 0.32 0.35 0.39 0.38 0.47 0.22 0.04 0.04 0.04 0.09 0.06 0.07 0.09 0.12 0.08 0.08 0.06 0.27 0.30 0.33 0.66 0.46 0.59 0.77 0.88 0.70 0.81 0.59 0.04 0.04 0.05 0.11 0.07 0.09 0.11 0.11 0.11 0.11 0.10 0.8 0.9 0.7 1.7 0.8 1.5 4.3 4.1 3.5 4.0 1.6 3.2 2.2 1.3 2.0 2.4 3.4 61.4 116.0 98.2 78.8 51.2 2.8 2.7 2.1 2.9 1.5 1.6 3.4 3.9 4.8 5.4 0.8 2.1 8.0 1.5 3.2 1.4 2.0 18.4 16.6 10.8 5.8 2.3 5.4 6.6 6.8 6.9 6.5 6.2 1.6 0.8 1.1 1.3 1.6 18.5 19.3 21.8 19.2 17.4 15.1 6.2 5.6 5.9 5.3 14.2 38.6 34.7 40.0 35.8 42.9 48.9 45.4 42.6 60.3 45.5 54.2 666.7 168.9 144.4 167.9 419.4 261.3 1,440.6 900.2 841.5 563.1 370.7 7.50 8.87 9.56 9.76 5.93 6.52 3.66 5.70 6.20 5.33 3.75 0.56 0.43 0.52 0.27 0.56 0.44 0.37 0.42 0.43 0.61 0.48 1.03 0.96 0.94 0.93 1.08 0.94 0.94 1.04 1.06 1.16 1.00 0.38 0.28 0.32 0.33 0.31 0.31 0.30 0.38 0.31 0.18 0.28 March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 11 Genesis of Askartor Be-Mo Deposit Yong et al. Molybdenite Re-Os Dating (REE 27–114.0 ppm), medium ﬁne-grained muscovite granite ( REE 4–10 ppm) to beryl-bearing muscovite granite ( REE 4–6 ppm); The rock samples all have negative Eu anomalies (δ Eu 0.27–0.61). the Lanthanide tetrad effects in the REE patterns are observed in muscovite granite. The degrees of the tetrad effects (TE3,4) in the muscovite and beryl-bearing muscovite granite are in the range of 1.08∼1.19 and 1.24∼1.34, respectively, (Table 2; Figures 8A,B). ratios in the range of 0.06–0.30 (Table 1). Seventeen spots were selected from domains of homogenous composition and clear oscillatory zoning for U-Pb analysis. Of these, eleven yield an age population with a weighted mean 206Pb/238U age of 231.4 ± 2.0 Ma (Figure 6D), representing the crystallization age of zircons. Four spots yield younger ages (213–226), deviating from the concordia due to a small amount of Pb loss or inclusions. The remaining spot yields an older age of 252 Ma, most likely representing the formation age of xenocrysts. y p g g y Pegmatite (AKG06): The zircon grains from the banded pegmatite collected for U-Pb dating are similar to those from sample AKG05 in color, transparency, crystal habit, and grain size. Some zircons show metasomatic relict textures in CL photos, and are characterized by microtaxitic mantle-core with nonuniform luminescences and relict rim with uniform luminescence, indicating evident metasomatic alteration after crystallization. The primary zircon domains are featured by weak and uniform luminescences, embayed texture, and regular or irregular shapes, and are enclosed by spongy domains (Figure 5E). All these features suggest that the zircons have suffered from metamictimation, metasomatism, and recrystallization to different extents. The zircons have Th and U contents of 291∼3,936 ppm and 4,034∼7,608 ppm, respectively, with varying Th/U ratios from 0.05 to 0.61. Twenty-four analytical points were conducted on the primary zircon domains and twenty-two of them yielded consistent and concordant 206Pb/238U ages with a weighted mean age of 220.6 ± 1.6 Ma (MSWD 0.80) (Figure 6E), which represents the crystallization time of the pegmatite. The other two points yielded inconcordant ages, indicating the incorporation of common lead after crystallization or inﬂuence of mineral inclusion in zircon. In the normalized spider web diagram of trace element primitive mantle, all of the rock samples show enrichment of large ion lithophile elements, and have obvious negative anomalies of Ba, Sr, and Ti (Figures 8C,D). The Genesis of Askartor Be-Mo Deposit The Genesis of Askartor Be-Mo Deposit The two-mica granite, muscovite granite, beryl-bearing muscovite granite, and pegmatoid shell have a close spatial and temporal relationship, with similar zircon Hf isotopic composition (Wang et al., 2015). These features indicate that these rocks may be the products of the same magmatic system at different evolution stages. DISCUSSION The Formation age of Be Deposit in Altay The zircon U-Pb age of pegmatoid orebody is 220.6 ± 1.6 Ma which coincides with the molybdenite Re-Os isochron age of 228.7 ± 7.1. This age is slightly younger than that of muscovite granite (231.4 ± 2.0 Ma). The zircon U-Pb age of the Askartor Be- Mo deposit obtained in this paper is consistent with that of the previous study (Wang et al., 2015) (Table 4). Recent studies have revealed that the geodynamic change in the Triassic period played a key role in the formation of rare- metal deposits in Altay, NW China. These deposits include (see Table 4): Koktokay No.3 pegmatite-type Li-Be-Ta-Nb-Cs deposit (220–209 Ma, Chen, 2011; Wang et al., 2007); Koktokay No.1 and Koktokay No.11 pegmatite-type Li-Be- Ta-Nb (208 and 212 Ma, respectively, Ren et al., 2011); Kelumute No.112 pegmatite-type Li-Be-Nb-Ta deposit (238–211 Ma, Lv et al., 2012), and Fuhai No.11 pegmatite- type Li-Be-Nb-Ta deposit (203 Ma, Lv et al., 2012). Some medium-scale Be-Nb-Ta deposits and small-scale pegmatite- type Be in Altay, such as Kalaerqisi, Qiebielin-xiaokalasu, and Jiamanhaba pegmatites were dated at 250–209 Ma (Ren et al., 2011; Ma et al., 2015). Molybdenite Re-Os Ages The Re-Os isotope results of the 7 molybdenite samples are shown in Table 2. The Re content of molybdenite is between 38 and 56 ppm with an average of 47 ppm. The Re-Os model ages of 7 molybdenite samples range from 224.6 ± 3.1 to 235 ± 3.4 Ma, with a weighted mean age of 229.0 ± 3.0 Ma (Table 3). ISOPLOT software was used to calculate the isochron age for the seven data points, and derived an isochron age of 228.7 ± 7.1 Ma (Figure 6D). The isochron age represents the time of molybdenite crystallization. Thus, the formation of Askartor Be-Mo deposit is not an accidental metallogenic event, but is representative of the Triassic rare metal metallogenic event in Altay. Geochemical Characteristic The two-mica granite, muscovite granite, and beryl-bearing muscovite granite are rich in SiO2 (71.92–75.41 wt%), Al2O3 (13.43–15.98 wt%), and total alkali Na2O+ K2O (7.18–10.35 wt %), poor in CaO (0.07–0.76 wt%), TFe2O3 (0.11–1.14 wt%), and TiO2 (0.01–0.25 wt%) (Table 2). In the TAS diagram, all the samples fall in the granite area (Figure 7A); on the K2O-SiO2 diagram, most of the samples fall in the range of high-k calc alkaline series (Figure 7B); the A/CNK index varies from 1.1 to 1.3 (Figure 7C), with strong peraluminous characteristics. Compared with the former two, beryl-bearing muscovite granite has higher Na2O/K2O ratio (Na2O/K2O 1.2–2.0). The REE content decreases gradually from two mica granite As shown in Figure 9, although the beryl-bearing muscovite granite has higher Na2O content and Na2O/K2O ratio, the evolutionary relationship among the major elements of the three types of granites is not obvious. The proportion of rock- forming minerals (such as quartz and feldspars) of the three types of granites is basically the same, which is close to the proportion of the minimum composition, Ab37Ms10Or20Qtz (Icenhower and London, 1995). Therefore, these rocks can be the products of March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 12 Genesis of Askartor Be-Mo Deposit Yong et al. FIGURE 7 \| Geochemical characterization of the different intrusions from the Askartor Be-Mo deposit. (A) SiO2 vs. Na2O + K2O [after Middlemost (1994)]; (B) SiO2 vs. K2O [after Rickwood (1989)]; (C) A/CNK vs. A/NK [after M i d Pi li (1989)] TABLE 3 \| Re-Os data of molybdenite from the Arskartor Be deposit. Sample Re (ppm) 187Re (ppm) 187Os (ppb) Ages (Ma) Measured 1σ Measured 1σ Measured 1σ Measured 1σ QHA12–02 49.7 2.3 31.1 1.4 117.1 2.3 226.5 4.5 QHA12–03 43.3 2.0 27.1 1.3 103.8 2.2 230.4 4.8 QHA12–04 45.7 0.7 28.6 0.4 109.3 1.9 229.6 3.9 QHA12–05 38.3 0.7 24.0 0.4 93.90 1.4 235.7 3.4 QHA12–07 56.4 0.7 35.3 0.4 136.2 3.2 231.8 5.5 QHA12–08 46.8 0.8 29.3 0.5 109.4 1.5 224.6 3.1 QHA12–10 50.9 1.5 31.7 1.0 120.1 2.2 227.8 4.2 TABLE 3 \| Re-Os data of molybdenite from the Arskartor Be deposit. model continuous fractional crystallization in the same magmatic systems. One consequence of Goldschmidt’s Rule is that isovalent trace elements of equal or very similar ionic radii should remain tightly coupled in geological processes. Geochemical Characteristic If, however, in certain geological systems other properties become important, this coherency is likely to break down. In octahedral coordination, Y3+, Ho3+, Zr 4+, Hf4+, Nb5+, and Ta5+ show effective ionic radii of 1.04, 1.04, 0.86, 0.85, 0.78, and 0.78, respectively, (Shannon, 1976). Hence, the element pairs Y-Ho, Zr-Hf, and Nb-Ta are geochemical twin pairs. The fractionation of these trace element pairs occurred in muscovite granites, with the increase degree of the tetrad effect (Figure 10). The fractionation of isovalent trace element and the lanthanide tetrad are known to indicate magmatic-hydrothermal transition environments (e.g., Bau, 1996; Irber, 1999; Monecke et al., 2002; Monecke et al., 2007; Ballouard et al., 2016). The characteristics of trace elements indicate that muscovite granites have undergone self-metasomatism. This self-metasomatism refers to the reaction between the ﬂuid, excluding from the granitic magma, and incompletely crystallized granite. Based on the partition coefﬁcients of Be and the results of this study, we preliminarily established the mineralization model of Askartor Be-Mo deposit (Figure 11). The model is mainly established on the basis of the Rayleigh fractional crystallization equation: CRL CLf Ds−1 (1) CS CL × 1 −f DS 1 −f (2) (1) (2) CL: concentration of Be in initial magma. CL: concentration of Be in initial magma. RL CRL: instant concentration of Be in residual magma. S CS: average concentration of Be in crystalline solid DS: The total partition coefﬁcient of Be DS: The total partition coefﬁcient of Be f: ratio of residual magma to initial magma. f: ratio of residual magma to initial magma. The inclusion data show that the emplacement pressure and temperature of the initial magma of the Askartor Be-Mo deposit are 500 MPa and 750 ± 50°C (Ding, 2016). At these temperature and pressure conditions, the minimum and maximum water contents in the initial magma are 4 wt% and 10 wt%, respectively, (Holtz et al., 2001). When we assume that the water content of the initial is 4%, according to the balance calculation:4 wt% (1−f) × 0 wt% + f × 10 wt FIGURE 7 \| Geochemical characterization of the different intrusions from the Askartor Be-Mo deposit. (A) SiO2 vs. Na2O + K2O [after Middlemost (1994)]; (B) SiO2 vs. K2O [after Rickwood (1989)]; (C) A/CNK vs. A/NK [after Maniar and Piccoli (1989)]. FIGURE 7 \| Geochemical characterization of the different intrusions from the Askartor Be-Mo deposit. (A) SiO2 vs. Geochemical Characteristic Na2O + K2O [after Middlemost (1994)]; (B) SiO2 vs. K2O [after Rickwood (1989)]; (C) A/CNK vs. A/NK [after Maniar and Piccoli (1989)]. March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 13 Yong et al. Yong et al. Genesis of Askartor Be-Mo Deposit FIGURE 8 \| The Chondrite-normalized rare earth element (REE) patterns and Primitive-mantle-normalized spider diagram of the different intrusions from the Askartor Be-Mo deposit. (A, C) two-mica and muscovite granite; (B, D) beryl-bearing muscovite granite. Symbols are as in Figure 7. Chondrite and primitive-mantle values are from Sun and McDonough (1989). FIGURE 8 \| The Chondrite-normalized rare earth element (REE) patterns and Primitive-mantle-normalized spider diagram of the different intrusions from the Askartor Be-Mo deposit. (A, C) two-mica and muscovite granite; (B, D) beryl-bearing muscovite granite. Symbols are as in Figure 7. Chondrite and primitive-mantle values are from Sun and McDonough (1989). TABLE 4 \| Age of rare metal deposit in the Chinese Altay orogenic belt. Location Deposit type Deposit name Mineralization type Age (Ma) Dating method Reference Koktokay Pegmatite KKT03 Li-Be-Ta-Nb-Cs 220–209 U-Pb Chen (2011) U-Pb Wang et al. (2007) Pegmatite KKT01 Li-Be-Ta-Nb 208.1 ± 0.8 U-Pb Ren et al. (2011) Pegmatite KKT11 Li-Be-Ta-Nb 212.7 ± 2.5 U-Pb Ren et al. (2011) Kelumute-jideke Pegmatite FH11 Be-Ta-Nb 202.9 ± 0.8 U-Pb Lv et al. (2012) Pegmatite KLMT112 Li-Be-Ta-Nb 238–211 U-Pb Ren et al. (2011) Kalaerqisi Pegmatite KKLG650 Li-Be-Ta-Nb 228–211 U-Pb Ma et al. (2015) Pegmatite JMK02 Li-Be-Ta-Nb 212.2 ± 1.7 U-Pb Ren et al. (2011) Pegmatite QK02 Be-Ta-Nb 206.8 ± 1.6 U-Pb Pegmatite FH08 Be-Ta-Nb 244.3 ± 1.1 U-Pb Pegmatite ALT01 Be-Ta-Nb 246.8 ± 1.2 U-Pb Qiebielin-xiaokalasu Pegmatite BRJ10 Be 249.7 ± 0.7 U-Pb Pegmatite BRJ07 Be 240.5 ± 1.4 U-Pb Jiamanhaba Pegmatite HB07 Be 237.5 ± 2.6 U-Pb Askartor Granite ± pegmatite Askartor Be-Mo 220.6 ± 1.6 U-Pb This study 218.2 ± 3.9 U-Pb Wang et al. (2015) 228.7 ± 7.1 Re-Os This study 218.6 ± 1.3 Re-Os Wang et al. (2015) TABLE 4 \| Age of rare metal deposit in the Chinese Altay orogenic belt. March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 14 Genesis of Askartor Be-Mo Deposit Yong et al. Yong et al. FIGURE 9 \| Bivariate plots of the oxides Al2O3, K2O, Na2O, Fe2O3, CaO, and Na2O/K2O ratios vs. SiO2 of the different intrusions from the Askartor Be-Mo deposit. Geochemical Characteristic FIGURE 9 \| Bivariate plots of the oxides Al2O3, K2O, Na2O, Fe2O3, CaO, and Na2O/K2O ratios vs. SiO2 of the different intrusions from the Askartor Be-Mo deposit. oxides Al2O3, K2O, Na2O, Fe2O3, CaO, and Na2O/K2O ratios vs. SiO2 of the different intrusions from the Askartor Be-Mo deposit. FIGURE 9 \| Bivariate plots of the oxides Al2O3, K2O, Na2O, Fe2O3, CaO, and Na2O/K2O ratios vs. SiO2 of the different intrus %, when the water content in magma will reach 10 wt%, the f equals to 0.4. Even if we consider hydrous minerals like mica, because the content of mica is low, it has little effect on the results. Trace element data show that the two-mica granite should be the product of pure magmatic crystallization, so at this stage, the fractional degree of initial magma will not exceed 0.6. The Be content in two-mica granite is ∼13 ppm. If f 0.4, from Eq. 2, we can calculate that the Be content in the initial magma is ∼30 ppm, and then we can work out that the Be content in residual magma is ∼60 ppm based on Eq. 1. Then we can %, when the water content in magma will reach 10 wt%, the f equals to 0.4. Even if we consider hydrous minerals like mica, because the content of mica is low, it has little effect on the results. Trace element data show that the two-mica granite should be the product of pure magmatic crystallization, so at this stage, the fractional degree of initial magma will not exceed 0.6. The Be content in two-mica granite is ∼13 ppm. If f 0.4, from Eq. 2, we can calculate that the Be content in the initial magma is ∼30 ppm, and then we can work out that the Be content in residual magma is ∼60 ppm based on Eq. 1. Then we can use the Be content in the beryl-bearing muscovite granite to calculate the fractional degree in the second stage. The mass of the ﬂuid and the Be content in ﬂuid are gained via the mass balance. Rayleigh fractional calculations show that: a. the content of Be in the initial magma forming the Askartor deposit is as high as 30 ppm, which is 10 times of the average Be content in the upper crust (Be ∼3 ppm in upper crust, Taylor and McLennan, 1995). Geochemical Characteristic This value is similar to the Be a. the content of Be in the initial magma forming the Askartor deposit is as high as 30 ppm, which is 10 times of the average Be content in the upper crust (Be ∼3 ppm in upper crust, Taylor and McLennan, 1995). This value is similar to the Be March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 15 Genesis of Askartor Be-Mo Deposit Yong et al. FIGURE 10 \| The fractionation of isovalent trace elements of the different intrusions from the Askartor Be-Mo deposit. (A) Zr/Hf vs Nb/Ta, the different rare metal mineralization zones are deﬁned by Ballouard et al. (2016); (B) Zr/Hf content of the less-altered tuffs and pumices from the Macusani Province, Peru (28 ± 9 ppm, with a high of 37 ppm, Noble et al., 1984). These rhyolites are S-type in overall chemical character with high 87Sr/86Sr, strongly peraluminous composition, and trace element patters that follow from an abundance of argillic or micaceous material at the source (Pichavant et al., 1988). Partial melting of a typical pelitic mica schist with 3 ppm Be, which is composed of muscovite, biotite, plagioclase, and quartz, will yield a content of 4∼6 ppm Be in melt (Evensen and London, 2002), so, the pre-enrichment of Be in the source will restrict the mineralization of Be. FIGURE 10 \| The fractionation of isovalent trace elements of the different intrusions from the Askartor Be-Mo deposit. (A) Zr/Hf vs Nb/Ta, the different rare metal mineralization zones are deﬁned by Ballouard et al. (2016); (B) Zr/Hf vs Y/Ho, the “CHARAC ﬁeld with 24 < Y/Ho < 34 and 26 < Zr/Hf < 46” is deﬁned by Bau (1996); (C) TE3-4 vs Nb/Ta, Degree of tetrad effect (TE3−4) has been calculated using equation of Irber (1999). b. The initial magma is ﬂuid saturated after the crystallization of 60%, and the residual magma contains ∼60 ppm Be. After that, the system changed from pure magma (+crystal) stage to magma-hydrothermal transition stage. FIGURE 10 \| The fractionation of isovalent trace elements of the different intrusions from the Askartor Be-Mo deposit. (A) Zr/Hf vs Nb/Ta, the differe rare metal mineralization zones are deﬁned by Ballouard et al. Geochemical Characteristic (2016); (B) Zr/ vs Y/Ho, the “CHARAC ﬁeld with 24 < Y/Ho < 34 and 26 < Zr/Hf < 46” is FIGURE 11 \| A simple mineralization model for Askartor Be-Mo deposit. This model is based on the fractionation crystallization. The partition coefﬁcients of Be between minerals and melts are cited from Evensen and London (2002): DK-feldspar/melt 0.14; Dquartz/melt 0.24; Dalbite/melt 0.19; Dmuscovite/melt 1.35; Dbiotite/melt 0.54; the mineral proportions are close to the real mineral proportions of rocks, which are typical of mica granite systems. Based on the inclusions, the emplacement pressure and temperature of the rocks from Askartor Be-Mo deposit are ∼500 MPa and 800°C Ding (2016), so the H2O content in the initial magma was ∼4 wt% and the H2O solubility in melt at this condition was 10 wt% Holtz et al. (2001). FIGURE 11 \| A simple mineralization model for Askartor Be-Mo deposit. This model is based on the fractionation crystallization. The partition coefﬁcients of Be between minerals and melts are cited from Evensen and London (2002): DK-feldspar/melt 0.14; Dquartz/melt 0.24; Dalbite/melt 0.19; Dmuscovite/melt 1.35; Dbiotite/melt 0.54; the mineral proportions are close to the real mineral proportions of rocks, which are typical of mica granite systems. Based on the inclusions, the emplacement pressure and temperature of the rocks from Askartor Be-Mo deposit are ∼500 MPa and 800°C Ding (2016), so the H2O content in the initial magma was ∼4 wt% and the H2O solubility in melt at this condition was 10 wt% Holtz et al. (2001). FIGURE 11 \| A simple mineralization model for Askartor Be-Mo deposit. This model is based on the fractionation crystallization. The partition coefﬁcients of Be between minerals and melts are cited from Evensen and London (2002): DK-feldspar/melt 0.14; Dquartz/melt 0.24; Dalbite/melt 0.19; Dmuscovite/melt 1.35; Dbiotite/melt 0.54; the mineral proportions are close to the real mineral proportions of rocks, which are typical of mica granite systems. Based on the inclusions, the emplacement pressure and temperature of the rocks from Askartor Be-Mo deposit are ∼500 MPa and 800°C Ding (2016), so the H2O content in the initial magma was ∼4 wt% and the H2O solubility in melt at this condition was 10 wt% Holtz et al. (2001). FIGURE 11 \| A simple mineralization model for Askartor Be-Mo deposit. This model is based on the fractionation crystallization. REFERENCES Cai, K., Sun, M., Yuan, C., Xiao, W., Zhao, G., Long, X., et al. (2012b). Carboniferous mantle-derived felsic intrusion in the Chinese Altai, NW China: implications for geodynamic change of the accretionary orogenic belt. Gondwana Res. 22 (2), 681–698. doi:10.1016/j.gr.2011.11.008 Andersen, T. (2002). Correction of common lead in U-Pb analyses that do not report 204Pb. Chem. Geol. 192 (1), 59–79. doi:10.1016/s0009-2541(02)00195-x Cai, K., Sun, M., Jahn, B.-M., Xiao, W., Long, X., Chen, H., et al. (2016). Petrogenesis of the permian intermediate-maﬁc dikes in the Chinese Altai, northwest China: implication for a postaccretion extensional scenario. J. Geol. 124 (4), 481–500. doi:10.1086/686464 Ballouard, C., Poujol, M., Boulvais, P., Branquet, Y., Tartèse, R., and Vigneresse, J.- L. (2016). Nb-Ta fractionation in peraluminous granites: a marker of the magmatic-hydrothermal transition. Geology 44 (3), 231–234. doi:10.1130/ g37475.1 Chen, B., and Jahn, B.-M. (2002). Geochemical and isotopic studies of the sedimentary and granitic rocks of the Altai orogen of Northwest China and their tectonic implications. Geol. Mag. 139 (1), 1–13. doi:10.1017/ S0016756801006100 Barton, M. D., and Young, S. (2002). Non-pegmatitic deposits of beryllium: mineralogy, geology, phase equilibria and origin. Rev. Mineral. Geochem. 50 (1), 591–691. doi:10.2138/rmg.2002.50.14 Chen, J.-F. (2011). Geochemistry of the plate part in Altai No.3 pegmatite and its formation and evolution. Master’s thesis. Beijing (China): University of Chinese Academy of Sciences. Bau, M. (1996). Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contrib. Mineral. Petrol. 123 (3), 323–333. doi:10.1007/ s004100050159 Ding, X., Li, J.-K., Ding, J.-G., Wang, S., Liu, Y.-C., and Wang, X. (2016). Molybdenite Re-Os isochron age and geological implication in Asikaerte Be- Nb-Mo deposit of Xinjiang. J. Guilin Univ. Technol. 36 (1), 60–65. doi:10.3969/j. issn.1674-9057.2016.01.009 [in Chinese, with English summary]. Broussolle, A., Aguilar, C., Sun, M., Schulmann, K., ˇStípská, P., Jiang, Y., et al. (2018). Polycyclic Palaeozoic evolution of accretionary orogenic wedge in the southern Chinese Altai: evidence from structural relationships and U-Pb geochronology. Lithos 314–315, 400–424. doi:10.1016/j.lithos.2018.06.005 n.1674-9057.2016.01.009 [in Chinese, with English summary]. Ding, X. (2016). Ore-forming mechanism of Askartor granite type berlyllium deposit in Xinjiang, China. Master’s thesis. Wuhan (China): China University of Geosciences. Cai, K., Sun, M., Yuan, C., Zhao, G., Xiao, W., Long, X., et al. (2010). Geochronological and geochemical study of maﬁc dykes from the northwest Chinese Altai: implications for petrogenesis and tectonic evolution. Gondwana Res. 18 (4), 638–652. Geochemical Characteristic The partition coefﬁcients of Be between minerals and melts are cited from Evensen and London (2002): DK-feldspar/melt 0.14; Dquartz/melt 0.24; Dalbite/melt 0.19; Dmuscovite/melt 1.35; Dbiotite/melt 0.54; the mineral proportions are close to the real mineral proportions of rocks, which are typical of mica granite systems. Based on the inclusions, the emplacement pressure and temperature of the rocks from Askartor Be-Mo deposit are ∼500 MPa and 800°C Ding (2016), so the H2O content in the initial magma was ∼4 wt% and the H2O solubility in melt at this condition was 10 wt% Holtz et al. (2001). FIGURE 10 \| The fractionation of isovalent trace elements of the different intrusions from the Askartor Be-Mo deposit. (A) Zr/Hf vs Nb/Ta, the different rare metal mineralization zones are deﬁned by Ballouard et al. (2016); (B) Zr/Hf vs Y/Ho, the “CHARAC ﬁeld with 24 < Y/Ho < 34 and 26 < Zr/Hf < 46” is deﬁned by Bau (1996); (C) TE3-4 vs Nb/Ta, Degree of tetrad effect (TE3−4) has been calculated using equation of Irber (1999). March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 16 Yong et al. Genesis of Askartor Be-Mo Deposit c. When this smaller volume of magma has reached ∼80% solidiﬁcation, then the residual magma could achieve beryl saturation (Be > 200 ppm), assuming that it could be efﬁciently extracted to produce beryl-bearing muscovite granite and its overlying pegmatoid shell. (3) Rayleigh fractional calculations show that the Askartor Be- Mo deposit is the product of multistage fractional crystallization of initial Be-enriched magma. d. The Be content of the exsolution ﬂuid is as high as 160 ppm. The partition coefﬁcient of Be in coexisting ﬂuid and melt is 0.8, which is consistent with that obtained from the inclusion data (e.g., 0.37 ± 0.43, Zajacz et al., 2008). d. The Be content of the exsolution ﬂuid is as high as 160 ppm. The partition coefﬁcient of Be in coexisting ﬂuid and melt is 0.8, which is consistent with that obtained from the inclusion data (e.g., 0.37 ± 0.43, Zajacz et al., 2008). DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors. FUNDING (2) The highly fractionated element ratios of Y/Ho, Zr/Hf and Nb/Ta, plus the REE tetrad effect indicate the ﬂuid exsolution occurs at the late stage of magma evolution and the muscovite granite experienced strong self-metasomatism. The study is supported by the National Science Foundation of China (No. 91962222). The study is supported by the National Science Foundation of China (No. 91962222). AUTHOR CONTRIBUTIONS (1) the zircon U-Pb ages of the granodiorite, biotite granite, two- mica granite, muscovite granite and pegmatoid are 386.8 ± 2.6, 385.4 ± 4.4, 247.5 ± 2.2, 231.4 ± 2.0, and 220.6 ± 1.6 Ma, respectively. The molybdenite Re-Os isochron age is 228.7 ± 7.1, which coincides with the zircon U-Pb age of 220.6 ± 1.6 Ma of the pegmatoid orebody. The two-mica granite, muscovite granite, beryl-bearing muscovite granite and pegmatoid shell have close spatial and temporal relationships. (1) the zircon U-Pb ages of the granodiorite, biotite granite, two- mica granite, muscovite granite and pegmatoid are 386.8 ± 2.6, 385.4 ± 4.4, 247.5 ± 2.2, 231.4 ± 2.0, and 220.6 ± 1.6 Ma, respectively. The molybdenite Re-Os isochron age is 228.7 ± 7.1, which coincides with the zircon U-Pb age of 220.6 ± 1.6 Ma of the pegmatoid orebody. The two-mica granite, muscovite granite, beryl-bearing muscovite granite and pegmatoid shell have close spatial and temporal relationships. TY, ZH, and LZ-H designed the study and Participated in ﬁeld work, TY and LZ-H analysize the sample and acquired the data, All authors contributed to the writing of the manuscript, and read and approved the ﬁnal version. REFERENCES The lanthanide tetrad effect and its correlation with K/Rb, Eu/ Eu*, Sr/Eu, Y/Ho, and Zr/Hf of evolving peraluminous granite suites. Geochim. et Cosmochim. Acta 63 (3–4), 489–508. doi:10.1016/s0016-7037(99)00027-7 Smoliar, M. I., Walker, R. J., and Morgan, J. W. (1996). Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271 (5252), 1099–1102. doi:10. 1126/science.271.5252.1099 Jiang, Y., Sun, M., Zhao, G., Yuan, C., Xiao, W., Xia, X., et al. (2010). The 390 Ma high-T metamorphic event in the Chinese Altai: a consequence of ridge- subduction?. Am. J. Sci. 310 (10), 1421–1452. doi:10.2475/10.2010.08 Stepanov, A. S., and Hermann, J. (2013). Fractionation of Nb and Ta by biotite and phengite: implications for the “missing Nb paradox”. Geology 41 (3), 303–306. doi:10.1130/g33781.1 Li, P., Yuan, C., Sun, M., Long, X., and Cai, K. (2015). Thermochronological constraints on the late Paleozoic tectonic evolution of the southern Chinese Altai. J. Asian Earth Sci. 113, 51–60. doi:10.1016/j.jseaes.2014.11.004 Sun, M., Yuan, C., Xiao, W., Long, X., Xia, X., Zhao, G., et al. (2008). Zircon U-Pb and Hf isotopic study of gneissic rocks from the Chinese Altai: progressive accretionary history in the early to middle Palaeozoic. Chem. Geol. 247 (3), 352–383. doi:10.1016/j.chemgeo.2007.10.026 Liu, Y., Hu, Z., Gao, S., Günther, D., Xu, J., Gao, C., et al. (2008). In situ analysis of major and trace elements of anhydrous minerals by LA-ICP-MS without applying an internal standard. Chem. Geol. 257 (1), 34–43. doi:10.1016/j. chemgeo.2008.08.004 Sun, S.-S., and McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Geol. Soc. London Spec. Publ. 42 (1), 313–345. doi:10.1144/gsl.sp.1989.042.01.19 London, D., and Evensen, J. M. (2002). Beryllium in silicic magmas and the origin of beryl-bearing pegmatites. Rev. Mineral. Geochem. 50 (1), 445–486. doi:10. 2138/rmg.2002.50.11 Taylor, S. R., and McLennan, S. M. (1995). The geochemical evolution of the continental crust. Rev. Geophys. 33 (2), 241–265. doi:10.1029/95rg00262 Ludwig, K. (2003). User’s manual for Isoplot 3.00: a geochronological toolkit for microsoft excel. Berkeley, CA: Berkeley Geochronology Center Special Publication, 74. Tong, Y., Wang, T., Jahn, B.-M., Sun, M., Hong, D.-W., and Gao, J.-F. (2014). Post- accretionary permian granitoids in the Chinese Altai orogen: geochronology, petrogenesis and tectonic implications. Am. J. Sci. 314 (1), 80–109. doi:10.2475/ 01.2014.03 Lv, Z.-H., Zhang, H., and Tang, Y. (2021). Anatexis origin of rare metal/earth pegmatites: evidences from the Permian pegmatites in the Chinese Altai. Lithos 380–381, 105865. REFERENCES doi:10.1016/j.lithos.2020.105865 Wang, C.-L., Qin, K.-Z., Tang, D.-M., Zhou, Q.-F., Shen, M.-D., Guo, Z.-L., et al. (2015). Geochronology and Hf isotope of zircon for the Arskartor Be-Nb-Mo deposit in Altay and its geological implication. Acta Petrol. Sin. 31 (8), 2337–2352. Lv, Z.-H., Zhang, H., Tang, Y., and Guan, S.-J. (2012). Petrogenesis and magmatic- hydrothermal evolution time limitation of Kelumute No. 112 pegmatite in Altay, Northwestern China: evidence from zircon UPb and Hf isotopes. Lithos 154 (0), 374–391. doi:10.1016/j.lithos.2012.08.005 Wang, T., Hong, D. W., Jahn, B. M., Tong, Y., Wang, Y. B., Han, B. F., et al. (2006). Timing, petrogenesis, and setting of Paleozoic synorogenic intrusions from the Altai Mountains, Northwest China: implications for the tectonic evolution of an accretionary orogen. J. Geol. 114 (6), 735–751. doi:10.1086/507617 Ma, Z.-L., Zhang, H., Tang, Y., Lv, Z.-H., Zhang, X., and Zhao, J.-Y. (2015). Zircon U-Pb geochronology and Hf isotopes of pegmatites from the Kaluan mining area in the Altay, Xinjiang and their genetic relationship with the Halong granite. Geochimica 44 (1), 9–26. doi:10.19700/j.0379-1726.2015.01.002 [in Chinese, with English summary]. Wang, T., Jahn, B.-M., Kovach, V. P., Tong, Y., Hong, D.-W., and Han, B.-F. (2009). Nd-Sr isotopic mapping of the Chinese Altai and implications for continental growth in the central asian orogenic belt. Lithos 110 (1), 359–372. doi:10.1016/j.lithos.2009.02.001 Maniar, P. D., and Piccoli, P. M. (1989). Tectonic discrimination of granitoids. Geol. Soc. Am. Bull. 101, 635–643. doi:10.1130/0016-7606(1989)101<0635: tdog>2.3.co;2 Wang, T., Tong, Y., Jahn, B.-M., Zou, T.-R., Wang, Y.-B., Hong, D.-W., et al. (2007). SHRIMP U-Pb Zircon geochronology of the Altai No. 3 Pegmatite, NW China, and its implications for the origin and tectonic setting of the pegmatite. Ore Geol. Rev. 32 (1–2), 325–336. doi:10.1016/j.oregeorev.2006.10.001 Middlemost, E. A. K. (1994). Naming materials in the magma/igneous rock system. Earth Sci. Rev. 37 (3), 215–224. doi:10.1016/0012-8252(94)90029-9 Monecke, T., Dulski, P., and Kempe, U. (2007). Origin of convex tetrads in rare earth element patterns of hydrothermally altered siliceous igneous rocks from the Zinnwald Sn-W deposit, Germany. Geochim. et Cosmochim. Acta 71 (2), 335–353. doi:10.1016/j.gca.2006.09.010 Windley, B. F., Kröner, A., Guo, J., Qu, G., Li, Y., and Zhang, C. (2002). Neoproterozoic to Paleozoic geology of the Altai orogen, NW China: new zircon age data and tectonic evolution. J. Geol. 110 (6), 719–737. doi:10.1086/ 342866 Monecke, T., Kempe, U., Monecke, J., Sala, M., and Wolf, D. (2002). REFERENCES doi:10.1016/j.gr.2010.02.010 Du, A., Wu, S., Sun, D., Wang, S., Qu, W., Markey, R., et al. (2004). Preparation and certiﬁcation of Re-Os dating reference materials: molybdenites HLP and JDC. Geostand. Geoanalyt. Res. 28 (1), 41–52. doi:10.1111/j.1751-908X.2004. tb01042.x Cai, K., Sun, M., Yuan, C., Zhao, G., Xiao, W., Long, X., et al. (2011a). Geochronology, petrogenesis and tectonic signiﬁcance of peraluminous granites from the Chinese Altai, NW China. Lithos 127 (1), 261–281. doi:10. 1016/j.lithos.2011.09.001 Evensen, J. M., and London, D. (2002). Experimental silicate mineral/melt partition coefﬁcients for beryllium and the crustal Be cycle from migmatite to pegmatite. Geochim. et Cosmochim. Acta 66 (12), 2239–2265. doi:10.1016/s0016-7037(02) 00889-x Cai, K., Sun, M., Yuan, C., Zhao, G., Xiao, W., Long, X., et al. (2011b). Prolonged magmatism, juvenile nature and tectonic evolution of the Chinese Altai, NW China: evidence from zircon U-Pb and Hf isotopic study of Paleozoic granitoids. J. Asian Earth Sci. 42 (5), 949–968. doi:10.1016/j.jseaes.2010.11.020 Foley, N. K., Jaskula, B. W., Piatak, N. M., and Schulte, R. F., II (2017). “Beryllium,” in Critical mineral resources of the United States-economic and environmental geology and prospects for future supply. Editors K. J. Schulz, J. H. DeYoung, R. R. Jr Seal, and D. C. Bradley (Reston, VA: U.S. Geological Survey Professional Paper), 797. Cai, K., Sun, M., Yuan, C., Zhao, G., Xiao, W., and Long, X. (2012a). Keketuohai maﬁc-ultramaﬁc complex in the Chinese Altai, NW China: petrogenesis and geodynamic signiﬁcance. Chem. Geol. 294–295, 26–41. doi:10.1016/j.chemgeo. 2011.11.031 Holtz, F., Johannes, W., Tamic, N., and Behrens, H. (2001). Maximum and minimum water contents of granitic melts generated in the crust: a March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 17 Yong et al. Genesis of Askartor Be-Mo Deposit reevaluation and implications. Lithos 56 (1), 1–14. doi:10.1016/s0024-4937(00) 00056-6 Mineral. Sin. 31 (3), 587–596. doi:10.16461/j.cnki.1000-4734.2011.03.041 [in Chinese, with English summary]. Icenhower, J., and London, D. (1995). An experimental study of element partitioning among biotite, muscovite, and coexisting peraluminous silicic melt at 200 MPa (H2O). Am. Mineral. 80 (11–12), 1229–1251. doi:10.2138/ am-1995-11-1213 Rickwood, P. C. (1989). Boundary lines within petrologic diagrams which use oxides of major and minor elements. Lithos 22 (4), 247–263. doi:10.1016/0024- 4937(89)90028-5 Shannon, R. D. (1976). Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Cryst. Sect. A 32 (5), 751–767. doi:10.1107/s0567739476001551 Irber, W. (1999). REFERENCES Tetrad effect in rare earth element distribution patterns: a method of quantiﬁcation with application to rock and mineral samples from granite-related rare metal deposits. Geochim. et Cosmochim. Acta 66 (7), 1185–1196. doi:10.1016/ s0016-7037(01)00849-3 Xiao, W., Han, C., Yuan, C., Sun, M., Lin, S., Chen, H., et al. (2008). Middle cambrian to permian subduction-related accretionary orogenesis of northern Xinjiang, NW China: implications for the tectonic evolution of central asia. J. Asian Earth Sci. 32 (2), 102–117. doi:10.1016/j.jseaes.2007.10.008 Noble, D. C., Vogel, T. A., Peterson, P. S., Landis, G. P., Grant, N. K., Jezek, P. A., et al. (1984). Rare-element-enriched, S-type ash-ﬂow tuffs containing phenocrysts of muscovite, andalusite, and sillimanite, southeastern Peru. Geology 12 (1), 35–39. doi:10.1130/0091-7613(1984)12<35:rsatcp>2.0.co10. 1130/0091-7613(1984)12<35:rsatcp>2.0.co;2 Xiao, W. J., Windley, B. F., Huang, B. C., Han, C. M., Yuan, C., Chen, H. L., et al. (2009). End-Permian to mid-Triassic termination of the accretionary processes of the southern Altaids: implications for the geodynamic evolution, Phanerozoic continental growth, and metallogeny of Central Asia. Int. J. Earth Sci. 98 (6), 1189–1217. doi:10.1007/s00531-008-0407-z Pichavant, M., Kontak, D. J., Herrera, J. V., and Clark, A. H. (1988). The miocene- pliocene Macusani volcanics, SE Peru. Contrib. Mineral. Petrol. 100 (3), 300–324. doi:10.1007/BF00379741 Xiao, W., Windley, B. F., Han, C., Liu, W., Wan, B., Zhang, J. e., et al. (2018). Late Paleozoic to early Triassic multiple roll-back and oroclinal bending of the Mongolia collage in Central Asia. Earth Sci. Rev. 186, 94–128. doi:10.1016/j. earscirev.2017.09.020 Qi, L., Hu, J., and Gregoire, D. C. (2000). Determination of trace elements in granites by inductively coupled plasma mass spectrometry. Talanta 51 (3), 507–513. doi:10.1016/s0039-9140(99)00318-5 Xiao, W., Windley, B. F., Sun, S., Li, J., Huang, B., Han, C., et al. (2015). A tale of amalgamation of three permo-triassic collage systems in central asia: oroclines, sutures, and terminal accretion. Annu. Rev. Earth Planet. Sci. 43 (1), 477–507. doi:10.1146/annurev-earth-060614-105254 Qin, K.-Z., Guo, Z.-l., Shen, M.-D., Tang, D.-M., Zhou, Q.-F., Wang, C.-L., et al. (2013). Typer, intrusive and mineralization ages of pegmatite rare-element deposits in Chinese Altay. Xinjiang Geol. 31, 1–7. [in Chinese, with English summary]. Yang, F.-Q., Zhang, Z.-l., Wang, R., Li, Q., Ding, J.-G., Su, Z.-h., et al. (2018). Geological characteristics and metallogenesis of rare metal deposits in Altay, Xinjiang. Geotecton. et Metallog. 42 (6), 1010–1026. doi:10.16539/j.ddgzyckx. 2018.06.006 [in Chinese, with English summary]. Ren, B.-Q., Zhang, H., Tang, Y., and Lv, Z.-H. (2011). REFERENCES LA-ICPMS U-Pb zircon geochronology of the Altai pegmatites and its geological signiﬁcance. Acta March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 18 Genesis of Askartor Be-Mo Deposit Yong et al. in Qinghe County, Northern Xinjiang. Mineral. Deposits 36 (3), 643–658. doi:10.16111/j.0258-7106.2017.03.007 [in Chinese, with English summary]. in Qinghe County, Northern Xinjiang. Mineral. Deposits 36 (3), 643–658. doi:10.16111/j.0258-7106.2017.03.007 [in Chinese, with English summary]. Yin, R., Wang, R. C., Zhang, A.-C., Hu, H., Zhu, J. C., Rao, C., et al. (2013). Extreme fractionation from zircon to hafnon in the Koktokay No. 1 granitic pegmatite, Altai, Northwestern China. Am. Mineral. 98 (10), 1714–1724. doi:10.2138/am.2013.4494 Zou, T.-R., Chao, H.-Z., and Wu, B.-Q. (1988). Orogenic and anorogenic granitoids of Altay Mountains of Xinjiang and their discrimiantion criteria. Acta Geol. Sin. 2 (1), 45–64. doi:10.1111/j.1755-6724.1989.mp2001005.x Zou, T.-R., Chao, H.-Z., and Wu, B.-Q. (1988). Orogenic and anorogenic granitoids of Altay Mountains of Xinjiang and their discrimiantion criteria. Acta Geol. Sin. 2 (1), 45–64. doi:10.1111/j.1755-6724.1989.mp2001005.x Yuan, C., Sun, M., Xiao, W., Li, X., Chen, H., Lin, S., et al. (2007). Accretionary orogenesis of the Chinese Altai: insights from paleozoic granitoids. Chem. Geol. 242 (1), 22–39. doi:10.1016/j.chemgeo.2007.02.013 Zou, T.-R., and Li, Q.-C. (2006). Rare and rare earth metallic deposits in Xinjiang. Benjing, China: Beijing: Geological Pubilishing House [in Chinese, with English summary]. Zajacz, Z., Halter, W. E., Pettke, T., and Guillong, M. (2008). Determination of ﬂuid/melt partition coefﬁcients by LA-ICPMS analysis of co-existing ﬂuid and silicate melt inclusions: controls on element partitioning. Geochim. et Cosmochim. Acta 72 (8), 2169–2197. doi:10.1016/j.gca.2008.01.034 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Zhang, A., Wang, R., Li, Y., Hu, H., Lu, X., Ji, J., et al. (2008). Tourmalines from the Koktokay No.3 pegmatite, Altai, NW China: spectroscopic characterization and relationships with the pegmatite evolution. EJM 20 (1), 143–154. doi:10.1127/ 0935-1221/2008/0020-1779 Copyright © 2021 Yong, Hui and Zheng-Hang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Earth Science \| www.frontiersin.org REFERENCES Zhang, H., Lv, Z.-H., and Tang, Y. (2019). Metallogeny and prospecting model as well as prospecting direction of pegmatite-type rare metal ore deposits in Altay orogenic belt, Xinjian. Mineral. Deposits 38 (4), 792–814. doi:10.16111/j.0258- 7106.2019.04.008 [in Chinese, with English summary]. Zhang, Y.-F., Lin, X.-W., Zhao, Y.-M., Guo, Q.-M., and Zhao, D.-C. (2017). Geochronology and geochmeistry of granitoids of Ascalt beryllium deposit March 2021 \| Volume 9 \| Article 646772 Frontiers in Earth Science \| www.frontiersin.org 19
https://openalex.org/W4312164537	https://digital.csic.es/bitstream/10261/304133/1/roboticelasto.pdf	English	null	Robotic Pick‐and‐Place Operations in Multifunctional Liquid Crystal Elastomers	Advanced intelligent systems	2,022	cc-by	6,753	Pengrong Lyu, Mert Orhan Astam, Carlos Sánchez-Somolinos, and Danqing Liu* Pengrong Lyu, Mert Orhan Astam, Carlos Sánchez-Somolinos, and Danqing Liu* tailored for delicate applications, such as surgery or prosthetics.[2] Within these robotic applications, pick-and-place opera- tions to precisely transport objects are common and essential functions. Typical robotic arm designs are inspired by the human arm, consisting of a hard metal frame with multiple protruding “ﬁngers” and joints.[3–5] Recently, soft materials, such as silicone, have been applied in robotic arms to enhance their adaptability to various environments.[6–8] Typically, these systems are operated with controlled air pressure and are easy to fabricate[9–11] but require external valves, tubes, and bulky motors. Meanwhile, robotic motions such as contracting,[12,13] bending,[14–16] and rolling[17] functions have been demon- strated by stimuli–responsive materials. However, unlike natural systems, state- of-the-art stimuli–responsive material actuators are typically designed for a single function, which renders them unable to perform multifactored tasks. Pick-and-place operations for transporting objects precisely to a target position are a prominent function of (soft-) robotic systems. Therefore, there is great interest in industry to improve the characteristic gripping, holding, and releasing methods involved in pick-and-place operations. Within living organisms such as octopi, nature demonstrates that multiple types of conjointly working actuators are required for ﬂexible pick-and-place operations. Herein, a multifunctional soft robotic arm is developed, capable of transporting an object within 3D space. The soft robotic arm consists of two structural actuators (rotating base and lifting unit) and a suction cup-based gripper. The structural actuator acts as both the load bearing and actuating components of the robotic system. Yet, the gripper is the crucial innovation within the robotic arm. A cephalopod-limb-inspired gripper functioning through the reversible ﬂat-to-conical deformation of azimuthally aligned liquid crystal elastomer (LCE) ﬁlms is proposed. The pressure-generating actuation mechanism of the gripper means that no external device is needed to operate the gripping function. Akin to natural systems, the in-tandem operation of the actuators in the soft robotic arm allows for multifactored tasks. Yet, the design achieves this through the use of a single material, which is not innate in natural archetypes. Therefore, in this work, we developed and designed a soft robotic arm capable of multifactored pick-and-place operations, incorporating the function of the stimuli–responsive material rotating base, lifting unit, and suction cup-based gripper (Figure 1a). The soft robotic arm can operate within 3D space, without the need for integration with any external motor. 1. Introduction The ORCID identiﬁcation number(s) for the author(s) of this article can be found under https://doi.org/10.1002/aisy.202200280. © 2022 The Authors. Advanced Intelligent Systems published by Wiley- VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Pengrong Lyu, Mert Orhan Astam, Carlos Sánchez-Somolinos, and Danqing Liu* Each actuator in the soft robotic arm can be individually controlled via electrical signals, enabling RESEARCH ARTICLE RESEARCH ARTICLE www.advintellsyst.com 1. Introduction (Micro-) robots are globally ubiquitous.[1] Capable of working under extreme, human-adverse conditions, robots have extended the potential of industry to unprecedented heights. Robots can be used to perform heavy or repetitive tasks, such as shifting com- ponents between production lines. Yet, robots can also be P. Lyu, M. O. Astam, D. Liu Institute for Complex Molecular Systems Eindhoven University of Technology Den Dolech 2, 5612 AZ Eindhoven, The Netherlands E-mail: d.liu1@tue.nl P. Lyu, M. O. Astam, D. Liu Department of Chemical Engineering and Chemistry Eindhoven University of Technology Den Dolech 2, 5612 AZ Eindhoven, The Netherlands The ORCID identiﬁcation number(s) for the author(s) of this article can be found under https://doi.org/10.1002/aisy.202200280. © 2022 The Authors. Advanced Intelligent Systems published by Wiley- VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1002/aisy.202200280 C. Sánchez-Somolinos Instituto de Nanociencia y Materiales de Aragón (INMA) Departamento de Física de la Materia Condensada CSIC-Universidad de Zaragoza Zaragoza 50009, Spain C. Sánchez-Somolinos CIBER in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN) Spain D. Liu SCNU-TUE Joint Lab of Device Integrated Responsive Materials (DIRM) National Center for International Research on Green Optoelectronics South China Normal University Guangzhou Higher Education Mega Center, No 378, West Waihuan Road, 510006 Guangzhou, China Adv. Intell. Syst. 2022, 4, 2200280 2200280 (1 of 8) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licen P. Lyu, M. O. Astam, D. Liu Department of Chemical Engineering and Chemistry Eindhoven University of Technology Den Dolech 2, 5612 AZ Eindhoven, The Netherlands C. Sánchez-Somolinos CIBER in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN) Spain D. Liu SCNU-TUE Joint Lab of Device Integrated Responsive Materials (DIRM) National Center for International Research on Green Optoelectronics South China Normal University Guangzhou Higher Education Mega Center, No 378, West Waihuan Road, 510006 Guangzhou, China D. Liu SCNU-TUE Joint Lab of Device Integrated Responsive Materials (DIRM) National Center for International Research on Green Optoelectronics South China Normal University Guangzhou Higher Education Mega Center, No 378, West Waihuan Road, 510006 Guangzhou, China The ORCID identiﬁcation number(s) for the author(s) of this article can be found under https://doi.org/10.1002/aisy.202200280. DOI: 10.1002/aisy.202200280 © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH Adv. Intell. Syst. 2022, 4, 2200280 2200280 (1 of 8) 2200280 (1 of 8) Adv. Intell. Syst. 2022, 4, 2200280 © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH 26404567, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aisy.202200280 by Csic Organización Central Om (Oficialia Mayor) (Urici), www.advintellsyst.com Figure 1. Principle of the multifunctional soft robotic arm. a) The schematic of the soft robotic arm and its components. b) Equivalent electrical circuit of the electrical-driving set-p within the soft robotic arm. R1 and R2 correspond to the resistance of the nichrome heating wire embedded in the rotating base and lifting unit respectively. c) The driving electric signal inputs for controlling the different elements of the soft robotic arm: signal 1 for actuating the rotating base, signal 2 for actuating the lifting unit, and signal 3 for actuating the gripper. Figure 1. Principle of the multifunctional soft robotic arm. a) The schematic of the soft robotic arm and its components. b) Equivalent electrical circuit of the electrical-driving set-p within the soft robotic arm. R1 and R2 correspond to the resistance of the nichrome heating wire embedded in the rotating base and lifting unit respectively. c) The driving electric signal inputs for controlling the different elements of the soft robotic arm: signal 1 for actuating the rotating base, signal 2 for actuating the lifting unit, and signal 3 for actuating the gripper. degrees of freedom, namely, in and out of the plane. The rotating base is responsible for the motion in X- and Y-axes. It consists of a nichrome heating element embedded in an LCE pillar with helical alignment, established by rolling a uniaxially aligned LCE ﬁlm around the nichrome wire and subsequently twisting it by 720°. The actuation of the LCE rotating base can be activated by triggering the Joule effect with current, which is enhanced by the positioning of the heating element within the LCE. The prin- ciple of the actuation lies in the anisotropic thermal expansion of the helically aligned LC mesogens. The angle of rotation (Δθ) can be deﬁned as a function of current (I1). On the other hand, the uniaxially aligned LCE lifting unit enables motion in the Z-axis and is operated in the same manner as the rotating base. The maximum lifting height (ΔZ) can also be deﬁned as a function of a separate current (I2). DOI: 10.1002/aisy.202200280 Nichrome wire, as the passive compo- nent, hinders deformation of LCEs. However, it further reinfor- ces the rigidity of the LCE and thus makes the system more robust. Most innovatively, we designed a suction cup-based grip- per, whose function is based on the reversible ﬂat-to-conical deformation of a photothermal-responsive LCE ﬁlm with azi- muthal alignment.[27] The underlying principle is the reversible vacuum generation caused by LCE ﬁlm deformation. Irradiating the gripper with light triggers LCE deformation and negative pressure generation that attaches the target object to the LCE suc- tion cup. As shown in Figure 1b, the two electrical inputs to the soft robotic arm can be visualized as parallel branches within a the capacity to program its movement and operate it repetitively. The actuation behavior of each component is pre-designed by controlling molecular alignment architecture, which is a funda- mental concept in the ﬁeld of liquid crystals. Liquid crystal elastomers (LCEs) consist of polymerized liquid crystals, networked by bonds to chain-extending molecules. These materials have gained recognition as stimuli–responsive actuators, leading to their application in artiﬁcial muscles,[18–22] artiﬁcial walkers,[15,23,24] responsive topography,[25,26] etc. Inspired by prior art, we constructed a photothermal-responsive LCE suction cup. Upon deformation of the azimuthally aligned (Figure 1a) LCE from its ﬂat to conical state, negative pressure is generated, adhering the target object to the LCE cup. Meanwhile, to transport the object in 3D space, a rotating base with helical alignment is responsible for the motion in X- and Y-axes. On the other hand, motion in the Z-axis is achieved through an addi- tional lifting unit with uniaxial alignment. Furthermore, as our designs do not require integration of external devices, it ena- bles the construction of a standalone, compact, single-material soft robotic arm. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH www.advintellsyst.com www.advintellsyst.com 1:0.91(Figure 2d). Commercial visible light dye with broad absorption in the visible light region (Figure S1, Supporting Information), Disperse Red 1 (Molecule 3), was selected as the photothermal conversion medium to endow the elastomer with light-responsive properties. Molecule 4 was used to initiate the photopolymerization after printing. The liquid crystal mix- ture exhibits smectic-to-nematic and nematic-to-isotropic phase transition temperatures at 40 and 90 °C, respectively (Figure S2, Supporting Information). To establish a ﬂat-to-coni- cal state transition, we utilize the azimuthal alignment, as schematically illustrated in Figure 2e (inset). To achieve this alignment, concentric annular extrusion paths were designed. After dissolving the polyvinyl alcohol (PVA) sacriﬁcial layer, a photothermal-responsive suction cup was obtained (as shown in Figure 2b) and then assembled with the LCE lifting unit to complete the soft robotic system. The crossed polarizer micro- scope (POM) was used to check the alignment of LC mesogens. As shown in Figure 2e, under POM, the photothermal- responsive suction cup exhibits the expected fan-shaped pattern. The regions where the molecules are parallel to the crossed polar- izers appear black. After rotating the crossed polarizers by 45°, the fan-shaped pattern is also rotated 45°. This conﬁrms the circuit. This allows the actuation amplitude of each element in the soft robotic arm to be deﬁned in terms of the magnitude and activation time of the electric input signal. The trajectory of the gripper can be mapped to the actuation of the rotating base and lifting unit by inverse kinematics. As shown in Figure 1c, the motion of the soft robotic arm can be preprogrammed in terms of two independent input electric signals and an incident light intensity, enabling the system to be run a repetitive motion. The gripper is the most crucial innovation of our soft robotic arm. Inspired by the branchial sucker of an octopus, as shown in Figure 2a, it was developed on the principle of suction-based adhesion. We manufactured the suction cup gripper (Figure 2b) by the direct-ink-writing printing procedure. As it involves responsive material, this process is referred to as 4D printing,[27–30] an important aspect of additive manufacturing. Figure 2c shows the schematic representation of the printing process. During extrusion and deposition, shear forces and elon- gational ﬂow inﬂuence the LC polymer chain alignment, estab- lishing a molecular order parallel to the direction of the shear ﬂow. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH 2. Results and Discussion As shown in Figure 1a, our multifunctional soft robotic arm con- sists of three different types of actuators. They provide two 2200280 (2 of 8) Adv. Intell. Syst. 2022, 4, 2200280 www.advancedsciencenews.com www.advancedsciencenews.com www.advintellsyst.com www.advancedsciencenews.com center and along the radial direction. The former is the desired conﬁguration for our soft robotic arm. azimuthal orientation. We further investigated the surface roughness of the printed LCE suction cup. 3D interferometer results suggest that the initial surface is irregular, as shown in Figure 2f, where small surface reliefs of 10 s of nanometers are observed. Furthermore, the data exhibits a double wedge- shaped proﬁle along the diameter direction. Upon heating, this azimuthally aligned LCE wedge can deform into a cone due to the anisotropic deformation, as shown in Figure 3a. There is no pre- ferred direction in the ﬁlm with a uniform cross section due to the symmetrical expansion of LC mesogens in the z-axis (Figure S3, Supporting Information). azimuthal orientation. We further investigated the surface roughness of the printed LCE suction cup. 3D interferometer results suggest that the initial surface is irregular, as shown in Figure 2f, where small surface reliefs of 10 s of nanometers are observed. Furthermore, the data exhibits a double wedge- shaped proﬁle along the diameter direction. Upon heating, this azimuthally aligned LCE wedge can deform into a cone due to the anisotropic deformation, as shown in Figure 3a. There is no pre- ferred direction in the ﬁlm with a uniform cross section due to the symmetrical expansion of LC mesogens in the z-axis (Figure S3, Supporting Information). Upon illumination by an light-emitting diode (LED) lamp (455 nm), the initial ﬂat ﬁlm deformed into a conical state within 7 s, as shown in Figure 3d and Movie S1, Supporting Information. This is due to the anisotropic expansion of LC mesogens in reaction to the photothermal effect of the visible light dye (Molecule 3). When the LED lamp is switched off, the ﬁlm relaxes back into its ﬂat state within 6 s. These delays between switching the LED lamp on/off and the complete ﬁlm actuation have a direct relation with the heating and cooling rates of the ﬁlm respectively. When the cooling or heating rate increases, the delay decreases. For instance, using a stronger LED light would allow the ﬁlm to heat up faster and reduce the actuation lag. The viscoelastic nature of the LCE also contrib- utes to activation–actuation mismatch. Beyond kinetics, we fur- ther characterized the deformation process by tracking the In order to control this deformation direction, we have engi- neered the LCE ﬁlm cross section. www.advintellsyst.com Before printing, the acrylate-terminated main-chain LCE was synthesized through a Michael addition reaction,[31] using Molecule 1 and Molecule 2 with a molar ratio of Figure 2. LCE photothermal-responsive suction cup as a gripper. a) Annotated illustrations of an octopus and its branchial limb sucker. Patmrepresents the atmospheric pressure. b) Photograph of the photothermal-responsive suction cup-based gripper. c) Schematic representation of the printing process. d) Structural formulae of the chemicals used to synthesize the utilized liquid crystal oligomer. e) The initial LCE ﬁlm observed by optical microscopy between crossed polarizers. The inset schematic depicts the alignment of LC mesogens. f) 3D interferometer image of the photothermal-responsive suction cup surface topography and its corresponding 2D proﬁle along the diameter direction. (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Figure 2. LCE photothermal-responsive suction cup as a gripper. a) Annotated illustrations of an octopus and its branchial limb sucker. Patmrepresents the atmospheric pressure. b) Photograph of the photothermal-responsive suction cup-based gripper. c) Schematic representation of the printing process. d) Structural formulae of the chemicals used to synthesize the utilized liquid crystal oligomer. e) The initial LCE ﬁlm observed by optical microscopy between crossed polarizers. The inset schematic depicts the alignment of LC mesogens. f) 3D interferometer image of the photothermal-responsive suction cup surface topography and its corresponding 2D proﬁle along the diameter direction. Figure 2. LCE photothermal-responsive suction cup as a gripper. a) Annotated illustrations of an octopus and its branchial limb sucker. Patmrepresents the atmospheric pressure. b) Photograph of the photothermal-responsive suction cup-based gripper. c) Schematic representation of the printing process. d) Structural formulae of the chemicals used to synthesize the utilized liquid crystal oligomer. e) The initial LCE ﬁlm observed by optical microscopy between crossed polarizers. The inset schematic depicts the alignment of LC mesogens. f) 3D interferometer image of the photothermal-responsive suction cup surface topography and its corresponding 2D proﬁle along the diameter direction. 2200280 (3 of 8) 2200280 (3 of 8) Adv. Intell. Syst. 2022, 4, 2200280 www.advancedsciencenews.com www.advancedsciencenews.com As shown in Figure 3b, the ﬁlm consistently bends upward due to the linearly decreasing ﬁlm thickness toward the center and along the radial direction. In contrast, as shown in Figure 3c, the ﬁlm consistently bends downward with a linearly increasing ﬁlm thickness toward the Figure 3. Controlling suction cup-based gripper deformation direction by manipulating ﬁlm cross section. a) The schematic of actuation behavior in the ﬁlm with azimuthal alignment. The black arrows show the direction of the LC mesogens contraction. b) The LCE ﬁlm with a linearly decreasing ﬁlm thickness toward the center, along the radial direction, leads to upwards deformation. The color bar shows the normalized height. c) The LCE ﬁlm with a linearly increasing ﬁlm thickness toward the center, along the radial direction, leads to downward deformation. d) Photothermally triggered actuation behavior of an azimuthal elastomer ﬁlm upon irradiation by blue light. e) Graphical representation of bending angle and its corresponding temperature against applied light intensity. Oficialia Mayor) (Urici), Wiley Online Library on [23/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA article Figure 3. Controlling suction cup-based gripper deformation direction by manipulating ﬁlm cross section. a) The schematic of actuation behavior in the ﬁlm with azimuthal alignment. The black arrows show the direction of the LC mesogens contraction. b) The LCE ﬁlm with a linearly decreasing ﬁlm thickness toward the center, along the radial direction, leads to upwards deformation. The color bar shows the normalized height. c) The LCE ﬁlm with a linearly increasing ﬁlm thickness toward the center, along the radial direction, leads to downward deformation. d) Photothermally triggered actuation behavior of an azimuthal elastomer ﬁlm upon irradiation by blue light. e) Graphical representation of bending angle and its corresponding temperature against applied light intensity. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH 2200280 (4 of 8) Adv. Intell. Syst. 2022, 4, 2200280 www.advancedsciencenews.com www.advintellsyst.com was illuminated with blue light (445 nm) to deform it to its coni- cal state. This formed an enclosed space (Figure 4a(ii)) between the object and the suction cup. Third, illumination by blue light was sustained, increasing LCE deformation and subsequently increasing the bending angle ðθ2Þ and the volume ðV2Þ of the enclosed space. This generated an increasing negative pressure (P2) inside the suction cup. www.advancedsciencenews.com Meanwhile, the atmospheric pres- sure enhances the attachment of the ﬁlm to the object, which causes sufﬁciently high friction to resist slippage. Consequently, as shown in Figure 4a(iii), after applying an upward pulling force, the object is lifted. In the last stage (Figure 4a(iv)), after removing the blue light, the bent membrane spontaneously returns to the initial ﬂat state, reducing the position of the printed LCE ﬁlm when fully actuated under dif- ferent light intensities. As shown in Figure 3e, increasing the light intensity increases the maximum bending angle θ. Below 40 °C, some deviation between the temperature curve and defor- mation curve occurred due to the layered structure formed in the smectic phase, which hampers the expansion rate in the direc- tion perpendicular to the director. After successfully controlling the direction of deformation, we explored the capacity of the actuator as a responsive gripper for pick-and-place operations. As shown in Figure 4a(i–iv), this oper- ation consists of four stages. First, the actuator was placed on the plastic round-topped cylinder (diameter ¼ 6 mm, 335 mg) to ensure complete contact at the interface. Second, the sample 2/aisy.202200280 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [23/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditio Figure 4. Suction-cup-based gripper for gripping and holding the object. a) The schematics of the gripper engaging an object with surface curvature: initial state i), forming an enclosed space between the suction cup and object ii), generating negative pressure inside the suction cup to grip the object iii), relaxing deformation, and releasing the object iv). The θ, V, and P parameters correspond to the actuated angle, the volume of closed space in the interface, and the air pressure inside the suction cup respectively. b) The pick-and-place procedure of the gripper, which corresponds to the procedure in Figure 4a(i–iv). c) The gripper engaging an object with a ﬂat surface, with insets schematics showing the different states of the LCE gripper. d) The schematic of the home-built setup for measuring attachment force and negative pressure. e) Graphical representation of the attachment force and the pressure difference inside the suction cup with respect to elapsed time while peeling the LCE ﬁlm off from the glass substrate. l Om (Oficialia Mayor) (Urici), Wiley Online Library on [23/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA Figure 4. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH www.advintellsyst.com www.advintellsyst.com suction cup was placed on the substrate with pressure and ten- sion sensors. The photothermal-responsive suction cup was also ﬁxed to a stepper motor, which provided force toward pulling it off the substrate. After irradiating the sample with blue light for 10 s to induce deformation, the photothermal-responsive suction cup was pulled upward at 1 mm s1, until it was free from the substrate. The pressure and tension sensors were used to moni- tor the evolution of pressure inside the suction cup and the pull- ing force in situ. As shown in Figure 4e, during the detachment process, the deformed suction cup can generate up to 3.5 kPa negative pressure. The corresponding attachment force from atmospheric pressure is 39.7 mN. This was calculated consider- ing that the diameter of the actuated suction cup is 3.8 mm. Moreover, the tension sensor measures about 130 mN of force at the detachment event, which suggests that the adhesion forces between the LCE element and the object also contribute to the attachment. negative pressure inside the LCE cone and releasing the object from the suction cup. The actual pick-and-place procedures of the gripper are shown in Figure 4b, which correspond to the proce- dure in Figure 4a. To demonstrate its function as a gripper, we apply the LCE suction cup to grip a spherically topped cylinder with smaller curvature of diameter¼5 mm (Movie S2, Supporting Information) as well as extremely large curvature (that is ﬂat). The result of grasping the ﬂat substrate is shown on Figure 4c. As this version of the gripper adhering to the ﬂat object contains no azobenzene dye (Molecule 3), the system was thermally activated by heating alone the adhered glass slide to 90 °C on a hotplate prior to contact (Figure 4c(i)). Its grasping process is similar to that achieved through the photothermal suc- tion cup. The solely thermal LCE gripper deformed into its conic state, generating negative pressure inside after contact with the hot glass (Figure 4c(ii)). Upon applying an upward pulling force, the ﬂat glass is lifted successfully. This also causes the gripper to cool rapidly as it is moved away from the hotplate, reducing actu- ation and subsequently the negative pressure inside. Consequently, the adhered ﬂat glass was released after 8 s. Yet, the suction cup gripper on its own can only achieve a sin- gular function. www.advancedsciencenews.com Suction-cup-based gripper for gripping and holding the object. a) The schematics of the gripper engaging an object with surface curvature: initial state i), forming an enclosed space between the suction cup and object ii), generating negative pressure inside the suction cup to grip the object iii), relaxing deformation, and releasing the object iv). The θ, V, and P parameters correspond to the actuated angle, the volume of closed space in the interface, and the air pressure inside the suction cup respectively. b) The pick-and-place procedure of the gripper, which corresponds to the procedure in Figure 4a(i–iv). c) The gripper engaging an object with a ﬂat surface, with insets schematics showing the different states of the LCE gripper. d) The schematic of the home-built setup for measuring attachment force and negative pressure. e) Graphical representation of the attachment force and the pressure difference inside the suction cup with respect to elapsed time while peeling the LCE ﬁlm off from the glass substrate. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH Adv. Intell. Syst. 2022, 4, 2200280 www.advancedsciencenews.com www.advintellsyst.com www.advintellsyst.com Information. As shown in Figure 5c, the ﬁnal rotation angle is controllable by the input current. The direction of rotation can be selected by changing the twist direction during sample prepara- tion. On the other hand, the lifting unit was fabricated in a simi- lar process to the rotating base, yet the diameter of the embedded nichrome wire was reduced to 50 μm to produce a more compli- ant heating element. Moreover, the polydomain precured elasto- mer was stretched instead of twisted. This induces a uniaxial alignment, which is then ﬁxed through photopolymerization. Details on the preparation process and actuation behavior can be viewed in Figures S6 and S7, Supporting Information. Upon current input, actuation is triggered and the lifting unit begins to contract. As shown in Figure 5d, the lifting unit con- tracted by 52% of its initial length with 0.1 A current input, yet this maximum lifting height can be adjusted by altering the input current. By ﬁxing the length of the lifting unit while heating, we also investigated its actuation tension with different input cur- rents. This experiment was a modiﬁed version of the experiment measuring the attachment force of the suction cup. As shown in Figure 5e, the tension can be selected by changing the input cur- rent, up to a maximum tension of 350 mN. However, this force is not equivalent to the maximum load capacity of the soft robotic arm, as the attachment force of the suction cup and the ﬂexural stiffness of the rotating base can act as limiting factors. (1,4-Bis-[4-(6-acryloyloxyhexyloxy) benzoyloxy]-2-methylbenzene) was purchased from Merck. Molecule 2 (3,6-Dioxa-1,8-octanedithiol), Molecule 3 (N-Ethyl-N-(2-hydroxyethyl)-4-(4-nitrophenylazo) aniline), and basic catalyst (1,8-diazabicyclo[5.4.0]undec-7-ene, DBU) were obtained from Sigma-Aldrich. Molecule 4 (Irgacure 819) was purchased from Ciba. Polyvinyl alcohol with an average molecular weight of 25 000 g mol1 was obtained from Sigma-Aldrich. Dichloromethane was obtained from Biosolve. For fabricating rotating base and lifting unit, pentaerythritol tetrakis(3-mercaptopropionate) (PETMP) and dipropylamine (DPA) were obtained from Sigma–Aldrich. g Synthesis and Processing of the Photothermal-Responsive suction cup: A direct ink writing procedure was used to fabricate an LCE ﬁlm with azi- muthal alignment. First, we synthesized the oligomer via a base-catalyzed thiol-acrylate Michael addition reaction. Molecule 1 (4036.62 mg), Molecule 3 (24.5 mg), and Molecule 4 (49 mg) were added into dichloro- methane (3 mL) and the mixture was then heated and stirred at 50 °C to dissolve all solid material. www.advintellsyst.com Then, Molecule 2 (875.04 mg) and based cata- lyst DBU (20 μL) were added to the solution. After stirring for 3 h on a hot plate at 40 °C, the mixture was poured into a poly(tetraﬂuoroethylene) (PTFE) evaporation dish and then placed in a vacuum oven at 40 °C for 3 h to evaporate the DCM. Second, the oligomer was ﬁlled into an ink syringe and then put into a vacuum oven at 80 °C for 20 min to remove any air bubbles. After that, the ink syringe was loaded into the commercial 3D printer (EHR, Hyrel 3D). The oligomer was extruded into a substrate coated with a PVA layer using a nozzle (diameter ¼ 335 μm) at 68 °C. The printing speed was set as 300 mm min1 and the printing path was preprogrammed by G-code generated by the printer software. After printing, an Exfo Omnicure S2000 light source was used to photopolymerize the aligned oligomer at room temperature for 2 h, while the sample was ﬂipped every 30 min. After photopolymerization, the sample was immersed in water at room temperature for 20 min to dissolve the PVA layer. Finally, after further drying the sample at room temperature, a free-standing ﬁlm with azimuthal alignment was obtained. Next, considering the speciﬁc task of the soft robotic arm is to displace an object from the 0° to 90° positions, the operation of the gripper would involve three consecutive stages, as shown in Figure 5ab. Current with different magnitudes can be directed through each element independently, allowing for each element to be activated individually or simultaneously as desired. In the ﬁrst stage, 0.07 A of current was run through the embedded nichrome wire in the lifting unit, activating material contraction and lifting the held object. In the second stage, 0.55 A of current was applied to actuate rotating base. In the last stage, the circuit running through the rotating base and lifting unit was switched off. This triggers the soft robotic arm to move back to its initial position, ready for the next task. This entire process is recorded in Movie S4, Supporting Information. Synthesis and Processing of the Rotating Base: The rotating base was manufactured through a two-step crosslinking procedure, following previ- ously reported work[32] with little modiﬁcation. The procedure of preparing the rotating base is shown in Figure S4, Supporting Information. www.advintellsyst.com First, reactive liquid crystal mesogens RM 257 (1200 mg) and photoinitiator Irgacure 819 (12 mg) were added into dichloromethane (1000 μL), and the mixture was heated and stirred at 50 °C to dissolve all the solids. Then, the solution was placed in a vacuum oven at 70 °C for 3 h to evapo- rate all the DCM and avoid monomer crystallization. Next, PETMP (91.67 mg), Molecule 2 (273.6 mg) and DPA (3 μL) were immediately added into the solution. After stirring, the solution was poured into the rectangular mold (thickness ¼ 500 μm) and left to dry at room tempera- ture for 24 h. Second, nichrome wire (diameter ¼ 200 μm) was placed on the edge of the precrosslinked ﬁlm, and then the ﬁlm was rolled up to form a pillar with a diameter of about 2.5 mm. Finally, after the pillar was twisted 720°, an Exfo Omnicure S2000 light source was used to photopolymerize the aligned oligomer at room temperature for 3 h, while the sample was ﬂipped every 30 min. Thereby, an electric responsive actuator for rotating was obtained. In conclusion, we demonstrate the pick-and-place operation capacity of our soft robotic arm. Our LCE photothermal- responsive suction cup gripper is the overarching factor in its function, eliminating the need for any external assistive devices. The LCE photothermal-responsive suction cup was recorded lift- ing objects 100 times its own weight, yet can theoretically lift objects weighing 40 times more. The attachment force is compa- rable to a live octopus’s sucker under water.[33] This project high- lights the potential of constructing complex soft robots by integrating different types of actuators into one device. Yet, this multifunctional soft robotic arm is constructed from a single material, advertising the prospect of single-material devices. We anticipate that this approach can further improve device mobility, advancing the soft robotics ﬁeld toward building devices that are applicable in industrial and consumer products and uses. Synthesis and Processing of the Lifting Unit: The lifting unit was fabricated in a similar procedure to the rotating base. The procedure of preparing the lifting unit is shown in Figure S6, Supporting Information. Differences include thinner nichrome wire (diameter ¼ 50 μm) being embedded in the precured oligomer to get a more ﬂexible actuator. Then, the precured oligomer was stretched to get uniaxial alignment. Finally, photopolymeri- zation was used to ﬁx the uniaxial alignment. www.advintellsyst.com Characterization Methods: Differential scanning calorimetry (DSC) measurements were employed to determine the transition temperature of oligomers in TA Instruments DSC Q1000. The POM images of the suc- tion cup were taken from a Leica DM 2700M equipped with two polarizers. The thickness and surface topography was determined via the 3D Optical Proﬁler (Sensofar Metrology). The photothermal-responsive suction cup was actuated by collimated visible light (455 nm, Thorlabs M455L2), in which a controller (ThorLabs DC4104) was used to tune the intensity © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH www.advintellsyst.com Therefore, it was attached to a soft robotic arm consisting of multiple LCE actuators to achieve multifactored pick-and-place functions. The schematic and the sample of the assembled robotic arm are shown in Figure 5a,b. In our soft robotic arm, the rotating base was synthesized through the two-step crosslinking process.[32] The ﬁnal actuation behavior can be viewed in Figure S5 and Movie S3, Supporting To further illustrate the potential of the photothermal- responsive suction cup as a gripper, we customized a pressure- sensitive device to quantify the attachment forces in the interface between the LC ﬁlm and object. The schematic of the custom device is shown in Figure 4d. The photothermal-responsive Figure 5. Soft robotic arm for transport and pick-and-place operations in 3D space. a) The schematic of the work procedure of electric signal-driven soft robotic arm: i) initial state, ii) gripping and lifting the object, and iii) ﬁnally transporting and releasing the object. The insets show the state of the gripper. b) Photographs of the work procedure of the soft robotic arm, which is corresponding to the schematic in Figure 5a(i–iii). The insets show the cor- responding top view in the different states. c) The actuation behavior of the rotating base. d) The actuation behavior of the lifting unit. e) The load capacity of the lifting unit in different applied currents. (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Figure 5. Soft robotic arm for transport and pick-and-place operations in 3D space. a) The schematic of the work procedure of electric signal-driven soft robotic arm: i) initial state, ii) gripping and lifting the object, and iii) ﬁnally transporting and releasing the object. The insets show the state of the gripper. b) Photographs of the work procedure of the soft robotic arm, which is corresponding to the schematic in Figure 5a(i–iii). The insets show the cor- responding top view in the different states. c) The actuation behavior of the rotating base. d) The actuation behavior of the lifting unit. e) The load capacity of the lifting unit in different applied currents. © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH 2200280 (6 of 8) Adv. Intell. Syst. 2022, 4, 2200280 www.advancedsciencenews.com www.advintellsyst.com www.advintellsyst.com www.advancedsciencenews.com [5] N. Vasios, A. J. Gross, S. Soifer, J. T. B. Overvelde, K. Bertoldi, Soft Robot. 2020, 7, 1. of the lights. The rotating base and lifting unit were actuated by a DC power source (TENMA 72-2720). [6] F. Ilievski, A. D. Mazzeo, R. F. Shepherd, X. Chen, G. M. Whitesides, Angew. Chem. Int. Ed. 2011, 50, 1890. Data Availability Statement [22] Z. Cheng, T. Wang, X. Li, Y. Zhang, H. Yu, ACS Appl. Mater. Interfaces 2015, 7, 27494. The data that support the ﬁndings of this study are available from the corresponding author upon reasonable request. [23] X. Lu, H. Zhang, G. Fei, B. Yu, X. Tong, H. Xia, Y. Zhao, Adv. Mater. 2018, 30, 1706597. [24] B. Zuo, M. Wang, B.-P. Lin, H. Yang, Nat. Commun. 2019, 10, 4539. [25] D. Liu, C. W. M. M. Bastiaansen, J. M. J. J. Den Toonder, D. J. Broer, Macromolecules 2012, 45, 8005. Supporting Information [7] R. Deimel, O. Brock, Int. J. Rob. Res. 2016, 35, 161. [8] H. Jiang, Z. Wang, Y. Jin, X. Chen, P. Li, Y. Gan, S. Lin, X. Chen, Int. J. Rob. Res. 2021, 40, 411. Supporting Information is available from the Wiley Online Library or from the author. [9] G. Oliveri, L. C. van Laake, C. Carissimo, C. Miette, J. T. B. Overvelde, Proc. Natl. Acad. Sci. 2021, 118, 2017015118. [10] S. Li, D. M. Vogt, D. Rus, R. J. Wood, Proc. Natl. Acad. Sci. USA 2017, 114, 13132. Conﬂict of Interest [19] J. Sun, Y. Wang, W. Liao, Z. Yang, Small 2021, 17, 1. The authors declare no conﬂict of interest. [20] M. Barnes, R. Verduzco, Soft Matter 2019, 15, 870. [21] C. Zhu, Y. Lu, L. Jiang, Y. Yu, Adv. Funct. Mater. 2021, 31, 1. Acknowledgements [11] B. Gorissen, E. Milana, A. Baeyens, E. Broeders, J. Christiaens, K. Collin, D. Reynaerts, M. De Volder, Adv. Mater. 2019, 31, 1804598. The authors thank Professor Dirk J. Broer for his support and guidance for this project. The authors also thank Dr. Thierry K. Slot for helping build the electronic setup used to measure negative pressures within the liquid crys- tal elastomer suction cup. The authors also thank ﬁnancial support from CRT Starterspakket from ICMS, Eindhoven University of Technology, the Netherlands Organization for Scientiﬁc Research (NWO OCENW.KLEIN. 10854, START-UP 8872), the European Union’s Horizon 2020 Research and Innovation Programmer under the Marie Sklodowska-Curie grant agreement no. 956150 (STORM-BOTS). C.S.S. thanks the Spanish “Ministerio de Ciencia, Innovación y Universidades (MCIU)” through AEI/FEDER(UE) project PID2020-118485RB-I00. [12] H. Wermter, H. Finkelmann, E-polymers 2001, 1, 111. [13] B. Jin, J. Liu, Y. Shi, G. Chen, Q. Zhao, S. Yang, Adv. Mater. 2022, 34, 2107855. [14] Y. Yu, M. Nakano, T. Ikeda, Nature 2003, 425, 145. [15] S. Ma, X. Li, S. Huang, J. Hu, H. Yu, Angew. Chem. Int. Ed. 2019, 58, 2655. [16] Y. Zhang, Z. Wang, Y. Yang, Q. Chen, X. Qian, Y. Wu, H. Liang, Y. Xu, Y. Wei, Y. Ji, Sci. Adv. 2020, 6, eaay8606. [16] Y. Zhang, Z. Wang, Y. Yang, Q. Chen, X. Qian, Y Y. Wei, Y. Ji, Sci. Adv. 2020, 6, eaay8606. [17] X. Lu, S. Guo, X. Tong, H. Xia, Y. Zhao, Adv. Mater. 2017, 29. [18] H. Zeng, O. M. Wani, P. Wasylczyk, R. Kaczmarek, A. Priimagi, Adv. Mater. 2017, 29, 1701814. [1] M. A. K. Bahrin, M. F. Othman, N. H. N. Azli, M. F. Talib, J. Teknol. 2016, 78, 6. [2] B. Preising, T. C. Hsia, B. Mittelstadt, IEEE Eng. Med. Biol. Mag. 1991, 10, 13. [3] H. Cruse, Biol. Cybern. 1986, 54, 125. [4] K. Suzumori, S. Iikura, H. Tanaka, IEEE Control Syst. Mag. 1992, 12, 21. [4] K. Suzumori, S. Iikura, H. Tanaka, IEEE Control Syst. Mag. 1992, 12, 21. D. Liu, Adv. Sci. 2004749. electrically driven, liquid crystal elastomers, multifunctional single materials, pick-and-place operations, soft robotic arms [27] M. López-Valdeolivas, D. Liu, D. J. Broer, C. Sánchez-Somolinos, Macromol. Rapid Commun. 2017, 39, 1700710. Received: August 23, 2022 Revised: September 28, 2022 Published online: November 26, 2022 [28] T. H. Ware, Z. P. Perry, C. M. Middleton, S. T. Iacono, T. J. White, ACS Macro Lett. 2015, 4, 942. [29] M. Barnes, S. M. Sajadi, S. Parekh, M. M. Rahman, P. M. Ajayan, R. Verduzco, ACS Appl. Mater. Interfaces 2020, 12, 28692. [29] M. Barnes, S. M. Sajadi, S. Parekh, M. M. Rahman, [30] M. Fang, T. Liu, Y. Xu, B. Jin, N. Zheng, Y. Zhang, Q. Zhao, Z. Jia, T. Xie, Adv. Mater. 2021, 33, 2105597. [31] T. H. Ware, M. E. McConney, J. J. Wie, S. K. Ahn, V. P. Tondiglia, T. J. White, Science. 2015, 347, 982. [32] Y. Wang, J. Sun, W. Liao, Z. Yang, Adv. Mater. 2022, 34, 2107840. [3] H. Cruse, Biol. Cybern. 1986, 54, 125. [33] H. Bagheri, A. Hu, S. Cummings, C. Roy, R. Casleton, A. Wan, N. Erjavic, S. Berman, M. M. Peet, D. M. Aukes, X. He, S. C. Pratt, R. E. Fisher, H. Marvi, Adv. Intell. Syst. 2020, 2, 1900154. 2200280 (8 of 8) 3. Experimental Section Materials: An overview of the materials for the photothermal- responsive suction cup is provided in Figure 2d. Molecule 1 2200280 (7 of 8) Adv. Intell. Syst. 2022, 4, 2200280 © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH [2] B. Preising, T. C. Hsia, B. Mittelstadt, IEEE Eng. Med. Biol. Mag. 1991, 10, 13. [3] H. Cruse, Biol. Cybern. 1986, 54, 125. Keywords [26] P. Lv, Y. You, J. Li, Y. Zhang, D. J. Broer, J. Chen, G. Zhou, W. Zhao, D. Liu, Adv. Sci. 2004749. [ 6] . v, . ou, J. , . a D. Liu, Adv. Sci. 2004749. 2200280 (8 of 8) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH Adv. Intell. Syst. 2022, 4, 2200280
https://openalex.org/W2559468428	http://agritrop.cirad.fr/582371/1/Verhegghen_RS_2016.pdf	English	null	The Potential of Sentinel Satellites for Burnt Area Mapping and Monitoring in the Congo Basin Forests	Remote sensing	2,016	cc-by	15,989	Astrid Verhegghen 1,, Hugh Eva 1,, Guido Ceccherini 2, Frederic Achard 1, Valery Gond 3, Sylvie Gourlet-Fleury 3 and Paolo Omar Cerutti 4 Astrid Verhegghen 1,, Hugh Eva 1,, Guido Ceccherini 2, Frederic Achard 1, Valery Gond 3, Sylvie Gourlet-Fleury 3 and Paolo Omar Cerutti 4 1 Bio-Economy Unit, Sustainable Resources Directorate, Joint Research Centre (JRC), European Comm Bio-Economy Unit, Sustainable Resources Directorate, Joint Research Centre (JRC), European Commis Via E. Fermi, 2749, TP 261, Ispra VA 21027, Italy; frederic.achard@jrc.ec.europa.eu y , , J (J ), p Via E. Fermi, 2749, TP 261, Ispra VA 21027, Italy; frederic.achard@jrc.ec.europa.eu Via E. Fermi, 2749, TP 261, Ispra VA 21027, Italy; frederic.achard@jrc.ec.europa.eu 2 Water and Marine Resources Unit, Sustainable Resources Directorate, Joint Research Centre (JRC), European Commission, Via E. Fermi, 2749, TP 120, Ispra VA 21027, Italy; guido.ceccherini@ext.jrc.ec.europa.eu 2 Water and Marine Resources Unit, Sustainable Resources Directorate, Joint Research Centre (JRC), European Commission, Via E. Fermi, 2749, TP 120, Ispra VA 21027, Italy; guido.ceccherini@ext.jrc.ec.europa.eu g j p 3 Centre de coopération internationale en recherche agronomique pour le développement (CIRAD), UPR 105 Forêts et Sociétés, F-34398 Montpellier Cedex 5, France; valery.gond@cirad.fr (V.G.); sylvie.gourlet-ﬂeury@cirad.fr (S.G.-F.) 3 Centre de coopération internationale en recherche agronomique pour le développement (CIRAD), UPR 105 Forêts et Sociétés, F-34398 Montpellier Cedex 5, France; valery.gond@cirad.fr (V.G.); sylvie.gourlet-ﬂeury@cirad.fr (S.G.-F.) 4 Center for International Forestry Research (CIFOR), United Nations Avenue, Gigiri, P.O. Box 30677-00100, Nairobi 00601, Kenya; P.Cerutti@cgiar.org * Correspondence: astrid.verhegghen@jrc.ec.europa.eu (A.V.); hugh.eva@jrc.ec.europa.eu (H.E.); Tel.: +39-0332-785-016 (A.V.) Academic Editors: Clement Atzberger, Ioannis Gitas and Prasad S. Thenkabail g Received: 31 August 2016; Accepted: 21 November 2016; Published: 30 November 2016 Abstract: In this study, the recently launched Sentinel-2 (S2) optical satellite and the active radar Sentinel-1 (S1) satellite supported by active ﬁre data from the MODIS sensor were used to detect and monitor forest ﬁres in the Congo Basin. In the context of a very strong El Niño event, an unprecedented outbreak of ﬁres was observed during the ﬁrst months of 2016 in open forests formations in the north of the Republic of Congo. The anomalies of the recent ﬁres and meteorological situation compared to historical data show the severity of the drought. Burnt areas mapped by the S1 SAR and S2 Multi Spectral Instrument (MSI) sensors highlight that the ﬁres occurred mainly in Marantaceae forests, characterized by open tree canopy cover and an extensive tall herbaceous layer. The maps show that the origin of the ﬁres correlates with accessibility to the forest, suggesting an anthropogenic origin. Astrid Verhegghen 1,, Hugh Eva 1,, Guido Ceccherini 2, Frederic Achard 1, Valery Gond 3, Sylvie Gourlet-Fleury 3 and Paolo Omar Cerutti 4 The combined use of the two independent and fundamentally different satellite systems of S2 and S1 captured an extent of 36,000 ha of burnt areas, with each sensor compensating for the weakness (cloud perturbations for S2, and sensitivity to ground moisture for S1) of the other. Keywords: Sentinel-2; Sentinel-1; tropical rainforests; ﬁres; burnt areas; Republic of Congo; logging roads remote sensing remote sensing remote sensing 1. Introduction The period 2015–2016 has been determined as an El Niño year [1]. A severe “drought” and an associated unprecedented outbreak of ﬁres have been observed in the north of the Republic of Congo from January to April 2016. First observations report [2] that the ﬁres took place in speciﬁc open forest formations, the Marantaceae forests. While the forests of the Congo Basin are second only to the Amazon Basin in extent, unlike the latter, these forests have, in general, been spared important deforestation due to large scale exploitation driven by commercial food production and farming [3–5]. However, selective logging activities, where speciﬁc species of trees are extracted from the forest, are widespread in most of the basin [6]. Of concern, to both the national forest service and to the local management, is the occurrence of ﬁres in active Remote Sens. 2016, 8, 986; doi:10.3390/rs8120986 www.mdpi.com/journal/remotesensing 2 of 22 Remote Sens. 2016, 8, 986 logging concessions, threatening current commercial extraction of round-wood, future regeneration of stock and biodiversity. Fire is a very common event on the African continent, however, the Congo Basin rain forests are usually too humid to burn [7]. In southeast Asia and in the Amazon Basin, tropical rainforests ﬁres are observed more frequently, as a consequence of current land-use practices [8]. Under normal conditions, ﬁre remains rare in tropical rainforests and, consequently, most plants and animals are poorly adapted [9]. Pan-tropical forest ﬁres are often more severe during El Niño events [8]. In 1997–1998, ﬁres associated with an exceptional drought caused by El Niño devastated large areas of tropical rain forests in Indonesia [10]. Wooster et al., 2012 [11] observed a relationship between monthly rainfall deﬁcit and ﬁres in Borneo during El Niño events for the years 1980–2000. In Amazonia, Nepstad et al., 1999 [12] highlight an increased vulnerability of the Brazilian Amazonian forests to ﬁre during drought events caused by El Niño. In addition, the hypothesis of a positive feedback between recent logging activities and ﬁre is made for Amazonian [13] and Indonesian forests [10]. In contrast, very little is known about increased vulnerability of Central African tropical forests to ﬁre due to El Niño related droughts and/or logging activities. In the last decades, remote sensing has been widely used to monitor active ﬁres and burnt areas at the global scale. 1. Introduction Sensors like Advanced Very High Resolution Radiometer (AVHRR), VEGETATION (VGT), Moderate Resolution Imaging Spectroradiometer (MODIS) and MEdium Resolution Imaging Spectrometer (MERIS) with spatial resolutions between 300 m and 1 km are generally used [14,15]. Their daily revisit cycle is useful to capture active ﬁre signatures or burn scars, however, the relatively coarse spatial resolution causes underestimations in the case of small burnt area extent [16]. Comparisons between different burnt area products generally show high variation in results [16]. Coarse spatial resolution satellite data also present methodological challenges in terms of assigning the correct vegetation class to the burnt area, which is important either for estimating emissions or for evaluating the ecological impact of ﬁres [17]. For tropical rainforests speciﬁcally, more knowledge is needed about which vegetation formation is burning in order to understand why the ﬁre is occurring at speciﬁc locations and times. Fusion between coarser and medium to high spatial resolution (5–30 m) satellite data have been used e.g. by Boschetti et al., [18] to overcome the problem of reduced temporal resolution of higher resolution satellite such as Landsat, ASTER or RapidEye. In tropical evergreen forests, an important challenge is the persistent cloud cover, requiring frequent acquisitions of optical data. Synthetic Aperture Radar (SAR) sensors can be used to overcome this problem because of their ability to penetrate clouds and haze [8,19]. High spatial resolution SAR data have been used together with coarse optical data in a few studies [19,20]. However, the use of SAR data in an operational way has been limited by the availability of time series of SAR acquisitions and the pre-processing requirements [21]. The new European Space Agency (ESA) Sentinel satellites (funded by the European Union and ESA member states) are expected to provide useful information for a wide range of land applications [22]. The two ﬁrst missions, Sentinel-1 (S1) and Sentinel-2 (S2) operational since October 2014 and December 2015, respectively, provide time series that can already be assessed for monitoring tropical forests. For the mapping of burnt areas, optical satellite data from the Multi Spectral Instrument (MSI) on S2 with a potential ﬁve-day temporal resolution (when the two satellites are operational) ﬁts the requirement for a higher spatial resolution burnt area product. Images from the radar C-SAR sensor on S1 can overcome cloud cover limitations in tropical evergreen forests. 2.1. Study Area 2.1. Study Area The study area is situated in the north of the Republic of Congo in the Sangha department. The climate across the study area is tropical humid characterized by an annual precipitation of 1600–2000 mm per year, with two dry seasons, one in June to August and a stronger one from December through to February [23–25]. The area is dominated by forests, which cover over 95% of the region. The different forest classes occurring in the study area are detailed by Gond et al., [26] (illustrated in Figure 1). This map classiﬁes humid forests into sub-classes according to their structure and phenology. Sub-classes such as swamp forests are found along the river network, while open to very open Marantaceae forests cover the majority of the study area. At the southern edge of the study area, savanna formations start to dominate. Fires in the region are relatively infrequent due to the high percentage of humid forest cover, and occur almost exclusively in the savanna regions in the south, the aquatic grasslands in the east and in and around villages where ﬁelds are being cleared. The study area is situated in the north of the Republic of Congo in the Sangha department. The climate across the study area is tropical humid characterized by an annual precipitation of 1600–2000 mm per year, with two dry seasons, one in June to August and a stronger one from December through to February [23–25]. The area is dominated by forests, which cover over 95% of the region. The different forest classes occurring in the study area are detailed by Gond et al., [26] (illustrated in Figure 1). This map classifies humid forests into sub-classes according to their structure and phenology. Sub-classes such as swamp forests are found along the river network, while open to very open Marantaceae forests cover the majority of the study area. At the southern edge of the study area, savanna formations start to dominate. Fires in the region are relatively infrequent due to the high percentage of humid forest cover, and occur almost exclusively in the savanna regions in the south, the aquatic grasslands in the east and in and around villages where fields are being cleared. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. 1. Introduction The complementarity of the two sensors can be used for the design of a joint passive and active burnt area mapping products at regular intervals for this region. The objective of this study is to review the recent occurrence of forest ﬁres in the north of the Republic of Congo and to evaluate the potential of a joint use of S1 and S2 satellites to detect and monitor burnt areas in tropical evergreen forests. The exceptional character of the ﬁre event for this region and the rainfall anomaly will be illustrated by reviewing climatic data and historical ﬁre record. The vegetation types in the study area 3 of 22 Remote Sens. 2016, 8, 986 will be investigated to assess which type of forest has been burnt. Active ﬁres detected by the MODIS sensor on board the Aqua and Terra satellites are used as a ﬁrst evidence of the location and timing of this event. S1 and S2 time series are used to map the location, extent and speed at which these ﬁres have progressed. The burnt area maps produced are used to validate the hypothesis that the ﬁres happened almost exclusively in Marantaceae forests. The proximity of the burnt areas to logging roads is also evaluated. MODIS sensor on board the Aqua and Terra satellites are used as a first evidence of the location and timing of this event. S1 and S2 time series are used to map the location, extent and speed at which these fires have progressed. The burnt area maps produced are used to validate the hypothesis that the fires happened almost exclusively in Marantaceae forests. The proximity of the burnt areas to logging roads is also evaluated. 2.1. Study Area 2.1. Study Area Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. Figure 1. Vegetation map of Gond et al., (2013) [26]. In black, the extent of the logging concessions (Source: Interactive Forest Atlas of the Republic of the Congo [27]). In red, the study area. 4 of 22 Remote Sens. 2016, 8, 986 The main commercial activity in the area is the extraction of round wood from concessions which are leased from the national authorities by private companies. These concessions cover an extensive part of forest in the north of the country, with wood quotas being extracted according to a management plan set out by concession managers in conjunction with the national authorities. Three UFA (unité forestière d’aménagement) encompass the Marantaceae forests as illustrated in Figure 1: Pokola and Ngombé in the Sangha department and Loundoungou Toukalaka in the Likouala department. The Ngombé concession contains the largest proportion of the Marantaceae forests and the subsequent ﬁres in 2016. The analyses presented in this study are focused on this area (the red box in Figure 1). The population density is very low (less than 1 inhabitant per km2 [28,29]). Indigenous populations occupy speciﬁc areas allocated in the logging concessions. 2.1. Study Area 2.1. Study Area The network of logging roads in the study area [30,31] is used as a proxy of human accessibility to the forest and related risk of human induced ﬁre outbreak. In concessions managed for timber production, secondary roads are usually closed after exploitation and a rapid regrowth of vegetation is observed after abandonment. Kleinschroth et al., [30] observed a median persistence of four years for open roads before reverting to a vegetated state and 20 years before complete disappearance (no longer distinguishable from the surrounding forest). 2.2. Climatic Data Information on rainfall is available from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) product [32]. Starting in 1981, CHIRPS is a 30-year quasi-global rainfall dataset. Spanning 50◦S to 50◦N (and all longitudes) CHIRPS incorporates 0.05◦resolution satellite imagery with in-situ station data to create gridded monthly time series for seasonal drought monitoring and water resource management [33]. The monthly mean average of the CHIRPS dataset for 1981–2015 highlights the two dry seasons; a short dry period in July and a longer dry season in December–February characteristic of the north of the Republic of Congo. To highlight rainfall anomalies from these data, the Standardized Precipitation Index (SPI) is used. The SPI is the standard index for quantifying and reporting meteorological drought. SPI quantiﬁes precipitation as a standardized departure from a selected probability distribution function that models the precipitation records [34]. Precipitation records are ﬁtted to a gamma distribution, and then transformed to a normal distribution. The SPI values represent the number of standard deviations that the observed value would deviate from the long-term mean for a normally distributed variable. The SPI can be retrieved for different periods, generally ranging from 1 to 36 months, using monthly input data. For this study, the SPI is calculated at monthly time scales. On short timescales, the SPI is closely related to soil moisture. The strength of the 2015–2016 El Niño climatic event is characterized by the value of the Oceanic Niño Index (ONI). The ONI represents the three-month moving average of the Sea Surface Temperature (SST) anomalies in the Niño 3.4 regions (5◦N–5◦S, 120◦–170◦W). For the 2015–2016 period, the ONI indicates an El Niño event from February 2015 to June 2016 with maximum anomaly of 2.3 ◦C which is considered as a very strong event. In the last decade, two weak El Niño events have been observed in 2004–2005 and 2006–2007 and two moderate events in 2002–2003 and 2009–2010. The last time an El Niño event was classiﬁed as very strong was during the 1997–1998 period [1]. 2.3. Marantaceae Forests The Marantaceae forests, central to the study, are conspicuous by an unusual physiognomy with a giant herbaceous continuous stratum in association with a scattered tree component. This vegetation, arranged as a heterogeneous patchwork within the lowland semi-deciduous forests, is widely distributed in northern Republic of Congo [35]. The Marantaceae are part of the arrowroot family, ﬂowering plants known for their large starchy rhizomes. The preferential propagation mode of these herbs is vegetative, through the scattering of underground stems or rhizomes [35,36]. 5 of 22 Remote Sens. 2016, 8, 986 The presence of Marantaceae forests is not easily explained; it is not linked to the availability of soil water content and there are no signs of recent changes in their distribution. One hypothesis is that anthropogenic perturbations in earlier times (1550 Before Present (BP)) such as cutting or burning of the understory is at the origin of the colonization of the forest by the Marantaceae. The herbaceous layer would then dominate and inhibit woody regeneration [35]. Maley et al., [37] make the hypothesis that a fragmentation of forest formations caused by a climatic drought 2500 years ago also favored the development of these open forest formations. Marantaceae forests are often assigned low conservation and research priorities; however, they not only have great importance to several large mammal species (gorillas, and forest elephants) [38] by providing a dependable year-round supply of vegetative food, but also store large quantities of carbon per unit area [39]. 2.4. Active Fire Data Active ﬁre data over the area from the Moderate Resolution Imaging Spectroradiometer (MODIS) were acquired from the Fire Information for Resource Management System (FIRMS) MODIS Fire Archive Download webpage [40]. The data are from the Collection 5 Near Real Time (NRT), extracted from the standard MCD14ML ﬁre product produced at the MODIS Fire Science Computing Facility (SCF). The MODIS data are available from 2000 to date, comprising of 1 km detections from the combined 4 overpasses of the sensors on the Aqua and Terra satellites. The ﬁres that are actively burning at the time of the satellite overpass are detected using thresholds in the middle and thermal infra-red channels [41]. A quality control channel accompanies the data. Although such data have a number of sampling issues, the comparison of the current number of ﬁre counts in a given region with the long term (i.e., 2000–2015) average can be useful to highlight possible anomalies. 2.5. Burnt Areas Detection Burnt area mapping is carried out using S1 and S2 datasets independently and all pixels detected as burnt will be combined into a map composite. Independent burnt area maps will be generated from the images acquired respectively by the sensors on S1 and S2 during the period of the 1 November 2015 to the 30 April 2016 for S1 and the 3 December 2015 to the 21 May 2016 for S2. The maps will then be combined at a 20-m resolution. Three different scenarios are foreseen when generating the map composite: (i) a pixel is detected as burnt in either S1 or S2 maps; (ii) both S1 and S2 detected a pixel as burnt; and (iii) neither S1 nor S2 detect a pixel as burnt. The different results are presented and discussed but the ﬁnal goal is to combine S2-based and S1-based maps to take advantage of the sensors complementarity. The S1 and S2 images are processed using the Google Earth Engine cloud environmental analysis platform. A ﬁeld visit to one of the sites was carried out in May 2016, noting the damage to trees and the regrowth of the burned herbaceous vegetation. Sentinel 2 However, a visual check revealed that this quality band does not give sufficiently reliable results for the study area. A cloud mask therefore is based on band 1 (sensitive to aerosols), the blue band 2 (sensitive to atmospheric contaminations), the near infrared band 8A and the short-wave infrared band 11. Thresholds are defined on the combination of these bands to eliminate clouds and cloud shadows from the S2 images. The Level 1C S2 images are contaminated with cloud and aerosols which need to be masked out before further processing. No robust cloud masking for S2 was available at the time of processing. Together with the Level 1C product, ESA provides a quality band (QA60) masking cloud and cirrus. However, a visual check revealed that this quality band does not give sufﬁciently reliable results for the study area. A cloud mask therefore is based on band 1 (sensitive to aerosols), the blue band 2 (sensitive to atmospheric contaminations), the near infrared band 8A and the short-wave infrared band 11. Thresholds are deﬁned on the combination of these bands to eliminate clouds and cloud shadows from the S2 images. g The detection of burnt areas is based on the difference of two vegetation indices computed on the 10-days time series of S2 cloud-free images and a pre-fire baseline value. The first step of the burnt area detection is a baseline cloud free mosaic which is built taking the median value of the first four images (3 and 23 December 2015, 1 and 12 January 2016). In the second step, the acquisitions from the 12 January 2016 to the 21 May 2016 are compared to the baseline mosaic to determine the precise timing and location of the fire events. The detection of burnt areas is based on the difference of two vegetation indices computed on the 10-days time series of S2 cloud-free images and a pre-ﬁre baseline value. The ﬁrst step of the burnt area detection is a baseline cloud free mosaic which is built taking the median value of the ﬁrst four images (3 and 23 December 2015, 1 and 12 January 2016). In the second step, the acquisitions from the 12 January 2016 to the 21 May 2016 are compared to the baseline mosaic to determine the precise timing and location of the ﬁre events. Sentinel 2 The S2-A satellite was successfully launched on the 23 June 2015, as part of the European Copernicus program (European Commission funding, ESA implementation) and the ﬁrst images were delivered four days later [42]. S2 satellites carry an innovative wide-swath, high-resolution Multi Spectral Instrument with 13 spectral bands spanning from the visible and the near infrared to the short wave infrared. The spatial resolution varies from 10 m to 60 m depending on the spectral band. The swath is 290 km. The temporal resolution of the S2-A satellite is 10 days. A revisit period of ﬁve-days at the equator will be ensured when the S2-B twin satellite becomes operational [43]. Over the study area, a time series of ﬁfteen Level 1C (Top of Atmosphere reﬂectance) S2 images is available from the 3 December 2015 to the 21 May 2016. These show the progression of the ﬁres 6 of 22 Remote Sens. 2016, 8, 986 in the area (Figure 2). A gap in the 10-day temporal acquisition occurred during two weeks (from the 21 February to 2 March 2016) due to an anomaly in the ground segment production chain. This anomaly affects the radiometric quality of the S2 L1C products. Remote Sens. 2016, 8, 986 6 of 22 Figure 2. The progression of burnt areas (appear in red-purple color on the images) in the Marantaceae forests in the north of the Republic of Congo imaged by the Sentinel 2 Multi Spectral Instrument. Figure 2. The progression of burnt areas (appear in red-purple color on the images) in the Marantaceae forests in the north of the Republic of Congo imaged by the Sentinel 2 Multi Spectral Instrument. Figure 2. The progression of burnt areas (appear in red-purple color on the images) in the Marantaceae forests in the north of the Republic of Congo imaged by the Sentinel 2 Multi Spectral Instrument. Figure 2. The progression of burnt areas (appear in red-purple color on the images) in the Marantaceae forests in the north of the Republic of Congo imaged by the Sentinel 2 Multi Spectral Instrument. The Level 1C S2 images are contaminated with cloud and aerosols which need to be masked out before further processing. No robust cloud masking for S2 was available at the time of processing. Together with the Level 1C product, ESA provides a quality band (QA60) masking cloud and cirrus. Sentinel 2 g Burnt areas are identified according to the difference in the value of two vegetation indices between each date and the base mosaic. Both the Normalized Difference Water Index (NDWI; also known as the Normalized Difference Moisture Index, NDMI) [44,45] and the Normalized Difference Vegetation Index (NDVI) [46] are used to detect burnt areas on the S2 images. The NDWI is calculated according to the formula of Gao [44] and adapted to S2, using the NIR band 8 (842 nm) and the SWIR band 11 (1610 nm). The use of the band 11 constrains the spatial resolution of the product to 20 meters. The NDVI is calculated using the NIR band 8 (842 nm) and the Red band 4 (665 nm). The recent passage of fire causes a drop in the two indices’ values (the respective thresholds for the difference value are set to 0.18 for NDWI and 0.19 for NDVI). Very few fires occurred before the 12 January 2016 and the few burnt areas present on the 1 January 2016 image have been delimitated on that image with a specific threshold for NDWI and NDVI values. Burnt areas are identiﬁed according to the difference in the value of two vegetation indices between each date and the base mosaic. Both the Normalized Difference Water Index (NDWI; also known as the Normalized Difference Moisture Index, NDMI) [44,45] and the Normalized Difference Vegetation Index (NDVI) [46] are used to detect burnt areas on the S2 images. The NDWI is calculated according to the formula of Gao [44] and adapted to S2, using the NIR band 8 (842 nm) and the SWIR band 11 (1610 nm). The use of the band 11 constrains the spatial resolution of the product to 20 meters. The NDVI is calculated using the NIR band 8 (842 nm) and the Red band 4 (665 nm). The recent passage of ﬁre causes a drop in the two indices’ values (the respective thresholds for the difference value are set to 0.18 for NDWI and 0.19 for NDVI). Very few ﬁres occurred before the 12 January 2016 and the few burnt areas present on the 1 January 2016 image have been delimitated on that image with a speciﬁc threshold for NDWI and NDVI values. Sentinel 1 With a temporal resolution of 10 days, S2 images should allow a precise mapping of the timing and extent of ﬁre event. However, the Congo Basin region has extensive cloud cover at various times of the year [25] limiting the use of optical sensors. Observations from S1 can complement S2 optical acquisitions due to the high sensitivity to structural properties of vegetation and day and night observing capabilities of the radar sensor. S1 is a two satellite constellation with prime objectives of Land and Ocean monitoring. The satellites carry a C-SAR sensor, which offers medium and high resolution imaging in all weather conditions [47]. The C-band SAR instrument supports operation in single polarization (HH or VV) and dual polarization (HH + HV or VV + VH). The ﬁrst satellite, Sentinel-1A, currently has a revisit time of once every ~15 days over Congo. After its sister Sentinel-1B (launched in April 2016) comes into service, coverage will be further improved, allowing for more frequent updates. Severe ﬁres remove vegetation constituents such as trunks and branches which results in changes in the SAR intensity signal. Multi-temporal SAR images have been often used for burned area detection by computing the difference in backscatter before and after ﬁre events [19,21,48–52]. Kurum [53] developed a method to detect ﬁre affected areas using SAR series in C band (such as S1) with the co-polarized VV channel. Similarly, Main et al., [54] used cross-polarized (i.e., VH) C-band to assess woody structure and changes in deciduous savannas. The basic assumption of this study is that most of the C-band radar backscatter of intact forests comes from surface, volume and multiple volume-surface scattering. After a ﬁre, the surface scattering becomes the major contributor to the return signal [55,56]. At this stage, the difference between radar backscatter before and after the ﬁre can be used to delineate the ﬁre-affected area from its surroundings. However, changes in the ground moisture can also contribute to the SAR backscatter signal [57]. Hence limitations in the use of SAR are the soil moisture changes and the fact that the vegetation structure must be severely modiﬁed in order for a ﬁre to be detected. The S1 images available in this study were acquired with the Interferometric Wide swath (IW) acquisition mode at 25 m spatial resolution. In total, 24 images are available from December 2015 to April 2016. Sentinel 2 ܰܦܹܫ= ܰܫܴ−ܹܵܫܴ ܰܫܴ+ ܹܵܫܴ (1) ܰܦܸܫ= ܰܫܴ−ܴܧܦ ܰܫܴ+ ܴܧܦ (2) NDWI = NIR −SWIR NIR + SWIR (1) NDVI = NIR −RED NIR + RED (2) ܰܦܹܫ= ܰܫܴ−ܹܵܫܴ ܰܫܴ+ ܹܵܫܴ (1) ܰܦܸܫ= ܰܫܴ−ܴܧܦ ܰܫܴ+ ܴܧܦ (2) NDWI = NIR −SWIR NIR + SWIR (1) NDVI = NIR −RED NIR + RED (2) (1) (1) (2) (2) 7 of 22 Remote Sens. 2016, 8, 986 7 of 22 Sentinel 1 The methodology employed to delineate the ﬁre-affected area is carried out in the following steps and summarized in Figure 3: (1) map Gamma ﬁlter correction; (2) temporal aggregation of S1 imagery; (3) calculation of backscatter differences between consecutive periods; and (4) threshold application and consistency test. All SAR images are ﬁrst corrected using Map Gamma ﬁlter technique described in Lopes et al., [58] to reduce speckle. Map Gamma is an adaptive ﬁlters that reduces the variance (i.e., speckle) within the image while preserving both radiometric and spatial resolution. The ﬁlter is based on the assumption that the (unspeckled) intensity of the scene is gamma distributed. The ﬁlter minimizes the loss of texture information and it is thus suitable for a wide range of gamma distributed scenes, such as forest and agriculture areas. The S1 temporal aggregation is performed on bi-monthly basis calculating the average of the SAR stack. The 2-month time span is necessary to further reduce the noise of SAR data records. The calculation of backscatter differences between consecutive periods allow to detect ﬁre scars that exhibit large differences in backscatter. The difference in the signal between different periods is simply computed by subtracting the backscatter of the second period to the backscatter of the ﬁrst one. Pixels that display a difference greater than a speciﬁc threshold are labeled as ﬁre scars. The threshold is land cover dependent. The variation of backscatter in evergreen forest is relatively stable due to the homogeneous forest canopy. Conversely, croplands and other land uses exhibit important variation in backscatter signal over time. In order to quantify the S1 forest-ﬁres backscatter threshold, artiﬁcial test cases have been designed over the study area. The test was carried out using different threshold differences between S1 backscatter from different periods ranging from −0.1 to −2.0 dB with a step of −0.1 dB. A backscatter difference of −1.0 dB provided the optimal threshold value for the assessment of ﬁre scars in the evergreen forests of the study area. A lower threshold 8 of 22 Remote Sens. 2016, 8, 986 p y To remove i l value would have increased the data noise, whereas, at higher values, the ﬁre scars effects would have been less evident. pp The procedure has been applied with both VV and VH polarizations, but only VV shows results coherent with S2 results. Figure 3. Methodological scheme for processing of Sentinel 1 data. Figure 3. Sentinel 1 Methodological scheme for processing of Sentinel 1 data. Figure 3. Methodological scheme for processing of Sentinel 1 data. Figure 3. Methodological scheme for processing of Sentinel 1 data. 2.6. Forest Cover Percentage Using VHR Data In order to assess the opening of the forest canopy and whether the fires were limited to the most open areas, a very high resolution image (VHR) was analysed to calculate the percentage tree cover of the forest over a small part (100 km2) of the study. The image is a GeoEye-1 four band pan sharpened image at a nominal spatial resolution of 0.4 m acquired the 28 November 2011. GeoEye-1 A simple threshold difference approach can however introduce commission errors. S1 data, although ﬁltered through the Map Gamma, still suffer from speckle noise and the backscatter strongly depends on the soil moisture content of the ground. A difference between two consecutive periods, e.g., due to soil moisture across intact forests may thus be detected as ﬁre scar. spectral bands are the following, Blue (450–510 nm), Green (510–580 nm), Red (655–690 nm) and Near Infra-Red (780–920 nm). The opening of the canopy is evaluated by segmenting the image into homogeneous textural objects. The data were segmented with the eCognition software to an average of 0.2 ha objects with a scale factor of 70 (shape and compactness of 0.5 and 0.7). A decision tree classifier was used to separate out four classes, bare soil (mainly logging roads), grass, shrubs and tree cover. Apart from the original spectral channels, the NDVI and the standard deviations of the red and near infra-red To mitigate the negative effect of these commission errors a “coherency test” between three consecutive periods (i.e., the S1 bi-monthly aggregations described above) is introduced. The test is carried out by detecting ﬁre scars between: (1) the ﬁrst and the second period (i.e., March–April aggregation vs. January–February one); and (2) the ﬁrst and the third period (i.e., March–April aggregation vs. December–January one). Fire scars found for the ﬁrst pair that are not subsequently conﬁrmed by the second pair are rejected. Therefore, pixels classiﬁed as burned that do not display a time-consistent change in backscatter signals are considered as outliers. This procedure is applied to ensure that changes in backscatter signal are due to the vegetation structure rather than to temporary differences in soil moisture. Sentinel 1 To remove possible errors caused by topography, SAR data are usually ﬁrst corrected using a maximum slope threshold. As the terrain is ﬂat in the study area, such correction is neither needed nor applied. The procedure has been applied with both VV and VH polarizations, but only VV shows results coherent with S2 results. 2.6. Forest Cover Percentage Using VHR Data In order to assess the opening of the forest canopy and whether the ﬁres were limited to the most open areas, a very high resolution image (VHR) was analysed to calculate the percentage tree cover of the forest over a small part (100 km2) of the study. The image is a GeoEye-1 four band pan Remote Sens. 2016, 8, 986 9 of 22 Remote Sens. 2016, 8, 986 sharpened image at a nominal spatial resolution of 0.4 m acquired the 28 November 2011. GeoEye-1 spectral bands are the following, Blue (450–510 nm), Green (510–580 nm), Red (655–690 nm) and Near Infra-Red (780–920 nm). The opening of the canopy is evaluated by segmenting the image into homogeneous textural objects. The data were segmented with the eCognition software to an average of 0.2 ha objects with a scale factor of 70 (shape and compactness of 0.5 and 0.7). A decision tree classiﬁer was used to separate out four classes, bare soil (mainly logging roads), grass, shrubs and tree cover. Apart from the original spectral channels, the NDVI and the standard deviations of the red and near infra-red channels were used for the segmentation and the classiﬁcation. The decision tree classiﬁer was trained using 100 randomly selected points visually interpreted into the 4 classes. A second level of segmentation with a scale factor of 200 (shape and compactness of 0.5 and 0.9) was undertaken to obtain 0.5 ha polygons. The relative proportion of forests in each 0.5 ha segment was calculated from the ﬁrst level segmentation of 0.2 ha. Remote Sens. 2016, 8, 986 9 of 22 channels were used for the segmentation and the classification. The decision tree classifier was trained using 100 randomly selected points visually interpreted into the 4 classes. A second level of segmentation with a scale factor of 200 (shape and compactness of 0 5 and 0 9) was undertaken to The joint S2-S1 map of burnt areas was compared to this estimation of tree cover to evaluate the occurrence of ﬁres in the more open forests types. segmentation with a scale factor of 200 (shape and compactness of 0.5 and 0.9) was undertaken to obtain 0.5 ha polygons. The relative proportion of forests in each 0.5 ha segment was calculated from the first level segmentation of 0.2 ha. 2.6. Forest Cover Percentage Using VHR Data Th j i t S2 S1 f b t d t thi ti ti f t t l t th 3. Results occu 3.2. Active Fires An exceptional number of active ﬁres were detected in the MODIS data for the region compared to the average of ﬁres observed during the 2000–2015 period. Normally, ﬁres in the Congo Basin are conﬁned to savanna regions and areas in and around villages, where clearance and crop residue burning are occurring. Fires are also observed in the aquatic grassland situated in the east of the study area. Fires within the forest domain area are exceedingly rare, due to the high humidity. Remote Sens. 2016, 8, 986 10 of 22 Examining the ﬁres by land cover classes shows that the three months from December 2015 to February 2016 were characterized by an exceptional number of ﬁre outbreaks, with nearly seven times as many ﬁres occurring in January 2016 as would normally be expected from the long term average. The monthly average May to April calculated from the 2000 to 2015 period is shown in Figure 5, along with the period from May 2015 to April 2016, both for all land cover types and speciﬁcally the Marantaceae forests. On average, for the 2000–2015 period, Marantaceae forests were affected by one or two ﬁres at the turn of the year. In January and February of 2016, 173 and 123 ﬁre detections were counted, respectively. A few ﬁres are observed in March 2016. Over half of the increase in ﬁres this year occurred in these forests. Examining the fires by land cover classes shows that the three months from December 2015 to February 2016 were characterized by an exceptional number of fire outbreaks, with nearly seven times as many fires occurring in January 2016 as would normally be expected from the long term average. The monthly average May to April calculated from the 2000 to 2015 period is shown in Figure 5, along with the period from May 2015 to April 2016, both for all land cover types and specifically the Marantaceae forests. On average, for the 2000–2015 period, Marantaceae forests were affected by one or two fires at the turn of the year. In January and February of 2016, 173 and 123 fire detections were counted, respectively. A few fires are observed in March 2016. Over half of the increase in fires this year occurred in these forests. Figure 5. The number of active fires detected by the MODIS sensors on-board Aqua and Terra for the study area. 3.1. Rainfall Trends 3. Results Analysis of the recent rainfall trends with respect to the long term (1981–2010) average, show an excess in precipitation in late 2015, followed by a drought at the time of the ﬁres. In Figure 4, SPI values more negative than −1 indicate a condition of drought. The more negative the value the more severe the situation. SPI values higher than +1 indicate more humid conditions compared to a normal situation. When the SPI has a value between −1 and +1 the situation is identiﬁed as normal. The SPI results support the initial observation. An unusually wet November is followed by severe droughts for the months of December to February 2016 in the North of the Republic of Congo. 3.1. Rainfall Trends Analysis of the recent rainfall trends with respect to the long term (1981–2010) average, show an excess in precipitation in late 2015, followed by a drought at the time of the fires. In Figure 4, SPI values more negative than −1 indicate a condition of drought. The more negative the value the more severe the situation. SPI values higher than +1 indicate more humid conditions compared to a normal situation. When the SPI has a value between −1 and +1 the situation is identified as normal. The SPI results support the initial observation. An unusually wet November is followed by severe droughts for the months of December to February 2016 in the North of the Republic of Congo. Figure 4. The Standardized Precipitation Index (SPI) for the Republic of Congo from November 2015 to April 2016. 3 2 A i Fi Figure 4. The Standardized Precipitation Index (SPI) for the Republic of Congo from November 2015 to April 2016. Figure 4. The Standardized Precipitation Index (SPI) for the Republic of Congo from November 2015 to April 2016. Figure 4. The Standardized Precipitation Index (SPI) for the Republic of Congo from November 2015 to April 2016. 10 of 22 Remote Sens. 2016, 8, 986 10 of 22 3.2. Active Fires The long term (2000–2014) average is under 50 fires per month, compared to a peak in January 2016 of over 250 fire events, of which a large proportion occurred in the Marantaceae open forests. 0 50 100 150 200 250 300 Number of fires detected Month of the year Full study area average 2000-2014 Full study area average 2015- 2016 Marentaceae forests 2015-2016 Figure 5. The number of active ﬁres detected by the MODIS sensors on-board Aqua and Terra for the study area. The long term (2000–2014) average is under 50 ﬁres per month, compared to a peak in January 2016 of over 250 ﬁre events, of which a large proportion occurred in the Marantaceae open forests. Month of the year Figure 5. The number of active fires detected by the MODIS sensors on-board Aqua and Terra for the study area. The long term (2000–2014) average is under 50 fires per month, compared to a peak in January 2016 of over 250 fire events, of which a large proportion occurred in the Marantaceae open forests. Figure 5. The number of active ﬁres detected by the MODIS sensors on-board Aqua and Terra for the study area. The long term (2000–2014) average is under 50 ﬁres per month, compared to a peak in January 2016 of over 250 ﬁre events, of which a large proportion occurred in the Marantaceae open forests. 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 2016, 8, 986 Figure 6. Spatial coherence in the location of the burnt areas as detected by: Sentinel 2 (S2) (A); and Sentinel 1 (S1) (B) and MODIS active fire product (black dots). (C) Map obtained by combining S2 and S1 based maps. S2 data (27 January 2016) are in the background. Figure 6. Spatial coherence in the location of the burnt areas as detected by: Sentinel 2 (S2) (A); and Sentinel 1 (S1) (B) and MODIS active ﬁre product (black dots). (C) Map obtained by combining S2 and S1 based maps. S2 data (27 January 2016) are in the background. Figure 6. Spatial coherence in the location of the burnt areas as detected by: Sentinel 2 (S2) (A); and Sentinel 1 (S1) (B) and MODIS active fire product (black dots). (C) Map obtained by combining S2 and S1 based maps. S2 data (27 January 2016) are in the background. Figure 6. Spatial coherence in the location of the burnt areas as detected by: Sentinel 2 (S2) (A); and Sentinel 1 (S1) (B) and MODIS active ﬁre product (black dots). (C) Map obtained by combining S2 and S1 based maps. S2 data (27 January 2016) are in the background. The final combined S2-S1 map is presented in Figure 7. In terms of area, the combined S2-S1 t ti ff t d b fi f 36 247 h i th t d f th i d J A il Table 1. Comparison of the detection of burnt areas with Sentinel 2 and Sentinel 1 derived maps. The final combined S2-S1 map is presented in Figure 7. In terms of area, the combined S2-S1 map s detecting an area affected by fire of 36,247 ha in the study area for the period January–April 2016 Table 2). Of this area, more than 32,382 ha are localized in the Ngombé logging concession. The burnt areas represent about 3% of the concession total area. Some burnt areas are also observed in the Pokala and Loundoungou Toukalaka logging concessions, situated outside the study area, but epresent a much smaller spatial extent Table 1. Comparison of the detection of burnt areas with Sentinel 2 and Sentinel 1 derived maps. 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 Maps of burnt areas at a 20 and 25 m resolution are obtained from S2 and S1 images. Burnt areas from S2 are mapped on a 10-day basis, from the 12 January to the 21 May 2016. The gap in S2 images from mid-February to mid-March is compensated by a cloud free acquisition mid-March. Two maps of burnt areas are produced from S1 data, one capturing the difference between December 2015 and January 2016 and the second one for the period December 2015 to March 2016. This mapping is based on the difference of the backscatter signal during those periods of time. Maps of burnt areas at a 20 and 25 m resolution are obtained from S2 and S1 images. Burnt areas from S2 are mapped on a 10-day basis, from the 12 January to the 21 May 2016. The gap in S2 images from mid-February to mid-March is compensated by a cloud free acquisition mid-March. Two maps of burnt areas are produced from S1 data, one capturing the difference between December 2015 and January 2016 and the second one for the period December 2015 to March 2016. This mapping is based on the difference of the backscatter signal during those periods of time. As illustrated in Figure 6, the maps of burnt areas derived from S1 and S2 images are not identical. The burnt area estimations by S1 and S2 are presented in Table 1. The estimations vary from around 26,000 ha for S1 to 33,000 ha for S2. Some areas are not detected by S2 because of constant high cloud coverage. Overall, S1 detects a smaller extent of burnt area than S2 due to the choice of a conservative threshold to avoid overestimation related to the change of water content in the t ti g g p As illustrated in Figure 6, the maps of burnt areas derived from S1 and S2 images are not identical. The burnt area estimations by S1 and S2 are presented in Table 1. The estimations vary from around 26,000 ha for S1 to 33,000 ha for S2. Some areas are not detected by S2 because of constant high cloud coverage. Overall, S1 detects a smaller extent of burnt area than S2 due to the choice of a conservative threshold to avoid overestimation related to the change of water content in the vegetation. 11 of 22 11 f 22 Remote Sens. 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background. Figure 7. Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background. Figure 7. Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background. Figure 7. Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background. Timing of the Fires The timing of the fire can be derived from the active fire product derived from MODIS data Table 2. Estimation of burnt area in the forest classes with the Sentinel 2 and Sentinel 1 combined product. emented by the S2 maps. As illustrated in Figure 8, coherence between the two in ral pattern is visible. he active fire product gives us the precise date of the fire observation with a daily tion. The timing detected by S2 strongly depends on the time of acquisition of th tically every 10 days, and the absence of clouds at the time of acquisition. From the ctive fires, it is observed that fires occur from the 26 December 2015 to the 10 April 20 Burnt Area (in ha) for the Combined S2 and S1 Map Marantaceae forest 30,128 (83.1%) Border of Marantaceae forest 4575 (12.6%) Dense forests 678 (1.9%) In the logging concession (Ngombé) 32,382 (89%) S2 based maps it can be derived that the most important progression of fire happened between end of January and February 2016. From MODIS active fire detection and S2 maps (Figure 8), it is evident that fires started in different locations rather than expanding from a single origin. The combination of both products shows that the first fires happened at the southern edge of the Ngombé logging concession, in late December 2015–early January 2016 (Figure 8B). This part of forest is adjacent to a savanna. Following that, active fires are detected in the north of the area where a network of old logging roads exists. 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 Burnt Area (in ha) Sentinel 2 (S2) Sentinel 1 (S1) S2 and S1 Combined Intersection of S2 and S1 32,997 26,124 36,247 22,874 In terms of forest type affected by the fires, it is crucial to notice that the fires are restricted to a very specific type of forests formation. A spatial comparison of the extent of burnt areas and the vegetation map of Gond et al., [26] showed that 83% of the burnt area is localized in the core of Marantaceae forests and another 13% in the border of these forests Another 2% is found in dense The maps obtained from S1 and S2 present a 63% spatial coherence (Figure 6). A common area of 22,874 ha is found between S1 and S2 (Table 1). The location of active ﬁres (MODIS) is coherent with maps of burnt areas derived from S1 and S2 imagery. Marantaceae forests and another 13% in the border of these forests. Another 2% is found in dense forest classes. The rest are found in a mixture of classes, probably due to the coarse spatial resolution of the vegetation map creating a geometric mismatch. In conclusion, 96% of the burnt areas are localized in very open to open forest formations. Table 2. Estimation of burnt area in the forest classes with the Sentinel 2 and Sentinel 1 combined product The ﬁnal combined S2-S1 map is presented in Figure 7. In terms of area, the combined S2-S1 map is detecting an area affected by ﬁre of 36,247 ha in the study area for the period January–April 2016 (Table 2). Of this area, more than 32,382 ha are localized in the Ngombé logging concession. The burnt areas represent about 3% of the concession total area. Some burnt areas are also observed in the Pokala and Loundoungou Toukalaka logging concessions, situated outside the study area, but represent a much smaller spatial extent. 12 of 22 Remote Sens. 2016, 8, 986 Figure 7. Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background Figure 7. Burnt areas mapped with Sentinel-2 and Sentinel-1 images acquired between November 2015 and April 2016 (S1) and December 2015 and May 2016 (S2). Vegetation map of Gond et al., [26] in the background Figure 7. 3.4. Timing of the Fires The timing of the ﬁre can be derived from the active ﬁre product derived from MODIS data and complemented by the S2 maps. As illustrated in Figure 8, coherence between the two in term of temporal pattern is visible. Remote Sens 2016 8 986 13 of 22 The timing of the ﬁre can be derived from the active ﬁre product derived from MODIS data and complemented by the S2 maps. As illustrated in Figure 8, coherence between the two in term of temporal pattern is visible. Remote Sens. 2016, 8, 986 13 of 22 Figure 8. Timing of the burnt areas detections with Sentinel 2 together with the MODIS active fire product in the north of the Republic of Congo. The logging roads come from [31]. (A) In the north of the study area. (B) In the south of the study area. Figure 8. Timing of the burnt areas detections with Sentinel 2 together with the MODIS active ﬁre product in the north of the Republic of Congo. The logging roads come from [31]. (A) In the north of the study area. (B) In the south of the study area. Figure 8. Timing of the burnt areas detections with Sentinel 2 together with the MODIS active fire product in the north of the Republic of Congo. The logging roads come from [31]. (A) In the north of the study area (B) In the south of the study area Figure 8. Timing of the burnt areas detections with Sentinel 2 together with the MODIS active ﬁre product in the north of the Republic of Congo. The logging roads come from [31]. (A) In the north of the study area. (B) In the south of the study area. Burnt areas are observed on the S2 image of 22 January in three distinct locations shown in dark blue in Figure 8A. Two of these areas are located in a previously logged area, close to old logging roads. The third one is located along the main road. From the MODIS active fires data, it can be deduced that these areas started burning from 16 January. The fire continued spreading in these three locations but also started in new ones. 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 3.3. Mapping of Burnt Areas with Sentinel-1 and Sentinel-2 In terms of forest type affected by the ﬁres, it is crucial to notice that the ﬁres are restricted to a very speciﬁc type of forests formation. A spatial comparison of the extent of burnt areas and the vegetation map of Gond et al., [26] showed that 83% of the burnt area is localized in the core of Marantaceae forests and another 13% in the border of these forests. Another 2% is found in dense forest classes. The rest are found in a mixture of classes, probably due to the coarse spatial resolution of the vegetation map creating a geometric mismatch. In conclusion, 96% of the burnt areas are localized in very open to open forest formations. 13 of 22 13 of 22 Remote Sens. 2016, 8, 986 3.4. Timing of the Fires In 10 days, from 22 January to 1 February, the estimated burnt area increased from less than 2000 ha to more The active ﬁre product gives us the precise date of the ﬁre observation with a daily temporal resolution. The timing detected by S2 strongly depends on the time of acquisition of the image, theoretically every 10 days, and the absence of clouds at the time of acquisition. From the MODIS data active ﬁres, it is observed that ﬁres occur from the 26 December 2015 to the 10 April 2016. From S2 based maps it can be derived that the most important progression of ﬁre happened between end of January and February 2016. Remote Sens. 2016, 8, 986 14 of 22 From MODIS active ﬁre detection and S2 maps (Figure 8), it is evident that ﬁres started in different locations rather than expanding from a single origin. The combination of both products shows that the ﬁrst ﬁres happened at the southern edge of the Ngombé logging concession, in late December 2015–early January 2016 (Figure 8B). This part of forest is adjacent to a savanna. Following that, active ﬁres are detected in the north of the area where a network of old logging roads exists. Burnt areas are observed on the S2 image of 22 January in three distinct locations shown in dark blue in Figure 8A. Two of these areas are located in a previously logged area, close to old logging roads. The third one is located along the main road. From the MODIS active ﬁres data, it can be deduced that these areas started burning from 16 January. The ﬁre continued spreading in these three locations but also started in new ones. In 10 days, from 22 January to 1 February, the estimated burnt area increased from less than 2000 ha to more than 18,000 ha. As no observations are available between those two dates, it is impossible to determine from the S2 maps if the ﬁre spread from these places to the other ones or if there were new sources of ﬁre in the meantime. However, from the active ﬁre data, some ﬁres are detected between 27 and 29 January in the northeast of Figure 8A (yellow dots), close to the logging road. Following this, the ﬁre spread to the rest of the area 30–31 January (orange dots in Figure 8A). 3.4. Timing of the Fires After 1 February, the ﬁre continued spreading but at a lower rate. Ten days later (11 February,) the burnt area has increased of about 4000 ha. The next S2 observation is 12 March where 8300 ha of additional burnt areas are observed. A gap in the acquisition of S2 prevents knowing when those ﬁres exactly happened but the MODIS active ﬁres product helps to determine the temporal dynamic of the ﬁre. In the south of the concession (Figure 8B), most of the ﬁres following the 11 February S2 acquisition seems to have happened around 15–17 February, while in the north (Figure 8A), they were observed 17–19 February. It seems than most of the burnt areas detected with S2 in March and April are events that happened earlier but were not captured by S2 due to the cloud coverage. An area in the south west also burnt during the period of 16–19 February and was only captured in the S2 images of April. Remote Sens. 2016, 8, 986 14 of 22 After 1 February, the fire continued spreading but at a lower rate. Ten days later (11 February,) the burnt area has increased of about 4000 ha. The next S2 observation is 12 March where 8300 ha of additional burnt areas are observed. A gap in the acquisition of S2 prevents knowing when those fires exactly happened but the MODIS active fires product helps to determine the temporal dynamic of the fire. In the south of the concession (Figure 8B), most of the fires following the 11 February S2 acquisition seems to have happened around 15–17 February, while in the north (Figure 8A), they were observed 17–19 February. It seems than most of the burnt areas detected with S2 in March and April are events that happened earlier but were not captured by S2 due to the cloud coverage. An area in the south west also burnt during the period of 16–19 February and was only captured in the S2 i f A il 3.5. Forest Cover Density of the Maranthaceae Forests g p 3 5 Forest Cover Density of the Maranthaceae Forests From the VHR image, results show that the percentage of tree cover has high variation, with open forests dominating in the north and central west of the GeoEye image. These areas correspond to the Marantaceae forests as mapped by Gond et al., [26]. Figure 9 highlights that the main ﬁre fronts initiated in more open forest formations and stopped when the forest becomes denser. The burnt areas cover mostly the open part of the forests but also extend partly into more closed canopy forest. 3.5. Forest Cover Density of the Maranthaceae Forests From the VHR image, results show that the percentage of tree cover has high variation, with open forests dominating in the north and central west of the GeoEye image. These areas correspond to the Marantaceae forests as mapped by Gond et al., [26]. Figure 9 highlights that the main fire fronts initiated in more open forest formations and stopped when the forest becomes denser. The burnt areas cover mostly the open part of the forests but also extend partly into more closed canopy forest. Figure 9. Cont. Figure 9. Cont. 15 of 22 Remote Sens. 2016, 8, 986 Figure 9. (A) Percentage of tree cover derived from a 2011 GeoEye image; (B) Burnt areas in 2016 detected on Sentinel images; (C) Zoomed in GeoEye image showing the transition from very open canopy to open forest canopy; and (D) Timing of the detection of burnt area for the magnified image. Fire happened first in the more open areas. Figure 9. (A) Percentage of tree cover derived from a 2011 GeoEye image; (B) Burnt areas in 2016 detected on Sentinel images; (C) Zoomed in GeoEye image showing the transition from very open canopy to open forest canopy; and (D) Timing of the detection of burnt area for the magniﬁed image. Fire happened ﬁrst in the more open areas. Figure 9. (A) Percentage of tree cover derived from a 2011 GeoEye image; (B) Burnt areas in 2016 detected on Sentinel images; (C) Zoomed in GeoEye image showing the transition from very open canopy to open forest canopy; and (D) Timing of the detection of burnt area for the magnified image. Fire happened first in the more open areas. Figure 9. 4.1. Exceptional Fire Outbreak From December 2015 to February 2016, a drought is observed in the Republic of Congo, leading to an unprecedented ﬁre outbreak in the Marantaceae forests in the north of the country. A review of the meteorological data (Figure 3) clearly shows the arrival of this anomaly, and with hindsight could have been used to predict a potential outbreak of ﬁre in the forest ecosystem. More than 36,000 ha of forest have burnt in less than three months. The most important ﬁres are observed in January and February 2016. Some active ﬁres were observed at the end of December 2015 and in March–April 2016. The end of the ﬁre outbreaks can be linked to the arrival of rains in mid-March. The magnitude of the anomaly is demonstrated by the exceptional number of active ﬁres that have been detected for the region compared to the average of ﬁres observed during the 2000–2015 period. It is the only time that ﬁres have been detected in these formations since November 2000 when the MODIS sensor became operational on the Terra satellite. While observations of ﬁre in tropical forests are frequent in Amazonia or Indonesia, few records of ﬁre are found for Central Africa. To our knowledge, it is the ﬁrst time that type of event is describe and documented in Central Africa. Gillet [35] reports an outbreak of ﬁre in 1970 in the southern part of the Marantaceae forest within the Ngombé logging concession, supposedly following an El Niño event. This event is reported by the timber company now operating in the concession, but no satellite image or maps are available to corroborate this. 3.5. Forest Cover Density of the Maranthaceae Forests g p 3 5 Forest Cover Density of the Maranthaceae Forests (A) Percentage of tree cover derived from a 2011 GeoEye image; (B) Burnt areas in 2016 detected on Sentinel images; (C) Zoomed in GeoEye image showing the transition from very open canopy to open forest canopy; and (D) Timing of the detection of burnt area for the magniﬁed image. Fire happened ﬁrst in the more open areas. 4.2. Potential of Sentinel 1 and Sentinel 2 for Burnt Areas Monitoring To monitor and map ﬁre events in tropical forests, the sensors installed on the S1 and S2 satellites show some potential. The arrival of the S2 satellite provides a far higher frequency of Earth Observation data than was previously possible at this ﬁne (10–20 m) spatial resolution. Due to its novelty, literature about the use of time series of S2 images is scarce and mainly covers the use of single or pair of images but ﬁrst experiences conﬁrm a good potential for land cover mapping capabilities [59,60]. The frequent observations by S1 over the Congo Basin and its independence from cloud cover also make the mission an indispensable asset. The use of the two sensors (supported by the active ﬁres from MODIS) allows a fuller picture of what has actually occurred. Remote Sens. 2016, 8, 986 16 of 22 16 of 22 The two sensors are sensitive to different phenomena, the S2 optical sensor detecting changes in surface reﬂectance, while the S1 SAR is detecting changes in vegetation structure and moisture. It may therefore be that the S1 SAR is only highlighting areas where the ﬁres were more intense, and the vegetation structure was more affected, which can only be assessed with ﬁeld data. With S1 data, some technical constrains have to be respected. In order to remove speckle and noise, monthly composites have to be used. Due to the high rainfall in November and the drought of December to February, the change of soil moisture had an impact on the changes in backscattering, requiring a higher threshold for the detection of burnt areas. Some small errors of commission were observed most probably due to the change in soil moisture. Errors due to the topography are not expected here, as the study area is quite ﬂat. With S2 optical data, cloud cover limits the data available and unlike other regions the dry season in Africa does not mean an absence of clouds [25]. However, changes in reﬂectance can be captured for a few months, depending on the pace of vegetation regrowth. The limitations of the individual systems are reduced by using both S1 and S2. Figure 10 illustrates the extreme cases for each sensor. In Figure 10A, S2 detects burnt areas (in blue) while S1 does not, probably because of a less severe ﬁre. 4.3. Spatial Distribution of the Burnt Areas In reviewing the spatial distribution of the ﬁres in the forest domain, it is clear that the ﬁres occurred mainly in open forests, more speciﬁcally in the very open Marantaceae forests. From the density of roads, it is deducted that while mainly located inside a logging concession, ﬁres occurred in old logged areas and in areas that were never exploited. Around 80% of the burnt areas detected with S1 and S2 data occur in the core of the very open Marantaceae forest. Another 13% is observed in the open forests situated at the border of the very open forests formations. According to the analysis, the dense forest was only slightly affected, with less than 2% of this class mapped as burnt. However, this low percentage is only indicative as the vegetation map has a coarser spatial resolution compared to the burnt area maps. Only a certain proportion of Marantaceae forests were affected by ﬁres. According to Gond et al., [26], Marantaceae formations in the study area occupy more than 300,000 ha. To understand the distribution of the ﬁre, we looked at the logging roads network and the accessibility in general. For the burnt areas observed in historically logged areas in the north (Figure 8A), the origin of the ﬁres is adjacent to old logging roads. The roads were identiﬁed by Kleinschroth et al., [31] as not yet entirely recolonized by vegetation in 2015. Although these roads have been closed and are no more used for active logging exploitation, they may have remained partly accessible for people on foot. In Amazonian and Indonesian tropical forests, the presence of recent logging roads has been reported as a potential facilitator for ﬁre in forests [8,10,61]. For the burnt areas observed in areas not previously logged, some are located along the main road (visible as a black thick line in Figure 8A), where there are a number of small to large villages. Some areas are located up to a distance of 8 km from the road. The areas accessed by local communities and indigenous people for hunting and gathering is thought to reach up to 30 km into the forest. Another extent of burnt area is situated in the south of the study area (Figure 8B), close to savannas. 4.2. Potential of Sentinel 1 and Sentinel 2 for Burnt Areas Monitoring In Figure 10B, however, S1 (in orange) detects burnt areas that were not detected by S2 due to high cloud coverage. Remote Sens. 2016, 8, 986 16 of 22 Figure 10. Examples of complementary use of Sentinel 1 and Sentinel 2 data in two specific zones of the study area (A) and (B). Areas detected only by Sentinel 2 are in blue, only by Sentinel 1 in orange and areas detected as burnt by both sensors are in dark red. Active fires (MODIS) are represented by the black dots. Figure 10. Examples of complementary use of Sentinel 1 and Sentinel 2 data in two speciﬁc zones of the study area (A) and (B). Areas detected only by Sentinel 2 are in blue, only by Sentinel 1 in orange and areas detected as burnt by both sensors are in dark red. Active ﬁres (MODIS) are represented by the black dots. Figure 10. Examples of complementary use of Sentinel 1 and Sentinel 2 data in two specific zones of the study area (A) and (B). Areas detected only by Sentinel 2 are in blue, only by Sentinel 1 in orange and areas detected as burnt by both sensors are in dark red. Active fires (MODIS) are represented by the black dots. Figure 10. Examples of complementary use of Sentinel 1 and Sentinel 2 data in two speciﬁc zones of the study area (A) and (B). Areas detected only by Sentinel 2 are in blue, only by Sentinel 1 in orange and areas detected as burnt by both sensors are in dark red. Active ﬁres (MODIS) are represented by the black dots. 4.3. Spatial Distribution of the Burnt Areas In reviewing the spatial distribution of the fires in the forest domain, it is clear that the fires d i l i f t ifi ll i th M t f t F th Clearly, the combination of the Sentinel data with that of Landsat would further improve the temporal coverage. 17 of 22 Remote Sens. 2016, 8, 986 17 of 22 4.3. Spatial Distribution of the Burnt Areas From the analysis of a the VHR image together with the timing of the ﬁre derived from S2 burnt area maps and MODIS active ﬁre data (see Figure 9), the hypothesis can be made that ﬁres started close to roads, both in logged and non-logged forests, or close to a savanna, but all ﬁres started in very open forests, then propagated to the border of the ecosystem and into denser canopy cover forests. From the S2 and MODIS data, it is clear that ﬁre started at the same time in different locations. Wind and dry vegetation may well have facilitated the spread of the ﬁre to such a widespread extent. There is, as yet, no ﬁrm explanation for the origin of these ﬁres, however, they are unlikely to be of natural origin (i.e., lightning) [10] and more likely to be related to the presence of local communities and indigenous people in the forest. Human activities have probably initiated ﬁre in an environment more prone to ﬁre after the drought of the beginning of the year. The ﬁres observed along the main road could be related to shifting cultivation activities by the villagers living there. Poaching activities are known to occur in the area, given that fauna such as gorillas and elephants are present in the open forests. However, to conﬁrm the origin of the ﬁres, a more detailed accessibility study and surveys with local people are needed. 4.4. Long Term Ecological Implications? The actual impact of the ﬁres on the tree mortality, tree cover density and longer term effects are difﬁcult to assess at this stage. However, a ﬁeld visit to one of the affected areas three months after the ﬁres showed that the ﬂames had reached the crowns of the trees, some 20–30 m high. The regrowth of the Marantaceae was already clearly visible as illustrated in Figure 11. Marantaceae, being pioneer plants, easily invade new open areas (following natural or anthropogenic disturbances). Doing so, they prevent the regeneration of trees and lead to an opening of the forest canopy cover [35]. 18 of 22 e forest 18 of 22 he forest Remote Sens. 2016, 8, 986 disturbances). Doing [35] Figure 11. Regrowth of Marantaceae in an area affected by fire (photo taken on 29 May 2016). Figure 11. Regrowth of Marantaceae in an area affected by ﬁre (photo taken on 29 May 2016). Figure 11. Regrowth of Marantaceae in an area affected by fire (photo taken on 29 May 2016). Figure 11. Regrowth of Marantaceae in an area affected by ﬁre (photo taken on 29 May 2016). The intrusion of fire into the borders of the Marantaceae forests can present a risk in terms of densification of this herbaceous species into the understory of denser forest formations. Cuni-Sanchez et al., [39] have documented that Marantaceae forests may be maintained by occasional fire events. Specific attention in the coming years should be put in following the vegetation regeneration in the transition zones between Marantaceae forests and other forests types. Most species in closed canopy evergreen forests have thin bark and are more vulnerable to fires The intrusion of ﬁre into the borders of the Marantaceae forests can present a risk in terms of densiﬁcation of this herbaceous species into the understory of denser forest formations. Cuni-Sanchez et al., [39] have documented that Marantaceae forests may be maintained by occasional ﬁre events. Speciﬁc attention in the coming years should be put in following the vegetation regeneration in the transition zones between Marantaceae forests and other forests types. Most species in closed canopy evergreen forests have thin bark and are more vulnerable to fires than drier forests [8,62]. It is known that fires can lead to tree mortality up to three years after the fire event [63], so the full impact of this episode cannot yet be determined. 4.4. Long Term Ecological Implications? Most species in closed canopy evergreen forests have thin bark and are more vulnerable to ﬁres than drier forests [8,62]. It is known that ﬁres can lead to tree mortality up to three years after the ﬁre event [63], so the full impact of this episode cannot yet be determined. Field survey would be important to complement the satellite study and to understand better what exactly has burnt (trees, Marantaceae herbs), and also to monitor in time the vegetation regeneration. This would be important to determine what would happen in case of a recurring scenario of ﬁres in this ecosystem, a major ecological question—if El Niño events are to become more frequent and access to the forest continues to increase. According to climatic scenarios, length and intensity of dry seasons might indeed increase in the future in Central Africa [64]. Concerning accessibility, roads are considered as a proximate driver of ﬁre [61]. In the study area, a rapid road network expansion has occurred in north of Republic of Congo since 1999 [65]. In Amazonia, a greater fragmentation of the forest has been shown to decrease the resilience of the forest to typical dry seasons, predisposing the fragmented forest to ﬁres [66,67]. The impact of logging on the propensity to ﬁre of such formations needs to be assessed. A review of ﬁres outbreaks and impacts in managed vs. non-managed concessions could help shed some light on such impacts. In theory, well-designed and fully implemented sustainable forest management plans reduce the collateral damage to the forests, especially through improved road-planning, effective road closure and post-harvesting monitoring, and as a result limit their susceptibility to ﬁre. The crux of the ﬁre problem in tropical rainforests is not so much the introduction of ﬁre into these ecosystems but the frequency with which it returns [8]. It will therefore be important in the coming years to see if this event reoccurs. A repetition of this event, together with assessing the proportion of trees affected and their associated carbon emissions is also of concern for REDD+ implications. 5. Conclusions This paper demonstrates the potential for S2 and S1 data to detect and monitor ﬁre outbreaks in forest ecosystems at ﬁner temporal and spatial resolution than was previously possible. Burnt areas totalizing an area of 36,000 ha were detected in the north of the Republic of Congo thanks to the Remote Sens. 2016, 8, 986 Remote Sens. 2016, 8, 986 19 of 22 combined used of S1 and S2 time series. At its peak in early 2016, the ﬁre was spreading at a rate of 1600 ha per day. Indications of the forthcoming event were in evidence in late December, both from the high detection of active ﬁres in the region and the low SPI rainfall values. combined used of S1 and S2 time series. At its peak in early 2016, the ﬁre was spreading at a rate of 1600 ha per day. Indications of the forthcoming event were in evidence in late December, both from the high detection of active ﬁres in the region and the low SPI rainfall values. The ﬁfteen S2 images analyzed from December 2015 to May 2016 represent the ﬁrst time series of S2 data released over the study area and show a great potential for tropical forest monitoring applications. The burnt area maps derived from each S2 image are adding a precise spatial extent to the point location of the MODIS active ﬁre product and provide a better documentation of the ﬁre propagation in the Marantaceae forests. The two burnt areas maps derived from S1 bi-monthly composites provides a gap ﬁlling capacity (in case of cloud cover) and brings conﬁdence to the S2 results, with over 60% concurrence between the two data sets. The arrival of the S1 and S2 satellites, soon supported by the Sentinel 3 SLSTR sensor with bands optimized for ﬁre monitoring, can provide forest managers and national agencies with a rapid and accurate assessment of ﬁre events. From the methodology proposed here, some points have to be addressed to move from a case study to an operational monitoring tool. The combination of the radar and optical information is robust in terms of burnt area detections but the respective S1 and S2 methodologies should evolve into algorithms that are independent from locally tuned thresholds. 5. Conclusions A review of climatic data and vegetation susceptibility to ﬁre could also be employed to highlight anomalies which may lead to ﬁre outbreaks and provide managers with an early warning of the possibility of such events. Of concern to all parties, forest managers, national agencies, etc., is the long-term perspective for these ecosystems which have been subjected to this exceptional ﬁre event. Is this just a one-off event that in fact helps to maintain these open forests, or is this a precursor to more frequent revisits of ﬁre events which may induce a change in the landscape of Central African forests and their future uses? Acknowledgments: The authors would like to thank Tom Van Loon from Interholco AG for his comments about the ﬁre outbreak in the Ngombé logging concession. The authors would also like to thank the team from Interholco AG working in the Ngombé logging concession for the making a ﬁeld visit possible. Author Contributions: All authors contributed to the design of the experiment. The analyses with Sentinel 2 data were mainly conducted by A.V., the analyses with Sentinel 1 and climatic data by G.C., and the analysis of the MODIS active ﬁres product as well as the processing of the Very High Resolution image by H.E. The Methodology and Results Sections of the manuscript were mainly written by A.V., G.C. and H.E. The rest of the authors mainly contributed to the Discussion and Conclusion Sections. Conﬂicts of Interest: The authors declare no conﬂict of interest. 5. Mayaux, P.; Pekel, J.F.; Desclée, B.; Donnay, F.; Lupi, A.; Achard, F.; Clerici, M.; Bodart, C.; Brink, A.; Nasi, R.; et al. State and evolution of the African rainforests between 1990 and 2010. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 2013, 368. [CrossRef] [PubMed] References 1. National Oceanic and Atmospheric Administration (NOAA). NOAA Climate Prediction Center, El Niño Southern Oscillation (ENSO) Indices. Available online: www.cpc.ncep.noaa.gov/products/analysis_ monitoring/ensostuff/ensoyears.shtml (accessed on 12 July 2016). 2. Interholco. Effets de El Niño à l´IFO. Available online: http://www.interholco.ch/News.3344.0.html (accessed on 7 April 2016). 2. Interholco. Effets de El Niño à l´IFO. Available online: http://www.interholco.ch/News.3344.0.html (accessed on 7 April 2016). 3. Achard, F.; Eva, H.D.; Stibig, H.J.; Mayaux, P.; Gallego, J.; Richards, T.; Malingreau, J.P. Determination of deforestation rates of the world’s humid tropical forests. Science 2002, 297, 999–1002. [CrossRef] [PubMed] 4. Achard, F.; Beuchle, R.; Mayaux, P.; Stibig, H.J.; Bodart, C.; Brink, A.; Carboni, S.; Desclée, B.; Donnay, F.; Eva, H.D.; et al. Determination of tropical deforestation rates and related carbon losses from 1990 to 2010. Glob. Chang. Biol. 2014, 2010, 1–15. [CrossRef] [PubMed] g 5. Mayaux, P.; Pekel, J.F.; Desclée, B.; Donnay, F.; Lupi, A.; Achard, F.; Clerici, M.; Bodart, C.; Brink, A.; Nasi, R.; et al. State and evolution of the African rainforests between 1990 and 2010. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 2013, 368. [CrossRef] [PubMed] 20 of 22 20 of 22 Remote Sens. 2016, 8, 986 6. Bayol, N.; Demarquez, B.; de Wasseige, C.; Eba’a Atyi, R.; Fisher, J.F.; Nasi, R.; Pasquier, A.; Rossi, X.; Steil, M.; Vivien, C. Forest management and the timber sector in Central Africa. In The Forests of the Congo Basin: State of the Forest 2010; de Wasseige, C., de Marcken, P., Bayol, N., Hiol, F., Mayaux, P., Desclée, B., Nasi, R., Billand, A., Defourny, P.E., Eba’a Atyi, R., Eds.; Publications Ofﬁce of the European Union: Luxembourg, 2012; pp. 43–62. 7. Cochrane, M.A. Tropical Fire Ecology: Climate Change, Land Use, and Ecosystem Dynamics; Springer: Berlin/Heidelberg, Germany, 2009; p. 644. g y p 8. Cochrane, M.A. Fire science for rainforests. Nature 2003, 421, 913–919. [CrossRef] [PubMed] 9. Laurance, W.F.; Williamson, G.B. Positive feedbacks among forest fragmentation, drought, and climate change in the Amazon. Conserv. Biol. 2001, 15, 1529–1535. [CrossRef] 10. Siegert, F.; Ruecker, G.; Hinrichs, A.; Hoffmann, A.A. Increased damage from ﬁres in logged forests during droughts caused by El Niño. Nature 2001, 414, 437–440. [CrossRef] [PubMed] 11. Wooster, M.J.; Perry, G.L.W.; Zoumas, A. Fire, drought and El Niño relationships on Borneo (southeast Asia) in the pre-MODIS era (1980–2000). Biogeosciences 2012, 9, 317–340. [CrossRef] 12. References Nepstad, D.C.; Verssimo, A.; Alencar, A.; Nobre, C.; Lima, E.; Lefebvre, P.; Schlesinger, P.; Potter, C.; Moutinho, P.; Mendoza, E.; et al. Large-scale impoverishment of Amazonian forests by logging and ﬁre. Nature 1999, 398, 505–508. [CrossRef] 13. Cochrane, M.A.; Alencar, A.; Schulze, M.D.; Souza, C.M.; Nepstad, D.C.; Lefebvre, P.; Davidson, E.A.; Saldarriaga, J.G.; West, D.C.; Sanford, R.L.; et al. Positive feedbacks in the ﬁre dynamic of closed canopy tropical forests. Science 1999, 284, 1832–1835. [CrossRef] [PubMed] 14. Mouillot, F.; Schultz, M.G.; Yue, C.; Cadule, P.; Tansey, K.; Ciais, P.; Chuvieco, E. Ten years of global burned area products from spaceborne remote sensing—A review: Analysis of user needs and recommendations for future developments. Int. J. Appl. Earth Obs. Geoinf. 2014, 26, 64–79. [CrossRef] 15. Alonso-Canas, I.; Chuvieco, E. Global burned area mapping from ENVISAT-MERIS and MODIS active ﬁre data. Remote Sens. Environ. 2015, 163, 140–152. [CrossRef] 16. Padilla, M.; Stehman, S.V.; Ramo, R.; Corti, D.; Hantson, S.; Oliva, P.; Alonso-Canas, I.; Bradley, A.V.; Tansey, K.; Mota, B.; et al. Comparing the accuracies of remote sensing global burned area products using stratiﬁed random sampling and estimation. Remote Sens. Environ. 2015, 160, 114–121. [CrossRef] p g 17. Eva, H.; Lambin, E.F. Fires and land-cover change in the tropics: A remote sensing analysis at the landscape scale. J. Biogeogr. 2000, 27, 765–776. [CrossRef] 18. Boschetti, L.; Roy, D.P.; Justice, C.O.; Humber, M.L. MODIS–Landsat fusion for large area 30m burned area mapping. Remote Sens. Environ. 2015, 161, 27–42. [CrossRef] 19. Siegert, F.; Ruecker, G. Use of multitemporal ERS-2 SAR images for identiﬁcation of burned scars in South-East Asian tropical rainforest. Int. J. Remote Sens. 2000, 21, 831–837. [CrossRef] 20. Page, S.E.; Siegert, F.; Rieley, J.O.; Boehm, H.D.; Jaya, A.; Limin, S. The amount of carbon released from peat and forest ﬁres in Indonesia during 1997. Nature 2002, 420, 61–65. [CrossRef] [PubMed] 21. Gimeno, M.; San-Miguel-Ayanz, J.; Schmuck, G. Identiﬁcation of burnt areas in Mediterranean forest environments from ERS-2 SAR time series. Int. J. Remote Sens. 2004, 25, 4873–4888. [CrossRef] 22. Malenovský, Z.; Rott, H.; Cihlar, J.; Schaepman, M.E.; García-Santos, G.; Fernandes, R.; Berger, M. Sentinels for science: Potential of Sentinel-1, -2, and -3 missions for scientiﬁc observations of ocean, cryosphere, and land. Remote Sens. Environ. 2012, 120, 91–101. [CrossRef] 23. White, F. The Vegetation of Africa. References A Descriptive Memoir to Accompany the Unesco/AEFTFAT/UNSO Vegetation Map of Africa; United Nations Educational, Scientiﬁc and Cultural Organization (UNESCO): Paris, France, 1983. 24. Bégué, L. Les forêts du nord de la République du Congo (Brazzaville). Bois. For. Trop. 1967, 63–76. 25. Philippon, N.; de Lapparent, B.; Gond, V.; Sèze, G.; Martiny, N.; Camberlin, P.; Cornu, G.; Morel, B.; Moron, V.; Bigot, S.; et al. Analysis of the diurnal cycles for a better understanding of the mean annual cycle of forests greenness in Central Africa. Agric. For. Meteorol. 2016, 223, 81–94. [CrossRef] 26. Gond, V.; Fayolle, A.; Pennec, A.; Cornu, G.; Mayaux, P.; Camberlin, P.; Doumenge, C.; Fauvet, N.; Gourlet-Fleury, S. Vegetation structure and greenness in Central Africa from MODIS multi-temporal data. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 2013, 368. [CrossRef] [PubMed] 21 of 22 21 of 22 Remote Sens. 2016, 8, 986 27. World Resources Institute (WRI) and the Ministère de l’Economie Forestière et du Développement Durable de la République du Congo (MEFDD). Interactive Forest Atlas of the Republic of the Congo—Version 3.0; WRI: Washington, DC, USA, 2011. 8. AfriPop. WorldPop Website. Available online: http://www.worldpop.org.uk/ (accessed 30 August 2016). 29. Linardw, C.; Gilbert, M.; Snow, R.W.; Noor, A.M.; Tatem, A.J.; Carruthers, R.; Krishnamani, R.; Murray, S.; Porter, G.; Roberts, P.; et al. Population distribution, settlement patterns and accessibility across Africa in 2010. PLoS ONE 2012, 7. [CrossRef] [PubMed] 30. Kleinschroth, F.; Gourlet-Fleury, S.; Sist, P.; Mortier, F.; Healey, J.R. Legacy of logging roads in the Congo Basin: How persistent are the scars in forest cover? Ecosphere 2015, 6. [CrossRef] 31. Kleinschroth, F.; Healey, J.R.; Gourlet-Fleury, S. Sparing forests in Central Africa: Re-use old logging roads to avoid creating new ones. Front. Ecol. Environ. 2016, 14, 9–10. [CrossRef] 32. Climate Hazards Group InfraRed Precipitation with Station Data. Available online: http://chg.geog.ucsb. edu/data/chirps/ (accessed on 15 April 2016). 33. Funk, C.; Peterson, P.; Landsfeld, M.; Pedreros, D.; Verdin, J.; Shukla, S.; Husak, G.; Rowland, J.; Harrison, L.; Hoell, A.; et al. The climate hazards infrared precipitation with stations—A new environmental record for monitoring extremes. Sci. Data 2015, 2. [CrossRef] [PubMed] 34. Mckee, T.B.; Doesken, N.J.; Kleist, J. The relationship of drought frequency and duration to time scales. In Proceedings of the 8th Conference on Applied Climatology, Anaheim, CA, USA, 17–22 January 1993; American Meteorological Society (AMS): Boston, MA, USA, 1993; pp. 179–184. g y pp 35. Gillet, J.F. References Les forêts à Marantaceae au sein de la Mosaïque Forestière du Nord de la République du Congo: Origines et Modalités de Gestion. Ph.D. Thesis, Université de Liège, Liège, Belgium, 2013. 36. Van de Weghe, J.P. Forêts d’Afrique Centrale, La Nature et l’Homme; Edition Lannoo SA: Tielt, Belgium, 2004. 36. Van de Weghe, J.P. Forêts d’Afrique Centrale, La Nature et l’Homme; Edition Lannoo SA: Tielt, Belgium, 2004. 37 Male J Giresse P Doumenge C Fa ier C Comment on “intensif ing weathering and land use in Iron 36. Van de Weghe, J.P. Forêts d Afrique Centrale, La Nature et l Homme; Edition Lannoo SA: Tielt, Belgium, 2004. 37. Maley, J.; Giresse, P.; Doumenge, C.; Favier, C. Comment on “intensifying weathering and land use in Iron Age Central Africa”. Science 2012, 337. [CrossRef] [PubMed] 37. Maley, J.; Giresse, P.; Doumenge, C.; Favier, C. Comment on “ Age Central Africa”. Science 2012, 337. [CrossRef] [PubMed] 38. Devos, C.; Sanz, C.; Morgan, D.; Onononga, J.R.; Laporte, N.; Huynen, M.C. Comparing ape densities and habitats in northern Congo: Surveys of sympatric gorillas and chimpanzees in the Odzala and Ndoki regions. Am. J. Primatol. 2008, 70, 439–451. [CrossRef] [PubMed] 39. Cuni-Sanchez, A.; White, L.J.T.; Calders, K.; Jeffery, K.J.; Abernethy, K.; Burt, A.; Disney, M.; Gilpin, M.; Gomez-Dans, J.L.; Lewis, S.L. African savanna-forest boundary dynamics: A 20-year study. PLoS ONE 2016, 11. [CrossRef] [PubMed] 40. Fire Information for Resource Management System (FIRMS) MODIS Fire Archive Download Web Page. Available online: https://ﬁrms.modaps.eosdis.nasa.gov/download/request.php (accessed on 15 April 2016). 41. Giglio, L.; Descloitres, J.; Justice, C.O.; Kaufman, Y.J. An enhanced contextual ﬁre detection algorithm for MODIS. Remote Sens. Environ. 2003, 87, 273–282. [CrossRef] 42. European Space Agency (ESA). Sentinel-2 Delivers First Images. Available online: http://www.esa.int/ Our_Activities/Observing_the_Earth/Copernicus/Sentinel-2/Sentinel-2_delivers_ﬁrst_images (accessed on 23 August 2016). 43. Drusch, M.; del Bello, U.; Carlier, S.; Colin, O.; Fernandez, V.; Gascon, F.; Hoersch, B.; Isola, C.; Laberinti, P.; Martimort, P.; et al. Sentinel-2: ESA’s optical high-resolution mission for GMES operational services. Remote Sens. Environ. 2012, 120, 25–36. [CrossRef] 44. Gao, B.C. NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sens. Environ. 1996, 58, 257–266. [CrossRef] 45. Silva, J.M.N.; Sá, A.C.L.; Pereira, J.M.C. Comparison of burned area estimates derived from SPOT-VEGETATION and Landsat ETM+ data in Africa: Inﬂuence of spatial pattern and vegetation type. Remote Sens. Environ. 2005, 96, 188–201. [CrossRef] 46. Tucker, C.J. References Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 1979, 8, 127–150. [CrossRef] 47. European Space Agency (ESA). Sentinel-1: ESA’s Radar Observatory Mission for GMES Operational Services; ESA Communications: Noordwijk, The Netherlands, 2012. 48. Menges, C.H.; Bartolo, R.E.; Bell, D.; Hill, G.J.E. The effect of savanna ﬁres on SAR backscatter in northern Australia. Int. J. Remote Sens. 2004, 25, 4857–4871. [CrossRef] 22 of 22 Remote Sens. 2016, 8, 986 49. Huang, S.; Siegert, F. Backscatter change on ﬁre scars in Siberian boreal forests in ENVISAT ASAR wide-swath images. IEEE Geosci. Remote Sens. Lett. 2006, 3, 154–158. [CrossRef] 50. Bourgeau-Chavez, L.L.; Harrell, P.A.; Kasischke, E.S.; French, N.H.F. The detection and mapping of Alaskan wildﬁres using a spaceborne imaging radar system. Int. J. Remote Sens. 1997, 18, 355–373. [CrossRef] 51. Bourgeau-Chavez, L.L.; Kasischke, E.S.; Brunzell, S.; Mudd, J.P.; Tukman, M. Mapping ﬁre scars i boreal forests using imaging radar data. Int. J. Remote Sens. 2002, 23, 4211–4234. [CrossRef] 51. Bourgeau Chavez, L.L.; Kasischke, E.S.; Brunzell, S.; Mudd, J.P.; Tukman, M. Mapping ﬁre scars in global boreal forests using imaging radar data. Int. J. Remote Sens. 2002, 23, 4211–4234. [CrossRef] 52. Ulaby, F.T.; Sarabandi, K.; McDonald, K.; Whitt, M.; Dobson, M.C. Michigan microwave canopy scattering model. Int. J. Remote Sens. 1990, 11, 1223–1253. [CrossRef] 53. Kurum, M. C-band SAR backscatter evaluation of 2008 Gallipoli forest ﬁre. IEEE Geosci. Remote Sens. Lett. 2015, 12, 1091–1095. [CrossRef] 54. Main, R.; Mathieu, R.; Kleynhans, W.; Wessels, K.; Naidoo, L.; Asner, G. Hyper-temporal C-band SAR for baseline woody structural assessments in deciduous savannas. Remote Sens. 2016, 8. [CrossRef] 54. Main, R.; Mathieu, R.; Kleynhans, W.; Wessels, K.; Naidoo, L.; Asner, G. Hyper temporal C band SAR fo baseline woody structural assessments in deciduous savannas. Remote Sens. 2016, 8. [CrossRef] 55. Kalogirou, V.; Ferrazzoli, P.; Della Vecchia, A.; Foumelis, M. On the SAR backscatter of burned forests: A model-based study in C-band, over burned pine canopies. IEEE Trans. Geosci. Remote Sens. 2014, 52, 6205–6215. [CrossRef] 56. Tanase, M.A.; Santoro, M.; Aponte, C.; de la Riva, J. Polarimetric properties of burned forest areas at C- and L-band. IEEE J. Sel. Topics Appl. Earth Obs. Remote Sens. 2014, 7, 267–276. [CrossRef] 57. Kasischke, E.S.; Bourgeau-Chavez, L.L.; French, N.H.F.; Harrell, P.; Christensen, N.L., Jr. Initial observations on using SAR to monitor wildﬁre scars in boreal forests. Int. J. Remote Sens. 1992, 13, 3495–3501. [CrossRef] 58. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). References Lopes, A.; Nezry, E.; Touzi, R.; Laur, H. Maximum a posteriori speckle ﬁltering and ﬁrst order texture models in SAR images. In Proceedings of the 10th Annual International Symposium on Geoscience and Remote Sensing, College Park, MD, USA, 20–24 May 1990; IEEE: Piscataway, NJ, USA, 1990; pp. 2409–2412. 59. Immitzer, M.; Vuolo, F.; Atzberger, C. First experience with Sentinel-2 data for crop and tree species classiﬁcations in Central Europe. Remote Sens. 2016, 8. [CrossRef] 60. Radoux, J.; Chomé, G.; Jacques, D.; Waldner, F.; Bellemans, N.; Matton, N.; Lamarche, C.; d’Andrimont, R.; Defourny, P. Sentinel-2’s potential for sub-pixel landscape feature detection. Remote Sens. 2016, 8. [CrossRef] 61. Laurance, W.F.; Clements, G.R.; Sloan, S.; O’Connell, C.S.; Mueller, N.D.; Goosem, M.; Venter, O.; Edwards, D.P.; Phalan, B.; Balmford, A.; et al. A global strategy for road building. Nature 2014, 513, 229–232. [CrossRef] [PubMed] 62. Uhl, C.; Kauffman, J.B. Deforestation, ﬁre susceptibility, and potential tree responses to ﬁre in the eastern Amazon. Ecology 1990, 71, 437–449. [CrossRef] 63. Barlow, J.; Peres, C.A.; Lagan, B.O.; Haugaasen, T. Large tree mortality and the decline of forest biomass following Amazonian wildﬁres. Ecol. Lett. 2002, 6, 6–8. [CrossRef] 64. Zelazowski, P.; Malhi, Y.; Huntingford, C.; Sitch, S.; Fisher, J.B. Changes in the potential distribution of humid tropical forests on a warmer planet. Philos. Trans. Ser. A Math. Phys. Eng. Sci. 2011, 369, 137–160. [CrossRef] [PubMed] 65. Kleinschroth, F.; Healey, J.R.; Gourlet-Fleury, S.; Mortier, F.; Stoica, R.S. Effects of logging on roadless space in intact forest landscapes of the Congo Basin. Conserv. Biol. 2016. [CrossRef] [PubMed] 66. Briant, G.; Gond, V.; Laurance, S.G.W. Habitat fragmentation and the desiccation of forest canopies: A case study from eastern Amazonia. Biol. Conserv. 2010, 143, 2763–2769. [CrossRef] 67. Cochrane, M.A.; Laurance, W.F. Fire as a large-scale edge effect in Amazonian forests. J. Trop. Ecol. 2002, 18, 311–325. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2992910907	https://www.nature.com/articles/s41598-019-55015-y.pdf	English	null	Lesion Network Localization of Seizure Freedom following MR-guided Laser Interstitial Thermal Ablation	Scientific reports	2,019	cc-by	8,868	Lesion Network Localization of Seizure Freedom following MR- guided Laser Interstitial Thermal Ablation Karim Mithani 1, Alexandre Boutet 2,3, Jurgen Germann 2, Gavin J. B. Elias2, Karim Mithani 1, Alexandre Boutet 2,3, Jurgen Germann 2, Gavin J. B. Elias2, Alexander G. Weil4, Ashish Shah5, Magno Guillen 6, Byron Bernal6, Justin K. Achua 5, John Ragheb5, Elizabeth Donner7, Andres M. Lozano8, Elysa Widjaja9 & George M. Ibrahim 10,11* , , g , , Alexander G. Weil4, Ashish Shah5, Magno Guillen 6, Byron Bernal6, Justin K. Achua 5, h h b5 li b h 7 d 8 l idj j 9 b hi g Alexander G. Weil4, Ashish Shah5, Magno Guillen 6, Byron Bernal6, Justin K. Achua 5, John Ragheb5, Elizabeth Donner7, Andres M. Lozano8, Elysa Widjaja9 & George M. Ibrahim Treatment-resistant epilepsy is a common and debilitating neurological condition, for which neurosurgical cure is possible. Despite undergoing nearly identical ablation procedures however, individuals with treatment-resistant epilepsy frequently exhibit heterogeneous outcomes. We hypothesized that treatment response may be related to the brain regions to which MR-guided laser ablation volumes are functionally connected. To test this, we mapped the resting-state functional connectivity of surgical ablations that either resulted in seizure freedom (N = 11) or did not result in seizure freedom (N = 16) in over 1,000 normative connectomes. There was no difference seizure outcome with respect to the anatomical location of the ablations, and very little overlap between ablation areas was identified using the Dice Index. Ablations that did not result in seizure-freedom were preferentially connected to a number of cortical and subcortical regions, as well as multiple canonical resting-state networks. In contrast, ablations that led to seizure-freedom were more functionally connected to prefrontal cortices. Here, we demonstrate that underlying normative neural circuitry may in part explain heterogenous outcomes following ablation procedures in different brain regions. These findings may ultimately inform target selection for ablative epilepsy surgery based on normative intrinsic connectivity of the targeted volume. Epilepsy is a common and debilitating neurological illness, affecting nearly 1% of the world’s population1. Up to 40% of patients are refractory to medications and may benefit from surgical intervention2. It remains unclear why some patients with focal epilepsy undergoing nearly identical resective procedures have different outcomes. For example, patients with temporal lobe epilepsy are significantly more likely to achieve seizure freedom than those with extratemporal epilepsies, by virtue of the location of the lesion, irrespective of the underlying epileptogenic pathology3–7. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y Lesion Network Localization of Seizure Freedom following MR- guided Laser Interstitial Thermal Ablation Recent studies have leveraged a technique known as lesion network mapping to identify connectomes associ- ated with specific conditions and diseases, based on the assumption that heterogeneous lesions in differing loca- tions that result in a shared phenotype can be localized to specific brain networks8. That is, disruption of intrinsic brain connectivity at any given node of a network will result in a clinical phenotype that is more attributable to the network itself than the individual brain lesion. This method has been successfully used to map the network basis of a variety of neuropsychiatric conditions, including akinetic mutism, alien limb, and free will, using normative data from healthy populations8–11. 1Faculty of Medicine, University of Toronto, Toronto, ON, Canada. 2University Health Network, Toronto, ON, Canada. 3Joint Department of Medical Imaging, University of Toronto, Toronto, ON, Canada. 4Division of Neurosurgery, CHU- Ste Justine, Université de Montréal, Montréal, Canada. 5Division of Neurosurgery, Brain Institute, Nicklaus Children’s Hospital, Miami, USA. 6Department of Radiology, Nicklaus Children’s Hospital, Miami, USA. 7Division of Neurology, Hospital for Sick Children, Toronto, Canada. 8Division of Neurosurgery, Department of Surgery, Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada. 9Department of Diagnostic Imaging, Hospital for Sick Children, Toronto, Canada. 10Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Canada. 11Division of Neurosurgery, Hospital for Sick Children, Department of Surgery, University of Toronto, Toronto, Canada. *email: george.ibrahim@sickkids.ca Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ Epilepsy may similarly be viewed as a disorder of large-scale brain networks, rather than a particular region of the brain12–15. Several studies have suggested that structural, functional, and electrophysiological connectivity of purported epileptogenic zones can portend post-operative seizure outcome16–25. Notably, a recent study identified epileptogenic zones and networks on resting state fMRI (rs-fMRI) using Independent Component Analysis that correlate strongly with zones identified with gold-standard intracranial-EEG recordings26. Surgical destruction of all rs-fMRI-identified epileptogenic zones resulted in significantly better seizure outcomes than ablation or resection of a single area. Furthermore, post-operative disruption of these rs-fMRI networks by ≥80% – with postoperative normalization of the patient’s rs-fMRI scans was significantly associated with seizure outcomes27. Methods bl Laser ablations. Twenty-seven children undergoing MRgLITT for refractory epilepsy at three separate institutions – The Hospital for Sick Children (Toronto, Canada), Centre Hospitalier Universitaire Sainte-Justine (Montréal, Canada), and Nicklaus Children’s Hospital (Miami, USA) – were included. At one-year post-operative follow-up, eleven children satisfied the criteria for Engel Class I and were labelled as “seizure-free” (SF), while the remaining sixteen were identified as “not-seizure-free” (NSF). This study involves the use of de-identified retrospective data, complies with the principles outlined in the Declaration of Helsinki, and was approved by the Research Ethics Board at the Hospital for Sick Children.l p Details of the MRgLITT procedure have been previously reported28,29. Briefly, a Leksell stereotactic frame is placed and a thin-cut CT or CT-angiogram scan is obtained and merged with pre-operative MRI scans to map frame-based coordinates. Depending on the institution, a Medtronic Visualase® (Minneapolis, MN) or Monteris (Neurablate) laser fiber is then guided to the target using a predetermined trajectory. The patient is then brought to an MRI suite where thermal ablations are performed. Ablative doses (65–85% of maximum power) are always preceded by lower energy test doses (35% of maximum power). The procedure is completed under real-time MRI-guidance, ensuring that a complete ablation of the hypothesized epileptogenic region is achieved in each case. g g p yp p p g g Details regarding image acquisition and processing have also been reported previously30. Briefly, high-resolution anatomical T1-weighted images (0.86 × 0.86 × 0.86 mm isotropic voxels) were collected on a 3 T scanner with a single channel transmit/receive head coil (Philips, the Netherlands), before and after the operation during the same anesthetic. The ablation sites for all children were identified and delineated on de-identified post-operative scans using semiautomatic segmentation with ITK-SNAP v1.4.131. Gadolinium contrast was administered in post-operative scans to facilitate robust demarcation of the ablation volume. Segmentations were manually and independently inspected by two authors to ensure accuracy. Importantly, these scans were de-identified prior to segmentation and all individuals involved in generating and reviewing them were blinded to patient outcomes. Next, the segmented ablation masks were linearly then non-linearly transformed to stand- ard space using the MNI-152 standard-space T1-weighted average structural template (2 × 2 × 2 mm). These registrations were completed using the FMRIB’s Linear Image Registration Tool (FLIRT) and then the FMRIB’s Non-linear Image Registration Tool (FNIRT), with nearest neighbor interpolation. Lesion Network Localization of Seizure Freedom following MR- guided Laser Interstitial Thermal Ablation In this study, the value of resting-state connectivity was highlighted by the finding that 97% of participants with post-operative rs-fMRI normalization became seizure-free compared to only 3% without rs-fMRI normalization.i p p p y Converging findings therefore suggest that the functional connectivity of hypothesized epileptogenic zones can influence their amenability to surgical treatment, and in turn the resulting postoperative outcomes. In the current study, we sought to investigate the relation between normative brain regions to which epileptogenic vol- umes are connected and seizure outcomes following ablation of these volumes. To achieve this, we characterized differences in intrinsic functional connectivity of surgical ablations that do and do not result in seizure freedom. Focal surgical ablation volumes were seeded in a large normative rs-fMRI data-set to map networks associated with post-surgical seizure-freedom. The relationships between ablation volume location/connectivity and clini- cal outcomes following Magnetic Resonance guided laser interstitial thermal ablation therapy (MRgLITT) were studied. This procedure results in minimal disruption of brain tissue and allows the quantification of the spatial extent of a therapeutic lesion with high fidelity. Networks associated with seizure response following MRgLITT were analyzed in order to better understand the association between the intrinsic functional connectivity of focal epileptogenic pathologies and outcomes following surgical ablation. Methods bl Although the ideal method for analyzing neuroimaging data from young adults and children is controversial, the use of the standardized MNI- 152 templated and associated atlases is an accepted and widely employed approach30,32–35. p p y p y pp In order to measure the extent to which the segmented ablation masks overlapped with each other, Dice indi- ces were computed. Dice indices evaluate the extent of overlap of a binarized volume with any other binarized volume, as follows: Dice Index volume of overlap volume volume 2 ( ) ( #1) ( #2) = ∗ + Dice Index volume of overlap volume volume 2 ( ) ( #1) ( #2) = ∗ + Dice Index volume of overlap volume volume 2 ( ) ( #1) ( #2) = ∗ + Lesion network mapping. In order to perform lesion network mapping and identify neural networks con- nected to the ablation volume that portended seizure outcome, each segmented lesion mask was used to seed a normative rs-fMRI dataset assembled from the 1000 healthy Brain Genomics Superstruct Project (https://data- verse.harvard.edu/dataverse/GSP)36, similar to prior analyses37,38 (in-house MATLAB script, The MathWorks, Inc., Version R2018a. Natick, MA, USA). A connectivity r-map was thus obtained for each individual lesion mask (Fig. 1). To identify brain regions with significant functional connectivity to the ablation volume, these maps were then converted to t-maps and thresholded by t = 5.1; this corrected for multiple comparisons (Bonferroni corrections) across the whole brain at pcor < 0.0538. The statistically significant thresholded maps were then bina- rized to obtain meaningful spatial patterns of connectivity associated with each ablation volume. Conscious of the Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ Figure 1. Lesion network mapping methodology and workflow. Segmented lesions were used as seed voxels in a normative dataset of resting-state functional MRI in healthy subjects. Individual connectivity maps were thresholded (p < 0.05) to identify meaningful voxels, binarized to inspect spatial patterns, and summed for SF and NSF groups. The summed maps were then compared to yield voxelwise odds ratios. Figure 1. Lesion network mapping methodology and workflow. Segmented lesions were used as seed voxels in a normative dataset of resting-state functional MRI in healthy subjects. Individual connectivity maps were thresholded (p < 0.05) to identify meaningful voxels, binarized to inspect spatial patterns, and summed for SF and NSF groups. The summed maps were then compared to yield voxelwise odds ratios. limitations of normative healthy data, this conservative approach ensured that subsequent analyses assessed only the topography of thresholded (pcor < 0.05) connectivity, rather than the magnitude of correlations themselves, as the latter is known to differ considerably between healthy and epileptic brains39–41. The thresholded (pcor < 0.05) binarized connectivity masks for each individual were then summed across each group (SF or NSF) to yield overall connectivity patterns associated with lesions that did and did not result in seizure freedom, respectively. A higher voxel value in these summed images indicates a greater number of individual, thresholded connectivity maps overlapping in that particular area of the brain. Dice Index volume of overlap volume volume 2 ( ) ( #1) ( #2) = ∗ + For example, if the binarized connectivity maps of four sep- arate ablations overlapped at a particular voxel, that voxel would have a value of 4 in the summed image. Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ Clinical Variable Seizure Free (N = 11) Not-Seizure-Free (N = 16) P-value Sex 6 Females 10 Females 0.71 5 Males 6 Males Age of surgery 13.09 years (range: 8–40) 18.29 years (range: 12–57) 0.14 Seizure etiology Focal Cortical Dysplasia 2 7 0.49 Tumour 3 2 Tuberous sclerosis 2 3 Hypothalamic hamartoma 3 1 Microgyria 1 2 Unknown 0 1 Lesion location Frontal 3 5 0.88 Parietal 2 3 Temporal 4 7 Hypothalamic 2 1 Lesion volume 3529 mm3 4752 mm3 0.33 Table 1. Clinical and demographic variables of twenty-seven individuals that underwent MRgLITT for intractable epilepsy. Table 1. Clinical and demographic variables of twenty-seven individuals that underwent MRgLITT for intractable epilepsy. These summed connectivity maps for each of the two groups were then used to calculate voxelwise odds ratios (VORs) in order to identify voxels that were more likely associated with lesions that either did or did not result in seizure freedom. VOR maps are generated by comparing the number of lesions that resulted in a particular outcome (e.g. seizure freedom) that are associated with a particular brain region or voxel, to the number of lesions that did not result in that outcome that are also associated with that same region/voxel. This approach has been used in previous lesion mapping studies to identify brain areas associated with risk of a particular outcome, such as central post-stroke pain or adverse effects of focused ultrasound thalamotomy42,43. VOR V N V V N V ( ) ( ) P C C C P P = − − To calculate the VORs for ablations that resulted in seizure freedom, NP is the number of lesions that resulted in seizure freedom, NC is number of lesions that did not result in seizure freedom, VP is the number of lesions that resulted in seizure freedom with thresholded functional connectivity at a specific voxel, and VC is number of lesions that did not result in seizure freedom with thresholded functional connectivity at a specific voxel. Dice Index volume of overlap volume volume 2 ( ) ( #1) ( #2) = ∗ + Conversely, to calculate the VORs for ablations that did not result in seizure freedom, NP is the number of lesions that did not result in seizure freedom, NC is number of lesions that did result in seizure freedom, VP is the number of lesions that did not result in seizure freedom with thresholded functional connectivity at a specific voxel, and VC is number of lesions that did result in seizure freedom with thresholded functional connectivity at a specific voxel. i Finally, VOR maps were individually overlaid onto each component of the Automated Anatomical Labelling (AAL) atlas44 and the Stanford FIND atlas45 to assess mean VORs in specific brain regions and canonical resting-state networks, respectively. The datasets analysed in the current study are available from the correspond- ing author upon request. Results Region SF Mean VOR SD SF Max VOR NSF Mean VOR Right middle frontal gyrus, orbital part 2.11 1.32 5.83 0.73 Left inferior frontal gyrus, pars opercularis 2.30 1.84 8.57 0.78 Right inferior frontal gyrus, pars triangularis 2.10 1.67 8.57 0.87 Table 2. Voxelwise odds ratios for key brain regions functionally associated with seizure freedom, based on the AAL atlas. Table 2. Voxelwise odds ratios for key brain regions functionally associated with seizure freedom, based on the AAL atlas. lesions that both did and did not result in seizure freedom (FDR-corrected p < 0.05). For both SF and NSF abla- tions, multiple lesions had regions of overlapping functional connectivity in the prefrontal cortex, cingulate cor- tex, parietal lobe, and parts of the temporal lobe. NSF ablations demonstrated additional functional connectivity to the bilateral precuneus, hippocampi, temporal lobes and poles, and deep nuclei. Given the significant overlap between functional maps associated with ablations resulting in SF and those associated with NSF, voxelwise odds ratios were computed to compare the likelihood of functional activity in each area being associated with these dichotomous outcomes. Ablation targets that resulted in SF were at least twice as likely to be functionally connected to the right orb- itofrontal cortex (Mean VOR = 2.11, SD = 1.32, Max VOR = 5.83), left pars opercularis (Mean VOR = 2.23, SD = 1.84, Max VOR = 8.57), and left pars triangularis (Mean VOR = 2.09, SD = 1.67, Max VOR = 8.57) (Table 2). Results Clinical data. Amongst 27 individuals that underwent MRgLITT for intractable epilepsy, 11 achieved seizure freedom (SF) and 16 did not (NSF). The SF and NSF groups did not show statistically significant differences in any key clinical variables, including: sex, age, seizure etiology, pre-operative seizure frequency, or prior epilepsy surgeries (Table 1). Lesion characteristics. Ablation targets were identified using multimodal pre-operative investigations aimed at delineating epileptic foci, involving a combination of electrophysiological, metabolic, and structural investigations. The seizure semiologies and epileptogenic zones were variable, with no significant difference between seizure-free and non-seizure-free groups in: temporal vs. extra-temporal localization (p = 0.97), mesial (i.e. mesial frontal, mesial temporal, cingulate, or precuneus) vs. non-mesial localization (p = 0.46), or ablation volume (p = 0.22; Fig. 2A). Specific characteristics of each ablation volume can be found in the Supplementary Material. pi pp y The average dice index was 0.0381(SD: 0.0703) for seizure-free ablations, and 0.0883 (SD: 0.141) for non-seizure-free ablations. Other than the left hippocampus where 3 lesions overlapped, this indicates that there was minimal overlap in the location of the individual ablation masks for either group. Whole-Brain connectivity patterns. A large, state-of-the-art normative rs-fMRI dataset was leveraged to understand epileptogenic network connectivity associated with seizure outcome. Summed connectivity maps demonstrated diffuse, statistically significant functional connectivity between various regions of the brain and Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ Figure 2. (A) Summed ablation masks associated with seizure freedom (SF; red) and non-seizure-freedom (NSF; blue). There is no systematic bias in the localization of ablations in seizure freedom vs. non-freedom groups. The maximum value is 3, in the left hippocampus of the NSF group. (B) Dice coefficients quantifying the degree of overlap between lesions in each group. There is no significant difference in average dice index between the two groups (p = 0.235). Figure 2. (A) Summed ablation masks associated with seizure freedom (SF; red) and non-seizure-freedom (NSF; blue). There is no systematic bias in the localization of ablations in seizure freedom vs. non-freedom groups. The maximum value is 3, in the left hippocampus of the NSF group. (B) Dice coefficients quantifying the degree of overlap between lesions in each group. There is no significant difference in average dice index between the two groups (p = 0.235). www.nature.com/scientificreports/ www.nature.com/scientificreports/ Region NSF Mean VOR SD NSF Max VOR SF Mean VOR Left medial frontal gyrus 2.47 1.29 7.78 0.57 Left medial orbitofrontal cortex 2.18 1.47 12.86 0.61 Left posterior cingulate gyrus 2.23 1.12 3.85 0.64 Left Hippocampus 2.02 1.13 6.00 0.67 Right Hippocampus 2.13 1.23 6.00 0.67 Left Amygdala 2.34 1.31 6.00 0.57 Right Amygdala 2.34 1.21 6.00 0.54 Right superior parietal lobule 2.91 2.04 12.86 0.65 Right inferior parietal lobule 2.00 1.04 7.78 0.70 Left angular gyrus 2.60 1.81 9.90 0.61 Left transverse temporal gyrus (Heschl’s gyrus) 2.06 0.71 3.33 0.55 Left superior temporal gyrus 2.78 1.70 10.00 0.62 Right superior temporal gyrus 2.22 1.61 6.00 0.81 Left middle temporal gyrus 3.02 2.06 13.50 0.64 Left middle temporal pole 4.35 1.35 5.79 0.26 Right middle temporal pole 4.01 1.53 7.50 0.31 Left inferior temporal gyrus 2.47 1.84 7.50 0.87 Right crus I of cerebellum 2.27 1.83 12.86 0.70 Right crus II of cerebellum 2.10 1.16 12.86 0.65 Right lobule IX of cerebellum 2.20 1.76 10.00 0.78 Lobule VII of vermis 2.11 1.49 7.78 0.76 Table 3 Voxelwise odds ratios for key brain regions functionally associated with non-seizure Table 3. Voxelwise odds ratios for key brain regions functionally associated with non-seizure-freedom, based on the AAL atlas. Table 3. Voxelwise odds ratios for key brain regions functionally associated with non-seizure-freedom, based on the AAL atlas. Figure 3. Voxelwise odds ratios of functional activity in various brain regions associated with seizure freedom (A) and non-seizure-freedom (B). Figure 3. Voxelwise odds ratios of functional activity in various brain regions associated with seizure freedo (A) and non-seizure-freedom (B). Results In contrast, ablated regions resulting in NSF were at least twice as likely to be functionally associated with the left orbitofrontal cortex (Mean VOR = 2.18, SD = 1.47, Max VOR = 12.86), left medial frontal gyrus (Mean VOR = 2.47, SD = 1.29, Max VOR = 7.78), left posterior cingulate (Mean VOR = 2.23, SD = 1.12, Max VOR = 3.85), bilateral hippocampi (Mean VOR = 2.08, SD = 1.18, Max VOR = 6.00), bilateral amygdala (Mean VOR = 2.34, SD = 1.26, Max VOR = 6.00), right parietal lobe (Mean VOR = 2.46, SD = 1.53, Max VOR = 12.86), left angular gyrus (Mean VOR = 2.60, SD = 1.81, Max VOR = 9.90), bilateral temporal lobe (Mean VOR = 2.62, SD = 1.80, Max VOR = 13.50), bilateral temporal poles (Mean VOR = 4.18, SD = 1.44, Max VOR = 7.50), and the right cerebellar hemisphere (Mean VOR = 2.19, SD = 1.58, Max VOR = 12.86), amongst others (Table 3). The greatest VOR in the SF group (18.17) was located in the left dorsolateral prefrontal cortex, and the greatest VOR in the NSF group (16.83) was found in the left middle temporal gyrus. Overall, NSF ablations demonstrated a significantly greater spatial distribution of functionally associated areas and larger range of VOR values than SF ablations (Fig. 3). Resting-State connectomic differences between seizure freedom and Non-freedom. In addi- tion to examining specific brain regions functionally associated with SF and NSF ablations, we also assessed differences between SF and NSF cohorts in terms of ablation volume connectivity to canonical resting-state net- works (Table 4). Lesions that do not result in seizure freedom are at least twice as likely to be functionally con- nected to the dorsal default mode network (DMN), the language network, and visuospatial network (Fig. 4B). In contrast, ablation targets that result in seizure freedom are twice as likely to be functionally connected to only the primary visual network (Fig. 4A). Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ Resting state networks associated with seizure-freedom (A) and non-seizure-freedom (B). All voxels are at least twice as likely to be functionally associated with each group. Figure 4. Resting state networks associated with seizure-freedom (A) and non-seizure-freedom (B). All voxels are at least twice as likely to be functionally associated with each group. to surgical ablation are more likely to be connected to a brain network comprising the right orbitofrontal cor- tex, left pars opercularis, and left pars triangularis, as well as nodes of the primary visual network. Conversely, surgical lesions less likely to abolish seizures show increased connectivity with several cortical and subcortical brain regions – most notably in the temporal lobe – as well as nodes of the dorsal DMN, language network, and visuospatial network. These findings are concordant with prior investigations on the lateralization and location of structural and functional networks in epilepsy26,27,46.hif p p y The most striking finding in this analysis is the substantial difference in the spatial extent of intrinsic func- tional connectivity between targeted regions that do and do not result in seizure freedom. Specifically, NSF ablations exhibit robust intrinsic functional connectivity across numerous major cortical and subcortical brain regions, including the nodes of several well-known resting-state networks. In contrast, SF ablations are only func- tionally associated with a few, specific, relatively well-demarcated foci. Furthermore, NSF ablations also exhibit a considerably larger range of the relative strength of the associations – as quantified by voxelwise odds ratios – than SF ablations. Finally, NSF ablations were functionally associated with deep nuclei, including the caudate and thalami, whereas SF ablations were not. Together, these findings suggest that epileptogenic foci located in areas that are constitutively connected to a wider range of brain regions, and therefore part of a relatively more widespread functional network, are less responsive to ablative surgery. Conversely, epileptogenic foci located in areas with more focused intrinsic functional connectivity are more amenable to ablation. This is consistent with the purpose of MRgLITT epilepsy surgeries, which are aimed at treating focal seizures with precise targets, as well as with prior studies documenting a strong association between the extent of resection of a seizure network and surgical outcomes27. A lesion with intrinsically greater spatial extent of functional connectivity is less likely to undergo destruction of its entire functional network, making it less likely to exhibit seizure freedom after MRgLITT ablation. Discussion Intrinsic brain connectivity is increasingly understood to link common clinical phenotypes that manifest follow- ing lesions in different brain regions. Presurgical and resting-state connectivity is also increasingly implicated in seizure-freedom following resective and ablative surgical procedures for epilepsy16,26,27,30,46–48. To better understand the network basis of heterogeneous seizure outcomes following MRgLITT, we performed lesion network mapping to identify interconnected brain networks that portend more favorable outcomes. Indeed, we report notable differ- ences in intrinsic connectivity of the epileptogenic foci that correspond to favourable and unfavourable outcomes following ablative epilepsy surgery. Lesions resulting in seizure freedom are more likely to be functionally connected to parts of the prefrontal cortex. In contrast, lesions that do not abolish seizures are preferentially connected to a far greater number of cortical and subcortical regions, as well as multiple canonical resting-state networks. Towards a “seizure freedom network”. Given the minimal overlap amongst ablation volumes, our find- ings suggest that the outcomes of ablative epilepsy surgery may be associated with the resting-state functional connectivity profile – rather than the location – of the targeted region. Epileptogenic foci that are more amenable Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ www.nature.com/scientificreports/ Network NSF Mean VOR NSF SD SF Mean VOR SF SD Auditory Network 1.51 1.05 1.05 0.65 Basal Ganglia Network 1.35 0.81 1.16 1.01 Left Executive Control Network 1.33 0.84 1.11 0.98 Language Network 2.34 1.66 0.70 0.59 Precuneus Network 1.07 0.71 1.42 1.04 Right Executive Control Network 1.30 0.86 1.29 1.29 Sensorimotor Network 0.95 0.51 1.26 0.47 Visuospatial Network 2.04 1.40 0.84 0.83 Anterior Salience Network 1.78 1.24 0.84 0.60 Dorsal Default Mode Network 2.18 1.11 0.63 0.49 Higher Visual Network 1.02 0.45 1.15 0.44 Posterior Salience Network 1.47 0.89 0.94 0.58 Primary Visual Network 0.55 0.20 2.07 0.92 Ventral Default Mode Network 1.14 1.03 1.41 1.11 Table 4. Voxelwise odds ratios for canonical functional networks, based on the Stanford FIND Atlas. Network NSF Mean VOR NSF SD SF Mean VOR SF SD Auditory Network 1.51 1.05 1.05 0.65 Basal Ganglia Network 1.35 0.81 1.16 1.01 Left Executive Control Network 1.33 0.84 1.11 0.98 Language Network 2.34 1.66 0.70 0.59 Precuneus Network 1.07 0.71 1.42 1.04 Right Executive Control Network 1.30 0.86 1.29 1.29 Sensorimotor Network 0.95 0.51 1.26 0.47 Visuospatial Network 2.04 1.40 0.84 0.83 Anterior Salience Network 1.78 1.24 0.84 0.60 Dorsal Default Mode Network 2.18 1.11 0.63 0.49 Higher Visual Network 1.02 0.45 1.15 0.44 Posterior Salience Network 1.47 0.89 0.94 0.58 Primary Visual Network 0.55 0.20 2.07 0.92 Ventral Default Mode Network 1.14 1.03 1.41 1.11 Table 4. Voxelwise odds ratios for canonical functional networks, based on the Visuospatial Network 2.04 1.40 0.84 0.83 Anterior Salience Network 1.78 1.24 0.84 0.60 Dorsal Default Mode Network 2.18 1.11 0.63 0.49 Higher Visual Network 1.02 0.45 1.15 0.44 Posterior Salience Network 1.47 0.89 0.94 0.58 Primary Visual Network 0.55 0.20 2.07 0.92 Ventral Default Mode Network 1.14 1.03 1.41 1.11 Table 4. Voxelwise odds ratios for canonical functional networks, based on the Stanford FIND Atlas. Figure 4. Resting state networks associated with seizure-freedom (A) and non-seizure-freedom (B). All voxels are at least twice as likely to be functionally associated with each group. Table 4. Voxelwise odds ratios for canonical functional networks, based on the Stanford FIND Atlas. Table 4. Voxelwise odds ratios for canonical functional networks, based on the Stanford FIND Atlas. Figure 4. Resting state networks associated with seizure-freedom (A) and non-seizure-freedom (B). All voxels are at least twice as likely to be functionally associated with each group. Figure 4. Table 4. Voxelwise odds ratios for canonical functional networks, based on the Stanford FIND Atlas. www.nature.com/scientificreports/ lobe is highly epileptogenic and easily kindled, as evidenced by a number of electrophysiological, metabolic, and molecular studies49. Therefore, epileptogenic lesions that are highly functionally connected to parts of the tem- poral lobe may result in more frequent, severe, and/or intractable seizures and ablating such lesions is evidently less likely to result in seizure freedom. Another potential explanation of this phenomenon is that these patients underwent incomplete or insufficient destruction of the entire functional network connected to the epileptogenic lesion, which is known to be associated with poor surgical outcomes27. Both of these explanations emphasize the importance of understanding the functional connectivity of epileptogenic lesions and incorporating this infor- mation into surgical planning and targeting. g g g g In addition, our study highlights the importance of canonical resting state networks in propagating epilepsy, which has been emphasized in prior studies30,50,51. Specifically, our analysis revealed that epileptogenic lesions associated with the primary visual network are more amenable to ablative surgery, whereas those associated with the language network, visuospatial network, and – perhaps most importantly – the DMN, are not. Functional neuroimaging of patients with numerous forms of drug-resistant epilepsy has consistently identified reduced connectivity and hubness of the DMN in epilepsy compared to healthy controls52–55. Furthermore, alterations in DMN activity and connectivity have been reported in focal, idiopathic generalized and absence seizures39,51,56–58. In addition, DMN is widely regarded to subserve stimulus-independent emotional processing, self-referential mental activity, and social cognitive functions; this is in keeping with theories that the network may underlie states of altered consciousness seen in certain types of epilepsy59. Given the central role of the DMN in intrin- sic brain activity, as well as its extensive involvement in mediating various forms of epilepsy, it is possible that increased functional connectivity between its components and epileptogenic lesions is more likely to propagate intractable seizures. Accordingly, lesions in regions that are strongly functionally connected to the DMN would be less amenable to surgical ablation, as demonstrated by our findings.h g yi g The other resting-state networks associated with NSF ablations include the language network and the visu- ospatial network. Notably, several forms of epilepsy – temporal lobe epilepsy in particular – have been associ- ated with impairment of language60,61 and visuospatial functioning62,63. Implications for clinical practice l d d bl Implications for clinical practice. Children who undergo nearly identical ablative surgeries for refrac- tory epilepsy demonstrate considerable heterogeneity in outcomes65–67. The source of this heterogeneity is not fully elucidated, but remains a critically important question. One potential explanation is under-targeting of functional networks associated with epileptogenic lesions, insufficient destruction of which is associated with a relatively poor surgical outcome27. Differences in the intrinsic, normative connectivity of lesions that do and do not respond to surgical ablation can inform the basis of this phenomenon, and further delineate the source of post-surgical htereogeneity after MRgLITT ablation. Characterizing brain connectivity that predisposes to seizure-recurrence following resective or ablative procedures may inform presurgical counseling and potentially guide clinical decision-making. Here, we find that heterogeneity in epilepsy surgery may be explained, at least in part, by the functional networks to which the epileptogenic brain regions are intrinsically connected. This under- scores the importance of incorporating functional connectivity into making decisions about surgical candidacy. Future studies investigating the predictive potential of these connectomes can inform the utility of pre-operative functional neuroimaging, either in each individual patient or in publicly available normative databases, in mini- mally disruptive neurosurgery for epilepsy. www.nature.com/scientificreports/ g Interestingly, many of the ablation volumes that did not result in seizure-freedom were functionally connected to temporal structures. Resection of the temporal lobe in temporal lobe epilepsy has the highest likelihood of seizure freedom. Our findings suggest that ablation of extratemporal seizure foci in locations that would nor- matively be connected to the temporal lobe leads to disappointing outcomes. This may be because the temporal Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ www.nature.com/scientificreports/ This again highlights the aforementioned resistant nature of epileptogenic foci that are highly functionally connected to the temporal lobe, even to surgical ablation.hi The increased functional connectivity of SF lesions to the primary visual network is of unclear significance, and poses an interesting line of further investigation. It does reinforce the finding that lesions that are amenable to surgical ablation have substantially more focal connectivity than those that are not, given the narrow spatial distribution of the primary visual network. The occipital lobe is also thought to be less epileptogenic than other parts of the brain, such as the temporal lobe, and occipital seizures are relatively less common than many other epilepsy syndromes64. Therefore, ablations with greater intrinsic functional connectivity to the primary visual network, as opposed to more widespread networks like the DMN, would be more amenable to surgical ablation, as demonstrated by our findings. www.nature.com/scientificreports/ www.nature.com/scientificreports/ using specialized MRI hardware and acquisition parameters, which may result in a more reliable and repro- ducible connectivity patterns than native resting-state fMRI. Furthermore, as previously mentioned, we were cognizant of this limitation and opted to use a stringent statistical threshold to establish meaningful connec- tivity maps. Moreover, we only examined spatial patterns of functional connectivity, rather than attempting to identify individual brain regions associated with seizure outcomes. This is a more conservative approach and is less likely to result in spurious conclusions. Importantly, the purpose of this study was to elucidate the functional connectivity associated with surgical ablations that did and did not respond to seizure freedom, and the results cannot be used to infer causality. y In sum, these results suggest that variability in surgical outcomes can be explained at least in part by the functional connectivity of targeted lesions, even in the absence of differences in gross lesion location, seizure characteristics, and demographic factors. Furthermore, analyzing healthy brains is valuable in identifying true functional connectomes as they are not exposed to disease-related aberrancies and compensatory changes. The current study also pose several new lines of investigation to better understand and clinically translate the connec- tomics of ablative surgery for epilepsy. Eventually, these and future findings may be useful in predicting whether lesions in specific locations are likely to abolish seizures based on the neuronal networks to which they are intrin- sically connected. Received: 9 July 2019; Accepted: 22 November 2019; Published: xx xx xxxx Published: xx xx xxxx References Holmes, M. D. & Tucker, D. M. Identifying the epileptic network. Front. Neurol. 4, 84 (2013). F. & Vulliemoz, S. Functional brain networks in epilepsy. Curr. Opi 14. Bernhardt, B. C., Bonilha, L. & Gross, D. W. Network analysis for a network disorder: The emerging role of graph theory in the of epilepsy. Epilepsy Behav. 50, 162–170 (2015). p p y p p y , ( ) 15. Kramer, M. A. & Cash, S. S. Epilepsy as a Disorder of Cortical Network Organization. Neurosci. 18, 360–372 (2012). 15. Kramer, M. A. & Cash, S. S. Epilepsy as a Disorder of Cortical N 16. He, X. et al. Presurgical thalamic “hubness” predicts surgical outcome in temporal lobe epilepsy. Neurology 88, 22 (2017). 7. Bonilha, L. et al. Presurgical connectome and postsurgical seizure control in temporal lobe epilepsy. Neurology 81, 1704–1710 (2013). 18. Liao, W. et al. Functional Connectome before and following Temporal Lobectomy in Mesial Temporal Lobe Epilepsy. Sci. Rep. 6, 23153 (2016).h ( ) 9. Bernhardt, B. C., Chen, Z., He, Y., Evans, A. C. & Bernasconi, N. Graph-Theoretical Analysis Reveals Disrupted Small-World Or niz ti n f C rti l Thi kn C rr l ti n N t rk in T mp r l L b Epil p C b C t 21 2147 2157 (2011) 19. Bernhardt, B. C., Chen, Z., He, Y., Evans, A. C. & Bernasconi, N. Graph-Theoretical Analysis Reveals Disrupted Small-World Organization of Cortical Thickness Correlation Networks in Temporal Lobe Epilepsy. Cereb. Cortex 21, 2147–2157 (2011). gh p p p y 0. Negishi, M., Martuzzi, R., Novotny, E. J., Spencer, D. D. & Constable, R. T. Functional MRI connectivity as a predictor of the surgica outcome of epilepsy. Epilepsia 52, 1733–1740 (2011). 21. Zhang, Z. et al. Altered spontaneous neuronal activity of the default-mode network in mesial temporal lobe epilepsy. Brain Res. 1323, 152–160 (2010).t 22. Morgan, V. L., Gore, J. C. & Abou-Khalil, B. Functional epileptic network in left mesial temporal lobe epilepsy detected using resting fMRI. Epilepsy Res. 88, 168–178 (2010). 3. Bettus, G. et al. Role of resting state functional connectivity MRI in presurgical investigation of mesial temporal lobe epilepsy. J Neurol. Neurosurg. Psychiatry 81, 1147–1154 (2010). g y y 4. RamachandranNair, R. et al. MEG Predicts Outcome Following Surgery for Intractable Epilepsy in Children with Normal o Nonfocal MRI Findings. Epilepsia 48, 149–57 (2007). 25. Lee, H. W. et al. References 1. Fiest, K. M. et al. Prevalence and incidence of epilepsy: A systematic review and meta-analysis of international studies. Neurology 88, 296–303 (2017). 2. French, J. A. Refractory Epilepsy: Clinical Overview. Epilepsia 48, 3–7 (2007). 3. Delev, D. et al. Surgical Treatment of Extratemporal Epilepsy: Results and Prognostic Factors. Neurosurgery 84, 242–252 (2019 4 H ák á P t l L t t d di t f ti i i ti t ith f t t t lev, D. et al. Surgical Treatment of Extratemporal Epilepsy: Results g y g g y 4. Hanáková, P. et al. Long-term outcome and predictors of resective surgery prognosis in patients with refractory extratempora epilepsy. Seizure 23, 266–273 (2014).i 5. Hardy, S. G. et al. Factors predicting outcome of surgery for intractable epilepsy with pathologically verified mesial tempora sclerosis. Epilepsia 44, 565–8 (2003). p p 6. Vakharia, V. N. et al. Getting the best outcomes from epilepsy surgery. Ann. Neurol. 83, 676–690 (2018).ft 7. Bell, G. S. et al. Factors affecting seizure outcome after epilepsy surgery: an observational series. J. Neurol. Neurosurg. Psych 933–940 (2017). 7. Bell, G. S. et al. Factors affecting seizure outcome after epilepsy surgery: an observational series. J. Neurol. Neurosurg. Psychiatry 88 933–940 (2017). 8 F M D M i S t t B i N t k ith th H C t N E l J M d 379 2237 2245 (2018) 933–940 (2017). 8 Fox M D Mapping Symptoms to Brain Networks with the Human Connectome N Engl J Med 379 2237 2245 (2018) 933 940 (2017). 8 Fox M D Mapping Symptoms to Brain Networks with the Human Connectome N Engl J Med 379 2237–2245 (2018) 8. Fox, M. D. Mapping Symptoms to Brain Networks with the Human Connectome. N. Engl. J. Med. 379, 2237–2245 (2018). 9. Joutsa, J., Horn, A., Hsu, J. & Fox, M. D. Localizing parkinsonis J , J , , , , J , g p k , ( ) 10. Darby, R. R., Horn, A., Cushman, F. & Fox, M. D. Lesion network localization of criminal behavior. Proc. Natl. Acad. Sci. 115, 601–606 (2018). 1. Darby, R. R., Laganiere, S., Pascual-Leone, A., Prasad, S. & Fox, M. D. Finding the imposter: brain connectivity of lesions causing delusional misidentifications. Brain 140, 497–507 (2017). i 12. Holmes, M. D. & Tucker, D. M. Identifying the epileptic network. Front. Neurol. 4, 84 (2013). i 12. Limitations and Conclusionh The purpose of ablative or resective epilepsy surgery is to obliterate the epileptogenic zone from the brain, and from any large-scale networks it may be a part of. Accordingly, one interpretation of post-operative seizure recur- rence is that the epileptogenic zone was not adequately identified or ablated. However, each of these surger- ies was performed with the explicit intention of ablating the entire epileptogenic focus, as verified by real-time MRI-guidance. Given the curative intent of these operations, our findings suggest that in children who are NSF, intrinsic connectivity networks are more likely to be intimately connected to the ablation focus and together comprise the “epileptogenic zone”. Therefore, although post-operative seizure recurrence may be in part a result of inadequate ablation, our analysis strongly implicates differences in intrinsic connectivity of targeted regions in seizure outcomes. Importantly, this is concordant with previous studies demonstrating that insufficient or incom- plete destruction of seizure networks connected to an epileptogenic zone, in addition to the lesion itself, are less likely to result in freedom for seizures27.h y The purpose of this study was to characterize the resting-state functional connectivity of lesions that result in distinct seizure outcomes after ablative surgery, in an attempt to understand clinical heterogeneity. The analyses were all conducted in non-diseased brains, using normative data from 1000 healthy individuals. Functional connectivity between certain brain regions differ in the diseased state, limiting the direct translata- bility of our findings68. However, normative connectivity datasets are derived from a large number of subjects Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y References Mapping seizure pathways in the temporal lobe. 49. McIntyre, D. C. & Gilby, K. L. Mapping seizure pathways in the 0. van Houdt, P. J. et al. Are Epilepsy-Related fMRI Components Dependent on the Presence of Interictal Epileptic Discharges in Scalp EEG? Brain Topogr. 28, 606–618 (2015).f 51. Fahoum, F., Zelmann, R., Tyvaert, L., Dubeau, F. & Gotman, J. Epileptic discharges affect the default mode network–FMRI and intracerebral EEG evidence. PLoS One 8, e68038 (2013). 52. Maccotta, L. et al. Impaired and facilitated functional networks in temporal lobe epilepsy. NeuroImage Clin. 2, 862–872 (2013) l d l l f b k h b l l l b l l 52. Maccotta, L. et al. Impaired and facilitated functional networks 53. Lee, K. et al. Disruption, emergence and lateralization of brain network hubs in mesial temporal lobe epilepsy. NeuroImage Clin 71–84 (2018). 4. Liao, W. et al. Default mode network abnormalities in mesial temporal lobe epilepsy: A study combining fMRI and DTI. Hum. Brai Mapp. 32, 883–895 (2011). pp 55. Cataldi, M., Avoli, M. & de Villers-Sidani, E. Resting state networks in temporal lobe epilepsy. Epilepsia 54, 2048–2059 (2013) y p y p p y g ( 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epilepsy Behav. 23, 353–9 (2012). 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epi 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epilepsy Behav. 23, 353–9 (2012). 58 K B P t l R d d d f lt d t k ti it i t t t i t t idi thi li d il E il i 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epilepsy Behav. 23, 353–9 (2012). 58 Kay B P et al Reduced default mode network connectivity in treatment resistant idiopathic generalized epilepsy Epilepsia 5 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epilepsy Behav. 23, 353 9 (2012). 58. Kay, B. P. et al. Reduced default mode network connectivity in treatment-resistant idiopathic generalized epilepsy. Epilepsia 54 57. McGill, M. L. et al. Default mode network abnormalities in idiopathic generalized epilepsy. Epilepsy Behav. 23, 353 9 (2012). 58. Kay, B. P. et al. Reduced default mode network connectivity in treatment-resistant idiopathic generalized epilepsy. Epilepsia 54 461 70 (2013) 58. References Altered functional connectivity in seizure onset zones revealed by fMRI intrinsic connectivity. Neurology 83, 2269–2277 (2014). 26. Boerwinkle, V. L. et al. Correlating Resting-State Functional Magnetic Resonance Imaging Connectivity by Independent Component Analysis-Based Epileptogenic Zones with Intracranial Electroencephalogram Localized Seizure Onset Zones and Surgical Outcomes in Prospective Pediatric Intractable Epilepsy Study. Brain Connect. 7, 424–442 (2017). Component Analysis-Based Epileptogenic Zones with Intracranial Electroencephalogram Localized Seizure Onset Zones and Surgical Outcomes in Prospective Pediatric Intractable Epilepsy Study. Brain Connect. 7, 424–442 (2017). 7. Boerwinkle, V. L. et al. Network‐targeted approach and postoperative resting‐state functional magnetic resonance imaging are associated with seizure outcome. Ann. Neurol. 86, ana.25547 (2019). 8. Jermakowicz, W. J. et al. Laser thermal ablation for mesiotemporal epilepsy: Analysis of ablation volumes and trajectories. Epilepsia 58, 801–810 (2017).hh 29. Cajigas, I. et al. MRI-Guided Laser Interstitial Thermal Therapy for Mesial Temporal Epilepsy: A Case Series Analysis of Outcomes and Complications at Two-Year Follow-Up. World Neurosurg, https://doi.org/10.1016/j.wneu.2019.03.057 (2019).t p p g p g j 30. Ibrahim, G. M. et al. Presurgical hyperconnectivity of the ablation volume is associated w resonance-guided laser interstitial thermal therapy. Seizure 61, 89–93 (2018). 0. Ibrahim, G. M. et al. Presurgical hyperconnectivity of the ablation volume is associated with seizure-freedom after magnetic resonance-guided laser interstitial thermal therapy. Seizure 61, 89–93 (2018).ifi g py , ( ) 31. Yushkevich, P. A. et al. User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability. Neuroimage 31, 1116–1128 (2006). Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y www.nature.com/scientificreports/ 2. Mankinen, K. et al. Connectivity disruptions in resting-state functional brain networks in children with temporal lobe epilepsy Epilepsy Res. 100, 168–178 (2012). 32. Mankinen, K. et al. Connectivity disruptions in resting-state functional brain networks in children with temporal lobe epilepsy. Epilepsy Res. 100, 168–178 (2012). 33 Supekar K Musen M & Menon V Development of Large Scale Functional Brain Networks in Children PLoS Biol 7 e1000157 p p y 3. Supekar, K., Musen, M. & Menon, V. Development of Large-Scale Functional Brain Networks in Children. PLoS Biol. 7, e1000157 (2009). ( ) 34. Ibrahim, G. M. et al. Presurgical thalamocortical connectivity is associated with response to vagus nerve stimulation in children with intractable epilepsy. NeuroImage. Clin. 16, 634–642 (2017).h p p y g 5. Ibrahim, G. M. et al. Thalamocortical connectivity is enhanced following functional hemispherotomy for intractable lateralized epilepsy. Epilepsy Behav. Competing interestsh p g The authors declare no competing interests. p g The authors declare no competing interests. References Hum. Brain Mapp. 35, 5686–5700 (2014). 68. Ibrahim, G. M. et al. Impaired development of intrinsic connectivity networks in children with medically intractable localiza related epilepsy. Hum. Brain Mapp. 35, 5686–5700 (2014). References Kay, B. P. et al. Reduced default mode network connectivity in treatment-resistant idiopathic generalized epilepsy. Epileps 461–70 (2013).h 59. Raichle, M. E. The Brain’s Default Mode Network. Annu. Rev. Neurosci. 38, 433–447 (2015).h h ( ) 60. Davey, D. & Thompson, P. Interictal language functioning in chronic epilepsy. J. Neurolinguistics 6, 381–399 (1991) h 60. Davey, D. & Thompson, P. Interictal language functioning in chronic epilepsy. J. Neurolinguistics 6, 381–399 (1991). 62. Lv, Z. et al. Alteration of functional connectivity within visuospatial working memory-related brain network in patient temporal lobe epilepsy: A resting state fMRI study Epilepsy Behav 35 64 71 (2014) 62. Lv, Z. et al. Alteration of functional connectivity within visuospatial working memory-related brain network in patients with right temporal lobe epilepsy: A resting-state fMRI study. Epilepsy Behav. 35, 64–71 (2014). 62. Lv, Z. et al. Alteration of functional connectivity within visuospatial working memory-rel temporal lobe epilepsy: A resting-state fMRI study. Epilepsy Behav. 35, 64–71 (2014). 62. Lv, Z. et al. Alteration of functional connectivity within visuospatial working memory related brain network in pa temporal lobe epilepsy: A resting-state fMRI study. Epilepsy Behav. 35, 64–71 (2014). y p g y temporal lobe epilepsy: A resting-state fMRI study. Epilepsy Behav. 35, 64–71 (2014). emporal lobe epilepsy: A resting-state fMRI study. Epilepsy Behav. 63. Yang, H. et al. Brain network alteration in patients with temporal lobe epilepsy with cognitive impairment. Epilepsy Behav. 81, 4 (2018).fii 64. Taylor, I., Scheffer, I. E. & Berkovic, S. F. Occipital epilepsies: identification of specific and newly recognized syndromes. Brain 126, 753–769 (2003). ( ) 65. Liu, S.-Y. et al. Clinical outcomes and quality of life following surgical treatment for refractory epilepsy: a systematic review and meta-analysis. Medicine (Baltimore). 94, e500 (2015).h h g y g y 67. Devlin, A. M. et al. Clinical outcomes of hemispherectomy for epilepsy in childhood and adolescence. Brain 126, 556–566 (2003). 68. Ibrahim, G. M. et al. Impaired development of intrinsic connectivity networks in children with medically intractable localization- related epilepsy. Hum. Brain Mapp. 35, 5686–5700 (2014). h 67. Devlin, A. M. et al. Clinical outcomes of hemispherectomy for epilepsy in childhood and adolescence. Brain 126, 556–566 (20 b h l d d l f k h ld h d ll bl l l p y p p y 68. Ibrahim, G. M. et al. Impaired development of intrinsic connectivity networks in children with medically intractable localization- related epilepsy. Author contributions K.M., E.W. and G.M.I. conceived of the study. K.M., A.B., J.G., J.A., A.S. and G.B.E. designed and performed analyses. K.M., A.G.W., M.G., B.B., J.R. and G.M.I. wrote the manuscript and drafted figures and tables. E.D., A.M.L. and G.M.I. contributed to and reviewed the manuscript. K.M., E.W. and G.M.I. conceived of the study. K.M., A.B., J.G., J.A., A.S. and G.B.E. designed and performed analyses. K.M., A.G.W., M.G., B.B., J.R. and G.M.I. wrote the manuscript and drafted figures and tables. E.D., A.M.L. and G.M.I. contributed to and reviewed the manuscript. References 51, 281–285 (2015).h p p y p p y 36. Yeo, B. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011). l d d b l k d l ( ) 37. Horn, A. et al. Connectivity Predicts deep brain stimulation outcome in Parkinson disease. Ann. Neurol. 82, 67–78 (2017).i 38. Darby, R. R., Joutsa, J. & Fox, M. D. Network localization of heterogeneous neuroimaging findings. Brain 142, 70–79 (2019). i 9. Luo, C. et al. Altered functional connectivity in default mode network in absence epilepsy: A resting-state fMRI study. Hum. Brain Mapp. 32, 438–449 (2011). 40. Luo, C. et al. Disrupted Functional Brain Connectivity in Partial Epilepsy: A Resting-State fMRI Study. PLoS One 7, e28196 (2012). ( ) 41. Pittau, F., Grova, C., Moeller, F., Dubeau, F. & Gotman, J. Patterns of altered functional connectivity in mesial temporal lobe epilepsy. Epilepsia 53, 1013–1023 (2012). p p 2. Sprenger, T. et al. Assessing the risk of central post-stroke pain of thalamic origin by lesion mapping. Brain 135, 2536–2545 (2012) 42. Sprenger, T. et al. Assessing the risk of central post-stroke pain of thalamic origin by lesion mapping. Brain 135, 2536–2545 (2012). 43. Boutet, A. et al. Focused ultrasound thalamotomy location determines clinical benefits in patients with essential tremor. Brain 141, 3405–3414 (2018). 4. Tzourio-Mazoyer, N. et al. Automated Anatomical Labeling of Activations in SPM Using a Macroscopic Anatomical Parcellation o the MNI MRI Single-Subject Brain. Neuroimage 15, 273–289 (2002). g j g 5. Shirer, W. R., Ryali, S., Rykhlevskaia, E., Menon, V. & Greicius, M. D. Decoding Subject-Driven Cognitive States with Whole-Brain Connectivity Patterns. Cereb. Cortex 22, 158–165 (2012).ii 46. Mithani, K. et al. Connectomic Profiling Identifies Responders to Vagus Nerve Stimulation. Ann. Neurol. ana.25574, https://doi. org/10.1002/ana.25574 (2019). g 7. Boerwinkle, V. L. et al. Subcentimeter epilepsy surgery targets by resting state functional magnetic resonance imaging can improve outcomes in hypothalamic hamartoma. Epilepsia 59, 2284–2295 (2018). yp p p 48. Boerwinkle, V. L., Wilfong, A. A. & Curry, D. J. Resting-State Functional Connectivity by Independent Component Analysis-Based Markers Corresponds to Areas of Initial Seizure Propagation Established by Prior Modalities from the Hypothalamus. Brain Connect. 6, 642–651 (2016). , ( ) 49. McIntyre, D. C. & Gilby, K. L. Mapping seizure pathways in the temporal lobe. Epilepsia 49, 23–30 (2008). D. C. & Gilby, K. L. Additional information Additional information Correspondence and requests for materials should be addressed to G.M.I. Correspondence and requests for materials should be addressed to G.M.I. Correspondence and requests for materials should be addressed to G.M.I. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y 10 www.nature.com/scientificreports nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © Th A th ( ) 2019 www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2019 © The Author(s) 2019 Scientific Reports \| (2019) 9:18598 \| https://doi.org/10.1038/s41598-019-55015-y
https://openalex.org/W2884347753	https://europepmc.org/articles/pmc5904189?pdf=render	English	null	Corrigendum: Precision Medicine and Global Health: The Good, the Bad, and the Ugly	Frontiers in medicine	2,018	cc-by	516	Correction published: 11 April 2018 doi: 10.3389/fmed.2018.00095 Correction published: 11 April 2018 doi: 10.3389/fmed.2018.00095 Approved by: Frontiers in Medicine Editorial Office, Frontiers, Switzerland 1 The Johns Hopkins University, Advanced Academic Programs (AAP), Baltimore, MD, United States, 2 Department of Microbiology, University Hospital of Larissa, University of Thessaly, Larissa, Greece, 3 Department of Neurology, University Hospital of Larissa, University of Thessaly, Larissa, Greece Correspondence: Alexios-Fotios A. Mentis amentis1@jhu.edu Keywords: precision medicine, genomics, sequencing, big data, global health, health policy, health equity, social determinants of health †Present address: Alexios-Fotios A. Mentis, Department of Microbiology, University Hospital of Larissa, University of Thessaly, Larissa, Greece Copyright © 2018 Mentis, Pantelidi, Dardiotis, Hadjigeorgiou and Petinaki. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Alexios-Fotios A. Mentis1†, Kleoniki Pantelidi2, Efthimios Dardiotis3, Georgios M. Hadjigeorgiou3 and Efthimia Petinaki2 Alexios-Fotios A. Mentis1*†, Kleoniki Pantelidi2, Efthimios Dardiotis3, Georgios M. Hadjigeorgiou3 and Efthimia Petinaki2 Specialty section: Specialty section: This article was submitted to Precision Medicine, a section of the journal Frontiers in Medicine Specialty section: This article was submitted to Precision Medicine, a section of the journal Frontiers in Medicine Received: 20 March 2018 Accepted: 23 March 2018 Published: 11 April 2018 In the affiliations section, the phrase “American Academy of Pediatrics (AAP)” was mistakenly placed, and it has now been replaced with the correct phrase “Advanced Academic Programs (AAP).” The authors sincerely apologize for the error.hi h This error does not change the scientific conclusions of the Opinion Article in any way. The original article has been updated. Received: 20 March 2018 Accepted: 23 March 2018 Published: 11 April 2018 The original article has been updated. The original article has been updated. A corrigendum on †Present address: Alexios-Fotios A. Mentis, Department of Microbiology, University Hospital of Larissa, University of Thessaly, Larissa, Greece Precision Medicine and Global Health: The Good, the Bad, and the Ugly by Mentis AA, Pantelidi K, Dardiotis E, Hadjigeorgiou GM, Petinaki E. Front Med (2018) 5:67. doi: 10.3389/fmed.2018.00067. h g y by Mentis AA, Pantelidi K, Dardiotis E, Hadjigeorgiou GM, Petinaki E. Front Med (2018) 5:67. doi: 10.3389/fmed.2018.00067. In the original Opinion article, there was an error in the affiliation address of the first and cor- responding author.fi In the original Opinion article, there was an error in the affiliation address of the first and cor- responding author.fi Citation: Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Mentis AA, Pantelidi K, Dardiotis E, Hadjigeorgiou GM and Petinaki E (2018) Corrigendum: Precision Medicine and Global Health: The Good, the Bad, and the Ugly. Front. Med. 5:95. doi: 10.3389/fmed.2018.00095 April 2018 \| Volume 5 \| Article 95 Frontiers in Medicine \| www.frontiersin.org
https://openalex.org/W2115499704	https://europepmc.org/articles/pmc2701417?pdf=render	English	null	The effect of dietary red palm oil on the functional recovery of the ischaemic/reperfused isolated rat heart: the involvement of the PI3-Kinase signaling pathway	Lipids in health and disease	2,009	cc-by	6,602	Published: 29 May 2009 Lipids in Health and Disease 2009, 8:18 doi:10.1186/1476-511X-8-18 This article is available from: http://www.lipidworld.com/content/8/1/18 © 2009 Engelbrecht et al; licensee BioMed Central Ltd. g ; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BioMed Central BioMed Central Lipids in Health and Disease Open Access The effect of dietary red palm oil on the functional recovery of the ischaemic/reperfused isolated rat heart: the involvement of the PI3-Kinase signaling pathway Address: 1Department of Physiological Sciences, University of Stellenbosch, Stellenbosch, 7600, South Africa, 2Department of Medical Physiology, University of Stellenbosch, Stellenbosch, 7600, South Africa, 3Cardiovascular Research Group, Department of Biochemistry, University of Szeged and Pharmahungary™ Group, Szeged, Hungary and 4Faculty of Health and Wellness Sciences, Cape Peninsula University of Technology, 7535, Bellville, South Africa mail: Anna-Mart Engelbrecht - ame@sun.ac.za; Louise Odendaal - odendaall@cput.ac.za; Eugene F Du Toit - efdt@ ristina Kupai - krisztina.kupai@pharmahungary.com; Tamás Csont - tamas.csont@pharmahungary.com; F di d f di d @ h h J R * j@ Email: Anna-Mart Engelbrecht - ame@sun.ac.za; Louise Odendaal - odendaall@cput.ac.za; Eugene F Du Toit - efdt@sun.ac.za; Kristina Kupai - krisztina.kupai@pharmahungary.com; Tamás Csont - tamas.csont@pharmahungary.com; Peter Ferdinandy - peter.ferdinandy@pharmahungary.com; Jacques van Rooyen* - vanrooyenj@cput.ac.za * Corresponding author Received: 6 March 2009 Accepted: 29 May 2009 Received: 6 March 2009 Accepted: 29 May 2009 Published: 29 May 2009 The effect of RPO and wortmannin (Wn) on FKHR (Ser256) phosphorylation (Figure 4) There was a significant decrease in FKHR phosphorylation in the RPO+Wn group when compared with the RPO group (130.9 ± 3.5 pixels vs 73.00 ± 14.1 pixels; p < 0.05). Background increased % RPP at the same time point, when compared to the control group (65.5 ± 6.3% in control vs 95.0 ± 8.0% in C+Wn; p < 0.05). After 30 minutes of reperfusion, the % RPP in the RPO group was still significant higher compared to the control group (71.03 ± 5.82% in control vs 92.36 ± 4.48% in RPO; p < 0.05). However, at this time point, there was also a significant difference between the RPO and the RPO+Wn groups (92.36 ± 4.48% in RPO vs 75.21 ± 5.23% in RPO+Wn; p < 0.05). Table 1 shows the pre-ischaemic values of heart rate, developed pressure, rate pressure product and coronary flow in control and red palm oil groups. These values were used to calculate the % RPP. g Cardiovascular disease is one of the major causes of death in the Western world. It is believed to account for more than 12 million deaths annually [1]. Although it was pre- viously demonstrated that dietary RPO protects the heart against ischemia/reperfusion-induced injury [2], the pre- cise molecular mechanisms are still unclear. RPO is a natural oil obtained from oil palm fruit (Elaeis guineensis). It is high in palmitic acid (44%) and oleic acid (40%) with natural fat-soluble tocopherol, tocotrienol and carotonoids, which may act as antioixants. Despite the high saturated fat content of RPO, several studies have demonstrated that RPO is associated with better recovery and protection of hearts submitted to ischaemia/reper- fusion [2-4]. RPO-supplementation also caused differen- tial phosphorylation of the MAPKs which were associated with improved functional recovery and reduced apoptosis [3,4]. This indicated that the improved physiological function associated with RPO-supplementation, was due to the cellular signaling effects of RPO, both through the NO-cGMP pathway or the pro-survival Akt pathway. These studies suggest that a combination of carotonoids, lycopene, pro-vitamin E and fatty acids provide more pro- tections than one individual component [5-7]. The effect of RPO and wortmannin (Wn) on the regulatory subunit (p85) of PI3-Kinase (Figure 2) RPO+Wn significantly decreased PI3-K (p85) activity compared to RPO alone (37.9 ± 3.4 pixels vs 59.2 ± 1.8 pixels; p < 0.05). The effect of RPO and wortmannin (Wn) on the regulatory subunit (p85) of PI3-Kinase (Figure 2) RPO+Wn significantly decreased PI3-K (p85) activity compared to RPO alone (37.9 ± 3.4 pixels vs 59.2 ± 1.8 pixels; p < 0.05). The effect of RPO and wortmannin (Wn) on Caspase-3 cleavage (Figure 5) However, as far as we know, no evidence exits for the interaction between RPO and the PI3-K signaling pathway during reperfusion. In order to assess the possible mecha- nisms of protection, the isolated perfused rat heart model with the aid of a PI3-K inhibitor, wortmannin (Wn) was used to assess signaling proteins after RPO-supplementa- tion. Wortmannin significantly increased caspase-3 cleavage compared with the control group (38.21 ± 4.14 pixels vs 56.27 ± 3.63 pixels; p < 0.05). Interestingly, wortmannin also caused a significant increase in caspase-3 cleavage in the RPO group compared with the RPO control group (26.75 ± 2.6 pixels vs 59.79 ± 2.1 pixels; p < 0.05). The effect of RPO and wortmannin (Wn) on PKB/Akt (Ser473) phosphorylation (Figure 3) Wortmannin administration during perfusion caused a significant decrease in PKB/Akt (Ser473) phosphorylation compared to the control group (33.2 ± 10.7 pixels for C+Wn vs 106.4 ± 7.8 pixels for C; p < 0.05). RPO counter- acted this decrease in phosphorylation when the RPO-fed animals received wortmannin during perfusion (33.2 ± 10.7 pixels in C+Wn vs 75.87 ± 10.3 pixels in RPO+Wn; p < 0.05). The serine/threonine protein kinase, protein kinase B (a member of the PI3-K pathway), is a crucial regulator of widely divergent cellular processes including apoptosis (programmed cell death), cell proliferation, differentia- tion and metabolism [8,9]. Several researchers have reported that activation of PI3-K and Akt play an impor- tant role in promoting cardiomyocyte survival and func- tion in models of cardiac injury [10,11]. Therefore, disruption of normal PI3-K signaling pathway during ischaemia/reperfusion-induced injury should therefore be considered as a potential target for therapy. Abstract We have previously shown that dietary red palm oil (RPO) supplementation improves functional recovery in hearts subjected to ischaemia/reperfusion-induced injury. Unfortunately, the cellular and molecular mechanisms responsible for this phenomenon are still poorly understood and no knowledge exists regarding the effects of RPO supplementation on the phosphoinositide 3-kinase (PI3-K) signaling pathway and apoptosis during ischaemia/reperfusion injury. Therefore, the aims of the present study were three fold: (i) to establish the effect of RPO on the functional recovery of the heart after ischaemia/reperfuion injury; (ii) to determine the effect of the PI3-K pathway in RPO-induced protection with the aid of an inhibitor (wortmannin); and (iii) to evaluate apoptosis in our model. Wistar rats were fed a standard rat chow control diet or a control diet plus 7 g RPO/ kg for six weeks. Hearts were excised and mounted on a Langendorff perfusion apparatus. Mechanical function was measured after a 25 min period of total global ischaemia followed by 30 minutes of reperfusion. Hearts subjected to the same conditions were freeze-clamped for biochemical analysis at 10 min during reperfusion to determine the involvement of the PI3-Kinase signaling pathway and apoptosis in our model. Dietary RPO supplementation significantly increased % rate pressure product recovery during reperfusion (71.0 ± 6.3% in control vs 92.36 ± 4.489% in RPO; p < 0.05). The % rate pressure product recovery was significantly reduced when wortmannin was added during perfusion (92.36 ± 4.489% in the RPO group vs 75.21 ± 5.26% in RPO + Wm). RPO + Wm also significantly attenuated PI3-K induction compared with the RPO group (59.2 ± 2.8 pixels in RPO vs 37.9 ± 3.4 pixels in RPO + Wm). We have also demonstrated that PI3-K inhibition induced PARP cleavage (marker of apoptosis) in the hearts during ischaemia/reperfusion injury and that RPO supplementation counteracted this effect. Page 1 of 8 (page number not for citation purposes) Page 1 of 8 (page number not for citation purposes) (page number not for citation purposes) (page number not for citation purposes) http://www.lipidworld.com/content/8/1/18 http://www.lipidworld.com/content/8/1/18 Lipids in Health and Disease 2009, 8:18 Page 2 of 8 (page number not for citation purposes) Percentage Rate Pressure Product Recovery (% RPP) (Figure 1) There was an increase in cleaved PARP in the control+Wn group when compared to the control group (42.12 ± 7.0 pixels vs 98.9 ± 9.3 pixels; p < 0.05. PARP cleavage was sig- nificantly reduced in the RPO+Wn group compared with the control wortmannin group (98.9 ± 9.3 vs 50.23 ± 6.9 pixels; p < 0.05. RPO-supplementation caused an increase in % RPP recov- ery at 10 min during reperfusion when compared with the control group (65.5 ± 6.3% in control vs 93.5 ± 5.2% in RPO; p < 0.05) confirming results in previous similar studies. Our results also indicate that C+Wn had an Page 2 of 8 (page number not for citation purposes) Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 Rate Pressure Product Recovery (%) Figure 1 Rate Pressure Product Recovery (%). The effect of RPO-supplementation and Wortmannin administration on RPP recov- ery during reperfusion (n = 7–10 per group) (mean +/- SEM), (* p < 0.05 for indicated groups). 0 20 40 60 80 100 120 5 min 10 min 30 min Time (Reperfusion) % RPP recovery C (n=7) RPO (n=8) C+Wn (n=7) RPO+Wn (n=10) * * * * Rate Pressure Product Recovery (%) Figure 1 Rate Pressure Product Recovery (%). The effect of RPO-supplementation and Wortmannin administration on RPP recov- ery during reperfusion (n = 7–10 per group) (mean +/- SEM), (* p < 0.05 for indicated groups). In response to ischemia/reperfusion-induced injury, cells activate various signal transduction pathways which either may be harmful or allow adaptation to this stressful environment. There is an increasing body of evidence showing that PKB/Akt may represent a nodal point to coordinate growth factor signaling in the early phase of ischemia/reperfusion-induced injury in the heart [12,13]. However, no knowledge exists about the possible involve- ment of the PI3-K/Akt pathway in RPO-induced protec- tion during ischemia/reperfusion injury in the heart. Therefore, another aim of this study was to determine whether PI3-Kinase inhibition induced changes in protein The effect nin admini jected to i SEM) (* p Figure 4 The effect exerted by RPO supplementation and Wortman nin administration on FKHR phosphorylation in hearts sub jected to ischaemia/reperfusion (n 3 7 per group) (mean ± SEM) ( p 0.05 for indicated groups) Figure 4 The effect exerted by RPO-supplementation and Wortmannin administration on FKHR phosphoryla- tion in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). phosphorylation and apoptosis in hearts subjected to ischemia/reperfusion injury of animals which received dietary RPO supplementation. Upon growth factor activation of receptor tyrosine kinases, PI3-K is recruited to the receptor in the plasma membrane and phosphorylates phosphatidylinositol-4,5- bisphosphate (PIP2) on the 3-OH group, generating phos- phatidylinositol-3,4,5-trisphosphate (PIP3). PI3-kinase is considered one of the intracellular signals responsible for the transmission of anti-apoptotic signals for controlling cell survival. Over expression of PI3-kinase in cells has been shown to cause a significant increase in survival of cells exposed to ionizing radiation [14]. Kennedy and co- workers also reported that inhibition of PI3-K causes an phatidylinositol-3,4,5-trisphosphate (PIP3). PI3-kinase is considered one of the intracellular signals responsible for the transmission of anti-apoptotic signals for controlling cell survival. Over expression of PI3-kinase in cells has been shown to cause a significant increase in survival of cells exposed to ionizing radiation [14]. Kennedy and co- workers also reported that inhibition of PI3-K causes an Page 4 of 8 cells exposed to ionizing radiation [14]. Kennedy and co- workers also reported that inhibition of PI3-K causes an The effect of RPO-supplementation and Wortmannin admin- istration on PKB/Akt (Ser473) phosphorylation in hearts sub- jected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 3 The effect of RPO-supplementation and Wortman- nin administration on PKB/Akt (Ser473) phosphor- ylation in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indi- cated groups). The effect nin admini jected to i SEM) (* p Figure 4 0 20 40 60 80 100 120 140 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PKB/Akt (Ser473) (Pixels) * * P-PKB/Akt Ser473 Total PKB/Akt The effect exerted by RPO-supplementation and Wn admin- istration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 5 The effect exerted by RPO-supplementation and Wn administration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). 0 10 20 30 40 50 60 70 80 90 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups Caspase-3 (Asp175) (Pixels) * * Cleaved caspase-3 0 20 40 60 80 100 120 140 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PKB/Akt (Ser473) (Pixels) * * P-PKB/Akt Ser473 Total PKB/Akt Page 4 of 8 (page number not for citation purposes) The effect exerted by RPO-supplementation and Wn admin- istration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 5 The effect exerted by RPO-supplementation and Wn administration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). 0 10 20 30 40 50 60 70 80 90 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups Caspase-3 (Asp175) (Pixels) * * Cleaved caspase-3 0 10 20 30 40 50 60 70 80 90 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups Caspase-3 (Asp175) (Pixels) * * Cleaved caspase-3 0 20 40 60 80 100 120 140 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PKB/Akt (Ser473) (Pixels) * * The effect istration o jected to i SEM) (* p Figure 3 The effect istration o ischaemia/ < 0.05 for Figure 5 The effect exerted by RPO-supplementation and Wn admin- istration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 5 The effect exerted by RPO-supplementation and Wn administration on the cleavage of caspase-3 in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). Discussion h d We have demonstrated that dietary RPO supplementation offered significant protection against ischemia/reper- fusion-induced injury in the isolated perfused working heart as reflected by improving functional recovery after 10 and 30 minutes of reperfusion. Furthermore, we have demonstrated for the first time that this effect of RPO was significantly attenuated when these hearts were perfused with the PI3-Kinase inhibitor, wortmannin. Therefore, we can conclude that the RPO-induced protection of the heart during ischemia/reperfusion-induced injury is, at least in part, mediated by the PI3-kinase pathway. Page 3 of 8 (page number not for citation purposes) Table 1: Pre-ischaemic values of heart rate, developed pressure, rate pressure product and coronary flow in control and red palm oil groups Heart Rate (beats/min LVDev P (mmHg) Coronary Flow (ml/min) Rate Pressure Product (HR X LVDevP) Control 296.6 ± 13.5 92.4 ± 3.9 16.7 ± 1.8 26314 ± 1973 Control + W 290.0 ± 10.8 75.1 ± 5.7& 12.5 ± 2.3 21778 ± 1940 RPO 312.0 ± 10.7 75.7 ± 3.5$ 17.2 ± 1.0 23630 ± 1330 RPO + W 276.2 ± 13.9 67.2 ± 2.4# 12.2 ± 1.5 19104 ± 1389* heart rate, developed pressure, rate pressure product and coronary flow in control and red palm oil Table 1: Pre-ischaemic values of heart rate, developed pressure, rate pressure product and coronary flow in control and red palm oil groups Heart Rate (beats/min LVDev P (mmHg) Coronary Flow (ml/min) Rate Pressure Product (HR X LVDevP) Control 296.6 ± 13.5 92.4 ± 3.9 16.7 ± 1.8 26314 ± 1973 Control + W 290.0 ± 10.8 75.1 ± 5.7& 12.5 ± 2.3 21778 ± 1940 RPO 312.0 ± 10.7 75.7 ± 3.5$ 17.2 ± 1.0 23630 ± 1330 RPO + W 276.2 ± 13.9 67.2 ± 2.4# 12.2 ± 1.5 19104 ± 1389* Table 1: Pre-ischaemic values of heart rate, developed pressure, rate pressure product and co groups Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 The effect of RPO-supplementation and Wortmannin admin- istration on PI3-K (p85) activity in hearts subjected to ischae- mia/reperfusion (n = 3–7 per group) (mean +/- SEM), (* p < 0.05 for indicated groups) Figure 2 The effect of RPO-supplementation and Wortman- nin administration on PI3-K (p85) activity in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean +/- SEM), (* p < 0.05 for indicated groups). Discussion h d -actin 0 10 20 30 40 50 60 70 80 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PI3-Kinase (p85) (Pixels) * p85 subunit of PI3-K phosphorylation and apoptosis in hearts subjected to ischemia/reperfusion injury of animals which received di RPO l i The effect of RPO-supplementation and Wortmannin admin- istration on PI3-K (p85) activity in hearts subjected to ischae- mia/reperfusion (n = 3–7 per group) (mean +/- SEM), (* p < 0.05 for indicated groups) Figure 2 The effect of RPO-supplementation and Wortman- nin administration on PI3-K (p85) activity in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean +/- SEM), (* p < 0.05 for indicated groups). -actin 0 10 20 30 40 50 60 70 80 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PI3-Kinase (p85) (Pixels) * p85 subunit of PI3-K The effect exerted by RPO-supplementation and Wortman- nin administration on FKHR phosphorylation in hearts sub- jected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 4 The effect exerted by RPO-supplementation and Wortmannin administration on FKHR phosphoryla- tion in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). 0 20 40 60 80 100 120 140 160 180 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups FKHR (Ser256) (Pixels) * P-FKHR -actin The effect exerted by RPO-supplementation and Wortman- nin administration on FKHR phosphorylation in hearts sub- jected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 4 The effect exerted by RPO-supplementation and Wortmannin administration on FKHR phosphoryla- tion in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). 0 20 40 60 80 100 120 140 160 180 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups FKHR (Ser256) (Pixels) * P-FKHR -actin -actin 0 10 20 30 40 50 60 70 80 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PI3-Kinase (p85) (Pixels) * p85 subunit of PI3-K 0 20 40 60 80 100 120 140 160 180 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups FKHR (Ser256) (Pixels) * P-FKHR -actin The effect istration o mia/reperf 0.05 for in Figure 2 pp (p ) y j p ( p g p) ( ) ( p g p ) g The effect of RPO-supplementation and Wortman- nin administration on PI3-K (p85) activity in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean +/- SEM), (* p < 0.05 for indicated groups). The effect exerted by RPO-supplementation and Wortman- nin administration on FKHR phosphorylation in hearts sub- jected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups) Figure 4 The effect exerted by RPO-supplementation and Wortmannin administration on FKHR phosphoryla- tion in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indicated groups). The cleava fusion that administra for indicat Figure 6 g j p pp ( ) ( p g p) ( p g p ) g The cleavage of PARP in hearts subjected to ischae- mia/reperfusion that received RPO-supplementation and Wortmannin administration (mean ± SEM) (n = 3–7 per group) (* p < 0.05 for indicated groups). increase in apoptosis and a decrease in cell survival [15]. PI3-kinases are composed of a catalytic subunit (p110) and a regulatory subunit (p85). PI3-kinase inhibition with wortmannin significantly reduced the regulatory subunit (p85) when compared with the RPO control group. This decrease in PI3-K activity also correlates with reduced function after 30 minutes of perfusion in the iso- lated hearts in the RPO+Wn group compared to the RPO control group. Results in previous studies (2, 3, 4) showed that the signaling pathways were activated early in reper- fusion whilst the true functional effect of these biochemi- cal changes were only observed at later time points. Another hallmark of the apoptotic pathway is the cleavage of caspase-3. Wortmannin caused significant increases in caspase-3 cleavage in the control and in the RPO group, thereby promoting apoptosis in these groups. Further- more, wortmannin also induced a significant increase in PARP cleavage to its proteolyzed products, a phenome- non that is well known to result from caspase-3 activation [19,20]. Interestingly, this increase cleavage of PARP was not observed in the RPO+Wn group. However, it is possi- ble that an increase in this group may be seen if the reper- fusion period is extended. PKB/Akt is one of the most important targets of PI3-K because it phosphorylates and regulates a wide variety of proteins implicated in cell survival/death decisions. Acti- vation of PKB requires binding to PIP3 via the pleckstrin homology domain and phosphorylation of Thr308 in the activation loop as well as phosphorylation of Ser473 within the carboxy-terminal [16]. The present results reveal that wormannin significantly reduced PKB (Ser473) phosphorylation when compared to the control group and that this reduction was partly counteracted in the RPO+Wn group. It was previously demonstrated (3) that PKB could play a role in RPO protection. In the current study the protective effect of RPO was abolished in the RPO+Wn group at 30 minutes of reperfusion. This indi- cates that PI3-K pathway may have had an effect on the RPO-induced protection. The cleava fusion that administra for indicat Figure 6 Although there was no signifi- cant decrease in PI3-K in the RPO group, PI-3K was signif- icantly reduced in the RPO+Wn group. Currently there is no clear evidence that a single substance in the red palm oil is responsible for the protection or effect on the signaling pathways. Previous studies suggest that a combination of carotonoids and vitamin E in the presence of lycopene in a natural food supplement have a far more potent anti-oxidative effect [5-7] than when con- sumed in an isolated form. http://www.lipidworld.com/content/8/1/18 Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 The cleavage of PARP in hearts subjected to ischaemia/reper- fusion that received RPO-supplementation and Wortmannin administration (mean ± SEM) (n = 3–7 per group) (* p < 0.05 for indicated groups) Figure 6 The cleavage of PARP in hearts subjected to ischae- mia/reperfusion that received RPO-supplementation and Wortmannin administration (mean ± SEM) (n = 3–7 per group) (* p < 0.05 for indicated groups). 0 20 40 60 80 100 120 140 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PARP (Asp214) (Pixels) * * Cleaved PARP 0 20 40 60 80 100 120 140 C (n=3) RPO (n=5) C+Wn (n=6) RPO+Wn (n=7) Groups PARP (Asp214) (Pixels) * * Cleaved PARP Although PKB promotes cell survival, the mechanisms involved have only recently begun to emerge. One means by which PKB may promote cell survival is by directly phosphorylating transcription factors that control the expression of pro- and anti-apoptotic genes. PKB appears to both negatively regulate factors that promote the expression of death genes and positively regulate factors that induce survival genes [17,18]. An example is the fam- ily of forkhead transcription factors (FKHR). All the mem- bers of the FKHR family contain a PKB phosphorylation sequence which can be effectively phosphorylated by PKB in vitro [19,20]. Phosphorylation of FKHR by PKB alters its subcellular location. FKHR phosphorylation was signifi- cantly inhibited when RPO+Wn was compared to the RPO group. This leads to forkhead proteins residing pre- dominantly in the nucleus where they are able to promote transcription of pro-apoptotic genes such as Fas-L and Bim through specific DNA elements in their promotor regions [21,22]. This result also correlates with the attenuation in function recovery in the RPO+Wn group compared to the control RPO group. Phosphorylation of FKHR by PKB leads to the export of FKHR from the nucleus and its accu- mulation and sequestration by 14-3-3 proteins in the cytoplasm [22]; thus inhibiting apoptosis. The effect nin admini jected to i SEM) (* p Figure 4 The effect of RPO supplementation and Wortmannin admin istration on PKB/Akt (Ser473) phosphorylation in hearts sub jected to ischaemia/reperfusion (n 3 7 per group) (mean ± SEM) ( p 0.05 for indicated groups) Figure 3 The effect of RPO-supplementation and Wortman- nin administration on PKB/Akt (Ser473) phosphor- ylation in hearts subjected to ischaemia/reperfusion (n = 3–7 per group) (mean ± SEM) (* p < 0.05 for indi- cated groups). Mechanical Function Parameters measured Table 2: The components of Carotino Premium red palm oil (per 100 ml) Total fats 92 g Monounsaturates 43 g Polyunsaturates 12 g Saturates 37 g Trans fat 0 g Cholesterol, Sodium 0 g Protein, Carbohydrate, Dietary fibre 0 g Natural Carotenes 46 mg Natural Vitamin E 74 mg Co-Enzyme Q10 4 mg LVDevP = Left ventricular developed pressure; RPO = red palm oil; W = wortmanin P < 0.05* RPO+W vs Control # RPO+W vs Control $RPO vs Control &Control + W vs Control Table 2: The components of Carotino Premium red palm oil (per 100 ml) Total fats 92 g Monounsaturates 43 g Polyunsaturates 12 g Saturates 37 g Trans fat 0 g Cholesterol, Sodium 0 g Protein, Carbohydrate, Dietary fibre 0 g Natural Carotenes 46 mg Natural Vitamin E 74 mg Co-Enzyme Q10 4 mg LVDevP = Left ventricular developed pressure; RPO = red palm oil; W = wortmanin P < 0.05* RPO+W vs Control # RPO+W vs Control $RPO vs Control &Control + W vs Control Table 2: The components of Carotino Premium red palm oil Table 2: The components of Carotino Premium red palm oil (per 100 ml) Functional measurements were taken during pre-ischae- mia (20 min perfusion) and at 5 min, 10 min and 30 min into reperfusion. Heart rate and left ventricle developed pressure (LVDevP) were measured. LVDevP was calcu- lated as the difference between left ventricular systolic (LVSP) and diastolic (LVDP) pressures. The rate pressure product (RPP) was calculated as the product of heart rate and LVDevP. Table 2: The components of Carotino Premium red palm oil (per 100 ml) Methods All animals received humane care in accordance with the Principles of Laboratory Animal Care of the National Society of Medical Research and the Guide for the Care and use of Lab- oratory Animals of the National Academy of Sciences (National Institutes of Health publications no. 80-23, revised 1978). Perfusion Protocol Male Wistar groups were randomly divided into four groups: two control groups receiving standard rat chow and two experimental groups receiving standard rat chow plus 2 ml RPO (Carotina Premium) for 4 weeks. The com- position of Carotino Premium red palm oil (per 100 ml) is given in Table 2. Red palm oil was mixed with one pellet of the chow every morning. Rats were only fed the rest of the daily rat chow allowance after they consumed the pel- let with the red palm oil. The study was divided into two perfusion protocols. In the first protocol, hearts were perfused for 10 min stabiliza- tion, followed by 20 min, during which mechanical func- tion was documented. Hearts were then subjected to 25 min of total gloabal ischaemia. After the ischaemic period, hearts were reperfused for 30 min and mechanical function was again documented. To reduce the incidence of arrhythmias during reperfusion, a 2% lignocaine solu- tion was used for the last min of pre-ischaemia perfusion as well as the initial 3 min of reperfusion in all hearts. In the second protocol, hearts were stabilized for 10 min and perfused for 15 min, before being subjected to a wortman- nin solution (100 nM) for 5 min pre-ischaemia. After the 25 min total global ischaemic period, hearts were reper- fused for 3 min with the wortmannin solution, before reverting to the drug-free Krebs-Henseleit buffer for the rest of the 27 min reperfusion period. Functional and bio- chemical measurements were taken. Conclusion The present study demonstrates that the beneficial effect of RPO during ischemia/reperfusion-induced injury is partially mediated by the PI3-kinase signaling pathway. PI3-K inhibition attenuated functional recovery of the hearts during reperfusion. This attenuation in functional recovery when PI3-K was inhibited also correlated with reduced PKB and FKHR phosphorylation. This, in turn, leads to increased apoptosis as indicated by increased cas- Page 5 of 8 (page number not for citation purposes) Page 5 of 8 (page number not for citation purposes) http://www.lipidworld.com/content/8/1/18 Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 http://www.lipidworld.com/content/8/1/18 The aorta was cannulated and retrograde perfusion was initiated. Hearts were kept in a water-jacketed chamber to maintain temperature at 37°C. Immediately after cannu- lation, excess tissue and the left atrium was removed. A water-filled balloon (made from transparent sandwich wrap), connected to a pressure transducer, was inserted through the opening of the left atrium into the left ventri- cle. The pressure transducer was connected to a Powerlab system (ADInstruments Pty Ltd. Castle Hill, Australia) on a computer. After insertion, the balloon was inflated to 2 mmHg, and the contraction force of the heart against the balloon causes pressure on the fluid filled balloon. This pressure is then registered on the Powerlab system. Thus, systolic pressure, diastolic pressure and heart rate were measured. The first 10 min of perfusion was used to stabi- lize the heart. pase-3 and PARP cleavage. The beneficial effect of RPO during ischemia/reperfusion-induced injury is thus asso- ciated with the PI3-K/PKB signaling pathway and thus points towards this pathway as a potential therapeutic tar- get. The effects of RPO on cardiac function should be fur- ther characterized for the purpose of development as an agent for the management of ischemic injury. Heart Perfusion Rats weighing 300–400 g were anaesthetized with sodium pentobarbital, before hearts were rapidly excised and briefly rinsed by immersion in ice-cold Krebs-Henseleit buffer. Hearts were transferred to a Langendorff perfusion apparatus and perfused with a Krebs-Henseleit buffer equilibrated with 95% O2 and 5% CO2 at 37°C (118, 5 mM NaCl; 4.75 mM KCL; 1.2 mM MgCl • 6 H2O; 1.36 mM CaCl2; 25, 0 mM NaHCO3; 1.2 mM KH2PO4; 11, 0 mM glucose). Pressure was kept constant at 100 cm H2O. Competing interests p g The authors declare that they have no competing interests. p g The authors declare that they have no competing interests. References 1. Ho KK, Pinsky JL, Kannel WB, Levy D: The epidemiology of heart failure: the Framingsham study. J Am Coll Cardiol 1993, 22:6A-13A. 2. Esterhuyse AJ, du Toit EF, Benade AJS, van Rooyen J: Dietary red palm oil improves reperfusion cardiac function in the iso- lated perfused rat heart of animals fed a high cholesterol diet. Prostaglandins, Leukotrienes and Essential Fatty Acids 2005, 72:153-161. 3. Engelbrecht AM, Esterhuyse J, du Toit EF, Lochner A, van Rooyen J: p38-MAPK and PKB/Akt, possible role players in red palm oil-induced protection of the isolated perfused rat heart? Journal of Nutritional Biochemistry 2006, 17(4):265-71. Data analysis l Values are expressed as mean ± standard error of the mean (SEM). Some functional values are presented as percent- age change from the baseline values. Results were com- pared by using a one-way ANOVA with a Bonferoni Multiple Comparison as a post hoc test. P < 0.05 was con- sidered as statistically significant. J f y ( ) 4. Kruger M, Engelbrecht AM, Esterhuyse J, du Toit EF, van Rooyen J: Dietary Red Palm Oil (RPO) reduces ischaemia/reperfusion injury in a hypercholesterolemic diet. Br J Nutr 2007, 97(4):653-60. ( ) 5. Van Rooyen J, Esterhuyse AJ, Engelbrecht AM, Du Toit EF: Health Benefits of a natural carotonoid rich oil: a proposed mecha- nisms of protection against ischaemia/reperfusion injury. Asia Pac J Clin Nutr 2008, 17(S1):316-319. J ( ) 6. Stahl W, Sies H: Bioactivity and protective effects of natural carotonoid. Biochim Biophys Acta. 2005, 1740(2):101-107. Authors' contributions AME contributed to the interpretation of results, drafted and finalized the manuscript. LO fed the rats, performed perfusions and western blots, initial preparation of man- uscript. EFDT participated in design and interpretation of results. KK performed initial western blots. TC coordi- nated biochemical analysis and study design. PF contrib- uted to the interpretation of the results concerning inhibitors and draft of manuscript. JvR conceived the study, participated in design, coordination, and interpre- tation of results and final preparation of manuscript. Biochemical Analysis To assess myocardial biochemical function, hearts, from all groups, were freeze clamped 10 min into reperfusion with Wollenberger clamps precooled in liquid nitrogen. Cardiac proteins were extracted with a lysis buffer contain- ing: 20 mM Tris; 20 mM p-nitrophenylphosphate; 1 mM EGTA; 50 mM NaF; 0.1 sodium orthovanadate; 1 mM Page 6 of 8 (page number not for citation purposes) Page 6 of 8 (page number not for citation purposes) Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 phenylmethyl sulfonyl fluoride (PMSF); 1 mM dithiothre- itol (DTT); 10 μg/ml aprotinin. The tissue lysates were diluted in Laemmli sample buffer, boiled for 5 min and 60 μg protein was separated by 10% PAGE-SDS gel elec- trophoresis. The lysate protein content was determined using the Bradford technique [23]. The separated proteins were transferred to a PVDF membrane (Immobilon P, Millipore). These membranes were routinely stained with Ponceau Red for visualization of proteins. Nonspesific binding sites on the membranses were blocked with 5% fat-free milk in Tris-buffered saline – 0.1% Tween 20 (TBST) and then incubated with the primary antibodies that recognize PKB/Akt (Ser473 and Thr308) and total PKB/ Akt, PI3-K (p85), PDK1 (Ser241), FKHR (Ser256), GSK-3β (Ser9), cleaved caspase-3 (Asp175), cleaved PARP (Asp214) and PTEN (Ser380). Membranes were subsequently washed with large volumes of TBST (5 × 5 min) and the immobilized antibody conjugated with a diluted horse- radish peroxidase-labaled secondary antibody (Amer- sham, LIFE SCIENCE). After thorough washing with TBST, membranes were covered with ECL detection reagents and quickly exposed to an autoradiography film (Hyperfilm ECL, RPN 2103) to detect light emission through a non- radioactive method (ECL Western blotting). Films were densitometrically analyzed (UN-SCAN-IT, Silkscience) and phoshorylated protein values were corrected for minor differences in protein loading, if required. Antibod- ies were purchased from Cell Signalling Technology and all other chemicals were obtained from Sigma (St Louis, Missouri, USA). Acknowledgements Carotino Palm Oil was supplied by Carotino SDN BHD (company no. 69046-T), Johor Bahru, Malaysia. This work was supported by the National Office for Research and Technology (5LET-2-2008-0008, Jedlik A1-2006- 029, Jedlik "Med-Food"). L.O. was a visiting fellow at the University of Szeged, and K.K. at Cape Peninsula University of Technology in the scope of a Collaborative Research Grant ZA-35/2006 between Hungary and South Africa. Lipids in Health and Disease 2009, 8:18 http://www.lipidworld.com/content/8/1/18 13. Matsui T, Tao J, del Monte F, Lee KH, Li L, Picard M, Force TI, Francke TF, Hajjar RJ, Rosenzweig A: Akt activation preserves cardiac function and prevents injury after transient cardiac ischemia in vivo. Circulation 2001, 104:330-335. 14. Cataldi A, Zauli G, Di Pietro R: Involvement of the pathway phosphatidylinositol-3-kinase/Akt-1 in the establishment of the survival response to ionizing radiation. Cell Signal 2001, 13:369-375. 15. Kennedy SG, Wagner AJ, Conzen SD, Jordan J, Bellacosa A, Tsichilis PN, Hay N: The PI3-kinase/Akt signalling pathway delivers an anti-apoptotic signal. Genes Dev 1997, 11:701-713. p p g 16. James SR, Downes CP, Giggs R, Grove SJ, Holmes AB, Alessi DR: Specific binding of the Akt-1 protein kinase to phosphatidyli- nositol 3,4,5-trisphosphate without subsequent activation. Biochem J 1996, 315:709-713. 17. Biggs WH, Meisenhelder J, Hunter T, Cavenee WK, Arden KC: Pro- tein kinase B/Akt-mediated phosphorylation promotes nuclear exclusion of the winged helix transcription factor FKHR1. Proc Natl Acad Sci USA 1999, 96:7421-7426. 18. Burgering BM, Coffer PJ: Protein kinase B (c-Akt) in phosphati- dylinositol-3-OH kinase signal transduction. Nature 1995, 376(6541):553-554. ( ) 19. Rena G, Guo S, Cichy SC, Unterman TG, Cohen P: Phosphoryla- tion of the transcription factor forkhead family member FKHR by protein kinase B. J Biol Chem 1999, 24:17179-17183. 20. Szabo C: Cardioprotective effects of poly(ADP-ribose) polymerase inhibition. Pharmacol Res 2005, 52(1):34-43. ( ) 19. Rena G, Guo S, Cichy SC, Unterman TG, Cohen P: Phosphoryla- tion of the transcription factor forkhead family member FKHR by protein kinase B. J Biol Chem 1999, 24:17179-17183. y p J 20. Szabo C: Cardioprotective effects of poly(ADP-ribose) polymerase inhibition. Pharmacol Res 2005, 52(1):34-43. 21. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J, Greenberg ME: Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription fac- tor. Cell 1999, 6:857-868. 22. Zheng WH, Kar S, Quirioin R: Insulin-like growth factor-1- induced phosphorylation of the forkhead family transcrip- tion factor FKHRL1 is mediated by Akt kinase in PC12 cells. J Biol Chem 2000, 275:39152-39154. 23. Bradford MM: A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976, 71:248-54. Abbreviations A: beta; C: control; cm: centimeter; CO2: carbon dioxide; DTT: Dithiothreitol; FKHR: Forkhead transcription factor; g: gram; LVDevP: Left ventricular developed pressure; Min: minutes; ml: milliliter; mM: millimolar; O2: oxygen; %: percentage; PMSF: Phenylmethyl sulfonyl fluoride; PTEN: Phoshoinositide-lipid-3-phosphotase; PIP3: Phos- phatidylinositol-3,4,5-trisphosphate; PDK-1: Phosphoi- nositide-dependent kinase-1; PI-3K: Phosphatidylinositol 3-kinase; PARP: Poly(ADP-ribose) polymerase RPP: rate pressure product; ROS: reactive oxygen species; RPO: red palm oil; PKB/Akt: Serine/threonine protein kinase, pro- tein kinase B or AKT; SEM: Standard error of the mean; H2O: water; Wn: wortmannin. p y ( ) 7. Zhang LX, Cooney RV, Bertram JS: Carotonoid up-regulate connexin43 gene expression independent of pro-vitamin A or or anti-oxidant properties. Cancer Res 1992, 52:5707-5712. p p , 8. Kandel ES, Hay N: The regulation and activities of the multi- functional serine/threonine kinase Akt/PKB. Exp Cell Res 1999, 253:210-229. 9. Downward J: Mechanisms and consequences of activation of protein kinase B/Akt. Curr Opin Cell Biol 1998, 10:262-267. 10. Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR, Tsichlis PN: The protein kinase encoded by the Akt proto-oncogene is a target of the PDGF-activated phosphati- dylinositol 3-kinase. Cell 1995, 81(5):727-736. y ( ) 11. Alessi DR, Anjelkovic M, Caudwell FB, Cron P, Morrice N, Cohen P, Hemmings BA: Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J 1996, 15(23):6541-6551. J ( ) 12. Kajstura J, Cheng W, Reiss K, Clark WA, Sonneblick EH, Krajewski S, Reed JC, Olivetti G, Anversa P: Apoptotic and necrotic myocyte cell deaths are independent contributing variables of infarct size in rats. Lab Invest 1996, 74(1):86-107. Page 7 of 8 (page number not for citation purposes) Page 7 of 8 (page number not for citation purposes) Lipids in Health and Disease 2009, 8:18 Lipids in Health and Disease 2009, 8:18 Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 8 of 8 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge
https://openalex.org/W4327603756	https://www.frontiersin.org/articles/10.3389/fonc.2023.1006172/pdf	English	null	Precise prediction of the sensitivity of platinum chemotherapy in SCLC: Establishing and verifying the feasibility of a CT-based radiomics nomogram	Frontiers in oncology	2,023	cc-by	7,434	OPEN ACCESS Yanping Su 1,2,3,4†, Chenying Lu 1,4†, Shenfei Zheng 1,4, Hao Zou 1,4, Lin Shen 1,4, Junchao Yu 1,4, Qiaoyou Weng 1,4, Zufei Wang 1,4, Minjiang Chen 1,4, Ran Zhang 5, Jiansong Ji 1,4* and Meihao Wang 2,3* Yanping Su 1,2,3,4†, Chenying Lu 1,4†, Shenfei Zheng 1,4, Hao Zou 1,4, Lin Shen 1,4, Junchao Yu 1,4, Qiaoyou Weng 1,4, Zufei Wang 1,4, Minjiang Chen 1,4, Ran Zhang 5, Jiansong Ji 1,4* and Meihao Wang 2,3* 1Key Laboratory of Imaging Diagnosis and Minimally Invasive Intervention Research, Institute of Imaging Diagnosis and Minimally Invasive Intervention Research, The Fifth Afﬁliated Hospital of Wenzhou Medical University, Lishui, Zhejiang, China, 2Department of Radiology, Key Laboratory of Intelligent Medical Imaging of Wenzhou, Institute of Aging, The First Afﬁliated Hospital of Wenzhou Medical University, Wenzhou Medical University, Wenzhou, Zhejiang, China, 3Key Laboratory of Alzheimer’s Disease of Zhejiang, Wenzhou, Zhejiang, China, 4Clinical College of The Afﬁliated Central Hospital, School of Medicine, Lishui University, Lishui, Zhejiang, China, 5AI Research Department, Huiying Medical Technology Co., Ltd, Beijing, China Objectives: To develop and validate a CT-based radiomics nomogram that can provide individualized pretreatment prediction of the response to platinum treatment in small cell lung cancer (SCLC). CITATION Su Y, Lu C, Zheng S, Zou H, Shen L, Yu J, Weng Q, Wang Z, Chen M, Zhang R, Ji J and Wang M (2023) Precise prediction of the sensitivity of platinum chemotherapy in SCLC: Establishing and verifying the feasibility of a CT-based radiomics nomogram. Front. Oncol. 13:1006172. doi: 10.3389/fonc.2023.1006172 Materials: A total of 134 SCLC patients who were treated with platinum as a ﬁrst- line therapy were eligible for this study, including 51 patients with platinum resistance (PR) and 83 patients with platinum sensitivity (PS). The variance threshold, SelectKBest, and least absolute shrinkage and selection operator (LASSO) were applied for feature selection and model construction. The selected texture features were calculated to obtain the radiomics score (Rad- score), and the predictive nomogram model was composed of the Rad-score and the clinical features selected by multivariate analysis. Receiver operating characteristic (ROC) curves, calibration curves, and decision curves were used to assess the performance of the nomogram. © 2023 Su, Lu, Zheng, Zou, Shen, Yu, Weng, Wang, Chen, Zhang, Ji and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). TYPE Original Research PUBLISHED 16 March 2023 DOI 10.3389/fonc.2023.1006172 TYPE Original Research PUBLISHED 16 March 2023 DOI 10.3389/fonc.2023.1006172 Precise prediction of the sensitivity of platinum chemotherapy in SCLC: Establishing and verifying the feasibility of a CT-based radiomics nomogram OPEN ACCESS EDITED BY Kaustav Bera, Maimonides Medical Center, United States REVIEWED BY Zhongxiang Ding, Zhejiang University, China Bilgin Kadri Aribas, Bülent Ecevit University, Türkiye Wenjie Liang, Zhejiang University, China Zhenzhou Yang, Chongqing Medical University, China CORRESPONDENCE Meihao Wang wzwmh@wmu.edu.cn Jiansong Ji jijiansong@zju.edu.cn †These authors have contributed equally to this work SPECIALTY SECTION This article was submitted to Cancer Imaging and Image-directed Interventions, a section of the journal Frontiers in Oncology RECEIVED 29 July 2022 ACCEPTED 06 March 2023 PUBLISHED 16 March 2023 CITATION Su Y, Lu C, Zheng S, Zou H, Shen L, Yu J, Weng Q, Wang Z, Chen M, Zhang R, Ji J and Wang M (2023) Precise prediction of the sensitivity of platinum chemotherapy in SCLC: Establishing and verifying the feasibility of a CT-based radiomics nomogram. Front. Oncol. 13:1006172. doi: 10.3389/fonc.2023.1006172 COPYRIGHT © 2023 Su Lu Zheng Zou Shen Yu Weng OPEN ACCESS EDITED BY Kaustav Bera, Maimonides Medical Center, United States REVIEWED BY Zhongxiang Ding, Zhejiang University, China Bilgin Kadri Aribas, Bülent Ecevit University, Türkiye Wenjie Liang, Zhejiang University, China Zhenzhou Yang, Chongqing Medical University, China CORRESPONDENCE Meihao Wang wzwmh@wmu.edu.cn Jiansong Ji jijiansong@zju.edu.cn †These authors have contributed equally to this work OPEN ACCESS The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Results: The Rad-score was calculated using 10 radiomic features, and the resulting radiomics signature demonstrated good discrimination in both the training set (area under the curve [AUC], 0.727; 95% conﬁdence interval [CI], 0.627–0.809) and the validation set (AUC, 0.723; 95% CI, 0.562–0.799). To improve diagnostic effectiveness, the Rad-score created a novel prediction nomogram by combining CA125 and CA72-4. The radiomics nomogram showed good calibration and discrimination in the training set (AUC, 0.900; 95% CI, 0.844-0.947) and the validation set (AUC, 0.838; 95% CI, 0.534-0.735). The radiomics nomogram proved to be clinically beneﬁcial based on decision curve analysis. 01 Frontiers in Oncology frontiersin.org Su et al. Su et al. 10.3389/fonc.2023.1006172 10.3389/fonc.2023.1006172 Conclusion: We developed and validated a radiomics nomogram model for predicting the response to platinum in SCLC patients. The outcomes of this model can provide useful suggestions for the development of tailored and customized second-line chemotherapy regimens. radiomics, computed tomography, small cell lung cancer, chemotherapy, platinum 2.1 Patients The study was approved by the Institutional Review Board and Human Ethics Committee of the Fifth Afﬁliated Hospital of Wenzhou Medical University, and the requirement for informed consent was waived. Patients who were diagnosed with pathologically conﬁrmed SCLC from February 2014 to November 2021 were enrolled. A total of 134 patients were included according to the following inclusion criteria: (1) they underwent a CT examination before treatment; (2) they used platinum derivatives on a regular basis in ﬁrst-line chemotherapy and had never received any other treatment before; (3) dynamic CT follow-up was performed during treatment; and (4) endpoint events occurred. A total of 133 patients were excluded due to the following factors: (1) they were not treated in our hospital (n = 74); (2) they underwent other chemotherapy regimens or treatments (n = 30); (3) they underwent surgical resection (n = 16); (4) there was no follow-up after treatment (n = 9); and (5) no endpoint events occurred during follow-up (n = 4). Finally, 134 patients were selected for the present study. The ﬂow of the case identiﬁcation process is shown in Figure 1. Several studies have sought to use serum indicators and genetic tissue features to predict the responses to platinum in SCLC. SCLC is composed of four distinct subtypes, each of which reacts differently to platinum-based chemotherapy. The percentage of each subtype in the tumor inﬂuences how sensitive it is to platinum-based chemotherapy as a whole. However, the majority of SCLC tissue test samples are collected using needle biopsy, which unavoidably results in test variance and instability of prediction results (13). Other studies have attempted to use peripheral blood indices such as LDH and the systemic immune-inﬂammation index to predict the OS and PFS of SCLC (14, 15). However, the basic peripheral blood information is unconvincing, and these studies do not account for the tumor’s size, shape, location, or other relevant factors. Compared with the above methods, radiomics nomograms can be combined with radiomic and clinical features for noninvasive diagnosis, prognosis evaluation, and treatment response prediction. Previous studies have demonstrated that features based on radiomics are inextricably linked to underlying genomic patterns across a range of cancer types (16–18). Several studies using radiomics to predict platinum resistance in non-small cell lung Frontiers in Oncology 1 Introduction cancer have been reported, and their radiomics models have shown excellent diagnostic efﬁcacy (19–23). Nonetheless, there is no radiomics model for predicting platinum resistance in SCLC. Small cell lung cancer (SCLC), the most aggressive kind of lung cancer, accounts for approximately 14% of all lung cancer types and has a 5-year overall survival (OS) rate of just 6.7%. Due to its strong invasiveness, medication resistance, and the fact that no new, effective treatments have been developed in recent years (1, 2). Etoposide and platinum (EP) chemotherapy are the standard ﬁrst- line therapies for SCLC, with initial response rates of 70–80% and high chemotherapeutic sensitivity. However, almost all patients will experience progression (3, 4). According to current studies, platinum-sensitive (PS) patients have a 15% to 20% better response rate to conventional second-line platinum chemotherapy than platinum-resistant (PR) patients, and their OS can be increased by 2-3 months (2, 5–7). For PS patients, the median PFS from the time of EP rechallenge as second-line treatment was 5.5 months, but PR patients had limited efﬁcacy. Therefore, platinum reactivation is recommended for PS patients as second- line treatment, while PR patients are recommended to undergo topotecan treatment and other clinical trials. Thus, individualized second-line therapy based on an evaluation of platinum sensitivity is essential for improving the overall survival of SCLC patients (8–12). In this study, we aimed to develop and validate a CT-based radiomics nomogram that can provide individualized pretreatment prediction of the response to platinum treatment in SCLC, while effectively integrating image texture features and clinical factors. Using this nomogram, clinicians can enhance the treatment plan before initiating platinum-based chemotherapy and direct second- line therapy, optimize existing treatment combinations, and increase patient survival. 2.2 Endpoints We evaluated the tumor response of SCLC patients who received CT examinations during platinum chemotherapy based Frontiers in Oncology 02 frontiersin.org Su et al. 10.3389/fonc.2023.1006172 within 6 months. Representative CT images for PR and PS patients are shown in Figure 2. within 6 months. Representative CT images for PR and PS patients are shown in Figure 2. FIGURE 1 Flowchart of study enrollment. FIGURE 1 Flowchart of study enrollment. 2.3 CT image acquisition and interpretation The patients underwent nonenhanced CT scans with a 256-slice Philips Brilliance iCT system prior to treatment (Philips Medical Systems). The following are the detailed acquisition parameter settings: tube voltage 120 kV, reference tube current 113 mAs, automatic millisecond technology, scanning ﬁeld of view (SFOV) 15-20 cm, tube rotation time 0.75 s/circle, collimation width 80 mm (128×0.625 mm), reconstruction thickness 0.9 mm, reconstruction interval 0.45 mm, reconstruction matrix 1024× 1024, using the iDose3 iterative reconstruction algorithm. Two thoracic radiologists with 5 and 15 years of experience (Y.S. and C.L.) independently conducted retrospective reviews. Disagreements were settled by a third radiologist who had 25 years of experience (J.J.). The image features included the following: (1) number of lesions and (2) volume, measured using the Extended Brilliance Workspace and Lung Nodule Assessment software (Philips); (3) location: central or peripheral; (4) morphology: regular or irregular; (5) shape: regular or irregular; (6) lobulation (present/absent); (7) necrosis (present/absent); (8) hydrothorax (present/absent); (9) intratumoral calciﬁcation (present/absent); (10) staging (limited- stage/extensive-stage); and (11) metastasis (lymph den/bone/ parenchyma organ/cardiovascular/pleural and pericardium). on the modiﬁed Response Evaluation Criteria in Solid Tumors (mRECIST). The corresponding mRECIST responses were as follows: (1) complete response (CR): complete tumor disappearance; (2) partial response (PR): a minimum of 30% decrease in the sum of target lesion diameters; (3) progressive disease (PD): a minimum of 20% increase in the sum of target lesion diameters; and (4) stable disease (SD): neither PR nor PD. In this study, all patients underwent CT before and after platinum treatment, and the endpoint event was deﬁned as the occurrence of PD. The patients were divided into PR and PS groups according to whether the time from platinum chemotherapy to the ﬁrst PD was B A FIGURE 2 Representative CT images for PR and PS of SCLC patients according to the mRECIST criteria. (A) A 65-year-old female SCLC patient with a lesion diameter of 71 mm underwent CT scanning 1 week before EP chemotherapy, followed by CT scanning 2 months later. The lesion diameter increased to 97 mm, and the results showed that the patient presented with PR. (B) A 54-year-old male SCLC patient with a lesion diameter of 95 mm. CT scanning was performed 1 week before EP chemotherapy, and follow-up CT examinations were performed regularly after EP chemotherapy. Ten months later, the lesion diameter decreased to 56 mm, and the results showed that the patient presented with PS. 2.4 Tumor segmentation of volumes of interest and extraction of radiomic features In the imaging and storage of medical images, to make the intensity information consistent, the following formula was used to normalize all the radiomic features of CT images. The radiomics workﬂow is shown in Figure 3. Tumors and mediastinal lymph nodes fused with tumors in the mediastinal window were included in the volume of interest (VOI). First, a radiologist (reader 1, Y. S, a radiologist with ﬁve years of chest imaging experience) manually annotated 3D tumor VOIs around the largest lesion using the Radcloud platform (Huiying Medical Technology Co., Ltd, http://mics.radcloud.cn). To evaluate the reproducibility of the extracted features, reader 2 (C. L, a radiologist with 15 years of chest imaging experience) independently segmented 10% of lesions randomly selected from both the PR and PS groups. f (x) = s(x −ux) sx Where f(x) is the normalized intensity, x is the original intensity, µ and s are the mean value and variance, respectively, and s represents an optional scaling whose default is 1. The samples were randomly divided into a training cohort (n=58, 70%) and a validation cohort (n=25, 30%). To reduce the redundant features, the feature selection methods included the variance threshold, SelectKBest, and the least absolute shrinkage and selection operator (LASSO). For the variance threshold method, the threshold is 0.8, so that the eigenvalues of the variance smaller than 0.8 are removed. The SelectKBest method, which is a single-variable feature selection method, uses the p value to analyze the relationship between the features and the classiﬁcation results; all the features with a p value smaller than 0.05 are used. For the LASSO model, L1 regularization is used as the cost function, the error value of cross-validation is 10, and the maximum number of iterations is 1,000. Subsequently, the radiomic parameters with nonzero coefﬁcients in the LASSO model generated by the entire training cohort with the optimal a were selected. The radiomics signature (i.e., Rad-score) was computed for each lesion by a linear combination of the selected features as weighted by their respective quotient. For each VOI on our CT images, 1,409 radiomic features were extracted using a tool from the Radcloud platform, which extracted radiomic features from medical image data with a large panel of engineered hard-coded feature algorithms (https:// pyradiomics.readthedocs.io/en/latest/features.html). 2.3 CT image acquisition and interpretation FIGURE 2 Representative CT images for PR and PS of SCLC patients according to the mRECIST criteria. (A) A 65-year-old female SCLC patient with a lesion diameter of 71 mm underwent CT scanning 1 week before EP chemotherapy, followed by CT scanning 2 months later. The lesion diameter increased to 97 mm, and the results showed that the patient presented with PR. (B) A 54-year-old male SCLC patient with a lesion diameter of 95 mm. CT scanning was performed 1 week before EP chemotherapy, and follow-up CT examinations were performed regularly after EP chemotherapy. Ten months later, the lesion diameter decreased to 56 mm, and the results showed that the patient presented with PS. 03 03 Frontiers in Oncology frontiersin.org Su et al. Su et al. 10.3389/fonc.2023.1006172 2.4 Tumor segmentation of volumes of interest and extraction of radiomic features The 1,409 features obtained were divided into four main categories: ﬁrst- order statistics, shape, texture [gray-level cooccurrence (GLCM), gray-level run length (GLRLM), gray-level size zone (GLSZM), neighboring gray tone difference (NGTDM), gray-level dependence (GLDM), Matrices], and higher-order statistics (Laplacian of Gaussian, wavelet, square, square root, logarithm, exponential, gradient, and local binary pattern ﬁlters) features. The intraclass correlation coefﬁcient (ICC) was used to validate the reproducibility of extracted features from the two radiologists. Radiomic features with intra-ICCs >0.75 were selected for the subsequent statistical analysis. FIGURE 3 Flowchart of the study. Flowchart of the study. 04 04 Frontiers in Oncology frontiersin.org Su et al. Su et al. 10.3389/fonc.2023.1006172 10.3389/fonc.2023.1006172 2.6 Construction and internal validation of the nomogram model median and interquartile range. The c2 test was used to analyze the categorical variables, the t test was applied to analyze the continuous variables with a normal distribution, and the Mann −Whitney U test was used for variables with an abnormal or unknown distribution. Multivariable logistic regression analysis was used to select the independent prognostic factors. The performance of the model was assessed in the primary and validation cohorts. The discrimination of the signature was measured by the area under the curve (AUC). The variables, including clinical factors, conventional CT ﬁndings, and Rad-scores between the samples of platinum- resistant groups and platinum-sensitive groups with signiﬁcant differences, were analyzed via multivariate logistic regression to build the radiomics nomogram. The performance of the nomogram was evaluated by plotting receiver operating characteristic curves. The ICC was graded as follows: poor (<0.20), moderate (0.20– 0.40), fair (0.40–0.60), good (0.60–0.80), or very good (0.80–1.00). Statistical analyses were performed using SPSS software (Ver. 25, IBM, Armonk, New York), SigmaPlot (Ver. 14.0), R software package (Ver. 3.5.2, R Development Core Team: https://www.r- project.org/), and the Python scikit-learn package (Ver. 3.7, scikit- learn Ver. 0.21, http://scikit-learn.org/). The Hosmer−Lemeshow test was used to evaluate the goodness- of-ﬁt of the nomogram. The classiﬁcation accuracy between the predicted probability and the observed results was evaluated using calibration curves. The diagnostic performance of the nomogram was assessed by evaluating the AUC, sensitivity, speciﬁcity, and accuracy. The AUC between the optimized signature and the nomogram was evaluated by using the DeLong test. Decision curve analysis (DCA) was performed to evaluate the clinical utility of the nomogram. 2.7 Statistical analysis All quantitative features were analyzed with SPSS 25. P<0.05 was considered as statistically signiﬁcant. The baseline clinical characteristics of the patients are summarized in Table 1. A total of 134 patients were enrolled in this study: 51 patients with PR and 83 patients with PS. The mean Categorical variables are shown as frequencies, and continuous variables are presented as the mean and standard deviation or TABLE 1 Baseline characteristics of the patients in the PR and PS groups. TABLE 1 Baseline characteristics of the patients in the PR and PS groups. TABLE 1 Baseline characteristics of the patients in the PR and PS groups. Variables PR (51) PS (83) t/c2/U P Sex 0.063 0.802 Male 45 (88%) 72 (87%) Female 6 (12%) 11 (13%) Age 62.71 ± 9.38 61.28 ± 7.33 0.983 0.327 BMI/kg·m-2 22.06 (20.20, 25.00) 22.53 (20.57, 24.61) 0.472 0.637 Smoking 38 (75%) 62 (75%) <0.001 0.981 Superior vena cava syndrome 3 (6%) 5 (6%) <0.001 1.000 Spinal cord compression 3 (6%) 5 (6%) <0.001 1.000 Ki67 80% (70%, 85%) 80% (70%, 85%) 0.576 0.565 Tumor number 0.737 0.692 1 41 (80%) 69 (83%) 2 1 (2%) 3 (4%) ≥3 9 (18%) 11 (13%) Tumor volume 115.06 (14.50, 247.30) 65.99 (18.72, 160.40) 1.191 0.233 Intratumoral calciﬁcation 3 (6%) 2 (2%) 0.314 0.575 Tumor location <0.001 0.987 Central 40 (78%) 65 (78%) Peripheral 11 (22%) 18 (22%) (Continued) 05 Frontiers in Oncology frontiersin.org Su et al. Frontiers in Oncology se; CEA, carcinoembryonic antigen; pro-GRP, progastrin-releasing peptide; CA125, carbohydrate antigen 125; CA72-4, carbohydrate antigen 72-4; CA199 ER, ferroprotein; SCC, squamous cell carcinoma; ApoB, apolipoprotein B; ApoA, apolipoprotein A; HDL, high-density lipoprotein; LDL, low-density lipoprotein 2.7 Statistical analysis 10.3389/fonc.2023.1006172 TABLE 1 Continued Variables PR (51) PS (83) t/c2/U P Tumor morphology 1.615 0.532 Regular 16 (31%) 25 (30%) Irregular 35 (69%) 58 (70%) Lobulated <0.001 1.000 Absent 2 (4%) 4 (5%) Present 49 (96%) 79 (95%) Necrosis 0.285 0.594 Absent 27 (53%) 41 (48%) Present 24 (47%) 43 (52%) Hydrothorax 1.941 0.164 Absent 27 (53%) 55 (65%) Present 24 (47%) 29 (35%) Staging 0.202 0.653 LS 25 (49%) 45 (53%) ES 26 (51%) 39 (47%) Metastasis Lymph den 49 (96%) 73 (88%) 1.659 0.198 Bone 9 (18%) 6 (7%) 3.449 0.063 Parenchyma organ 12 (24%) 14 (17%) 0.897 0.344 Cardiovascular 12 (24%) 16 (19%) 0.346 0.557 Pleural and pericardium 7 (14%) 8 (10%) 0.531 0.466 NSE 37.30 (22.20, 103.60) 31.80 (18.20, 55.90) 1.991 0.046 CEA 7.30 (2.90, 18.60) 3.80 (2.20, 6.50) 2.390 0.017 Pro-GRP 714.9 (129.90, 3212.20) 597.80 (156.90, 1953.10) 0.660 0.509 CYFRA-211 3.20 (2.30, 5.20) 2.90 (2.00, 4.40) 1.407 0.159 CA125 33.40 (18.20, 66.60) 17.30 (13.10, 31.10) 4.443 <0.001 CA72-4 1.60 (1.00, 2.60) 2.50 (1.20, 5.60) 2.726 0.006 CA199 23.40 (5.60, 48.80) 12.10 (4.80, 23.20) 2.809 0.005 FER 249.70 (156.00, 491.80) 280.50 (188.30, 387.90) 0.332 0.740 SCC 0.90 (0.50, 1.10) 0.70 (0.60, 1.00) 0.600 0.549 ApoB/ApoA 0.76 (0.63, 0.82) 0.67 (0.60, 0.80) 1.241 0.215 HDL 1.12 (0.90, 1.32) 1.12 (0.97, 1.27) 0.133 0.894 LDL 2.47 (2.12, 3.10) 2.37 (2.12, 3.10) 0.500 0.617 TG 0.98 (0.77, 1.43) 1.40 (1.12, 2.01) 4.015 <0.001 NSE, neuron-speciﬁc enolase; CEA, carcinoembryonic antigen; pro-GRP, progastrin-releasing peptide; CA125, carbohydrate antigen 125; CA72-4, carbohydrate antigen 72-4; CA199, 0.977-0.998, P =0.022) and CA72-4 (OR: 1.172, 95% CI: 1.023- 1.341, P =0.022) were independent predictors of SCLC with PS (Table 2). The ROC curves of CA125, CA72-4 and the clinical model are shown in Figure S1. ages were 62.71 ± 9.38 and 61.28 ± 7.33, respectively. Univariate analysis showed that NSE, CEA, CA125, CA72-4, CA199, and TG were signiﬁcantly different between the two groups. Subsequently, multivariate analysis suggested that CA125 (OR: 0.98, 95% CI: 06 frontiersin.org Su et al. 10.3389/fonc.2023.1006172 TABLE 2 Univariate and multivariate analyses of clinical factors. Characteristic Univariate Multivariate OR 95% CI P OR 95% CI P NSE 0.994 0.988-0.999 0.024 0.994 0.987-1.000 0.063 CEA 0.982 0.962-1.002 0.078 CA125 0.986 0.975-0.997 0.014 0.987 0.977-0.998 0.022 CA72-4 1.187 1.037-1.359 0.013 1.172 1.023-1.341 0.022 CA199 0.995 0.987-1.003 0.212 TG 1.043 0.913-1.190 0.538 3.2 Feature extraction, selection, and radiomic signature building Of the 1409 radiomic features extracted from CT images, 1186 were demonstrated to have good interobserver agreement, with intra-ICCs >0.75. A total of 1107 radiomic features by variance threshold were enrolled in SelectKBest to select the most valuable 60 features. Finally, 10 features were screened out by LASSO to build the radiomic signature model. The optimal parameter l of each fold and the selected features of the corresponding fold are shown in Figure 4. The ROC curves of the 10 radiomic features and radiomics model are shown in Figure 5. The ROC and decision curves of the nomogram model are shown in Figures 6A–D. The CA125, CA72-4, and Rad-score were incorporated into the construction of the radiomics nomogram (Figure 6E). Figures 6F, G shows the calibration curve of the nomogram. The Nomo-scores for each patient are shown in Figure S2. The AUC of the clinical model was higher than that of the radiomics model in the training cohort, whereas the AUC value of the radiomics model was higher than that of the clinical model in the validation cohort. The AUC value of the nomogram model was signiﬁcantly higher than that of the clinical and radiomics models in the training cohort and veriﬁcation cohort. The calibration curve Based on these 10 features and their regression coefﬁcients, the radiomics score (Rad-score) formula was constructed as follows: Rad-score = feature * coefﬁcient (Table 3). B C D A FIGURE 4 Radiomic feature selection using the variance threshold, SelectKBest and selection operator (LASSO) regression model. LASSO regression model on CT images. The mean square error on each fold in the tenfold cross-validation method and the optimal value of the lasso tuning parameter (-log (a) =1.574,a= 2.978) were found (A). The vertical line was plotted with 10 selected radiomic features (B). The 10 radiomic features were selected after dimension reduction (C, D). B B A B C D C D FIGURE 4 Radiomic feature selection using the variance threshold, SelectKBest and selection operator (LASSO) regression model. LASSO regression model on CT images. The mean square error on each fold in the tenfold cross-validation method and the optimal value of the lasso tuning parameter (-log (a) =1.574,a= 2.978) were found (A). The vertical line was plotted with 10 selected radiomic features (B). The 10 radiomic features were selected after dimension reduction (C, D). 07 Frontiers in Oncology frontiersin.org Su et al. 4 Discussion showed good calibration in the training cohort and validation cohort (Figure S3). The radiomics nomogram showed the highest net beneﬁt of the three models. The standard ﬁrst-line therapy for SCLC is platinum-based chemotherapy, which has a 70–80% success rate and often a very pronounced early effect (3, 4). However, the disease will progress quickly, on average, six months after the ﬁrst treatment has been administered (2). Current management advice is that PR patients should try clinical trial medication such as topotecan or lurbinectedin because they would gain little from an EP regimen, whereas PS patients should be restimulated with an EP regimen (6). To increase the overall survival rate, it is crucial to evaluate the tumor’s response to platinum chemotherapy and choose a suitable second-line treatment prior to ﬁrst-line therapy (24, 25). In this study, we established a CT-based, noninvasive radiomics nomogram model that incorporates the radiomics signature and clinical factors to predict a customized platinum response in SCLC patients. Overall, our study serves as an example of precision medicine and can inﬂuence treatment options. The radiomics signatures based on the nomogram model showed high performance in differentiating between platinum- resistant groups and platinum-sensitive groups, with an AUC of 0.900 (95% CI, 0.844-0.947; sensitivity, 83.61%; speciﬁcity, 78.13%; accuracy, 81.72%) in the training cohort and 0.838 (95% CI, 0.534- 0.735; sensitivity, 68.57%; speciﬁcity, 83.33%; accuracy, 70.73%) in the validation cohort. The AUC of the CT image model was 0.727 (95% CI, 0.627-0.809; sensitivity, 73.85%; speciﬁcity, 64.29%; accuracy, 70.97%) in the training cohort and 0.723 (95% CI, 0.562-0.799; sensitivity, 71.88%; speciﬁcity, 77.78%; accuracy, 73.17%) in the validation dataset. The AUC of the clinical model was 0.734 (95% CI, 0.637–0.814; sensitivity, 65.82%; speciﬁcity, 57.14%; accuracy, 64.52%) in the training cohort and 0.715 (95% CI, 0.514–0.756; sensitivity, 70.00%; speciﬁcity, 63.64%; accuracy, 68.29%) in the test dataset. For the combined radiomics signature, the Hosmer−Lemeshow test yielded P values of 0.219 and 0.308 in the training and validation cohorts, respectively, indicating no departure from a good ﬁt. In our study, mediastinal window texture characteristics in patients with SCLC were extracted using radiological methods, and 1409 potential radiological features were chosen for further TABLE 3 Description of the selected radiomic features with their associated feature group and ﬁlter. TABLE 3 Description of the selected radiomic features with their associated feature group and ﬁlter. 3.2 Feature extraction, selection, and radiomic signature building 10.3389/fonc.2023.1006172 B A FIGURE 5 ROC curves of the radiomics model. The ROC curves of the radiomics model in the training (A) and validation (B) cohorts. A B s model. The ROC curves of the radiomics model in the training (A) and validation (B) cohorts. FIGURE 5 ROC curves of the radiomics model. The ROC curves of the radiomics model in the training (A) and validation (B) cohorts. Imc2, Informational Measure of Correlation 2; MCC, Maximal Correlation Coefﬁcient; glrlm, gray level tun length matrix; glcm, gray-level cooccurrence matrix. mc2, Informational Measure of Correlation 2; MCC, Maximal Correlation Coefﬁcient; glrlm, gray level tun length matrix; glcm, gray-level cooccurrence matrix. Frontiers in Oncology 4 Discussion p g p Radiomic feature Radiomic class Filter Coefﬁcient Skewness ﬁrstorder wavelet-HLL -0.0778821372989 Kurtosis ﬁrstorder wavelet-LLH -0.048467472885 LongRunLowGrayLevelEmphasis glrlm original -0.0168388737795 Imc2 glcm wavelet-HHL -0.0243113156389 10Percentile ﬁrstorder square 0.0276043499311 MCC glcm wavelet-LHH -0.00170586466044 RunPercentage glrlm wavelet-HLL -0.0129710574977 ShortRunHighGrayLevelEmphasis glrlm wavelet-HLH -0.0156495283052 RootMeanSquared ﬁrstorder wavelet-LLL 0.0201114297855 RootMeanSquared ﬁrstorder original 0.0190099678839 08 frontiersin.org 10.3389/fonc.2023.1006172 Su et al. B C D E F G A FIGURE 6 ROC and curve decision curve analysis of the nomogram model. The radiomics nomogram and calibration curves for the radiomics nomogram. The ROC curves of the nomogram model in the training (A) and validation (B) sets and the decision curve analysis for the nomogram model in the training (C) and validation (D) sets. The radiomics nomogram, combining CA125, CA72-4, and Rad-score, was developed in the training cohort (E). Calibration curves for the radiomics nomogram in the training (F) and validation (G) cohorts. Calibration curves indicate the goodness-of-ﬁt of the nomogram. The 45° gray line represents the ideal prediction, and the blue lines and red lines represent the performance of the corrected and apparent bias, respectively. The closer the line approaches the ideal prediction line, the better the predictive efﬁcacy of the nomogram. E B A B A B D F G C F G F C D FIGURE 6 ROC and curve decision curve analysis of the nomogram model. The radiomics nomogram and calibration curves for the radiomics nomogram. The ROC curves of the nomogram model in the training (A) and validation (B) sets and the decision curve analysis for the nomogram model in the training (C) and validation (D) sets. The radiomics nomogram, combining CA125, CA72-4, and Rad-score, was developed in the training cohort (E). Calibration curves for the radiomics nomogram in the training (F) and validation (G) cohorts. Calibration curves indicate the goodness-of-ﬁt of the nomogram. The 45° gray line represents the ideal prediction, and the blue lines and red lines represent the performance of the corrected and apparent bias, respectively. The closer the line approaches the ideal prediction line, the better the predictive efﬁcacy of the nomogram. investigation. We were able to greatly enhance the number of texture characteristics by utilizing 3D annotation, which allowed us to avoid missing any crucial aspects altogether. Wavelet-based characteristics have been proposed as a tool for illness diagnosis and predicting therapy response (26, 27). 4 Discussion GLCM and GLRLM are both matrix-based features: GLCM describes the pairwise arrangement of voxels with the same gray value and is used to highlight local heterogeneity information; GLRLM is used to measure the distribution of high gray values, and the GrayLevelEmphasis value is expected to be larger for images with higher gray values. Our Rad- score includes two GLSZM features, MCC and GLCM. The MCC represents the complexity of the texture, and the lower the value, the more complex the texture. In this study, the MCC value of the sensitive group was lower, indicating that the lesion heterogeneity in the sensitive group was higher, and thus, the probability of a response to the treatment outcome was higher. recommendations to enhance PFS, which only offered minimal clinical support (23). These preliminary studies have further conﬁrmed that the texture feature-based radiomics method of SCLC is feasible for predicting platinum responsiveness. Additionally, our study expands on these ﬁndings to achieve more signiﬁcant outcomes with regard to clinical requirements and increased patient survival. To improve the prediction efﬁcacy, predictors beyond radiomics should also be incorporated with the radiomics signature to further increase the power of the decision support model. In previous studies, patient prognosis was inﬂuenced by characteristics such as patient sex, smoking history, tumor stage, and other variables; however, in our study, these variables had no impact on the tumor’s sensitivity to platinum-based chemotherapy. NSE, Pro-GRP, and CYFRA 21-1 are all linked to the prognosis of SCLC; however, they also cannot predict platinum resistance. As a result, CA125 and CA72-4 with corresponding odds ratios of 0.987 and 1.172 were selected by multivariable logistic regression analysis. The ORs suggest that the higher the CA125 and CA72-4 levels are, the greater the probability of a favorable response to platinum treatment in SCLC. The clinical phases of SCLC were linked to CA125 (29). According to the literature, a higher level of CA125 can indicate a better impact of ﬁrst-line treatment (30). Although Ca72- 4 can predict the degree of differentiation in gastric cancer (31–33), no studies have found a link between it and small cell lung cancer. The baseline expression of CA 125 and CA72-4 in SCLC can predict platinum resistance, according to our ﬁndings. Frontiers in Oncology frontiersin.org 8. Alvarado-Luna G, Morales-Espinosa D. Treatment for small cell lung cancer, where are we now?-a review. Transl Lung Cancer Res (2016) 5(1):26–38. doi: 10.3978/ j.issn.2218-6751.2016.01.13 Data availability statement All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Funding This study was supported by Zhejiang Medical and Health Science Project (2023KY418, 2020KY1085, 2020KY1080), Natural Science Foundation of Zhejiang (LGF21H180002). In summary, we developed and validated a radiomics model that incorporates the pretreatment CT-based radiomics signature and clinical variables for the prediction of the response to platinum treatment in patients with SCLC. This study can assist patients in customizing second-line chemotherapy, improve clinical decision- making, and increase patient survival. Additionally, this research could be utilized to forecast second-line therapy responsiveness and support the development of third-line treatment approaches. It offers a wide range of potential applications and is also applicable to different tumor types. 4. Gazdar AF, Bunn PA, Minna JD. Small-cell lung cancer: What we know, what we need to know and the path forward. Nat Rev Cancer (2017) 17(12):765. doi: 10.1038/ nrc.2017.106 4 Discussion Meanwhile, the potential 1409 candidate radiomic features were ﬁnally reduced to 10 potential predictors by shrinking the regression coefﬁcients with the LASSO method for further integration to form the Rad-score, which contains effective biological information and could reﬂect the heterogeneity of the tumor. The radiomics signature demonstrated good discrimination in both the training set (AUC, 0.727; 95% CI, 0.627-0.809) and the validation set (AUC, 0.723; 95% CI, 0.562-0.799). Several previous studies have demonstrated that the Rad-score can effectively predict the prognosis of patients due to its high correlation with tumor biological characteristics (28). Several radiomic model prediction algorithms have been developed in the past to predict tumor response to medications, including platinum- based chemotherapeutics, in a variety of cancers (16, 17, 19, 20). A recent study showed that the computed tomography-based radiomics signature was closely associated with the PFS of SCLC; however, this study primarily concentrated on PFS prediction and made no After selecting candidate predictors using multivariate logistic regression analysis, a nomogram model was built that included radiomics signatures, CA125, and CA72-4. Of note, our radiomics nomogram showed favorable discrimination (AUC 0.900) in the training cohort, which was further validated in the internal validation cohorts (AUC 0.834). Furthermore, DCA showed a higher overall net 09 frontiersin.org Su et al. Su et al. 10.3389/fonc.2023.1006172 Conﬂict of interest Author RZ is/was employed by Huiying Medical Technology Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Supplementary material The studies involving human participants were reviewed and approved by Institutional Review Board and Human Ethics Committee of Lishui Central Hospital. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fonc.2023.1006172/ full#supplementary-material Author contributions beneﬁt of the radiomics model, thus highlighting its value as a better tool for assisting in clinical decision-making. Using the radiomics nomogram model, if a patient is predicted to have a favorable response to platinum, second-line platinum chemotherapy should be recommended; if not, immune checkpoint inhibitors are a good alternative (11, 34, 35). This is particularly important for those with PR, since doctors can choose other treatment options at an earlier stage to prevent tumor progression due to drug resistance and improve recurrence-free survival. However, our study has several limitations. First, given the retrospective nature of this study, selection bias may exist. Second, the training/testing cohort is tiny. Due to morbidity, the sample size is smaller than other tumor type radiomics research samples but similar to those in SCLC radiomics studies. Larger datasets are needed to verify and improve our results, and external validation of our model’s performance with an independent cohort from other institutions is necessary. MW and JJ designed the study. SZ, HZ, LS and JY collected the data and drafted the paper. RZ, YS, QW, CL and ZW analyzed the data and made the ﬁgures. YS, MC, CL, JJ and MW revised the paper. All authors contributed to the article and approved the submitted version. 5. Mao MM, Li HM, Shi J, Qiu QS, Feng F. Prediction of platinum-based chemotherapy sensitivity for epithelial ovarian cancer by multi-sequence mri-based radiomic nomogram. Zhonghua Yi Xue Za Zhi (2022) 102(3):201–8. doi: 10.3760/ cma.j.cn112137-20210816-01844 7. Wilson MK, Pujade-Lauraine E, Aoki D, Mirza MR, Lorusso D, Oza AM, et al. Fifth ovarian cancer consensus conference of the gynecologic cancer intergroup: Recurrent disease. Ann Oncol (2017) 28(4):727–32. doi: 10.1093/annonc/mdw663 3. Schild SE, Sio TT, Daniels TB, Chun SG, Rades D. Prophylactic cranial irradiation for extensive small-cell lung cancer. J Oncol Pract (2017) 13(11):732–8. doi: 10.1200/ JOP.2017.026765 6. Ettinger DS, Wood DE, Akerley W, Bazhenova LA, Borghaei H, Camidge DR, et al. Nccn guidelines insights: Non-small cell lung cancer, version 4.2016. J Natl Compr Canc Netw (2016) 14(3):255–64. doi: 10.6004/jnccn.2016.0031 2. Tariq S, Kim SY, Monteiro de Oliveira Novaes J, Cheng H. Update 2021: Management of small cell lung cancer. Lung (2021) 199(6):579–87. doi: 10.1007/ s00408-021-00486-y 1. Das M, Padda SK, Weiss J, Owonikoko TK. Advances in treatment of recurrent small cell lung cancer (Sclc): Insights for optimizing patient outcomes from an expert roundtable discussion. Adv Ther (2021) 38(11):5431–51. doi: 10.1007/s12325-021- 01909-1 Frontiers in Oncology References High systemic immune-inﬂammation index (Sii) represents an unfavorable prognostic factor for small cell lung cancer treated with etoposide and platinum-based chemotherapy. Lung (2020) 198(2):405–14. doi: 10.1007/s00408-020-00333-6 30. Fu J, Fang S, Wen Y, Wang Y, Yin X, Wang D. The efﬁcacy and safety of irinotecan combined with nedaplatin in the treatment of small cell lung cancer. J BUON (2020) 25(4):1707–13. 16. Yi X, Liu Y, Zhou B, Xiang W, Deng A, Fu Y, et al. Incorporating Sulf1 polymorphisms in a pretreatment ct-based radiomic model for predicting platinum resistance in ovarian cancer treatment. BioMed Pharmacother (2021) 133:111013. doi: 10.1016/j.biopha.2020.111013 31. Wu T, Wang CH, Wang W, Liu LL, Yun JP, Zhou ZW. Association of preoperative and postoperative Ca72-4 with gastric cancer outcome. J Surg Oncol (2021) 123(8):1699–707. doi: 10.1002/jso.26446 17. Khorrami M, Khunger M, Zagouras A, Patil P, Thawani R, Bera K, et al. Combination of peri- and intratumoral radiomic features on baseline ct scans predicts response to chemotherapy in lung adenocarcinoma. Radiol Artif Intell (2019) 1(2): e180012. doi: 10.1148/ryai.2019180012 32. Tong Y, Zhao Y, Shan Z, Zhang J. Ca724 predicts overall survival in locally advanced gastric cancer patients with neoadjuvant chemotherapy. BMC Cancer (2021) 21(1):4. doi: 10.1186/s12885-020-07666-8 18. Chen Y, Fang Z, Tang Y, Jin Y, Guo C, Hu L, et al. Integrative analysis of multi- omics data reveals the heterogeneity and signatures of immune therapy for small cell lung cancer. Clin Transl Med (2021) 11(12):e620. doi: 10.1002/ctm2.620 33. Liu S, Xu M, Qiao X, Ji C, Li L, Zhou Z. Prediction of serosal invasion in gastric cancer: Development and validation of multivariate models integrating preoperative clinicopathological features and radiographic ﬁndings based on late arterial phase ct images. BMC Cancer (2021) 21(1):1038. doi: 10.1186/s12885-021-08672-0 19. Rekhtman N. Lung neuroendocrine neoplasms: Recent progress and persistent challenges. Mod Pathol (2022) 35(Suppl 1):36–50. doi: 10.1038/s41379-021-00943-2 34. Mak MP, Tong P, Diao L, Cardnell RJ, Gibbons DL, William WN, et al. A patient-derived, pan-cancer emt signature identiﬁes global molecular alterations and immune target enrichment following epithelial-to-Mesenchymal transition. Clin Cancer Res (2016) 22(3):609–20. doi: 10.1158/1078-0432.CCR-15-0876 20. He L, Li Z, Chen X, Huang Y, Yan L, Liang C, et al. A radiomics prognostic scoring system for predicting progression-free survival in patients with stage iv non- small cell lung cancer treated with platinum-based chemotherapy. Chin J Cancer Res (2021) 33(5):592–605. doi: 10.21147/j.issn.1000-9604.2021.05.06 35. References 1. Das M, Padda SK, Weiss J, Owonikoko TK. Advances in treatment of recurrent small cell lung cancer (Sclc): Insights for optimizing patient outcomes from an expert roundtable discussion. Adv Ther (2021) 38(11):5431–51. doi: 10.1007/s12325-021- 01909-1 3. Schild SE, Sio TT, Daniels TB, Chun SG, Rades D. Prophylactic cranial irradiation for extensive small-cell lung cancer. J Oncol Pract (2017) 13(11):732–8. doi: 10.1200/ JOP.2017.026765 4. Gazdar AF, Bunn PA, Minna JD. Small-cell lung cancer: What we know, what we need to know and the path forward. Nat Rev Cancer (2017) 17(12):765. doi: 10.1038/ nrc.2017.106 10 Frontiers in Oncology frontiersin.org 10.3389/fonc.2023.1006172 Su et al. 10.3389/fonc.2023.1006172 Su et al. 9. von Pawel J, Jotte R, Spigel DR, O'Brien ME, Socinski MA, Mezger J, et al. Randomized phase iii trial of amrubicin versus topotecan as second-line treatment for patients with small-cell lung cancer. J Clin Oncol (2014) 32(35):4012–9. doi: 10.1200/ JCO.2013.54.5392 9. von Pawel J, Jotte R, Spigel DR, O'Brien ME, Socinski MA, Mezger J, et al. Randomized phase iii trial of amrubicin versus topotecan as second-line treatment for patients with small-cell lung cancer. J Clin Oncol (2014) 32(35):4012–9. doi: 10.1200/ JCO.2013.54.5392 22. Wei H, Yang F, Liu Z, Sun S, Xu F, Liu P, et al. Application of computed tomography-based radiomics signature analysis in the prediction of the response of small cell lung cancer patients to ﬁrst-line chemotherapy. Exp Ther Med (2019) 17 (5):3621–9. doi: 10.3892/etm.2019.7357 23. Jain P, Khorrami M, Gupta A, Rajiah P, Bera K, Viswanathan VS, et al. Novel non-invasive radiomic signature on ct scans predicts response to platinum-based chemotherapy and is prognostic of overall survival in small cell lung cancer. Front Oncol (2021) 11:744724. doi: 10.3389/fonc.2021.744724 10. O'Brien ME, Ciuleanu TE, Tsekov H, Shparyk Y, Cucevia B, Juhasz G, et al. Phase iii trial comparing supportive care alone with supportive care with oral topotecan in patients with relapsed small-cell lung cancer. J Clin Oncol (2006) 24(34):5441–7. doi: 10.1200/JCO.2006.06.5821 10. O'Brien ME, Ciuleanu TE, Tsekov H, Shparyk Y, Cucevia B, Juhasz G, et al. Phase iii trial comparing supportive care alone with supportive care with oral topotecan in patients with relapsed small-cell lung cancer. J Clin Oncol (2006) 24(34):5441–7. doi: 10.1200/JCO.2006.06.5821 24. Wheeler HE, Maitland ML, Dolan ME, Cox NJ, Ratain MJ. Cancer pharmacogenomics: Strategies and challenges. Nat Rev Genet (2013) 14(1):23–34. doi: 10.1038/nrg3352 11. Trigo J, Subbiah V, Besse B, Moreno V, Lopez R, Sala MA, et al. References Lurbinectedin as second-line treatment for patients with small-cell lung cancer: A single-arm, open- label, phase 2 basket trial. Lancet Oncol (2020) 21(5):645–54. doi: 10.1016/S1470-2045 (20)30068-1 11. Trigo J, Subbiah V, Besse B, Moreno V, Lopez R, Sala MA, et al. Lurbinectedin as second-line treatment for patients with small-cell lung cancer: A single-arm, open- label, phase 2 basket trial. Lancet Oncol (2020) 21(5):645–54. doi: 10.1016/S1470-2045 (20)30068-1 25. Weinshilboum RM, Wang L. Pharmacogenomics: Precision medicine and drug response. Mayo Clin Proc (2017) 92(11):1711–22. doi: 10.1016/j.mayocp.2017.09.001 12. Genestreti G, Tiseo M, Kenmotsu H, Kazushige W, Di Battista M, Cavallo G, et al. Outcomes of platinum-sensitive small-cell lung cancer patients treated with Platinum/Etoposide rechallenge: A multi-institutional retrospective analysis. Clin Lung Cancer (2015) 16(6):e223–8. doi: 10.1016/j.cllc.2015.04.006 26. Zhou J, Lu J, Gao C, Zeng J, Zhou C, Lai X, et al. Predicting the response to neoadjuvant chemotherapy for breast cancer: Wavelet transforming radiomics in mri. BMC Cancer (2020) 20(1):100. doi: 10.1186/s12885-020-6523-2 Cancer (2015) 16(6):e223–8. doi: 10.1016/j.cllc.2015.04.006 13. Plaja A, Moran T, Carcereny E, Saigi M, Hernandez A, Cucurull M, et al. Small- cell lung cancer long-term survivor patients: How to ﬁnd a needle in a haystack? Int J Mol Sci (2021) 22(24):13508. doi: 10.3390/ijms222413508 27. Chun SH, Suh YJ, Han K, Park SJ, Shim CY, Hong GR, et al. Differentiation of left atrial appendage thrombus from circulatory stasis using cardiac ct radiomics in patients with valvular heart disease. Eur Radiol (2021) 31(2):1130–9. doi: 10.1007/s00330-020-07173-1 14. He M, Chi X, Shi X, Sun Y, Yang X, Wang L, et al. Value of pretreatment serum lactate dehydrogenase as a prognostic and predictive factor for small-cell lung cancer patients treated with ﬁrst-line platinum-containing chemotherapy. Thorac Cancer (2021) 12(23):3101–9. doi: 10.1111/1759-7714.13581 28. Sala E, Mema E, Himoto Y, Veeraraghavan H, Brenton JD, Snyder A, et al. Unravelling tumour heterogeneity using next-generation imaging: Radiomics, radiogenomics, and habitat imaging. Clin Radiol (2017) 72(1):3–10. doi: 10.1016/ j.crad.2016.09.013 29. Wang L, Wang D, Zheng G, Yang Y, Du L, Dong Z, et al. Clinical evaluation and therapeutic monitoring value of serum tumor markers in lung cancer. Int J Biol Markers (2016) 31(1):e80–7. doi: 10.5301/jbm.5000177 15. Wang C, Jin S, Xu S, Cao S. High systemic immune-inﬂammation index (Sii) represents an unfavorable prognostic factor for small cell lung cancer treated with etoposide and platinum-based chemotherapy. Lung (2020) 198(2):405–14. doi: 10.1007/s00408-020-00333-6 15. Wang C, Jin S, Xu S, Cao S. References Gay CM, Stewart CA, Park EM, Diao L, Groves SM, Heeke S, et al. Patterns of transcription factor programs and immune pathway activation deﬁne four major subtypes of sclc with distinct therapeutic vulnerabilities. Cancer Cell (2021) 39 (3):346–60.e7. doi: 10.1016/j.ccell.2020.12.014 21. Fave X, Zhang L, Yang J, Mackin D, Balter P, Gomez D, et al. Delta-radiomics features for the prediction of patient outcomes in non-small cell lung cancer. Sci Rep (2017) 7(1):588. doi: 10.1038/s41598-017-00665-z 11 Frontiers in Oncology frontiersin.org
https://openalex.org/W2403517106	https://europepmc.org/articles/pmc4877980?pdf=render	English	null	Recurrent pulseless electrical activity in a patient with coronary vasospasm and supravalvular aortic stenosis: a case report	BMC cardiovascular disorders	2,016	cc-by	2,648	© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lother et al. BMC Cardiovascular Disorders (2016) 16:100 DOI 10.1186/s12872-016-0284-5 Lother et al. BMC Cardiovascular Disorders (2016) 16:100 DOI 10.1186/s12872-016-0284-5 DOI 10.1186/s12872-016-0284-5 Open Access Background We describe here for the first time the case of a pa- tient with the rare coincident finding of coronary vaso- spasm and supravalvular aortic stenosis leading to recurrent pulseless electrical activity cardiac arrests. Cardiac arrest due to pulseless electrical activity (PEA) is associated with poor survival and neurological outcome when compared to a shockable rhythm [1, 2]. To provide successful resuscitation it is fundamental to identify po- tentially reversible causes for cardiac arrest. The reasons for PEA are considered to be more often non-cardiac in- cluding hyperkalemia, hypoglycemia, bleeding or hypovol- emia, pulmonary embolism, pneumothorax and cardiac tamponade [2–4]. This is reflected by current treatment guidelines that recommend the use of laboratory testing or ultrasound during resuscitation [3]. However, car- diac disease needs to be considered as an alternative cause of refractory PEA. Extracorporeal cardiac life support is more and more commonly available and can be applied to allow diagnostic and treatment of potentially reversible causes [3]. These include also uncommon findings that might be missed by routine measures during resuscitation. Abstract Background: Pulseless electrical activity cardiac arrest is associated with poor outcomes and the identification of potentially reversible reasons for cardiac arrest is fundamental. Case presentation: We describe the case of a 46-year-old male with the rare coincidental finding of supravalvular aortic stenosis and coronary vasospasm leading to recurrent pulseless electrical activity cardiac arrest. Extracorporeal life support was successfully applied for hemodynamic stabilization. Supravalvular aorticstenosis underwent surgical repair. The patient survived five time resuscitation and was discharged after full neurological recovery. Conclusions: Coronary vasospasm and supravalvular aortic stenosis are rare but potentially reversible causes of pulseless electrical activity cardiac arrest. Extracorporeal life support allows accurate diagnostic and possibly therapy even of uncommon reasons for cardiac arrest. Keywords: Pulseless electrical activity, Cardiac arrest, Coronary vasospasm, Supravalvular aortic stenosis, ECLS Case presentation A 46-year-old male was successfully resuscitated for out- of-hospital cardiac arrest due to ventricular fibrillation (Fig. 1). Coronary angiography immediately after admis- sion to the county hospital revealed coronary vasospasm but no coronary artery disease (Fig. 2a). After intracor- onary administration of glycerol trinitrate the coronary system exhibited regular flow without any relevant sten- osis (Fig. 2b). Pulmonary embolism, aortic dissection and other potentially reversible causes were ruled out by computer-assisted tomography (CT) of the chest and lab-testing. Mild therapeutic hypothermia was estab- lished for 24 h. For aspiration pneumonia he was treated with ampicillin and sulbactam. Because of the development of a severe acute respira- tory distress syndrome five days after admission our ECMO team established a veno-venous extra-corporeal membrane oxygenation (vvECMO) system using a 31 Fr bi-caval cannula and transferred the patient to our * Correspondence: achim.lother@universitaets-herzzentrum.de 1Heart Center, Department of Cardiology and Angiology I, University of Freiburg, Hugstetter Str. 55, 79106 Freiburg, Germany Full list of author information is available at the end of the article * Correspondence: achim.lother@universitaets-herzzentrum.de 1Heart Center, Department of Cardiology and Angiology I, University of Freiburg, Hugstetter Str. 55, 79106 Freiburg, Germany Full list of author information is available at the end of the article Page 2 of 4 Lother et al. BMC Cardiovascular Disorders (2016) 16:100 Fig. 1 Time course. Schematic representation of relevant events. For cardiac arrests initial heart rhythm and time to return of spontaneous circulation are given. vv-ECMO, veno-venous extracorporeal membrane oxygenation; vva-ECMO, veno-veno-arterial extracorporeal membrane oxygenation; PEA, pulseless electrical activity; VF, ventricular fibrillation; ROSC, return of spontaneous circulation Fig. 1 Time course. Schematic representation of relevant events. For cardiac arrests initial heart rhythm and time to return of spontaneous circulation are given. vv-ECMO, veno-venous extracorporeal membrane oxygenation; vva-ECMO, veno-veno-arterial extracorporeal membrane oxygenation; PEA, pulseless electrical activity; VF, ventricular fibrillation; ROSC, return of spontaneous circulation Fig. 1 Time course. Schematic representation of relevant events. For cardiac arrests initial heart rhythm and time to return of spontaneous circulation are given. vv-ECMO, veno-venous extracorporeal membrane oxygenation; vva-ECMO, veno-veno-arterial extracorporeal membrane oxygenation; PEA, pulseless electrical activity; VF, ventricular fibrillation; ROSC, return of spontaneous circulation Fig. 2 Coronary vasospasm and supravalvular aortic stenosis leading to pulseless electrical activity cardiac arrest. Coronary angiography of the right coronary artery revealing coronary vasospasm (a) that was relieved by application of glycerol trinitrate (b). Case presentation ECG recording showing marked intraventricular block during pulseless electrical activity cardiac arrest (c) and a restored QRS (d) after vasodilatory therapy. 3D reconstruction of computer-assisted tomography (e) displaying the supravalvular aortic stenosis (black arrow) and double lumen ECMO cannula (white arrow). Ao, Aorta; LV, left ventricle; RA, right atrium Fig. 2 Coronary vasospasm and supravalvular aortic stenosis leading to pulseless electrical activity cardiac arrest. Coronary angiography of the right coronary artery revealing coronary vasospasm (a) that was relieved by application of glycerol trinitrate (b). ECG recording showing marked intraventricular block during pulseless electrical activity cardiac arrest (c) and a restored QRS (d) after vasodilatory therapy. 3D reconstruction of computer-assisted tomography (e) displaying the supravalvular aortic stenosis (black arrow) and double lumen ECMO cannula (white arrow). Ao, Aorta; LV left ventricle; RA right atrium Fig. 2 Coronary vasospasm and supravalvular aortic stenosis leading to pulseless electrical activity cardiac arrest. Coronary angiography of the right coronary artery revealing coronary vasospasm (a) that was relieved by application of glycerol trinitrate (b). ECG recording showing marked intraventricular block during pulseless electrical activity cardiac arrest (c) and a restored QRS (d) after vasodilatory therapy. 3D reconstruction of computer-assisted tomography (e) displaying the supravalvular aortic stenosis (black arrow) and double lumen ECMO cannula (white arrow). Ao, Aorta; LV, left ventricle; RA, right atrium Fig. 2 Coronary vasospasm and supravalvular aortic stenosis leading to pulseless electrical activity cardiac arrest. Coronary angiography of the right coronary artery revealing coronary vasospasm (a) that was relieved by application of glycerol trinitrate (b). ECG recording showing marked intraventricular block during pulseless electrical activity cardiac arrest (c) and a restored QRS (d) after vasodilatory therapy. 3D reconstruction of computer-assisted tomography (e) displaying the supravalvular aortic stenosis (black arrow) and double lumen ECMO cannula (white arrow). Ao, Aorta; LV, left ventricle; RA, right atrium Lother et al. BMC Cardiovascular Disorders (2016) 16:100 Page 3 of 4 Page 3 of 4 angina pectoris are at higher risk of sudden cardiac death and most often cardiac arrest occurs due to tachy- cardic ventricular arrhythmia [5]. Nevertheless, as re- ported in this case, presentation might be different. There are only few previous reports of PEA associated with coronary vasospasm [5–8], one of them describing recurrent cardiac arrests [6]. Abbreviations Abbreviations CT, computer-assisted tomography; ECLS, extracorporeal life support; ECMO, extracorporeal membrane oxygenation; PEA, pulseless electrical activity; ROSC, return of spontaneous circulation. Discussion We report a case of recurrent cardiac arrests due to cor- onary vasospasm and supravalvular aortic stenosis. Acknowledgements The article processing charge was funded by the German Research Foundation (DFG) and the University of Freiburg in the funding programme Open Access Publishing. Additional file Additional file 1: Animated 3D reconstruction of computer-assisted tomography displaying the supravalvular aortic stenosis and double lumen ECMO cannula. (MP4 10783 kb) The patient had full neurological recovery and was discharged to follow-up treatment. A cardiac defibrillator was implanted for secondary prevention of the initial VF. Case presentation The patient described in this case had a supravalvular aortic stenosis which might have a) aggravated low cardiac output during coronary vasospasm and b) triggered coronary vasospasm due to elevated pressures in the aortic bulbus. Supravalvular aortic stenosis is rare in general population but frequent in Williams-Beuren syndrome, a genetic disorder that is associated with an increased rate of cardiovascular events and death [9]. We did not find one of the known Williams-Beuren mutations or other syndromic presen- tation in this patient. center (Fig. 1). Another CT of the chest showed progres- sive bi-pulmonary infiltration but no pulmonary embolism. Subsequently sufficient oxygenation was delivered by com- bined ECMO treatment and lung protective ventilation. g During his stay in the intensive care unit the patient underwent recurrent episodes of cardiac arrest with pulseless electrical activity (PEA, Fig. 1, Fig. 2c) that were not associated with hypoxemia as he was on full ECMO support. There was no evidence for other causes for PEA like tension pneumothorax or embolism in the bedside ultrasonography and echocardiography. Distur- bances in electrolytes could be ruled out be point-of-care testing. As we could not reach a return of spontaneous circulation after 15 min of resuscitation we successfully modified the established vvECMO to a veno-veno-arterial ECMO system by introduction of a 17 Fr delivering can- nula into the right femoral artery in order to provide extracorporeal life support (ECLS). This procedure could be performed within 15 min without complications under ongoing CPR. During the following episode PEA was suc- cessfully terminated when coronary vasospasm was re- lieved by i.v. administration of glycerol trinitrate and verapamil while circulation was maintained by ECLS (Fig. 1, Fig. 2d). In total, the patient had experienced five episodes of resuscitation till then. A combined long-acting nitrate and calcium channel blocker vasodilatory therapy was established and successfully prevented further events. ECMO therapy could be stopped 14 days after implant- ation, weaning from mechanical ventilation was successful after 35 days. Veno-arterial ECMO is considered as the preferred method for extracorporeal life support to allow further diagnosis and therapeutic measures when standard life support fails to achieve ROSC [3]. Percutaneous cannu- lation can be performed fast and with a low rate of com- plications in the intensive care unit or the cath lab by experienced teams [10]. In this case veno-veno-arterial ECMO was applied to treat concurrent ARDS and car- diac arrests. Conclusion In conclusion, the coincidence of coronary vasospasm and supravalvular aortic stenosis lead to recurrent pulseless electrical activity cardiac arrest in this case. Unusual causes for refractory cardiac arrest should be considered during and after resuscitation. Extracorporeal life support provides the opportunity for accurate diagnostic and ther- apy when routine diagnostic measures are not sufficient. The patient has no family history of cardiac death. He underwent patch-repair for supravalvular aortic stenosis during childhood. On day 6 transthoracic and transesoph- ageal echocardiography revealed a remaining supravalvu- lar aortic stenosis (Vmax 4 m/s, mean pressure gradient 37 mmHg) that was confirmed by CT (Fig.1e, Additional file 1: Video S1). Older medical records showed, that the re-stenosis had been already detected during a rehabilita- tion exam 20 years ago but did not cause any symptoms. Since hemodynamics remained stable after initiation of vasodilatory therapy surgical repair was postponed until respiratory recovery. The stenosis was successfully re- lieved by supracoronary ascending aortic replacement on day 44 (Fig. 1). Competing interests The authors declare that they have no competing interests. Consent for publication f d Written informed consent was obtained from the patient for publication of this case report including clinical data and any accompanying images. Written informed consent was obtained from the patient for publication of this case report including clinical data and any accompanying images. Authors’ contributions AL and TW created the concept, performed acquisition, analysis and interpretation of data and wrote the manuscript. HHO provided image content. FB, HHO and CB were involved in drafting the manuscript and revising it critically for important intellectual content. All authors gave final The patient survived five episodes of cardiac arrest, four of them with PEA which is an uncommon compli- cation of vasospastic angina. Patients with vasospastic Page 4 of 4 Lother et al. BMC Cardiovascular Disorders (2016) 16:100 Lother et al. BMC Cardiovascular Disorders (2016) 16:100 approval of the version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Author details 1 1Heart Center, Department of Cardiology and Angiology I, University of Freiburg, Hugstetter Str. 55, 79106 Freiburg, Germany. 2Heart Center, 1Heart Center, Department of Cardiology and Angiology I, University of Freiburg, Hugstetter Str. 55, 79106 Freiburg, Germany. 2Heart Center, Department of Cardiovascular Surgery, University of Freiburg, Freiburg, Germany 3District Hospital Loerrach Medical Clinic Loerrach Germany Department of Cardiovascular Surgery, University of Freiburg, Freiburg, Germany. 3District Hospital Loerrach, Medical Clinic, Loerrach, Germany. Department of Cardiovascular Surgery, University of Freiburg, Freiburg, Germany. 3District Hospital Loerrach, Medical Clinic, Loerrach, Germany. Received: 29 January 2016 Accepted: 14 May 2016 Received: 29 January 2016 Accepted: 14 May 2016 References 1. Meaney PA, Nadkarni VM, Kern KB, Indik JH, Halperin HR, Berg RA. Rhythms and outcomes of adult in-hospital cardiac arrest. Crit Care Med. 2010;38(1):101–8. 2. Wallmuller C, Meron G, Kurkciyan I, Schober A, Stratil P, Sterz F. Causes of in- hospital cardiac arrest and influence on outcome. Resuscitation. 2012;83(10): 1206–11. 3. Soar J, Nolan JP, Bottiger BW, Perkins GD, Lott C, Carli P, Pellis T, Sandroni C, Skrifvars MB, Smith GB, et al. European resuscitation council guidelines for resuscitation 2015: section 3. Adult advanced life support. Resuscitation. 2015;95:100–47. 3. Soar J, Nolan JP, Bottiger BW, Perkins GD, Lott C, Carli P, Pellis T, Sandroni C, Skrifvars MB, Smith GB, et al. European resuscitation council guidelines for resuscitation 2015: section 3. Adult advanced life support. Resuscitation. 2015;95:100–47. 4. Desbiens NA. Simplifying the diagnosis and management of pulseless electrical activity in adults: a qualitative review. Crit Care Med. 2008;36(2):391–6. 5. Matsue Y, Suzuki M, Nishizaki M, Hojo R, Hashimoto Y, Sakurada H. Clinical implications of an implantable cardioverter-defibrillator in patients with vasospastic angina and lethal ventricular arrhythmia. J Am Coll Cardiol. 2012;60(10):908–13. 6. Martin J, Sanfilippo F, Bradlow W, Sabharwal NK. Severe vasospastic angina complicated by multiple pulseless electrical activity arrests. Lancet. 2013; 382(9890):478. 6. Martin J, Sanfilippo F, Bradlow W, Sabharwal NK. Severe vasospastic angina complicated by multiple pulseless electrical activity arrests. Lancet. 2013; 382(9890):478. 7. Gaikwad NJ, McNamara M, Batra R, Aroney G, Jayasinghe R. An unusual case of pulseless electrical activity arrest associated with Prinzmetal’s angina. Crit Care Resusc. 2010;12(4):269–72. 7. Gaikwad NJ, McNamara M, Batra R, Aroney G, Jayasinghe R. An unusual case of pulseless electrical activity arrest associated with Prinzmetal’s angina. Crit Care Resusc. 2010;12(4):269–72. 8. Seniuk W, Mularek-Kubzdela T, Grygier M, Grajek S, Cieslinski A. Cardiac arrest related to coronary spasm in patients with variant angina: a three-case study. J Intern Med. 2002;252(4):368–76. 9. Pober BR. Williams-Beuren syndrome. N Engl J Med. 2010;362(3):239–52. y g 10. Conrad SA, Grier LR, Scott LK, Green R, Jordan M. Percutaneous cannulation for extracorporeal membrane oxygenation by intensivists: a retrospective single-institution case series. Crit Care Med. 2015;43(5):1010–5. 7. Gaikwad NJ, McNamara M, Batra R, Aroney G, Jayasinghe R. An unusual case of pulseless electrical activity arrest associated with Prinzmetal’s angina. Crit Care Resusc. 2010;12(4):269–72. 8. Seniuk W, Mularek-Kubzdela T, Grygier M, Grajek S, Cieslinski A. Cardiac arrest related to coronary spasm in patients with variant angina: a three-case study. J Intern Med. 2002;252(4):368–76. Ethics approval and consent to participate No ethics approval was required in this case. Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit and we will help you at every step: • We accept pre-submission inquiries
https://openalex.org/W3215891694	https://univoak.eu/islandora/object/islandora:126402/datastream/PDF/download/citation.pdf	English	null	Hybrid renewable energy systems based on micro-cogeneration	Energy reports	2,022	cc-by	5,439	Abstract Hybrid renewable energy systems (HRES) are seeing as a solution to overcome the fluctuation and randomness of certain renewable energy sources, such as solar and wind power. Coupling the fluctuating renewable energy sources with the controllable sources, such as biomass-fueled micro-cogeneration, constitute an HRES that significantly reduces CO2 emissions and primary energy consumption. The purpose of this work is to review research works on hybrid renewable energy systems based on micro- cogeneration and to present a case study of optimizing a solar-based micro-cogeneration system. First, renewable energy-fueled micro-cogeneration systems are presented according to the prime mover technology: Stirling engine, organic Rankine cycle and photovoltaic-thermal (PVT). The different prime movers are assessed according to their advantages, disadvantages and market availability. Next, several research works on hybrid renewable energy systems including solar and micro-cogeneration technologies are summarized and key findings are highlighted. Finally, the results of the case study are presented for reasoning the necessity of system hybridization. The results indicated that more experimental data on HRES and research effort on energy management strategies and stochastic optimization models are required. The results of the case study showed maximum thermal and electrical reliability of 68% and 70%, respectively. The optimized PVT/battery/thermal storage system was not able to cover all energy demand of the case study but supporting heat and electricity sources are required. © 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems Engineering, CPESE, 2021. Keywords: Renewable energy sources; Energy conversion; Solar energy; Hybrid renewable energy systems; Micro- management Keywords: Renewable energy sources; Energy conversion; Solar energy; Hybrid renewable energy systems; Micro-cogeneration; Microgrid energy management Available online at www.sciencedirect.com ScienceDirect 2021 8th International Conference on Power and Energy Systems Engineering (CPESE 2021), 10–12 September 2021, Fukuoka, Japan 10–12 September 2021, Fukuoka, Japan Hybrid renewable energy systems based on micro-cogeneration Sonja Kallio, Monica Siroux∗ INSA Strasbourg ICUBE, University of Strasbourg, Strasbourg, France Received 26 October 2021; accepted 8 November 2021 Available online xxxx Sonja Kallio, Monica Siroux∗ INSA Strasbourg ICUBE, University of Strasbourg, Strasbourg, France Received 26 October 2021; accepted 8 November 2021 Available online xxxx https://doi.org/10.1016/j.egyr.2021.11.158 2352-4847/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/). Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems Available online at www.sciencedirect.com ScienceDirect Energy Reports 8 (2022) 762–769 www.elsevier.com/locate/egyr Available online at www.sciencedirect.com ScienceDirect Energy Reports 8 (2022) 762–769 www.elsevier.com/locate/egyr 1. Introduction The building sector has a high impact on greenhouse gas emissions and final energy consumption in the European Union (EU) by being the largest energy end-use sector with a share of 41% [1]. The micro combined heat and power (micro-CHP), or cogeneration, units produce simultaneously decentralized heat and power from a single fuel source at high efficiency. The building integrated micro-cogeneration systems are in the key role in reaching the primary energy and pollutant emissions reduction targets of the EU [2]. ∗Corresponding author. E-mail addresses: sonja.kallio@insa-strasbourg.fr (S. Kallio), monica.siroux@insa-strasbourg.fr (M. Siroux). Corresponding author. E-mail addresses: sonja.kallio@insa-strasbourg.fr (S. Kallio), monica.siroux@insa-strasbourg.fr (M. Siroux). g resses: sonja.kallio@insa-strasbourg.fr (S. Kallio), monica.siroux@insa-strasbourg.fr (M. Siroux). https://doi.org/10.1016/j.egyr.2021.11.158 2352-4847/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/). 2352-4847/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/). Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems Engineering, CPESE, 2021. Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems Engineering, CPESE, 2021. Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Nomenclature Q thermal power (W) R reliability SOC state of charge T temperature (K, ◦C) η efficiency Hybrid renewable energy systems integrate two or more renewable energy sources with or without conventional energy source to produce heat and electricity to satisfy a certain end-use demand [3]. The common features for these systems are decentralized set-up, high renewable energy share, flexible operation to follow local energy demand and small- or micro-scale system components. Due to the multitude of possible combinations, hybrid energy systems can deliver highly efficient solutions for energy generation. The targets of using hybrid energy systems are to increase reliability, to support fluctuating renewable energy sources, to improve overall efficiency and fuel flexibility. Additionally, these systems reduce costs, fossil fuels, primary energy use and CO2 emissions. HRES can operate as a grid-connected system or stand-alone system in rural areas without an existing power grid. The HRES are widely studied in the literature. Shivarama et al. (2015) [4] presented a comprehensive review on hybrid renewable energy systems. They considered modeling, analysis, optimal sizing, energy management and control aspects of the HRES. 1. Introduction If the technology types of the HRES are selected appropriately, the system can be used to produce cooling, heating and power, simultaneously. Applying combined cooling, heating and power (CCHP) systems can reach an overall efficiency even more than 90% compared to a conventional centralized power plant, which has an efficiency of 30%–45% [5]. Gu et al. (2014) [5] presented the review of modeling, planning and optimal energy management of CCHP micro-grids. They defined that a typical CCHP system consists of four main elements: the renewable energy source, prime mover, energy storage and load. Wang et al. (2020) reviewed current research of small-scale hybrid solar-assisted CCHP systems. They highlighted that more research is required on system integration evaluation and advanced energy management strategies. Wegener et al. (2018) [6] presented a review on small-scale CCHP systems covering biomass-fired systems and solar extensions. The results indicated the high energetic and environmental potential of biomass-fired CHP systems with solar extensions. However, they pointed out that more experimental data on the system operation is required. In this given framework, a research gap has been indicated and more research effort is required in biomass-fired micro-scale CHP systems with solar support to enable highly efficient and 100% renewable energy systems in the building sector as well. The present work provides a research contribution by summarizing possible renewable energy-based micro-cogeneration prime movers and by reviewing research works on HRES based on micro- cogeneration and solar technologies. Finally, a case study of solar-based micro-cogeneration system with energy storages is presented for reasoning the necessity of system hybridization. 2. Renewable energy based micro-cogeneration Prime movers are used to transform fuel energy to mechanical work and heat as a by-product. The prime movers of the micro-CHP systems can be divided into the following three categories: internal, external and no combustion process [7]. The most developed, used and researched technology is internal combustion engine (ICE) based micro- CHP systems [6]. However, the external combustion-based systems, such as Stirling and organic Rankine cycle (ORC) have better fuel adaptability to use different renewable sources, such as solid biomass, waste heat and solar energy [7]. Fuel cells and photovoltaic-thermal (PVT) collectors are based on the no combustion process. In the fuel cells, the chemical energy of a fuel is converted to electricity through a chemical reaction with oxygen. The PVT collectors convert solar radiation into heat and electricity simultaneously from the same installed area. An assessment of different micro-CHP prime movers presented in this review is shown in Table 1. The selection of the considered prime movers was based on their adaptability to use solid biomass and solar as fuel and market penetration in the building sector. The assessment is based on the literature findings. 763 Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Table 1. Assessment of renewable energy based micro-CHP prime mover technologies. Prime mover Stirling engine ORC PVT Advantages Low maintenance, noise, vibration and emissions [8] High fuel flexibility [7,9] Noise level and electricity/heat ratio suitable for residential use [7,10] High thermal efficiency [9] Good partial load performance [6] Low-grade heat use starting from 50 ◦C [6,11] High fuel flexibility [8] Easy to retrofit the existing heating device [12] High partial load efficiency [6] Well developed as flat plate collector [13] Simple operation without moving parts [13] Wide range of possible applications [13] Disadvantages Low electrical efficiency [10,14] High investment costs [6,8] High investment costs [6] Relatively low electrical efficiency and waste heat recovery [6,12] Challenges in efficient downscaling [8] Difficulty of sun tracking with flat plate [13] Fuel types Biomass, natural gas, solar radiation [15] Waste heat, geothermal, biomass, solar radiation [8,12,16] Solar energy State of technology Commercially available [16,17] R&D and some commercially available [16,17] R&D and commercially available [13] Market availability ÖkoFEN, Whispergen, BAXI, Sunmachine, Genoastirling S.r.l. Rank®, Kaymacor ORChidea, Viking Development Group DualSun, Abora Solar, Naked Energy, Solarus Table 1. Assessment of renewable energy based micro-CHP prime mover technologies. Prime mover Stirling engine ORC 3. Hybrid renewable energy system Various technology types are identified in the literature for the planning and design of hybrid energy systems. Their suitability depends on available fuel sources on the generation side, the availability of a grid connection or a heating network, and the considered application scenario. Within the scope of the HRES, the utilization of the following technologies can be considered: solar technologies: photovoltaic (PV), solar thermal and PVT systems, mini-wind turbines, micro combined heat and power (micro-CHP) systems, heat pumps and thermal chillers. The following storage technologies are often deployed in the HRES to facilitate energy management in the system: sensible and latent heat storages, electric battery storages, thermally activated building systems and electric vehicles as flexible electric storages. l The above-listed components can form different configurations for the HRES depending on the considered application and load demand. As an example, photovoltaic electricity production can be used to support the electricity consumption of the heat pump. On the other hand, the combustion-based micro-CHP can support fluctuating solar energy production and activate thermal chiller during the cooling period. Different configurations for the HRES have been studied in the literature. Zabalaga et al. (2020) [9] presented the performance analysis of the HRES including photovoltaic (PV), biomass Stirling engine and battery system. They compared the HRES with the PV/diesel engine/battery hybrid system and performed the analysis in terms of energy efficiency, environmental sustainability and economic feasibility. Their results revealed that the HRES performed better in terms of each indicator in the analysis. Calise et al. (2016) [18] studied a novel solar-assisted CCHP system, which coupled PVT collectors with a reversible heat pump and adsorption chiller. The aim was to design, dynamic simulation and analyzing the energetic, environmental and economic performance of the system. In this study, the solid biomass-fueled micro-CHP systems supported by solar technologies are considered as components to form a hybrid renewable energy system with energy storages. The selection is based on the exergy approach which indicates the highest overall exergy efficiency for the HRES components based on the combined heat and power production [19,20]. Additionally, the literature review has shown that coupling the micro-CHP system with solar technologies can overcome the intermittency issues of solar energy production, increase self-consumption and reduce primary energy use and fuel costs. Table 2 summarizes the cases of the hybrid solar supported micro-CHP systems in the literature. 4. Case study The preliminary study was conducted considering residential electricity and domestic hot water production by a photovoltaic-thermal system integrated with thermal and electrical storages. The hourly demand profiles with 764 Energy Reports 8 (2022) 762–769 Energy Reports 8 (2022) 762–769 Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Table 2. The cases in the literature of the hybrid renewable energy systems based on micro-CHP and solar support. Prime mover Power output Fuel Solar support Methods Evaluation Key findings Ref. Stirling engine Pel = 1 kW, Pth = 7 kW Natural gas Biomass possibility PV Flat plate solar thermal To produce heat and electricity Experimental study Operation data of 36 days in Spain Matlab Comparison to a reference system including electric grid and diesel oil boiler Typical residential domestic hot water and electricity demand profiles ηel ηth CO2 emissions Higher performance is expected if heat demand is extended to space heating and cooling The system reliability increased with thermal and electrical storages, and micro-CHP unit CO2 emissions reduced by 36% 76% of total energy demand is covered Biomass use proposed for further CO2 reduction [21] Stirling engine Pel = 1 kW, Pth = 26 kW Natural gas Biomass possibility PV To produce electricity Comparison of different PV powers and battery capacities with Stirling micro-CHP Experimental data of PV and micro-CHP installation Demand self-coverage index Sold share index Energy stored share Self-consumption index Net present value The system enables significant improvement in self-sufficiency High investment costs High ecological effect Research on energy management and control required [22] ORC Pel = 40 kW, Pth = 310 kW Biomass PVT To produce heat and electricity Matlab Transient techno-economic modeling Design optimization Sport center in Bari, Italy Thermo-economic analysis Payback time 11.5 years 100% renewable energy supply with hybrid system Electricity production capability increased significantly [23] ORC Pel = 14 kW, Pth = 70 kW Biomass PV To produce electricity System modeling and analysis Residential building block Electricity, space heating and domestic hot water Energy analysis Self-consumption Fuel savings Yearly electrical self-consumption 84.9% with hybrid system, 32.5% with PV only. 4. Case study 54.9% of thermal energy demand covered Significant natural gas savings ORC overcomes the fluctuation of solar energy source [24] PVT Natural gas boiler as back-up Dynamic simulation in TRNSYS Optimization of the system configuration Hotel in Naples, Italy Energy, economic and environmental analysis Sensitivity analysis Primary energy savings 57.66% CO2 emissions reduction of 51.63% Simple payback time 11.64 years [25] daily average demands of 24.6 kWh and 7.5 kWh for electricity and thermal energy, respectively, were used. The study was conducted under meteorological conditions of Strasbourg, France. To this purpose, the dynamic system model was built into Matlab/Simulink environment and a multi-objective optimization with genetic algorithm wasi ses in the literature of the hybrid renewable energy systems based on micro-CHP and solar support. Solar support Methods Evaluation Key findings daily average demands of 24.6 kWh and 7.5 kWh for electricity and thermal energy, respectively, were used. The study was conducted under meteorological conditions of Strasbourg, France. To this purpose, the dynamic system model was built into Matlab/Simulink environment and a multi-objective optimization with genetic algorithm was performed. The aim of the optimization was to find a Pareto optimal set of system designs to maximize self- consumption and autonomy of the system over year with the minimum initial investment costs. The considered system is presented in Fig. A.1 in Appendix. daily average demands of 24.6 kWh and 7.5 kWh for electricity and thermal energy, respectively, were used. The study was conducted under meteorological conditions of Strasbourg, France. To this purpose, the dynamic system model was built into Matlab/Simulink environment and a multi-objective optimization with genetic algorithm was performed. The aim of the optimization was to find a Pareto optimal set of system designs to maximize self- consumption and autonomy of the system over year with the minimum initial investment costs. The considered system is presented in Fig. A.1 in Appendix. The dynamic model of the PVT collector have been presented in the literature [26–30] and used to evaluate the thermal and electrical performance of the collector. In this case study, glazed water cooled PVT collectors with 765 Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Table 3. The decision variables for the multi-objective optimization problem. Decision variable Bounds Unit Coolant mass flow rate, ˙m 60 ≤x(1) ≤130 kg/h Nr. of PVT in series, N 1 ≤x(2) ≤6 – Nr. 4. Case study of PVT rows, M 1 ≤x(3) ≤6 – Battery capacity, CB 4 ≤x(4) ≤14 kWh Thermal storage volume 0.1 ≤x(5) ≤0.4 m3 Table 3. The decision variables for the multi-objective optimization problem. sheet-and-tube thermal absorbers were used and modeled with the main heat transfer mechanisms between the PVT layers. The key governing equations of the PVT collector and thermal storage are presented as follows: mg ∗cg ∗dTg/dt = Qg,s + Qg−e,CV + Qg−sky,RD + Qg−pv,CV + Qg−pv,RD (1) m pv ∗cpv ∗dTpv/dt = Q pv,s + Q pv−g,CV + Q pv−g,RD + Q pv−f,CV −E (2) m f ∗c f ∗dT f /dt = Q f −pv,CV + Q f (3) mT S ∗cw ∗dTT S/dt = Q f −T S + Qaux + Q DHW + QT S,loss (4) mg ∗cg ∗dTg/dt = Qg,s + Qg−e,CV + Qg−sky,RD + Qg−pv,CV + Qg−pv,RD (1) m pv ∗cpv ∗dTpv/dt = Q pv,s + Q pv−g,CV + Q pv−g,RD + Q pv−f,CV −E (2) m f ∗c f ∗dT f /dt = Q f −pv,CV + Q f (3) mT S ∗cw ∗dTT S/dt = Q f −T S + Qaux + Q DHW + QT S,loss (4) (3) (4) (4) In Eqs. (1)–(4) m, c, T , Q and E indicate mass, specific heat capacity, temperature, heat transfer rate and electricity production, respectively. The following subscripts: “g”, “s”, “e”, “CV ”, “sky”, “RD”, “pv”, “ f ”, “T S”, “w”, “aux”, “DHW” and “loss” refer to glass cover, solar energy, environment, convection, sky, radiation, PV module, fluid, thermal storage, water, auxiliary, domestic hot water and heat losses. The more detailed description of the PVT governing equations and model validation are presented in the Authors’ previous work [28].i The PV converts a fraction of the solar radiation into the electricity. The electrical efficiency of the PV depends linearly on the PV temperature Tpv, the temperature coefficient βPV and on the efficiency ηSTC at standard conditions Tref . The electrical efficiency ηel is calculated according to the following relation: ηel(T ) = ηSTC [ 1 −βPV ( Tpv −Tref )] (5) The battery model was based on the load balance between the electricity production and demand. If the electricity production was higher than demand, the battery was charged and discharged if the demand was higher. The surplus electricity was sold to the grid. 4. Case study However, selling and buying from the grid was aimed to be minimized in order to maximize self-consumption and autonomy of the system. The battery state of charge SOCB is presented as follows: SOCB (t) = SOCB (t −∆t) + ηch Pch (t) ∆t/CB −ηdis Pdis (t) ∆t/CB (6) SOCB (t) = SOCB (t −∆t) + ηch Pch (t) ∆t/CB −ηdis Pdis (t) ∆t/CB (6) where Pch, Pdis, CB, ηch and ηdis are charging power, discharging power, battery capacity, charging and discharging efficiency, respectively. i The multi-objective design optimization problem was formulated to maximize the thermal and electrical reliability Rthermal and Relectric, respectively, and minimize the initial investment costs of the considered system, simultaneously. The inflation was not considered in the case study. The Matlab function called gamultiobj of the Global Optimization Toolbox was applied to create a set of optimal solutions in the space of decision variables on the Pareto front. The function uses a controlled, elitist genetic algorithm, which is a variant of non-dominated sorting genetic algorithm-II (NSGA-II). The detailed description of the algorithm is presented by Deb (2001) in [31]. Table 3 presents the selected decision variables to be optimized by minimizing the objective functions with NSGA-II optimization algorithm. The bounds of the variables were selected based on the previous research on the PVT collectors [28] and on the market availability of the components for residential buildings. Table 4 present the estimated prices of the system components used in the optimization. As a result, the Pareto front was obtained for the optimization problem and is presented in Fig. 1. The optimization algorithm found 18 non-dominating optimal solutions for the component sizing of the considered renewable energy system. The optimal solutions can be used by the decision-makers to select the final optimal solution depending on the application and constraints, such as available roof area or budget. Three solutions were selected from Fig. 1 based on the increasing investment costs. These solutions are presented in Table 5. Within the optimal solutions the investment costs varied from 8172 C to 40877 C while the thermal reliability increased from 0.48 to 0.68 and electrical reliability from 0.02 to 0.7. 766 Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Table 4. The component prices of the system. Component Price Ref. PVT collector 325 C/m2 DualSun company Thermal storage 2.95 C/dm3 [32] Electrical storage 1143 C/kWh Enphase company Fig. 1. 4. Case study The Pareto front of the optimal solutions in the objective space. Fig. 1. The Pareto front of the optimal solutions in the objective space. Table 5. The selected optimal solutions from the Pareto front. Solution ˙m (kg/h) PVT in series PVT rows CB (kWh) VT S (m3) Rthermal Relectric Inv. (C) 1 108 6 6 14 0.5 0.68 0.7 40877 2 64 6 2 9 0.2 0.57 0.41 18677 3 111 1 3 4 0.1 0.5 0.09 6817 Table 5. The selected optimal solutions from the Pareto front. The results show that the thermal and electrical energy demands of the considered case study under residential constraints, such as maximum available roof area, cannot be covered by the PVT system only. The results show also that the thermal and electrical storages increase the reliability of the system significantly. However, auxiliary heat and electricity sources are required to support fluctuating solar energy production and to cover the remaining part of the energy demand. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. 5. Conclusion In the present work, the renewable energy-based micro-cogeneration prime movers (Stirling engine, ORC and PVT collector) were presented. The advantages and disadvantages of these technologies were assessed. Based on the assessment, the Stirling engine micro-CHP and PVT collectors have the highest potential for renewable cogeneration in the residential building sector. The literature review on the HRES including micro-cogeneration and solar support on the building sector was conducted and the key findings were highlighted. It was seen that photovoltaic and solar thermal technologies were integrated with Stirling engine and ORC systems. In more advanced hybridization, the PVT was used as a solar extension. Based on the literature, the system reliability was increased with the hybridization and CO2 emissions were reduced. Additionally, the electricity production capability was increased significantly with PVT and the fuel-fired micro-CHP overcame the fluctuation of solar energy sources. However, more experimental data on hybridization and research on advanced energy management strategies are required. The results of the case study including PVT/battery/thermal storage system revealed that hybridization with high efficient and fuel-fired heat and electricity generation is required. In this case, the coupling of the PVT system 767 Energy Reports 8 (2022) 762–769 S. Kallio and M. Siroux Fig. A.1. The system layout of the case study. Fig. A.1. The system layout of the case study. with the biomass-fueled Stirling engine micro-CHP system should be considered to cover residential electricity, space heating and domestic hot water demands. The Stirling micro-CHP can provide a stable baseload and support fluctuating solar energy production. The coupling of these technologies can form a hybrid renewable energy system that provides 100% renewable energy for buildings. with the biomass-fueled Stirling engine micro-CHP system should be considered to cover residential electricity, space heating and domestic hot water demands. The Stirling micro-CHP can provide a stable baseload and support fluctuating solar energy production. The coupling of these technologies can form a hybrid renewable energy system that provides 100% renewable energy for buildings. Acknowledgment The authors would like to thank Interreg V Rhin sup´erieur ACA-MODES project for their support and funding of this research. Appendix See Fig. A.1. See Fig. A.1. References [1] Doroudchi E, Alanne K, Ö Okur, Kyyrä J, Lehtonen M. Approaching net zero energy housing through integrated EV. Sustain Cities Soc 2018;38:534–42. http://dx.doi.org/10.1016/j.scs.2018.01.042. [1] Doroudchi E, Alanne K, Ö Okur, Kyyrä J, Lehtonen M. Approaching net zero energy housing through integrated EV. Sustain Cities Soc 2018;38:534–42. http://dx.doi.org/10.1016/j.scs.2018.01.042. [1] Doroudchi E, Alanne K, Ö Okur, Kyyrä J, Lehtonen M. Approaching net zero energy housing through integrated EV. Sustain Cities Soc 2018;38:534–42. http://dx.doi.org/10.1016/j.scs.2018.01.042. p g j [2] González-Pino I, Pérez-Iribarren E, Campos-Celador A, Terés-Zubiaga J, Las-Heras-Casas J. Modelling and experimental characterization of a stirling engine-based domestic micro-CHP device. Energy Convers Manag 2020;225:113429. http://dx.doi.org/10.1016/j.enconman. 2020.113429. [2] González-Pino I, Pérez-Iribarren E, Campos-Celador A, Terés-Zubiaga J, Las-Heras-Casas J. Modelling and experimental characterization of a stirling engine-based domestic micro-CHP device. Energy Convers Manag 2020;225:113429. http://dx.doi.org/10.1016/j.enconman. 2020.113429. [3] Bajpai P, Dash V. Hybrid renewable energy systems for power generation in stand-alone applications: A review. Renew Sustain Energy Rev 2012;16:2926–39. http://dx.doi.org/10.1016/j.rser.2012.02.009. [3] Bajpai P, Dash V. Hybrid renewable energy systems for power generation in stand-alone applications: A review. Renew Sustain Energy Rev 2012;16:2926–39. http://dx.doi.org/10.1016/j.rser.2012.02.009. [4] Krishna KShivarama, Kumar KSathish. A review on hybrid renewable energy systems. Renew Sustain Energy Rev 2015;52:907–16. http://dx.doi.org/10.1016/j.rser.2015.07.187. [4] Krishna KShivarama, Kumar KSathish. A review on hybrid renewable energy systems. Renew Sustain Energy Rev 2015;52:907–16. http://dx.doi.org/10.1016/j.rser.2015.07.187. [5] Gu W, Wu Z, Bo R, Liu W, Zhou G, Chen W, et al. Modeling, planning and optimal energy managem heating and power microgrid: A review. Int J Electr Power Energy Syst 2014;54:26–37. http://dx.doi.org/10 [6] Wegener M, Malmquist A, Isalgué A, Martin A. Biomass-fired combined cooling, heating and power for small scale applications – A review. Renew Sustain Energy Rev 2018;96:392–410. http://dx.doi.org/10.1016/j.rser.2018.07.044. [6] Wegener M, Malmquist A, Isalgué A, Martin A. Biomass-fired combined cooling, heating and power for small scale applications – A review. Renew Sustain Energy Rev 2018;96:392–410. http://dx.doi.org/10.1016/j.rser.2018.07.044. gy ; p g j [7] Martinez S, Michaux G, Salagnac P, Bouvier JL. Micro-combined heat and power systems (micro-CHP) based on renewable energy sources. Energy Convers Manag 2017;154:262–85. http://dx.doi.org/10.1016/j.enconman.2017.10.035. gy p g j [7] Martinez S, Michaux G, Salagnac P, Bouvier JL. Micro-combined heat and power systems (micro-CHP) based on renewable energy sources. Energy Convers Manag 2017;154:262–85. http://dx.doi.org/10.1016/j.enconman.2017.10.035. [7] Martinez S, Michaux G, Salagnac P, Bouvier JL. Micro-combined heat and power systems (micro-CH sources. Energy Convers Manag 2017;154:262–85. http://dx.doi.org/10.1016/j.enconman.2017.10.035.i [8] Padinger R, Aigenbauer S, Schmidl C. References http://dx.doi.org/10.1016/j.apenergy.20 [16] Zhu S, Yu G, Jongmin O, Xu T, Wu Z, Dai W, et al. Modeling and experimental investigation of a free piston stirling engine based micro-combined heat and power system. Appl Energy 2018;226:522–33. http://dx.doi.org/10.1016/j.apenergy.2018.05.122. cro-combined heat and power system. Appl Energy 2018;226:522–33. http://dx.doi.org/10.1016/j.apenergy.2018.0 [17] Mascuch J, Novotny V, Spale J, Vodicka V, Zeleny Z. Experience from set-up and pilot operation of an in-house developed biomass-fired ORC microcogeneration unit. Renew Energy 2021;165:251–60. http://dx.doi.org/10.1016/j.renene.2020.11.021. 18] Calise F, d’Accadia MDentice, Figaj RD, Vanoli L. A novel solar-assisted heat pump driven by photovoltaic/t simulation and thermoeconomic optimization. Energy 2016;95:346–66. http://dx.doi.org/10.1016/j.energy.201 19] Wall G. Exergy – a useful concept within resource accounting. Report no 77-42, 1977. 20] Gong M, Wall G. Life cycle exergy analysis of solar energy systems. J Fundam Renew Energy Appl 201 [21] Auñón Hidalgo JA, Sidrach-de Cardona M, Auñón Rodríguez F. Performance and CO2 emissions assessment of a novel combined solar photovoltaic and thermal, with a stirling engine micro-CHP system for domestic environments. Energy Convers Manag 2021;230:113793. http://dx.doi.org/10.1016/j.enconman.2020.113793. [22] Kotowicz J, Uchman W. Analysis of the integrated energy system in residential scale: Photovoltaics, micro-cogeneration and electrical energy storage. Energy 2021;227:120469. http://dx.doi.org/10.1016/j.energy.2021.120469. [23] Wang K, Pantaleo AM, Oyewunmi OA, Markides CN. Flexible PVT-ORC hybrid solar-biomass cogeneration systems: the case study of the university sports centre in Bari, Italy. In: 5th international seminar on ORC power systems. 2019. p. 1–9. [24] Settino J, Morrone P, Algieri A, Sant T, Micallef C, Farrugia M, et al. Integration of an organic rankine cycle and a photovoltaic unit for micro-scale CHP applications in the residential sector. Energy Procedia 2017;126:597–604. http://dx.doi.org/10.1016/j.egypro.2017. 08.223. [24] Settino J, Morrone P, Algieri A, Sant T, Micallef C, Farrugia M, et al. Integration of an organic rankine cycle and a photovoltaic unit for micro-scale CHP applications in the residential sector. Energy Procedia 2017;126:597–604. http://dx.doi.org/10.1016/j.egypro.2017. 08.223. [25] Buonomano A, Calise F, Cappiello FL, Palombo A, Vicidomini M. Dynamic analysis of the integration of electric vehicles in efficient buildings fed by renewables. Appl Energy 2019;245:31–50. http://dx.doi.org/10.1016/j.apenergy.2019.03.206. [25] Buonomano A, Calise F, Cappiello FL, Palombo A, Vicidomini M. Dynamic analysis of the integration of electric vehicles in efficient buildings fed by renewables. Appl Energy 2019;245:31–50. http://dx.doi.org/10.1016/j.apenergy.2019.03.206. [26] Chow TT. Performance analysis of photovoltaic-thermal collector by explicit dynamic model. Sol Energy 2003;75:143–52. http: //dx.doi.org/10.1016/j.solener.2003.07.001. [26] Chow TT. Performance analysis of photovoltaic-thermal collector by explicit dynamic model. Sol Energy 2003;75:143–52. http: //dx.doi.org/10.1016/j.solener.2003.07.001. References Best practise report on decentralized biomass fired CHP plants and status of biomass fired small-and micro scale CHP technologies. 2019, IEA Bioenergy, 32. [8] Padinger R, Aigenbauer S, Schmidl C. Best practise report on decentralized biomass fired CHP plants and status of biomass fired small-and micro scale CHP technologies. 2019, IEA Bioenergy, 32. [9] Zabalaga PJ, Cardozo E, Campero LAC, Ramos JAA. Performance analysis of a stirling engine hybrid power system. Energies 2020;13(980). http://dx.doi.org/10.3390/en13040980. [9] Zabalaga PJ, Cardozo E, Campero LAC, Ramos JAA. Performance analysis of a stirling engine hybrid power system. Energies 2020;13(980). http://dx.doi.org/10.3390/en13040980. 10] Alanne K, Saari A. Sustainable small-scale CHP technologies for buildings: The basis for multi-perspectiv Sustain Energy Rev 2004;8:401–31. http://dx.doi.org/10.1016/j.rser.2003.12.005. [10] Alanne K, Saari A. Sustainable small-scale CHP technologies for buildings: The basis for multi-perspective decision-making. Renew Sustain Energy Rev 2004;8:401–31. http://dx.doi.org/10.1016/j.rser.2003.12.005. [11] Rank nd. https://www.rank-orc.com/rank-micro-2/ (Accessed 17 May 2021). [ ] p ( y ) [12] Pereira JS, Ribeiro JB, Mendes R, Vaz GC, André JC. ORC based micro-cogeneration systems for residential application - a state of the art review and current challenges. Renew Sustain Energy Rev 2018;92:728–43. http://dx.doi.org/10.1016/j.rser.2018.04.039. [12] Pereira JS, Ribeiro JB, Mendes R, Vaz GC, André JC. ORC based micro-cogeneration systems for residential application - a state of the art review and current challenges. Renew Sustain Energy Rev 2018;92:728–43. http://dx.doi.org/10.1016/j.rser.2018.04.039. 768 S. Kallio and M. Siroux Energy Reports 8 (2022) 762–769 Energy Reports 8 (2022) 762–769 Energy Reports 8 (2022) 762–769 [14] Angrisani G, Roselli C, Sasso M. Distributed microtrigeneration systems. Prog Energy Combust Sci 2012;38:502–21. http://dx.doi.org/ 10.1016/j.pecs.2012.02.001. [15] Ferreira AC, Silva J, Teixeira S, Teixeira JC, Nebra SA. Assessment of the stirling engine performance comparing two renewable energy sources: Solar energy and biomass. Renew Energy 2020;154:581–97. http://dx.doi.org/10.1016/j.renene.2020.03.020. Zhu S, Yu G, Jongmin O, Xu T, Wu Z, Dai W, et al. Modeling and experimental investigation of a free-pisto micro-combined heat and power system. Appl Energy 2018;226:522–33. http://dx.doi.org/10.1016/j.apenergy.20 u G, Jongmin O, Xu T, Wu Z, Dai W, et al. Modeling and experimental investigation of a free-piston stirling e [16] Zhu S, Yu G, Jongmin O, Xu T, Wu Z, Dai W, et al. Modeling and experimental investigation of a free-piston stirling engine-based micro-combined heat and power system. Appl Energy 2018;226:522–33. http://dx.doi.org/10.1016/j.apenergy.2018.05.122.i Zhu S, Yu G, Jongmin O, Xu T, Wu Z, Dai W, et al. Modeling and experimental investigation of a free-pisto micro-combined heat and power system. Appl Energy 2018;226:522–33. References g j [27] Tamayo Vera J, Laukkanen T, Sirén K. Performance evaluation and multi-objective optimization of hybrid photovoltaic-thermal collectors. Sol Energy 2014;102:223–33. http://dx.doi.org/10.1016/j.solener.2014.01.014. [27] Tamayo Vera J, Laukkanen T, Sirén K. Performance evaluation and multi-objective optimization of hybrid photovoltaic-thermal collectors. Sol Energy 2014;102:223–33. http://dx.doi.org/10.1016/j.solener.2014.01.014. . Energy analysis and exergy optimization of photovoltaic-thermal collector. Energies 2020;13(5106). http n13195106. [28] Kallio S, Siroux M. Energy analysis and exergy optimization of photovoltaic-thermal collector. //dx.doi.org/10.3390/en13195106. [28] Kallio S, Siroux M. Energy analysis and exergy optimization of photovoltaic-thermal collector. Energies 2020;13(5106). http: //dx.doi.org/10.3390/en13195106. [29] da Silva RM, Fernandes JLM. Hybrid photovoltaic/thermal (PV/T) solar systems simulation with Simulink/Matlab. Sol Energy g [29] da Silva RM, Fernandes JLM. Hybrid photovoltaic/thermal (PV/T) solar systems simulation with Simulink/Matlab. Sol Energy 2010;84:1985–96. http://dx.doi.org/10.1016/j.solener.2010.10.004. [30] Bhattarai S, Oh JH, Euh SH, Krishna Kafle G, Hyun Kim D. Simulation and model validation of sheet and tube type photovoltaic thermal solar system and conventional solar collecting system in transient states. Sol Energy Mater Sol Cells 2012;103:184–93. http://dx.doi.org/10.1016/j.solmat.2012.04.017. [30] Bhattarai S, Oh JH, Euh SH, Krishna Kafle G, Hyun Kim D. Simulation and model validation of sheet and tube type photovoltaic thermal solar system and conventional solar collecting system in transient states. Sol Energy Mater Sol Cells 2012;103:184–93. http://dx.doi.org/10.1016/j.solmat.2012.04.017. p g j 31] Deb K. Multi-objective optimization using evolutionary algorithms, vol. 2001. Chichester, England: John W [31] Deb K. Multi-objective optimization using evolutionary algorithms, vol. 2001. Chichester, England: John Wiley and Sons; 2001. [32] Vera JTamayo, Laukkanen T, Sirén K. Multi-objective optimization of hybrid photovoltaic-thermal collectors integrated in a DHW heating system. Energy Build 2014;74:78–90. http://dx.doi.org/10.1016/j.enbuild.2014.01.011. [32] Vera JTamayo, Laukkanen T, Sirén K. Multi-objective optimization of hybrid photovoltaic-thermal collectors integrated in a DHW heating system. Energy Build 2014;74:78–90. http://dx.doi.org/10.1016/j.enbuild.2014.01.011. 769
https://openalex.org/W4286559984	https://www.frontiersin.org/articles/10.3389/fcell.2022.906042/pdf	English	null	Artificial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study	Frontiers in cell and developmental biology	2,022	cc-by	7,690	Artiﬁcial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study Xing Wu 1†, Di Xu 2†, Tong Ma 2, Zhao Hui Li 1, Zi Ye 1, Fei Wang 3, Xiang Yang Gao 3, Bin Wang 2, Yu Zhong Chen 2, Zhao Hui Wang 4, Ji Li Chen 5, Yun Tao Hu 6, Zong Yuan Ge 2, Da Jiang Wang 1* and Qiang Zeng 3* 1Senior Department of Ophthalmology, The Third Medical Center of Chinese PLA General Hospital, Beijing, China, 2Beijing Airdoc Technology Co., Ltd., Beijing, China, 3Health Management Institute, The Second Medical Center & National Clinical Research Center for Geriatric Diseases, Chinese PLA General Hospital, Beijing, China, 4IKang Guobin Healthcare Group Co., Ltd., Beijing, China, 5Department of Ophthalmology, Shanghai Shibei Hospital of Jing’an District, Shanghai, China, 6Department of Ophthalmology, Beijing Tisnghua Changgung Hospital, Beijing, China ORIGINAL RESEARCH published: 22 July 2022 doi: 10.3389/fcell.2022.906042 ORIGINAL RESEARCH published: 22 July 2022 doi: 10.3389/fcell.2022.906042 Edited by: Arjun Singh, Edited by: Arjun Singh, Memorial Sloan Kettering Cancer Center, United States Background: Cataract is the leading cause of blindness worldwide. In order to achieve large-scale cataract screening and remarkable performance, several studies have applied artiﬁcial intelligence (AI) to cataract detection based on fundus images. However, the fundus images they used are original from normal optical circumstances, which is less impractical due to the existence of poor-quality fundus images for inappropriate optical conditions in actual scenarios. Furthermore, these poor-quality images are easily mistaken as cataracts because both show fuzzy imaging characteristics, which may decline the performance of cataract detection. Therefore, we aimed to develop and validate an antiinterference AI model for rapid and efﬁcient diagnosis based on fundus images. Reviewed by: Azhar Imran, Air University, Pakistan Daniela Quaglino, University of Modena and Reggio Emilia, Italy Correspondence: Da Jiang Wang Wangdajiang301@163.com Qiang Zeng zq301@126.com Correspondence: Da Jiang Wang Wangdajiang301@163.com Qiang Zeng zq301@126.com †These authors have contributed equally to this work and share the ﬁrst authorship. Materials and Methods: The datasets (including both cataract and noncataract labels) were derived from the Chinese PLA general hospital. The antiinterference AI model consisted of two AI submodules, a quality recognition model for cataract labeling and a convolutional neural networks-based model for cataract classiﬁcation. The quality recognition model was performed to distinguish poor-quality images from normal- quality images and further generate the pseudo labels related to image quality for noncataract. Through this, the original binary-class label (cataract and noncataract) was adjusted to three categories (cataract, noncataract with normal-quality images, and noncataract with poor-quality images), which could be used to guide the model to distinguish cataract from suspected cataract fundus images. In the cataract classiﬁcation stage, the convolutional-neural-network-based model was proposed to classify cataracts based on the label of the previous stage. The performance of the model was internally validated and externally tested in real-world settings, and the evaluation indicators included area under the receiver operating curve (AUC), accuracy (ACC), sensitivity (SEN), and speciﬁcity (SPE). †These authors have contributed equally to this work and share the ﬁrst authorship. Specialty section: This article was submitted to Molecular and Cellular Pathology, a section of the journal Frontiers in Cell and Developmental Biology Received: 28 March 2022 Accepted: 21 June 2022 Published: 22 July 2022 INTRODUCTION categories, including normal, immature, mature, and hyper mature using the hybrid model (Imran et al., 2020; Simanjuntak et al., 2022). In the study, 1239 fundus images were used to train the model, and three independent feature sets (i.e., wavelet-, sketch-, and texture-based features) were extracted from each fundus image. Two learning models were established, and then, the ensemble method combining the double models was used to classify the fundus image. The best performance of the ensemble method for cataract classiﬁcation and four-level grading tasks was 93.2% and 84.5%, respectively. Zhang et al. (2019) proposed a six-level cataract grading method that focused on multifeature fusion based on stacking. They extracted two kinds of features that can distinguish the level of cataracts from 1352 fundus images and created a frame consisting of two supported vector machine classiﬁers and a fully connected neural network to grade cataracts. The average accuracy of the six-level grading model was 92.66%. Previous studies have focused on the use of AI for the identiﬁcation and grading of age-related cataracts based on normal-quality fundus images, and they are less likely to consider the quality of fundus images (Gao et al., 2015; Guo et al., 2015; Dong et al., 2017). Cataracts are the leading cause of blindness worldwide (Flaxman et al., 2017). According to the etiological classiﬁcation, the most common type is age-related cataracts (Tang et al., 2016). In China, the incidence rate of cataracts is as high as 80% in 60–89-year-old people and is almost 90% in the elderly over 90 years old (National Health Commission, 2020). With the acceleration of population aging, the prevalence of cataracts is expected to increase (Song et al., 2018). Early diagnosis and timely surgery can effectively treat cataracts to improve the vision and quality of life of patients (Limwattananon et al., 2018; Wu et al., 2019). However, due to the uneven distribution of medical resources, the shortage of ophthalmologists, and the increase in the number of cataract patients, many cataract patients cannot receive early diagnosis and effective treatment, particularly in the primary medical facilities of low- and middle-income countries (Ramke et al., 2017). At present, slit lamp camera images are widely applied for cataract diagnosis due to their optical feature and legibility (Zhang et al., 2019). However, there are some limitations to a certain degree in rural areas, i.e., the nonportability of slit lamp devices and the shortage of medical device technicians. INTRODUCTION In comparison, fundus photographs have several advantages in their efﬁciency and their handleability. Meanwhile, with the universal application of artiﬁcial intelligence (AI) for disease diagnosis, some work focus on automatic cataract detection using AI technique (Patel et al., 2009; Castaneda et al., 2015; Harini and Bhanumathi, 2016; Long et al., 2017). Therefore, combining fundus images and AI-based methods is regarded as a more feasible scheme for automatic cataract detection in actual applications (Park et al., 2020). g However, for cataract detection, the issue of image quality must be considered in actual scenarios since the existence of poor-quality fundus images caused by inappropriate optical scenes in the real world is likely to be mistaken as cataract images, which may cause performance degradation and false positives of cataract diagnosis to some extent. Figure 1 showed the typical noncataract with poor-quality image, noncataract with normal-quality image, and cataract image, respectively. The ﬁgure shows that noncataract with poor- quality image is easily mistaken as cataract, which will bring some challenges to cataract identiﬁcation. In the study, the criteria to distinguish normal and poor-quality noncataract images mainly depended on the lighting and exposure of fundus images. For normal noncataract images, they were under moderate exposure, in which the junction between the rim of the optic disk and the optic cup, the small blood vessels on the surface of the optic disk, and the normal retinal nerve ﬁber layer were clearly distinguishable as a reference. As for poor-quality noncataract images, there were two main manifestations, including underexposure and light leaking. The ﬁrst type was that fundus images were generally blurred and dark due to underexposure. In addition, the other one showed a yellow edge, a light leakage-like edge, or a water drop-like reﬂective band of the surrounding area of the fundus image. Several studies work on AI-assisted diagnosis models of cataracts based on fundus images. Li et al. (2010) divided the fundus images into normal and abnormal, which was an earlier method to apply machine learning to fundus image classiﬁcation. Xu et al. (2020) proposed a deep learning approach to integrate global and local cataract features to construct a hybrid global–local feature representation model. Ran et al. (2018) extracted the initial feature using deep convolutional neural networks (CNNs) and detected the level of cataracts through random forests. Yang et al. (2016) proposed an ensemble learning-based method to improve the accuracy of cataract diagnosis. Triyadi et al. Citation: Wu X, Xu D, Ma T, Li ZH, Ye Z, Wang F, Gao XY, Wang B, Chen YZ, Wang ZH, Chen JL, Hu YT, Ge ZY, Wang DJ and Zeng Q (2022) Artiﬁcial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study. Front. Cell Dev. Biol. 10:906042. doi: 10.3389/fcell.2022.906042 Wu X, Xu D, Ma T, Li ZH, Ye Z, Wang F, Gao XY, Wang B, Chen YZ, Wang ZH, Chen JL, Hu YT, Ge ZY, Wang DJ and Zeng Q (2022) Artiﬁcial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study. Chen JL, Hu YT, Ge ZY, Wang DJ and Zeng Q (2022) Artiﬁcial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study. Results: In the internal and external validation, the antiinterference AI model showed robust performance in cataract diagnosis (three classiﬁcations with AUCs >91%, Front. Cell Dev. Biol. 10:906042. doi: 10.3389/fcell.2022.906042 July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org Wu et al. AI Model to Diagnose Cataract ACCs >84%, SENs >71%, and SPEs >89%). Compared with the model that was trained on the binary-class label, the antiinterference cataract model improved its performance by 10%. ACCs >84%, SENs >71%, and SPEs >89%). Compared with the model that was trained on the binary-class label, the antiinterference cataract model improved its performance by 10%. Conclusion: We proposed an efﬁcient antiinterference AI model for cataract diagnosis, which could achieve accurate cataract screening even with the interference of poor-quality images and help the government formulate a more accurate aid policy. Keywords: cataract, artiﬁcial intelligence, auxiliary diagnosis, fundus image, convolution neural network Keywords: cataract, artiﬁcial intelligence, auxiliary diagnosis, fundus image, convolution neural network MATERIALS AND METHODS Therefore, in our study, to alleviate the issue of diagnostic performance degradation due to the interference of poor-quality fundus images, we proposed a hybrid structure based on a two- stage AI model to achieve accurate cataract image recognition with the distraction of image quality. The results show that our proposed method has increased the robustness of the model and achieved accurate cataract detection even with numerous interferences. Furthermore, it can assist doctors in cataract diagnosis more efﬁciently. Dataset Collection and Labeling for Artiﬁcial Intelligence Model INTRODUCTION (2022) processed the two-class cataract classiﬁcation using VGG-19, Resnet-50, and Resnet-100, whereas several studies improved the cataract classiﬁcation into four July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 2 Wu et al. AI Model to Diagnose Cataract E 1 \| Overall training pipeline for the cataract artiﬁcial intelligence model. (A) the dataset included 33,965 images of 30,668 participants. Each image was dently labeled by two experienced ophthalmologists, and a third ophthalmologist was consulted if a disagreement arose between the initial ophthalmologists. (B) 5 images with binary-class diagnosis labels were adjusted and reassigned to three categories of labels by the quality recognition model. (C) all 33,965 images ut to the convolutional neural networks-based model for training and validating the antiinterference cataract artiﬁcial intelligence classiﬁcation model. E 1 \| Overall training pipeline for the cataract artiﬁcial intelligence model. (A) the dataset included 33,965 images of 30,668 participants. Each image was ndently labeled by two experienced ophthalmologists, and a third ophthalmologist was consulted if a disagreement arose between the initial ophthalmologists. (B 65 images with binary-class diagnosis labels were adjusted and reassigned to three categories of labels by the quality recognition model. (C) all 33,965 image put to the convolutional neural networks-based model for training and validating the antiinterference cataract artiﬁcial intelligence classiﬁcation model. \| Overall training pipeline for the cataract artiﬁcial intelligence model. (A) the dataset included 33,965 images of 30,668 participants. Each image was ntly labeled by two experienced ophthalmologists, and a third ophthalmologist was consulted if a disagreement arose between the initial ophthalmologists FIGURE 1 \| Overall training pipeline for the cataract artiﬁcial intelligence model. (A) the dataset included 33,965 images of 30,668 participants. Each image was independently labeled by two experienced ophthalmologists, and a third ophthalmologist was consulted if a disagreement arose between the initial ophthalmologists. (B) all 33,965 images with binary-class diagnosis labels were adjusted and reassigned to three categories of labels by the quality recognition model. (C) all 33,965 images were input to the convolutional neural networks-based model for training and validating the antiinterference cataract artiﬁcial intelligence classiﬁcation model. Introduction of the Artiﬁcial Intelligence Model Training and Validation examinations such as slit lamp images, and fundus images were anonymized and acquired from the hospital information system. The fundus images were excluded from the study if the participants had congenital cataract, intraocular lens, aphakic eye, severe eye trauma, or corneal opacity. The 14,820 participants with fundus images and medical information were simply randomized into the development dataset (88%) and internal validation dataset (12%). The 16,200 fundus images were split randomly into mutually exclusive sets for training dataset (development dataset) and internal testing dataset (internal validation dataset) of the AI model at an 8:1 ratio. To validate the availability of the antiinterference AI model in a real-world scenario, 17,765 fundus images of 15,848 participants were prospectively collected (from June 2021 to December 2021) from three real- world settings (i.e., iKang Guobin Healthcare Group Co., Ltd., Shanghai Shibei Hospital of Jing’an District and Beijing Tsinghua Changgung Hospital) as external test dataset (Shown in Figure 2). In our study, three different nonmydriatic fundus cameras (Canon, Syseye, and Topcon) were used. All fundus images were macula and optic disk-centered 45-color fundus photographs. g For our model training stage, the model consisted of two AI submodules, a quality recognition model for cataract labeling and a CNN-based model for cataract classiﬁcation, which were both trained on the development dataset. The AI model used in this study was based on deep neural network architecture. In order to obtain an effective model for real clinical usage, the whole model learning process needed to involve two steps, model training and model validation. In the model training stage, the goal was to train an AI model to ﬁt the training data points and be competent for the speciﬁc problem. One common approach was to let the model learn with labeled data sets that were annotated by professionals. The labels were used as supervisory signals to guide models to have better capabilities to recognize cataracts. The validation stage aimed to predict the results of the input images and validate the performance of the model. During the validation stage, labels were not available during the prediction and were used to measure the performance after the model output its predictions. In our experiment, the model was tested in the internal and external validation dataset, and the results output by the AI model were then compared with the ground truth to evaluate the performance of the model. Dataset Collection and Labeling for Artiﬁcial Intelligence Model In the study, the dataset which included 14,820 participants and 16,200 fundus images of cataract and noncataract was retrospectively derived from the Chinese PLA general hospital from September 2018 to May 2021. The participants’ basic information (age, sex), brief medical history with related July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 3 AI Model to Diagnose Cataract Wu et al. FIGURE 2 \| Flow chart describing the datasets and methods used for our artiﬁcial intelligence model. FIGURE 2 \| Flow chart describing the datasets and methods used for our artiﬁcial intelligence model. Frontiers in Cell and Developmental Biology \| www.frontiersin.org Introduction of the Artiﬁcial Intelligence Model Training and Validation Each image was independently evaluated and labeled by two experienced ophthalmologists, and a third ophthalmologist was consulted in case of disagreement (shown in Figure 1A). The medical information corresponding to each recruited fundus image was provided to the ophthalmologists to improve diagnostic accuracy. The fundus images were labeled into two categories, namely, cataract and noncataract. Quality Recognition Stage y g g In the practical scene, the performance of the cataract AI classiﬁcation model could be largely affected by the image quality, especially when the cataract images are easily confused July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 4 Wu et al. AI Model to Diagnose Cataract with poor-quality images of noncataract; thus, the correct distinction between the two is crucial for the performance of the following cataract AI recognition model. Therefore, in the ﬁrst stage, we built a quality recognition model that aimed to assess the fundus images according to the image quality and generated the three-category cataract labeling for the next step. FIGURE 3 \| Accuracy and loss curve of antiinterference cataract artiﬁcial intelligence diagnosis model in the training process. g g y g Based on this situation, we proposed a label-based method to better distinguish cataract from noncataract with suspected fundus images using the quality recognition model. The quality recognition model was trained with a seven-category task, including normal, slight darkness, slight brightness, slight blur, severe darkness, severe brightness, and severe blur. This quality recognition model is the quality control tool for fundus images developed by Hu et al. (2021) in Airdoc company in 2019, which can be widely used in many retinal disease recognition tasks on fundus images. The model consists of two steps, the ﬁrst one uses a generative adversarial network to determine whether the images are fundus images, and the second step is applied to output the probabilities of each quality grade (seven in total) of images. By combining the fundus image recognition model with the image quality classiﬁcation network model, the overall accuracy can be improved by ﬁltering out the interference from nonfundus images while obtaining a more accurate fundus image quality classiﬁcation. The area under the curve (AUC) of the quality model was 99%, which was suitable for quality recognition of noncataract images; thus, we have applied this method to our work. Moreover, we employed the model to generate pseudo labels for noncataract, distinguishing between normal quality and poor quality. Quality Recognition Stage Therefore, by labeling the categories (slight darkness, slight brightness, slight blur, severe darkness, severe brightness, and severe blur) as poor-quality noncataract class, the original annotations were further reﬁned into three categories (cataract, noncataract with normal-quality images, and noncataract with poor-quality images), which would give guidance to our cataract AI diagnosis model on the aspect of the label, so that the AI model could improve the performance of cataract classiﬁcation (shown in Figure 1B). Th li i i d l i h d i l FIGURE 3 \| Accuracy and loss curve of antiinterference cataract artiﬁcial intelligence diagnosis model in the training process. Cataract Classiﬁcation Stage g Based on the quality recognition model, an antiinterference AI classiﬁcation model with a CNN was trained to predict the label of images (cataract, noncataract with normal-quality images, and noncataract with poor-quality images). p q y g CNN is a kind of feedforward neural network with depth structure and convolution calculation (Sun et al., 2019). It is one of the representative algorithms of deep learning. Because of its depth and massive layers, CNN has huge representation power to learn visual features of ophthalmic diseases and discriminate them effectively. The structure of the cataract AI classiﬁcation model was mainly based on Inception-Resnet pretrained on ImageNet (Szegedy et al., 2017). As shown in Figure 1C, a CNN consisted of several convolution layers, a pooling layer, and a fully connected layer. The convolution and pooling layers cooperated to form multiple convolution groups, extract features layer by layer, and ﬁnally complete the classiﬁcation through the fully connected layer. More speciﬁcally, random rotation, a data argumentation method, was applied in the data preprocessing stage, and then, each image was resized to 300px × 300px as input into the model. After CNN analysis, the model output three values in the range 0–1, each representing the probability of the corresponding category for each image. At last, the category corresponding to the largest value was selected using the model, which was the predicted classiﬁcation result of the image. The quality recognition model in our method was mainly based on Inception-Resnet pretrained on ImageNet. During training, the model was optimized with the cross-entropy loss, which was deﬁned as follows. L −1 N N n1 C i1y(n) i logp(n) i (1) (1) N was the batch size, y(n) i represented the label in of sample n (if the sample n belonged to class i, the value of y(n) i was equal to 1; otherwise, y(n) i 0), C was the class number, and p(n) i meant the predicted probability in class i of sample n. In the quality recognition stage, for hyperparameters conﬁguration, the model of seven-class classiﬁcation was trained for 200 epochs with a batch size of 24, a dropout of 0.5, and an initial learning rate of 0.0001. In addition, we used stochastic gradient descent as an optimizer, SoftMax as the last activation function, and Pytorch ReduceLRonPlateau with factor 0.2 and patience 6 as a scheduler. Heatmap Visualization Figure 6 provided the visual feature map both in our proposed method and control experiment. We analyzed the heatmaps in two aspects. First, by comparing our approach with the control experiment, it could be seen that for the cataract and noncataract with normal-quality images, the attention regions of both methods were similar, with cataract focusing on the blurred areas, optic disk, and great vessels and noncataract with normal-quality image concerning the small and medium vessels (Figure 6A). However, for the noncataract with poor- quality images, our proposed method focused more on the important part, which showed a better ability to distinguish Cataract Classiﬁcation Stage The model determined the diagnosis of cataracts with an AUC of 82.22% and an ACC of 64.33%. From the confusion matrix, shown in Figure 5A, we could see that there were many noncataract images were mistakenly classiﬁed as cataract due to the interference of poor-quality fundus images; thus, the performance of cataract recognition degraded to a certain degree. To verify the robustness of our proposed method, we set up a control experiment, in which the cataract diagnosis model was trained on the binary-class labels, including normal and cataract two categories. In addition, the setting of the control experiment was consistent with the proposed method. Then, we tested the performance of the cataract binary-class AI model in the control experiment and the antiinterference cataract AI diagnosis model concerning standard diagnosis based on ophthalmologists’ evaluation in the same dataset (internal validation set and external test dataset, which contained a certain amount of poor-quality fundus images). Furthermore, we tested the performance of the antiinterference model in the external test dataset. For the results of the binary-class model in the external test dataset, the receiver operating characteristic curve and confusion matrix diagram were exhibited in Table 3; Figures 4B, 5B. The AUC and ACC in the external test dataset were 81.33% and 65.34%. In addition, similar to the performance in the internal validation dataset, the performance of the model was also affected to some extent by poor-quality images. RESULTS Basic Information of Recruited Cases A total of 30,668 participants with 33,965 fundus images were recruited for this study (Table 1). Among them, 15,804 (51.53%) are male and 14,864 (48.47%) are female. In the development, internal validation, and external test dataset, the participants in different sex groups were relatively evenly distributed, respectively. Statistical Analysis The indices used for evaluation were calculated using the accuracy of the formula (ACC) = (TP + TN)/(TP + TN + FP + FN), sensitivity (SEN) = TP/(TP + FN), and speciﬁcity (SPE) = TN/(TN + FP), where TP is true positive, TN is true negative, FP is false positive, and FN is false negative. Asymptotic two-sided 95% CIs presented as the AUC and were calculated by using bootstrap analysis with 100,000 random seed sampling. Receiver operating characteristic curves were created using the R statistical package, V.3.2.4. To visualize the decision ways of the model, we applied the Grad-CAM to generate heatmaps. Receiver operating characteristic curves and confusion matrix diagrams of the antiinterference cataract AI diagnosis model was shown in Figures 4C,D; Figures 5C,D. The model determined diagnosis of cataract, noncataract with normal-quality images, or noncataract with poor-quality images with AUC of 91.84%, 96.76%, and 96.83% in the internal validation dataset, respectively (Table 2). Compared with the model that was trained on the binary-class label, the antiinterference cataract model improved its performance by 10%. The model determined diagnosis of cataract, noncataract with normal-quality images, or noncataract with poor-quality images with AUCs of 91.62%, 96.12%, and 97.00% in the external test dataset, respectively (Table 3). Cataract Classiﬁcation Stage N was the batch size, y(n) i represented the label in of sample n (if the sample n belonged to class i, the value of y(n) i was equal to 1; otherwise, y(n) i 0), C was the class number, and p(n) i meant the predicted probability in class i of sample n. In the quality recognition stage, for hyperparameters conﬁguration, the model of seven-class classiﬁcation was trained for 200 epochs with a batch size of 24, a dropout of 0.5, and an initial learning rate of 0.0001. In addition, we used stochastic gradient descent as an optimizer, SoftMax as the last activation function, and Pytorch ReduceLRonPlateau with factor 0.2 and patience 6 as a scheduler. During the training stage, we used the fundus images from the development dataset as input to train the models. After one hundred training epochs, a cataract classiﬁcation model can be obtained, and the accuracy and loss curve during the training process were shown in Figure 3. As for implementation details, we trained the network on our dataset for 100 epochs with a batch size of 24. Stochastic gradient descent with a momentum of 0.9 and a weight decay of 10−4 was used as the optimizer. In addition, the initial learning rate was set as 0.0001, and SoftMax was chosen as the last activation function. In the internal validation and external test stage, fundus images were mixed July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 5 AI Model to Diagnose Cataract Wu et al. TABLE 1 \| Characteristics of the development, internal validation, and external test dataset. Characteristics Development dataset Internal validation dataset External test dataset Male Female Male Female Male Female No. of participants 6,745 6,295 947 833 8,112 7,736 Age 53.00 ± 15.09 52.81 ± 14.80 52.16 ± 14.96 53.89 ± 14.87 52.92 ± 15.19 53.03 ± 15.02 No. of images 7,498 6,902 960 840 9,201 8,564 Cataract 2,298 2,502 273 327 2,831 3,166 Noncataract with normal-quality images 2,469 2,331 326 274 3,008 2,992 Noncataract with poor-quality images 2,731 2,069 361 239 3,362 2,406 TABLE 1 \| Characteristics of the development, internal validation, and external test dataset. and used as input, and the cataract AI classiﬁcation model could predict and output classiﬁcation labels directly. shown in Table 2; Figure 4A, Figure5A. Evaluation of Cataract Artiﬁcial Intelligence Diagnosis Model Based on the Original Binary-Class Labels (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) Antiinterference model in the external test dataset. FIGURE 5 \| Confusion matrix of the two cataract artiﬁcial intelligence diagnosis models. (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) Antiinterference model in the external test dataset. FIGURE 5 \| Confusion matrix of the two cataract artiﬁcial intelligence diagnosis models. (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) Antiinterference model in the external test dataset. TABLE 3 \| erformance of the two cataract artiﬁcial intelligence diagnosis models in the external test dataset. Classiﬁcation AUC (%) ACC (%) SEN (%) SPE (%) Binary-class model Cataract 81.33 65.34 94.03 50.72 Anti-interference model Cataract 91.62 84.37 71.20 90.84 Noncataract with normal-quality images 96.12 89.82 85.37 91.11 Noncataract with poor-quality images 97.00 90.97 91.50 89.39 AUC, area under the curve; ACC, accuracy; SEN, sensitivity; SPE, speciﬁcity. identify the details such as blood vessels in the fundus images, which makes it feasible for automatic diagnosis through the fundus images. Several studies have been reported using an AI model based on fundus images for cataract diagnosis. In the fundus images taken from subjects with mild cataracts, small retinal vessels are visible but slightly blurred. With the aggravation of cataracts, more structures will be invisible until nothing can be seen. on the blur degree of whole fundus images and the visibility of the vessels and optic disk to determine the cataract (Figure 6C). Therefore, through the visual analysis, it showed that our method had a better ability to diagnose cataracts than the control experiment. Evaluation of Cataract Artiﬁcial Intelligence Diagnosis Model Based on the Original Binary-Class Labels The diagnostic performance of the cataract AI diagnosis model in the control experiment on the internal validation dataset was July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 6 AI Model to Diagnose Cataract Wu et al. TABLE 2 \| erformance of the two cataract artiﬁcial intelligence diagnosis models in the internal validation dataset. Classiﬁcation AUC (%) ACC (%) SEN (%) SPE (%) Binary-class model Cataract 82.22 64.33 93.67 49.67 Anti-interference model Cataract 91.84 85.06 73.17 90.75 Noncataract with normal-quality images 96.76 90.44 85.67 91.97 Noncataract with poor-quality images 96.83 91.06 91.00 89.91 AUC, area under the curve; ACC, accuracy; SEN, sensitivity; SPE, speciﬁcity. 2 \| erformance of the two cataract artiﬁcial intelligence diagnosis models in the internal validation dataset. FIGURE 4 \| Receiver operating characteristic curves of the two cataract artiﬁcial intelligence diagnosis models. (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) antiinterference model in the external test dataset. FIGURE 4 \| Receiver operating characteristic curves of the two cataract artiﬁcial intelligence diagnosis models. (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) antiinterference model in the external test dataset. the device. However, our method paid more attention to the light change and darker area rather than the blur and suspected cataract part. Second, comparing the attention area that our model focused on and the cataract criterion of professional doctors, they had the consistent regulations that were based cataracts with the suspected cataract images. For instance, in Figure 6B, a noncataract with poor-quality images was misclassiﬁed by the control experiment as the cataract class because it expressed a similar yellow margin of the peripheral area with cataract images, which was due to the light leakage from July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 7 AI Model to Diagnose Cataract Wu et al. FIGURE 5 \| Confusion matrix of the two cataract artiﬁcial intelligence diagnosis models. (A) binary-class model in the internal validation dataset. (B) binary-class model in the external test dataset. (C) antiinterference model in the internal validation dataset. (D) Antiinterference model in the external test dataset. FIGURE 5 \| Confusion matrix of the two cataract artiﬁcial intelligence diagnosis models. DISCUSSION In the previous study, the cataract AI model developed from a single learning model to multiple learning models and improved the prediction accuracy substantially. Yang et al. (2013) built a neural network classiﬁer that consists of three parts: Cataract is the most common cause of blindness worldwide with the characteristic of lenticular opacity. With the rapid development of AI in medicine, the AI model can effectively July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 8 Wu et al. AI Model to Diagnose Cataract FIGURE 6 \| Heatmap visualization examples. Left column: the original fundus images; middle column: general heatmap of antiinterference method; right column: general heatmap of control experiment. (A) noncataract with normal-quality image. (B) noncataract with poor-quality image. (C) cataract. FIGURE 6 \| Heatmap visualization examples. Left column: the original fundus images; middle column: general heatmap of antiinterference method; right column: general heatmap of control experiment. (A) noncataract with normal-quality image. (B) noncataract with poor-quality image. (C) cataract. FIGURE 6 \| Heatmap visualization examples. Left column: the original fundus images; middle column: general heatmap of antiinterference method; right column: general heatmap of control experiment. (A) noncataract with normal-quality image. (B) noncataract with poor-quality image. (C) cataract. preprocessing, feature extraction, and classiﬁer construction. Imran et al. (2021) proposed a novel hybrid convolutional and recurrent neural network for cataract classiﬁcation and increased an average accuracy of 97.39% for four-class cataract classiﬁcation. Yadav and Yadav (2022) studied computer-aided cataract detection and grading by extracting and fusing features and integrating the predictions through machine learning methods, which achieved 96.25% four-class classiﬁcation accuracy. The six-level classiﬁcation of cataracts could enable ophthalmologists to accurately understand the patient’s condition, and the average accuracy of the six-grading model was up to 92.66%. However, all above this, the fundus images with good quality are the key points (Zhang et al., 2019). Here, we reported that the accuracy of the antiinterference model was approximately 83%, which is lower than that in the previous study. This may be due to our inclusion of fundus images with poor quality in the study, resulting in complicated interference. Studies have shown that the fuzziness and scanning quality index of fundus images are related to cataract AI identiﬁcation and grading (Xu et al., 2010; Chen et al., 2021). Poor-quality images can greatly affect the accuracy of the results (Zhang et al., 2019). Frontiers in Cell and Developmental Biology \| www.frontiersin.org DISCUSSION The common limitation of previous studies is that there are high requirements for the quality of fundus images. However, in the actual cataract screening scenario, it is difﬁcult to ensure that all of the fundus images meet the quality requirements because of the uneven technical level of operators and inadequate cooperation of patients. The interference of poor- quality images caused by shooting encountered in the research has not been solved. In the study, we developed and validated an antiinterference cataract diagnosis model that can identify quality problems of fundus images to be better applied in the real world. The model that can achieve accurate cataract recognition mainly included a quality recognition model for adjusting cataract labels and a CNN-based model for cataract classiﬁcation. The quality recognition model aimed to distinguish between cataract images and noncataract with poor-quality images that are easy to be misclassiﬁed by converting the original binary-class label into a triplet. Then, we trained cataract diagnosis models based on dichotomy and trichotomy respectively and compared their performance. According to the results of this study, we found that the antiinterference cataract model improved the July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 9 Wu et al. AI Model to Diagnose Cataract ETHICS STATEMENT The Ethical Committee of Chinese PLA General Hospital waived the need for informed consent from the participants. The study was conducted by the Declaration of Helsinki and approved by the ethics committee of the Chinese PLA General Hospital. DATA AVAILABILITY STATEMENT performance by 10% compared with the model that was trained on the binary-class label; hence, the quality recognition model can enhance the robustness of the cataract AI diagnosis model. The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation. In the primary medical facilities of low- and middle-income countries, a cataract cannot be diagnosed until it develops to an advanced stage and even the symptoms can be observed with the naked eye. Our cataract AI diagnosis model is helpful for the early detection of cataracts. If participants are diagnosed with cataracts using the AI model, they need to go to the hospital and follow the doctor’s advice for further examination, such as slit lamp and ophthalmic B-type ultrasound. If the output of the AI model is noncataract with poor-quality images, it would be best to retake new fundus images with normal quality or after mydriasis as early as possible. Assuming that the output of the AI model is noncataract with normal-quality images, the participants need to be retested in 12 months. AUTHOR CONTRIBUTIONS Considering that the dataset used for internal validation has similar characteristics to the dataset used for model training, it may lead to better accuracy and stability of the research results than the real situation. Therefore, further external validation was conducted, and our proposed model showed good performance. In the future, a larger sample size database will help to optimize the antiinterference cataract model (Chen et al., 2021). XW, YC, DW, and QZ contributed to the conception of the study. XW, DX, BW, ZG, and QZ contributed to study design and model training. XW, ZY, FW, XG, ZW, JC, YH, and ZL contributed to the acquisition of the data. XW, DX, TM, and ZL contributed to the analysis and interpretation of the data. XW, DX, and TM wrote the draft. DW and QZ revised the draft. All the authors read and approved the ﬁnal manuscript. In conclusion, we developed an AI model for antiinterference cataract automatic diagnosis based on fundus images. The antiinterference model achieved cataract diagnosis with high accuracy even in the presence of poor-quality image interference, allowing the model to provide cataract screening and help the government formulate a more accurate aid policy. In recent 10 years, the application of AI for cataract identiﬁcation and grading based on fundus images has developed rapidly. To better achieve cataract classiﬁcation, further research needs to be done to develop a cataract grading AI model based on our proposed method. FUNDING The research was supported by the Natural Science Foundation of Beijing (7212092), Capital’s Funds for Health Improvement and Research (2022-2-5041), the Bigdata Project Foundation of PLA general hospital (2019MBD-037) Military Healthcare Program (19BJZ24), and the National Natural Science Foundation of China (81872920). The research was supported by the Natural Science Foundation of Beijing (7212092), Capital’s Funds for Health Improvement and Research (2022-2-5041), the Bigdata Project Foundation of PLA general hospital (2019MBD-037) Military Healthcare Program (19BJZ24), and the National Natural Science Foundation of China (81872920). Image Analysis. Comput. Industry 69, 72–80. doi:10.1016/j.compind.2014. 09.005 REFERENCES Harini, V., and Bhanumathi, V. (2016). “Automatic Cataract Classiﬁcation System,” in Proceedings of the 2016 International Conference on Communication and Signal Processing (ICCSP), Melmaruvathur, India, 06- 08 April 2016, 0815–0819. doi:10.1109/ICCSP.2016.7754258 Castaneda, C., Nalley, K., Mannion, C., Bhattacharyya, P., Blake, P., Pecora, A., et al. (2015). Clinical Decision Support Systems for Improving Diagnostic Accuracy and Achieving Precision Medicine. J. Clin. Bioinform 5, 4. doi:10. 1186/s13336-015-0019-3 Hu, X., Zhao, X., He, C., and Zhang, D. (2021). Fundus Image Quality Control Method, Device, Electronic Equipment and Storage Medium. China patent application CN202110616409.6. Chen, S., Wang, k., and Wan, X. (2021). Artiﬁcial Intelligence in Diagnosis and Classiﬁcation of Cataract. Int. Rev. Ophthalmol. 45, 231–236. doi:10.3760/cma.j. issn.1673-5803.2021.03.011 Huiqi Li, H., Joo Hwee Lim, J. H., Jiang Liu, J., Mitchell, P., Tan, A. G., Jie Jin Wang, J. J., et al. (2010). A Computer-Aided Diagnosis System of Nuclear Cataract. IEEE Trans. Biomed. Eng. 57, 1690–1698. doi:10.1109/TBME.2010.2041454 Dong, Y., Zhang, Q., Qiao, Z., and Yang, J.-J. (2017). “Classiﬁcation of Cataract Fundus Image Based on Deep Learning,” in Proceeding of the 2017 IEEE International Conference on Imaging Systems and Techniques (IST), Beijing, China, 18-20 October 2017, 1–5. doi:10.1109/IST.2017. 8261463 Imran, A., Li, J., Pei, Y., Akhtar, F., Mahmood, T., and Zhang, L. (2021). Fundus Image-Based Cataract Classiﬁcation Using a Hybrid Convolutional and Recurrent Neural Network. Vis. Comput. 37 (8), 2407–2417. doi:10.1007/ s00371-020-01994-3 Flaxman, S. R., Bourne, R. R. A., Resnikoff, S., Ackland, P., Braithwaite, T., Cicinelli, M. V., et al. (2017). Global Causes of Blindness and Distance Vision Impairment 1990-2020: a Systematic Review and Meta-Analysis. Lancet Glob. Health 5, e1221–e1234. doi:10.1016/s2214-109x(17)30393-5 Imran, A., Li, J., Pei, Y., Akhtar, F., Yang, J.-J., and Dang, Y. (2020). Automated Identiﬁcation of Cataract Severity Using Retinal Fundus Images. Comput. Methods Biomechanics Biomed. Eng. Imaging & Vis. 8 (6), 691–698. doi:10. 1080/21681163.2020.1806733 Gao, X., Lin, S., and Wong, T. Y. (2015). Automatic Feature Learning to Grade Nuclear Cataracts Based on Deep Learning. IEEE Trans. Biomed. Eng. 62, 2693–2701. doi:10.1109/TBME.2015.2444389 Limwattananon, C., Limwattananon, S., Tungthong, J., and Sirikomon, K. (2018). Association between a Centrally Reimbursed Fee Schedule Policy and Access to Cataract Surgery in the Universal Coverage Scheme in Thailand. JAMA Ophthalmol. 136, 796–802. doi:10.1001/jamaophthalmol.2018.1843 Guo, L., Yang, J.-J., Peng, L., Li, J., and Liang, Q. (2015). REFERENCES A Computer-Aided Healthcare System for Cataract Classiﬁcation and Grading Based on Fundus July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 10 Wu et al. AI Model to Diagnose Cataract Long, E., Lin, H., Liu, Z., Wu, X., Wang, L., Jiang, J., et al. (2017). An Artiﬁcial Intelligence Platform for the Multihospital Collaborative Management of Congenital Cataracts. Nat. Biomed. Eng. 1, 0024. doi:10.1038/s41551-016-0024 Triyadi, A. B., Bustamam, A., and Anki, P. (2022). “Deep Learning in Image Classiﬁcation Using VGG-19 and Residual Networks for Cataract Detection,” in Proceeding of the 2022 2nd International Conference on Information Technology and Education (ICIT&E), Malang, Indonesia, 22-22 January 2022, 293–297. doi:10.1109/ICITE54466.2022.9759886 Meimei Yang, M. M., Yang, J.-J., Qinyan Zhang, Q. Y., Yu Niu, Y., and Jianqiang Li, J. Q. (2013). “Classiﬁcation of Retinal Image for Automatic Cataract Detection,” in Proceeding of the 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services (Healthcom 2013), Linsbon, 09- 12 October 2013, 674–679. doi:10.1109/HealthCom.2013.6720761 Wu, X., Huang, Y., Liu, Z., Lai, W., Long, E., Zhang, K., et al. (2019). Universal Artiﬁcial Intelligence Platform for Collaborative Management of Cataracts. Br. J. Ophthalmol. 103, 1553–1560. doi:10.1136/bjophthalmol-2019-314729 National Health Commission. (2020). The White Paper on Eye Health in china. Available at: https://www.nhc.gov.cn/xcs/s3574/202006/ 1f519d91873948d88a77a35a427c3944.shtml [Accesed Jan 13, 2022]. Xu, L., Yang, C., Yang, H., Wang, S., Shi, Y., and Song, X. (2010). The Study of Predicting the Visual Acuity after Phacoemulsiﬁcation According to the Blur Level of Fundus Photography. Ophthalmol. CHN 19, 81–83. Park, S. J., Lee, E. J., Kim, S. I., Kong, S.-H., Jeong, C. W., and Kim, H. S. (2020). Clinical Desire for an Artiﬁcial Intelligence-Based Surgical Assistant System: Electronic Survey-Based Study. JMIR Med. Inf. 8, e17647. doi:10.2196/17647 Xu, X., Zhang, L., Li, J., Guan, Y., and Zhang, L. (2020). A Hybrid Global-Local Representation CNN Model for Automatic Cataract Grading. IEEE J. Biomed. Health Inf. 24, 556–567. doi:10.1109/JBHI.2019.2914690 Yadav, J. K. P. S., and Yadav, S. (2022). Computer-aided Diagnosis of Cataract Severity Using Retinal Fundus Images and Deep Learning. Comput. Intell. doi:10.1111/coin.12518 Patel, V. L., Shortliffe, E. H., Stefanelli, M., Szolovits, P., Berthold, M. R., Bellazzi, R., et al. (2009). The Coming of Age of Artiﬁcial Intelligence in Medicine. Artif. Intell. Med. 46, 5–17. doi:10.1016/j.artmed.2008.07.017 Ramke, J., Zwi, A. B., Lee, A. C., Blignault, I., and Gilbert, C. E. (2017). REFERENCES Inequality in Cataract Blindness and Services: Moving beyond Unidimensional Analyses of Social Position. Br. J. Ophthalmol. 101, 395–400. doi:10.1136/bjophthalmol- 2016-309691 Yang, J.-J., Li, J., Shen, R., Zeng, Y., He, J., Bi, J., et al. (2016). Exploiting Ensemble Learning for Automatic Cataract Detection and Grading. Comput. Methods Programs Biomed. 124, 45–57. doi:10.1016/j.cmpb.2015.10.007 Zhang, H., Niu, K., Xiong, Y., Yang, W., He, Z., and Song, H. (2019). Automatic Cataract Grading Methods Based on Deep Learning. Comput. Methods Programs Biomed. 182, 104978. doi:10.1016/j.cmpb.2019.07.006 Ran, J., Niu, K., He, Z., Zhang, H., and Song, H. (2018). “Cataract Detection and Grading Based on Combination of Deep Convolutional Neural Network and Random Forests,” in Proceeding of the 2018 International Conference on Network Infrastructure and Digital Content (IC-NIDC), Guiyang, China, 22- 24 August 2018, 155–159. doi:10.1109/ICNIDC.2018.8525852 Conﬂict of Interest: Authors DX, TM, BW, YC, and ZG were employed by the company Beijing Airdoc Technology Co., Ltd., China. Author ZW was employed by the company iKang Guobin Healthcare Group Co., Ltd., China. Simanjuntak, R. B. J., Fu’adah, Y., Magdalena, R., Saidah, S., Wiratama, A. B., and Ubaidah, I. D. W. S. (2022). Cataract Classiﬁcation Based on Fundus Images Using Convolutional Neural Network. JOIV Int. J. Inf. Vis. 6 (1), 33–38. doi:10.30630/joiv.6.1.856 The remaining authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as potential conﬂicts of interest. Song, P., Wang, H., Theodoratou, E., Chan, K. Y., and Rudan, I. (2018). The National and Subnational Prevalence of Cataract and Cataract Blindness in China: a Systematic Review and Meta-Analysis. J. Glob. Health 8, 010804. doi:10.7189/jogh.08-01080410.7189/jogh.08.010804 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations or those of the publisher, the editors, and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Sun, X., Li, Y., Kang, H., and Shen, Y. (2019). Automatic Document Classiﬁcation Using Convolutional Neural Network. J. Phys. Conf. Ser. 1176, 032029. doi:10. 1088/1742-6596/1176/3/032029 Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A. A. (2017). “Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning,” in Proceeding of the Thirty-First AAAI Conference on Artiﬁcial Intelligence, San Francisco, USA, February 04-09, 2017, 4278–4284. Frontiers in Cell and Developmental Biology \| www.frontiersin.org July 2022 \| Volume 10 \| Article 906042 REFERENCES Copyright © 2022 Wu, Xu, Ma, Li, Ye, Wang, Gao, Wang, Chen, Wang, Chen, Hu, Ge, Wang and Zeng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Tang, Y., Wang, X., Wang, J., Huang, W., Gao, Y., Luo, Y., et al. (2016). Prevalence of Age-Related Cataract and Cataract Surgery in a Chinese Adult Population: The Taizhou Eye Study. Invest. Ophthalmol. Vis. Sci. 57, 1193–1200. doi:10. 1167/iovs.15-18380 July 2022 \| Volume 10 \| Article 906042 Frontiers in Cell and Developmental Biology \| www.frontiersin.org 11
https://openalex.org/W4324124276	https://journal.iainkudus.ac.id/index.php/QIJIS/article/download/14661/pdf	English	null	Religious Psychosocial Healing for the Trauma of the Children of the Terrorists	Qudus International Journal Of Islamic Studies/Qudus International of Journal Islamic Studies	2,022	cc-by	9,665	Religious Psychosocial Healing for the Trauma of the Children of the Terrorists Miftahuddin IAIN Salatiga, Central Java, Indonesia miftahuddin@iainsalatiga.ac.id Hanung Triyoko IAIN Salatiga, Central Java, Indonesia hanungtriyoko@iainsalatiga.ac.id Burhan Yusuf Habibi Suez Canal University, Ismailia, Egypt burhanyusufhabibi@suez.edu.eg Muhammad Nazil Iqdami Monash University, Victoria, Australia Muhammad.iqdami@monash.edu Religious Psychosocial Healing for the Trauma of the Children of the Terrorists Miftahuddin IAIN Salatiga, Central Java, Indonesia miftahuddin@iainsalatiga.ac.id Hanung Triyoko IAIN Salatiga, Central Java, Indonesia hanungtriyoko@iainsalatiga.ac.id Burhan Yusuf Habibi Suez Canal University, Ismailia, Egypt burhanyusufhabibi@suez.edu.eg Muhammad Nazil Iqdami Monash University, Victoria, Australia Muhammad.iqdami@monash.edu Abstract This study aims to reveal the trauma-healing process with a religious psychosocial approach to the children of the terrorists. Qualitative method is used to describe the trauma healing process analytically. The subjects in this study were 10 children of perpetrators of acts of terrorism in Surabaya, Qudus International Journal of Islamic Studies (QIJIS) Volume 10, Number 2, 2022 (PP : 319-356) https://journal.iainkudus.ac.id/index.php/QIJIS/index http://dx.doi.org/10.21043/qijis.v10i2.14661 Qudus International Journal of Islamic Studies (QIJIS) Volume 10, Number 2, 2022 (PP : 319-356) https://journal.iainkudus.ac.id/index.php/QIJIS/index http://dx.doi.org/10.21043/qijis.v10i2.14661 Abstract This study aims to reveal the trauma-healing process with a religious psychosocial approach to the children of the terrorists. Qualitative method is used to describe the trauma healing process analytically. The subjects in this study were 10 children of perpetrators of acts of terrorism in Surabaya, Magetan, Banten, and Bogor, who were cared for at the Islamic Boarding School, the Central Java Fostering House, Indonesia. Data collection techniques include in-depth interviews, 319 QIJIS, Volume 10, Number 2, 2022 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami observation, and documentation. This study found that the children of the terrorists have traumas in the form of feelings of excessive threat, remembering repeated and continuous events, and feeling helpless. To get rid of this trauma, the Central Java Islamic Boarding School conducted trauma healing with a religious psychosocial approach through three phases. First, creating a sense of security and stabilizing emotions. Second, training the children to make peace with the trigger of the trauma. Third, training the children to naturally socialize with the pesantren, school, and community environment. The novelty of this research lies in the integration of psychological, social, and religious approaches in trauma healing. The implication of this research is to enrich the trauma-healing model of the children of the terrorists. Keywords: The Children of the Terrorists, Violent Trauma, Trauma Healing, Religious Psychosocial Approach. A. Introduction The 2019 Global Terrorism Index report by the Australian Institute of Economic & Peace (IEP) stated that the number of people who died due to terrorist crimes was 15,952 people (Global Terrorism Index, 2019). The most recent suicide bombing in Pakistan’s Peshawar Mosque on March 4, 2022, killed 56 people and injured 194 (Iswara, 2022). In Indonesia, acts of terrorism are still a real threat (Galamas, 2011). For example, the suicide bombings at 3 churches in Surabaya on May 13, 2018, killed at least 10 people and injured more than 40 people, in addition to 3 suicide bombers who also died. One of the bombers was even a young mother holding her two toddlers (Kriswanto, 2018). A suicide bombing was also carried out by a husband and wife at the Makassar Cathedral Church on March 28, 2021, causing 20 people to be injured QIJIS, Volume 10, Number 2, 2022 320 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... (Rahma, 2021). These incidents imply that the danger of terrorism, both at the international and national levels, still exists and even escalates. Acts of terrorism have a negative impact on society. The impact of terrorism is not only experienced by the families of the victims, but also by the children of the bombers (Kruglandski, 2009; Albrecht, 2007). The children of terrorists experience psychological trauma, especially concerning violence committed by their parents (Boulanger, 2008; Miller, 2006; Shah et al., 2021). Psychologically, children who are in a vortex of violence have a deep traumatic impression of terror incidents (Bracher, 2004; Ulman & Brothers, 2013). These children are likely to have difficulty in adjusting to society. The difficulty of adjustment is because terrorists’ children have been given a negative stigma in society. Stigma is the process of identifying a person’s attributes which are then attached to these people (Scheid & Brown, 2010; Stafford & Scott, 1986). Negative stigma on a person will leave an imprint for a long time and even tends to be permanent (Scheid & Brown, 2010). Psychological trauma of children affected by terrorism has an impact on the subconscious memory of violence that can appear at any time (Crespo & Fernandez-Lansac, 2016; Lubit et al., 2003). Trauma, both physical and psychological injuries, will cause unfavorable conditions and cause responses to certain conditions differently (Asmundson & Katz, 2009; Okifuji & Skinner, 2009; Vincent, 2003). QIJIS, Volume 10, Number 2, 2022 QIJIS, Volume 10, Number 2, 2022 A. Introduction In the psychological trauma of physical and verbal violence by the community against children who are victims of terrorism, their responses are different and can change according to environmental attitudes toward these children. Thus the QIJIS, Volume 10, Number 2, 2022 321 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami memory of violence can re-appear at any time (Boulanger, 2007; Scaer, 2014). In this context, trauma healing which seeks to internalize anti-violence values for children victims of terrorism is urgent work to do (Gayatri & Kosasih, 2019). Trauma healing aims to eliminate the trauma of violence and eliminate negative stigma through value transformation, value transactions, and trans- internalization (Suprapto et al., 2021). One of the institutions that carry out this method is the Foster House Central Java Indonesia Islamic Boarding School. At this Islamic boarding school, trauma healing was carried out in the form of nurturing 10 children of terrorists from several regions in Indonesia. Arimbi’s research revealed that terrorists’ children experienced negative stigma so that some closed themselves off from the environment, but others responded with achievements and wide association in society (Arimbi, 2016). Nugraha complemented Arimbi’s research, especially on aspects of the impact of acts of terrorism on collective punishment for women and children, as well as boarding schools to protect them (Nugraha, 2016). This research strengthens the study of Arimbi and Nugraha on the integration of religious psychosocial approaches on the negative stigma against the children of the terrorists. Aliyah and Mulawarman revealed that terrorism caused trauma to children. They offered to counsel using the Systematic Literature Review (SLR) method (Aliyah & Mulawarman, 2020). Hidayat and Husna added that the role of community leaders and ulama (Islamic scholars) at the local level played a significant role in the resilience of the families of the terrorists (Hidayat & Husna, 2021). Firmasyah discussed the deradicalization QIJIS, Volume 10, Number 2, 2022 322 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... program carried out by National Counterterrorism Agency (BNPT) in the form of guidance on religious insight, national insight, and general social rehabilitation (Firmansyah, 2019). Firmansyah’s study was more juridical in terms of its implementation in the field. This research offers a new approach, different from the SLR of Aliyah and Mulawarman while at the same time sharpening Hidayat and Husna in terms of the role of the local ulama. QIJIS, Volume 10, Number 2, 2022 A. Introduction The new approach is in the form of handling three aspects of post-traumatic healing, namely fear, recurring memories, and helplessness in a more comprehensive manner. This research also reveals a new pattern of deradicalization, which is different from what was done by BNPT, as stated in Firmansyah’s study. The pattern of deradicalization of the BNPT was based on formal state policies, while the religious psychosocial pattern was more based on the local wisdom of the community. The theory used to analyze the trauma of violence experienced by the children of the terrorists is the theory of Judith Lewis Herman. This theory states that trauma inevitably results in a loss. Even those who are lucky to be able to physically avoid injury still lose their internal psychological structure in the form of a loss of security and love (Herman, 1997). He reveals that there are three main symptoms of trauma, namely (a) hyperarousal or feeling excessively threatened; (b) intrusion or remembering events repeatedly and continuously; and (c) constriction or feeling helpless (Herman, 1997). This theory is used to explain the feelings of loss of parental affection experienced by the bombers’ children. This theory is also adequate to explain the feelings of terrorists’ children. Children feel QIJIS, Volume 10, Number 2, 2022 323 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami overly threatened, easily remember the past, and experience hopelessness and even helplessness. Furthermore, Herman states that trauma healing is carried out in three phases: (a) security and stability, namely building a sense of security and stability; (b) remembrance and acceptance, namely remembering the past to be accepted; and (c) reconstruction of relationships, namely building relationships with other people (Herman, 1997). This theory is chosen because it is relevant to describing children’s mental condition and the pattern of parenting carried out by Islamic boarding schools. This theory emphasizes both the pre-condition aspects in the form of maintaining emotional stability, conditioning in the form of accepting values, and post-conditioning in the form of maintaining social relationships. This study aims to reveal the trauma healing process in eliminating the trauma of violence in terrorists’ children, to be constructed into a pattern of trauma healing for children who are victims of social violence in general. Thus, its contribution can be developed in the scientific development of education, especially parenting for groups marginalized by social, political, and economic systems. QIJIS, Volume 10, Number 2, 2022 A. Introduction This research also contributes to the National Counterterrorism Agency (BNPT) in implementing deradicalization and counter-radicalization for the children of the terrorists. Several orphanages for abandoned children and the children of the terrorists can benefit from this research. This type of research is qualitative with the main characteristics of an analytical depiction of phenomena (Miles et al., 2014; Geertz, 1973). This study analytically describes trauma healing in the children of the terrorists, as well as QIJIS, Volume 10, Number 2, 2022 QIJIS, Volume 10, Number 2, 2022 324 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... providing deep meaning for it. The main subjects of this study were 10 children of the terrorists from Surabaya, Magetan, Banten, and Bogor. Three data collection methods are used: in- depth interviews, observation, and documentation. In-depth interviews were used to collect direct data on 10 children of the terrorists, 2 caregivers, 2 administrators, 3 foster siblings, 1 counselor, 1 teacher, and 1 guardian of these children. The interviews took place from August 2021 to October 2021. Observations were used to directly observe the activities of the children and their caregivers at the pesantren (Islamic boarding school). The documentation method was used to examine the children’s population documents, pesantren documents, and other relevant documents. The main data sources in this study are the words, actions and attitudes of the research subjects, so that in-depth interviews and observation are the main data collection methods. Data analysis used the following steps: (1) data reduction; (2) data presentation; (3) drawing conclusions (Miles Matthew et al., 2014). Data on the trauma of violence against the children of the terrorists, and data on the implementation of trauma healing, are reduced by selecting and categorizing data based on the focus of the problem. Furthermore, the data resulting from the reduction is presented in the form of data exposure and concluded as a result of the analysis. The validity of the data is checked through triangulation of sources and methods (Moleong, 2021). Triangulation of sources is done by checking and comparing data from children, caregivers, administrators, foster siblings, teacher, counselor, and guardian. While the triangulation method is done by checking the data with three different methods: in-depth QIJIS, Volume 10, Number 2, 2022 325 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami interviews, observation, and documentation. Interview data were rechecked with observation and documentation data, or vice versa. A. Introduction The factual conditions in the pesantren are used as a database and validated according to the triangulation rules. B. Condition of the Central Java Foster House Islamic Boarding School The Foster House Islamic Boarding School is located in the eastern part of Semarang City, Central Java Province. This Islamic boarding school was founded in 2007. Initially, this pesantren was an orphanage that accommodated orphans and neglected children. Along with the passage of time and demands from the community to develop the orphanage into a boarding school, in 2015, the Foster House Islamic Boarding School was officially declared. Thus, the pesantren is the result of the transformation of the orphanage. Departing from such a journey, the pesantren puts its vision, mission, and goals to remain on the path of caring for neglected children. In the concept of pesantren management, neglected children have several clusters: (a) biologically neglected, namely orphans; (b) economically neglected, namely underprivileged children; (c) lovingly neglected, namely those whose families are broken homes; and ideologically neglected children, namely children from families of terrorist perpetrators (Gf, interview, August 28, 2021). This boarding school is affiliated with the Nahdlatul Ulama (NU) organization, where all of its administrators and caregivers are NU activists. The nuances of NU’s can be seen in the formulation of the vision, mission, and goals of the pesantren as well as the program of activities developed. 326 QIJIS, Volume 10, Number 2, 2022 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... In the pesantren document, it is stated that the vision of the Foster House Islamic Boarding School is to create a boarding house that becomes a reference in assisting orphans, dhu’afa, and neglected children in their efforts to promote Islam, rahmatan lil alamin ‘ala tariqah Ahlissunnah wal Jamaah Nahdlatul Ulama. According to Ms, one of the administrators, the vision was set because the founders wanted to embody the humanitarian vision of NU’s aswaja values (Ms, interview, August 21, 2021). The goals and objectives of the pesantren were formed based on this concept, namely “to care for the children of the poor, both orphans, abandoned children, children who are victims of violence and terrorism, so that they can become successful people, become pioneers of progress and humanity in the midst of society” (Visi, Misi, dan Tujuan Pesantren Rumah Asuh, 2019). The Foster House Islamic Boarding School has 57 students from various social backgrounds. There are orphans, children from underprivileged families, children whose parents have problems, and some children from terrorists. QIJIS, Volume 10, Number 2, 2022 B. Condition of the Central Java Foster House Islamic Boarding School The santri (someone who follows Islamic religious education in pesantren) simultaneously in the pesantren tradition, live and socialize together with fellow students, caregivers, foster siblings, and the surrounding community. This boarding school is integrated with the community environment. There is no fence or barrier between the pesantren complex and the community. Like other NU pesantren, the programs and activities of this pesantren are thick with traditional-moderate religious styles such as recitation of the kitab kuning, bahsul masail, memorizing the Qur’an, reading salawat, tahlil, yasin, QIJIS, Volume 10, Number 2, 2022 327 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami manaqib, and others. Pesantren also periodically hold general recitations, such as commemorating the Prophet’s birthday, Isra’ Mi’raj, the Nuzulul Qur’an, and others. Besides containing activities as usual in NU typical pesantren, this pesantren provides additional soft skills lessons, entrepreneurship training, English, and computational thinking. The Foster House Islamic Boarding School develops a tradition of religious moderation. The religious moderation developed is characterized by four main pillars: love for the homeland, anti-violence, tolerance, and respect for tradition (Badan Litbang dan Diklat Kementerian Agama RI, 2019). Specifically, the pesantren fosters a sense of love for the homeland in the santri through activities to commemorate national holidays, getting santri used to singing the song “syubbanul wathon” at every pesantren event, and organizing a speech competition on the theme of love for the homeland. Meanwhile, the attitude of anti-violence and tolerance is taught through the teaching of books and habituation in association. The strengthening of the attitude of respect for tradition is carried out by visiting the graves of the saints and participating in community activities such as slametan, mertideso, and sadranan (Gf, interview, August 28, 2021). The Pesantren stands on an area of approximately 1,800 m2, integrated with the residential community. There are 8 rooms in the male student complex and 12 rooms in the female student complex which is about 500 meters away. Each room is occupied by 4-5 students. The children of the terrorists are not isolated in their placement but mingled randomly in the general santri rooms. In every 2 rooms, there is one foster brother, except for the room occupied by the QIJIS, Volume 10, Number 2, 2022 328 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... terrorist’s children, one foster brother is placed. QIJIS, Volume 10, Number 2, 2022 B. Condition of the Central Java Foster House Islamic Boarding School The foster brother accompanies the children of the terrorists almost 24 hours a day. According to Th, one of the foster brothers, with such a pattern, those children can socialize with other students at any time without being constrained by negative stigma, as well as psychological and religious assistance at any time (Th, interview, October 7, 2021). The existence of the children of the terrorists in this pesantren is a collaboration with BNPT, the Indonesian Ministry of Religion, and the Indonesian Ministry of Social Affairs since the end of 2018. The roles of each in the collaboration are as follows. Islamic boarding schools provide religious guidance, raise children to normally live in a social environment, and provide psychological assistance. Islamic boarding schools also provide basic needs for children such as food, drink, shelter, and daily equipment, including the cost of transportation to and from school every day and school fees. The role of the Ministry of Religion is to facilitate and ensure that these children can attend school in educational institutions under the auspices of the Ministry of Religion, namely Madrasah Ibtidaiyah and Madrasah Tsanawiyah. The role of the Ministry of Social Affairs is more related to formal legal responsibility for the existence of these children. The role of the BNPT is to provide protection for these parenting activities. The specific reason why the BNPT has entrusted these children is due to the relations of several pesantren personnel who happen to be academics observing terrorism. The BNPT has a trust that the pesantren can become a foster home for children. Regarding funding from BNPT, there is no specific and explicit agreement. In practice, over time, QIJIS, Volume 10, Number 2, 2022 329 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami BNPT personnel came to make donations voluntarily and incidentally several times. The Pesantren accepts care as part of its efforts to deradicalize the children’s religious ideology. The main reason the pesantren accepts these children is for humanitarian reasons. One of the administrators gave an argument that the children of the terrorists were not responsible for the sins of their parents. If no one paid attention, their futures would be bleak, and they could even fall into the path of terrorism like their parents, so they must be fostered (Zr, interview, August 24, 2021). QIJIS, Volume 10, Number 2, 2022 C. Violence Trauma of the Children of the Terrorists The violence experienced by terrorist children results in trauma in the form of a loss of security and love. There is a difference in the trauma symptoms in the age category. Trauma experienced by children under the age of 12 shows the main symptom of loss of affection. Whereas children over 12 years of experience undergo more trauma in the form of a deep fear of the memory of violence. According to Herman (1997), there are three symptoms of trauma experienced by victims of violence: (a) hyperarousal; (b) intrusion; and (c) constriction (Herman, 1997). B. Condition of the Central Java Foster House Islamic Boarding School Acceptance of children was carried out in three periods: 7 children in November 2018, 1 child in January 2019, and 2 children in March 2021. The acceptance process is based on the agreement between the BNPT and the Islamic Boarding School with the approval of the children’s families. At the time of handing over the children, the BNPT representative stated that the children, apart from receiving negative stigma from the community, also experienced the trauma of violence due to witnessing acts of terrorism committed by their parents. Even some children had been exposed to the ideology of terrorism by their parents (Zr, interview, August 24, 2021). The trauma of acts of terrorism was quite imprinted on their memory. The Pesantren took care of the children of the terrorists to relieve the trauma they had endured. 330 QIJIS, Volume 10, Number 2, 2022 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... Table 1. Identity of the Children of the Terrorists No. Name (Initial) Gender (M/F) Age (y.o) Scholl & Grade Name of the Father (Initial) Additional Information 1 Dv F 15 State Voca tional High School Grade XI Is The Father is the Bomber in Magetan 2 Fr F 12 Private Islamic Elementa ry School Grade VI Is Dv’s younger sister 3 Hk M 9 Private Islamic Elementa ry School Grade III Is Fr’s younger brother 4 An M 17 Vocational High School Grade XI Af-Ps The parents were the ter rorists at an apartment in Surabaya 5 Fz F 14 State Islamic Junior High School Grade IX Af-Ps An’s younger sister 5 Hd M 12 Private Islamic Elementa ry School Grade VI Af-Ps Fz’s younger brother 7 Ai F 10 Private Islamic Elementa ry School Grade V Tm-Te Her parents were the suicide bomber at the police headquar ter in Surabaya Table 1. Identity of the Children of the Terrorists 331 QIJIS, Volume 10, Number 2, 2022 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami 8 Rt F 15 Private Is lamic Junior High School Grade IX Ar Her father is the stabber of the Minister Wiran to in Pande gelang Banten 9 Um M 14 School dropped out In His parents were terrorists seized in Bogor 10 Hr M 12 School dropped out In Um’s younger brother 1. Hyperarousal Hyperarousal means feeling excessively threatened. The children of the terrorists who are being cared for at the pesantren are seen to be traumatized by violence. The trauma of violence begins with excessive fear of a violent situation, both physical and verbal, that has been seen or experienced. QIJIS, Volume 10, Number 2, 2022 332 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... The terrorist children when they first came to the pesantren seemed afraid of the people they met. One of the caregivers at the pesantren stated that when the children first arrived, they seemed afraid to meet other people and were closed and behaved like people who had a certain trauma (Mz, interview, August 30, 2021). They show a moody, aloof, and difficult-to-socialize personality (Ir, interview, September 8, 2021). In line with Mz and Ir, a foster brother named Dz stated that at the beginning of the first year when the children came, they were frightened when they saw guests at the pesantren in police uniforms, because the parents of terrorist children had a lot of dealings with the police (Dz, interview, October 9, 2021). The state of fear when seeing the police lasted for a long time has not even completely disappeared until now. Gf, one of the caregivers, said: Honestly, until now they are (some) still traumatized. Still really hate the police. For them the police are their enemy. Thus, some are still afraid. Sometimes there are babins (police unit) who are asked by the Densus (Anti-terrorist Specific Detasement) to just visit them and give them a snack to make them happy so that their blood pressure is a little lower. So they are still traumatized by the police. (Gf, interview, August 28, 2021) Some children who had seen the trauma their parents suffered or who had personally experienced violence in the form of being yelled at or beaten still retained traumatic memories associated with this violence (Th, interview, October 7, 2021). These events QIJIS, Volume 10, Number 2, 2022 333 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami cause excessive anxiety and fear. Even if the fear is so excessive, when they see incidents of non-physical violence, they feel afraid, like seeing a schoolmate being scolded by their teacher. Violent incidents or witnessing violence against children are things that continue to make them afraid of triggering conditions excessively (van Der Kolk & Fisler, 1995). This can be seen in their awkward behaviors and even fears of seeing new people they have not known previously. Some children of the terrorists in the pesantren in the first 3 months did not want to meet people they did not know (Dz, interview, October 9, 2021). RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... When asked why those children were afraid, the reason was that those new people for them were bad people (Rt, interview, September 24, 2021). Children perceived all unknown people as a threat to themselves. This kind of fear of excessive threats was the result of the physical and verbal abuse he had experienced in the past. 2. Intrusion Children who are victims of child abuse remember events over and over again. They find it difficult to forget their past experiences. For children, psychological trauma due to violence will be carried over into adulthood (Bushman & Huesmann, 2006) Terrorist children are deeply traumatized by the violence of acts of terrorism. Terrorists’ children are traumatized by the terrors perpetrated by their parents. At first, the children did not think that their parents would commit acts of terrorism. The child’s surprise at the actions of his parents makes QIJIS, Volume 10, Number 2, 2022 334 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... his or her memory record events relatively permanently (Fonagy, 2004). Ai, took part in the action by her parents when they blew themselves up at the Surabaya Police Station. Luckily Ai bounced back and was saved before the bomb exploded. Ai stated that the trauma still lingered in her memory (Ai, interview, September 24, 2021). Psychological trauma due to violence will be difficult for children to remove and even carry over for a long time. When children recall past events that caused trauma, the trauma will reappear, leading to psychological disorders and problems (Ayers, 2007). Among these psychological issues include anger, vindictiveness, and even a tendency to cover up or a preference for lying. In this case, the children of the terrorists show apathy, lie, unfriendly, and get angry easily (Ir, interview, September 8, 2021). Trauma-triggering factors determine the frequency of recurring memories and influence behavior in the future (Bariah, 2019). Hd, for example, if questioned about his parents, would spontaneously show an angry face and avoid answering. This attitude was an expression of disapproval of his parents’ past story being revealed again (Th, interview, October 7, 2021). Memories of past violence cause a person to be demonstrative, easily provoked by emotions and other mental disorders (Sumartiningsih, 2019; Fallot, 2013). 3. Constriction The feeling of helplessness over the difficult situation experienced by the terrorists’ children is the culmination of the trauma of violence they suffer. This QIJIS, Volume 10, Number 2, 2022 335 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami triggers a desire for revenge. The feeling of wanting revenge can arise at any time. Children who witness high- intensity violence such as acts of terrorism cause deep psychological trauma and can sometimes push their subconscious to commit acts of violence again. (Stein, 2013; Weissman et al., 2010) Revenge can lead to demonstrative attitudes for children. An, a son of a suicide bomber who witnessed his father’s actions, until several years after the incident, still remembered the past events, even expressively, he still showed a grudge against the police. (An, interview, September 21, 2021; Rt, interview, September 24, 2021). Mn once witnessed a child being alone while drawing his father’s face while writing longing sentences to his father. These children held a grudge against the police, but felt powerless, so they just wrote it down in a note. Mn said, “I once met a child, he drew the faces of three people, when I asked whose picture there were, the child answered father, mother and brother, I miss and want to help father” (Mn, Interview, August 29, 2021). Feelings of helplessness will result in an attitude of despair. At first, the children seemed resigned to the situation that befell them. This could be seen in the symptoms that seemed like not talking much, following what the foster brother said even though sometimes it was not in accordance with his wishes, and being indifferent to every event. According to Th, the children were sometimes ‘angry’ they did not want to do something that their foster brother asked them to do, but in the end, they were forced to do it (Th, interview, October 7, 2021). QIJIS, Volume 10, Number 2, 2022 336 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... The feeling of helplessness of the terrorists’ children deepens when they receive negative stigma from society (Yang et al., 2007). Ai is an example of this case. When her grandfather was about to take her home and sent her to a private school in her hometown, the school refused to accept her on the grounds that she was a child of a terrorist. 3. Constriction Likewise, the community where the child came from, also objected to receiving her back for the same reason (Gf, interview, August 28, 2021). Ai’s experience was also experienced by several other terrorist children. The strength of this stigma further exacerbated children’s sense of helplessness in dealing with community culture. Pesantren offers treatment as part of long-term trauma rehabilitation utilizing a mix of psychological, social, and religious approaches to eliminate this trauma. QIJIS, Volume 10, Number 2, 2022 D. Trauma Healing: A Combination of Psychological, Social, and Religious Approaches Trauma healing is a post-traumatic healing process experienced by a person, so that he can continue his life properly without the shadow of the incident (Kusumandari, et.al., 2019). Trauma can be caused by several events such as: serious illness or injury, death of a loved one, experiencing violence or witnessing violence. These events can cause a person to experience post-traumatic stress disorder (PTSD) (Trappler, & Newville, 2007). PTSD must be treated quickly and appropriately so as not to interfere with the survival of the victim. PTSD experienced by the terrorists’ children is mainly related to experiencing acts of violence or witnessing QIJIS, Volume 10, Number 2, 2022 337 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami violence in the form of acts of terrorism. Herman (1997) states that the handling of PTSD in the form of trauma healing is carried out in three phases: (a) security and stability; (b) remembrance and acceptance; and (c) reconstruction of relationship (Herman, 1997). QIJIS, Volume 10, Number 2, 2022 1. Security and Stability Security and stability is the stage of building a sense of security and emotional stability for the victim. The children of the terrorists psychologically experience the emptiness of love from their parents. They feel insecure and emotionally unstable. To eliminate this, the pesantren assigns foster siblings to accompany them every day. Foster siblings are chosen by the pesantren from among students who have a volunteer spirit (Ir, interview, September 8, 2021). The foster brothers or sisters become a place to express feelings. Foster siblings also direct their interactions with fellow students, and the community around the pesantren. The role of foster siblings is to provide religious guidance, in the form of worship and accompanying religious activities (Interview, Dz, October 9, 2021). With a 24 hours per day psychological assistance, the emptiness of love from their parents can begin to be resolved. Each foster sibling lives together in one room with 2-3 children. Thus, daily needs, including pocket money, study needs, worship habits, reading materials for religious teaching, and the daily behavior of children are under the foster sibling care. In this way, foster siblings are free to provide psychological assistance, especially when QIJIS, Volume 10, Number 2, 2022 QIJIS, Volume 10, Number 2, 2022 338 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ... children experience symptoms of trauma, for example, by calming, advising, inviting them to pray, and inviting recitations in the community. Besides that, some caregivers replace the role of their parents in the pesantren. The caregivers at this pesantren are the guardians for all students. The students place caregivers as their parents while in the pesantren. Likewise, the terrorist children in this pesantren feel protected by the presence of caregivers. One of the children stated that our needs are met by the caretaker father and mother (Ai, interview, September 24, 2021). Caregivers also provide an example of daily life with love. They teach the values of tolerance, non- violence, respect for community culture, and love for the homeland. The method used by the pesantren in teaching moderation among children is through teaching and habituation (Mz, interview, August 30, 2021). (Gf, interview, August 28, 2021). At the pesantren, Children are taught how to get along well, how to worship, and have noble character. The study of religious books is also chosen with Islamic nuances, the nuances of love or rahmatan lil ‘alamin. 1. Security and Stability Some basic religious books such as the Qur’an, hadith, fiqh, creed, and morals are taught to them without being distinguished whether the students are from the terrorists’ families or from ordinary people (Gf, interview, August 28, 2021). The specific study added to them is the cultivation of love for the homeland and love of peace with the approach of religious arguments. Caregivers specifically guide how to understand the postulates of love for the homeland and peace. In conveying the arguments for love for the homeland, caregivers sometimes encounter difficulties. Especially for children who have been exposed to a deep ideology of terrorism, the process of internalizing the values of patriotism is initially mental. This is because, from the start, their parents have indoctrinated those children that the Republic of Indonesia (NKRI) is an infidel state (Gf, interview, August 28, 2021). The terrorists believe that Islam must be defended from the misconduct of the government (Afriyanti, 2012). Some basic religious books such as the Qur’an, hadith, fiqh, creed, and morals are taught to them without being distinguished whether the students are from the terrorists’ families or from ordinary people (Gf, interview, August 28, 2021). The specific study added to them is the cultivation of love for the homeland and love of peace with the approach of religious arguments. Caregivers specifically guide how to understand the postulates of love for the homeland and peace. In conveying the arguments for love for the homeland, caregivers sometimes encounter difficulties. Especially for children who have been exposed to a deep ideology of terrorism, the process of internalizing the values of patriotism is initially mental. This is because, from the start, their parents have indoctrinated those children that the Republic of Indonesia (NKRI) is an infidel state (Gf, interview, August 28, 2021). The terrorists believe that Islam must be defended from the misconduct of the government (Afriyanti, 2012). Through religious habituation in the style of NU, the internalization of peaceful and tolerant religious values in children can be formed. In the NU pesantren tradition, accommodation of local culture is a matter of concern. In this pesantren, grave pilgrimage activities, nyadran, halal bihalal, and so on are carried out. Children are involved in these various activities (Mz, interview, August 30, 2021). With this involvement, children can get used to respecting the culture of the community. This kind of thing is not usually taught by their parents. 1. Security and Stability These books are commonly used among NU, including: ‘aqidah al-’awam, safinah al-najah, fathu al-qarib, ta’lim al-muta’allim, and akhlak li al-banin. The existence of foster siblings and caregivers in carrying out assistance to the terrorists’ children is supported by the role of counselors. Periodically, foster siblings coordinate with counselors to provide accurate psychological treatment. Counselors provide input to foster siblings regarding specific matters in maintaining QIJIS, Volume 10, Number 2, 2022 339 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami the children’s emotional stability. Counselor Wh said that the children need to be protected from their feelings so that it is not easy to remember their past (Wh, interview, September 10, 2021). To maintain emotional stability, the pesantren also carry out various activities such as outbound, speech exercises, and skills training. Another activity to keep emotions in check is to involve them in sports competitions between Islamic boarding schools. During outbound and sports competitions, children feel happy and comfortable (Hd, interview, September 21, 2021). These kinds of social-recreational activities can help children in the stability-building phase. In building this emotional stability, in addition to psychological and social approaches, religious approaches are also emphasized. Religious approach is used to strengthen emotional and spiritual control of children (Grangvist, 2014). The Pesantren provides Islamic teachings that promote compassion or Islam rahmatan lil ‘alamin. Islam rahmatan lil ‘alamin in the Islamic tradition is also known as wasathiyyah Islam which balances individual piety with social piety and gets used to being tolerant (Shihab, 2019). Islam rahmatan lil ‘alamin is manifested in a pesantren environment by practicing religious traditions such as istighatsah, tahlilan, yasinan, grave pilgrimage, the commemoration of religious holidays, and others (Mz, interview, August 30, 2021). The curriculum given to the terrorists’ children in this pesantren is generally the same as for other students. QIJIS, Volume 10, Number 2, 2022 340 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... Some basic religious books such as the Qur’an, hadith, fiqh, creed, and morals are taught to them without being distinguished whether the students are from the terrorists’ families or from ordinary people (Gf, interview, August 28, 2021). The specific study added to them is the cultivation of love for the homeland and love of peace with the approach of religious arguments. 1. Security and Stability With the involvement of this kind of activity, the pesantren hopes that the children QIJIS, Volume 10, Number 2, 2022 341 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami can internalize the values of love for ancestral traditions. Love for tradition is expected to make people wise in religion (Gf, interview, August 28, 2021). This first phase takes a relatively long time, almost a year (Gf, interview, August 28, 2021). Indicators of the success of this phase include: the children seeming to feel comfortable hanging out with their foster siblings and caregivers. They also do not show a closed attitude anymore. They also want to go to school and can participate in learning activities (Mz, interview, August 30, 2021). After the first phase is considered successful, where a sense of security and emotional stability has been maintained, the next phase is carried out. 2. Remembrance and Acceptance In this second phase, terrorist children are trained to get used to the triggers of trauma and begin to be taught to accept the situation as normal. This is done gradually because it is not easy for them to immediately accept the reality of violence in their past (Gf, interview, August 28, 2021). In implementing this phase, pesantren uses psychological, social, and religious approaches. The psychological approach is carried out by preparing the emotions and stabilizing the children’s psychology so that they are mentally ready to meet trauma triggers. How to do with individual assistance to them? Assistance, in this case, is also combined with a religious approach, namely the provision of religious advice that contains love, not to hate, let alone hold grudges. To QIJIS, Volume 10, Number 2, 2022 342 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... build readiness to make peace with conflict triggers, children are invited to carry out routine zikr activities at Islamic boarding schools. This is felt to help the children in maintaining emotional stability, especially in accepting past circumstances or events because it is believed to be God’s provision (Mz, interview, August 30, 2021). As for the social approach in this second phase, the pesantren carries out several activities, such as getting used to meeting people who are not yet known. Almost every week, guests from various backgrounds visit to make donations at the pesantren. The Pesantren also often receive visits from BNPT officers. Children generally know them as policemen. Children are trained to get used to meeting them in order to get used to the triggers of the trauma. The method taken by the pesantren is that the children are accustomed to shaking hands and serving drinks to the police guests (Mz, interview, August 30, 2021). In the end, the children are able to accept the existence of the police. This is a sign that these children are able to live side by side with one of the triggers of the trauma. Another activity to familiarize children with accepting their past is to accept requests for visits to several government agencies at compensation events. The students are also included in the compensation activities at the police station several times. One time, the local police chief took the children for a vacation to the Mall (Gf, interview, August 28, 2021). 2. Remembrance and Acceptance At first, some children did not want to follow because they were afraid. However, in the end, with the direction of the foster brothers, those children were brave to join (Dz, interview, October 9, 343 QIJIS, Volume 10, Number 2, 2022 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami 2021). The process of remembrance and acceptance is a relatively difficult phase because it requires mental strengthening of those children. 3. Reconstruction of Relationship The final phase of trauma healing is relationship reconstruction. In this phase, children begin to be taught how to rebuild relationships with others. Initially, it was still difficult to build relationships with other santri friends. For example, they still like to be quiet, aloof, and even reluctant to be with their friends in one meeting (Th, interview, October 7, 2021). In this final phase, the pesantren emphasizes the social approach. The social approach is carried out by bringing terrorists’ children into a wider association. Starting with their friends at school. At school, they were introduced to new friends, including teachers. At first, it was still awkward. One of the teachers at the school said that when they first came, the children looked scared and awkward in communicating with teachers and friends (Kh, interview, September 12, 2021). As time went by, the terrorists’ children finally got to hang out at school, even though it still seemed awkward sometimes. Socialization for the terrorists’ children is also carried out by providing opportunities for children to get along with the community around the pesantren. Because the pesantren complex with the surrounding community does not have a guardrail, the children hang out with the surrounding community. Every congregational QIJIS, Volume 10, Number 2, 2022 344 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... prayer in the village mosque, they blend into one with the neighborhood (Ir, interview, September 8, 2021). In social encounters with the congregation in the mosque like this, the children feel accepted by their environment. More broadly, those children of the terrorists and other Islamic boarding school students often attend donation activities from several government and private organizations. In these kinds of activities, those children feel they have been accepted by society. On the other hand, the community has also become part of the social trauma healing process. Handling children of terrorism must be carried out together with a whole community approach (Owojori et al., 2020)conduct, scope, and outcomes of Nigeria’s counterterrorism efforts to a critical analysis. This work adopted qualitative research methods (primary and secondary data collection. The psychological and religious approaches in this last phase are integrated with the social approach. In the social approach, children are routinely included in recitation at the neighborhood mosque. When listening to religious lectures at the mosque, children’s psychology can usually be touched by moderate Islamic views, not extreme ones. QIJIS, Volume 10, Number 2, 2022 3. Reconstruction of Relationship In this pesantren tradition, after the children take part in the recitation, it is continued with bahtsul masail activities, namely discussing the studies that have been delivered during the recitation and other themes. One of the caregivers said that it is traditional for the children at this pesantren to join congregational prayers and take part in recitations at the mosque and bahtsul masail activities (Gf, interview, August 28, 2021). 345 QIJIS, Volume 10, Number 2, 2022 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami The indiscriminate treatment of the children of the terrorists has been able to gradually make those children feel that they have been accepted by the wider community, without any negative stigma. This condition has contributed to the children forgetting the trauma of their past violence. With this social approach, children seem more open in socializing, and not as awkward or closed as they were when they first arrived. By carrying out three phases of trauma healing through a combination of psychological, social, and religious approaches, the trauma of the terrorists’ children has been resolved relatively permanently. The trauma of violence from their parents’ acts of terrorism gradually began to disappear. After a span of three years, from 2018 to 2021, those children of the terrorists have been able to resume their lives in the midst of society. 346 QIJIS, Volume 10, Number 2, 2022 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ... QIJIS, Volume 10, Number 2, 2022 E. Conclusion This study found that terrorists’ children experienced deep trauma from the violence and acts of terrorism committed by their parents in the past. Such trauma was disclosed in several symptoms, such as hyperarousal, intrusion, and constriction (Herman, 1997). Hyperarousal appeared in the way these children isolated themselves, QIJIS, Volume 10, Number 2, 2022 347 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami were being closed to everyone, felt inferior, and were afraid of meeting people, especially the police. In contrast, intrusion was evident as they became petulant, vindictive, apathetic, and unfriendly. Constriction was seen as they tended to become more sensitive, desperate, and eager to take an act of revenge, especially to the police. The trauma could emerge at any time as long as there were triggers — symbols that lay deep in the trauma — such as seeing the police. This trauma also created memories to act on behalf of their past grudges. For example, there were children who, when remembering their father who had died due to being shot by the police, held deep hatred towards the police and wanted to take a revenge. The pattern of trauma healing for terrorists’ children as developed at the Foster House Islamic Boarding School integrated psychological, social, and religious approaches. The integration of the three approaches was carried out in three phases: security and stability, remembrance and acceptance, and reconstruction of relationship (Herman, 1997). Security and stability phase focused on giving the children psychological assistance by providing shelter to replace the role of parents and monitoring counselors as well as religious assistance through religious teaching with nuances of love. Meanwhile, the remembrance and acceptance phase was intended to help them gradually accept their past through some teachings, such as how to serve drinks to guests and meet with the police at various activities. The reconstruction of the relationship phase was carried out by providing opportunities for children to naturally socialize with their friends in their pesantren, at schools, and in the wider community. QIJIS, Volume 10, Number 2, 2022 348 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN ..... REFERENCES Afriyanti, D. (2014). Islamic education and youth extremism in Indonesia.. Journal of Policing, Intelligence and Counter Terrorism, 7(2), 134-146. https://doi.org/10.1080/183 35330.2012.719095 Albrecht, H. J., & Kilchling, M. (2007). Victims of terrorism policies: Should victims of terrorism be treated differently?. European Journal on Criminal Policy and Research, 13: 13-31. https://doi.org/10.1007/s10610- 007-9038-3 Aliyah, U. & Mulawarman, M. (2020). Kajian systematic literature review (slr) untuk mengidentifikasi dampak terorisme, layanan konseling dan terapi trauma pada anak-anak. ISLAMIC COUNSELING Jurnal Bimbingan Konseling Islam, 4(2), 209. https://doi.org/10.29240/ jbk.v4i2.1759 Arimbi, R. S. (2016). Dampak stigma sebagai “anak teroris”: studi kasus mengenai tumbuh kembang anak-anak narapidana kasus terorisme di Indonesia. Prosiding Seminar Nasional Psikollogi Indigenous Indonesia, 580– 596. Asmundson, G. J. G., & Katz, J. (2009). Understanding the co- occurrence of anxiety disorders and chronic pain: State- of-the-art. Depression and Anxiety, 26(10), 888–901. https://doi.org/10.1002/da.20600 Ayers, S. (2007). Thoughts and emotions during traumatic birth: A qualitative study. Birth, 34(3), 253–263. https:// doi.org/10.1111/j.1523-536X.2007.00178.x 349 QIJIS, Volume 10, Number 2, 2022 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami Badan Litbang dan Diklat Kementerian Agama RI. (2019). Moderasi Beragama. Kementerian Agama RI. Bariah, O., & Marlina, R. (2019). Tinjauan psikologi dan agama terhadap tindak kekerasan pada anak. Jurnal Studia Insania, 7(2), 92-107 Boulanger, G. (2007). Understanding and treating adult onset trauma (1st ed.). Routledge. https://doi.org/https:// doi.org/10.4324/9780203837795 Boulanger, G. (2008). Witnesses to reality: Working psychodynamically with survivors of terror. Psychoanalytic Dialogues, 18(5), 638–657. https://doi. org/10.1080/10481880802297673 Bracher, M. (2004). Healing trauma, preventing violence: A radical agenda for literary study. JAC, 515–561. Bushman, B. J., & Huesmann, L. R. (2006). Short-term and long-term effects of violent media on aggression in children and adults. Archives of Pediatrics and Adolescent Medicine, 160(4), 348–352. https://doi.org/10.1001/ archpedi.160.4.348 Crespo, M., & Fernandez-Lansac, V. (2016). Memory and narrative of traumatic events: A literature review. Psychological Trauma: Theory, Research, Practice, and Policy, 8(2), 149–156. https://doi.org/10.1037/ tra0000041 Fallot, R.D., & Blanch, A. K. (2013). Religious and spiritual dimensions of traumatic violence. In. K. I. Pargament, A & E.P. Shafranske (Eds.). APA Handbook of Psychology, 350 QIJIS, Volume 10, Number 2, 2022 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... Religion, and Spirituality: An Applied Psychology of Religion and Spirituality (2), 371-387. American Psychological Association. https://doi.org/10.1037/14046-019 Firmansyah, R. (2019). Rehabilitasi dan deradikalisasi terhadap anak pelaku tindak pidana terorisme. Jurist-Diction, 2(2), 669. https://doi.org/10.20473/ jd.v2i2.14258 Fonagy, P. (2004). Early-life trauma and the psychogenesis and prevention of violence. REFERENCES Annals of the New York Academy of Sciences, 1036, 181–200. https://doi.org/10.1196/ annals.1330.012 Galamas, F. (2011). Profiling bioterrorism: Present and potential threats. Comparative Strategy, 30(1), https:// doi.org/10.1080/01495933.2011.545689 Gayatri, E., & Kosasih, K. (2019). Break the chain of terrorism through deradicalization of (former) terrorists’ children: Narration from Al-Hidayah Pesantren, Medan. Analisa: Journal of Social Science and Religion, 4(01), 79–96. https://doi.org/10.18784/analisa.v4i01.792 Geertz, C. (1973). Thick description: Toward an interpretive theory of culture. In The Interpretation of Cultures: Selected Essays (1973). In Basic Books, Inc. (pp. 310– 323). https://philpapers.org/archive/geettd.pdf Global Terrorism Index, G. (2019). Global terrorism index, GTI. 1–56. www.economicsandpeace.org Grangvist, P. (2014). Mental health and religion from an attachment view point: Overview with implications for QIJIS, Volume 10, Number 2, 2022 351 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami future research. Mental Health, Religion, and Culture, 17 (8), Taylor & Francis Online, https://doi.org/10.1080/ 13674676.2014.908513 Hidayat, M., & Husna, S. (2021). Resiliensi keluarga teroris a: Kekuatan menghadapi stigma negatif, rasa malu dan psychological distress sebagai keluarga teroris. Sosio Konsepsia, 10(02), 159–176. https://ejournal.kemsos. go.id/index.php/SosioKonsepsia/article/view/2389 Herman, J.L. (1997). Trauma and recovery: The aftermath of violence-from domestic abuse to political terror. Basic Books. Iswara, A. J. (2022, March). Ledakan Bom Bunuh Diri di Masjid Pakistan, 56 Tewas dan 194 Luka-luka. Kompas.Com 26. https://www.kompas.com/global/ read/2022/03/05/113200470/ledakan-bom-bunuh- diri-di-masjid-pakistan-56-tewas-dan-194-luka-luka Kriswanto, J. (2018, May). Serangan bom di tiga gereja Surabaya: Pelaku bom bunuh diri “perempuan yang membawa dua anak.” BBC News. https://www.bbc.com/ indonesia/indonesia-44097913 Kruglandski, A. W. & Fishman, S. (2009). Psychological Factors in terrorism and counterterrorism: Individual, group, and organizational levels of analysis. Social Issues and Policy Review, 3(1), 1-44. https://doi.org/10.1111/ j.1751-2409.2009.01009.x Kusumandari, R. B., et.al. (2019). Game learning to optimize learning and trauma healing in children after disaster in Banyumas. Book Proceeding, 381-386. https:// QIJIS, Volume 10, Number 2, 2022 352 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... repository.stiesia.ac.id/id/eprint/2984/6/SU- AFBE_2018%20%28BOOK%20PROCEEDING%29. pdf#paage=389 Lubit, R., Rovine, D., Defrancisci, L., & Eth, S. (2003). Impact of trauma on children. Journal of Psychiatric Practice, 9(2), 128–138. Miles, M. B., Huberman, A. M., & Saldana, J. (2014). Qualitative data analysis, A methods sourcebook edisi ketiga. Sage publications inc: Los Angeles. Miller, L. (2006). The terrorist mind: I. A psychological and political analysis. International Journal of Offender Therapy and Comparative Criminology, 50(2), 121–138. https://doi.org/10.1177/0306624X05281405 Moleong, L. J. (2021). Metodologi penelitian kualitatif. PT Remaja Rosdakarya. Nugraha, M. T. (2016). Dampak aksi ekstrimisme dan terorisme terhadap collective punishment pada wanita dan anak- anak. REFERENCES Jurnal Harkat : Media Komunikasi Gender, 12(1), 49–55. https://doi.org/10.15408/harkat.v12i1.7579 Okifuji, A., & Skinner, M. (2009). Psychological aspects of pain. Current Therapy in Pain, 39, 513–518. https://doi. org/10.1016/B978-1-4160-4836-7.00070-5 Owojori, A. J., Azman, M. D., & Suppaiah, B. R. K. (2020). Vested interests and politicisation of terrorism in Nigeria: A critical terrorism studies perspective. Pertanika Journal of Social Sciences and Humanities, 28(4), 2823–2841. https://doi.org/10.47836/PJSSH.28.4.18 QIJIS, Volume 10, Number 2, 2022 353 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami Rahma, A., (2021, March). Pelaku bom bunuh diri di Gereja Katedral Makassar adalah suami istri. Tempo.Co. https:// nasional.tempo.co/read/1447085/pelaku-bom-bunuh- diri-di-katedral-makassar-adalah-suami-istri Scaer, R. (2014). The body bears the burden trauma, dissociation, and disease. Routledge. https://doi.org/ https://doi.org/10.4324/9780203081822 Scheid, T. L., & Brown, T. N. (2010). Social context, theoris, and systems, handbook for the study of mental health (2nd ed.). Cambridge University Press. Shah, A., Riaz, H., Sajjad, U., & Ahmad, S. (2021). “Irreversible damage to psyche”: Representation of trauma in Fatima Bhutto’s narratives. Ilkogretim Online, 20(4). https:// doi.org/ 10.17051/ilkonline.2021.04.153 Shihab. M. Q. (2019). Wasatiyah Islam tentang moderasi beragama. Lentera Hati. Stafford, M. C., & Scott, R. R. (1986). Stigma, deviance, and social control. In The dilemma of difference (77–91). Springer. Stein, A. (2013). Prologue to violence: Child abuse, dissociation, and crime. Routledge. Sumartiningsih, M. S., & Prasetyo, Y. E. (2019). Literature review: Pengaruh cognitive behaviour therapy terhadap posttraumatic stress disorder akibat kekerasan pada anak. Jurnal Pendidikan Keperawatan Indonesia, 5(2), 167-176. https://doi.org/10.17509/jpki.v5i2.17429 354 QIJIS, Volume 10, Number 2, 2022 RELIGIOUS PSYCHOSOCIAL HEALING FOR THE TRAUMA OF THE CHILDREN .... Suprapto, W., Maftuh, B., Sjamsuddin, H., & Malihah, E. (2021). Internalization of living value education program (lvep) as a based of developing conflict resolution model. Journal of Education, Teaching and Learning, 6(1), 31– 38. http://dx.doi.org/10.26737/jetl.v6i1.2305 Trappler, B., & Newville, H. (2007). Trauma healing via cognitive behaviour therapy in chronically hospitalized patients. Psychiatr Q, (78), 317-325. https://doi. org/10.1007/s11126-007-9049-8 Ulman, R. B., & Brothers, D. (2013). The shattered self: A psychoanalytic study of trauma. Routledge. van Der Kolk, B., & Fisler, R. (1995). Dissociation and the fragmentary nature of traumatic memories: Overview and exploratory study. Journal of Traumatic Stress, 8(4), 505–525. https://doi.org/10.1002/jts.v8:4 Vincent, C. (2003). Understanding and responding to adverse events. New England Journal of Medicine, 348(11), 1051–1056. https://doi.org/10.1056/nejmhpr020760 Visi, Misi, dan Tujuan Pesantren Rumah Asuh, (2019). Visi, Misi, dan Tujuan Pesantren Rumah Asuh, (2019). Weissman, S. H., Busch, K. G., & Schouten, R. (2010). The evolution of terrorism from 1914 to 2014. Behavioral Sciences & the Law, 28(2), 211–223. https://doi. org/10.1002/bsl Yang, L. H., Kleinman, A., Link, B. G., Phelan, J. C., Lee, S., & Good, B. (2007). Culture and stigma: Adding moral experience to stigma theory. Social Science and Medicine, 64(7), 1524–1535. https://doi.org/10.1016/j. socscimed.2006.11.013 QIJIS, Volume 10, Number 2, 2022 355 Miftahuddin, Hanung Triyoko, Burhan Yusuf Habibi, and M. Nazil Iqdami Interviews An interview with Ai, the terrorists’ children, Central Java, September 24, 2021. An interview with An, the terrorists’ children, Central Java, September 21, 2021. An interview with Dz, the foster sibling, Central Java, October 9, 2021. An interview with Gf, the caregiver, Central Java, August 28, 2021. An interview with Hd, the terrorists’ children, Central Java, September 21, 2021. An interview with Ir, the foster sibling, Central Java, September 8, 2021. An interview with Kh, the teacher, Central Java, September 12, 2021. An interview with Mn, the guardian, Central Java, August 29, 2021. An interview with Ms, the administrator, Central Java, August 21, 2021. An interview with Mz, the caregiver, Central Java, August 30, 2021. An interview with Rt, the terrorists’ children, Central Java, September 24, 2021. An interview with Th, the foster sibling, Central Java, October 7, 2021. An interview with Wh, the counselor, Central Java, September 10, 2021 . An interview with Zr, the administrator, Central Java, August 24, 2021. Interviews An interview with Ai, the terrorists’ children, Central Java, September 24, 2021. An interview with An, the terrorists’ children, Central Java, September 21, 2021. An interview with Dz, the foster sibling, Central Java, October 9, 2021. An interview with Gf, the caregiver, Central Java, August 28, 2021. An interview with Hd, the terrorists’ children, Central Java, September 21, 2021. An interview with Ir, the foster sibling, Central Java, September 8, 2021. An interview with Kh, the teacher, Central Java, September 12, 2021. An interview with Mn, the guardian, Central Java, August 29, 2021. An interview with Ms, the administrator, Central Java, August 21, 2021. An interview with Mz, the caregiver, Central Java, August 30, 2021. An interview with Rt, the terrorists’ children, Central Java, September 24, 2021. An interview with Th, the foster sibling, Central Java, October 7, 2021. An interview with Wh, the counselor, Central Java, September 10, 2021 . An interview with Zr, the administrator, Central Java, August 24 2021 An interview with Wh, the counselor, Central Java, September 10, 2021 . An interview with Zr, the administrator, Central Java, August 24, 2021. 356 QIJIS, Volume 10, Number 2, 2022
https://openalex.org/W4362542441	https://aacr.figshare.com/articles/journal_contribution/Table_TS2_from_Intertumoral_Differences_Dictate_the_Outcome_of_TGF-_Blockade_on_the_Efficacy_of_Viro-Immunotherapy/22546502/1/files/40009982.pdf	Azerbaijani	null	Table TS2 from Intertumoral Differences Dictate the Outcome of TGF-β Blockade on the Efficacy of Viro-Immunotherapy	null	2,023	cc-by	181	Table S2. List of primers used for RT-qPCR analysis. Gene Forward Reverse S4Q 5’-CGCTTTTGAAGGTCGTGTATCA-3’ 5’-CTGGCTGTGCTGAGATTGTTTT-3’ Ifit-1 5’-CTGGACAAGGTGGAGAAGGT-3’ 5’-AGGGTTTTCTGGCTCCACTT-3’ Ifit-2 5’-TGCTCTTGACTGTGAGGAGG-3’ 5’-ATCCAGACGGTAGTTCGCAA-3’ Ifit-3 5’-GTGCAACCAGGTCGAACATT-3’ 5’- AGGTGACCAGTCGACGAATT-3’ Irf7 5’-GACCGTGTTTACGAGGAACC-3’ 5’-GCTGTACAGGAACACGCATC-3’ Isg15 5’-GGAACGAAAGGGGCCACAGCA-3’ 5’-CCTCCATGGGCCTTCCCTCGA-3’ Oas1b 5’-AGCATGAGAGACGTTGTGGA-3’ 5’-GCGTAGAATTGTTGGTTAGGCT-3’ Ddx58 5’-AAGGCCACAGTTGATCCAAA-3’ 5’-TTGGCCAGTTTTCCTTGTCG-3’ Cxcl9 5’-TGGAGTTCGAGGAACCCTAGT-3’ 5’-AGGCAGGTTTGATCTCCGTT-3’ Cxcl10 5’-ACGAACTTAACCACCATCT-3’ 5’-TAAACTTTAACTACCCATTGATACATA-3’ Mx1 5’-GATGGTCCAAACTGCCTTCG-3’ 5’-TTGTAAACCTGGTCCTGGCA-3’ β2m 5’-CTCGGTGACCCTGGTCTTT-3’ 5’-CCGTTCTTCAGCATTTGGAT-3’ Bst2 5’-ACATGGCGCCCTCTTTCTATCACT-3’ 5’-TGACGGCGAAGTAGATTGTCAGGA-3’ Rsad2 5’-GGTGCCTGAATCTAACCAGAAG-3’ 5’-CCACGCCAACATCCAGAATA-3’ Ctgf 5’-GGCCTCTTCTGCGATTTCG-3’ 5’-CCATCTTTGGCAGTGCACACT-3’ Id-1 5’-ACCCTGAACGGCGAGATCA-3’ 5’-TCGTCGGCTGGAACACAT-3’ Mmp2 5’-TTCTGTCCCGAGACCGCTAT-3’ 5’-GTGTAGATCGGGGCCATCAG-3’ Serpin E1 5’-GCCAACAAGAGCCAATCACA-3’ 5’-AGGCAAGCAAGGGCTGAAG-3’ Snail 5’-AGCCCAACTATAGCGAGCTG-3’ 5’-CCAGGAGAGAGTCCCAGATG-3’ TGF-β1 5’-CAACAATTCCTGGCGTTACC-3’ 5’-TGCTGTCACAAGAGCAGTGA-3’ TGF-β2 5’-TCCCCTCCGAAAATGCCATC-3’ 5’-TGCTATCGATGTAGCGCTGG-3’ TGF-β3 5’-CGACCGGATGAGCACATAGC-3’ 5’-TTTGCTTCTTGAGACGCCCC-3’ Mzt2 5’-TCGGTGCCCATATCTCTGTC-3’ 5’-CTGCTTCGGGAGTTGCTTTT-3’ Ptp4a2 5’-AGCCCCTGTGGAGATCTCTT-3’ 5’-AGCATCACAAACTCGAACCA-3’ Table S2. List of primers used for RT-qPCR analysis. Gene Forward Reverse S4Q 5’-CGCTTTTGAAGGTCGTGTATCA-3’ 5’-CTGGCTGTGCTGAGATTGTTTT-3’ Ifit-1 5’-CTGGACAAGGTGGAGAAGGT-3’ 5’-AGGGTTTTCTGGCTCCACTT-3’ Ifit-2 5’-TGCTCTTGACTGTGAGGAGG-3’ 5’-ATCCAGACGGTAGTTCGCAA-3’ Ifit-3 5’-GTGCAACCAGGTCGAACATT-3’ 5’- AGGTGACCAGTCGACGAATT-3’ Irf7 5’-GACCGTGTTTACGAGGAACC-3’ 5’-GCTGTACAGGAACACGCATC-3’ Isg15 5’-GGAACGAAAGGGGCCACAGCA-3’ 5’-CCTCCATGGGCCTTCCCTCGA-3’ Oas1b 5’-AGCATGAGAGACGTTGTGGA-3’ 5’-GCGTAGAATTGTTGGTTAGGCT-3’ Ddx58 5’-AAGGCCACAGTTGATCCAAA-3’ 5’-TTGGCCAGTTTTCCTTGTCG-3’ Cxcl9 5’-TGGAGTTCGAGGAACCCTAGT-3’ 5’-AGGCAGGTTTGATCTCCGTT-3’ Cxcl10 5’-ACGAACTTAACCACCATCT-3’ 5’-TAAACTTTAACTACCCATTGATACATA-3’ Mx1 5’-GATGGTCCAAACTGCCTTCG-3’ 5’-TTGTAAACCTGGTCCTGGCA-3’ β2m 5’-CTCGGTGACCCTGGTCTTT-3’ 5’-CCGTTCTTCAGCATTTGGAT-3’ Bst2 5’-ACATGGCGCCCTCTTTCTATCACT-3’ 5’-TGACGGCGAAGTAGATTGTCAGGA-3’ Rsad2 5’-GGTGCCTGAATCTAACCAGAAG-3’ 5’-CCACGCCAACATCCAGAATA-3’ Ctgf 5’-GGCCTCTTCTGCGATTTCG-3’ 5’-CCATCTTTGGCAGTGCACACT-3’ Id-1 5’-ACCCTGAACGGCGAGATCA-3’ 5’-TCGTCGGCTGGAACACAT-3’ Mmp2 5’-TTCTGTCCCGAGACCGCTAT-3’ 5’-GTGTAGATCGGGGCCATCAG-3’ Serpin E1 5’-GCCAACAAGAGCCAATCACA-3’ 5’-AGGCAAGCAAGGGCTGAAG-3’ Snail 5’-AGCCCAACTATAGCGAGCTG-3’ 5’-CCAGGAGAGAGTCCCAGATG-3’ TGF-β1 5’-CAACAATTCCTGGCGTTACC-3’ 5’-TGCTGTCACAAGAGCAGTGA-3’ TGF-β2 5’-TCCCCTCCGAAAATGCCATC-3’ 5’-TGCTATCGATGTAGCGCTGG-3’ TGF-β3 5’-CGACCGGATGAGCACATAGC-3’ 5’-TTTGCTTCTTGAGACGCCCC-3’ Mzt2 5’-TCGGTGCCCATATCTCTGTC-3’ 5’-CTGCTTCGGGAGTTGCTTTT-3’ Ptp4a2 5’-AGCCCCTGTGGAGATCTCTT-3’ 5’-AGCATCACAAACTCGAACCA-3’ Table S2. List of primers used for RT-qPCR analysis.
https://openalex.org/W1970320808	https://europepmc.org/articles/pmc3820086?pdf=render	English	null	From Sociocultural Disintegration to Community Connectedness Dimensions of Local Community Concepts and Their Effects on Psychological Health of Its Residents	Psychiatry journal	2,013	cc-by	12,601	Hindawi Publishing Corporation Psychiatry Journal Volume 2013, Article ID 872146, 13 pages http://dx.doi.org/10.1155/2013/872146 Hindawi Publishing Corporation Psychiatry Journal Volume 2013, Article ID 872146, 13 pages http://dx.doi.org/10.1155/2013/872146 Hindawi Publishing Corporation Psychiatry Journal Volume 2013, Article ID 872146, 13 pages http://dx.doi.org/10.1155/2013/872146 Tom Sørensen,1,2 Robert Kleiner,3 Paul Ngo,4 Andreas Sørensen,5 and Nils Bøe Tom Sørensen,1,2 Robert Kleiner,3 Paul Ngo,4 Andreas Sørensen,5 and Nils Bø 1 Division of Mental Health and Addiction, Institute of Clinical Medicine, University of Oslo, Norway 2 Nordliveien 6, 1482 Nittedal, Norway 3 Emeritus, Department of Sociology, Temple University, Philadelphia, PA, USA 4Psychology, St. Norbert College, De Pere, WI, USA 5 Research Assistant, Division of Research, North Coast Psychiatry, Nittedal, Norway 5 Research Assistant, Division of Research, North Coast Psychiatry, Nittedal, Norway 5 Research Assistant, Division of Research, North Coast Psychiatry, Nittedal, Norway Correspondence should be addressed to Tom Sørensen; tom.sorensen@medisin.uio.no Received 22 November 2012; Revised 8 April 2013; Accepted 30 April 2013 Received 22 November 2012; Revised 8 April 2013; Accepted 30 April 2013 Academic Editor: Petros Skapinakis Academic Editor: Petros Skapinakis Copyright © 2013 Tom Sørensen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. In a series of community mental health promotion studies in Lofoten, Norway, the concept of sociocultural integration is used to describe properties of a local community that are related to people’s psychological health. Starting with Durkheim’s description of a cohesive society, we compare different concepts that are related to sociocultural integration, for example, sense of community, social capital, and social cohesion. We then examine the relationship of various individual oriented social psychological concepts to sociocultural integration. These concepts often share theoretical and operational definitions. The concept of sociocultural integration in the Lofoten studies was proved to be very valuable in understanding how the properties of a community can affect people’s mental health and their social psychological properties. It has also shown its value in the planning of mental health services and demonstrating its success in concrete community-based mental health promotion projects. Thus they could make important contributions to further studies and actions in local communities where the intersection between the individual, their social network, and their local community occurs. 3. Sociocultural Integration-Disintegration and Its Application in Lofoten Leighton, in “My name is Legion” [1], the first book on the stirling County Study of psychiatric disorders and sociocul- tural environment, gives a detailed theoretical framework for the connection between the sociocultural structure of a local community and the amount of psychiatric disorders among its inhabitants. He looked at a community as a self- integrating and self-regulating unit or “quasiorganism.” This “functional” point of view reflected an orientation deriving in part from anthropology, for example, Radcliffe-Brown [11] and Malinowski [12], and also from sociology, for example, Merton’s [13] and Hyman and Singer [14] discussions of anomie and reference groups, and Parson’s [15] emphasis on the “systems approach.” Other writers discussed this strategy in terms of consensus (trend towards agreement on orienta- tions or meanings) and functional interdependence (comple- mentary interactions and exchange of goods and services) in social systems [16, 17] or used concepts like organization- disorganization [18]. i The present studies from Lofoten in North Norway are in debt to Leighton’s comprehensive approach to the mental health of communities, (a) mapping the need for psychiatric attention/services, (b) analytically relating them to commu- nity functioning, (c) developing and describing mental health services in a small rural community [9], and (d) carrying out mental health promotion projects through community development [10]. The Lofoten projects were, in the main, evaluated by surveys designed to interview total adult popu- lations. Indicators have been developed that show the effects of project priorities, the program targets, and measures of progress, that is, having mental health promotion indicators at individual, organizational, and community levels. The ultimate objective of such a focus is to identify the social and psychological capital (or resources) in each community that can be mobilized to deal with the psychological problems that exist in their respective communities, that is, factors of impor- tance for promoting mental health and preventing men- tal illness. g In the work done by Leighton, integration-disintegration is treated as one dimension, and communities (in Nova Scotia) could be ordered according to their sociocultural inte- gration-disintegration. He listed types of factors that affect the integration-disintegration. Initially, “causes” for disinte- grating the community were (a) a recent history of disaster, (b) widespread ill health, (c) extensive poverty, (d) cultural confusion, (e) widespread secularization, (f) extensive in and out migration, and (g) rapid and widespread social change. 1. Introduction from Leighton’s Stirling County study, Dalgard et al. [6, 7] and Sørensen [8] combined a psychiatric service perspective with an analytical focus on the local community that itself contributed to the mental health of a population.h The seminal Stirling County study [1] took social psychiatry to a position where mental health promotion could be trans- lated into concrete community involvement. From a social psychiatry that linked social stratification and social class to prevalence of psychiatric disorders [2, 3], Leighton amplified the ecological approach to include concrete community processes and to operational measures of the phenomena, thereby linking the context of geographic areas to specific characteristics, that is, the sociocultural integration of a local community to the occurrence and distribution of mental disorders. A main finding from the Stirling County studies was the stronger effects on mental health of higher levels of sociocultural integration of a local community than higher social class within those same communities [4, 5]. Taking off The major thrust of our basic and applied research project has been to contribute effective and sustainable strategies for treatment, intervention, prevention, and promotion of mental health and well-being, especially at the local commu- nity level. Unfortunately, the definition of “community” can vary from one social science discipline to another and even vary in the same discipline, making it difficult to evaluate and generalize from research using the concept. We defined communities in terms of geographical boundaries, that is, local communities, ethnic composition, or enclaves of people that may or may not have systematic interpersonal relations and obligations to each other. The main aim is to relate Psychiatry Journal 2 of their culture, social life, work, and, for some, even family within the same geographic area. the concept of sociocultural integration-disintegration, as it evolved in the Lofoten mental health promotion studies to other concepts used to describe the local community and its linkage to mental health. For bad and good, inhabitants of small communities as in Lofoten are more affected by the level of sociocultural inte- gration. In more urbanized areas people would usually have their work and recreational activities outside their residential neighborhood, and their personal network would often have many locations. If a crisis occurs, it would not necessary hit all important supportive structures at the same time. 2. Lofoten and the Lofoten Mental Health Promotion Studies The Lofoten region in North Norway has about 24,000 inhabitants and consists of a series of islands stretching out into the Gulf Stream north of the Arctic Circle. The region is composed of six municipalities. The main income of this region has historically come from the Norwegian Arctic cod fisheries in the wintertime. Work connected with the fisheries has been important in all of the local communities we have studied. In the six Lofoten municipalities, peopled employed in the fisheries varied from 7,7% to 30,7% when the Lofoten mental health promotion studies were started in the early 1990. But in the two largest municipalities, more people were employed in both industry and health and social services. In the last year of the 1980s, the usual migration of cod that come to spawn along the coasts of Lofoten in February-March failed to occur. The authorities introduced limited quotas that could be taken by each boat. That put the fishermen, boat owners, and the fishery companies ashore into an economically impoverished situation. The mental health promotion project was derived from this fishery crisis.h 1. Introduction On the other hand, efforts to change a neighborhood would not have the same total impact as could be done in smaller local com- munities like Lofoten. Hence, the generalizations from the Lofoten studies are most relevant for smaller local communi- ties with a relative high overlapping of social network sectors. 3. Sociocultural Integration-Disintegration and Its Application in Lofoten The first task of this group was to systematize the information that would be gathered, especially with regard to what was of value in the local community, what should be preserved, and what was wanted. The next initiative was to conduct open “popular meetings,” to invite all local inhabitants, and particularly to encourage attendance by people who would be stimulated in the project to actively participate in working groups. These “popular” meetings were to be “idea workshops” based on various themes, including culturally defined issues, family issues, local environmental problems, tourism, trade, and industry. The participants concluded the meetings by assign- ing priorities of the ideas brought forward. The third step was based on the work done in the “popular” meetings and focused on establishing a formal organization whose aims would be derived from the priorities, or themes, decided upon at the meetings. Working groups around thematic pro- jects were established. Reports from various initiatives were brought back to subsequent meetings. From our own experience, the core of the disintegration process is the disruption of social communication systems and structures of mutual assistance. The signs are the absence of functional leaders and people with ideas and initiatives, with a lack of people to follow such ideas when they are put forward. The community viability is reduced with economic problems, dissolving families and other formal and informal organizations, deteriorating interpersonal relations, and the weakening of norms about what is right or wrong undermin- ing their physical, psychological, and social security. These changes in a local community do not make a place for growth, self-realization, acknowledgment, recreation, and being to- gether with their significant social network(s). The total pic- ture is one of increasingly burdensome situations with in- creased risk of mental problems on the individual level.h p The “Local Community Development Project” in Lofoten focused on the entire population of seven selected local com- munities. The activities that evolved illustrate what emerged in the context of the seven Lofoten municipalities at a time of economic difficulties in the fisheries. The selection of sites for local community development had, as its point of depart- ure, the ongoing fishery crisis. Hence, fishery-related local communities were picked. The explicit criterion brought forward in the dialog between the municipality and the project leader was to have the local communities represent all dimensions and structures of the fisheries, including its links with farming. 3. Sociocultural Integration-Disintegration and Its Application in Lofoten Each of the selected communities had a history, an identity, a name, geographical boundaries, and an “informal” social structure. They actually constituted “the old school districts” that have (or have had) an elementary school within the borders of the community (the old school house could now be used as a community center). Often the com- munity also had a church building and a graveyard.h In each of the seven Lofoten local communities, surveys were carried out to acquire information from residents to be used in the community development programs, as well as data to evaluate possible changes in both personal mental health and in properties of the community structures, and hence three cross-sectional surveys of the same seven local communities were carried out. The followups were linked by community and not by individual. The T1 surveys of the local communities that participated in the mental health promotion project took place from June 1991 to May 1992, namely, within the first few months of the community acti- vities. The followup of all seven communities, that is, at T2, was carried out in 2000 and a third T3 survey took place in 2010. No externally organized community interventions took place after 1994. To varying degrees, project activities were continued on a local basis. The context and the planning process was an essential part of the health promotion strategy. The Lofoten project was given legitimacy from two sources, (a) from key authorities in the region, by having a series of meetings with people involved in health and social services, and (b) by meeting with politicians and administrators in each of the municipalities in the Lofoten region. Ideas for health promotion measures were discussed and translated into concrete plans. A second step was to have the overall project has its office based within the regional political liaison body, that is, “Lofotr˚adet.” The focus on the interrelation between mental health and the totality of people’s life context was also reflected in the composition of the steering group of “Liv Laga i Lofoten,” which included having the mayor in one of the municipalities as its leader. Several different initiatives occurred in the same population in the same time period, stimulating interactive processes. The activities were not to be fully defined at the start of the project. 3. Sociocultural Integration-Disintegration and Its Application in Lofoten Later, a second set of “effects” of disintegration included (i) high frequency of broken homes, (ii) few and weak associations, (iii) few and/or weak leaders, (iv) few recreation activities, (v) high frequency of hostility, (vi) high frequency of crime and delinquency, and (vii) weak and frag- mented networks of communication. Looking at these “causes” and “effects” showed to us that integration-disinte- gration is multidimensional and not unidimensional [19]. A group of mental health promotion projects named “Liv Laga in Lofoten” were carried out during the 1990s using as points of departure the Lofoten psychiatric outpatient clinic. Early in the project, “Local Community Development” focused on the entire adult population of seven selected local communities. The Lofoten studies were at the intersection of a theoretical approach to the community-individual mental health interaction and the practice of evaluating and setting into action mental health promotion in local community contexts. The impact of local communities and the relevance of sociocultural integration to living conditions and mental health can best be seen in communities like Lofoten where the present studies were carried out. Here, people have most Based on surveys of seven local communities in Lofoten, factor analyses of statements evaluating the local community identified eight factors or dimensions to the integration-dis- integration concept. Four dimensions were related to the core community integration functions: initiative and coop- eration, leadership, community identity, and contact and 3 Psychiatry Journal the project, and be responsive to real local needs, and thereby also promote leadership and cooperative competence.h communication, and four were related to risks and resources in the community: future economic viability, leisure time and recreation, milieu for children and youths, and security when sick or old. Each statement included the name of the respondent’s actual local community. p p p p The “Local Community Development Project” was clearly focused on such locals’ feelings of ownership and identifi- cation. After meeting the municipal leaders, the first initia- tive in the community development process was to contact key community members (in each of the chosen local com- munities) by visiting the community, getting to know people, and gathering information about their local community. Thereafter an informal local working party was established with members of the key local social network(s). 3. Sociocultural Integration-Disintegration and Its Application in Lofoten They had to allow for concrete input from the various segments of the community as an inherent part of local mobilization. In this way, one could secure the target groups’ or locals’ feelings of ownership, identification with 4. Comparing Concepts—The ‘‘Cohesive Community’’ in Perspective Currently relatively few researchers explicitly use Leighton’s integration-disintegration concept, but the literature is filled with references that make use of different concepts or terms that sound different but, in reality, have similar meanings to our perspective. Conversely, there are also many references that make use of the same concept or term but the actual meaning of the term differs from one group of researchers to another. This situation makes it difficult to integrate such material except at some vague theoretical level. Too often we find ourselves superimposing a vague general unproductive 4 Psychiatry Journal perspective on the literature. We need clearly defined con- cepts that have importance for our perspective even though they may be seen and used differently in the literature. Gusfield [33] said that “territorial” dimensions of com- munities are an oversimplification. One needs the relational dimensions as well. Some communities, like local commu- nities in this paper, are defined primarily according to ter- ritory, but even proximity or shared territory may not by itself constitute a community. Factor analyses of an urban neigh- borhood yield two distinct factors [34], named social bound- ing and physical rootedness, similar to Gusfield’s dimensions above. “Together with the family, the neighborhood is one of the few places where a community can emerge without exter- nal intervention [35]”. f Historically, the perspective of societal disorganization causing personal disorganization was put forth by the French philosopher Comte and applied specifically to psychiatry by Audifferend. The opposite position, namely, that personal disorganization caused disturbances of the society, was put forward by Bastide [20]. A starting point for defining a community could be Durkheim’s [21, 22] description of a cohesive society, that is, a society characterized by a multitude of mutual moral supports, which does not depend on individ- uals own resources but leads them to share their the collective energy and mutual support. Durkheim defined the cohesive society in terms of mutually defined and agreed upon role expectation and goal directed behavior which would lead to positive mental health. Tonnie’s division of social groups into “gemeinschaft” and “gesellschaft” [23] is also a forerunner of the concept of the “cohesive community.”h For us, social capital [36] emerges as a meaningful con- cept because it embraces individual (internal) and social (external) resources, the degree of sociocultural integration, and the interaction of both types of resources. 4. Comparing Concepts—The ‘‘Cohesive Community’’ in Perspective Social capital is often used in a narrower and simpler way than the way we intend to use it. Although in the literature, social capital may be defined in terms of any of the following properties: (a) val- ues of social networks, (b) the bonding of similar people, (c) linking diverse people with norms of reciprocity [37], or (d) properties that facilitate collective action [38–40]. But defi- nitions can also consider internal cognitive factors aspects of social capital which also describe community characteristics, that is, degree of trust, reciprocity of norms, and the quality of interpersonal bonds [41].i The Chicago School of Sociology [24] conducted early significant empirical investigations using the community ecological approach focusing on the effects of selected social properties of environment. Faris and Dunham [25] mapped the distribution of mental disorders by various characteristics of communities in the urban environment. Influenced by the Field theory perspective, Parker and Kleiner [26] have demonstrated the potent effects of considering simultane- ously the interaction of psychological, social, and objective properties of the situational context on mental health status and mental disorder. Weinman and Kleiner [27] have shown how the deliberate manipulation of the community context in which groups of patients (i.e., patient networks) live influ- ences therapeutic success and the mental health status of the patients. The variables they explicitly dealt with included group cohesion, social support systems, and methods of con- flict resolution. In our view, a geographically defined community is socially integrated or cohesive to the degree that it has inter- nal and external social capital. A cohesive community can be described as a society with a high shared sense of morality and common purpose, aspects of social control and social order, high level of social interactions within communities, and a high sense of belonging to a place [42]. Coleman [43] introduces the concept of social capital and underscores its effects on the formation of human capital in a community, and how social capital, especially social cohesion, is linked to the health of communities. Other studies have also dealt with social capital and health [44–46].i l Leighton emphasized that increased economic and edu- cational opportunities were not enough to bring about a turn for the better in a disintegrated community. 4. Comparing Concepts—The ‘‘Cohesive Community’’ in Perspective One needed the development and enhancement of socially important pro- cesses: leadership, followership, and practice in cooperatively working together, thus enabling people to gain confidence that they could do things to better their lot. This could perhaps explain the findings from research on regenerated neighborhoods in the UK [28], which illustrate how difficult it is to shift the perception of a community’s residents and the intermediary agencies, and the finding of no mental health improvement following the urban regeneration project [29]; that is, bettering in the standard of living is not enough. l d ld b d fi d h “ b l In sum, researchers mainly describe five domains of social cohesion in geographically delimited rural communities and urban neighborhoods. They include (a) common values and civic culture, (b) social order and social control, (c) social solidarity and reductions in wealth disparities, (d) social networks and social capital, and (e) place attachment and identity. Hence, social cohesion, broadly considered, would encompass most of Lofoten study’s four core dimensions related to community integration, as well as their relation to mental health. 5. Sense of Community and Other Place Related Concepts More developed local communities encourage the development of a sense of community among their community residents. The absence of a sense of community has been found to bring about feelings of alienation, isolation, and loneliness [51], while a strong sense of community has been linked to a range of positive outcomes including improved well-being, empowerment, sense of efficacy, life satisfaction, and hap- piness [52–54, 54, 55].i Other definitions are given in the literature that vary in complexity of operational definition but are analogous and even partially the same. For example, McMillan and Chavis [56] included four dimensions: (a) membership under which resubsumed boundaries, emotional safety, a sense of belong- ing, and identification), (b) ability to influence others or being influenced by the others (c) integration and fulfillment of needs bytheir participation in the community, and (d) shared emotional and participating linkages to a shared history. Obst et al. [57] added “conscious identification” as a fifth dimen- sion. Unger and Wandersman [48] suggested that sense of community includes three components: a social component (emotional and instrumental support and social networks), a cognitive component (cognitive mapping of the physical environment and symbolic communication), and an affective component (the emotional attachment individuals have to persons living around them). This specification seems clearly related to Leighton’s concept of integration-disintegration. With a community focus, social support, in terms of concrete collective assistance, also acts to cope with the emer- gence of risks, stresses, the escalation of problems, and struc- tures that influence such processes. With adequate resources within a local community, a possible risk situation may be resolved and not be experienced as a stressful life situa- tion. If this network action is recognized by the individual, it would manifest itself as a positive main effect of social sup- port in relation to mental health [76]. In the surveys in Lofo- ten, we noted substantial shared contributions of personal social support and the dimensions of sociocultural integra- tion in relation to mental health and well-being [73]. Two items in the social support index, that is, “belonging to a group,” and “help when sick”, were most potent among the local community properties. Place attachment can be described in the same terms as the bonding that occurs between individuals and their impor- tant environments [58, 59]. 5. Sense of Community and Other Place Related Concepts Social disorganization could be defined as the “inability of a community structure to realize the common values of its residents and maintain effective social controls” [30]. This approach also sees local communities and neighborhoods as complex systems of friendship, kinship networks, and formal and informal associations, rooted in ongoing family life and socialization processes [31, 32]. Clearly they have a lot in common with Leighton’s view and the views of other researchers that have focused on the cohesive commu- nity. A concept clearly related to social integration is the “sense of community” [47–49]. It has been used in local community studies and in projects dealing with “planned healthy com- munities” [50]. Farrell et al. suggested that this construct includes the perception of similarity to others, an acknowl- edged interdependence by giving to or doing for others, expecting the same from others, the feeling that one is part of a larger dependable and stable structure, that is, the capacity Psychiatry Journal 5 6.1. Social Support. The perception of “social support” in one’s informal social networks and its relation to the integration of the network is important. Originally the construct was seen only from the individual perspective and not reflecting on the local community or social networks social support as a system in its own right. For example, Keyes and Shapiro [71] define social well-being as an individual’s self-report of the quality of his or her relationship with other people, the neighborhood, and the community. Lindenberg [72] also speaks of community when individuals realize multiple well- being goals within their group. His description of social well- being, as feeling recognized and accepted by others and being liked, and receiving confirmation for one’s behavior is similar to the definition of social support we have used in the Lofoten study [73]. The distribution of personal expectations of social support could also be used as an indicator of social capital in a community when individual data is collectively analyzed and seen as a reflection of property of the community [74]. Social support enables individuals to use more effective coping strategies that facilitate problem comprehension and action, thereby reducing emotional distress [75]. At the community level this, could measure potential resources for meeting life events that affect a significant part of its population, and/or the community as a whole. of community members to influence each other, and the interdependence of the individual and community. 5. Sense of Community and Other Place Related Concepts Some authors use sense of place as a somewhat broader concept, that is, sense of community as one dimension [60] and place attachment as the other. Here sense of community is defined as the residents’ place attach- ment. Overlapping place related descriptions would be place dependence and place identity. Place attachment has been used in studies of mental health behavior among adolescents [61]. Other related concepts are “sustainable communities” [62] and “community connectedness” [63, 64]. 6. Individual Capital and Social Capital The latter can have favorable potential for key networks within a local community. Subse- quent analysis from Lofoten found social support and sense of coherence to have independent, overlapping, and interactive components with regard to mental health [76]. 6.3. Empowerment. Rappaport linked the concept of “em- powerment” to community psychology. It leads to collective actions to improve the quality of life in communities and to important linkages among community organizations [82, 83]. For Rappaport the aim of empowerment was to enhance the possibility for people to control their own lives. Thus empowerment was defined as a process: the mechanism by which people, organizations, and communities gain mastery over their own lives [84]. Psychological empowerment may generally be described as the connection between a sense of personal competence, a desire for, and a willingness to take action in the public domain [85, 86] or as a social-action pro- cess that promotes the participation of people, organizations, and communities towards the goals of increased individual and community control, political efficacy, improved quality of life, and social justice [87]. The various approaches in the field tend to depend on methods that are specific to a discipline and where the data may be empirically narrow or superficial. Earlier, we made the point that Leighton’s studies depended on anthropological methods, but it is also important to indicate that other methods were necessary to make sure that this position was justified. For example, they used individual interview surveys based on sampling procedures, and medical records to investigate people’s mental health. This was clearly nec- essary, as well, in the Lofoten studies. However, the main method in Lofoten depended on survey data that gathered the total adult population in each community. We analyzed the changes in the data from three points in time in each commu- nity. However, we found that other sources of data were necessary to evaluate the effectiveness of our projects. More specifically, popular meetings in each community occurred at the beginning of the period and during the project period. Detailed notes were kept and audio tapes were made during the entire process and at all community and committee meet- ings. Newsletters were published in the same period. After the project terminated, there were also systematic interviews at meetings with key persons in each community to give feed- back about their experiences during the follow-up periods. 6. Individual Capital and Social Capital 6.2. Sense of Coherence. Antonovsky introduced “sense of coherence” and defined it as “the extent to which one has a pervasive, enduring though dynamic feeling of confidence . . .” [77], that is, a personality-related factor. It is clear from this definition that it is the person’s relationship with both his/her internal and external environments that determines the degree of this factor. He also proposed that this feeling of control is shaped by the influence of collective evaluation of one’s important social networks [78]. Other authors proposed the reverse direction of causality, that is, people with a strong sense of personal control use their social network(s) more often and more efficiently, thereby affecting their perceived social support [79, 80]. A stable community, providing stable social support, could directly enhance the development of a stronger sense of coherence among its inhabitants [81]. In sum, the sense of coherence, even if primarily a personality- related resource can be influenced by the integration prop- erties of the community, but even more important is that an individual’s psychological makeup can contribute to Social capital is clearly a community level concept. The way the Lofoten studies have measured it also sees it as a com- munity level concept. All descriptive statements refer to the actual local community (using its name), and the data is ana- lyzed in terms of aggregate properties. Human capital, that is, individual psychological resources, is analyzed together with social capital. In the Lofoten study, social support [65], sense of coherence [66, 67], empowerment [68], self-esteem [68], self-efficacy [69], and mastery [70] would be examples of such human capital related to psychosocial coping behavior. Social capital is defined in terms sociocultural integration of a local community. Social capital could influence and interact with human capital expressed as sociopsychological resources. In mental health promotion projects, such individual factors could be important intervening variables that enhance the development of the community’s capacity to deal with such problems. Psychiatry Journal 6 in ecological psychology [95, 96]. Measurement of “sense of community” focused on how people experienced their com- munity [97]. They did this by asking questions about indi- viduals’ perception, understanding, attitudes, feelings, and so forth about community, and how they relate to it. This is for example, measured by the Neighborhood Cohesion Index (NCI) [98, 99]. But this scale does not so much enquire for risks and resources.hi community development. 6. Individual Capital and Social Capital The project coordinator for the total project was interviewed systematically about the project and other developments in each of the communities. Rogers et al., using a factor analytic procedure with their “Making Decisions Empowerment Scale,” identified five dimensions of empowerment, that is, self-efficacy/self- esteem, power/powerlessness, community activism, righ- teous anger, and optimism/control over the future as key components [88, 89]. This was a point of departure for the empowerment concept in the Lofoten study. In our analyses of the Lofoten surveys, we identified four factors, resembling, but not identical with Rogers’ factors. These were community activism, power/powerlessness, self-determination, and self- esteem [90].i More recent definitions have made important distinctions between the subjective experience of psychological empow- erment and the objective reality of empowered community structures. Psychological empowerment can be defined as a feeling of greater control over one’s life and may occur with- out participation in collective political action. Community empowerment includes a heightened level of psychological empowerment amongst its members influenced by collective political action, or the achievement of favorable redistribu- tions of decision-making resources in the community. 8. Community Promotion and Mental Health Targeting the local community in mental health promotion was the optimal place for dealing with intimate personal concerns and the wider sociopolitical context [100]. In the Norwegian context, psychiatric health services have treat- ment responsibilities related to fairly small geographic areas and hence focusing on community mental health promotion was feasible and could be part of the practice. A study of how a particular local community, “The Road” in Nova Scotia, manifested a decrease in mental problems when it was deliberately transformed from a disintegrated to an integrated community [10] argues for the validity of a causal relationship between sociocultural integration of a local community and mental disorder. Leighton emphasized that increased eco- nomic and educational opportunities are not enough to bring about a turn for the better in a disintegrated community. One also needs the development and enhancement of socially important processes, that is, leadership, followership, and practice in working together cooperatively. This would enable people to gain confidence that they can do things to better their lot.f 7. Measuring ‘‘Community’’ Dalgard, in a series of studies [107, 108], has developed community health profiles that were either presented to authorities of the given communities or directly used in projects with risk groups that manifested increased symptom levels [109, 110]. y p In a study by Latkin and Curry [111], perceptions of one’s local community, as indicators of social disorganization and social stress, predicted depressive symptoms at a 9-month follow-up interview, suggesting the need for structural inter- vention in local communities. On the individual level, social integration was related to plasma concentration of fibrinogen among elderly men, which may help explain the association between social integration and mortality in men [112]. A study that examined the interaction of structural character- istics of a neighborhood and the subjective experience of social cohesion revealed that high levels of community social cohesion reduce the adverse effects of small-area deprivation on mental health [113–115]. The perception of higher levels of cognitive social capital (e.g., trust in neighbors) is associated with a lower risk of developing major depression during a 2-3 year followup [116]. g g Looking at two other communities in the project which were highly integrated at T1, that is, in the early phase of the intervention project, one remained integrated, even increas- ing, over the nearly 20-year follow-up period, and there are indications that the mental health promotion project contri- buted to this development [102]. The process of how the objective reality targeted the social, and subjective reality could clearly be seen in the second of these communities. The decline of this community seemed to follow the construction of a bridge that connected the community with the municipal center, that is, how a particular geographical change could disturb the economic base and social organization of this local community. The first community could be seen as ref- lecting the opposite causal process. The cohesive social rela- tions helped a community make the most of its geographic context and developed the community with positive changes in all three realities. A main asset in this first community y p A report from the University of Glasgow [42, 117] describes eight domains of social capital. In relation to these domains the researchers list appropriate policies to support them. 7. Measuring ‘‘Community’’ During recent years, the popu- lation had declined and many services had moved to the more central parts of the municipality. In sum (at T1), we could observe a process of societal disorganization. In the inter- vening time, T1-T2, this local community was exposed to two opposing fields of forces. Resources like school, shops, and other services were closed down or moved to more central parts of the municipality. On the other hand, through actions to save and further develop a home for the elderly, many of the inhabitants came together and through cooperation succeeded in their “struggle.” People from this community participated in courses about care for the elderly, including training to create organisations and for developing leadership skills. The home for the elderly also developed into a commu- nity centre, bringing together people for voluntary communal work, including an organisation to attend to the needs of the residents of the institution, as well as provide recreational activities for the general population. The development was compared to six other local communities in Lofoten. Of the seven investigated local communities in the first series of surveys (T1), this community had the highest mean (or score) on nervous symptoms, the second lowest on well-being, and the lowest perceived social/support. At T2, all these three indicators positively increased and got close to the mean Lofoten scores. Compared to the other investigated commu- nities in Lofoten at T1, the inhabitants also placed themselves low on most of the indexes of community integration. This indicated a low level of sociocultural integration. From T1 to T2, we observed an increase on the majority of the indexes. Although the increase was often not significant when we looked at the indexes one by one, but the tendency seemed clear. At T2 this community had moved closer to the level of the other local communities in Lofoten, hence, a development towards higher integration. Looking at two other communities in the project which Authors who have used the disintegration concept in mental health promotion have had different targets to change [103]. To some, it has been the (a) entire population, especially in terms of developing leadership and cooperation as in the Leighton and present project [104], (b) community clubs aimed at community image and community decision making [105], or (c) engineering projects in isolated communities with a simultaneous focus on community participation and social integration [106]. 7. Measuring ‘‘Community’’ Of the seven investigated local communities in the first series of surveys (T1), this community had the highest mean (or score) on nervous symptoms, the second lowest on well-being, and the lowest perceived social/support. At T2, all these three indicators positively increased and got close to the mean Lofoten scores. Compared to the other investigated commu- nities in Lofoten at T1, the inhabitants also placed themselves low on most of the indexes of community integration. This indicated a low level of sociocultural integration. From T1 to T2, we observed an increase on the majority of the indexes. Although the increase was often not significant when we looked at the indexes one by one, but the tendency seemed clear. At T2 this community had moved closer to the level of the other local communities in Lofoten, hence, a development towards higher integration. k h h h h was a very collective leadership group. Many of the original project leader group meets with us at follow up meeting years later. But in the process after the formal end of the project, other people were also taking part as leaders in different local organizations. An important aspect of the social reality was that this second community was not dependent on one person to make the project work; that is, there was reduced dependency on any single person. In the second community, leadership leaned heavily on one person and some few in her network. When this group had personal difficulties or moved, other did not take over and brought the activities further. In sum, it points to the need to know the characteristics of the community at the start of the project and emphasizes the need to involve the community as a whole in the project as quickly as possible in the planning and decision making. higher sociocultural integration and a concomitant aug- mentation in psychological health [101]. In addition to the six communities that participated in the local community development projects at T1, one local community approached the project leaders on its own. They asked for help to con- struct a community profile of its problems and assets. They felt their local community was in danger, because they could lose their school and other institutions that they felt was part of a living community. 7. Measuring ‘‘Community’’ Communities can be studied by focusing on people’s emo- tional sentiments towards their community and/or by using a more rational community evaluation [91, 92]. Leighton’s anthropological approach of measuring communities’ degree of integrated focused on (a) emotional sentiments towards the community, that is, social cohesion, and (b) more rational evaluations of risk and resource factors. A similar approach was used in the Lofoten studies. Many approaches to the com- munity look at neighborhoods as communities measured by the number and quality of relations to neighbors [93, 94]. The most systematic behavioral approach has been the study of behavioral settings, places where people in a community meet and manifest social interaction as suggested by Barker’s work In Lofoten, following community integration efforts, we also had a community showing a parallel process towards 7 Psychiatry Journal higher sociocultural integration and a concomitant aug- mentation in psychological health [101]. In addition to the six communities that participated in the local community development projects at T1, one local community approached the project leaders on its own. They asked for help to con- struct a community profile of its problems and assets. They felt their local community was in danger, because they could lose their school and other institutions that they felt was part of a living community. During recent years, the popu- lation had declined and many services had moved to the more central parts of the municipality. In sum (at T1), we could observe a process of societal disorganization. In the inter- vening time, T1-T2, this local community was exposed to two opposing fields of forces. Resources like school, shops, and other services were closed down or moved to more central parts of the municipality. On the other hand, through actions to save and further develop a home for the elderly, many of the inhabitants came together and through cooperation succeeded in their “struggle.” People from this community participated in courses about care for the elderly, including training to create organisations and for developing leadership skills. The home for the elderly also developed into a commu- nity centre, bringing together people for voluntary communal work, including an organisation to attend to the needs of the residents of the institution, as well as provide recreational activities for the general population. The development was compared to six other local communities in Lofoten. 7. Measuring ‘‘Community’’ Many of these would be consistent with the experience in Nova Scotia and with our own projects in Lofoten, provid- ing support to community groups, helping to provide solu- tions to problems, giving local people a voice and a role in pol- icy processes, establishing and/or supporting local activities and local organization, developing and supporting networks between organizations in the area, securing harmonious social relations, promoting community interests, encourag- ing trust in residents in their relationships with each other, bringing conflicting groups together, encouraging a sense of safety in residents, and boosting the identity of a place. In itself, insufficient sense of community is seen as a barrier Psychiatry Journal 8 8 on mental health vary along a “gemeinschaft-gesellschaft” dimension, that is, would community integration have a larger impact on people’s well-being and psychiatric status if it is nearer to the “gemeinschaft” end of such a dimension? Our preliminary answers to these questions are in the affirmative. Our principles for increasing the integration of a local community deriving mainly from Leighton’s studies were the guidelines for the health promotion activities and the nature of the research. The dynamic pattern of the “Liv Laga Model for Mental Health Promotion” is supported by an urban study [129], showing that individual factors like perceived stress, depressive moods, and situational determinants like undesirable life events were related to a decrease in social integration in the community. Interpersonal determinants such as positive emotional guidance and instrumental sup- port were associated with increases in social integration in the community. Hence the Lofoten project of combining activities that reduce stressful life burdens with enhancement of social interaction should be a comprehensive way of doing mental health promotion on the community level. to community development [52, 97, 118–120]. In a study from Northern Ireland [121], the successful intervention in rural communities, much like the experience in Lofoten, was characterized by “a partnership model of working; local coordinating structures and consultation mechanisms; use of structured planning model to guide program planning and implementation; mobilization of cross-community and inter-agency support; and a comprehensive logic evaluation framework to assess the input, process, impact and outcomes of the project as it unfold”.i A review of the literature [35] identified at least four major dimensions that stimulate the creation of community: (a) meeting opportunities, (b) individual motivation to invest in others in the group, (c) alternatives to realize individual goals, and (d) interdependencies. 7. Measuring ‘‘Community’’ Important for the creation of community was interdependence among neighbours, the intention to stay in the neighbourhood (e.g., people with young children), more facilities and facilities related to the market (e.g., shops), homogeneity with respect to income, and ownership of one’s home. On the other hand, having more relational alternatives outside of the neighbourhood can work against the creation of community. y In the Stirling County studies, Leighton evaluated the effects of sociocultural integration on mental health. His theoretical approach was the basis of a mental health project that transformed a small community, “The Road,” from a poorly integrated community into a well-integrated one. The qualitative measurement methods used by these anthropol- ogists working with Leighton may lack precision which may explain the hesitation to use them in other studies of mental health promotion projects. The strength of the data gathering procedures used in the Lofoten studies is the inclusion of subjective and collective evaluations about the community, that is, giving questionnaires constructed for self-completion to the total adult population. This gives collective rational community evaluations of the problems, risks, and resources. The use of self-completing questionnaires in a community is a feasible way of overcoming the problems of sampling methods. By delivering questionnaires to all adult inhabitants in a local community it was possible to get closer to the uni- verse of respondents in each community. Repeated surveys over time in the Lofoten project provided feedback on the effects of concrete actions initiated in promotion projects. The quantification of the community indexes allowed us to measure the degree of change in community promotion projects. The dimensional properties of the questionnaire open up for more specific actions and evaluations. g y In the Lofoten studies, Lewin’s concepts of life space and social space [122], even if not explicitly formulated at the planning stage, have been important for our understanding and study of the “psychology of local communities.” “Liv Laga i Lofoten” has been an action research project, having seeker conferences, quality circles, and the building of cooperative networks in its toolbox. Hence, the practical implementation of the projects in Lofoten is not only in debt to Leighton, but also to Lewin and his students. Lewin and his students laid the early foundation for understanding change in social situations [123–125]. 7. Measuring ‘‘Community’’ Lewin’s model of planned change [126– 128]; has added a useful analytic tool for the further devel- opment of the Lofoten mental health promotion approach. Furthermore, Lewin believed that organizations, with their associated patterns of attitudes, expectations, and behavioral forms, behave much like other living biological systems and are influenced by the concept of homeostasis. This has the same connotations as Leighton’s descriptions of the inte- grated/disintegrated community as a quasiorganism. Lewin’s planned change could be at template for local community mental health promotion. This approach brings together four integrated elements: field theory which seeks to map the totality of human behavior that is taking place; group dynamics which seeks to understand the behavior of groups; action research which requires analyzing the situation and choosing the best change for the situation; a stepwise change model (unfreezing present stagnant or negative community behavior; learning and effectuating new behavior; stabilizing the new equilibrium). i In the studies in Lofoten, it became clear that evaluations of the community projects depended on several types of data and analyses. We had to analyze the change over time in the importance of each dimension of community integration, the changes in each of the outcome measures, and the changes in community consensus. It was also necessary to obtain notes and records from group meetings and evaluative comments from the formal and informal leaders in the community on the perceived effects of the strategy in their community. Success (or failure) of a community project could show itself in a myriad of different ways. References [1] A. H. Leighton, My Name Is Legion, Basic Books, New York, NY, USA, 1959. [2] A. B. Hollingshead and F. C. Redlich, Social Class & Mental Illness, John Wiley & Sons, New York, NY, USA, 1958. [3] L. Srole, T. S. Langner, S. T. Michael, M. K. Opler, and T. Rennie, “Mental health in the metropolis: the midtown Manhattan study,” in Rennies Series in Social Psychiatry, A. C. Thomas, Ed., McGraw-Hill, New York, NY, USA, 1962. [4] A. H. Leighton, “Research directions in psychiatric epidemiol- ogy,” Psychological Medicine, vol. 9, no. 2, pp. 235–247, 1979. [5] M. D. Teehan, “Interview. In conversation with Alexander H. Leighton,” Psychiatric Bulletin, vol. 17, pp. 449–457, 1993. [6] O. S. Dalgard, Bomiljø og psykisk helse, Universitetsforlaget, Oslo, Norway, 1980. [7] O. S. Dalgard, S. Bjork, and K. Tambs, “Social support, negative life events and mental health—a longitudinal study,” The British Journal of Psychiatry, vol. 166, pp. 29–34, 1995. y In recent years, similar studies have begun to follow the strategy described here and have used some of our data gathering methods. A study of urban regeneration and mental health in Manchester [131, 132] used some of the most potent Lofoten items. They found that higher overall quality of life ratings were associated with greater sense of belonging to the community, less feelings of isolation, better leadership, and more leisure opportunities. Lower subjective quality of life was associated with the feeling that the area was in decline. Several questions measuring community integration and connectedness from the Lofoten studies were also used in a Norwegian study called HUNT2, conducted in the county of Nord-Trøndelag. Their index of community connectedness was also shown to be related to subjective well-being and self- rated health [133].h [8] T. Sørensen, “Commuting, community disintegration and psy- chiatric disorder,” in Psychiatric Disorders in the Social Domain, T. Sørensen, P. Abrahamsen, and S. Torgersen, Eds., Norwegian University Press, Oslo, Norway, 1991. [9] A. H. Leighton, “Caring for mentally ill people,” in Psychological and Social Barriers in Historical Context, Cambridge University Press, Cambridge, UK, 1982. [10] A. H. Leighton, “Poverty and social change,” Scientific America, vol. 215, pp. 3–9, 1965. [11] A. R. Radcliffe-Brown, The Andamant Islanders, The Free Press, Glencoe, Ill, USA, 1948. [12] B. Malinowski, Argonauts of the Western Pacific, Dutton, New York, NY, USA, 1953. [13] R. 9. Some Concluding Reflections and Future Directions What have we learned from the mental health promotion project in Lofoten? Is this strategy a feasible way to promote mental health? What types of local communities can make the most of this method? Would the impact of local communities f In 2010, following the Lofoten project, 12 local commu- nities in Valdres in southern Norway were selected for study. Valdres consists of six municipalities in a rural mountainous Psychiatry Journal 9 region. In each of these municipalities, two communities were selected by local authorities. In each municipality those communities were chosen with regard to having contrasting resources and risk characteristics [130]. After the surveys, meetings with each of the six municipal councils were held. They focused on how the population scored on the measures of mental health and on the indexes of community integration which differed among the selected communities. In Valdres, they compared them to the communities in Lofoten. Partic- ular attention was paid to the meaning of different patterns of means and correlations in the communities, combinations that suggested possible targets for community interventions. The discussion with the members of local councils supported the use of means and correlations as a way of defining health promotion activities. Comparing Lofoten to Valdres also allowed us to determine the effects of different geographical characteristics on social and psychological processes. The comparison between the Lofoten programs with the planning for Valdres communities opened up new opportunities for translating theory and methods into more sophisticated and successful uses of community resources. very concrete link between theory and practical actions. This theory-action linkage is even more developed and directed at specific domains in the Lofoten project. References Lewis, “Environmental factors and sense of community in a planned town,” The American Journal of Com- munity Psychology, vol. 24, no. 1, pp. 109–143, 1996. [31] R. J. Sampson, “The community,” in Crime, J. Q. Wilson and J. Petersilia, Eds., pp. 193–216, Institute for Contemporary Studies, San Francisco, Calif, USA, 1996. [51] S. J. Farrell, T. Aubry, and D. Coulombe, “Neighborhoods and neighbors: do they contribute to personal well-being?” Journal of Community Psychology, vol. 32, no. 1, pp. 9–25, 2004. [32] I. Kawachi and L. Berkman, “Social cohesion, social capital, and health,” in Social Epidemiology, I. Kawachi and L. Berkman, Eds., Oxford University Press, New York, NY, USA, 2000. [52] D. M. Chavis and A. Wandersman, “Sense of community in the urban environment: a catalyst for participation and community development,” The American Journal of Community Psychology, vol. 18, no. 1, pp. 55–81, 1990. [33] J. R. Gusfield, The Community: A Critical Response, Harper Col- ophon, New York, NY, USA, 1975. [53] M. P. Prezza, M. Amici, T. Roberti, and G. Tedeschi, “Sense of community referred to the whole town: its relations with loneliness, life satisfaction and area of residence,” Journal of Community Psychology, vol. 29, no. 1, pp. 29–52, 2001. [34] S. Riger and P. J. Lavrakas, “Community ties: patterns of attach- ment and social interaction in urban neighborhoods,” The American Journal of Community Psychology, vol. 9, no. 1, pp. 55– 66, 1981. [54] M. L. Martinez, M. Black, and R. H. Starr, “Factorial structure of the perceived neighborhood scale (PNS): a test of longitudinal invariance,” Journal of Community Psychology, vol. 30, no. 1, pp. 23–43, 2002. [35] B. V¨olker, H. Flap, and S. Lindenberg, “When are neighbour- hoods communities? Community in Dutch neighbourhoods,” European Sociological Review, vol. 23, no. 1, pp. 99–114, 2007. [55] C. C. Sonn, “Immigrant adaptation. Understanding the process through sense of community,” in Psychological Sense of Commu- nity: Research, Applications and Implications, A. T. Fisher, C. C. Sonn, and B. J. Bishop, Eds., pp. 205–222, Plenum Press, New York, NY, USA, 2002. [36] R. D. Putnam, “The prosperous community—social capital and public life,” The American Prospects, vol. 13, pp. 35–42, 1993. [37] P. Dekker and E. M. Uslander, Eds., Social Capital and Partici- pation in Everyday Life, Routledge/ECPR, London, UK, 2001.h [38] J. S. Coleman, Foundation of Social Theory, Harvard University Press, Cambridge, Mass, USA, 1990. [56] D. W. McMillan and D. M. References Durkheim, Suicide: A Study in Sociology, The Free Press, New York, NY, USA, 1887–1997. [44] V. Cattell, “Poor people, poor places, and poor health: the med- iating role of social networks and social capital,” Social Science and Medicine, vol. 52, no. 10, pp. 1501–1516, 2001. [23] J. Harris, Ed., Ferdinan Tonnies. Community and Civic Society, Cambridge University Press, Cambridge, UK, 2001. [45] A. M. Almedom, “Social capital and mental health: an inter- disciplinary review of primary evidence,” Social Science and Medicine, vol. 61, no. 5, pp. 943–964, 2005. [24] E. W. Burges, Ed., The Urban Community, University of Chicago Press, Chicago, Ill, USA, 1926. [46] M. J. de Silva, K. McKenzie, T. Harpham, and S. R. A. Huttly, “Social capital and mental illness: a systematic review,” Journal of Epidemiology and Community Health, vol. 59, no. 8, pp. 619– 627, 2005. [25] R. E. L. Faris and H. W. Dunham, Mental Disorders in Urban Areas, The University of Chicago Press, Chicago, Ill, USA, 1934. [26] S. Parker and R. Kleiner, Mental Illness in the Urban Negro Com- munity, The Glencoe Press, New York, NY, USA, 1966. [47] S. B. Sarason, “The psychological sense of community,” in Pro- spects for a Community Psychology, Jossey-Bass, San Francisco, Calif, USA, 1974. [27] B. Weinman and R. J. Kleiner, “The impact of community living and community member intervention on the adjustment of the chronic psychotic patient,” in Alternatives to Mental Hospital Treatment, L. I. Stein and M. A. Test, Eds., Plenum Press, New York, NY, USA, 1978. [48] D. G. Unger and A. Wandersman, “The importance of neigh- bors: the social, cognitive, and affective components of neigh- boring,” The American Journal of Community Psychology, vol. 13, no. 2, pp. 139–169, 1985. [28] J. Dean and A. Hastings, Challenging Images: Housing Estates, Stigma and Regeneration, Policy Press, Bristol, UK, 2000. [49] D. M. Chavis and G. M. Pretty, “A review of the sense of com- munity index: current uses, factor structure, reliability, and fur- ther development,” Journal of Community Psychology, vol. 27, no. 6, pp. 635–642, 1999. [29] P. Huxley, S. Evans, M. Leese et al., “Urban regeneration and mental health,” Social Psychiatry and Psychiatric Epidemiology, vol. 39, no. 4, pp. 280–285, 2004. [30] R. Sampson and W. B. Groves, “Community structure and crime: testing social-disorganisation theory,” The American Journal of Sociology, vol. 94, no. 4, pp. 774–802, 1989. [50] J. M. Plas and S. E. References Merton, “Manifest and latent functions, toward the codifi- cation of functional analysis in sociology,” in Social Theory and Social Structure, pp. 21–81, The Free Press, Glencoe, Ill, USA, 1949. The Norwegian studies have shown that the concept and the measuring of sociocultural integration are very valuable in understanding how the properties of a community can affect people’s mental health and their relation to other social psychological concepts. It has shown its value in the planning of mental health services and has demonstrated success in concrete mental health promotion projects. The concepts and the methods used in the Lofoten studies could make impor- tant contributions to further studies and actions in local communities where the crossroad between the individual, their social network, and their local community occurs, that is, an extension of Dalgard’s concept “community mental health profile” [134]. [14] H. H. Hyman and E. Singer, Readings in Reference Group Theory and Research, The Free Press, New York, NY, USA, 1968.hh [15] T. Parson, The Social System, The Free Press, Glencoe, Ill, USA, 1951. [16] J. R. Williams, American Society, a Sociological Interpretation, Alfred A. Knopf, New York, NY, USA, 1951. [17] J. F. Cuber, Sociology, a Synopsis of Principles, Appelton- Century-Crofts, New York, NY, USA, 1951. [18] H. Blumer, “Social disintegration and individual disorganiza- tion,” The American Journal of Sociology, vol. 42, no. 6, pp. 871– 877, 1937. i A main strength of the sociocultural integration-disinte- gration concept is the possibility to integrate an emphasis on risk and resource factors of the community’s objective reality with a more long lasting experience of a community that has or has not sufficient structural capacity to support and maintain people’s need for social support. Leighton’s theoret- ical proposition of looking at a community as an entity that is more than the sum of its individual members strengthen a mental health promotion strategy that encompasses more than detection of individuals at risk. The strength is also the [19] T. Sørensen, R. J. Kleiner, N. Bøe, T. Moum, and I. Sandanger, “Sociocultural integration and disintegration,” in The Local- Community Approach to Mental Health: NordkySt PSykiatri 2000. 31 S, vol. 1 of North-Coast Series in Community Psychiatry, 2000. [20] R. Bastide, The Sociology of Mental Disorder, Routledge & Kegan Paul, London, UK, 1972. [21] E. Durkheim, The Division of Labor in Society, The Free Press, New York, NY, USA, 1997. Psychiatry Journal 10 [22] E. References Antonovsky, “The structure and properties of the sense of coherence scale,” Social Science and Medicine, vol. 36, no. 6, pp. 725–733, 1993. [62] W. T. Grunkemeyer and M. L. Moss, “The sustainable com- munity model approach to the development and use of multi- dimensional quality of life indicators,” in Community Quality of Life Indicators: Best Cases, M. J. Sirgy, D. Rath, and D. J. Lee, Eds., pp. 29–52, Kluwer Academic, Dorecht, The Netherlands, 2004. [78] A. Antonovsky, “The structural sources of salutogenetic strengths,” in Personality and Stress: Individual Differences in the Stress Process, L. Cooper and R. Payne, Eds., John Wiley & Sons, New York, NY, USA, 1991. [63] J. Whirlock, “The role of adults, public space, and power in adolescent community connectedness,” Journal of Community Psychology, vol. 35, no. 4, pp. 499–518, 2007. [79] J. Eckenrode, “The mobilization of social supports: some indi- vidual constraints,” The American Journal of Community Psy- chology, vol. 11, no. 5, pp. 509–528, 1983. [64] C. Yuksel and C. Turner, “Building community connectedness in Broadmeadows,” The Australian Community Psychologist, vol. 20, pp. 8–21, 2008. [80] I. N. Sandler and B. Lakey, “Locus of control as a stress mod- erator: the role of control perceptions and social support,” The American Journal of Community Psychology, vol. 10, no. 1, pp. 65–80, 1982. [65] T. Sørensen, “The feeling and anticipation of social support, life stress and their impact on mental health,” in Psychiatry at the Cross-Roads Between Social Science and Biology, S. Torgersen, P. Abrahamsen, and T. Sørensen, Eds., Norwegian University Press, Oslo, Norway, 1991. [81] H. Antonovsky and S. Sagy, “The development of a sense of coherence and its impact on responses to stress situations,” The Journal of social psychology, vol. 126, no. 2, pp. 213–225, 1986. [82] D. D. Perkins and M. A. Zimmerman, “Empowerment theory, research, and application,” The American Journal of Community Psychology, vol. 23, no. 5, pp. 569–579, 1995. [66] A. Antonovsky, Health, Stress, and Coping: New Perspectives on Mental and Physical Well-Being, Jossey-Bass, San Francisco, Calif, USA, 1979. [83] G. Laverack and N. Wallerstein, “Measuring community em- powerment: a fresh look at organizational domains,” Health Promotion International, vol. 16, no. 2, pp. 179–185, 2001. [67] O. M. Klepp, A. Mastekaasa, T. Sørensen, I. Sandanger, and R. References Kleiner, “Structure analysis of Antonovsky’s sense of coherence from an epidemiological mental health survey with a brief nine- item sense of coherence scale,” International Journal of Methods in Psychiatric Research, vol. 16, no. 1, pp. 11–22, 2007. [84] D. A. Torre, Empowerment: structured conceptualization and instrument development [Ph.D. thesis], Cornell University, Itacha, NY, USA, 1986. [68] J. Rappaport, “In praise of paradox: a social policy of empow- erment over prevention,” The American Journal of Community Psychology, vol. 9, no. 1, pp. 1–25, 1981. [85] M. A. Zimmerman and J. Rappaport, “Citizen participation, perceived control, and psychological empowerment,” The Amer- ican Journal of Community Psychology, vol. 16, no. 5, pp. 725– 750, 1988. [69] R. Schwartzer and M. Jerusalem, “Generalized Self-Efficacy scale,” in Measures in Health Psychology: A Users Portfolio. Causal and Control Beliefs, J. Weinman, S. Wright, and M. Johnston, Eds., pp. 35–37, NFER-Nelson, Windsor, UK, 1995. [86] N. Kari and P. Michels, “The Lazarus project: the politics of empowerment,” The American Journal of Occupational Therapy, vol. 45, no. 8, pp. 719–725, 1991. [70] L. I. Pearlin, M. A. Lieberman, E. G. Menaghan, and J. T. Mullan, “The stress process,” Journal of Health and Social Behavior, vol. 22, no. 4, pp. 337–356, 1981. [87] N. Wallerstein, “Powerlessness, empowerment, and health: implications for health promotion programs,” The American Journal of Health Promotion, vol. 6, no. 3, pp. 197–205, 1992. [71] C. L. M. Keyes and A. D. Shapiro, “Social well-being in the United States: a descriptive epidemology,” in How Healthy Are We? A National Study of Well-Being at Midlife, O. G. Brim, C. D. Ryff, and R. C. Kessler, Eds., pp. 350–372, University of Chicago Press, Chicago, Ill, USA, 2004. [88] E. S. Rogers, J. Chamberlin, M. L. Ellison, and T. Crean, “A con- sumer-constructed scale to measure empowerment among users of mental health services,” Psychiatric Services, vol. 48, no. 8, pp. 1042–1047, 1997. [89] S. Ramon, B. Healy, and N. Renouf, “Recovery from mental illness as an emergent concept and practice in Australia and the UK,” International Journal of Social Psychiatry, vol. 53, no. 2, pp. 108–122, 2007. [72] S. Lindenberg, “Continuities in the theory of social production function,” in Verklarende Sociologie: Opstellen voor Reihard Wippler, H. Ganzeboom and S. Lindenberg, Eds., pp. 169–184, Thesis, Amsterdam, The Netherlands, 1996. [90] O. M. Klepp, T. Sørensen, and R. References Chavis, “Sense of community: a definition and theory,” Journal of Community Psychology, vol. 14, no. 1, pp. 6–23, 1986. [39] R. D. Putnam, Making Democracy Work: Civic Tradition in Modern Italy, Princeton University Press, Princeton, NJ, USA, 1993. [57] P. Obst, S. G. Smith, and L. Zinkiewicz, “An exploration of sense of community, part 3: dimensions and predictors of psycholog- ical sense of community in geographical communities,” Journal of Community Psychology, vol. 30, no. 1, pp. 119–133, 2002. [40] J. Macinko and B. Starfield, “The utility of social capital in research on health determinants,” Milbank Quarterly, vol. 79, no. 3, pp. 387–427, 2001. [58] S. M. Low and I. Altman, “Place attachment: a conceptual inquiry,” in Place Attachment, I. Altman and S. M. Low, Eds., Plenum Press, New York, NY, USA edition, 1992. [41] T. Harpham, E. Grant, and E. Thomas, “Measuring social capital within health surveys: key issues,” Health Policy and Planning, vol. 17, no. 1, pp. 106–111, 2002. [59] L. C. Manzo and D. D. Perkins, “Finding common ground: the importance of place attachment to community participation and planning,” Journal of Planning Literature, vol. 20, no. 4, pp. 335–350, 2006. [42] R. Forrest and A. Kearns, “Social cohesion, social capital and the neighbourhood,” Urban Studies, vol. 38, no. 12, pp. 2125–2143, 2001. [60] G. H. Pretty, H. M. Chipuer, and P. Bramston, “Sense of place amongst adolescents and adults in two rural Australian towns: the discriminating features of place attachment, sense of [43] J. S. Coleman, “Social capital in the creation of human capital,” The American Journal of Sociology Supplement, vol. 84, pp. S95– S120, 1988. Psychiatry Journal 11 helpful in motivating cardiac risk factor modification,” Heart and Lung, vol. 34, no. 1, pp. 22–29, 2005. community and place dependence in relation to place identity,” Journal of Environmental Psychology, vol. 23, no. 3, pp. 273–287, 2003. [76] [76] T. Sørensen, O. Klungsøyr, R. Kleiner, and O. M. Klepp, “Social support and sense of coherence: independent, shared and interaction relationships with life stress and mental health,” International Journal of Mental Health Promotion, vol. 13, no. 1, pp. 27–42, 2011. [61] L. Dallago, D. D. Perkins, M. Santinello, W. Boyce, M. Molcho, and A. Morgan, “Adolescent place attachment, social capital, and perceived safety: a comparison of 13 countries,” The Amer- ican Journal of Community Psychology, vol. 44, no. 1-2, pp. 148– 160, 2009. [77] A. References 16, no. 6, pp. 771–791, 1988.h [114] D. Fone, F. Dunstan, K. Lloyd, G. Williams, J. Watkins, and S. Palmer, “Does social cohesion modify the association between area income deprivation and mental health? A multilevel analysis,” International Journal of Epidemiology, vol. 36, no. 2, pp. 338–345, 2007. [99] D. Wilkinson, “The multidimensional nature of social cohesion: psychological sense of community, attraction, and neighbor- ing,” The American Journal of Community Psychology, vol. 40, no. 3-4, pp. 214–229, 2007. [115] S. R. Zubrick, “Commentary: area social cohesion, deprivation and mental health—does misery love company?” International Journal of Epidemiology, vol. 36, no. 2, pp. 345–347, 2007. [100] J. Raeburn, “Community approaches to mental health promo- tion,” International Journal of Mental Health Promotion, vol. 3, pp. 13–19, 2001. [116] T. Fujiwara and I. Kawachi, “A prospective study of individual- level social capital and major depression in the United States,” Journal of Epidemiology and Community Health, vol. 62, no. 7, pp. 627–633, 2008. [101] T. Sørensen, A. Mastekaasa, R. Kleiner, I. Sandanger, N. Bøe, and O. M. Klepp, “Local community mobilization and mental health promotion,” International Journal of Mental Health Pro- motion, vol. 6, pp. 5–16, 2004. [117] B. Burnes, “Kurt Lewin and the planned approach to change: a re-appraisal,” Journal of Management Studies, vol. 41, no. 6, pp. 977–1002, 2004. [102] T. Sørensen, N. Bøe, R. Kleiner, A. Sørensen, and P. Ngo, “Socio- cultural properties of communities and psychological health,” in Proceedings of the 41st Annual Meeting of SCCR, Las Vegas, Nev, USA, February 2012. [118] J. Carter, Sense of Place: An Interpretive Planning Handbook, Tourism & Environment Initiative, Inverness, UK, 1997. [119] J. M. Bolland and D. M. McCallum, “Neighboring and commu- nity mobilization in high-poverty inner-city neighborhoods,” Urban Affairs Review, vol. 38, no. 1, pp. 42–69, 2002. [103] O. S. Dalgard and J. N. Brevik, “Community health profile with special emphasis on social environment and mental health,” Nordic Journal of Psychiatry, vol. 50, no. 37, pp. 21–27, 1996. [120] B. P. Conway and D. S. Hachen Jr., “Attachments, grievances, resources, and efficacy: the determinants of tenant association participation among public housing tenants,” Journal of Urban Affairs, vol. 27, no. 1, pp. 25–52, 2005. [104] T. Sorensen, N. Boe, G. Ingebrigtsen, and I. Sandanger, “Indi- vidual-local community and mental health. Towards a compre- hensive community psychiatric model for prevention of mental problems and promotion of mental health,” Nordic Journal of Psychiatry, vol. References Kleiner, “Empowerment: addi- tive, overlapping and interactive relation to sense of coherence, with regard to mental health and its promotion,” International Journal of Mental Health Promotion, vol. 9, no. 3, pp. 5–26, 2007. [73] T. Sørensen, A. Mastekaasa, I. Sandanger et al., “Contribution of local community integration and personal social network support to mental health,” Norsk Epidemiologi, vol. 12, pp. 269– 274, 2002. [91] A. M. Guest and B. A. Lee, “Sentiment and evaluation as eco- logical variables,” Sociological Perspectives, vol. 26, no. 2, pp. 159– 184, 1983. [74] P. Hawe and A. Shiell, “Social capital and health promotion: a review,” Social Science and Medicine, vol. 51, no. 6, pp. 871–885, 2000. [92] R. E. Adams, “A tale of two cities: community sentiments and community evaluation in Indianapolis and Pittsburgh,” Socio- logical Focus, vol. 25, no. 3, pp. 217–240, 1992. [75] C. Boutin-Foster, “Getting to the heart of social support: a quali- tative analysis of the types of instrumental support that are most Psychiatry Journal 12 [93] G. Hillery, “Definitions of community: areas of agreement,” Rural Sociology, vol. 20, pp. 111–123, 1955. psychosocial perspective,” International Journal of Technology Assessment in Health Care, vol. 12, no. 4, pp. 604–617, 1996. [94] D. G. Unger and A. Wandersman, “Neighboring in an urban environment,” The American Journal of Community Psychology, vol. 10, no. 5, pp. 493–509, 1982. [111] C. A. Latkin and A. D. Curry, “Stressful neighborhoods and depression: a prospective study of the impact of neighborhood disorder,” Journal of Health and Social Behavior, vol. 44, no. 1, pp. 34–44, 2003. [95] R. G. Barker, Ecological Psychology, Stanford University Press, Stanford, Calif, USA, 1968. [112] E. B. Loucks, L. F. Berkman, T. L. Gruenewald, and T. E. Seeman, “Social integration is associated with fibrinogen concentration in elderly men,” Psychosomatic Medicine, vol. 67, no. 3, pp. 353– 358, 2005. [96] R. G. Barker and P. Schoggen, Qualities of Community Life, Jossey-Bass, San Francisco, Calif, USA, 1973. [97] W. B. Davison and P. R. Cotter, “Measurement of sense of com- munity within the sphere of city,” Journal of Applied Social Psy- chology, vol. 16, no. 7, pp. 608–619, 1986. [113] D. L. Fone, D. M. Farewell, and F. D. Dunstan, “An ecometric analysis of neighbourhood cohesion,” Population Health Met- rics, vol. 4, article 17, 2006. [98] J. C. Buckner, “The development of an instrument to measure neighborhood cohesion,” The American Journal of Community Psychology, vol. References 50, supplement 37, pp. 11–19, 1996. [121] M. M. Barry, “Researching the implementation of community mental health promotion programs,” Health Promotion Journal of Australia, vol. 18, no. 3, pp. 240–246, 2007. [105] J. C. Kern, “Sociocultural aspects of poverty: springboard for action,” Journal of Community Psychology, vol. 2, no. 1, pp. 5–10, 1974. [122] D. Cartwright, Ed., Field Theory in Social Science, Harper, New York, NY, USA, 1951. [106] G. H. Cabrera, “Propuestra para un programa de desarollo humano en campamentos de proyectos de ingeniera,” Revista Interamericana de Psicologia Ocupaciona, vol. 6, pp. 26–35, 1987. [123] R. Lippitt, J. Watson, and B. Wesley, The Dynamics of Planned Change, Hartcort, Brace & Company, New York, NY, USA, 1958.hh [124] A. J. Marrow, The Practical theorist: The Life and Work of Kurt Lewin, Basic Books, New York, NY, USA, 1969. [107] O. S. Dalgard, “Social epidemiology and mental health promo- tion,” in Social Psychiatry in Changing Times, E. Dragomirecka, A. Palcova, and H. Papezova, Eds., Praha Psychiatric Centrum, 2000. [125] E. H. Schein, Organizational Culture and Leadership, Jossey- Bass, San Francisco, Calif, USA, 1992. [126] K. Lewin, “Frontiers in group dynamics,” Human Relations, vol. 1, pp. 5–41, 1947. [108] O. S. Dalgard, “Psykisk helse, lokalsamfunn og forebyggende arbeid. Tre eksempler. I,” in Ubevisst sjeleliv og bevisst sam- funnsliv. Psykisk helse i en sammenheng, I. Sandanger, G. Ingebrigtsen, J. F. Nyg˚ard, and K. Sørgaard, Eds., Festskrift til Tom Sørensen p˚a hans 60-˚arsdag, Nordkyst Psykiatri, Nittedal, Norway, 2004. [127] D. Burns, R. Forrest, A. Kearns, and J. Flint, The Impact of Housing Associations on Social Capital: Interim to Scottish Homes, Department of Urban Studies, University of Glasgow, 2000. [109] O. S. Dalgard, T. Anstorp, K. Benum, T. Sørensen, and T. Moum, “Social psychiatric field studies in Oslo: some preliminary results,” in Advances in Field Theory, S. Wheelan, V. Abt, and E. Pepitone, Eds., pp. 230–243, Sage, Newbury Park, Calif, USA, 1990. [128] B. C. Medly and O. H. Akan, “Creating positive change in com- munity organizations. A case for rediscovering Lewin,” Non- profit Management & Leadership, vol. 18, pp. 485–496, 2008. [129] E. Gracia and J. Herrero, “Determinants of social integration in the community: an exploratory analysis of personal, inter- personal and situational variables,” Journal of Community and Applied Social Psychology, vol. 14, no. 1, pp. 1–15, 2004. [110] O. S. Dalgard, T. Sørensen, I. Sandanger, and J. I. References Brevik, “Psy- chiatric interventions for prevention of mental disorders: a 13 13 Psychiatry Journal [130] A. P. Sørensen, T. Sørensen, and B. Øygard, “Lokalsamfunn og mental helse i Valdres,” Rapport 1, Valdres Viser Veg. Sør-Aurdal kommune, 2012. [131] P. Huxley and A. Rogers, “Urban regeneration and mental health,” Health Variation, vol. 7, pp. 8–9, 2001. [132] A. Rogers, P. Huxley, R. Thomas et al., “Evaluating the impact of a locality based social policy intervention on mental health: conceptual and methodological issues,” International Journal of Social Psychiatry, vol. 47, no. 4, pp. 41–55, 2001. [133] Heimburg and D. Von, Health determinants of the Norwegian HUNT study. Psychosocial resistance resources: a salutogenic approach. Paper II: public health and health promotion: a salu- togenic approach [M.S. thesis in Health Science], Department of Social Work and Health Science, NTNU, Trondheim, Norway, 2010. [134] O. S. Dalgard, “Community health profile: a tool for psychiatric prevention,” in Promotion of Mental Health, D. R. Trent and C. Reeds, Eds., vol. 5, Avebury, Aldershot, UK, 1996.
https://openalex.org/W2922308072	https://journals.vgtu.lt/index.php/JCEM/article/download/9478/8272	English	null	THE INTERACTION OF SOUND FIELDS SEPARATED BY THE RESONANT SUSPENDED CEILING/GARSO LAUKŲ, ATSKIRTŲ REZONANSINĖMIS KABAMOSIOMIS LUBOMIS, SĄVEIKA	Journal of civil engineering and management	1,997	cc-by	4,913	Statyba ISSN: 1392-1525 (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tcem19 V. Stauskis V. Stauskis 1. Introduction premises. Kh. Shchirzhecki [4[ investigated theoreti- cally the time-dependence of the sound field attenuat- ion in the acoustically-interrelated large bodies separated by a wide opening. H. Kuttmff [5] was in- terested in the same dependence of three premises linked by openings. However, no experimental results are presented in these works. The frequency-depen- dence of the interacting sound fields and the change in these fields depending on varied geometrical paramet- ers of the openings were not examined. Designing of multipurpose halls almost always involves installation of suspended ceiling. Such ceiling exerts considerable influence on the hall acoustics. It may consist of even planes or structural elements with varied geometrical parameters. Sound may be reflected from such ceiling planes to the listeners' seats directionally or diffusely. Such subjective indicator of nrusical sound as the sound clarity index largely depends on the suspended ceiling. When forming such ceiling, slits of various form and width may be left between the ceiling planes. Therefore it is necessary to estimate the effect of such slits on the acoustic indicators of the hall. The suspended ceiling having rectangular-shaped slits located crosswise in respect of the hall divides the hall into two bodies: the upper body above the suspended ceiling and the lower one under it. After impulse excitation of the sound field in the hall, i.e. in the lower body, the sound waves will reach the upper body through the slits, thus forming a sound field in the upper body as well. Part of the energy will return back to the hall, i.e. to the listeners, through the same slits; then the sound retlectcd from the lloor and other planes will return to the upper body again. The process will occur over and over. Thus the muffling sound field in the hall will be influenced by the muffling sound field above the suspended ceiling and vice versa. The attenuation of the sound field above the ceiling will have an impact on the acoustic indicators of the hall. The process is schematically depicted in Fig I. It has been established in [1] that the reverbera- tion time is markedly reduced by such ceiling at low frequencies. Depending on the width of the slits and the distance between the ceiling and the rigid surface, the absorption coefficients of the hall and the general hall absorption may be increased or reduced. To cite this article: V. Stauskis (1997) THE INTERACTION OF SOUND FIELDS SEPARATED BY THE RESONANT SUSPENDED CEILING, Statyba, 3:11, 90-95, DOI: 10.1080/13921525.1997.10531359 To cite this article: V. Stauskis (1997) THE INTERACTION OF SOUND FIELDS SEPARATED BY THE RESONANT SUSPENDED CEILING, Statyba, 3:11, 90-95, DOI: 10.1080/13921525.1997.10531359 Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=tcem20 ISSN 1392-1525. STATYBA- CIVIL ENGINEERING- CTPOV1TEJlbCTBO, 1997, Nr. 3(11) . . . . ---- . . . . ---- 1. Introduction With slits between the planes of the suspended ceiling, an in- teraction of the sound energies of two bodies of air, the one under and the one above the suspended ceiling, will take place. The purpose of this work is to establish both theoretically and experimentally the interaction be- tween the sound energies of the upper and the lower bodies of air in the hall with suspended ceiling and slits between the planes as well as to determine the effect produced on the energies by changing geometri- cal and acoustic parameters of the ceiling. A case will be examined when the regime is settled both in the lower body, marked as V, and the upper body, marked a-; V The sound source P is in the lower body V The interaction of both bodies will be con- sidered to occur through thl· opening, which is equal to the overall area of the slits between the ceiling planes. In ca'iC of an ideally diffusive sound Geld in both bodies, the steady-state energy in the lower body may be writ- ten as follows: 2. Theory A number of authors have investigated the theo- retical problems of the interaction of the sound fields of acoustically-interrelated premises, both in the steady state and the transitional regime. V Furduycv [2] and G. Goldberg 131 studied theoretically the changes in the reverberation time in acoustically-interrelated - 90- Fig 1. The distribution of the first sound reflections in the lower and upper bodies of the hall bodies, and the third member - the energies returning to respective bodies. The following markings will be introduced: (7) (7) Then the energy in the lower body will be equal Then the energy in the lower body will be equal Fig 1. The distribution of the first sound reflections in the lower and upper bodies of the hall to: (8) (8) and in the upper body: The energy in the upper body will be equal to: (2) (2) (9) (9) where Pa is the power of the source; c is the sound velocity in the air; a 1 , S1 are the absorption coef- ficients of the lower body and the area of surfaces; F is the overall area of all slits in the ceiling; X(l), X(2), Y(l) and Y(2) are the coefficients to be found; Q1 and Q2 are the coefficients of link between the two bodies. where 81 is the constant of the attenuation of the lower body, which is equal to (10) (10) The upper body attenuation constant 82 is equal The link coefficient for the lower body is equal to: (11) (3) (3) (11) The following markings will be introduced: The link coefficient for the upper body is equal The link coefficient for the upper body is equal (12) Q - F 2 - a S +F · 2 2 (4) (12) (4) When t = 0, i.e. when the sound source has ceased operating and the reverberation process in both bodies has a perfectly exponent character, the sound field attenuation in both bodies will depend on the sound attenuation constants in these bodies. 3. Results The interaction of the sound fields should, first of all, exert effect on the reverberation times of both txxlies. When conducting experiments with the hall model scaled 1:25 [2], all walls and the floor in the lower body were made of sound-reflecting materials and only the orchestra rise 119 m2 in area was of sound-absorb- ing materials. All planes of the upper volume were made of materials reflecting sound well, just as the ceiling itself. The results of the investigation'> are pre- sented in Fig 2. In all cases of investigation, a distinction may be made of two attenuation periods: the initial pe- riod with still unsettled attenuation process, and the final periods with a settled attenuation process. The speed of the energy attenuation is different and the sound fields muffle according to two different expo- nents. When there is no sound-absorbing material above the ceiling, the attenuation of the sound fields of both bodies has the same character in the period up to 1500 rns, while after 1500 rns the attenuation becomes slower in the lower body than in the upper body. The attenuation of energy in the lower body is influenced by the attenuation in the upper body. When there is a sound-absorbing material above the suspended ceiling, the attenuation character under- goes a sharp change. The fields of both bodies muffle to 500 ms only, and then the attenuation of the sound field of the upper body becomes considerably slower, than that of the lower body. It means that the interact- ion of the sound fields differs with time and depends on whether the attenuation process is settled or not. When the hall has no suspended ceiling, its reverberation time is long and it rapidly decreases with the increase in frequency. In the case under consideration, the ceiling was installed at the distance of 1 m from the rigid surface. Then the lower body makes up 8273 m3, while the volume of the upper body is 748 m3 only. In such case, the reverberation times of both bodies should differ greatly, since these times are largely dependent on the volume of the hall. Quite to the contrary, no such results were obtained during the investigations. Though the lower body is 11 times as large as the upper body, their reverberation times are almost equal all over the frequency range. 2. Theory Then the sound energy in the lower body may be expressed as follows: (13) (13) Then the changes of the above-mentioned ener- gies in the lower body may be calculated from the following formula: Y(1) = X(1)[1 + _}J__(l + ~)-~(1 + ~)] (14) Y 12 v2 Y 12 v2 (5) (5) (14) The sound energy in the upper body may be expressed as: The energy changes in the upper body will be calculated from the formula V d£2 ca2S1 cF cF 2--=----£2--£2 +-E dt 4 4 4 l· (6) Y(2) = X(2)[1 + _}J__(l + ~)- 02 (1 + ~)]. (15) '¥12 v2 'Yr2 V2 (6) The first member in these formulas describes the absorbing energy in the upper and lower bodies, the second member - the energies outgoing from respective -91 - Fig 3 shows the attenuation of the nonfiltered signal sound energies in both bodies of the hall. 3. Results At 100 Hz and 160 Hz, the reverberation time of the upper volume is even longer than the reverberation time of the large lower body. This shows that there is an interaction between the energies of both volumes. The same regularity is also observed when the suspended ceiling is at the distance of 4 m from the rigid surface. Fig 4 depicts the change in the relative sound energy returning to the lower body depending on the width of the opening. The graph indicates that the sound returning to the lower body becomes more energetic with the in- crease in the area of the opening and is little affected E, dB 100 r---.----r-----r----"T----- 90 80 70 60 ' ' I I I I --~----~-----~----~----~--- ' 1000 1500 2000 2500 Time, s 3000 3 Fig 3. The. attenuation of the sound energy of a nonfiltered s1gnal in the upper and the lower bodies of the hall depending on the absorption in the upper body and the change in its volume. 1 and 2 - in the lower body v! and the upper body v2 respectively, when there is no sound-absorbing material above the suspended ceiling; 3 and 4 - same with the sound-absorbing material above the suspended ceiling rigid surface. Sr----------------------------- 7 ~ 6 E ~ 5 c 0 ~ 4 ftl ... 1l 3 ... Gl > 2 Gl 0:: 2 / I 0 Ol L() ,..._ 8 CXl Frequency, Hz Fig 2. The frequency-dependence of the reverberation time in the halls with varied volumes. 1 - without sus- pended ceiling, V = 9000 m3 ; 2 - for the lower body, with the suspended ceiling and V1 = 8273 mJ; 3 - same for the upper body, when V = 748 mJ E, dB 100 r---.----r-----r----"T----- 90 80 70 60 ' ' I I I I --~----~-----~----~----~--- ' 1000 1500 2000 2500 Time, s 3000 3 Fig 3. The. attenuation of the sound energy of a nonfiltered s1gnal in the upper and the lower bodies of the hall depending on the absorption in the upper body and the change in its volume. 1 and 2 - in the lower body v! 3. Results and the upper body v2 respectively, when there is no sound-absorbing material above the suspended ceiling; 3 and 4 - same with the sound-absorbing material above the suspended ceiling E, dB 100 r---.----r-----r----"T----- 90 80 70 60 ' ' I I I I --~----~-----~----~----~--- ' 1000 1500 2000 2500 Time, s 3000 3 Sr----------------------------- 7 ~ 6 E ~ 5 c 0 ~ 4 ftl ... 1l 3 ... Gl > 2 Gl 0:: 2 / I 0 Ol L() ,..._ 8 CXl Frequency Hz Sr----------------------------- 7 ~ 6 E ~ 5 c 0 ~ 4 ftl ... 1l 3 ... Gl > 2 Gl 0:: 2 / I 0 Ol L() ,..._ 8 CXl Frequency, Hz Fig 2. The frequency-dependence of the reverberation time in the halls with varied volumes. 1 - without sus- pended ceiling, V = 9000 m3 ; 2 - for the lower body, with the suspended ceiling and V1 = 8273 mJ; 3 - same for the upper body, when V = 748 mJ Time, s Time, s Fig 3. The. attenuation of the sound energy of a nonfiltered s1gnal in the upper and the lower bodies of the hall depending on the absorption in the upper body and the change in its volume. 1 and 2 - in the lower body v! and the upper body v2 respectively, when there is no sound-absorbing material above the suspended ceiling; 3 and 4 - same with the sound-absorbing material above the suspended ceiling Fig 3. The. attenuation of the sound energy of a nonfiltered s1gnal in the upper and the lower bodies of the hall depending on the absorption in the upper body and the change in its volume. 1 and 2 - in the lower body v! and the upper body v2 respectively, when there is no sound-absorbing material above the suspended ceiling; 3 and 4 - same with the sound-absorbing material above the suspended ceiling Frequency, Hz Fig 2. The frequency-dependence of the reverberation time in the halls with varied volumes. 1 - without sus- pended ceiling, V = 9000 m3 ; 2 - for the lower body, with the suspended ceiling and V1 = 8273 mJ; 3 - same for the upper body, when V = 748 mJ . 92. tion, when the process is still not settled. 3. Results When the area of the opening is small (10m2), the energy attenu- ation is very rapid in the large lower body during the first 800 rns; the attenuation in the small upper body, on the contrary, is very slow, with no attenuation up to 250 rns. This indicates that the upper body is influ- enced by the sound energy of the lower volume and there is almost no attenuation when the process is still not settled. When the area of the opening is 50 m2, the attenuation of energy of the lower body is slower and that of the upper body is faster, starting with 200 rns. The interaction of energies becomes less pronounced as the area of the opening increases. 16 14 ~ 12 e; 10 .. :; 8 " i 6 ~ 4 2 0 ./3 \2 - 1 ~ ~ ~ 8 ~ ~ ~ ~ ~ § ~ ~ ~ ~ ~ ~ ~ ~ Frequency, Hz Fig 4. Dependence of the relative sound energy returning from the upper body to the lower body on the area of the opening. 1 - area of the opening 10 m2; 2 - 100m2; 200 m2 16 14 ~ 12 e; 10 .. :; 8 " i 6 ~ 4 2 0 ./3 \2 - 1 ~ ~ ~ 8 ~ ~ ~ ~ ~ § ~ ~ ~ ~ ~ ~ ~ ~ Frequency, Hz Frequency, Hz Fig 4. Dependence of the relative sound energy returning from the upper body to the lower body on the area of the opening. 1 - area of the opening 10 m2; 2 - 100m2; 200 m2 Fig 4. Dependence of the relative sound energy returning from the upper body to the lower body on the area of the opening. 1 - area of the opening 10 m2; 2 - 100m2; 200 m2 The attenuation of energies of both bodies is different at a certain moment of time. Fig 6 demon- strates the attenuation of the sound fields over 4000 rns. on frequency. The investigations show that both the energy leaving the lower body and the upper body do not depend on frequency, therefore it is better to take the same frequency for the determination of their dependencies. 3. Results As the process becomes settled, the attenuation of energy approximately after 1000-1100 rns has an exponent character and a continuous interchange of energies takes place during the interaction of the sound fields of both bodies. Ao; the area of the opening increases, the moment at which the settled process starts and from which the attenuation of the lower body energy acquires an exponent character, is postponed. The larger the area of the opening, the faster is the attenuation of the sound fields of both bodies, with smaller reverberation. Fig 5 depicts the dependence of the energy attenu- ation in both bodies on the area of the opening. In all cases of investigation, the frequency value was taken as 200 Hz, the all'iorption coefficient of the lower volume a 1 , the absorption coefficient of the upper volume a 2 , H = 4 m and the energy of the lower volume a 1 obtained from the experimental data. In Fig 7 is shown the dependence of the attenua- tion of energy on the change in the sound absorption of the lower body. The area of the opening has a profound effect on the attenuation and interaction of energies. This is particularly noticeable at the initial stage of attenua- E, dB 65~-,---T---.~~--~--r-~r--. 55 50 45 ---~---~---~---~---~---~---~- . . . 40~~--~--~--~--~--~~~~ 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Time, s Fig 5. The dependence of the attenuation of the sound energy over 800 ms on the area of the opening. 1 and 2 - volume of the lower body V1 and of the upper body V2 when the area of the opening is 10 m2; 3 and 4 - same with the area of 50 m2 E, dB 70r--.---.--,---r-~---r--,---. 50 40 ~ ---~---~---~---~---~---~- ' I I I I I I I I I I I I I 0.5 1.5 2 2.5 3 3.5 4 Time, s Fig 6. The dependence of the attenuation of the sound energy over 4000 ms on the area of the opening. I and 2 - volume of the lower body V1 and of the upper body V2 when the area of the opening is 10 m2; 3 and 4 - same with the area of 50 m2 dB 65~-,---T---.~~--~--r-~r--. 55 50 45 ---~---~---~---~---~---~---~- . . . 40~~--~--~--~--~--~~~~ 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Time, s E, dB 70r--.---.--,---r-~---r--,---. 3. Results 50 40 ~ ---~---~---~---~---~---~- ' I I I I I I I I I I I I I 0.5 1.5 2 2.5 3 3.5 4 Time, s Time, s Time, s Fig 5. The dependence of the attenuation of the sound energy over 800 ms on the area of the opening. 1 and 2 - volume of the lower body V1 and of the upper body V2 when the area of the opening is 10 m2; 3 and 4 - same with the area of 50 m2 Fig 5. The dependence of the attenuation of the sound energy over 800 ms on the area of the opening. 1 and 2 - volume of the lower body V1 and of the upper body V2 when the area of the opening is 10 m2; 3 and 4 - same with the area of 50 m2 Fig 6. The dependence of the attenuation of the sound energy over 4000 ms on the area of the opening. I and 2 - volume of the lower body V1 and of the upper body V2 when the area of the opening is 10 m2; 3 and 4 - same with the area of 50 m2 . 93. /:·, dB &5~-----,-------.------,-------, GO 55 :~: : . ' ' I--- __ 4~-------~------ --....... , I . --........__ . . ·--..._ ' -~------~---- 50 45 -------~-------~-------~ . . 40~----~------~------~----~ 0 0.5 1.5 2 Time, ' Fig 8. The dependence uf the attenuation of the sound energy on the chan~e in the volume' of the lower and upper bodie~. when the an:a of the opening is 50 m2, a 1 = 0.1; a 2 =0.01. I and 2 - the volunu: of the lower body V = R273 m3 and of the upper body V2 = 748 m3; 3 and 4 - V1 = 6080 m1 and V2 = 2992 m1 E, dB &5~--~-----.-----.-----.-----, 55 50 0.2 0.4 Time, s Fig 7. The dependence of the attenuation of the sound energy on the change in the sound absorption in the lower body V1. 1 and 2 - the lower body V1 and the upper body V2, when a 1 = 0.3; 3 and 4 - same with a 1 = 0.1 /:·, dB &5~-----,-------.------,-------, GO 55 :~: : . ' ' I--- __ 4~-------~------ --....... , I . --........__ . . ·--..._ ' -~------~---- 50 45 -------~-------~-------~ . . 3. Results 40~----~------~------~----~ 0 0.5 1.5 2 Time, ' Fig 8. The dependence uf the attenuation of the sound energy on the chan~e in the volume' of the lower and upper bodie~. when the an:a of the opening is 50 m2, a 1 = 0.1; a 2 =0.01. I and 2 - the volunu: of the lower body V = R273 m3 and of the upper body V2 = 748 m3; 3 and 4 - V1 = 6080 m1 and V2 = 2992 m1 /:·, dB &5~-----,-------.------,-------, GO 55 :~: : . ' ' I--- __ 4~-------~------ --....... , I . --........__ . . ·--..._ ' -~------~---- 50 45 -------~-------~-------~ . . 40~----~------~------~----~ 0 0.5 1.5 2 Time, ' E, dB &5~--~-----.-----.-----.-----, 55 50 0.2 0.4 Time, s Fig 8. The dependence uf the attenuation of the sound energy on the chan~e in the volume' of the lower and upper bodie~. when the an:a of the opening is 50 m2, a 1 = 0.1; a 2 =0.01. I and 2 - the volunu: of the lower body V = R273 m3 and of the upper body V2 = 748 m3; 3 and 4 - V1 = 6080 m1 and V2 = 2992 m1 Fig 7. The dependence of the attenuation of the sound energy on the change in the sound absorption in the lower body V1. 1 and 2 - the lower body V1 and the upper body V2, when a 1 = 0.3; 3 and 4 - same with a 1 = 0.1 When the absorption of the lower body is small, strong interaction of the sound fields occurs at the initial stage of attenuation only. The fields of both bodies muffle slowly, because an energy interchange between both bodies takes place and the process becomes settled approximately after 800 rns. When the absorption of the lower volume increases, the attenuation of the field of the lower body is very fast, while that of the upper body is slow. In this case, the lower body gives up much energy to the upper body and absorbs more of it. As the sound absorption of the lower body increases, the exponent character of the energy attenuation has a later start- ing point, while the settled phase of the process starts earlier. continuous interchange of energies starts. As the vol- ume increases, the exponent attenuation of the sound field is postponed. Conclusions 1. The experimental evidence provided by the investigations show that the dividing of the hall volume into two bodies, with the volume of one body 11 times the volume of the other, results in almost identical reverberation times of both bodies aU over the frequency range, with little dependence on the change in the volume of these bodies. 2. 2. In the acow;tically-interrelatcd premises, both during the settled and un-;cttlcd phases of the process, it is only at certain frequencies that the sound fields muffle along different exponents with different con- stants. The attenuating energy of the lower body exerts stronger inllucncc on the upper lxldy only in the phase of the un<;ettlcd process. The influence of the change in the volumes of the bodies on the energy attenuation is shown in Fig 8. When the lower body is large and the upper one is small, the attenuation of the sound energy in both bodies has almost identical ex-ponent character. After 200-300 rns, a continuous energy interchange between both bodies starts. The attenuation of sound takes place with almost identical reverberatiom. When the distance from the suspended ceiling to the rigid surface is 41X) em, i.e. when the lower body becomes smaller and the upper one becomes larger, the exponent energy attenu- ation in the lower body starts only after 1000 m~. while that in the upper body - after 600 m'i only. A'i long as the process is unsettled, the energy of the lower body has strong impact on the upper body during 1000 m'i. From this moment, the process becomes settled and a 3. 3. The outcoming energies do not depend on frequency, while the returning ones arc little de- pendent on it. -l. As the opening between the bodies increases, the moment at which the settled process starts and up to which more energy is given up to the upper body by the lower one, occurs sooner. A<i the absorption of the lower body increases, the period over which the lower body exerts stronger influence on the upper one becomes longer and the attenuation of the lower lxldy energy is very fast while the pwccss is still unsettled. -l. As the opening between the bodies increases, the moment at which the settled process starts and up to which more energy is given up to the upper body by the lower one, occurs sooner. Jteikta 1997 02 07 Jteikta 1997 02 07 Conclusions -94- virsutinio turio energija slopsta kur kas leciau, negu apa- tinio turio energija. Energij'q slopimas nera eksponentinis. As the lower body is reduced and the upper one becomes larger, the unsettled phase of the process takes more time and during it the lower body gives up more energy to the upper body. Apatinis ir virsutinis garso laukai buvo aprasyti formu- Iemis, pagal kurias apskaiciuotos abiejq turiq absor- buotosios, iSejimo ir jejimo energijos. Skaiciavimais rasta, kad energijos, iSeinancios iS abiej'q turiq, nepriklauso nuo daznio, o j abu turius grjZtancios energijos mal3i priklauso nuo daznio. V. Stauskis Mazejant apatiniam turiui ir didejant virSutiniam tliriui eksponentinis garso Janko slopimas prasideda velesniu laiko momentu, o nenusistovejus procesui energij'q Sllveika tarp abiejv ruriq uzima platesnj laiko interval~!. References 1. V. Stauskis. Rezonansinil) kabamqjq Iubq jtaka sales akustiniams rodikliams II Statyba, 1996, Nr. 3(7). V.: Technika, 1996, p. 64-69. Skaiciavimais nustatyta, kad garso laukv s11veika priklauso nuo plysio plato. Skaiciavimams plysio plotas parinktas lygus visq plysiq tarp plokStumq suminiam plotui. Nenusistovejus procesui energijos slopimas esant vienam dazniui yra dang staigesnis esant mazesniam plysio plotui apatiniame dideliame turyje, ir jis slopsta pagal dvi skir- tingas eksponentes. Siame laiko intervale apatinis turis atiduoda daug energijos virsutiniam ruriui. Nusistovejus procesui abu turiai pasikeicia energijomis, bet slopimas vyksta su skirtingomis reverberacijomis. Didejant plysio plotui mazeja laiko tarpas, iki kurio apatinis ruris stipriai veikia virsutinj turj, o eksponentinis garso Janko slopimas prasideda anksciau. 2. B. CI>yp,eyeB. 3neKTpoaKycmKa. M.: focreXJn,nar, 1948. 315 c. 3. f. fon,n6epr. BpeMH peBep6epaUI'!lf B CBH3aHHhiX IIOMe- J.UeHlUIX I I V1 Bcecoi03HaH aKycntqecKaH KOHcpepeH- UHH. M., 1968, c. 1-5. 4. X. lliqnp:.KecKHii. YrrpomeHHhllf Mero.n pacqera peBep- 6epaunn cBH3aHHhiX rroMemeHuii I I C6opHHK HayqHhiX rpy,noB HHJ1CCI>, M., 1989. 38-44 c. 5. H. Kuttruff. Room Acoustics. New-York, 1991. 380 p. Didejant didelio apatinio tlirio absorbcijai jo garso Iauko slopimas darosi staigesnis, ir jis labiau veikia virSutinj garso lauk11. VirSutinio turio energijos slopimas yra gerokai letesnis, negu apatinio turio, ir apatinio turio absorbcijos kitimas jau turi nedidelc; jtak11. Kuo didesne pirmo tlirio absorbcija, tuo veliau prasideda eksponentinis garso lauko slopimas. Santrauka Eksperimentais ir teoriskai nagnneJama dviejq turiq, atskirtq rezonansinemis kabamosiomis Iubomis su sta- ciakampes formos plysiais tarp plokstumq garso laukq, sllveika. Eksperimentas atliktas su sales modeliu masteliu 1:25. Nustatyta, kad dviejq turiq, kurie skiriasi net 11 kartq, reverberacijos laikai yra vienodi visame dazniq diapazone. To neturetq buti, nes smarkiai skiriantis ruriams turi smarkiai skirtis ir reverberacijos Iaikai, nes jq reikSmems turis turi Iemiamos jtakos. Tai rodo, kad tarp abiej'q ruriq yra garso laukq SI!Veika. Nefiltruoto signalo energijos iki 1500 ms slopsta vienodai, o ve!iau virSutinio mazo turio enrgijos slopimas yra Ietesnis, negu didelio apatinio turio. Kada virS lubq yra gars11 absorbuojanti medziaga, tai slopimo pobudis stipriai keiciasi. Abiejq turiq energijos vienodai slopsta tile iki 500 ms, o paskui ilgejant laikui Vytautas STAUSKIS. Doctor, Associate Professor. Depa- rtment of Building Structures. Vilnius Gediminas Techni- cal University, 11 Sauletekio Ave, 2040 Vilnius, Lithuania. Doctor's degree, 1974. From 1974 at VTU Depart- ment of Building Structures. Scientific visits: Moscow Civil Engineering Institute, Sankt-Petersburg Politechnical In- stitute. Research interests: experimental testing of halls by primary hall models and on site, computer simulation of theoretic tasks, wave diffraction and reflections, direct sound and subjective acoustic indicators, large-dimension resonance structures, early attenuation of acoustic field and its relation to hall acoustics. Vytautas STAUSKIS. Doctor, Associate Professor. Depa- rtment of Building Structures. Vilnius Gediminas Techni- cal University, 11 Sauletekio Ave, 2040 Vilnius, Lithuania. Doctor's degree, 1974. From 1974 at VTU Depart- ment of Building Structures. Scientific visits: Moscow Civil Engineering Institute, Sankt-Petersburg Politechnical In- stitute. Research interests: experimental testing of halls by primary hall models and on site, computer simulation of theoretic tasks, wave diffraction and reflections, direct sound and subjective acoustic indicators, large-dimension resonance structures, early attenuation of acoustic field and its relation to hall acoustics. . 95.
https://openalex.org/W2613688818	https://infoscience.epfl.ch/record/228645/files/articles-srep46721.pdf	English	null	An ecologically-controlled exoskeleton can improve balance recovery after slippage	Scientific reports	2,017	cc-by	7,942	An ecologically-controlled exoskeleton can improve balance recovery after slippage V. Monaco1,2, P. Tropea1, F. Aprigliano1,†, D. Martelli1,3,†, A. Parri1, M. Cortese1, R. Molino-Lova2, N. Vitiello1,2,* & S. Micera1,4,* 6 7 N received: 20 December 2016 accepted: 23 March 2017 Published: 11 May 2017 The evolution to bipedalism forced humans to develop suitable strategies for dynamically controlling their balance, ensuring stability, and preventing falling. The natural aging process and traumatic events such as lower-limb loss can alter the human ability to control stability significantly increasing the risk of fall and reducing the overall autonomy. Accordingly, there is an urgent need, from both end-users and society, for novel solutions that can counteract the lack of balance, thus preventing falls among older and fragile citizens. In this study, we show a novel ecological approach relying on a wearable robotic device (the Active Pelvis Orthosis, APO) aimed at facilitating balance recovery after unexpected slippages. Specifically, if the APO detects signs of balance loss, then it supplies counteracting torques at the hips to assist balance recovery. Experimental tests conducted on eight elderly persons and two transfemoral amputees revealed that stability against falls improved due to the “assisting when needed” behavior of the APO. Interestingly, our approach required a very limited personalization for each subject, and this makes it promising for real-life applications. Our findings demonstrate the potential of closed-loop controlled wearable robots to assist elderly and disabled subjects and to improve their quality of life. Balance and postural control are achieved by implementing sophisticated motor control strategies to ensure sta- bility and prevent falling1,2. The natural aging process alters the ability to address unexpected perturbations of bal- ance, thus increasing the probability of falling3,4. Age-related diseases and traumatic events can further exacerbate the inability to maintain balance, leading to a dramatic increment of both the risk of balance loss and the severity of related accidents4–6. The consequences of falls are costly for patients and for society.i h In addition to prevention programs based on training7,8, and behavior and environmental modifications9, the effectiveness of which are generally acknowledged, advanced wearable robotic systems could also reduce the risk of falling. For example, novel approaches are currently under investigation to properly design and control transfemoral prostheses aimed at reducing falls in amputees10,11. However, these solutions cannot be transferred to all fragile and prone-to-fall subjects. On the other hand, robotic exoskeletons could represent a more general and useful solution for a larger cohort of subjects. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports An ecologically-controlled exoskeleton can improve balance recovery after slippage V. Monaco1,2, P. Tropea1, F. Aprigliano1,†, D. Martelli1,3,†, A. Parri1, M. Cortese1, R. Molino-Lova2, N. Vitiello1,2,* & S. Micera1,4,* 6 7 N Nonetheless, despite their potential benefits, stakeholders still notice that the minimization of fall risk is one of the most important factors not yet addressed in the design of these active orthosis12. One of the main problems of current exoskeletons, limiting their use, is the poor human-machine interac- tion13. This issue is particularly important when – as in our case – the users still retain important voluntary abili- ties. In fact, prone-to-fall subjects are still able to generate “countermeasures” to falling events but not strong and fast enough to avoid these events in most of the cases. In this situation, it is necessary to develop ecological and symbiotic solutions able to help only when necessary (onset of imminent fall) providing only an extra “delta” of response (not substituting by “assisting when needed” the subjects). The hybrid subject-exoskeleton must become a “symbiotic” union with the artificial system providing benefits with no (or very limited) disturbance to the subjects. 1The BioRobotics Institute, Scuola Superiore Sant’Anna, Pisa, Italy. 2IRCSS Don Carlo Gnocchi Foundation, Firenze, Italy. 3Department of Mechanical Engineering, Columbia University, New York, NY 10027 USA. 4Bertarelli Foundation Chair in Translational NeuroEngineering, Center for Neuroprosthetics and Institute of Bioengineering, School of Engineering, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland. †These authors contributed equally to this work. These authors jointly supervised this work. Correspondence and requests for materials should be addressed to V.M. (email: vito.monaco@santannapisa.it) or S.M. (email: silvestro.micera@santannapisa.it) SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 1 www.nature.com/scientificreports/ Figure 1. Control strategy. (A) The mechatronic platform (1) provides sudden and unexpected slipping-like perturbations while the subject was steadily walking (2). The subject’s balance control was hence challenged (3) and, accordingly, the Active Pelvis Orthosis (4) could supply the assistive strategy for stability recovery. (B) The balance loss is detected in real time by an algorithm running in the APO control unit and comparing the actual hip angles of the robot (i.e., θ) with those predicted by a pool of adaptive oscillators (i.e., θˆ). (C) Once the slipping-like perturbation was detected, counteracting torques were supplied by the APO at the hip joints to promote balance recovery. Panel C was created by Arch. Alessio Tommasetti Panel C was created by Mr. Francesco Giovacchini. Figure 1. Control strategy. (A) The mechatronic platform (1) provides sudden and unexpected slipping-like perturbations while the subject was steadily walking (2). Resultsh The proposed assistive strategy was tested with eight elderly persons (68.9 ± 5.1 years old) and two transfemoral amputees (65.5 ± 6.4 years old). They were asked to manage unexpected slipping-like perturbations (i.e., one of the most common causes of falling15) delivered while walking at their preferred speeds (elders: 0.89 ± 0.11 m/s; amputees: 0.69 ± 0.06 m/s; see also Table S1). The mechanical perturbations were delivered by a mechatronic platform designed to expose enrolled participants to well-controlled experimental conditions (see also Materials and Methods, Section A and Supplementary Materials and Methods, Section A)18. , pp y , ) During the experimental sessions, subjects managed unexpected slippage wearing or without wearing the APO. In the former case, the exoskeleton was used either in zero-torque modality (Z-mode trials), that is, no assistance was provided to the users, or to supply assisting torques at hip joints after detecting postural transitions (A-mode trials). The A-mode consisted in applying synchronous extensor and flexor torques (Fig. 1C) based on the subjects’ body weight, to the perturbed (PL) and the unperturbed (UL) limbs, respectively, in order to coun- teract the downward displacement of the center of mass (COM; Materials and Methods, Section B; as represent- ative examples, see Movie S1 and Movie S2). In the latter case, to test the unobtrusiveness of the APO during the balance recovery, a subset of the enrolled elderly subjects (n = 5) was asked to manage slipping-like perturbations without wearing the exoskeleton (no-APO trials), as well. Analysis of the biomechanical response during balance recovery. When a perturbation was applied, both elderly subjects and amputees showed a significant change of their lower limb kinematics. In par- ticular, the hip and the ankle of the PL were more flexed compared to the steady locomotion (Fig. 3A and B).i pl p y g When the APO was set in A-mode the early modifications of the leg kinematics allowed the detection algo- rithm to identify balance loss in about 350 ms (Fig. 2B) and to enable the mitigation strategy. This assistive strat- egy generally ended during the swing phase of the PL (~650 ms).hi The hip range of motion (ROM) of the PL significantly changed among conditions (i.e., no-APO, Z-mode and A-mode; one way repeated measures ANOVA, N = 8, p = 0.001 for elderly subjects; Fig. 3C). Resultsh The post-hoc Tukey test revealed that the hip ROM of the PL was significantly different during the A-mode trials compared to the other two conditions (p = 0.002 and p = 0.039, respectively for no-APO and Z-mode). Specifically, values related to A-mode were lower than the other two conditions (Z-mode and no-APO), which instead appeared comparable (Fig. 3C). ( g ) Conversely, the hip ROM of the UL was not modified among conditions (i.e., no-APO, Z-mode and A-mode; one way repeated measures ANOVA, N = 8, p = 0.635 for elderly subjects; Fig. 3C). y p p y j g ) Concerning the amputees, the behavior of these participants was comparable to that observed for elderly people. Specifically, the hip ROM of the PL, for amputees, was lower when APO was set in A-mode than during Z-mode (Fig. 3D). Differently, no modifications were observed at hip ROM of the UL (Fig. 3D). Stability against balance loss: effects of the proposed approach. To investigate whether our approach was effective in promoting balance recovery, the COM stability19–21 and the margin of stability (MOS)22,23 were computed. These metrics were calculated at the touch down of the perturbed leg (TDP) to ensure that the assistive action, when present, was completed. Additional details are reported in Materials and Methods, Section E. The time course of the COM motion state is shown in a representative case for both elderly subjects and amputees wearing the APO (Fig. 4A and B, respectively). During the first single support phase (I–II events, lift off to touch down of the unperturbed leg, LOU -TDU, red/cyan dashed lines), intra-subject differences were not observed between the Z- and A- modes because the mitigation strategy, when it was enabled, was not yet activated.tf During the second single support phase (III-IV events, lift off to touch down of the perturbed leg, LOP–TDP, red/cyan solid lines), we observed different behaviors between Z-mode and A-mode for both the healthy subject and the amputee. Specifically, when the APO was functioning in Z-mode, the COM motion state of the healthy subject moved outside the stability region and ended above the limit of the forward balance loss region; whereas the COM motion state of the amputee had a long transition to the instability region and ended close to its limit. By contrast, when the APO was functioning in A-mode, and the mitigation strategy was enabled (see yellow stars in Fig. An ecologically-controlled exoskeleton can improve balance recovery after slippage V. Monaco1,2, P. Tropea1, F. Aprigliano1,†, D. Martelli1,3,†, A. Parri1, M. Cortese1, R. Molino-Lova2, N. Vitiello1,2, & S. Micera1,4,* 6 7 N The subject’s balance control was hence challenged (3) and, accordingly, the Active Pelvis Orthosis (4) could supply the assistive strategy for stability recovery. (B) The balance loss is detected in real time by an algorithm running in the APO control unit and comparing the actual hip angles of the robot (i.e., θ) with those predicted by a pool of adaptive oscillators (i.e., θˆ). (C) Once the slipping-like perturbation was detected, counteracting torques were supplied by the APO at the hip joints to promote balance recovery. Panel C was created by Arch. Alessio Tommasetti Panel C was created by Mr. Francesco Giovacchini. Figure 2. Real time detection of slipping-like perturbations. (A) The adaptive-threshold based algorithm analyses the difference between measured (light green) and estimated (purple) hip joint angles. If the error signal (orange) is over the thresholds (dashed lines), a balance loss is detected. (B) Detection time obtained during the experimental trials for the elderly subjects and amputees (mean values ± SD). Figure 2. Real time detection of slipping-like perturbations. (A) The adaptive-threshold based algorithm analyses the difference between measured (light green) and estimated (purple) hip joint angles. If the error signal (orange) is over the thresholds (dashed lines), a balance loss is detected. (B) Detection time obtained during the experimental trials for the elderly subjects and amputees (mean values ± SD). In this study, we show that this approach can be implemented improving balance recovery in elderly people and transfemoral amputees. In particular, these two groups of subjects represent persons whose risks of fall ranges from moderate to severe3,4,14,15, thus they can be considered as potential users who can take advantages from a wearable assistive device. To achieve this goal a novel control algorithm16, leading an Active Pelvis Orthosis17 SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 2 www.nature.com/scientificreports/ (APO; Fig. 1 and Fig. S1, Supplementary Materials and Methods, Section B), was designed to: i) be transparent to the user’s intended movements during unperturbed motor tasks, ii) identify in real time the onset of an unex- pected slipping-like perturbation16 (Fig. 2A) and iii) generate countermeasures to restore user’s stability against falling. (APO; Fig. 1 and Fig. S1, Supplementary Materials and Methods, Section B), was designed to: i) be transparent to the user’s intended movements during unperturbed motor tasks, ii) identify in real time the onset of an unex- pected slipping-like perturbation16 (Fig. An ecologically-controlled exoskeleton can improve balance recovery after slippage V. Monaco1,2, P. Tropea1, F. Aprigliano1,†, D. Martelli1,3,†, A. Parri1, M. Cortese1, R. Molino-Lova2, N. Vitiello1,2,* & S. Micera1,4,* 6 7 N 2A) and iii) generate countermeasures to restore user’s stability against falling. Resultsh 4A and B), the COM motion state was confined within the stability region for all the duration of the second single support phase (III–IV events, LOP –TDP, red/cyan solid lines), thus facilitating the balance recovery of both healthy and amputee subjects (see Supplementary Movies S1 and S2).hi y j y The one-way ANOVA on COM stability and MOS yielded significant variation among the three conditions in elderly subjects (i.e., no-APO, Z-mode and A-mode; N = 8, p = 0.019 and p = 0.011 respectively; Fig. 4C and D, respectively). In this respect, the post hoc Tukey test revealed COM stability (p = 0.040) and MOS (p = 0.047) were significantly higher during A-mode condition than during the Z one, thus highlighting an improvement in overall balance against falls due to the assistive strategy of the exoskeleton. g gy Noticeably, the adopted mitigation strategy promoted balance recovery in transfemoral amputees, as well. As a matter of fact, both metrics (i.e., COM stability and MOS) were higher during A-mode than Z-mode (Fig. 4C and D). SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 3 www.nature.com/scientificreports/ Figure 3. Kinematic patterns at leg joints. (A and B) Joint angles (hip, knee and ankle) at PL and UL are depicted for three experimental conditions, for one elderly subject and one amputee: i) steady locomotion (mean values ± SD, grey bands); ii) Z-mode (cyan lines); iii) A-mode (red lines). Stick diagrams (on the top; PL and UL are indicated in green and blue, respectively) and APO torques (on the bottom) are shown for the A-mode condition. Pink vertical bands represent the time-intervals corresponding to the enabled assistive torques. (C) The perturbed and unperturbed hip ranges of motion during the no-APO, Z- and A-modes trials (orange, cyan and red bars, respectively) are shown for elderly subjects (mean values ± SD). The label * indicates a significant (p < 0.05) difference among trials. (D) The perturbed and unperturbed hip ranges of motion during the Z- and A-modes trials (cyan and red bars, respectively) are shown for amputee groups (mean values ± SD). The time axes in panels A and B start at the heel strike of the unperturbed gait cycle (grey/shadow area) and at the onset of the perturbed strides (blue and red lines). Figure 3. Kinematic patterns at leg joints. Resultsh (A and B) Joint angles (hip, knee and ankle) at PL and UL are depicted for three experimental conditions, for one elderly subject and one amputee: i) steady locomotion (mean values ± SD, grey bands); ii) Z-mode (cyan lines); iii) A-mode (red lines). Stick diagrams (on the top; PL and UL are indicated in green and blue, respectively) and APO torques (on the bottom) are shown for the A-mode condition. Pink vertical bands represent the time-intervals corresponding to the enabled assistive torques. (C) The perturbed and unperturbed hip ranges of motion during the no-APO, Z- and A-modes trials (orange, cyan and red bars, respectively) are shown for elderly subjects (mean values ± SD). The label * indicates a significant (p < 0.05) difference among trials. (D) The perturbed and unperturbed hip ranges of motion during the Z- and A-modes trials (cyan and red bars, respectively) are shown for amputee groups (mean values ± SD). The time axes in panels A and B start at the heel strike of the unperturbed gait cycle (grey/shadow area) and at the onset of the perturbed strides (blue and red lines). Analysis of the obtrusiveness of the APO. During unperturbed locomotion (i.e., before the onset of the perturbation), the kinematic patterns of the elderly subjects and amputees, while wearing the APO, were similar to those reported in literature (Fig. 3A and B, respectively, grey curves24,25). Thus, the exoskeleton appeared likely transparent while participants walked steadily, confirming previous results26. p p p yi g p As far as the compensatory stride after the perturbation is concerned, the pairwise Tukey’s comparison revealed that, during no-APO trials the ROM at both hip joints (i.e., PL and UL) was not significantly different (p = 0.080 and p = 0.640, respectively) than that observed when the APO was set in Z-mode (Fig. 3C). In addi- tion, the statistical analysis also revealed that both COM stability and MOS observed during the Z-mode were not significantly different than that seen when subjects were not wearing the exoskeleton (i.e., no-APO condition; p = 0.702 and p = 0.409, respectively; Fig. 4C and D, respectively). p p p y g p y According to these evidences, the reactive behavior of elderly participants was not altered by the use of the exoskeleton in Z-mode in terms of both kinematics and stability, thus revealing that the APO was unobtrusive with respect to the experimental conditions. Resultsh SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 4 www.nature.com/scientificreports/ ntificreports/ Figure 4. Analysis of the stability against fall. Subplots (A and B) show the COM motion state for one elderly subject and one amputee, during the Z- and A-modes (cyan and red lines, respectively). The green area represents the stability region: forward falls would be initiated if states exceeded the upper boundary; backward falls would be initiated if states dropped below the lower boundary. The two components of the COM motion state (i.e., position and velocity of the COM) were calculated relative to the base of support (BOS) and normalized by foot length and (gH)1/2, respectively, where g is gravitational acceleration and H is body height. Two single support phases, concerning the PL and UL, were identified from the lift-off and the touch-down of the unperturbed (I-II events, LOU and TDU) and the perturbed (III-IV events, LOP and TDP) feet, respectively. The effect of the fall mitigation action was observed during the second single support phase, and, accordingly, the dynamic stability was assessed at the end of the compensatory stride (i.e., IV event). (C) COM stability and (D) margin of stability (MOS) are depicted for the no-APO, Z- and A-modes (orange, cyan and red bars respectively), for both elderly and amputee groups (mean values ± SD). The label indicates a significant (p < 0.05) difference among trials. Figure 4. Analysis of the stability against fall. Subplots (A and B) show the COM motion state for one elderly subject and one amputee, during the Z- and A-modes (cyan and red lines, respectively). The green area represents the stability region: forward falls would be initiated if states exceeded the upper boundary; backward falls would be initiated if states dropped below the lower boundary. The two components of the COM motion state (i.e., position and velocity of the COM) were calculated relative to the base of support (BOS) and normalized by foot length and (gH)1/2, respectively, where g is gravitational acceleration and H is body height. Two single support phases, concerning the PL and UL, were identified from the lift-off and the touch-down of the unperturbed (I-II events, LOU and TDU) and the perturbed (III-IV events, LOP and TDP) feet, respectively. Resultsh The effect of the fall mitigation action was observed during the second single support phase, and, accordingly, the dynamic stability was assessed at the end of the compensatory stride (i.e., IV event). (C) COM stability and (D) margin of stability (MOS) are depicted for the no-APO, Z- and A-modes (orange, cyan and red bars respectively), for both elderly and amputee groups (mean values ± SD). The label indicates a significant (p < 0.05) difference among trials. Methods Participants and experimental setup. Ten subjects, 2 transfemoral amputees and 8 age- and anthropom- etry-matched elderly subjects (see Table S1) were enrolled for this study. The exclusion criteria were: age greater than 75 years old, falls documented in the last 6 months, relevant comorbidities (e.g., hemiplegia, degenerative nervous system diseases, chronic heart failure, chronic obstructive pulmonary disease, sever sensory deficits), poor cognitive skills (Mini-Mental State Examination score <24), inability to walk safely on a treadmill, and severe anxiety or depression. In addition, eligible amputees had, at a minimum, the ability to transfer using a household ambulator on a level surface (Medicare Functional Classification Levels K-level ≥K2). Enrolled sub- jects were informed about the purpose of the study and signed informed consent forms before the experimental sessions began. All research procedures were in accordance with the Declaration of Helsinki and were approved by the Institutional Ethics Committee of Don Gnocchi Foundation (Florence, Italy), where the experiments took place. p Subjects were asked to manage unexpected slipping-like perturbations delivered by a custom-made split-belt treadmill which belts can be independently controlled to provide multi-directional perturbations18, (Fig. 1 and Fig. S2, Supplementary Materials and Methods, Section A). Subjects donned a safety harness attached to an over- head track during the trials for safety-related issues.h g y The perturbation consisted of a forward movement of one belt at the heel strike of the foot being perturbed29,30. Two representative velocity profiles of the belts related to the perturbed and the unperturbed limbs are presented in Fig. S2. Perturbations were delivered on the prosthetic limb and on the right side for amputees and elderly sub- jects, respectively. Amputees donned their own standard prostheses and all participants used their own athletic shoes during the entire experimental session. Noticeably, the perturbation was delivered on the prosthetic side in order to allow the amputees to safety rely on the sound limb during the compensatory step. Otherwise, if the perturbation was delivered on the sound limb, the amputees would suddenly transfer their body weight from the sound limb to the prosthesis. In this condition, the prosthetic knee would be unlocked (i.e., it would be supposed to start a swing phase) thus collapsing under the amputee’s weight and leading to a fall. g p p g p g g During the experimental sessions, participants wore a novel APO, a powered wearable device17, designed to assist hip flexion-extension in the sagittal plane (Fig. Discussioni All these findings indicate that the light-weighted APO can detect in real-time the onset of a mechanical per- turbation and activate specific mitigation strategies to promote balance recovery. Specifically, our experimental design revealed that the proposed strategy effectively (p < 0.05) favorites the balance recovery in elderly people after a slippage. Moreover, our pilot tests on transfemoral amputees showed that these subjects did benefit from the APO assistance in the same perturbing conditions. We acknowledge that further experimental sessions are required to robustly confirm these latter results. However, these findings suggest that the proposed approach is easily adaptable to different groups of prone to fall subjects, and represents a valuable proof of concept concerning the attitude of wearable robotic devices to assist persons while managing challenging motor tasks. SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 5 www.nature.com/scientificreports/ Remarkably, our detection algorithm identified the lack of balance in about 350 ms (Fig. 2B), consistent with the performance estimated during our previous off-line study16 and compatible with a possible falling avoid- ance27,28. In addition, our approach does not require any subject-specific and training-based customizing proce- dure for successful implementation among different subjects. In fact, the algorithm for perturbation detection is based on the comparison between the actual leg joint angles and those predicted by a pool of adaptive oscillators (Fig. 2A). Once the algorithm’s tuning was optimized16, it was immediately able to detect postural transitions without any further subject-based training. Moreover, the amplitude of the torque was easily tuned taking into account only the body weight of the subjects (see Materials and Methods, Section B). y y g j Another very interesting characteristic of our approach is that it promotes users’ balance recovery only if required by providing an additional “impulsive” torque achieving the ecological and symbiotic approach previ- ously mentioned. Accordingly, our “assisting when needed” strategy suggests the development of a novel con- cept of wearable devices specifically designed to minimize the risk of falling. In this respect, we envisage that a more focused design would further reduce the bulkiness of wearable robotic platforms and increase usability and acceptance by senior and disabled users. This opportunity requires further investigations to identify the APO with the minimum performance necessary to effectively reduce fall risk.ti p yf y After this promising proof of concept, challenges lie ahead. Our results must be confirmed in additional sub- jects and with different types of mechanical perturbations. Discussioni It will be extremely important to understand the limits of both the detection and the mitigation algorithms under different experimental conditions and, subsequently, in real-life field tests. Indeed, a loss of balance can be induced by a number of causes, such as multi-directional slipping-like perturbations, tripping, and stumbling. In addition, the psycho-physical status of fragile persons further increases the risk of falls.i Overall, our findings demonstrate the potentials of ecologically-controlled wearable robots to assist the elderly and disabled subjects during slipping events, thus potentially improving their quality of life. www.nature.com/scientificreports/ displacement of the body. In particular, the reactive moment at the hips is responsible for the lowering of the foot onto the ground, while the reactive moment at the knees decelerates the trailing leg and absorbs part of the energy generated by the hip extensors31.f g y p According to these evidences, we hypothesized that an increment of the stiffness at hip joints against limb movements induced by the belts could promote the balance recovery. In this regard, when the A-mode is enabled, torque patterns acted in order to extend the perturbed limb and flex the unperturbed one. The assistive strategy implemented in the APO was defined as a time-dependent torque pattern which profile was designed based on two settable features: A [Nm], the amplitude; T [ms], the duration. Specifically, if the perturbation is detected at t, the set point of the counteracting torques was defined as follows for t ≤ t ≤ t* + T (elsewhere τdes was zero; see Supplementary Methods, Section B online): τ π = ± ⋅   −    ⋅ ⋅ −       ⁎ A T t t 2 1 cos 2 ( ) (1) des (1) where ± refers to flexor and extensor torques, respectively.i Despite the implemented strategy was defined in accordance with physiological countermeasure described in literature31, the duration and the intensity of the assistive torque are two critical issues which have never been investigated before. As a matter of fact, from the best of our knowledge, no authors have ever closed the loop from detecting an incipient fall to provide a suitable support by means of a wearable robotic platform. Therefore, to address these issues, we adopted following assumptions: (i). Since a fall occurs in 0.7–1 s29 and the detection time is expected to be about 0.3–0.4 s12, the duration of the assistive torque was set at 0.25 s. ii). Since the reactive response depends on the subject inertia, the APO provided a torque proportional to the gross (participant + exoskeleton) weight, that is, 0.2 Nm/kg. Data collection. The protocol consisted of two repetitions for each of the following experimental condi- tions: assisted (i.e., A-modes); no-assisted (i.e., Z-mode); without wearing the exoskeleton (i.e., no-APO) trials. www.nature.com/scientificreports/ To reduce the effect of the adaptation on the results: i) the full experimental session accounted for three addi- tional trials, in which no perturbation was delivered; ii) all (perturbed and no-perturbed) trials were randomly arranged; iii) participants did not know whether they would be perturbed or not; and iv) participants did not know whether they would be assisted or not.h y The 3D kinematics of a set of spherical markers located on suitable body landmarks of the whole body and on the APO were recorded at 250 Hz using a six-camera-based Vicon 512 Bonita 10 Motion Analysis System (Oxford, U.K.) (more details in Supplementary Materials and Methods, Section D). ) ( pp y ) Body kinematics records, belt movements and the APO were synchronized using a logic pulse generated by he split-belt treadmill during the delivery of the perturbations. Data processing. Raw data were firstly pre-processed to remove the effects of noise and artifacts. In par- ticular, high-frequency related noise was removed from digitized coordinates by low-pass filtering data (zero-lag, 4th order Butterworth low-pass filter) with cut off at 10 Hz. The cut off frequency was selected as described else- where32. Raw ground reaction force signals were also band-pass-filtered (Butterworth filter, 4th order, cutoff at 0.5–10 Hz) in accordance with previous authors33,34.h ) p A full body model accounting for 15 segments and 42 internal degrees of freedom was developed. The 15 seg- ments were: head/neck, chest, abdomen/pelvis, upper arms, forearms, hands, thighs, shanks and feet. All joints were approximated as spherical and their center was located in accordance with the literature30,35–38. For the ith body segment, an orthogonal local reference frame was located in its own center of mass and defined according to ISB recommendations39. The time course of limb joints angles into the sagittal plane were estimated by using the 3D approach described in literature39. The range of motion (ROM) at hip joints was computed as the difference between the maximum and the minimum values during the compensatory cycle (i.e., from the onset of the per- turbation to the touch down of perturbed leg, TDP). g Body segment inertial parameters (i.e., mass and center of mass position) were calculated using procedures described by Zatsiorsky and colleagues40, and modified by de Leva41. The inertial properties of the prosthesis were considered equal to those of the sound limb. Methods 1 and Fig. S1; more details in Supplementary Materials and Methods, Section B). The control strategy driving the APO can switch between two working modalities, i.e., zero-torque and assistive modes (Z- and A-modes, respectively). The former (Z-mode) is designed to allow the wearer to freely perform movements without being obstructed by the device. The latter (A-mode) is enabled only when a balance loss is detected and supplies torques at both hip joints to promote stability recovery (more details in Materials and Methods, Section B).h ) The postural transition was detected by a plug-and-play algorithm based on adaptive oscillators and run- ning in the APO’s control unit (Supplementary Materials and Methods, Section C)16. Once the perturbation was detected, the exoskeleton switched from Z-mode to A-mode, and the APO supplied the counteracting torques. The assistive strategy (A-mode) consisted in increasing the stiffness at hip joints against limb movements induced by the belts, thus counteracting the downward movement of the body COM. In this regard, when the A-mode was enabled, torque patterns acted in order to extend the perturbed limb and flex the unperturbed one (more details in Materials and Methods, Section B). , ) Furthermore, a subset of five elderly subjects performed additional trials without wearing the APO in order to test the hypothesis that the APO was unobtrusiveness with respect to the experimental conditions. Assistive strategy. Literature revealed that after a slip, reactive joint moments at the knee and hip joints are mostly aimed at slowing down the sliding motion of the foot and, possibly, at minimizing the downward SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 6 www.nature.com/scientificreports/ SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 www.nature.com/scientificreports/ Mass and center of mass of each of the body segments were used to estimate the whole body center of mass (COM).tfi y ( ) Touch down and lift off were identified by visual inspection of kinematics of the feet and time course of the vertical component of the ground reaction force.t For each subject and trial, data were subdivided in two subsets: data recorded before and after the onset of the perturbation. The former referred to the last three ipsilateral unperturbed strides, in which each cycle started and finished with the heel strike of the leg being perturbed. These data were subdivided into strides, individually time-interpolated over 250 points, and averaged in order to have a representative unperturbed gait cycle. After the touch down of the foot being perturbed, four consecutive time events were identified: (I) the lift off of the unperturbed foot (LOU); (II) the instant immediately before (i.e., 1 frame, consisting of 1/250 s) the touch down of the unperturbed foot (TDU); (III) the lift off of the perturbed foot (LOP); and (IV) the instant immediately before the touch down of the perturbed foot (TDP). Overall, these four consecutive time events allowed for investigating stability against balance loss, under the hypothesis that the balance recovery may require more than one step to be achieved. SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 7 www.nature.com/scientificreports/ Analysis of the stability against balance loss. To investigate the effectiveness of the assistive strategies while promoting balance recovery, the following metrics were calculated: margin of stability (MOS) and COM stability. y The MOS along the antero-posterior (AP) direction was computed according to Hof et al.23 as: (2) = − MOS U X (2) x MAX COM = − MOS U X x MAX COM where UMAX is the anterior boundary of the base of support (BOS) and XCOM is the position of the extrapolated COM in the AP direction estimated as follows: = + ˙ X COM COM / g/h (3) COM x x COM = + X COM COM / g/h COM x x COM (3) where COMx and  COMx are the AP components of position and velocity of the COM, hCOM is the distance between COM and floor while subjects were keeping the unperturbed upright stance, and g is the gravitational acceleration. www.nature.com/scientificreports/ The MOS was quantified at the end of the compensatory stride (i.e., TDP).hh hi The COM motion state was used to investigate the ongoing of the balance recovery. This variable has been introduced by previous authors19–21, and indicates the range of values for both COM position and velocity to allow a feasible control of subject’s stability. Specifically, based on these state variables it is possible to determine upper and lower boundaries identifying a feasible stability region against backward or forward balance loss. The COM stability was hence defined as the shortest distance from the instantaneous COM motion state to the limits against backward or forward balance loss21 and was calculated at TDP. Statistical analysis. Mean and standard deviation were the main descriptive statistics used to summarize the main features of data samples. With respect to elderly subjects, the effect of the exoskeleton (i.e., Z-mode vs. A-mode vs. no-APO) on hip ROM, COM stability and MOS was assessed by means of one-way repeated meas- ures ANalysis Of VAriance (ANOVA). If significant, this analysis was followed up with Tukey Honestly Significant Difference test in order to test two different hypothesis: i) the proposed assistive strategy (i.e., A-mode) favorites the balance recovery compared to the Z-mode; ii) the reactive response of the subjects is not altered when the exoskeleton works in Z-mode, i.e., the device is not obtrusive with respect to the experimental conditions.h p p Data processing was performed using custom written MATLAB (The MathWorks, Inc., Natick, MA, USA) scripts; the statistical analysis was carried out by means of Minitab (Minitab Inc., PA, USA). Significance of sta- tistical tests was set at α = 0.05. References 1. Marigold, D. S. & Misiaszek, J. E. Whole-body responses: neural control and implications for rehabilitation and fall prevention Neuroscientist 15, 36–46 (2009). 1. Marigold, D. S. & Misiaszek, J. E. Whole-body responses: neural control and implications for rehabilitation and fall prevention. Neuroscientist 15, 36–46 (2009). 2 Mi i k J E N l t l f lki b l if f lli th t l ti E i d p t i i 34 128 134 2. Misiaszek, J. E. Neural control of walking balance: if falling then react else continue. Exercise and sport sciences reviews 34, 128 (2006). ( ) 3. Berg, W. P., Alessio, H. M., Mills, E. M. & Tong, C. Circumstances and consequences of falls in independent community-dwelling older adults. Age Ageing 26, 261–268 (1997). g g g 4. Masud, T. & Morris, R. O. Epidemiology of falls. Age Ageing 30 gy g g g 5. Rubenstein, L. Z. Falls in older people: epidemiology, risk factors and strategies for prevention. Age Ageing 35 Suppl 2, ii37–ii41, doi 35/suppl_2/ii37 (2006).i pp ( ) 6. Martelli, D. et al. Stability against backward balance loss: Age-related modifications following slip-like perturbations of multiple amplitudes. Gait & posture 53, 207–214, doi: 10.1016/j.gaitpost.2017.02.002 (2017).i p j g 7. Grabiner, M. D., Bareither, M. L., Gatts, S., Marone, J. & Troy, K. L. Task-Specific Training Reduces Trip-Related Fall Risk in Women Med Sci Sport Exer 44, 2410–2414, doi: 10.1249/Mss.0b013e318268c89f (2012).f y port Exer 44, 2410–2414, doi: 10.1249/Mss.0b013e318268c89f (201 p 8. Hauer, K., Becker, C., Lindemann, U. & Beyer, N. Effectiveness of physical training on motor performance and fall prevention in cognitively impaired older persons: a systematic review. American journal of physical medicine & rehabilitation/Association of Academic Physiatrists 85, 847–857, doi: 10.1097/01.phm.0000228539.99682.32 (2006).f y p 9. Perula, L. A. et al. Effectiveness of a multifactorial intervention program to reduce falls incidence among community-living older adults: a randomized controlled trial. Archives of physical medicine and rehabilitation 93, 1677–1684, doi: 10.1016/j.apmr.2012.03.035 (2012).h 0. Thatte, N. & Geyer, H. Toward Balance Recovery with Leg Prostheses using Neuromuscular Model Control. IEEE Trans Biomed Eng doi: 10.1109/TBME.2015.2472533 (2015).fifi 1. Highsmith, M. J. et al. Safety, energy efficiency, and cost efficacy of the C-Leg for transfemoral amputees: A review of the literature Prosthet Orthot Int 34, 362–377, doi: 10.3109/03093646.2010.520054 (2010).ff 11. Highsmith, M. J. et al. www.nature.com/scientificreports/ (John Wiley & Sons, Inc., 1990).t . Winter, D. A. Biomechanics and Motor Control of Human Movemt 33. Quagliarella, L., Sasanelli, N. & Monaco, V. Drift in Posturography Systems Equipped With a Piezoelectric Force Platform: Analysis and Numerical Compensation. Instrumentation and Measurement, IEEE Transactions on 57, 997–1004, doi: 10.1109/ TIM.2007.913833 (2008).h ( ) 4. Winter, D. A. The biomechanics and motor control of human gait: normal, elderly and pathological. 2nd edn, (University of Waterloo Press, 1991). 35. Wang, X. Construction of arm kinematic linkage from external surface markers. Proceeding of the 4th International Symposium on 3D Analysis of Human Movement (1996). 36. Anglin, C. & Wyss, U. P. Arm motion and load analysis of sit-to-stand, stand-to-sit, cane walking and lifting. Clinical biomechanics 15, 441–448 (2000). 37. Bell, A. L., Pedersen, D. R. & Brand, R. A. A comparison of the accuracy of several hip center location prediction methods. Jou of biomechanics 23, 617–621 (1990). f ( ) 8. Davis, R. B., Ounpuu, S., Tyburski, D. & Gage, J. R. A gait analysis data collection and reduction technique. Hum. Mov. Sci. 10 575–587 (1991).h h y f y 40. Zatsiorsky, V. M., Seluyanov, V. N. & Chugunova, L. G. Methods of determining mass-inertial characteristics of human body segments in Contemporary Problems of Biomechanics (eds Chernyi, G. G. & Regirer, S. A.) 272–291 (CRC Press, Massachusetts, USA, 1990). 41. de Leva, P. Adjustments to Zatsiorsky-Seluyanov’s segment inertia parameters. Journal of biomechanics 29, 1223–1230 (1996). h 40. Zatsiorsky, V. M., Seluyanov, V. N. & Chugunova, L. G. Methods of determining mass-inertial characteristics of human body segments in Contemporary Problems of Biomechanics (eds Chernyi, G. G. & Regirer, S. A.) 272–291 (CRC Press, Massachusetts, USA, 1990). Acknowledgementsh g The authors are grateful to the subjects who enrolled in the experiments. In addition, they would like to thank Guido Pasquini for his clinical support, Federica Vannetti for her support during the experimentation, Francesco Giovacchini for his support with APO mechanics tuning during the experimentations. Subplot A in Figure 1 was created by Arch. AlessioTommasetti, D’Arc. Studio Architects Associates. This work was partly supported by: the EU Commission through the FP7 project CYBERLEGs (The CYBERnetic LowEr-Limb CoGnitive Ortho- prosthesis, Grant Agreement no. 287894), and the H2020 project CYBERLEGs Plus Plus (The CYBERnetic LowEr-Limb CoGnitive Ortho-prosthesis Plus Plus, Grant Agreement no. 731931), the private foundation “Fondanzione Pisa” through the project IUVO (“The wearable cognitive robot for the functional assistance of lower-limb motion”), the Italian Ministry of Education and Research (MIUR) through the relevant interest national project (PRIN) “A quantitative and multi factorial approach for estimation and preventing the risk of falls in the elderly people”, the Bertarelli Foundation, the Swiss National Centre of Competence in Research (NCCR) Robotics, and the institutional funds from The BioRobotics Institute, Scuola Superiore Sant’Anna, and IRCCS Don Carlo Gnocchi Foundation. Author Contributions V.M. designed and supervised the experiments and wrote the paper. P.T., F.A. and D.M. performed the experiments, processed the data, and wrote the paper. A.P. and M.C. developed and programmed the APO R.M.L. was responsible for the clinical aspects of the study. N.V. designed the experiments and developed the APO, and S.M. designed and supervised the experiments and wrote the paper. www.nature.com/scientificreports/ 1. Yang, F., Bhatt, T. & Pai, Y. C. Limits of recovery against slip-induced falls while walking. Journal of biomechanics 44, 2607–2613, doi 10.1016/j.jbiomech.2011.08.018 (2011). j j 2. Hak, L. et al. Stepping strategies used by post-stroke individuals to maintain margins of stability during walking. Clinica biomechanics 28, 1041–1048, doi: 10.1016/j.clinbiomech.2013.10.010 (2013).h j 3. Hof, A. L., Gazendam, M. G. & Sinke, W. E. The condition for dynamic stability. Journal of biomechanics 38, 1–8, doi: 10.1016/j jbiomech.2004.03.025 (2005).f j 4. Monaco, V., Rinaldi, L. A., Macri, G. & Micera, S. During walking elders increase efforts at proximal joints and keep low kinetics a the ankle. Clinical biomechanics 24, 493–498 (2009). 25. Segal, A. D. et al. Kinematic and kinetic comparisons of transfemoral amputee gait using C-Leg and Mauch SNS prosthetic knees. Journal of rehabilitation research and development 43, 857–870 (2006).hf g Journal of rehabilitation research and development 43, 857–870 (2 f p 6. Martelli, D. et al. The effects on biomechanics of walking and balance recovery in a novel pelvis exoskeleton during zero-torque control. Robotica 32, 1317–1330, doi: 10.1017/S0263574714001568 (2014).i 7. Becker, C. et al. Proposal for a multiphase fall model based on real-world fall recordings with body-fixed sensors. Z Gerontol Geriat 45, 707–715, doi: 10.1007/s00391-012-0403-6 (2012).f 28. Yang, F., Bhatt, T. & Pai, Y. C. Role of stability and limb support in recovery against a fall following a novel slip induced in different daily activities. Journal of biomechanics 42, 1903–1908, doi: 10.1016/j.jbiomech.2009.05.009 (2009). 29. Aprigliano, F., Martelli, D., Micera, S. & Monaco, V. Intersegmental coordination elicited by unexpected multidirectional slipping- like perturbations resembles that adopted during steady locomotion. Journal of neurophysiology 115, 728–740, doi: 10.1152/ jn.00327.2015 (2016). j 30. Martelli, D., Monaco, V., Bassi Luciani, L. & Micera, S. Angular momentum during unexpected multidirectional perturbations delivered while walking. IEEE Trans Biomed Eng 60, 1785–1795 (2013). g g , ( ) 31. Moyer, B. E., Redfern, M. S. & Cham, R. Biomechanics of trailing leg response to slipping - evidence coordination. Gait & posture 29, 565–570, doi: 10.1016/j.gaitpost.2008.12.012 (2009). g g ( ) 31. Moyer, B. E., Redfern, M. S. & Cham, R. Biomechanics of trailing leg response to slipping - evidence of interlimb and intra coordination. Gait & posture 29, 565–570, doi: 10.1016/j.gaitpost.2008.12.012 (2009). Biomechanics and Motor Control of Human Movement. (John Wiley 32. Winter, D. A. Biomechanics and Motor Control of Human Movement. References Safety, energy efficiency, and cost efficacy of the C-Leg for transfemoral Prosthet Orthot Int 34, 362–377, doi: 10.3109/03093646.2010.520054 (2010).ff 12. Wolff, J., Parker, C., Borisoff, J., Mortenson, W. B. & Mattie, J. A survey of stakeholder perspectives on exoskeleton technology. J Neuroeng Rehabil 11, 169, doi: 10.1186/1743-0003-11-169 (2014). g 13. Young, A. & Ferris, D. State-of-the-art and Future Directions for Robotic Lower Limb Exoskeletons. IEEE Trans Neural Syst Rehabil Eng, doi: 10.1109/TNSRE.2016.2521160 (2016).i 14. Wong, C. K., Rheinstein, J. & Stern, M. A. Benefits for Adults with Transfemoral Amputations and Peripheral Artery Disease Using Microprocessor Compared with Nonmicroprocessor Prosthetic Knees. American journal of physical medicine & rehabilitation / Association of Academic Physiatrists 94, 804–810, doi: 10.1097/PHM.0000000000000265 (2015).h f y 5. Yang, J. et al. The reaction strategy of lower extremity muscles when slips occur to individuals with trans-femoral amputation. Electromyogr Kinesiol 17, 228–240 (2007). 16. Tropea, P. et al. Detecting Slipping-Like Perturbations by Using Adaptive Oscillators. Annals of biomedical engineering, doi: 10.1007/ s10439-014-1175-5 (2014). 17. Giovacchini, F. et al. A light-weight active orthosis for hip movement assistance. Robotics and Autonomous Systems 73, 123–134, doi: http://dx.doi.org/10.1016/j.robot.2014.08.015 (2015). p g j 8. Bassi Luciani, L. et al. Design and evaluation of a new mechatronic platform for assessment and prevention of fall risks. J Neuroeng Rehabil 9, 51 (2012). 19. Pai, Y. C. & Patton, J. Center of mass velocity-position predictions for balance control. Journal of biomechanics 30, 347–354 (1997). 19. Pai, Y. C. & Patton, J. Center of mass velocity-position predictions for balance control. Journal of biomechanics 30, 347–354 (1997). 20. Yang, F., Anderson, F. C. & Pai, Y. C. Predicted threshold against backward balance loss following a slip in gait. Journal of biomechanics 41, 1823–1831, doi: 10.1016/j.jbiomech.2008.04.005 (2008). 19. Pai, Y. C. & Patton, J. Center of mass velocity-position predictions for balance control. Journal of biomechanics 30, 347–354 (1997). 20. Yang, F., Anderson, F. C. & Pai, Y. C. Predicted threshold against backward balance loss following a slip in gait. Journal of 19. Pai, Y. C. & Patton, J. Center of mass velocity-position predictio d d d h h ld SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 8 www.nature.com/scientificreports/ SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 Additional Information Supplementary information accompanies this paper at http://www.nature.com/srep Competing Interests: N.V. and S.M. have financial interests in the company IUVO SRL, which will bring on the market an engineered version of the APO; N.V. is a co-inventor of the torsional spring adopted in the APO actuation unit. Competing Interests: N.V. and S.M. have financial interests in the company IUVO SRL, which will bring on the market an engineered version of the APO; N.V. is a co-inventor of the torsional spring adopted in the APO actuation unit. How to cite this article: Monaco, V. et al. An ecologically-controlled exoskeleton can improve balance recov after slippage. Sci. Rep. 7, 46721; doi: 10.1038/srep46721 (2017). How to cite this article: Monaco, V. et al. An ecologically-controlled exoskeleton can improve balance recovery after slippage. Sci. Rep. 7, 46721; doi: 10.1038/srep46721 (2017). How to cite this article: Monaco, V. et al. An ecologically-controlled exoskeleton can improve balance recovery fter slippage. Sci. Rep. 7, 46721; doi: 10.1038/srep46721 (2017). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps an institutional affiliations. SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 9 www.nature.com/scientificreports/ This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ © The Author(s) 2017 www.nature.com/scientificreports/ This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ © The Author(s) 2017 SCIENTIFIC REPOrTS \| 7:46721 \| DOI: 10.1038/srep46721 10
https://openalex.org/W2596757643	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0173280&type=printable	English	null	Measuring complexity in Brazilian economic crises	PloS one	2,017	cc-by	4,552	Letı´cia P. D. Mortoza, Jose´ R. C. Piqueira* Laborato´rio de Automac¸ãoe Controle/Escola Polite´cnica, Universidade de São Paulo, Avenida Prof. Luciano Gualberto, travessa 3 - 158, São Paulo-Brazil * piqueira@lac.usp.br * piqueira@lac.usp.br a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Mortoza LPD, Piqueira JRC (2017) Measuring complexity in Brazilian economic crises. PLoS ONE 12(3): e0173280. https://doi.org/ 10.1371/journal.pone.0173280 Editor: Boris Podobnik, University of Rijeka, CROATIA Editor: Boris Podobnik, University of Rijeka, CROATIA Received: October 28, 2016 Accepted: February 4, 2017 Published: March 16, 2017 Copyright: © 2017 Mortoza, Piqueira. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: © 2017 Mortoza, Piqueira. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Abstract Capital flows are responsible for a strong influence on the foreign exchange rates and stock prices macroeconomic parameters. In volatile economies, capital flows can change due to several types of social, political and economic events, provoking oscillations on these parameters, which are recognized as economic crises. This work aims to investigate how these two macroeconomic variables are related with crisis events by using the traditional complex measures due to Lopez-Mancini-Calbet (LMC) and to Shiner-Davison-Landsberg (SDL), that can be applied to any temporal series. Here, Ibovespa (Bovespa Stock Exchange main Index) and the “dollar-real” parity are the background for calculating the LMC and SDL complexity measures. By analyzing the temporal evolution of these mea- sures, it is shown that they might be related to important events that occurred in the Brazilian economy. OPEN ACCESS Citation: Mortoza LPD, Piqueira JRC (2017) Measuring complexity in Brazilian economic crises. PLoS ONE 12(3): e0173280. https://doi.org/ 10.1371/journal.pone.0173280 Measuring complexity in Brazilian economic crises Letı´cia P. D. Mortoza, Jose´ R. C. Piqueira* Complexity and Brazilian crises Additionally, Shiner, Davison and Landsberg, by slightly changing the LMC definition, pro- posed the SDL measure [13] and both, LMC and SDL, are satisfactory to measure complexity for many problems [14]. Concerning economy, these measures are barely used [15] although they could be useful to model crises by the analysis of their temporal evolutions [16], as it was shown in former work; moreover they could be used as an alternative to the analysis of the tails of statistical distribu- tions [8, 17]. Here, slightly modified versions of the LMC and SDL measures are proposed for temporal series. Then, considering that capital fluxes are determinant to the economic health of an emerging country [18], oscillating drastically due to several political, social and economic changes, Brazilian macro-economical data are analyzed by calculating the LMC and SDL com- plexity measures, aiming to identify how their behavior is connected to crisis events. The chosen data are Ibovespa (Bovespa Stock Exchange main Index) and the “dollar-real” parity, generating the LMC and SDL complexity measures temporal series that are studied in five critical periods of the Brazilian economy. Since the end of dictatorship regime in 1985, Brazil has been alternating short cycles of economical phases, with macroeconomics indicators oscillating, apparently commanded by the political changes. To delineate the Brazilian political events, the analysis starts with Jose´ Sarney’s term, between 1985 and 1990, which is considered the transition period between dictatorship and democracy and in which several stabilization plans were practiced, such as freezing prices, causing an increase in demand and a strong monetary black market. In 1990, Fernando Collor de Mello was democratically elected and applied the “Collor” plan, radically freezing the finan- cial assets, creating the “commercial-dollar”, a parallel to the existing fluctuating rates. Latter, the “Collor” plan-II was enforced based on freezing prices [17]. The Collor plans promoted important changes in the Brazilian economy, mainly opening the market to foreign capital and privatizing important service companies formerly controlled by the state. Corruption suspicions and a bad economic scenario caused the impeachment of the president (Fernando Collor de Mello) who was replaced in September, 1992. A new stabili- zation plan was designed (“Real” plan), and started to be applied in 1993, with a new currency, linked to the dollar quotation and implying a strong inflation decrease [17]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 1 Introduction For the last forty years, complex thinking has influenced many research areas [1] contributing to model problems related to the onset of surprising or catastrophic events such as traffic jams, crowd behavior, tsunamis and earthquakes [2]. Data Availability Statement: Data are available from Ibovespa at http://www.bmfbovespa.com.br/ pt_br/produtos/indices/indices-amplos/. As the site is opened, click on the Ibovespa window and go to “Serie Retroativa do Ibovespa. The exchange rate can be viewed at http://www4.bcb.gov.br/pec/ taxas/port/ptaxnpesq.asp?id=txcotacao. As the site is opened, choose period and choose “dolar dos EUA”. Consequently, the concept of complexity has been largely debated [3], with generality implying the prevalence of qualitative arguments [4]. In this scene, concerning to computer science, Shannon [5] and Kolmogorov [6] proposed similar and useful complexity measures. In terms of economic applications, fractal dimension measures were proposed by Mandel- brot [7], which are largely used to characterize economic time series [8], including different fractal measures adaptable to different patterns of series [9]. Another well spread line of reasoning applied to economic problems is the Dynamical Sys- tem’s approach, aiming to detect order and chaos in processes [10]. Although the common sense is that complexity is a relative concept [11], Lo´pez-Ruiz, Man- cini and Calbet proposed a global measure for complexity by using the entropy definition, which is referred to as the LMC measure [12]. Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. 1 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises presentation of the results. A discussion section presents some conjectures regarding the methodology and the Brazilian crisis events followed by a brief conclusion. PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 2 Crisis events This analysis starts in January 1999, when Brazil finally abandoned the crawling peg exchange rate regime and adopted floating rates. In this period the Brazilian currency, which was already suffering a huge devaluation pressure, depreciated drastically (49.51% from January to March, 1999). In contrast, Ibovespa increased 59.34% in the same period [19]. This foreign exchange crisis is the first event analyzed in this study. The second crisis event is the Twin Towers attack, in September 2001, when terrorist acts lead to the complete destruction of the World Trade Center in New York and partial destruc- tion of the Pentagon, in Washington. Although it had a short term effect, it caused the Ameri- can Stock Exchanges closure during one week and raised the investors risk aversion, causing the removal of investment money from the emerging markets. The third event is specially related to investor confidence. By the middle of 2002, Luiz Ina- cio Lula da Silva, main leader of Brazilian Labour’s Party, appeared in the polls as the favorite candidate for the Presidental election. The uncertainty about economic policies lead to a dra- matic capital outflow, causing the Brazilian currency to reach its biggest devaluation since it started floating in 1999. The fourth event is the world’s well known sub-prime crisis. It started with sub-prime mort- gages defaults—hence its name—in 2007 and in one year, it contaminated global economy. The peak of this crisis happened in September 2008, when the Federal Reserve surprisingly refused to rescue the Lehman Brothers, triggering a historical liquidity crisis in the interbank money market. Consequently, global credit offer disappeared consequently and the whole World entered into a historic economic recession. It was the biggest global financial crisis since 1930. The fifth event is the reelection presidential campaign of Dilma Roussef, opposing the desires of the Brazilian elite radically against all the social programs, which started during her first term. The data related to the recent impeachment process are not analyzed here, but it may be insightful in the future. The main objective is to investigate how the LMC and SDL complexity measures change in the two time series representing Brazilian economy: dollar/real ratio and Ibovespa index. The observation of the dynamics of the complexity measures evidences how crisis events give sig- nals to occur. The exchange policy was pressed and, in 1999, the Brazilian currency was finally allowed to have free fluctuations, with a strong initial devaluation. The data sampling presented here starts from this period on, considering that the economic policy did not suffer any additional change. Between 1999 and 2003, liberal ideas, mainly represented by the new president Fernando Henrique Cardoso, governed the country and a satisfactory economic equilibrium was reached; however, with timid social actions. Supported by ideas regarding social justice and the democratic distribution of gains, the Labor Party, represented by Luiz Inacio Lula da Silva was elected in 2003, governing up to 2011. This period was characterized by optimism and, consequently, the Labor Party, represented by Dilma Roussef, was elected again in 2011 and in 2016. During Dilma’s second term, many corruption scandals erupted provoking her impeachment in October of 2016. The new period presented no political novelty, with important names of the Senate and the House of Represen- tatives supporting the impeachment also being involved in corruption. Aiming to verify how political events changed the Brazilian economical life, between 1999 and 2015, this article is divided into five sections, including this introduction. The next sec- tion describes the main financial, political and social crisis events that influenced the Brazil- ian economy in the recent past and its macroeconomics effects. Furthermore, the acquisition of the two time series and the calculation methodology are explained, followed by the 2 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 3 Data and methodology 3.1 Data The data collected are constituted of two different time series: the dollar-real exchange rates and the Ibovespa Index, both expressed on a daily basis. The Brazilian Real / American Dollar exchange rate is compiled by the Brazilian Central Bank and it is available on the INTERNET by accessing (http://www4.bcb.gov.br/pec/taxas/port/ptaxnpesq.asp?id=txcotacao), choosing the period and the option “dolar EUA”. This series is composed of 4,109 data points, from January 1999 to August 2015. The aver- age rate of this period was 2.2017 BRL/USD, with a 0.4811 standard deviation. The minimum value of 1.2071 BRL/USD occurred in January 1999 and the series reached its maximum of 3.9544 BRL/USD during the elections in October 2002. This time series is plotted in Fig 1, where five local maxima can be spotted: March 1999, September 2001, October 2002, October 2008 and January 2015, which are indicated by 3 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises Fig 1. Temporal series: USD/BRL and Ibovespa. https://doi.org/10.1371/journal.pone.0173280.g001 Fig 1. Temporal series: USD/BRL and Ibovespa. https://doi.org/10.1371/journal.pone.0173280.g001 https://doi.org/10.1371/journal.pone.0173280.g001 arrows. These peaks correspond to the 1999 Exchange Rate Crisis, the Twin Tower Crisis, 2002 Election Crisis, Sub-Prime Crisis and the reelection of Dilma Roussef reelection, respectively. On the other hand, Ibovespa is the main Bovespa (Brazilian Stock Exchange) index. It is compiled as a weighted average of a theoretical portfolio of stocks. It is designed to gauge the average performance of the stock market by tracking changes in the prices of the more actively traded and the better representative stocks of the Brazilian stock market. Ibovespa is a total return index and it is expressed in points. It was first compiled in January 1968, when it was represented by 100 points. As to the foreign exchange rates, 4,109 daily data points, from January 1999 to August 2015, compose this time-series, available on the INTERNET (http://www.bmfbovespa.com.br/pt-br/ produtos/indices/indices-amplos/) where the window “Ibovespa” must be opened and “Serie retroativa Ibovespa” must be selected. The average for this period is of 38,140 points, with a standard deviation of 20,720. The series minimum was of 5,057 points in the beginning of the studied period, January 1999; and the maximum of 73,516 points happened in May 2008, months before the peak of the Sub-Prime Crisis. PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 3.2 Methodology To calculate the LMC and SDL complexity measures for a temporal series, it is necessary to define the instantaneous order Δ analogously to the definition of disorder based on Shannon entropy [12, 13], demanding the division of the range of the variables into intervals, to calcu- late the instantaneous probability associated to the measured value. First both series were divided into N = 16 discrete parts, with a 0.175 BRL/USD amplitude for the foreign exchange rate and 4,279 stock points for the Ibovespa. The relative frequency of occurrence of each interval is considered to be its individual probability (pi) and the number N was chosen after several numerical experiments optimizing accuracy and computational time. By using the calculated probabilities, the instantaneous disorder, corresponding to the ther- modynamical equilibrium [12, 13], for the interval Hi is given by: First both series were divided into N = 16 discrete parts, with a 0.175 BRL/USD amplitude for the foreign exchange rate and 4,279 stock points for the Ibovespa. The relative frequency of occurrence of each interval is considered to be its individual probability (pi) and the number N was chosen after several numerical experiments optimizing accuracy and computational time. p y (pi) was chosen after several numerical experiments optimizing accuracy and computational time. 3 Data and methodology 3.1 Data The Ibovespa time series is also plotted in Fig 1, where its upward trend can be observed, from the beginning of the period until 2008, when it reached its maxima. Following, a huge depression was caused by the Sub-Prime Crisis and the recovery just happened by the end of 2009 and the beginning of 2010. Apart from this trend, the local minima can be spotted around August 1999, September 2001, October 2002, October 2008 and January 2015, which coincides with the crisis events studied here. 4 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises https://doi.org/10.1371/journal.pone.0173280.t001 https://doi.org/10.1371/journal.pone.0173280.t001 https://doi.org/10.1371/journal.pone.0173280.t001 3.2 Methodology By using the calculated probabilities, the instantaneous disorder, corresponding to the ther- modynamical equilibrium [12, 13], for the interval Hi is given by: By using the calculated probabilities, the instantaneous disorder, corresponding to the ther- modynamical equilibrium [12, 13], for the interval Hi is given by: Di ¼ pi ðlog2piÞ log2N ; ð1Þ ð1Þ and considering the order term (1 −Δi), corresponding to the thermodynamical disequilib- rium [13], the instantaneous SDL complexity measure is given by: and considering the order term (1 −Δi), corresponding to the thermodynamical disequilib- rium [13], the instantaneous SDL complexity measure is given by: ðSDLÞi ¼ Dið1 DiÞ: ð2Þ ð2Þ As proposed in [12], the LMC complexity is calculated by replacing the order term by a dis- equilibrium expression [12], given by; As proposed in [12], the LMC complexity is calculated by replacing the order term by a dis- equilibrium expression [12], given by; Di ¼ ðpi 1 NÞ 2; ð3Þ ð3Þ consequently, the instantaneous LMC is given by: consequently, the instantaneous LMC is given by: ðLMCÞi ¼ DiDi: ð4Þ ðLMCÞi ¼ DiDi: ð4Þ ð4Þ The calculations described were performed considering the data represented in Fig 1 and are summarized in Table 1 for the Dollar-Real ratio and in Table 2 for the Ibovespa index. The calculations described were performed considering the data represented in Fig 1 and are summarized in Table 1 for the Dollar-Real ratio and in Table 2 for the Ibovespa index. Table 1. Dollar-real ratio. Table 1. Dollar-real ratio. Table 1. Dollar-real ratio. Interval Min Max Occurrence Probability (%) SDL LMC H1 1.2071 1.3788 9 0.22 0.00008 0.00000 H2 1.3788 1.5505 4 0.10 0.00004 0.00000 H3 1.5505 1.7222 440 10.71 0.00264 0.00001 H4 1.7222 1.8939 866 21.08 0.00469 0.00002 H5 1.8939 2.0656 571 13.90 0.00329 0.00001 H6 2.0656 2.2373 525 12.78 0.00306 0.00001 H7 2.2373 2.4090 554 13.49 0.00320 0.00001 H8 2.4090 2.5808 220 5.36 0.00145 0.00001 H9 2.5808 2.7525 173 4.21 0.00118 0.00000 H10 2.7525 2.9242 269 6.55 0.00173 0.00001 H11 2.9242 3.0959 204 4.97 0.00136 0.00001 H12 3.0959 3.2676 123 2.99 0.00088 0.00000 H13 3.2676 3.4393 37 0.90 0.00030 0.00000 H14 3.4393 3.6110 65 1.58 0.00050 0.00000 H15 3.6110 3.7827 30 0.73 0.00025 0.00000 H16 3.7827 3.9544 18 0.44 0.00005 0.00000 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 5 / 12 Complexity and Brazilian crises Table 2. Ibovespa index. Interval Min Max Occurrence Probability (%) SDL LMC H1 5,057.19 9,335.92 66 1.61 0.00051 0.00000 H2 9,335.92 13,614.64 603 14.68 0.00344 0.00002 H3 13,614.64 17,893.37 496 12.07 0.00292 0.00001 H4 17,893.37 22,172.10 1814 4.41 0.00123 0.00000 H5 22,172.10 26,450.82 255 6.21 0.00165 0.00000 H6 26,450.82 30,729.55 92 2.24 0.00068 0.00000 H7 30,729.55 35,008.27 77 1.87 0.00058 0.00000 H8 35,008.27 39,287.00 227 5.53 0.00149 0.00000 H9 39,287.00 43,565.73 123 2.99 0.00088 0.00000 H10 43,565.73 47,844.45 141 3.43 0.00099 0.00000 H11 47,844.45 52,123.18 313 7.62 0.00197 0.00000 H12 52,123.18 56,401.91 486 11.83 0.00287 0.00001 H13 56,401.91 60,680.63 344 8.37 0.00214 0.00000 H14 60,680.63 64,959.36 308 7.50 0.00194 0.00000 H15 64,959.36 69,238.08 273 6.65 0.00175 0.00000 H16 69,238.08 73,516.81 1237 2.9 0.00088 0.00000 https //doi org/10 1371/jo rnal pone 0173280 t002 Table 2. Ibovespa index. https://doi.org/10.1371/journal.pone.0173280.t002 4.2 Changing currency policies To investigate the described periods of crisis, a zoom is provided for each one, starting with the changes in the Brazilian exchanging regime, shown in Fig 3. Ibovespa index suffer a strong change from a low to a high value in March-99, a retarded effect from governmental economic actions over currency regulations. The effects over the measures for dollar-real ratio are more sensitive and the complexity measure oscillates during January, February and March of 1999. Complexity and Brazilian crises Another interesting point highlighted is the presence of higher frequency oscillation for the complexity measures coinciding with the five crisis events described in section 2. 4.1 The whole picture After calculating the LMC and SDL complexity measure for each interval in each series, the data points were placed back into their original sequence to construct the complexity measures time evolution for each variable. Finally both complexity series were combined permitting a visual comparison for each measure (Fig 2a and 2b), with arrows indicating the crises to be studied here. The figures indicate that the temporal series for the two complexity measures (LMC and SDL) present the same qualitative behavior. Then, for the sake of simplicity, from this point on, only the SDL results are presented. Fig 2. Complexity measures temporal evolutions. (a) LMC complexity measure. (b) SDL complexity measure. https://doi.org/10.1371/journal.pone.0173280.g002 Fig 2. Complexity measures temporal evolutions. (a) LMC complexity measure. (b) SDL complexity measure. Fig 2. Complexity measures temporal evolutions. (a) LMC complexity measure. (b) SDL complexity measure. https://doi org/10 1371/journal pone 0173280 g002 https://doi.org/10.1371/journal.pone.0173280.g002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 6 / 12 4.5 Sub-prime crisis Paradoxically, the sub-prime crisis affected the Brazilian currency more than the Brazilian stock market, as shown in Fig 6. During this period, the complexity measures related to Ibovespa index suffered small oscil- lations, maintaining low levels. Concerning the dollar/real ratio complexity, it became high with strong oscillations. 4.4 Lula’s campaign The Ibovespa index shows to be more sensitive to Lula’s campaign than the dollar/real ratio, as shown in Fig 5. The Ibovespa index complexity oscillated between high and low levels all over the campaign and remained high after the election results. The dollar/real ratio oscillated in a lower level, maintaining a practically constant low level complexity. 4.3 Twin towers attack When the September 11th attack occurred, Brazilian stock market suffered very strong effects, as it can be deduced from observing Fig 4, which shows how the complexity measures were affected in this period. Ibovespa index complex measure remained high and constant in this period, indicating that the closing of American stock agencies produced important uncertainties in the Brazilian stock market. Concerning the dollar-real ratio, the complexity remained constant, but at lower level than of the Ibovespa index. Fig 3. The SDL complexity measure during changes in the Brazilian exchanging regime. Fig 3. The SDL complexity measure during changes in the Brazilian exchanging regime. https://doi.org/10.1371/journal.pone.0173280.g003 Fig 3. The SDL complexity measure during changes in the Brazilian exchanging regime. https://doi.org/10.1371/journal.pone.0173280.g003 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 7 / 12 Complexity and Brazilian crises Fig 4. The SDL complexity measure after September 11th. https://doi.org/10.1371/journal.pone.0173280.g004 Fig 4. The SDL complexity measure after September 11th. https://doi.org/10.1371/journal.pone.0173280.g004 Fig 4. The SDL complexity measure after September 11th. Fig 4. The SDL complexity measure after September 11th. https://doi.org/10.1371/journal.pone.0173280.g004 Fig 4. The SDL complexity measure after September 11th. https://doi.org/10.1371/journal.pone.0173280.g004 https://doi.org/10.1371/journal.pone.0173280.g004 5.1 Methodological issues The use of LMC and SDL complexity measures seem to show robust results, indicating that both are adequate to interpret macro-economic data. Additionally, either LMC or SDL might provide important qualitative hints about crisis events, being equivalent for data analysis. Apparently, stock market indexes are more sensitive to the complexity calculations with results providing more evidence when crisis events occur. Consequently, to maintain a real time diagnostic of the complexity measure of these indexes can be an important diagnostic economic tool. The complexity of currency indexes has given some diagnostics, but the robustness of the results is lower than the stock market complexity results with too many apparently spurious oscillations. 5 Discussion LMC and SDL complexity measures temporal series were defined for macro-economics vari- ables. The definitions, applied to Brazilian economic data, have given some conjectures that can be divided into two groups: general methodological issues and particular diagnostic issues. 4.6 Dilma’s reelection During this period, both Ibovespa index and Dollar/Real ratio had their complexities strongly affected, as shown in Fig 7. The measures presented high levels and oscillated atypically during this period. 8 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises Fig 5. The SDL complexity measure during Lula’s campaign. https://doi.org/10.1371/journal.pone.0173280.g005 Fig 5. The SDL complexity measure during Lula’s campaign. https://doi.org/10.1371/journal.pone.0173280.g005 Fig 5. The SDL complexity measure during Lula’s campaign. Fig 5. The SDL complexity measure during Lula’s campaign. https://doi.org/10.1371/journal.pone.0173280.g005 Fig 5. The SDL complexity measure during Lula’s campaign. https://doi.org/10.1371/journal.pone.0173280.g005 https://doi.org/10.1371/journal.pone.0173280.g005 https://doi.org/10.1371/journal.pone.0173280.g005 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Acknowledgments This work was revised by Luisa P. Franzen, from the EPUSP revision department. 6 Conclusion The LMC and SDL complexity measures provided some hints regarding Brazilian economy behavior. Although the work is exploratory, it can be useful to new analysis of economic data, attempting to explain how social and political events affect the economy. 5.2 Brazilian crisis events As Brazil is an important emerging economy, analyzing its data can give some hints regarding the social and political state of the country, as it was shown here. 9 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises Fig 6. The SDL complexity measure during sub-prime crisis. https://doi.org/10.1371/journal.pone.0173280.g006 Fig 6. The SDL complexity measure during sub-prime crisis. https://doi.org/10.1371/journal.pone.0173280.g006 Fig 6. The SDL complexity measure during sub-prime crisis. https://doi.org/10.1371/journal.pone.0173280.g006 For instance, the exchanging currency rule changes in 1999 produced an important insta- bility of the stock market, measured by the high complexity of the Bovespa index, decreasing only in several months. The twin-towers attack effects appeared in the high values of the Ibovespa index complexity, which remained high for several months after the event. Lula’s campaign and election caused high complexity for the Ibovespa index, which remained for about 6 months after the elections, when the market understood that Lula was following a liberal policy. The astonishing fact is that the sub-prime crisis has not strongly affected the Brazilian stock market, as shown by Ibovespa index complexity. The main motive is that the Brazilian market does not have complex derivative products. The effect appeared in the dollar-real ratio, proba- bly because Brazilians suffer from a strong risk aversion. Finally, when President Dilma Roussef initiated her reelection campaign, the Brazilian capi- tal owners became apprehensive and started to declare that social actions were conducting the economy to bankruptcy. The newspapers and media owners agreed and a strong counter cam- paign spread throughout the country. Regardless, Dilma Roussef won the elections, although the counter campaign continued, as it is evidenced by the complexity measures. Social scientists may tell the history in the next decades, with complexity measures helping in the analysis. 10 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises Fig 7. The SDL complexity measure during Dilma’s reelection. https://doi org/10 1371/journal pone 0173280 g007 Fig 7. The SDL complexity measure during Dilma’s reelection. https://doi.org/10.1371/journal.pone.0173280.g007 Fig 7. The SDL complexity measure during Dilma’s reelection. https://doi.org/10.1371/journal.pone.0173280.g007 https://doi.org/10.1371/journal.pone.0173280.g007 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Author Contributions Conceptualization: JRCP LPDM. PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Conceptualization: JRCP LPDM. Data curation: LPDM JRCP. Formal analysis: JRCP LPDM. Formal analysis: JRCP LPDM. Funding acquisition: LPDM. Funding acquisition: LPDM. Funding acquisition: LPDM. Investigation: LPDM. Investigation: LPDM. Methodology: JRCP. Methodology: JRCP. 11 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0173280 March 16, 2017 Complexity and Brazilian crises Software: JRCP. Supervision: LPDM JRCP. Validation: LPDM. Visualization: JRCP. Writing – original draft: JRCP LPDM. Writing – review & editing: JRCP LPDM. References 1. Morin E. Introduccio´n al pensamiento complejo, (2011), Gedisa: Spain. 2. Bunde A. Kropp J, Schellnhuber HJ. The Science of Disasters (2002), Springer-Verlag: Berlin- Germany. 3. E´ rdi P. Complexity Explained (2008), Springer-Verlag: Berlin-Germany. 4. Waldrop MM. Complexity: the emerging science at the edge of order and chaos (1992), Simon & Schuster Paperbacks: New York-USA. 5. Shannon CE, Weaver W. The Mathematical Theory of Communication, (1963), Illini Books Edition: Urbana and Chicago, USA. 6. Kolmogorov AN. “Three approaches to the definition of the concept “quantity of information””, Problemy Peredachi Informatsii, 1 (1965) pp. 3–11. 7. Chapman T. “Father of fractal complexity”, Quantitative Finance, 3(5) (2003) pp. C88–C90. https://doi. org/10.1088/1469-7688/3/5/601 8. Mandelbrot BB. “New Methods of Statistical Economics, Revisited: Short versus Long Tails and Gauss- ian versus Power-Law Distributions”, Complexity, 14(3) (2008) pp. 55–65. https://doi.org/10.1002/cplx. 20264 9. Rani TGGE, Jayalalitha G. “Complex Patterns in Financial Time Series through Higuchi’s Fractal Dimension”, Fractals, 24(4) (2016) 1650048 (7 pages). https://doi.org/10.1142/S0218348X16500481 10. Liu T, Granger CWJ, Heller WP. “Using the Correlation Exponent to Decide Whether an Economic Series is Chaotic”, Journal of Applied Econometrics, 7, (1992) pp. S25–S39. https://doi.org/10.1002/ jae.3950070504 11. Piqueira JRC. “A mathematical view of Biological complexity”, Communications in Nonlinear Science and Numerical Simulation, 14(6) (2009) pp. 2581–2586. https://doi.org/10.1016/j.cnsns.2008.10.003 12. Lo´pez-Ruiz R, Mancini HL, Calbet X. “A statistical measure of complexity”, Physics Letters A, 209 (1995) pp. 321–326. https://doi.org/10.1016/0375-9601(95)00867-5 13. Shiner J, Davison M, Landsberg P. “Simple measure for complexity”, Physical Review E, 59(2) (1999) pp. 1459–1464. https://doi.org/10.1103/PhysRevE.59.1459 14. Northrop RB. Introduction to complexity and complex systems, (2010), CRC Press: New Jersey, USA 15. Arthur WB. Complexity and the Economy, (2015), Oxford University Press: New York, USA. 16. Tenreiro Machado JA. “Complex dynamics of financial indeces”, Nonlinear Dynamics, 74 (2013) pp. 287–296. https://doi.org/10.1007/s11071-013-0965-x 17. Piqueira JRC, Mortoza LPD “Brazilian exchange rate complexity: Financial crisis effects”, Communica- tions in Nonlinear Science and Numerical Simulation, 17, (2012) pp. 1690–1695. https://doi.org/10. 1016/j.cnsns.2011.08.031 18. Boudias R. “Capital inflows, exchange rate regimes and credit dynamics in emerging market econo- mies”, International Economics, 143 (2015) pp. 80–97. https://doi.org/10.1016/j.inteco.2015.05.001 19. Tabak B. “The Dynamic Relationship Between Stock Prices and Exchange Rates: Evidence for Brazil”, Banco Central do Brasil Working Paper Series, 124 (2006) pp.1–37. 12 / 12
https://openalex.org/W3188989487	https://www.scielo.br/j/ciedu/a/BNMX8ZWcRQ949x7wRd4sbJq/?lang=es&format=pdf, http://old.scielo.br/pdf/ciedu/v27/1516-7313-ciedu-27-e21028.pdf, https://www.redalyc.org/journal/2510/251066798025/251066798025.pdf, https://dialnet.unirioja.es/descarga/articulo/8052366.pdf	es		Nominalización en Procesos de Condensación de Significados en la Enseñanza de Solución Saturada: un estudio de caso en una escuela secundaria	Ciência & Educação	2,021	cc-by	8,255	1 de 17 Ciência & Educação, Bauru, v. 27, e21028, 2021 https://doi.org/10.1590/1516-731320210028 ARTIGO ORIGINAL Nominalización en Procesos de Condensación de Significados en la Enseñanza de Solución Saturada: un estudio de caso en una escuela secundaria Nominalization in Processes of Condensation of Meanings in the Teaching of Saturated Solution: a case study in a secondary school Guillermo Eduardo Cutrera1 Marta Beatriz Massa2 Silvia Stipcich3 1Universidad Nacional de Mar del Plata, Facultad de Ciencias Exactas y Naturales, Departamento de Educación Científica, Mar del Plata, Buenos Aires, Argentina. Autor correspondiente: gcutrera@mdp.edu.ar 2Universidad Nacional de Rosario, Facultad de Ciencias Exactas, Ingeniería y Agrimensura, Rosario, Santa Fé, Argentina. 3 Universidad Nacional del Centro de la Provincia de Buenos Aires, Facultad de Ciencias Exactas y Naturales, Tandil, Buenos Aires, Argentina. Resumen: En el trabajo didáctico con los términos propios de un campo disciplinar, uno de los recursos más frecuentes es el uso recurrente de sustantivos en la estrategia de nominalización. Tal nominalización implica una condensación de significados. En este trabajo se comunican los movimientos de empaque y desempaque de significados identificados y los recursos lingüísticos utilizados por una futura profesora de química durante el trabajo didáctico para construir el concepto de solución saturada. El tramo discursivo analizado corresponde a una clase que aborda la temática soluciones en su período de residencia docente en un aula de la educación secundaria. La investigación es de corte cualitativo, centrado en un estudio de caso que se enmarca en las nociones de nominalización y formas congruentes para estudiar los procesos de condensación de significados durante las interacciones discursivas. Se identifican los recursos lingüísticos que la practicante privilegia para la condensación de significados y se discuten sus implicaciones didácticas. Palabras clave: Formación docente inicial; Discurso docente; Enseñanza de la química; Solución saturada. Abstract: In teaching work with specific terminology of each field of discipline, the frequent use of nouns in the strategy of nominalization is one of the most common resources. Nominalization implies condensation of meanings. This paper communicates movements of packing and unpacking of meanings identified and linguistic resources used by a future Chemistry teacher during her didactic work to construct the concept of saturated solution. The discursive section that was analyzed corresponds to a class focused on the theme solutions during her residence period in a secondary school classroom. The research uses a qualitative approach and it is centered on a case study. It is framed in the notions of nominalization and congruent forms in order to analyze the processes of condensation of meanings during the discursive interactions. The linguistic resources for the condensation of meanings preferred by the future teacher are identified and its teaching implications are discussed. Keywords: Pre-service teacher training; Teacher discourse; Chemistry teaching; Saturated solution. Recebido em: 13/06/2020 Aprovado em: 21/02/2021 e-ISSN 1980-850X. Todo o conteúdo deste periódico está sob uma licença Creative Commons (CC Atribuição 4.0 Internacional), exceto onde está indicado o contrário. Ciên. Educ., v. 27, e21028, 2021 2 de 17 Introducción Diferentes investigaciones han estudiado las prácticas de enseñanza en aulas de ciencia desde una perspectiva discursiva (SERDER; JAKOBSSON, 2016; WALLACE, 2004). En estos contextos, la educación científica constituye una actividad lingüística específica o un encuentro entre diferentes discursos (KAMBERELIS; WEHUNT, 2012; WALLACE, 2004). Esta perspectiva sugiere que en la enseñanza en las aulas de ciencia convergen diferentes tipos de lenguajes que se expresan en un híbrido de idiomas (BAJTIN, 1934) o un discurso específico, en el que las expresiones científicas se mezclan con el uso del lenguaje coloquial o cotidiano. Este tipo de uso del lenguaje en la educación científica, con sus características lingüísticas y estructurales distintivas, también se ha descrito en el marco de la lingüística funcional sistémica (HALLIDAY; MATTHIESSEN, 2014). Varios estudios han demostrado que las demandas lingüísticas impuestas a los estudiantes a menudo van más allá de los desafíos conceptuales (SEAH; CLARKE; HART, 2011). En este contexto, aprender ciencia significa desarrollar una comprensión tanto conceptual como discursiva, que incluya las relaciones semánticas o los patrones temáticos específicos (LEMKE, 1997) y las características lingüísticas y estructurales distintivas del discurso científico. Pero, ¿cómo actúa un futuro docente en sus experiencias iniciales en el aula para construir estructuras lingüísticas propias de la ciencia con sus significados? En este trabajo se identifican los movimientos de empaque y desempaque de significados y los recursos lingüísticos utilizados por una futura profesora de química durante el trabajo didáctico centrado en la temática 'soluciones'. Marco teórico Esta investigación propone una interpretación de la enseñanza de la ciencia en las aulas escolares considerando las perspectivas socioconstructivista y sociosemiótica. Desde la perspectiva socioconstructivista, se recupera la relación entre contextos sociales, históricos y culturales (ROGOFF, 1990) y la acción individual (WERTSCH, 1993). Esta relación puede abordarse analizando la acción mediada. El lenguaje puede conceptualizarse como un instrumento de mediación omnipresente y constituye un requisito previo para todos los restantes tipos de mediación por artefactos culturales. Esta perspectiva tiene su fundamento teórico en la ley genética general del desarrollo cultural y su énfasis en la internalización de las experiencias sociales mediadas por herramientas culturales ( VYGOTSKY, 1993). De estas herramientas culturales, el lenguaje fue visto por Vygotsky como la de mayor importancia, entendido como la "herramienta de herramientas" (WELLS, 1994). La relevancia del lenguaje reside no solo en su papel como instrumento de las interacciones sociales (herramienta cultural) sino también en tanto instrumento de pensamiento verbal (herramienta cognitiva) en el individuo ( VYGOTSKY, 1993). El conocimiento científico es visto como construido socialmente, validado y comunicado y el aprendizaje de las ciencias como un proceso de enculturación (MORTIMER; SCOTT, 2003). Para Driver et al. (1994), en el proceso de aprender ciencias, los alumnos internalizan las herramientas culturales, devenidas como herramientas cognitivas, para la creación de significados. Este proceso involucra la participación de Ciên. Educ., v. 27, e21028, 2021 3 de 17 estudiantes y docentes en actividades discursivas de las lecciones de ciencias a través de las cuales "los alumnos se socializan en las formas de conocimiento y prácticas de la ciencia escolar" (DRIVER et al., 1994, p. 11). Desde la perspectiva sociosemiótica, el lenguaje es entendido como una herramienta esencial para la construcción del conocimiento del individuo. Todo aprendizaje, incluido el científico experimental, se lleva a cabo únicamente porque existe el lenguaje como instrumento semiótico. El lenguaje es la condición esencial del conocimiento; el proceso por el cual la experiencia se convierte en conocimiento (HALLIDAY, 1989). Al proponer que el desarrollo del lenguaje es aprender a significar y que el aprendizaje en todas sus formas se caracteriza por el proceso de dar sentido, Halliday (1989) sostiene, además, que aprender el lenguaje es la base de todo ser humano. En esta teoría basada en el lenguaje, propuso una triple perspectiva de aprendizaje: "[...] aprender el lenguaje, aprender a través del lenguaje, aprender sobre el lenguaje" (HALLIDAY, 1993, p. 112, traducción nuestra). Simultáneamente, mientras está aprendiendo ciencias a través del lenguaje, está aprendiendo el lenguaje de la ciencia escolar y sobre ella, aunque no necesariamente a un nivel consciente. Por lo tanto, aprender el lenguaje de la ciencia escolar no se reduce, solamente, a aprender el significado de los conceptos científicos, sino que también incluye la capacidad de "[...] controlar las formas y estructuras lingüísticas únicas que construyen y comunican principios, conocimientos y creencias científicas" (FANG, 2005, p. 337, traducción nuestra). Desde estas exigencias se interpelan, además, a las prácticas de enseñanza. Palabras, relaciones semánticas y densidad semántica Las características gramaticales, la importancia de las palabras y los patrones semánticos típicos (LEMKE, 1997) que constituyen los lenguajes científicos pueden crear dificultades en su uso para los estudiantes (FANG, 2006). Además de términos científicos específicos (como equilibrio químico, solubilidad, reacción química y electrones), hay varias palabras cotidianas que también pueden tener una definición y uso específicos en la ciencia (como calor, trabajo, fuerza y energía). Diferentes investigaciones muestran que el empleo de este tipo de palabras frecuentemente crean confusión para los estudiantes (BROWN; SPANG, 2008; FANG, 2005; KARLSSON; LARSSON; JAKOBSSON, 2020). Además, varias palabras típicas del sujeto (como referencia, constante, patrón y factor) pueden tener distintos significados en diferentes disciplinas, lo que aumenta aún más la complejidad (SERDER; JAKOBSSON, 2016). Por ejemplo, en un estudio sobre el uso del lenguaje en aulas de ciencias, Gyllenpalm y Wickman (2011) encontraron que los estudiantes utilizaban el término experimento relacionado con tarea de laboratorio y refería, principalmente a una actividad pedagógica en contraste con una metodología de investigación. Estas consideraciones adquieren relevancia en las aulas de ciencias si se considera el uso situacional del lenguaje y, consecuentemente, sus variaciones según sea el texto informal/formal, concreto/abstracto, cotidiano/especializado y según quién es el destinatario y cómo se comunica el texto. De esta manera, los discursos en las aulas se desplazan en un continuo entre una comprensión cotidiana del mundo (uso del lenguaje cotidiano) y una relación más distante con el mundo (uso del lenguaje Ciên. Educ., v. 27, e21028, 2021 4 de 17 científico). Los lenguajes y textos utilizados en el campo de la ciencia se caracterizan por la densidad de información, abstracción y tecnicismo (MARKIC; CHILDS, 2016). Halliday y Webster (2004) argumentan que el uso de nominalizaciones y formas pasivas es una técnica gramatical funcional para permitir la tarea del lenguaje cotidiano. Pero para una nueva generación que va a aprender ciencias en la escuela, esto puede llevar a que el lenguaje de enseñanza se perciba como demasiado intensivo en información y abstracto y, por lo tanto, obtenga un rasgo de objetividad y autoridad que a menudo carece de agencia humana. Halliday y Martin (1993) enfatizan que la dificultad para comprender el lenguaje de la ciencia no solo existe a nivel de las palabras. Un lenguaje técnico (tecnicismo) va más allá del uso de palabras individuales y términos específicos. Los términos científicos derivan su significado y valor de estar taxonómicamente organizados, lo que implica que las relaciones semánticas se vuelven centrales para comprender los conceptos de la ciencia (LEMKE, 1997). Los términos técnicos y las entidades abstractas son centrales en la producción de conocimiento científico; una forma esencial de construir palabras científicas es mediante el uso de metáforas gramaticales como las nominalizaciones (HALLIDAY; MARTIN, 1993). La nominalización constituye una característica esencial del discurso científico e implica la sustantivación de procesos y propiedades (CINTO, 2009). La nominalización es el recurso más poderoso para crear metáforas gramaticales. A través de este dispositivo, los procesos (escritos congruentemente como verbos) y las propiedades (escritas congruentemente como adjetivos) se reescriben metafóricamente como sustantivos (HALLIDAY; MATTHIESSEN, 2014). En la versión nominalizada, se pierde la información acerca de los participantes y, normalmente, la de las circunstancias – empaque o condensación de la información –, de manera que la información que, congruentemente, se expresaría en una cláusula con sus componentes, se enuncia en forma resumida en un grupo nominal. Conceptualmente, se mantiene inmóvil a la realidad para permitir la observación y experimentación (HALLIDAY; MARTIN, 1993). Al proceso inverso (convertir la nominalización en una forma congruente) se le denomina desempaque (HALLIDAY; MATTHIESSEN, 2014). Esta densidad está relacionada con expresiones del lenguaje verbal con una alta proporción de palabras de contenido en las que la información se agrupa en nominalizaciones y grupos nominales extendidos (HALLIDAY; MARTIN, 1993). Estos procesos son recuperados por Maton (2014a, 2014b) a partir de la noción de densidad semántica que se refiere al grado en que el significado se condensa en símbolos (términos, conceptos, frases, expresiones, gestos, etc.). Esta noción tiene similitudes con el término condensación (MATON; DORAN, 2017a). Cuanto más fuerte es la densidad semántica, mayor es la cantidad de significados que se condensan; cuanto más débil es la densidad semántica, menos significados se condensan. El grado de condensación dentro de una práctica sociocultural determinada se relaciona con la estructura semántica en la que se encuentra. Esta condensación de significados puede ser de carácter epistémico, esto es, puede referirse a conceptos o a descripciones empíricas o puede ser de carácter axiológico. Este trabajo se centra en la condensación epistémica de las definiciones formales y las descripciones empíricas (MATON, 2014a, 2014b, 2016), a partir de la noción de densidad semántica epistemológica (en adelante, densidad semántica). Ciên. Educ., v. 27, e21028, 2021 5 de 17 Metodología La investigación realizada se inscribe en una metodología cualitativainterpretativa (GUBA; LINCOLN, 1994). En ésta, la fuente principal y directa de los datos son las situaciones naturales; los fenómenos se deben entender dentro de sus referencias espacio-temporales y su contexto (VALLES, 2000). Se recurrió a un estudio instrumental de casos (STAKE, 2013) que se propone la descripción de cómo se conforma el habla de futuros profesores de química en las situaciones de aula en que intervienen. En Argentina, la formación del profesor en Química se realiza, tanto en instituciones de educación superior universitarias como no universitarias, en una carrera específica Profesorado de Educación en Química, con una duración de cuatro años. Si bien el Plan de Estudios puede diferir en su estructura curricular específica, se caracteriza por comprender espacios de formación en los campos de conocimiento propios de la química y de formación para la inserción ocupacional: trayecto sociohistórico-político, trayecto de formación pedagógico-didáctica y el trayecto de la práctica docente. Es en este último trayecto y finalizando la carrera, en el que se ubica la Residencia Docente como espacio curricular donde los futuros profesionales docentes planifican y desarrollan prácticas de enseñanza en cursos de instituciones de educación secundaria y superior. El presente estudio analiza las intervenciones discursivas de una practicante, que cursa la Residencia en el Profesorado en Química de la Universidad Nacional de Mar del Plata (Argentina), cuando lleva adelante sus clases con estudiantes (14-15 años) en un aula de Fisicoquímica en un 2º año de la educación secundaria de un colegio de la Provincia de Buenos Aires, Argentina. Fue seleccionado como el caso de estudio por cuanto el profesor responsable del curso le asignó a la practicante, explícitamente, la tarea de enseñar la construcción de explicaciones científicas escolares mientras aborda la unidad didáctica Disoluciones Químicas ubicada en el eje temático La naturaleza corpuscular de la materia de la propuesta curricular oficial. Este hecho convierte su discurso en el aula en peculiar y relevante para esta investigación. La secuencia didáctica se desarrolló en un total de siete clases. En la primera de ellas se recuperaron las nociones de sistemas homogéneos, sustancia y soluciones. Durante la segunda y tercera clases, estudiantes y practicante trabajaron con la noción de concentración de las soluciones. En la clase siguiente los estudiantes resolvieron una actividad de evaluación y en la quinta clase se trabajó con las nociones de soluciones saturadas e insaturadas, comenzando a abordar la noción de solubilidad que se extendió durante las dos últimas clases en las que se trabajó con la dependencia del punto de saturación con la temperatura. Las clases fueron grabadas en audio y video y transcriptas en su totalidad. Además, se registraron las reflexiones que la practicante explicita durante clases de discusión sobre el episodio que nos ocupa como pos-actividad en la asignatura Residencia. Este trabajo se focaliza en la primera parte de la quinta clase que se inicia con una actividad de mezcla de cloruro de sodio en agua a temperatura ambiente hasta formar una solución saturada (1° episodio). Para el análisis, se recupera la división de la clase en episodios propuesta por la practicante quien utiliza como criterios tanto el Ciên. Educ., v. 27, e21028, 2021 6 de 17 cambio temático como el tipo de actividad desarrollado por los estudiantes (LEMKE, 1997), según los explicita en su diario de clase: Para diferenciar los episodios de la clase, se tuvieron en cuenta los siguientes criterios: cambio en el tipo de recurso (por ejemplo, para reconocer el comienzo del segundo episodio tuve en cuenta que se comenzó a trabajar con un recurso didáctico diferente al que se estaba utilizando, que es el uso de un simulador virtual); cambios en el tipo de actividad (si el grupo de alumnos trabaja a partir de actividades de la guía, si se realizó corrección grupal, si realicé una experiencia demostrativa, etc.); cambio en la forma de diálogo (si se crea una situación de debate para sacar conclusiones sobre la experiencia, por ejemplo). El episodio que se analiza en este artículo es aquel donde se trabajó con la caracterización de las soluciones según su clasificación en saturadas e insaturadas. Durante este episodio la practicante desarrolla tal clasificación y luego se centra en la noción de solubilidad. Para la diferenciación entre los tipos de soluciones utiliza un simulador como recurso didáctico. Para el análisis de las participaciones discursivas se consideró la transcripción de la clase a texto completo y las reflexiones de la practicante sobre sus intervenciones durante el trabajo didáctico con la clasificación de las soluciones. Esto último permitió recuperar la voz de la practicante para seleccionar como fuentes aquellos elementos que ella destaca como significativos sobre sus intervenciones: "[...] tanto en el diálogo extraído de las desgrabaciones como en las diapositivas utilizadas como recurso durante la presentación del patrón temático" [D] 1. Durante un proceso de codificación abierta, se identificaron temas clave que permitieron entrever algunas categorías y subcategorías preliminares, desarrollando un análisis microscópico (o frase por frase) de los datos. El análisis del discurso de la practicante se realizó buscando indicadores que den cuenta de los procesos de empaque y desempaque implicados en la nominalización (organización de grupos nominales). Los fragmentos del discurso seleccionados, en función de tales indicadores, fueron asociados a categorías preliminares, proceso que permitió, por la recurrencia de significados observados en cada frase codificada, definir y describir finalmente cada una de las categorías y subcategorías. Así se diferenció entre aquellos intercambios discursivos practicanteestudiantes en los cuales se priorizó el empleo de formas nominalizadas y aquellos en los que se priorizaron formas congruentes. Estas dos categorías – forma nominalizada y forma congruente – incluyen intervenciones de la practicante en las cuales refiere a solución saturada recurriendo a alguna de ambas modalidades. Los intercambios incluyeron intervenciones discursivas de la practicante durante las cuales recurrió a ambas formas inscribiéndolas en un proceso. Estas intervenciones son consideradas en la categoría denominada procesos que vinculan formas congruentes y nominalizadas bajo las subcategorías denominadas empaque y desempaque. Por otra parte, en la subcategoría desempaque se diferencian aquellos intercambios discursivos durante los cuales la practicante denota un referente empírico – generalmente en el contexto del trabajo con el simulador – de aquellas en las que, desempacando la información, no lo hace. Posteriormente, estas categorías y subcategorías fueron relacionadas entre sí delimitando la estructura jerárquica que se muestra en la ilustración 1. 1 D: documento de reflexión elaborado por practicante sobre sus intervenciones durante el trabajo didáctico con la clasificación de las soluciones. 7 de 17 Ciên. Educ., v. 27, e21028, 2021 Ilustración 1 – Sistema de categorías elaborado durante el análisis Fuente: elaborado por los autores. Durante el análisis también se elaboraron marcas de nubes presentando las nueve palabras más frecuentes en las intervenciones discursivas de la practicante y se consideraron coincidencia con palabras derivadas y longitud mínima igual a cinco letras. El número de palabras se determinó a partir de un proceso de reducción por saturación de palabras relevantes. Análisis episódico En la primera parte del episodio la practicante recupera junto a los estudiantes la noción de solución saturada, abordada durante la clase anterior. Esta revisión es utilizada como introducción al trabajo con la noción de solubilidad. En su turno de habla la practicante refiere a la noción de solución saturada a partir de diferentes formulaciones, cada una de las cuales representa distintas aproximaciones al concepto. En una primera intervención, la practicante recupera la noción de saturación centrando la atención en las cantidades relativas de soluto y de solvente, a efecto de diferenciar y establecer la condición de saturación. Focaliza su discurso en contrastar la idea de supera la cantidad de solvente con la de "el solvente que tengo no puede disolver más soluto que el que voy a ir agregando ¿sí?" (línea 12). 2,3 En esta intervención anticipa una conceptualización basada en las nociones de 'soluto' y de 'solvente' abstrayéndolas de la referencia contextualizada. Esta distinción, seguidamente, es reformulada por la practicante ("Entonces, como venían diciendo los chicos, el solvente ya no puede disolver más cantidad de soluto"; línea 12) y vinculándola con la noción de solución saturada ("Se tienen que acordar eso en las soluciones saturadas", línea 12). De esta forma, la condición establecida a partir de las cantidades relativas soluto/ solvente es empaquetada en el grupo nominal solución saturada. A continuación, la practicante presenta a este grupo como 'causa', inscribiéndolo en una estructura que le permite desempaquetar su significado ("Esto implica que se llega a un máximo valor de concentración", línea 12). En esta última intervención, además, se propone una nueva conceptualización a la noción de solución saturada introduciendo el término concentración trabajado en clases anteriores. Durante estas intervenciones la practicante desarrolla movimientos discursivos centrados en torno a un grupo nominal – solución saturada – que implican procesos de nominalización y desempaque centrados en el empleo de términos del modelo 2 3 Línea 12: corresponde a la línea de la transcripción correspondiente a la intervención. La disolución es un fenómeno de interacción soluto-disolvente y no una acción del disolvente sobre el soluto. No es correcto, por tanto, afirmar «el disolvente no disuelve más soluto». En este trabajo, el objeto del análisis es el discurso verbal de la practicante para identificar, en su organización, procesos de nominalización. Se destaca que, en función de los términos utilizados, el empaque puede incluir errores conceptuales. Ciên. Educ., v. 27, e21028, 2021 8 de 17 científico escolar trabajado en las clases anteriores. Utilizando un software como recurso didáctico, seguidamente, recupera un proceso simulado trabajado en clases anteriores con los estudiantes. La simulación permite visualizar el proceso de saturación de una solución a partir de la identificación perceptual de un depósito sólido en el fondo del recipiente. En este contexto, la practicante utiliza este criterio observacional como evidencia de saturación ("Habíamos visto en el simulador, no sé si se acuerdan, que una manera de reconocer a este tipo de soluciones, era cuando veíamos, por ahí está muy clarito pero, partículas de soluto depositadas en el fondo", línea 12). Esta intervención, caracterizada por una baja densidad semántica, precede a la nominalización ("¿Sí? entonces tenemos una solución saturada", línea 12) recuperando, finamente, el referente empírico simulado ("Observamos soluto depositado en el fondo", línea 12). En esta intervención, la practicante contextualiza y descontextualiza el contenido de su discurso utilizando un referente observable. Este proceso es analizado por la practicante, en su diario de clase, en términos de la transición entre niveles – observable y conceptual –, proceso al que refiere como traducción. El primero de ellos alude al trabajo didáctico a partir de la percepción de un fenómeno; el segundo, a la conceptualización que puede interpretarse, siguiendo la propuesta de Taber (2013), en términos de los niveles de conceptualización macroscópico y submicroscópico: "Durante el desarrollo del patrón temático y el diálogo con los alumnos existe una traducción desde un nivel a otro. Al hablar de 'traducción' estoy haciendo referencia a la manera en que 'ubico' la respuesta de los alumnos, situada generalmente en un nivel observable, en una respuesta correspondiente al nivel conceptual" (D). Durante los procesos de nominalización desarrollados por la practicante durante esta intervención discursiva, participantes – soluto y solvente –, proceso – disolver –, y circunstancias son eliminados del discurso, quedando los dos primeros condensados en el grupo nominal solución saturada. Cláusulas simples son comprimidas en un grupo nominal en la versión metafórica. En este proceso se desarrolla el empaque de la información. En particular, cuando refiere al reconocimiento de una solución saturada, recurriendo al criterio observable ("Habíamos visto en el simulador, no sé si se acuerdan, que una manera de reconocer a este tipo de soluciones, era cuando veíamos, por ahí está muy clarito pero, partículas de soluto depositadas en el fondo", línea 12), es posible reconocer la forma en la cual la practicante reconstruye la experiencia considerando a las partículas de soluto depositadas en el fondo del recipiente. En el cuadro 1 se representan los movimientos discursivos de la practicante durante una de sus participaciones (línea 12) en su evolución temporal. 9 de 17 Ciên. Educ., v. 27, e21028, 2021 Cuadro 1 – Escala de dependencia contextual y condensación de significado para la noción solución saturada en la intervención de la practicante correspondiente a la línea 12 + Dependencia contextual + Condensación de significados "[...] no es que supera la cantidad de solvente sino que el solvente que tengo no puede disolver más soluto que el que voy a ir agregando. ¿Sí? [...]" "Entonces, como venían diciendo los chicos, el solvente ya no puede disolver más cantidad de soluto. [...]" "[...] Se tienen que acordar eso en las soluciones saturadas. [...]" "[...] Esto implica que se llega a un máximo valor de concentración. [...]" "[...] Habíamos visto en el simulador, no sé si se acuerdan, que una manera de reconocer a este tipo de soluciones, era cuando veíamos, por ahí está muy clarito pero, partículas de soluto depositadas en el fondo. [...]" "[...] ¿Sí? entonces tenemos una solución saturada. [...]" "[...] Observamos soluto depositado en el fondo [...]" Fuente: elaborado por los autores. Siguiendo la propuesta de Larsson (2018), se recupera e invierte el modelo de Maton (2014a), para ilustrar en el cuadro 1 las intervenciones de la practicante según la naturaleza de su contenido, en una escala horizontal, de izquierda a derecha, en lugar de verticalmente. Estos movimientos constituyen una escala continua, sin límites exactos, entre un lenguaje cotidiano (izquierda) a un lenguaje científico escolar (derecha). Estas intervenciones son interpretadas según su mayor cercanía a un lenguaje contextual y cotidiano (izquierda) o a un lenguaje más condensado y específico (derecha). En algunos casos, las intervenciones discursivas de la practicante se ubican en la zona central del continuo, lo que puede interpretarse en términos del uso de un lenguaje intermedio (OLANDER, 2013) caracterizado por su carácter híbrido, en tanto constituido por términos pertenecientes a ambos lenguajes sociales. Las intervenciones discursivas de la practicante se ubican en alguna de estas instancias de enunciación; la sucesión temporal de estas intervenciones permite caracterizar los movimientos discursivos en términos de las transiciones entre enunciados que privilegian algunos de ambos lenguajes sociales o de aquellos que los incluyen simultáneamente. Por otra parte, el continuo presentado en el cuadro 1 puede ser leído en términos de un aumento progresivo de la densidad semántica conforme se avanza del extremo representado por una mayor dependencia contextual hacia el de mayor condensación de significados. La inscripción temporal de las transiciones entre enunciaciones caracterizadas por estas modalidades de enunciados, define el perfil semántico del discurso en el aula, caracterizado en términos de su densidad semántica. Al referir a la condensación epistemológica, Maton y Doran (2017a) distinguen diferentes tipos de herramientas. Entre ellas, la conexión ubica explícitamente significados dentro de constelaciones epistemológicas, por ejemplo, los esquemas Ciên. Educ., v. 27, e21028, 2021 10 de 17 clasificatorios. La clasificación describe a un término como un subtipo de otro: cada subtipo en la clasificación es un ejemplo del tipo. En un turno de habla posterior, en el mismo episodio, la practicante explicita la definición de soluciones insaturadas al desplegar un sistema clasificatorio que incluye a éstas junto a las soluciones saturadas. La aproximación conceptual a la noción de solución insaturada se realiza, por un lado, a partir de un proceso de metaforización, desde el contraste con solución saturada, recurriendo a una operación léxica – agregado del prefijo 'in' – a la nominalización previa: "[...] si decimos que en las saturadas nuestro solvente ya no puede disolver más soluto, en las insaturadas, por ende, o sea, por lógica, podemos suponer que sí admite más soluto. Todavía no llegó a la saturación ¿Lo ven eso?", línea 16. En esta última intervención, la practicante desempaqueta la noción de solución saturada, explicitando una significación que recupera en la caracterización para solución insaturada. La insaturación es significada tanto en términos de las cantidades relativas de soluto y solvente como a partir de la no-saturación. En una nueva intervención el término solución insaturada es resignificado a partir de la noción de concentración. "Admitir más soluto" (línea 16) implica la posibilidad de modificar la concentración de la solución y esta idea aumenta el contenido informativo del término solución insaturada ("[...] A medida que vas agregando más soluto, se va concentrando la solución, ¿sí? [...]", línea 30). Sucesivamente, la posibilidad de aumentar la cantidad de soluto y el consiguiente aumento de la concentración, permiten aumentar la densidad semántica del término solución insaturada en sendos procesos de desempaque conceptual. La nominalización inicial, conducente a solución saturada es, seguidamente, des-andada durante la delimitación conceptual de solución insaturada. En esta misma intervención discursiva, la practicante fija la atención de los estudiantes en un referente empírico – como es el soluto depositado en el fondo del recipiente, presentado como indicador perceptual de la ausencia de saturación –, y finaliza su intervención expresando el proceso metafóricamente mediante un sustantivo ("[...] ¿sí? [...] no llega al punto de saturación, sí, porque no hay soluto depositado en el fondo. Bien [...]", línea 30). Los términos saturada e insaturada aumentan la densidad semántica de la palabra solución (MATON, 2016), permitiendo una mayor diferenciación al explicitar el tipo (solución) en subtipos (saturada, insaturada). Los significados asociados con el tipo de orden superior son necesariamente adoptados por el subtipo de orden inferior (aunque no al revés) porque para que X sea un subtipo de Y debe poseer las características relevantes de Y (además de sus propias características distintivas). Así se crean significados cruzados permitiendo un aumento de la densidad semántica tanto del término-tipo como de cada término-subtipo. El despliegue de estas clasificaciones durante las intervenciones discursivas es recuperado por la practicante en una nueva instancia, centrada en la soluciones insaturadas ("Si tienen dudas, me frenan, pero dentro de las insaturadas vamos a clasificar a las soluciones, ¿tienen alguna duda? Pregunten [...] Bueno, dentro de las insaturadas podemos clasificar a las soluciones en diluidas y concentradas, ¿Les suenan esos términos?", línea 18). Recurre a la concentración, una propiedad de la solución para delimitar la diferenciación de cada término-subtipo ("[...] Va a estar concentrada porque tiene una concentración alta, podemos llamarlo así, y diluida, cuando está lejana a ese valor máximo. Es decir, que tiene poca cantidad de soluto [...]", línea 26) y al valor 11 de 17 Ciên. Educ., v. 27, e21028, 2021 máximo de concentración definido para la solución saturada ("[...] Concentrada va a ser cuando esté bastante cerca de ese valor máximo, cuando esté cerca de saturarse ¿lo ven?", línea 26). Luego recupera ambos subtipos de solución insaturada a partir de una cláusula congruente seguida de una nominalización ("A medida que vas agregando más soluto, se va concentrando la solución, ¿sí? Todavía no llega al punto de saturación, sí, porque no hay soluto depositado en el fondo. Bien ¿Alguna duda hasta acá?", línea 30). En la nube de palabras mostrada en la ilustración 2 se presentan las palabras más frecuentes durante las interacciones discursivas. Se priorizaron palabras técnicas para el patrón temático propuesto por la practicante y otras asumidas como relevantes durante el análisis de sus intervenciones. Entre estas últimas, fondo y máximo son utilizadas por la practicante para contextualizar los procesos nominalizados y diferenciar los tipos de soluciones, respectivamente. Ilustración 2 – Marca de nube correspondiente a las palabras más frecuentes en las intervenciones discursivas de la practicante Fuente: elaborada por los autores. Un término técnico es, frecuentemente, parte de una compleja red de significados que define su alcance conceptual. Los procesos de empaque y desempaque desarrollados por la practicante delimitan diferentes aproximaciones conceptuales para los grupos nominales propuestos durante este episodio. La practicante reconoce estas diferencias para el término solución saturada: "Se definió al término [solución saturada] a partir de diferentes relaciones semánticas. Lo mismo ocurrió con el concepto de solución insaturada. La siguiente tabla [cuadro 2, elaborado por la practicante], resume las relaciones semánticas que involucran a cada uno de estos conceptos". Cuadro 2 – Diferentes relaciones semánticas utilizadas durante el trabajo discursivo con la noción de solución saturada* Concepto Relaciones semánticas que involucran al término Solución saturada [si se le agrega más] cantidad de soluto [no se modifica su] concentración [llegan a un] máximo valor de concentración [aquellas en que] el solvente [ya no puede disolver más] cantidad de soluto [su] concentración [equivale al] valor de solubilidad [podemos observar] soluto depositado en el fondo [su] color [es muy] fuerte Las letras cursivas negritas están en los textos elaborados por la practicante. Fuente: elaborado por la practicante. Ciên. Educ., v. 27, e21028, 2021 12 de 17 Estas aproximaciones se diferencian entre sí en tanto suponen diferentes niveles de abstracción y su reconocimiento permite dar cuenta de las formas en que los significados se condensan en términos tales como solución saturada, solución insaturada, solución diluida y solución concentrada durante los intercambios discursivos practicante-estudiantes. Estas nominalizaciones delimitan condensaciones léxicas de significado, codificadas en un grupo nominal y constituyen un recurso lingüístico clave para explicar algunos fenómenos asociados a la construcción de conocimiento en las aulas de ciencias. La fortaleza de la densidad semántica no es intrínseca a una palabra, sino que depende de la constelación epistemológica de significados con los que se relaciona la palabra. Cuanto mayor sea la cantidad de relaciones establecidas con otros significados, más fuerte es la densidad semántica asociada a cada uno de estos grupos nominales. Estas relaciones presentadas por la practicante son diversas delimitando diferentes significados a cada uno de estos grupos; la misma palabra se puede utilizar como técnica o cotidiana. Esto es particularmente importante para la temática trabajada en este episodio considerando que algunos de los términos del patrón temático de la practicante resuenan en el lenguaje cotidiano de los estudiantes. En la ilustración 3 se muestran las diferentes conceptualizaciones empaquetadas en cada uno de estos grupos nominales. Ilustración 3 – Conceptualizaciones incluidas en diferentes grupos nominales propuestas por la practicante en sus intervenciones discursivas *Las citas para solución saturada se muestran en el cuadro 2. Fuente: elaborado por los autores. La practicante interpretó sus intervenciones en términos del trabajo didáctico entre los niveles observable y conceptual. En una reflexión general acerca del lenguaje utilizado y en referencia a los niveles con los que trabajó durante el episodio, reconoce, en su reflexión, la prevalencia del conceptual: "[...] en cuanto a los niveles de representación de la materia utilizados en la presentación del patrón temático se evidencia el privilegio en el uso del nivel conceptual por sobre el nivel observable" (D). Ciên. Educ., v. 27, e21028, 2021 13 de 17 A partir del análisis de la transcripción de la clase la residente reconoció las diversas aproximaciones conceptuales a la noción de solución saturada, por ejemplo, y lo expresó en los siguientes términos: "Al analizar las maneras de presentar estos términos durante el desarrollo del patrón temático se observó, en ambos casos, que cada uno de estos términos estaba involucrado en una importante cantidad de relaciones semánticas, que ofrecía diversidad en su conceptualización. Esta diversidad puede ser un factor de dificultad para el alumno en la comprensión de estos conceptos." (D). En el análisis del episodio, la practicante interpreta las conceptualizaciones sobre solución saturada mostradas en el cuadro 2, en términos de la siguiente clasificación: Del análisis de las relaciones semánticas que involucran al término soluciones saturadas se infiere la diversidad en las relaciones semánticas que incluyen a este concepto: éste es presentado a partir de la definición del mismo ('aquellas en que el solvente ya no puede disolver más cantidad de soluto'), de sus características ('llegan a un máximo valor de concentración') y utilizando relaciones semánticas con los términos: concentración, soluto y solvente. No necesariamente se utilizaron estos tres términos al mismo tiempo durante su definición ('si se le agrega más cantidad de soluto no se modifica su concentración'). Este término también es presentado a partir de sus variables observables ('podemos observar soluto depositado en el fondo'; 'su color es muy fuerte'). (D). Además de explicitar que este último término "[...] estaba involucrado en una importante cantidad de relaciones semánticas que ofrecía diversidad en su conceptualización [...]", la practicante reconoce que "esta diversidad puede ser un factor de dificultad para el alumno en la comprensión de estos conceptos" (D). La diversidad de relaciones semánticas en la que se ubica la noción de solución saturada expresa, en la perspectiva de la practicante, el trabajo con el patrón temático. En particular, destaca que "[...] la complejidad indicada se acentúa con el hecho de estar relacionado con el término 'concentración' (que suele ser un término difícil de asimilar)" (D). Esta lectura de la residente, por un lado, expresa la dificultad conceptual que presenta, para los estudiantes, la noción de solución saturada considerando su conceptualización relacionada a la noción de concentración. Esta última nominalización compacta una red de significados que abstraen la noción de solución saturada. Pero, además, la noción de concentración priorizada por el grupo de estudiantes, estaría vinculada, según la practicante, a variables con referentes observables que, finalmente, fueron priorizados por los estudiantes: "Esto podría deberse al uso del simulador de concentración previo al desarrollo del concepto de 'soluciones saturadas' ya que en el mismo se trabaja con variables del tipo observables: presencia de soluto depositado en el fondo y cambio en el color de las soluciones a medida que aumenta la concentración de las mismas". (D) La lectura de la noción de concentración, ligada a aspectos observables de la solución, expresaría tanto una forma privilegiada de significar el fenómeno por parte de los estudiantes como de las dificultades que les presentó la conceptualización asociada al patrón temático propuesto por la practicante. Según la residente, estas dificultades vinculadas a la conceptualización se expresaron en la persistencia de una fuerte referencia al contexto, expresada en las intervenciones discursivas de los estudiantes ("A pesar de la prevalencia del nivel conceptual durante el desarrollo del patrón, se observó que los alumnos se aferran a las variables observables en el momento de definir o reconocer un concepto", D). En sus intervenciones, los estudiantes privilegiaron la lectura del fenómeno – definido por la formación de la solución – delimitándolo a partir de referentes observables frente a la lectura centrada en una conceptualización en el nivel macroscópico, promovida por la practicante. En este contexto, las formas Ciên. Educ., v. 27, e21028, 2021 14 de 17 congruentes y nominalizadas para hablar sobre el fenómeno coexistieron durante los intercambios. Consideraciones finales La investigación sobre las prácticas en el aula durante la residencia docente se instala como un requisito constitutivo para el análisis de las situaciones de enseñanza, las actuaciones del docente, las intervenciones de los estudiantes, las propuestas didácticas y los recursos utilizados (STENHOUSE, 1998). En este contexto se enmarca el estudio de los recursos lingüísticos y la densidad semántica para analizar la manera en que se organiza la nominalización de solución saturada en el discurso de la practicante durante el episodio presentado. Los resultados permiten establecer algunos aspectos a tener en cuenta para la formación de futuros profesores en Química. Las nominalizaciones constituyen condensaciones léxicas de significado codificadas en un grupo nominal, y representan un recurso lingüístico clave para explicar fenómenos fisicoquímicos. En este estudio se observa que el lenguaje utilizado por la practicante se caracteriza por priorizar el empleo de términos técnicos que denotan procesos que se expresan metafóricamente mediante sustantivos. Con menor frecuencia, la practicante recurre al empleo de enunciados más familiares y a un lenguaje no especializado, indicado por la observación de fenómenos cotidianos que recupera de la clase anterior. La diversidad registrada en las conceptualizaciones propuestas por la practicante, se expresa en procesos que implican agrupamientos diferentes en un mismo grupo nominal que, si bien se homologa discursivamente a un mismo término, es provisto de una importante polisemia en tanto resume diferentes palabras y relaciones semánticas (como se observa en la ilustración 2). Este resultado permite reflexionar, durante la residencia docente, en instancias post-activas, acerca de la manera en que la misma etiqueta empleada en el grupo nominal expresa diferentes aproximaciones al concepto. Estas aproximaciones pueden permanecer atomizadas entre sí en tanto el docente no establezca explícitamente los nexos conceptuales pertinentes. En este contexto se instala la necesidad de desandar el proceso de nominalización disolviendo la alta densidad semántica concentrada en el grupo nominal y seleccionar adecuadamente qué conceptualizaciones han de ser privilegiadas. Otro aspecto que se deriva de este estudio atiende a la orientación para la construcción de significados en el discurso científico escolar. La noción de solución, durante el episodio de esta clase fue resignificada conceptualmente a partir de la clasificación propuesta por la practicante. La clasificación permitió aumentar la densidad semántica del término solución tanto por las relaciones tipo/subtipo involucradas (MATON; DORAN, 2017b) como por la conceptualización propuesta para cada subtipo. En este último caso, el aumento en la densidad semántica se presentó con la diversidad inherente a las diversas aproximaciones conceptuales abordadas. La tecnicidad, la nominalización y la clasificación se presentaron como recursos lingüísticos privilegiados por la practicante para la condensación de significados y para aumentar la densidad semántica. Se trata de recursos que posibilitan la condensación epistemológica (MATON; DORAN, 2017a). Como sostienen Chamorro, Mizuno y Moss (2003), con relación a los aprendizajes de los estudiantes, existe un factor Ciên. Educ., v. 27, e21028, 2021 15 de 17 clave en el manejo de las nominalizaciones: la concientización. En este trabajo se extiende esta importancia a la formación profesional docente: aprender a reconocer las nominalizaciones como procesos dinámicos, utilizados en sus discursos en el aula de ciencia. Esto es, además del reconocimiento de los procesos de empaque y desempaque, la necesidad de visibilizar que existe otro factor clave en el manejo de estos procesos: la toma de conciencia. En este contexto, Cinto (2009) desarrolla sugerencias para el manejo adecuado de las metáforas en el contexto pedagógico que, si bien fueron propuestas para el trabajo didáctico con textos escritos, pueden ser recuperadas en el contexto de las intervenciones discursivas docente. Entre ellas, des-ocultar la información presente en la nominalizaciones se presenta como particularmente valiosas en términos de des-naturalizar el empleo de estrategias discursivas tendientes a aumentar la densidad semántica en el discurso docente. Estas modalidades de intervención didáctica, además, favorecerían la movilidad discursiva (LARSSON; JAKOBSSON, 2017) entre los discursos cotidianos y científicos, así como entre significados concretos, abstractos, específicos y generales. En todo caso, el potencial didáctico tanto de esta movilidad como del empleo de nominalizaciones reside en la posibilidad de un trabajo reflexivo de parte del docente sobre sus intervenciones discursivas. En el contexto de la formación docente inicial, el análisis del propio discurso en el aula se presentó como un recurso didáctico valorado positivamente por la practicante, por ejemplo, en términos del reconocimiento de las dificultades didácticas durante la construcción de significados con los estudiantes ("El análisis realizado, centrado en el discurso, permite evidenciar estos niveles de dificultad", D). A partir de los resultados presentados en este trabajo se entiende que las instancias de reflexión centradas en identificar y explicitar los recursos lingüísticos utilizados durante la construcción de nominalizaciones se presentarían como un recurso privilegiado para la construcción del discurso docente en el aula de ciencia. El aumento de la densidad semántica en el proceso de nominalización sin que se expliciten las relaciones entre las diferentes formas nominalizadas se presentaría como una fuente de dificultades potenciales para la conceptualización de los estudiantes. En este contexto se instala la necesidad de una mediación docente que explicite, en caso de presentarse, las relaciones entre las diferentes formas nominalizadas. Finalmente, y recuperando recomendaciones dirigidas a sugerir que las prácticas de enseñanza de la ciencia deberían apelar más a la congruencia que a la metáfora gramatical en los niveles de escolaridad iniciales de la educación secundaria (CINTO, 2009), es interesante destacar el reconocimiento de la practicante a la tendencia de los estudiantes a ubicar los intercambios discursivos en referencia a eventos observables: Esta tendencia es justificada, en este estudio, por referencia al empleo de procesos simulados en clases anteriores: En este contexto, al ofrecer referentes empíricos explícitos, el trabajo didáctico con procesos simulados permitiría al docente recurrir al empleo de formas congruentes que facilitarían desocultar y mediar discursivamente en la construcción de formas nominalizadas. Referencias BAJTIN, M. La palabra en la novela. In: BAJTIN, M. Teoría y estética de la novela. Madrid: Taurus, 1934. p. 77-236. Ciên. Educ., v. 27, e21028, 2021 16 de 17 BROWN, B. A.; SPANG, E. Double talk: synthesizing everyday and science language in the classroom. Science Education, Hoboken, v. 92, n. 4, p. 708-732, 2008. DOI: https://doi.org/10.1002/ sce.20251. CHAMORRO, D.; MIZUNO, J.; MOSS, G. Tergiversaciones y correspondencias: la metáfora y sus bemoles. Revista Latinoamericana de Estudios del Discurso, Caracas, v. 3, p. 29-47, 2003. CINTO, M. T. La nominalización: obstáculo para la comprensión lectora. Educación, Lenguaje y Sociedad, La Pampa, v. 6, n. 6, p. 171-186, 2009. DRIVER, R.; ASOKO, H.; LEACH, J.; SCOTT, P.; MORTIMER, E. Constructing scientific knowledge in the classroom. Educational Researcher, Washington, v. 23, n. 7, p. 5-12, 1994. DOI: https://doi. org/10.3102/0013189X023007005. FANG, Z. Scientific literacy: a systemic functional linguistics perspective. Science Education, Hoboken, v. 89, n. 2, p. 335-347, 2005. DOI: https://doi.org/10.1002/sce.20050. FANG, Z. The language demands of science reading in middle school. International Journal of Science Education, London, v. 28, n. 5, p. 491-520, 2006. DOI: https://doi. org/10.1080/09500690500339092. GUBA, E. G.; LINCOLN, Y. S. Competing paradigms in qualitative research. In: DENZIN, N. K.; LINCOLN, Y. S. (ed.). Handbook of qualitative research, vol. 2. Thousand Oaks, CA: Sage, 1994. p. 105-117. GYLLENPALM, J.; WICKMAN, P. O. "Experiments" and the inquiry emphasis conflation in science teacher education. Science Education, Hoboken, v. 95, n. 5, p. 908-926, 2011. HALLIDAY, M. Some grammatical problems in scientific English. In: HALLIDAY, M. A. K.; MARTIN, J. R. (ed.). Writing science: literacy and discursive power. Washington: Falmer Press, 1989. p. 69-85. HALLIDAY, M. Towards a language-based theory of learning. Linguistics and Education, New York, v. 5, n. 2, p. 93-116, 1993. DOI: https://doi.org/10.1016/0898-5898(93)90026-7. HALLIDAY, M.; MARTIN, J. Writing science: literacy and discursive power. Washington: Falmer Press, 1993. HALLIDAY, M.; MATTHIESSEN, C. An introduction to functional grammar. London: Routledge, 2014. HALLIDAY, M.; WEBSTER, J. The language of science, vol. 5. London: Continuum International Publishing, 2004. KAMBERELIS, G.; WEHUNT, M. D. Hybrid discourse practice and science learning. Cultural Studies of Science Education, Dordrecht, v. 7, n. 3, p. 505-534, 2012. DOI: https://doi.org/10.1007/s11422012-9395-1. KARLSSON, A.; LARSSON, P. N.; JAKOBSSON, A. The continuity of learning in a translanguaging science classroom. Cultural Studies of Science Education, Dordrecht, v. 15, p. 1-25, 2020. DOI: https://doi.org/10.1007/s11422-019-09933-y. LARSSON, P. N. "We’re talking about mobility": discourse strategies for promoting disciplinary knowledge and language in educational contexts. Linguistics and Education, New York, v. 48, p. 61-75, 2018. LARSSON, P. N.; JAKOBSSON, A. Semantiska vågor–elevers diskursiva rörlighet i gruppsamtal. Nordic Studies in Science Education, v. 13, n. 1, p. 17-35, 2017. DOI: https://doi.org/10.5617/ nordina.2739. LEMKE, J. Aprender a hablar ciencia: lenguaje, aprendizaje y valores. Barcelona: Paidós, 1997. Ciên. Educ., v. 27, e21028, 2021 17 de 17 MARKIC, S.; CHILDS, P. E. Language and the teaching and learning of chemistry. Chemistry Education Research and Practice, London, v. 17, n. 3, p. 434-438, 2016. DOI: https://doi.org/10.1039/C6RP90006B. MATON, K. Building powerful knowledge: the significance of semantic waves. In: BARRETT, B.; RATA, E. (ed.). Knowledge and the future of the curriculum. London: Palgrave Macmillan, 2014a. p. 181-197. MATON, K. Knowledge and knowers: towards a realist sociology of education. London: Routledge, 2014b. MATON, K. Legitimation code theory: building knowledge about knowledge-building. In: HOOD, S.; SHAY, S. (ed.). Knowledge-building: educational studies in legitimation code theory. London: Routledge, 2016. p. 1-24. MATON, K.; DORAN, Y. J. Condensation: a translation device for revealing complexity of knowledge practices in discourse, part 2 – clausing and sequencing. Onomázein, Chile, p. 77-110, 2017a. (Número especial SFL). MATON, K.; DORAN, Y. J. Semantic density: A translation device for revealing complexity of knowledge practices in discourse, part 1 – wording. Onomázein, Chile, p. 46-76, 2017b. (Número especial SFL). MORTIMER, E. F.; SCOTT, P. Meaning making in secondary science classrooms. Philadelphia: Open University Press, 2003. OLANDER, C. Why am I learning evolution?: pointers towards enacted scientific literacy. Journal of Biological Education, New York, v. 47, n. 3, p. 175-181, 2013. DOI: https://doi.org/10.1080/00219 266.2013.821750. ROGOFF, B. Apprenticeships in thinking: cognitive development in social context. Oxford, UK: Oxford University Press, 1990. SEAH, L. H.; CLARKE, D. J.; HART, C. E. Understanding students' language use about expansion through analyzing their lexicogrammatical resources. Science Education, Hoboken, v. 95, n. 5, p. 852-876, 2011. DOI: https://doi.org/10.1002/sce.20448. SERDER, M.; JAKOBSSON, A. Language games and meaning as used in student encounters with scientific literacy test items. Science Education, Hoboken, v. 100, n. 2, p. 321-343, 2016. STAKE, R. E. Multiple case study analysis. New York: Guilford Press, 2013. STENHOUSE, L. La investigación como base de la enseñanza. Madrid: Morata, 1998. TABER, K. Revisiting the chemistry triplet: drawing upon the nature of chemical knowledge and the psychology of learning to inform chemistry education. Chemistry Education Research and Practice, London, v. 14, n. 2, p. 156-168, 2013. DOI: https://doi.org/10.1039/C3RP00012E. VALLES, M. Técnicas cualitativas de investigación social: reflexión metodológica y práctica profesional. Madrid: Síntesis Ed., 2000. VYGOTSKY, L. S. Pensamiento y lenguaje. In: VYGOTSKY, L. S. Obras escogidas: vol. II. Madrid: Visor, 1993. p. 11-348. WALLACE, C. Framing new research in science literacy and language use: authenticity, multiple discourses, and the "third space". Science Education, Hoboken, v. 88, n. 6, p. 901-914, 2004. WELLS, G. The complementary contributions of Halliday and Vygotsky to a "language-based theory of learning". Linguistics and Education, New York, v. 6, n. 1, p. 41-90, 1994. WERTSCH, J. Voces de la mente: un enfoque sociocultural para el estudio de la acción mediada. Madrid: Visor, 1993.
https://openalex.org/W2962309050	https://www.e-publicacoes.uerj.br/index.php/neiba/article/download/39247/29943	Portuguese	null	Gramsci e o negacionismo climático estadunidense: a construção do discurso hegemônico no Antropoceno	Revista Neiba	2,019	cc-by	9,905	Palavras-chaves: Negacionismo climático; hegemonia; Gramsci. Negacionismo climático; hegemonia; Gramsci. Vínculo Institucional:1 Programa de Pós-Graduação em Relações Internacionais na Universidade Federal de Uberlândia Vínculo Institucional:1 Programa de Pós-Graduação em Relações Internacionais na Universidade Federal de Uberlândia Fernanda Castro Gastaldi Fernanda Castro Gastaldi Resumo: Antropoceno é o termo utilizado por alguns cientistas para denominar o período geológico mais recente do planeta Terra - período este marcado pela ação humana prejudicial ao meio ambiente. Diante desse contexto, o aquecimento global emerge como um tema relevante nos debates científicos e políticos. O consenso científico reconhece a existência de um processo de aquecimento global antropogênico atualmente em curso, enquanto as vertentes negacionistas o questionam. Os Estados Unidos é centro preponderante de formulação e difusão das perspectivas negacionistas e, como superpotência, possui capacidade de estabelecer os termos do discurso hegemônico global. Este trabalho objetiva apresentar uma análise essencialmente gramsciana a respeito do negacionismo. O estudo concretiza-se por meio de pesquisa bibliográfica, avaliações qualitativas, análise teórica e método de abordagem hipotético-dedutivo. Conclui-se que o negacionismo configura-se como uma ideologia orgânica do bloco histórico corrente, sancionando um sistema produtivo coerente à valorização do capital e profícuo à classe hegemônica vigente. Palavras-chaves: VOLUME VII 2018 Dossiê: SimpoRI 2018 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 relevant issue in scientific and political debates. The scientific consensus recognizes the existence of an anthropogenic global warming process currently under way, while the denialist perspectives question it. The United States is the preponderant center of formulation and diffusion of denialist approaches and, as the superpower, it has the capacity to establish the terms of global hegemonic discourse. This article aims to present a Gramscian analysis of climate denialism. The study is carried out through bibliographical research, qualitative evaluations, theoretical analysis and hypothetico-deductive method. It is concluded that the denialism is configured as an organic ideology of the current historical bloc, sanctioning a productive system coherent to the capital appreciation and profitable to the hegemonic class. Keywords: Climate denialism; hegemony; Gramsci. Abstract: Anthropocene is the term used by some scientists to name the most recent geological period of planet Earth - a period marked by detrimental human action to the environment. Given this context, global warming emerges as a 1 1 O presente trabalho foi realizado com apoio da Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Código de Financiamento 001. 1 O presente trabalho foi realizado com apoio da Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Código de Financiamento 001. VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Introdução No que tange à questão climática, evidências corroboram o cenário apresentado. Junto ao aumento da temperatura terrestre verificam-se fenômenos dela decorrentes, como o derretimento das geleiras e o aumento do nível do mar. Dados apresentados por Konrad et al. (2018) mostram que o Oceano Antártico perdeu uma área de 1.463 quilômetros quadrados de gelo submarino entre 2010 e 2016. Essa área equivale à da cidade de Londres (Konrad et al., 2018, pp. 258-261). E os efeitos provavelmente se tornarão ainda mais evidentes no futuro. Somente em Bangladesh a previsão é que, nos próximos anos, dezenas de milhões de pessoas sejam obrigadas a fugir das planícies baixas devido à elevação do nível do mar, gerando uma crise imigratória sem precedentes naquela região (Chomsky, 2017, p. 321). O termo “Antropoceno” é utilizado por alguns cientistas para referir-se ao período geológico mais recente do planeta Terra, marcado pelo domínio da ação humana em meio à natureza. Segundo o conceito original, proposto por Crutzen e Stoermer (2000), o período estende-se do final do século XVIII aos dias atuais - intervalo de tempo em que o impacto da atividade antropogênica sobre o meio ambiente tornou-se claramente perceptível. Entretanto, não há consenso em torno de seu marco inicial, nem ao menos de sua validade científica. Ainda assim, em 2008, especialistas da Comissão de Estratigrafia da Sociedade Geológica de Londres considerou a terminologia válida como unidade formal de tempo geológico (Sayre, 2012). A humanidade vem ao longo de muitos anos submetendo o sistema terrestre ao atendimento de suas necessidades básicas e aos processos industriais vinculados à dinâmica capitalista. O uso imoderado dos recursos naturais fez eclodir - nas ciências naturais e nos debates políticos - temas como destruição da camada de ozônio, poluição urbana, extinção animal, devastação dos recursos hídricos, entre outras questões típicas do Antropoceno. E dentre as temáticas correntes, o aquecimento global destaca-se por sintetizar uma ampla variedade de fatores e, principalmente, por mobilizar posicionamentos antagônicos. Mesmo diante de evidências empíricas e comprovações científicas, há quem questione a veracidade do aquecimento global. Surgem então vertentes teóricas que afirmam que o aquecimento global não é um fenômeno real, ou que é um processo natural sobre o qual a ação antropogênica não possui qualquer influência ou culpa. 2 A referida publicação consiste em um boletim anualmente elaborado. Para fins comparativos, horário semelhante ao de 2018 só foi declarado em 1953, em meio ao desenvolvimento da bomba de hidrogênio pela União Soviética no contexto da Guerra Fria (Bulletin of the Atomic Scientists, 2018). Keywords: Keywords: Climate denialism; hegemony; Gramsci. 2 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Este trabalho objetiva apresentar uma análise essencialmente gramsciana a respeito do negacionismo, abordando-o enquanto discurso que visa colocar-se e preservar-se como hegemônico ou dominante em benefício do sistema capitalista vigente. Não há primazia de uma esfera sobre a outra. Ambas são igualmente relevantes à análise. Ideias e condições materiais estão sempre atreladas, influenciando- se mutuamente. Ideias precisam ser compreendidas em relação às condições materiais, e estas incluem tanto as relações sociais quanto os meios físicos de produção. Assim, o materialismo histórico de Gramsci evita o reducionismo, consumado pelo economicismo ou pelo idealismo (Cox, 1993, p. 56). Os argumentos apresentados encontram- se assim distribuídos: a seção 1 aborda o arcabouço teórico aplicado à análise; a seção 2 versa sobre o negacionismo, como uma construção discursiva constituída por elementos básicos da comunicação - entre os quais se destacam mensagem, emissor e receptor; a seção 3 realiza a aplicação de conceitos básicos da teoria gramsciana (apresentados na seção 1) ao objeto de estudo (descrito na seção 2). Por fim, são apresentadas as conclusões do presente estudo. A superestrutura do bloco histórico é formada por duas esferas essenciais: a da sociedade política e a da sociedade civil. A sociedade civil constitui a maior parte da superestrutura e deve ser considerada como “a hegemonia cultural e política de um grupo social sobre o conjunto da sociedade como conteúdo ético do Estado” (Portelli, 1977, pp. 19-22). 1 Gramsci: Bloco histórico e a construção do discurso hegemônico Por sua vez, a sociedade política agrupa o conjunto das atividades que se relaciona ao exercício da coerção (dominação direta ou de comando exercida através do Estado ou do governo jurídico). A sociedade política está a cargo de conformar as massas populares ao tipo de produção e à economia de um determinado momento, assegurando a disciplina dos grupos sociais a fim de garantir o funcionamento do sistema produtivo. Torna-se um prolongamento da sociedade civil, quando reconhecida como fruto da direção econômica e ideológica que a classe fundamental exerce sobre a sociedade. (Ibid., pp. 30-31). Os “Cadernos do Cárcere”, obra de Antonio Gramsci constituída por uma coleção de escritos desenvolvidos entre 1929 e 1937, oferecem arcabouço teórico de grande valia aos propósitos deste trabalho. Esta seção aborda os principais argumentos do pensamento gramsciano aplicados à análise em questão. Especificadamente, apresenta-se os conceitos de bloco histórico, bloco intelectual e hegemonia, respectivamente. Introdução Enfim, há ainda aqueles que, embora acreditem no processo tal como é descrito pelo consenso científico, consideram demasiada a importância concedida ao fenômeno e discordam que medidas devam ser tomadas para combatê- lo. Essas visões se resumem, neste artigo, ao termo “negacionismo”. Em janeiro de 2018, a publicação The bulletin of the Atomic Scientists avançou os ponteiros de seu famoso Relógio do Juízo Final para dois minutos antes da meia noite2. O horário é estabelecido com base nas maiores ameaças à sobrevivência humana, que hoje são basicamente duas: armas nucleares e mudança climática (Bulletin of the Atomic Scientists, 2018). Nos Estados Unidos, o discurso negacionista tem obtido êxito em influenciar processos políticos. Atualmente, o país é centro preponderante de formulação e difusão internacional das vertentes negacionistas, realidade resultante da atuação dos think tanks conservadores, impérios midiáticos, lobbies, partidos políticos e grandes corporações. Como superpotência, os Estados Unidos possuem capacidade de estabelecer termos do discurso global e, por isso, o estudo do tema proposto é de singular relevância nesse país. 3 no contexto da Guerra Fria (Bulletin of the Atomic Scientists, 2018). VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 3 Os intelectuais não constituem uma classe propriamente dita, mas formam grupos vinculados às diversas classes existentes. Cada grupo social possui uma camada de intelectuais própria, ou tende a formá-la. Os intelectuais agem em nome da classe que representam e à qual estão social e economicamente vinculados. Esses indivíduos dão ao grupo social homogeneidade e consciência de sua própria função nos planos econômico, político e social (Portelli, 1977, pp. 48-85). 4 Quando a classe dirigente agrega em torno de si os intelectuais de outros grupos sociais a fim de decapitar sua direção ideológica e política (Portelli, 1977, p. 69). O transformismo absorve ideias potencialmente contra- hegemônicas e as fazem consistentes com a doutrina hegemônica (Cox, 1993, p. 63). Dossiê: SimpoRI 2018 intelectual”. Há ainda, no interior do bloco histórico, os grupos aliados, que servem de base social à hegemonia e desempenham um papel secundário no sistema. Enfim, verifica-se a existência de grupos excluídos do sistema hegemônico, que são as classes subalternas (Ibid., pp. 45-80). Gill e Law (1993) afirmam que os intelectuais orgânicos teorizam as condições de existência do sistema como um todo. Eles sugerem políticas, as justificam e, quando necessário, as aplicam (Gill e Law, 1993, p. 110). Os intelectuais desempenham, portanto, papel chave na construção do bloco histórico. São eles os “emissários” do grupo dominante para o exercício da hegemonia social e do governo político (Cox, 1993, p. 57; Portelli, 1977, p. 87). O ponto essencial das relações estrutura- superestrutura reside no estudo do vínculo que constitui sua unidade. A este vínculo, Gramsci dá o nome de orgânico. A unidade formada corresponde a uma organização social concreta, isto é, o próprio bloco histórico (Ibid., pp. 1-15). Para Gramsci, “o aspecto essencial da hegemonia da classe dirigente reside em seu monopólio intelectual, isto é: na atração que seus próprios representantes suscitam nas demais camadas de intelectuais”. É preciso que a classe dirigente possua uma “verdadeira política para os intelectuais”, que subordine a si os intelectuais dos outros grupos sociais3, criando um sistema de solidariedade entre todos (Portelli, 1977, pp. 65- 66). O bloco é histórico, porque não é eterno. Possui início e fim, e sua duração está diretamente relacionada à capacidade de manutenção da hegemonia por parte de uma classe fundamental. Um bloco histórico não pode existir sem uma classe social hegemônica. Onde a classe hegemônica é dominante, o Estado mantém a coesão dentro do bloco, através da propagação de uma cultura comum. Os intelectuais desempenham papel chave nesse processo. Este é o objeto de estudo da próxima subseção (Cox, 1993, pp. 56-57). Por isso, as classes dominantes tentam permanentemente integrar os intelectuais dos grupos subalternos à classe política, particularmente através do recurso ao transformismo4. Tal estratégia traduz a política da classe dominante que recusa assumir qualquer compromisso com as classes subalternas. Nesse aspecto, a absorção das elites dos grupos opositores conduz ao que Gramsci denomina “decapitação pacífica” dos grupos inimigos. Isso explica, em parte, a vulnerabilidade das classes subalternas dentro de um sistema hegemônico. Uma vez ingresso num partido político de determinado grupo social do qual 1.1 Bloco histórico No que tange à estrutura do bloco histórico, esta se define basicamente pelo mundo da produção e pelo conjunto das forças sociais nele existente. Os agrupamentos sociais são formados com base no grau de desenvolvimento das forças materiais de produção, cada grupo representando uma função no processo produtivo. Nesse sentido, a classe dirigente encontra-se em uma “situação de dupla preeminência: a nível estrutural, porque é a classe fundamental no campo econômico; a nível superestrutural, porque tem a direção ideológica através do bloco Segundo Portelli (1977), bloco histórico é o conceito-chave para a compreensão da teoria de Gramsci. O termo refere-se a um sistema constituído por duas esferas interrelacionadas: a estrutura sócio-econômica e a superestrutura político-ideológica. Cox (1993) qualifica essas esferas como objetiva e subjetiva, e afirma que, devido a essa formação, o bloco histórico é um conceito dialético cujos elementos estão em constante interação (Portelli, 1977, pp. 1-15; Cox, 1993, p. 56). 4 Dossiê: SimpoRI 2018 originalmente não fazia parte, o intelectual passa a se confundir com os intelectuais orgânicos desse grupo, e assim sua “consciência de classe” é minada (Ibid., pp. 71-94). Hegemonia é um conceito utilizado para analisar as relações de forças numa sociedade específica. Para Gramsci, uma ordem hegemônica é aquela em que o consentimento, ao invés da coerção, caracteriza as principais relações entre classes, e entre o Estado e a sociedade civil (Gill e Law, 1993, p. 93). O poder hegemônico se expressa pela habitualização e internalização de práticas sociais. (Morton, 2007) . No entanto, é importante destacar que “não existe sistema social em que o consentimento seja a base exclusiva da hegemonia, nem Estado em que um mesmo grupo possa, somente por meio da coerção, continuar a manter de forma durável a sua dominação” (Portelli, 1977, pp. 32-34). Dentro do sistema hegemônico, os intelectuais se agrupam em um “bloco intelectual” (ou “bloco ideológico”). Cada função nele desempenhada (econômica, coercitiva, ideológica) contribui, de seu modo, para a unidade da classe fundamental. O bloco intelectual, em última instância, desenvolve e dirige o sistema hegemônico. Portanto, a destruição do bloco histórico passa, necessariamente, pela desagregação do bloco intelectual e coincide com o colapso da própria hegemonia (Ibid., pp. 87- 102). Na perspectiva gramsciana, o Estado nacional7 é a entidade básica de análise. Trata-se do local em que relações sociais são estabelecidas, conflitos sociais nascem e hegemonias são construídas. Consequentemente, a formação de um bloco histórico é um fenômeno essencialmente nacional. Entretanto, é também possível aplicar a teoria de Gramsci ao âmbito internacional. Isto porque uma hegemonia pode exceder os limites do território nacional e transnacionalizar-se, mediante atuação da classe fundamental, consonância das classes dominantes locais e incorporação de ideologias particulares caracterizadas como universais8 (Cox, 1993, pp. 58-61; Morton, 2007, p. 121). 7 Em seu sentido ampliado, que abrange fusão orgânica entre Estado e sociedade civil (Gill e Law, 1993, p.93). 8 Morton (2007) define hegemonia como a articulação e justificação de um conjunto particular de interesses como interesses gerais. Trata-se de administrar a ordem pelo consentimento amplo resultante de uma aparente universalidade dos princípios e interesses envolvidos (Morton, 2007, p. 113). 5 A referida publicação consiste em um boletim anualmente elaborado. Para fins comparativos, horário semelhante ao de 2018 só foi declarado em 1953, em meio ao desenvolvimento da bomba de hidrogênio pela União Soviética no contexto da Guerra Fria (Bulletin of the Atomic Scientists, 2018). 6 Para Lenin, a hegemonia é política, “porque a sociedade política é mais importante, em suas preocupações estratégicas, do que a civil; assim, desta ele só retém o aspecto político” (Portelli, 1977, p. 65). 1.2 Bloco Intelectual Como anteriormente mencionado, o bloco histórico é formado por uma estrutura social diretamente relacionada às forças produtivas, e também por uma superestrutura de natureza ideológica e política. O vínculo orgânico entre essas duas esferas é realizado pela ação de certos grupos sociais, cuja função primordial não é operar ao nível econômico (estrutural), mas sim ao nível superestrutural. Esses grupos, “funcionários da superestrutura”, são os chamados intelectuais orgânicos (Portelli, 1977). Através da atuação desses indivíduos e grupos, um sistema de valores culturais “impregna, penetra, socializa e integra um sistema social”, em benefício de uma classe dirigente fundamental. Forma-se, assim, um sistema hegemônico. E consolida-se um bloco histórico específico (Pizzorno, 1968, p. 166 apud Portelli, 1977, p. 16). 4 Quando a classe dirigente agrega em torno de si os intelectuais de outros grupos sociais a fim de decapitar sua direção ideológica e política (Portelli, 1977, p. 69). O transformismo absorve ideias potencialmente contra- hegemônicas e as fazem consistentes com a doutrina hegemônica (Cox, 1993, p. 63). 5 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 2 O discurso negacionista Esclarecidos os conceitos e pressupostos teóricos básicos, torna-se agora possível analisar o discurso negacionista sob a perspectiva gramsciana. Examinado como discurso, para fins didáticos, primeiramente faz-se necessário explorar o negacionismo em si, com base nos elementos básicos da comunicação (a saber: mensagem, emissor e receptor). 1.3 Hegemonia Dossiê: SimpoRI 2018 contraditórios através de um processo de síntese política9 (Gill, 2008, p. 194). Hegemonia em nível internacional não é meramente uma ordem entre Estados. É uma ordem dentro da economia mundial, com um modo de produção dominante que penetra em todos os países e liga-se a outros modos de produção subordinados. É também um complexo de relações sociais internacionais que conecta as classes sociais de diferentes países. Hegemonia mundial é descritível como uma estrutura social, uma estrutura econômica e uma estrutura política; e não pode ser simplesmente uma dessas coisas, mas deve ser todas as três. A hegemonia mundial, além disso, é expressa em normas universais, instituições e mecanismos que estabelecem regras gerais de comportamento para Estados e para aquelas forças da sociedade civil que atuam entre fronteiras nacionais - regras que amparam o modo de produção dominante (Cox, 1993, pp. 61-62; tradução nossa). 1.3 Hegemonia Segundo Gruppi (1970), é possível perceber vínculos estreitos entre o conceito de hegemonia desenvolvido por Lenin e a definição proposta por Gramsci5. Entretanto, essas duas concepções divergem em um ponto essencial: a preeminência da direção ideológica e cultural. “A hegemonia gramscista é a primazia da sociedade civil sobre a sociedade política. A análise leninista é exatamente inversa”6 (Portelli, 1977, pp. 52-65). Gramsci toma de Maquiavel a concepção de poder como centauro: meio fera, meio homem – uma combinação necessária entre força e consentimento. A coerção é sempre latente, mas só é aplicada em casos marginais, desviantes. Na maioria das vezes, a hegemonia é suficiente para assegurar o comportamento social esperado (Cox, 1993, p. 52). Historicamente, hegemonias globais nasceram da expansão de hegemonias internas (nacionais), fundadas pelas classes dominantes nos Estados mais desenvolvidos. Hegemonias transnacionais impactam diretamente a ordem internacional, na medida em que o modelo hegemônico é incorporado e antigas estruturas de poder são abaladas (Cox, 1993, pp. 58-61). 5 A referida publicação consiste em um boletim anualmente elaborado. Para fins comparativos, horário semelhante ao de 2018 só foi declarado em 1953, em meio ao desenvolvimento da bomba de hidrogênio pela União Soviética no contexto da Guerra Fria (Bulletin of the Atomic Scientists, 2018). 6 Para Lenin, a hegemonia é política, “porque a sociedade política é mais importante, em suas preocupações estratégicas, do que a civil; assim, desta ele só retém o aspecto político” (Portelli, 1977, p. 65). 7 Em seu sentido ampliado, que abrange fusão orgânica entre Estado e sociedade civil (Gill e Law, 1993, p.93). 8 Morton (2007) define hegemonia como a articulação e justificação de um conjunto particular de interesses como interesses gerais. Trata-se de administrar a ordem pelo consentimento amplo resultante de uma aparente universalidade dos princípios e interesses envolvidos (Morton, 2007, p. 113). 8 Morton (2007) define hegemonia como a articulação e justificação de um conjunto particular de interesses como interesses gerais. Trata-se de administrar a ordem pelo consentimento amplo resultante de uma aparente universalidade dos princípios e interesses envolvidos (Morton, 2007, p. 113). 6 Para Lenin, a hegemonia é política, “porque a sociedade política é mais importante, em suas preocupações estratégicas, do que a civil; assim, desta ele só retém o aspecto político” (Portelli, 1977, p. 65). 6 VOLUME VII 2018 Dossiê: SimpoRI 2018 Hegemonia em nível internacional não é contraditórios através de um processo de síntese política9 (Gill, 2008, p. 194). 12 Por exemplo, em termos simplistas, a Fox News dissemina a ideologia do Tea Party, segundo a qual a ciência climática resulta de uma construção essencialmente esquerdista, possivelmente atrelada à uma potencial conspiração socialista (Painter, 2011, pp. 8-24). existência e natureza antropogênica) (Norgaard, 2011, p. 11 apud Painter, 2011, p. 19; traduções nossas). década de 1940 exigira que as emissoras de televisão oferecessem cobertura adequada e imparcial para pontos de vista antagônicos nas questões de importância pública. Em grande medida devido a essa revogação, nos Estados Unidos, verifica-se hoje uma presença incomum - em comparação a outros países - do jornalismo opinativo nos canais de ampla difusão. E não existem regulamentações externas a respeito da veracidade das informações - este controle é realizado apenas internamente de acordo com as políticas e as regras de cada empresa. Além disso, a partir de 1996, o afrouxamento gradual das limitações sobre a quantidade de estações que uma empresa podia ter permitiu a formação e o crescimento de “impérios midiáticos”, a exemplo da Fox News e da CNN (Stone e Kuznick, 2015, pp. 292-293; Painter, 2011, p. 35). Entre esses grupos, uma parcela significativa dos indivíduos rejeita o título “negacionista”, preferindo a denominação “cético”. Isso ocorre principalmente em meio ao grupo implicatório. O problema é que o termo “ceticismo” não é desprovido de juízo de valor10. Então, para fins analíticos e concretização dos objetivos propostos neste trabalho, as diferentes perspectivas serão agrupadas ao termo “negacionismo climático” (Painter, 2011, p. 17). Enfim, é importante destacar que o grupo antagonista - os cientistas climáticos (adeptos à visão convencional) - também reivindica uma distinção intragrupo, entre os chamados “radicais” (ou “alarmistas”) e aqueles associados a interpretações menos apocalípticas do fenômeno. Outrossim, as diferenças existentes no interior do grupo, embora reconhecidas, não serão enfatizadas11 (Ibid., p. 20). O ambiente propício ao jornalismo opinativo nos Estados Unidos explica, em parte, a clara politização do debate entre negacionistas e convencionais. Ocorre que políticos e mídia partidária se aliam na construção e difusão do discurso que mais lhes convém12. Uma pesquisa publicada pelas universidades Yale e George Mason (2011) mostrou que apenas 34% dos apoiadores do Tea Party acreditam em um atual processo de aquecimento global, porcentagem esta que equivale a 53% entre os republicanos que não apoiam o Tea Party, e a 78% entre os democratas. Constata-se, desse modo, que o aquecimento global é frequentemente examinado sob a ótica política, em detrimento da visão científica (Painter, 2011, pp. 8-24). 10 Na ciência e no jornalismo, por exemplo, o ceticismo é uma característica bastante positiva, ao passo que a credulidade não é uma aspiração comum nesses âmbitos (Painter, 2011, p. 17). 11 E a veracidade das perspectivas mencionadas, seja convencionais ou negacionistas, foge do escopo deste estudo. 10 Na ciência e no jornalismo, por exemplo, o ceticismo é uma característica bastante positiva, ao passo que a credulidade não é uma aspiração comum nesses âmbitos (Painter, 2011, p. 17). 11 E a veracidade das perspectivas mencionadas, seja convencionais ou negacionistas, foge do escopo deste estudo. 2.1 A mensagem “Negacionismo climático” é aqui empregado em seu sentido mais amplo, abrangendo as diversas formas de ceticismo relacionadas à veracidade do aquecimento global ou ao seu caráter antropogênico. Incluem-se também aqueles que, embora acreditem que o fenômeno seja real e influenciado pela ação humana, questionam a forma como é usualmente representado (crêem, por exemplo, que os cientistas não-negacionistas exageram ao abordar o ritmo e a extensão dos impactos). Finalmente, o termo abarca aqueles que defendem que nenhuma ação humana deva ser tomada a fim de conter o aumento contínuo da temperatura terrestre, ainda que concordem com a representação consensual do fenômeno (Painter, 2011, p. 1). Sob a perspectiva gramsciana, é possível, portanto, utilizar o termo “bloco histórico internacional”. Segundo Gill e Law (1993), isto envolveria uma aliança de forças sociais vista como natural e legítima pela maioria de seus membros (Gill e Law, 1993, p. 97). Nesse sentido, destaca-se o papel exercido pelas “elites globalizantes” – grupo de intelectuais orgânicos e líderes políticos dentro do que se pode atualmente denominar “fração transnacional das classes capitalistas” do mundo. Essas elites localizam-se na interface entre aspectos territoriais e globalizantes da ordem mundial, e buscam reconciliar elementos Kari Norgaard (2011) distingue três grupos de negacionistas: o literal (que afirma que o aquecimento global não está acontecendo), o interpretativo (que acredita que a mudança climática está em voga, mas que a ação humana não é a causa desse processo) e o implicatório (que menospreza o fenômeno, a despeito de acreditar em sua 7 9 Vide Gill (2008). 9 Vide Gill (2008). VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 diretamente a membros do Congresso pelas grandes corporações, principalmente por aquelas alocadas nos ramos vinculados ao petróleo, gás natural e mineração. Ainda, o setor privado exerce influência sobre o Congresso através da atuação dos lobbies. De acordo com o US Center for Responsive Politics, em 2009 existiam mais de 13.000 lobistas oficialmente registrados, que receberam cerca de 3,5 bilhões de dólares em honorários. Segundo o The Center for Public Integrity, o número de lobistas trabalhando por questões relacionadas à mudança climática em 2009 (considerando representantes dos dois lados do debate político) correspondia a um aumento de 300% em relação aos cinco anos antecedentes - porcentagem equivalente a 2.340 lobistas, isto é, quatro lobistas para cada membro do Congresso à época (Ibid., p. 30). 2.3 O receptor O discurso negacionista visa, em última instância, disseminar o ceticismo no seio da opinião pública, de modo a conter qualquer ação governamental em prol da preservação ambiental. De forma geral, o discurso tem obtido êxito nesse quesito, sob o protagonismo de alguns veículos de comunicação de ampla difusão. Como anteriormente ressaltado, o jornalismo estadunidense é fortemente opinativo, e isso gera uma polarização evidente entre os canais adeptos ao negacionismo e aqueles seguidores do consenso científico. A opinião pública torna-se reflexo dessa polarização. Por exemplo, segundo uma pesquisa conduzida pelo World Public Opinion (2010), 60% das pessoas que afirmam assistir diariamente aos noticiários da Fox News acredita que a maioria dos cientistas não concorda que uma mudança climática esteja realmente acontecendo. Essa porcentagem é significativamente inferior entre os espectadores da CNN (25%), e entre os ouvintes da NPR ou da PBS (14%) (World Public Opinion, 2010, p. 23 apud Painter, 2011, p. 13). Por fim, as companhias privadas desempenham papel crucial na formação e difusão das vertentes negacionistas através do financiamento oferecido aos think tanks conservadores13. Nos Estados Unidos, essas organizações alcançaram o status de “academia alternativa”, e é comum que seus representantes sejam tratados como peritos independentes em questões relevantes para as políticas ambientais. Os think tanks empregam quadro interno e comissionado para produzir uma gama de materiais impressos, tais como op-eds, resumos políticos, livros e artigos para revistas. Além disso, seus representantes fazem aparições nas mídias, fornecem depoimentos congressionais e dão discursos a fim de promover o posicionamento conservador14 (Dunlap e Jacques, 2013, p. 701). Em relação à mídia impressa situação análoga se apresenta, por exemplo, na comparação entre os jornais The New York Times (NYT) e Wall Street Journal (WSJ). Em seus editoriais, o NYT tem consistentemente seguido a perspectiva do consenso científico, defendendo que há um aquecimento global antropogênico em processo e que uma ação governamental é urgentemente necessária para combatê-lo. Em oposição, o WSJ tem sustentado que as emissões de carbono podem ter algum impacto sobre o aquecimento global, mas que isso não é um fato suficientemente provado a ponto de justificar uma mudança massiva no uso de energia (Painter, 2011, p. 88). estadunidense. E entre estes, o vínculo com algum think tank pôde ser comprovado em 43, isto é, 65% da quantidade total dos livros publicados nos Estados Unidos. 13 CATO Institute, Competitive Enterprise Institute, American Enterprise Institute, Heritage Foundation, Heartland Institute, Marshall Institute, entre outros, são exemplos desse tipo de organização (Painter, 2011, p. 28; Dunlap e Jacques, 2013, pp. 699-700). 14 Dunlap e Jacques (2013) examinaram uma quantidade de 108 livros com temáticas centradas na defesa do negacionismo climático, no intuito de delinear associações entre essas produções e os think tanks conservadores. A análise restringiu-se aos livros publicados em Língua Inglesa, em primeira edição, entre 1980 e 2010. Entre os livros avaliados, 66 deles são de origem 2.2 O emissor O discurso negacionista estadunidense é elaborado e difundido por meio da atuação de uma série de agentes, dentre os quais se destacam as grandes corporações privadas, o Congresso Nacional, os think tanks, a mass media e os pensadores autônomos. Quanto à mídia, emissor mais direto e eficaz do discurso negacionista, algumas peculiaridades do sistema midiático estadunidense precisam ser mencionadas. Em 1987, a Comissão Federal de Comunicações dos Estados Unidos, com o auxílio do presidente Reagan, revogou a Fairness Doctrine (Princípio da Imparcialidade), que desde a Assim, constitui-se um cenário no qual a mídia, direta ou indiretamente, vincula-se à determinada concepção política de caráter partidário. Por sua vez, no âmbito do Congresso Nacional, os partidos políticos e seus representantes são influenciados pelo setor privado. Isso ocorre através das doações multimilionárias pagas 11 E a veracidade das perspectivas mencionadas, seja convencionais ou negacionistas, foge do escopo deste estudo. 8 VOLUME VII 2018 Dossiê: SimpoRI 2018 b d l d O 13 CATO Institute, Competitive Enterprise Institute, American Enterprise Institute, Heritage Foundation, Heartland Institute, Marshall Institute, entre outros, são exemplos desse tipo de organização (Painter, 2011, p. 28; Dunlap e Jacques, 2013, pp. 699-700). 2.3 O receptor Tais estatísticas evidenciam o significante papel desempenhado pelos think tanks na elaboração e disseminação das vertentes negacionistas (Dunlap e Jacques, 2013, pp. 702-709). 9 naram uma quantidade de 108 livros com o negacionismo climático, no intuito de produções e os think tanks conservadores. publicados em Língua Inglesa, em primeira os livros avaliados, 66 deles são de origem estadunidense. E entre estes, o vínculo com algum think tank pôde ser comprovado em 43, isto é, 65% da quantidade total dos livros publicados nos Estados Unidos. Tais estatísticas evidenciam o significante papel desempenhado pelos think tanks na elaboração e disseminação das vertentes negacionistas (Dunlap e Jacques, 2013, pp. 702-709). estadunidense. E entre estes, o vínculo com algum think tank pôde ser comprovado em 43, isto é, 65% da quantidade total dos livros publicados nos Estados Unidos. Tais estatísticas evidenciam o significante papel desempenhado pelos think tanks na elaboração e disseminação das vertentes negacionistas (Dunlap e Jacques, 2013, pp. 702-709). 14 Dunlap e Jacques (2013) examinaram uma quantidade de 108 livros com temáticas centradas na defesa do negacionismo climático, no intuito de delinear associações entre essas produções e os think tanks conservadores. A análise restringiu-se aos livros publicados em Língua Inglesa, em primeira edição, entre 1980 e 2010. Entre os livros avaliados, 66 deles são de origem 9 VOLUME VII 2018 Dossiê: SimpoRI 2018 Uma pesquisa intitulada Climate Change in h A i Mi d ( ) l % d Fonte: Gallup (2014). Dossiê: SimpoRI 2018 3.1 A dinâmica de classes e o bloco ideológico O bloco histórico foi apresentado como uma articulação interna de uma situação histórica precisa (Portelli, 1977, p. 103). Nesse sentido, é possível afirmar que o bloco hoje vigente é o sistema capitalista neoliberal. Particularmente nos Estados Unidos, esse sistema é orientado pelo liberalismo econômico, favorecendo o exercício da direção e da dominação por parte de uma classe fundamental específica, formada pelas grandes corporações do setor privado. Verifica-se, então, a forte presença do poder corporativo, como explica Chomsky (2017): Dossiê: SimpoRI 2018 aquecimento global, estendeu-se de 18% em 2001 para 44% em 2010. A ampliação dessa lacuna é reflexo do crescimento e do acirramento dos debates envolvendo negacionismo e consenso científico (Painter, 2011, p. 25; McCright e Dunlap, 2011 apud Painter, 2011, p. 25). Analiticamente, é possível afirmar que a polarização no âmbito da opinião pública tem colaborado com os interesses dos defensores assíduos do negacionismo. Isso ocorre porque a dúvida que permeia a sociedade estadunidense dificulta a organização social em benefício de uma ação efetiva destinada a combater o aquecimento global. consiste em um estudo imediato, “fotográfico”, cuja abordagem a respeito da estrutura concretiza-se de maneira indireta, através do exame dos movimentos realizados no âmbito superestrutural. Em suma, trata-se de um estudo essencialmente hipotético (Portelli, 1977, pp. 46-47). Fonte: Gallup (2014). Fonte: Gallup (2014). Uma pesquisa intitulada Climate Change in the American Mind (2013) revelou que 23% dos estadunidenses afirmam ter certeza de que a mudança climática não está acontecendo, enquanto 14% dizem não saber se a mudança está ou não em curso. Entre os que afirmam acreditar em mudança climática e aquecimento global, menos da metade atribui causa antropogênica aos fenômenos (Leiserowitz et al., 2013, p. 5 apud Park, 2015). A pesquisa ainda aponta que a porcentagem de pessoas “muitíssimo” preocupadas com o aquecimento global em 2014 diferia em apenas 1 ponto percentual das estatísticas apresentadas em 1989. A Figura 2 apresenta os valores calculados entre esse intervalo de tempo (Gallup, 2014). A Gallup Poll - uma empresa de pesquisa de opinião estadunidense - desenvolveu, em 2014, um estudo acerca da preocupação dos norte-americanos em relação à mudança climática. Entre as opções a great deal (muitíssimo), a fair amount (em uma medida justa), only a little (só um pouco) e not at all (de modo nenhum), pessoas entrevistadas deveriam responder qual o grau de preocupação delas próprias a respeito do aquecimento global. Os entrevistados foram classificados por idade, nível educacional e partido político simpatizante. Os resultados foram compilados na Figura 1 (Gallup, 2014). Figura 2. Pesquisa Gallup, 1989-2014 Figura 2. Pesquisa Gallup, 1989-2014 Fonte: Gallup (2014). Figura 1. Pesquisa Gallup, 2014 Figura 1. Pesquisa Gallup, 2014 Fonte: Gallup (2014). Fonte: Gallup (2014). O fato é que reconhecer e atribuir relevância à mudança climática implica apoiar medidas como a regulamentação governamental, o aumento de taxas, a interferência do Estado nos mercados e alterações no processo produtivo. Tais medidas constituem agenda oposta aos ideais neoliberais defendidos pelas classes hegemônicas conservadoras. Por essa razão, a mudança climática é frequentemente examinada sob a ótica política, em detrimento da científica. Trata-se de interesses relacionados ao modo de produção e à dinâmica do livre mercado. Em 2011, pesquisadores da Universidade do Estado de Michigan publicaram um estudo segundo o qual a lacuna entre conservadores e liberais, no que tange à crença no 10 10 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 dependentes do financiamento fornecido pelas grandes corporações.15 Desse modo, políticos tornam-se “servos do capital privado”, e representantes dos interesses da classe fundamental (Chomsky, 2017, pp. 85-86). A hegemonia é diretamente dependente do bloco ideológico que, por sua vez, depende da atuação dos intelectuais orgânicos. Neste grupo, é possível considerar os think tanks, a mídia conservadora e até mesmo os políticos que, ao desempenhar função social orgânica como intelectuais representantes da classe dirigente, formalizam- se como “funcionários da superestrutura”. Em síntese: Além disso, essa classe retém para si o exercício da função de hegemonia. Como visto, a hegemonia opera sobre a sociedade civil, e esta pode ser examinada sob três aspectos complementares: como ideologia da classe dirigente, como concepção do mundo e como direção ideológica. É nessa esfera que o discurso negacionista adquire importância em termos práticos e analíticos (Portelli, 1977, pp. 19-22). Atrás das guerras culturais e da mídia estão as corporações. Estas são protegidas pelos legisladores por teias intricadas de arranjos de interesses especiais. As relações entre os legisladores e as corporações estão incrustadas no sistema de financiamento das campanhas eleitorais (Pieterse, 2009, p. 205). O negacionismo pode ser classificado como uma ideologia orgânica constituinte do sistema hegemônico vigente. Isso se explica pelo fato de exercer papel necessário à estrutura, visto que influi diretamente sobre o modo de produção, pois a aceitação desta ideologia permite a operacionalização de um sistema produtivo menos sustentável (em termos ambientais) e mais coerente à valorização do capital. E, assim, consolida-se o monopólio intelectual da classe dirigente, elemento essencial da unidade social e da hegemonia em si. A manutenção do monopólio intelectual depende da capacidade de atração que os intelectuais orgânicos suscitam nas demais camadas de intelectuais, por meio do recurso ao transformismo e à decapitação pacífica dos grupos opositores (Portelli, 1977). No que se relaciona ao negacionismo climático, destaca-se, nesse sentido, a atuação dos intelectuais defensores das vertentes menos radicais, que atraem facilmente a si a parcela mais indecisa e moderada da sociedade. Na esfera da sociedade civil, o negacionismo atua como ideologia da classe dirigente, por exemplo, ao incorporar-se ao que é considerado ciência ou influir sobre o direito. Além disso, contribui à difusão da concepção de mundo, na medida em que sistematiza uma organização social e produtiva apropriada. 15 Por exemplo, Barack Obama recebeu das indústrias de combustíveis fósseis uma quantia de US$ 710.277 entre 2011 e 2012. No mesmo período, essas indústrias despenderam um montante de US$ 4.763.934 em auxílio à campanha de Mitt Romney (Collomb, 2014, p. 3). 3 Gramsci e o negacionismo climático estadunidense Explicado o arcabouço teórico utilizado neste artigo (seção 1), e apresentadas as características essenciais do objeto de estudo (seção 2), faz-se necessária a aplicação de conceitos básicos da teoria gramsciana ao objeto de estudo. Esta análise será realizada a partir de dois elementos: o bloco ideológico (ou intelectual) e o bloco histórico. Acerca do primeiro, examinar-se-á a construção do bloco mediante análise da dinâmica de classes e do vínculo orgânico existente entre estrutura e superestrutura. Sobre o segundo, serão formuladas inferências a respeito das dinâmicas futuramente possíveis perante desafios atualmente evidentes. A hegemonia é temática incorporada a essas questões. O triunfo ideológico das “doutrinas de livre mercado”, seletivas como sempre, aplicou golpes adicionais, na medida em que essas doutrinas se traduziram em desregulamentação, regras de governança corporativa vinculando polpudas recompensas pagas a altos executivos de empresas a lucros de curto prazo e outras decisões políticas afins [...] (Chomsky, 2017, p. 85). Primeiramente, é preciso ressaltar que o aspecto essencial da estrutura é o seu caráter relativamente estático em comparação à superestrutura do bloco histórico. Desse modo, uma fase estrutural não pode ser concretamente analisada antes de haver findado todo o seu processo de desenvolvimento. Durante esse processo, todo estudo deverá constituir-se por hipóteses e ser realizado sob a condição de explicitamente afirmar-se enquanto hipotético. Desse modo, a exposição realizada nesta seção O poder econômico resultante da crescente concentração de renda a nível estrutural, culminou no controle do aparelho de Estado pela classe fundamental, no âmbito da superestrutura. Isso se explica, em parte, pela influência decisiva sobre a sociedade política. Atualmente, os custos das campanhas eleitorais tornam os partidos, tanto o Republicano quanto o Democrata, intensamente 11 11 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 como a National Defense University - ou seja, envolvem a participação de um grupo muito específico de indivíduos (Pieterse, 2009, p. 208). discurso negacionista é, portanto, base relevante do consenso e da manutenção do capitalismo como hoje se apresenta. Trata-se de um recurso ideológico que permite a preservação das relações sociais no âmbito do processo produtivo, de forma a beneficiar a classe dirigente e dar continuidade à sua hegemonia. A linguagem de marketing é um outro mecanismo de manipulação do comportamento. Comunicações estratégicas dão “vantagens extras à força estrutural das corporações”. A linguagem de marketing se tornou rotineira na mídia e na política, e não raro apresenta- se de maneira bastante enganosa. Por exemplo, se um projeto legislativo é intitulado “Iniciativa pelas florestas saudáveis”, seus opositores podem ser acusados de querer florestas poluídas. De fato, o termo “Iniciativa pelas florestas saudáveis” foi criado pelo estrategista eleitoral Frank Luntz para descrever um conjunto de políticas do governo Bush. Na prática, entretanto, tais políticas favoreciam a expansão do desmatamento pela indústria da madeira. Em nada contribuíam para a preservação das florestas (Ibid., p. 209). Por fim, atua como direção ideológica da sociedade, pois além de abranger a ideologia em si, produz estrutura ideológica (organizações que criam e difundem a ideologia, como os think tanks) e material ideológico (instrumentos técnicos de difusão da ideologia, a exemplo da produção literária e midiática). Artifícios diversos são usados a fim de manipular as sociedades civil e política para adequá-las ao sistema hegemônico. A elitização da informação é um exemplo. Há debates restritos às elites e informações que não atingem o público geral. Assuntos de objetivos e estratégia nacionais, por exemplo, são discutidos nos institutos de pesquisa, nas revistas acadêmicas e em foros 12 campanha de Mitt Romney (Collomb, 2014, p. 3). VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 assistido pela remoção das barreiras governamentais à extração de recursos naturais e abrindo caminho para que as empresas transnacionais continuem a atender à crescente demanda global por seus produtos (Ibid.; traduções nossas). sociedade não se dispõe a realizar qualquer atividade cujos os meios necessários sejam inexistentes ou insuficientes. Nesse sentido, a ameaça fundamental ao meio ambiente é o próprio capitalismo. Medidas para combater a devastação ambiental são, por natureza, contrárias à essência de um sistema supostamente autorregulador (Moreira e Estevo, 2018, p. 41; Portelli, 1977, p. 48; Sayre, 2012, p. 66). Há ainda uma segunda condição que torna previsível um futuro colapso do bloco histórico vigente. Ao seguir seu curso evolutivo natural, o capitalismo - como qualquer outro bloco histórico - conduz ao desenvolvimento de todas as formas de vida contidas em suas relações, possibilitando a posterior dissolução ou substituição da sociedade (Portelli, 1977, p. 48). É um “ciclo vicioso e auto-perpetuante”, que eventualmente esgotará todo o capital natural do mundo se continuar inabalável. Logo, o cenário é preocupante. No sistema econômico global, todas as nações estão sujeitas a esse ciclo capitalista, porque cada país procura melhorar sua prosperidade percebida, seu prestígio global e sua riqueza acumulada (medida pelo PIB). Assim, embora grande parte dos países pratique tecnicamente a economia mista, os ideais do capitalismo direcionam as economias nacionais de maneira geral. As nações que participam do comércio global (ou seja, todas as nações da Terra) jogam atualmente de acordo com as regras do neoliberalismo. Os países desenvolvidos, pela atuação de suas elites, têm sido bem sucedidos na incorporação dos países menos desenvolvidos ao sistema hegemônico, criando uma rede de comércio global definida pelas métricas econômicas neoclássicas. Uma consequência da globalização neoliberal, ao menos até a crise econômica mundial de 2008, foi a expansão do livre comércio - que resultou em processos de desregulamentação e aumento do poder corporativo (Park, 2015; Pieterse, 2009, p. 146). A globalização neoliberal foi resultado da evolução do sistema capitalista. Particularmente no que tange ao sistema internacional, entre o imperialismo clássico e a globalização neoliberal, ocorreram mudanças significativas nos processos de internacionalização da produção e do capital. E isto conduziu ao desenvolvimento de formas diversas de relações sociais. Fronteiras foram abertas, novos atores emergiram no cenário internacional, a interdependência foi acentuada e regimes foram consolidados. As sociedades nunca estiveram tão conectadas umas às outras, seja no âmbito político, econômico ou social. 3.2 Inferências sobre o bloco histórico Qualquer análise do bloco histórico deve considerar dois princípios: 1) a sociedade não se dispõe a realizar qualquer tarefa para a qual não existam os meios necessários e suficientes, 2) nenhuma sociedade se dissolve nem pode ser substituída enquanto não tiver desenvolvido todas as formas de vida contidas nas suas relações (Portelli, 1977, p. 48). O bloco histórico capitalista cria dentro de si as condições favoráveis ao seu desmantelamento, pois o sistema capitalista não é projetado para inibir sua propensão ao crescimento insustentável. O crescimento econômico ocorre em função da população, da tecnologia e do consumo de recursos. Desse modo, crescimento econômico e sustentabilidade ambiental são objetivos irreconciliáveis dentro do modelo neoliberal contemporâneo (Park, 2015). Demonstra-se, portanto, que o bloco ideológico desempenha papel primordial na construção e na preservação do sistema hegemônico neoliberal. Contribuem para isso, ideologias orgânicas específicas, dentre as quais o negacionismo climático constitui um exemplo relevante. Em suma, nos Estados Unidos, o bloco histórico corrente é orientado pelo liberalismo econômico, que favorece a hegemonia de uma classe fundamental formada pelas grandes corporações do setor privado. No sistema capitalista vigente, a base social da hegemonia é formada pelos grupos adeptos ao American Way of Life e inseridos na cultura consumista. Essa é a materialização da concepção de mundo e da direção ideológica da classe dirigente no seio da sociedade civil. Esta é uma das grandes ironias das economias capitalistas: quanto mais recursos uma nação usa, mais essa nação cresce. Quanto mais essa nação cresce, menos recursos a nação tem. Quanto menos recursos a nação tem, menos a nação pode crescer, mas quanto maior a nação, mais recursos a nação deve consumir para continuar crescendo. É por isso que o capitalismo está tão intimamente relacionado à globalização. Assim, o livre mercado está em constante expansão, O negacionismo climático adquire importância ao se configurar como uma ideologia orgânica do sistema hegemônico. Em âmbito estrutural, o negacionismo sanciona um sistema produtivo coerente à valorização do capital. Na esfera da sociedade civil, propicia o comportamento social adequado a tal sistema produtivo. O 13 13 VOLUME VII 2018 Dossiê: SimpoRI 2018 16 Introdução de mudanças que não envolvem a ascensão de forças populares. O conceito de revolução passiva é uma contrapartida ao conceito de hegemonia, na medida em que descreve a condição de uma sociedade não- hegemônica – ou seja, situação em que nenhuma classe dominante foi capaz de estabelecer uma hegemonia no sentido gramsciano do termo (Cox, 1993, pp. 54-55). 17 É importante ressaltar que a estratégia de involução passiva é, geralmente, implementada em momentos de crise orgânica. Desse modo, se a estratégia não for bem sucedida, é possível que a crise se intensifique, ao fragilizar ainda mais o vínculo entre estrutura e superestrutura e expor problemas intrínsecos ao sistema. Entretanto, é possível afirmar que o sistema hegemônico vem tentando retardar seu processo evolutivo diante da crise orgânica eclodida em 2008, visando postergar o desmantelamento previsto ao atingir seu desenvolvimento pleno. Portanto, a existência de um bloco histórico transnacional é evidente. A hegemonia estadunidense é uma realidade percebida em meio à globalização neoliberal, pela existência de regras e instituições que sustentam a ordem econômica arquitetada pelos Estados Unidos desde o término da Segunda Guerra Mundial. O modelo hegemônico tem em seu modus operandi as bases de seu próprio colapso. Ao conduzir à devastação ambiental, o capitalismo corroi sua base de reprodução contínua, pois, como já mencionado, a A globalização e o livre comércio eram antes considerados sinais de progresso e relações pacíficas entre os povos, portanto valorados positivamente. Hoje, representantes da potência hegemônica os consideram responsáveis pela instabilidade interna pela qual os Estados Unidos vêm passando. Deste modo, os Estados Unidos vêm, nos últimos anos, adotando políticas protecionistas que, em 14 VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 16 Introdução de mudanças que não envolvem a ascensão de forças populares. O conceito de revolução passiva é uma contrapartida ao conceito de hegemonia, na medida em que descreve a condição de uma sociedade não- hegemônica – ou seja, situação em que nenhuma classe dominante foi capaz de estabelecer uma hegemonia no sentido gramsciano do termo (Cox, 1993, pp. 54-55). tese, opõem-se aos princípios neoliberais anteriormente prezados. As elites governantes investem no “simbolismo da megapotência”, o que implica roteiro unipolar de não- cooperação com tratados internacionais - paradigma oposto a tendências que prevalecem na Europa e outras regiões do mundo (Pieterse, 2009, pp. 130-149). A estratégia de postergar a derrocada do bloco histórico através da retirada de determinados progressos consolidados, como a globalização econômica e as políticas ambientais, é denominada “involução passiva”. Através dela, o capitalismo retornaria a um ponto anterior de seu modelo de acumulação, o que permitiria um ciclo progressivo futuro (nova revolução passiva16 conduzida pela classe dirigente). Isso retardaria o desenvolvimento pleno do sistema capitalista e, consequentemente, a sua derrocada (Curro, 2014, pp. 94-102). Seguindo perspectiva análoga, é possível que o negacionismo climático seja também interpretado como uma tentativa de conter o processo evolutivo do bloco histórico, retardando seu desmantelamento frente à crise orgânica de 2008. O estudo de Dunlap e Jacques (2013) dão base à essa visão, por exemplo, ao constatar o aumento excepcional das publicações de livros com temáticas negacionistas entre 2007 e 2009, conforme mostrado na Figura 3 (Curro, 2014, pp. 94-102; Dunlap e Jacques, 2013, p. 704). No entanto, a implementação da estratégia de “involução passiva” não ocorre sem efeitos potencialmente desastrosos para o sistema hegemônico. Afinal, objetiva-se salvar a hegemonia por alterações em seu próprio ciclo evolutivo, já que a hegemonia está diretamente atrelada ao bloco histórico. Na prática, isto pode tanto retardar quanto acelerar o colapso do bloco17, a depender de fatores muitas vezes não diretamente controlados pelas classes dominantes. Figura 3. Livros de negacionismo climático - número de publicações por ano (1982-2010) Fonte: Dunlap e Jacques (2013) Figura 3. Livros de negacionismo climático - número de publicações por ano (1982-2010) O negacionismo climático impulsiona a devastação ambiental e opõe-se ao que vem sendo defendido por um grande número de atores internacionais. Além disso, políticas unilaterais e protecionistas são vistas com desconfiança, afetando a base social sobre a qual a hegemonia se sustenta. “Quando os EUA se colocam do lado de fora do direito internacional e das instituições internacionais e renunciam à pretensão da legitimidade, o que resta é o governo pela força”. Ou seja, neste cenário, resta somente a coerção. A hegemonia é perdida. E ao Fonte: Dunlap e Jacques (2013) 16 Introdução de mudanças que não envolvem a ascensão de forças populares. 17 É importante ressaltar que a estratégia de involução passiva é, geralmente, implementada em momentos de crise orgânica. Desse modo, se a estratégia não for bem sucedida, é possível que a crise se intensifique, ao fragilizar ainda mais o vínculo entre estrutura e superestrutura e expor problemas intrínsecos ao sistema. desconsiderarem instituições internacionais e aliados, os Estados Unidos fornecem às elites governantes uma opção de desvinculação (Pieterse, 2009, p. 129). país do Acordo. A saída deverá efetivar-se em novembro de 2020 (Moreira e Estevo, 2018, p. 35). As escolhas dos aliados internos de Trump reforçaram a inclinação da atual gestão contra políticas ambientais de combate às mudanças climáticas. O secretário de energia nomeado foi o ex-governador do Texas Rick Perry, ligado à indústria de combustíveis fósseis e cético em relação à responsabilidade das ações humanas no aquecimento global. Scott Pruitt, administrador da Agência de Proteção Ambiental entre fevereiro de 2017 e julho de 2018, é também um negacionista declarado (Banerkjee, 2017 apud Moreira e Estevo, 2018, p. 41). O desmantelamento da ordem hegemônica faz parte do comportamento da hegemonia ao longo do tempo. O bloco histórico não pode durar eternamente, e sua desintegração requer a criação de um novo sistema hegemônico. Segundo Cox (1999), esse novo sistema originar-se-ia na sociedade civil, por meio das classes subalternas. Essas classes são aquelas excluídas da base social do sistema hegemônico, sobre as quais a classe dirigente exerce a dominação (e não a direção). No contexto corrente, os subalternos são basicamente os grupos sociais excluídos da cultura consumista. Em âmbito mundial, são as classes ligadas apenas parcialmente ao capitalismo global18. Nesse sentido, as comunidades indígenas constituem um exemplo apropriado19 (Cox, 1999; Gill, 2008, p. 197). Os Estados Unidos foram até 2007 os maiores emissores globais de gases de efeito estufa, sendo ultrapassados pela China naquele ano. Contudo, em termos per capita, suas emissões são ainda as maiores do mundo. O país é parte significativa do problema do aquecimento global, mas, atualmente, não é parte da solução. A classe fundamental estadunidense não tem na mudança climática um dos temas prioritários de sua agenda. Acontece que o combate ao aquecimento global exige mudanças nas formas de produção e consumo (Moreira e Estevo, 2018, pp. 40-43). Em última instância, requer o colapso das relações orgânicas que sustentam o sistema hegemônico, pois o caminho para a solução não pode ser o mesmo que conduziu ao problema. 18 É preciso, no entanto, identificar a formação objetiva das classes sociais analisando desenvolvimentos e transformações dentro da esfera da produção, mas sem sucumbir ao economicismo. Ou seja, é necessário também considerar as mentalidades e ideologias, a afiliação ativa ou passiva das classes subalternas às formas sociais dominantes de associação política, e, principalmente as formações políticas produzidas por essas próprias classes (Morton, 2007, p. 174). 19 O conceito de revolução passiva é uma contrapartida ao conceito de hegemonia, na medida em que descreve a condição de uma sociedade não- hegemônica – ou seja, situação em que nenhuma classe dominante foi capaz de estabelecer uma hegemonia no sentido gramsciano do termo (Cox, 1993, pp. 54-55). Fonte: Dunlap e Jacques (2013) 17 É importante ressaltar que a estratégia de involução passiva é, geralmente, implementada em momentos de crise orgânica. Desse modo, se a estratégia não for bem sucedida, é possível que a crise se intensifique, ao fragilizar ainda mais o vínculo entre estrutura e superestrutura e expor problemas intrínsecos ao sistema. 15 15 VOLUME VII 2018 Dossiê: SimpoRI 2018 VOLUME VII 2018 Dossiê: SimpoRI 2018 desconsiderarem instituições internacionais e aliados, os Estados Unidos fornecem às elites governantes uma opção d d i l ã (Pi ) país do Acordo. A saída deverá efetivar-se em novembro de 2020 (Moreira e Estevo, 2018, p. 35). Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 Conclusão Em 2015, nações do mundo inteiro uniram-se na Conferência das Partes (COP-21), em Paris, em torno de uma causa comum: as mudanças climáticas. Como resultado, os 195 Estados participantes formalizaram o Acordo de Paris, objetivando limitar o aumento da temperatura média global através da redução da emissão de gases de efeito estufa. Os Estados Unidos assinaram o Acordo ao final do governo Obama, em 2016. No entanto, no dia 1 de junho de 2017, o atual presidente - Donald Trump - anunciou a retirada do Portanto, o século XXI enfrenta desafios ecológicos “para os quais a hegemonia é irrelevante ou contraproducente” (Pieterse, 2009, p. 162). A nível global, os Estados Unidos apresentam-se como centro hegemônico. E este centro resiste às ameaças ao bloco histórico vigente. O negacionismo climático insere-se nesta dinâmica ao constituir uma base importante para a manutenção do sistema capitalista - o bloco histórico contemporâneo - como hoje se apresenta. Em âmbito estrutural, o negacionismo sanciona um sistema produtivo coerente à valorização do capital. Na esfera da sociedade civil, propicia o consumismo. Trata-se, portanto, de um recurso ideológico (discurso, ideia, 19 Vide Inoue e Moreira (2016). 19 Vide Inoue e Moreira (2016). 16 16 VOLUME VII 2018 Dossiê: SimpoRI 2018 VOLUME VII 2018 Dossiê: SimpoRI 2018 concepção de mundo) que contribui à preservação das relações sociais em meio ao processo produtivo, de forma a beneficiar a classe fundamental e assegurar sua hegemonia. Entretanto, é possível que o discurso negacionista não se sustente diante dos sinais diariamente evidentes, do desenvolvimento da pesquisa científica e da atuação de ativistas ambientais. O futuro dirá até quando o bloco capitalista continuará se reestruturando diante das eventuais crises orgânicas, e quando se dará o seu desmantelamento. A construção e o colapso de qualquer bloco histórico constituem processo diretamente atrelado à questão da hegemonia. Certamente, o discurso negacionista acompanhará esse processo. VOLUME VII 2018 Dossiê: SimpoRI 2018 Dossiê: SimpoRI 2018 concepção de mundo) que contribui à preservação das relações sociais em meio ao processo produtivo, de forma a beneficiar a classe fundamental e assegurar sua hegemonia. atuação de ativistas ambientais. O futuro dirá até quando o bloco capitalista continuará se reestruturando diante das eventuais crises orgânicas, e quando se dará o seu desmantelamento. A construção e o colapso de qualquer bloco histórico constituem processo diretamente atrelado à questão da hegemonia. Certamente, o discurso negacionista acompanhará esse processo. Entretanto, é possível que o discurso negacionista não se sustente diante dos sinais diariamente evidentes, do desenvolvimento da pesquisa científica e da 17 VOLUME VII 2018 Dossiê: SimpoRI 2018 BIBLIOGRAFIA VOLUME VII 2018 Dossiê: SimpoRI 2018 BIBLIOGRAFIA Bulletin of the Atomic Scientists (2018), “Timeline”, disponível em https://thebulletin.org/doomsday-clock/past-announcements/ (acesso em 09 jul. 2018). Chomsky, N. (2017), Quem manda no mundo?, São Paulo, Planeta. Chomsky, N. (2017), Quem manda no mundo?, São Paulo, Planeta. Collomb, J-D. (2014), “The Ideology of Climate Change Denial in the United States”, European Journal Of American Studies, Vol. 9, pp.1-20. Cox, R.W. (1999), “Civil society at the turn of the millenium: prospects for an alternative world order”, Review Of International Studies, Vol. 25, pp. 3-28. ______ (1993), “Gramsci, hegemony and international relations: an essay in method”, in Smith, S. et al. Gramsci, Historical Materialism and International Relations, Cambridge, Cambridge University Press, pp. 49-66. Curro, C.A.G. (2014), 'O pensamento de Antonio Gramsci e as possibilidades de revolução passiva ou efetiva no capitalismo do século XXI', dissertação de mestrado, Universidade Federal de Uberlândia, Uberlândia. Dunlap, R.E. e Jacques, P.J. (2013), “Climate Change Denial Books and Conservative Think Tanks”, American Behavioral Scientist, Vol. 57, pp. 699-731. Gallup (2014), “Americans Show Low Levels of Concern on Global Warming”, disponível em https://news.gallup.com/poll/168236/americans-show-low-levels-concern-global-warming.asp Gallup (2014), “Americans Show Low Levels of Concern on Global Warming”, disponível em https://news.gallup.com/poll/168236/americans-show-low-levels-concern-global-warming.aspx (acesso em 10 jul. 2018). gallup.com/poll/168236/americans-show-low-levels-concern-global-warming.aspx (acesso em 10 jul. 2018 Gill, S. (2008), Power and Resistance in the New World Order, Nova Iorque, Palgrave Macmillan. ower and Resistance in the New World Order, Nova Iorque, Palgrave Macmillan. Gill, S. e Law, D., (1993), “Global hegemony and the structural power of capital”, in Smith, S. et al. Gramsci, Historical Materialism and International Relations. Cambridge, Cambridge University Press, pp. 93-126. Inoue, C.Y.A. e Moreira, P.F. (2016), “Many worlds, many nature(s), one planet: indigenous knowledge in the Anthropocene”, Revista Brasileira de Política Internacional, Vol. 59, pp.1-19. Konrad, H. et al. (2018), “Net retreat of Antarctic glacier grounding lines”, Nature Geoscience, Vol. 11, pp. 258-262. Moreira, H.M. e Estevo, J.S. (2018), “A política dos EUA para as mudanças climáticas: análise da saída do Acordo de Paris”, Conjuntura Internacional, Vol. 14, pp. 32-45. Moreira, H.M. e Estevo, J.S. (2018), “A política dos EUA para as mudanças climáticas: análise da saída do Acordo de Paris”, Conjuntura Internacional, Vol. 14, pp. 32-45. Moreira, H.M. e Estevo, J.S. (2018), “A política dos EUA para as mudanças climáticas: análise da saída do Acordo de Paris”, Conjuntura Internacional, Vol. 14, pp. 32-45. Morton, A.D. (2007), Unravelling Gramsci: Hegemony and Passive Revolution in the Global Political Economy, Londres, Pluto Press. Morton, A.D. (2007), Unravelling Gramsci: Hegemony and Passive Revolution in the Global Political Economy, Londres, Pluto Press. Painter, J. (2011), Poles Apart: The international reporting of climate scepticism, Oxford, Universidade de Oxford. Painter, J. (2011), Poles Apart: The international reporting of climate scepticism, Oxford, Universidade de Oxford. Park, J.T. (2015), “Climate Change and Capitalism”, Consiliense: The Journal of Sustainable Development, Vol. 14, pp. 189-206. 18 Park, J.T. (2015), “Climate Change and Capitalism”, Consiliense: The Journal of Sustainable Development, Vol. 14, pp. 189-206. 18 VOLUME VII 2018 Dossiê: SimpoRI 2018 Pieterse, J.N. (2009), O fim do império americano?: os Estados Unidos depois da crise, São Paulo, Geração Editorial. Portelli, H. (1977), Gramsci e o Bloco Histórico, Rio de Janeiro, Paz e Terra. Sayre, N.F. (2012), “The Politics of the Anthropogenic”, Annual Review of Anthropology, Vol. 41, pp. 57-70. Stone, O. e Kuznick, P. (2015), A história não contada dos Estados Unidos, São Paulo, Faro Editorial. VOLUME VII 2018 Dossiê: SimpoRI 2018 Pieterse, J.N. (2009), O fim do império americano?: os Estados Unidos depois da crise, São Paulo, Geração Editorial. Dossiê: SimpoRI 2018 Pieterse, J.N. (2009), O fim do império americano?: os Estados Unidos depois da crise, São Paulo, Geração Editorial. Portelli, H. (1977), Gramsci e o Bloco Histórico, Rio de Janeiro, Paz e Terra. Sayre, N.F. (2012), “The Politics of the Anthropogenic”, Annual Review of Anthropology, Vol. 41, pp. 57-70. Stone, O. e Kuznick, P. (2015), A história não contada dos Estados Unidos, São Paulo, Faro Editorial. 19
https://openalex.org/W2035146075	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0041200&type=printable	English	null	Socioeconomic and Nutritional Factors Account for the Association of Gastric Cancer with Amerindian Ancestry in a Latin American Admixed Population	PloS one	2,012	cc-by	7,507	Abstract s is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherw y lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Funding: Fogarty International Center and National Cancer Institute (5R01TW007894) funded this study. The study and its participants also received funding and fellowships from the following Brazilian agencies: Brazilian National Research Council, Ministry of Education, Ministry of Health (PNPD-Sau´de Program), and the Minas Gerais State Reaearch Agency. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: edutars@icb.ufmg.br * E-mail: edutars@icb.ufmg.br Socioeconomic and Nutritional Factors Account for the Association of Gastric Cancer with Amerindian Ancestry in a Latin American Admixed Population Latife Pereira1, Roxana Zamudio1, Giordano Soares-Souza1, Phabiola Herrera2, Lilia Cabrera2, Catherine C. Hooper3, Jaime Cok4, Juan M. Combe5, Gloria Vargas6, William A. Prado7, Silvana Schneider8, Fernanda Kehdy1, Maira R. Rodrigues1, Stephen J. Chanock9, Douglas E. Berg10, Robert H. Gilman2,3,11, Eduardo Tarazona-Santos2* 1 Departamento de Biologia Geral, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 2 Asociacio´n Bene´fica PRISMA, Lima, Peru, 3 Laboratorios de Investigacion y Desarrollo, Facultad de Ciencias, Universidad Peruana Cayetano Heredia, Lima, Peru, 4 Departamento de Patologı´a, Hospital Nacional Cayetano Heredia, Lima, Peru, 5 Departamento de Gastroenterologia, Instituto Nacional de Enfermedades Neopla´scas, Lima, Peru, 6 Servicio de Gastroenterologia, Hospital Nacional Arzobispo Loayza, Lima, Peru, 7 Servicio de Gastroenterologia, Hospital Dos de Mayo, Lima, Peru, 8 Departamento de Estatı´stica, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 9 Laboratory of Translational Genomics of the Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Gaithersburg, Maryland, United States of America, 10 Department of Molecular Microbiology, Washington University Medical School, St Louis, Missouri, United States of America, 11 Department of International Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States of America Abstract Gastric cancer is one of the most lethal types of cancer and its incidence varies worldwide, with the Andean region of South America showing high incidence rates. We evaluated the genetic structure of the population from Lima (Peru) and performed a case-control genetic association study to test the contribution of African, European, or Native American ancestry to risk for gastric cancer, controlling for the effect of non-genetic factors. A wide set of socioeconomic, dietary, and clinic information was collected for each participant in the study and ancestry was estimated based on 103 ancestry informative markers. Although the urban population from Lima is usually considered as mestizo (i.e., admixed from Africans, Europeans, and Native Americans), we observed a high fraction of Native American ancestry (78.4% for the cases and 74.6% for the controls) and a very low African ancestry (,5%). We determined that higher Native American individual ancestry is associated with gastric cancer, but socioeconomic factors associated both with gastric cancer and Native American ethnicity account for this association. Therefore, the high incidence of gastric cancer in Peru does not seem to be related to susceptibility alleles common in this population. Instead, our result suggests a predominant role for ethnic-associated socioeconomic factors and disparities in access to health services. Since Native Americans are a neglected group in genomic studies, we suggest that the population from Lima and other large cities from Western South America with high Native American ancestry background may be convenient targets for epidemiological studies focused on this ethnic group. Citation: Pereira L, Zamudio R, Soares-Souza G, Herrera P, Cabrera L, et al. (2012) Socioeconomic and Nutritional Factors Account for the Association of Gastric Cancer with Amerindian Ancestry in a Latin American Admixed Population. PLoS ONE 7(8): e41200. doi:10.1371/journal.pone.0041200 Editor: Zongli Xu, National Institute of Environmental Health Sciences, United States of America Received March 1, 2012; Accepted June 18, 2012; Published August 3, 2012 icle, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for work is made available under the Creative Commons CC0 public domain dedication. This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, o any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Introduction Recavarren-Arce et al. and Correa proposed a progression model for the development of intestinal-type gastric adenocarci- noma, which consist of a transition from superficial gastritis to metaplasia to dysplasia and finally, gastric adenocarcinoma [2,3]. A plethora of socio-economic, environmental, and dietary factors modulate this progression and the individual risk of ultimately developing gastric cancer. Chronic infection of the stomach by the bacterium Helicobacter pylori leading to chronic inflammation is a major attributable risk factor [4], although less than 2% of H. pylori carriers develop gastric cancer [5]. Helicobacter pylori diversity also affects the risk of host gastric cancer, and the presence of the bacterial virulence factor cag+ is one of the most relevant risk factors. While this virulence factor has a frequency of ,60% in European and US populations, it attains more than 90% in the Gastric cancer is one of the most lethal types of cancer, accounting for approximately 800,000 deaths per year, but its incidence varies substantially worldwide [1]. The highest incidence of gastric cancer is observed in East Asia, Eastern Europe, and the Andean region of South America. Indeed, in the Peruvian population, gastric cancer ranks second in incidence among men and third among women (22.6 and 20 cases per 100,000 males and females respectively), being the type of cancer with the highest mortality. Comparatively, the incidence of gastric cancer in Peru is approximately five times higher than in the United States and twice that observed in Brazil [1]. PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 August 2012 \| Volume 7 \| Issue 8 \| e41200 1 Ancestry and Gastric Cancer in Peru intestinal metaplasia (n = 46) were included as controls on the assumption that this type of metaplasia does not increase the risk of developing gastric cancer [23,24,25]. Peruvian population [6]. Also, while Native American individuals from isolated populations are infected by mostly native strains that resemble Asian strains, due to the Pleistocene Asian origin of Native Americans, individuals living in medium and large urban centers, even if they may have a predominant Native American ancestry, are infected by largely European or hybrid strains brought to the Americas after the 15th century. These strains had largely replaced less virulent or vigorous native strains [7,8]. Introduction If there is a human genetic basis for the high incidence of gastric cancer in the Andean region, we expect the admixed population from Lima to harbor genetic variants accounting for this high incidence, and if these variants were more common in the Native American genetic background of this population, it would be possible to use the genome-wide strategy of admixture mapping to discover these variants. Admixture mapping studies have recently helped to identify variants associated with prostate cancer [22] in African-American popu- lations, but this approach is yet to be fully applied to Latin American or Latino/Hispanic US populations. In this context, the goals of this case-control genetic association study are: (i) to assess the ethnic composition and its related genetic structure of patients attending large hospitals in Lima; (ii) to test if individual Native American, European and African ancestries are risk factors for gastric cancer, controlling for the effect of non-genetic factors (i.e., socioeconomic, nutritional, and clinical). Socioeconomic, dietary, and clinical information was collected for each participant in the study and ancestry was estimated based on 103 ancestry- informative markers (AIMs). We determined that higher Native American individual ancestry is associated with gastric cancer, but that socioeconomic factors associated both with gastric cancer and ethnicity account for this association. Despite the high incidence of gastric cancer among Peruvians with predominantly Native American ancestry, our results do not point to a clear genetic basis for this discrepancy in incidence. Rather, they suggest a predominant role for ethnic-associated socioeconomic and human ecologic factors and disparities in access to health services. As expected, variables that are proxies for poverty (low education level, home quality characteristics such as the use of low quality materials, lack of good appliances, and poor sanitary conditions) were associated with gastric cancer (Table 1). Also, some digestive-related symptoms such as burning (p,0.0001), nausea (p,0.0001), vomiting (p,0.0001), and heaviness (p,0.0001) were more frequent in cases than in controls, but these symptoms are likely a consequence of the disease (Table 1). The variable age was not normally distributed (p,0.01, Kolmogorov–Smirnov test), the medians (and deviation inter- quartile) for controls and cases were 61 (Standard deviation (SD): 21) and 65 (SD: 25) years respectively (p = 0.042, Mann-Whitney test); therefore, this variable was considered as covariate in further logistic regression analyses. Introduction p g g We used a validated set of 103 AIMs [26] to estimate Native American, European and Native American individual ancestry for each of the 241 gastric cancer cases and 300 controls recruited for this study. The Principal Component Analysis (PCA, Figure 1) of the individual genotypes for our Peruvian samples (including 296 Native Americans) and for European, African, and Mexican individuals from HapMap-III shows that the AIMs used discrim- inate between African, European, and Peruvian Native American parental populations, and that the admixed Mexicans (resident in Los Angeles) and the Peruvian gastric cases and controls attending Lima hospitals are placed between Europeans and Native Americans. Moreover, Peruvian gastric cases and controls are relatively closer to the Peruvian Native American parental populations. Although the urban population from Lima is considered as mestizo (i.e., typically admixed), the groups studied herein showed a very high Native American ancestry (78.4% for the cases and 74.6% for the controls, Box in Figure 1), with a low African ancestry (,5%, Figure S1). Interestingly, there is a positive association between Native American ancestry and gastric cancer (logistic regression, OR = 3.69, 95%CI of the OR: 1.34- 10.09, p = 0.011, R2 = 0.016), and consequently a negative association with European ancestry. Poverty also correlates with gastric adenocarcinoma [9] and while elevated consumption of processed or smoked food and salt are risk factors, frequent intake of fresh fruits and vegetables is protective [10,11,12,13]. Human genetic diversity is also relevant [14]. The observed differences in the incidence of gastric cancer worldwide may be due to environmental factors or to the presence of susceptibility genetic variants that are more frequent in populations with high incidence of the disease, but the identifi- cation and discrimination of these factors is challenging. Although common susceptibility genetic variants have been identified in European and Chinese populations by genome-wide and candi- date-gene association studies in genes such as PLCE [15], IL1B [14], IL8 [16,17,18], IL1RN [19], and PTGS2 [20], these variants account for a small portion of the genetic variance associated with sporadic gastric cancer. Peru, with its high incidence of gastric cancer, has the largest Native American population in South America [21] and large cities such as Lima are populated by people classified as mestizo (i.e., individuals with admixture from Africans, Europeans, and Native Americans). Introduction We synthesized the wide set of non-genetic variables collected in cases and controls using a multivariate factor analysis, to reduce the dimensionality of these 43 non-genetic variables by capturing the correlation among them (Table 1 and Table S2). Interestingly, the first factor (16.38% of the total variance) is dominated by socioeconomic variables but includes a subset of correlated nutritional variables, higher values of the factor corresponding to wealthier conditions. The second factor (5.95% of total variance) includes subsets of socioeconomic and nutritional variables, as well as complaints, such as pain. The third factor (5.26% of the total variance) is dominated by digestive-related symptoms. We used the individual coordinates for each of these three factors to synthetically represent the original set of 43 variables. Both the first ‘‘socioeconomic’’ factor and the second factor are associated with gastric cancer (OR = 0.68, p = 0.0002 and OR = 0.7, p = 0.0008, respectively) and also with ethnicity (p = 0.02 and p = 0.00003 respectively, where higher values of the factor correspond to better socioeconomic conditions). The third ‘‘digestive symptoms’’ factor is associated with gastric cancer (OR = 2.16, p,0.0001). PLoS ONE \| www.plosone.org Results We recruited individuals attending Gastroenterology Divisions and prescribed for an endoscopy in three large hospitals in Lima (Table S1). Cases were adults referred for endoscopy and whose biopsies were confirmed positive for gastric cancer by histopath- ological analyses. The control group was composed of individuals whose biopsies were negative for gastric cancer. Individuals with The observed association between Native American ancestry and gastric cancer may be due to the effect of confounding socioeconomic or nutritional variables associated with both gastric cancer and ancestry. It is well known that high Native American or African ancestry is associated with poverty in many populations in PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 August 2012 \| Volume 7 \| Issue 8 \| e41200 2 Ancestry and Gastric Cancer in Peru Figure 1. Principal Component Analysis of our Peruvian samples of gastric cancer cases, their controls, and Native Americans in the context of HapMap-III European, African, and Mexican individuals, and distribution of Native American ancestry in cases and controls (box). Each individual was genotyped for 103 ancestry informative markers validated by Yaeger et al. (2008). We represent the first (horizontal) and second (vertical) principal components, which capture the 35.3% and 7.7% of total variance, respectively. HapMap individuals: YRI: Yoruba from Nigeria, LWK: Luhya from Kenya, ASW: African American from Southwest USA, MKK: Maasai from Kenya, CEU: Utah residents with European ancestry, TSI: Toscani from Italy, MEX: Mexican ancestry resident in Los Angeles. Peruvian Native Americans: Shimaa (SHI) and Ashaninkas (ASH) from the Matsiguenga ethnic group, and individuals from Puno in the Andes (PU). The box within the figure shows the density plot and means (vertical lines) of Native American ancestry in the gastric cancer cases (yellow) and controls (brown). doi:10.1371/journal.pone.0041200.g001 Figure 1. Principal Component Analysis of our Peruvian samples of gastric cancer cases, their controls, and Native Americans in the context of HapMap-III European, African, and Mexican individuals, and distribution of Native American ancestry in cases and controls (box). Each individual was genotyped for 103 ancestry informative markers validated by Yaeger et al. (2008). We represent the first (horizontal) and second (vertical) principal components, which capture the 35.3% and 7.7% of total variance, respectively. Results HapMap individuals: YRI: Yoruba from Nigeria, LWK: Luhya from Kenya, ASW: African American from Southwest USA, MKK: Maasai from Kenya, CEU: Utah residents with European ancestry, TSI: Toscani from Italy, MEX: Mexican ancestry resident in Los Angeles. Peruvian Native Americans: Shimaa (SHI) and Ashaninkas (ASH) from the Matsiguenga ethnic group, and individuals from Puno in the Andes (PU). The box within the figure shows the density plot and means (vertical lines) of Native American ancestry in the gastric cancer cases (yellow) and controls (brown). doi:10.1371/journal.pone.0041200.g001 the Americas (see also [27]). Consistently, we observed that Native American ancestry is associated with variables that are indicators of poverty (many of which were also associated with gastric cancer) as well as their synthetic ‘‘socioeconomic’’ first factor of the multivariate analysis (Table 1). When we controlled for all covariates (i.e., the three factors of the multivariate analysis and age), the association between gastric cancer and Native American ancestry does not persist (OR = 1.28, 95%CI of the OR: 0.37– 4.47, p = 0.69). Likewise, when we separately controlled for the effect of socioeconomic conditions (the first factor of the multivariate analysis) or age, the association between gastric cancer and Native American ancestry also does not persist (OR = 2.58, 95%CI of the OR: 0.83–8.07 for factor 1 and OR = 1.01, 95%CI of the OR: 0.99–1.02 for age). August 2012 \| Volume 7 \| Issue 8 \| e41200 PLoS ONE \| www.plosone.org 3 Ancestry and Gastric Cancer in Peru Table 1. Socioeconomic, nutritional, and digestive-symptom-related variables and their association with gastric cancer and Native American ancestry. Discussion We performed a case-control study in the urban admixed population from Lima (Peru) and determined that Native American individual ancestry is associated with gastric cancer. However, this association seems primarily to be due to the association of socioeconomic variables both with gastric cancer and with Native American ancestry. Consistently, although the association of ancestry with gastric cancer is significant (p = 0.011), ancestry only explains 1.6% of the variance in disease status. When non-genetic covariates are included, their joint effect with ancestry explains 22.3% of the variance in disease status. Therefore, the high incidence of gastric cancer in Peru [1,28] does not seem to be due to the presence of common susceptibility genetic variants more frequent in Native American populations, but rather to a combination of socioeconomic factors present in this population. However, further studies with larger sample sizes are needed to explore this observation, since the power to detect ancestry genetic effects was limited in the current study. This result is consistent with the relative decrease in gastric cancer incidence in the United States during the last decades, due to the improvement of socioeconomic conditions [29]. Accuracy of ancestry estimations depends on several issues. The first is the number and the nature of markers used to estimate admixture. The 103 AIMs used in this study contain enough information to produce acceptable admixture estimates [26,30]. Galanter et al. have also showed that a panel of more than 88 AIMs contains enough information to estimate individual admix- ture with accuracy [31]. A second pervasive methodological issue in estimating admixture is the difficulty in using data from the most representative parental African, European, and Native American populations of the admixed group. In this case, we included as proxy for the parental populations European and African individuals from the HapMap project, and a set of Peruvian Native Americans from the Peruvian Andes and neighboring Eastern areas. While this choice may not be optimal, the 103 SNPs used, being AIMs, mitigated this issue because their frequencies are very different among the parental ethnic groups and highly homogeneous within them [26]. Thus, the use of markers with these characteristics renders our results robust to the choice of suboptimal parental populations. Third, different methods to estimate individual admixture may produce slightly different results even starting out from the same dataset. Results Variables P-value of association test with gastric cancer P-value of association test with Native-American Ancestry Personal Variables Gender 0.0416b 0.8621e Ethnicity (self-identification) 0.8692b ,0.0001 e Civil status 0.0017b 0.0305f Birth in lima 0.3371b 0.0007e Socioeconomic Variables Education level 0.0013b 0.0025f Property of household 0.1241b 0.7714f Material of household walls ,0.0001b 0.0029e Material of household floor 0.0009b 0.0590e Material of household ceiling ,0.0001b 0.0023e Type of water supply 0.0007b 0.0036e Type of sanitary service 0.0010b ,0.0001e Type of garbage collection service 0.0002b 0.0056e Fuel used for cooking 0.0149b 0.0023e Possession of a refrigerator ,0.0001b 0.0020e Possession of a freezer 0.2776b 0.5412e Type of energy in the household 0.0057b 0.4103e Type of water treatment 0.0069b 0.2028f Number of adults in the household 0.0911c 0.6643g Number of rooms in the household 0.0665c 0.0001g Number of bathroom in the household 0.0009c 0.0003g Number of children in the household 0.4894c ,0.0001g Number of meals per day 0.9407c 0.8532g Number of windows in the household 0.0001c ,0.0001g Frequency of eating in a restaurant 0.0690c 0.0170f Frequency of eating at the street 0.2345c 0.4067f Frequency of eating at home 0.4426c 0.9031f Household localization 0.0001b 0.0080f Nutritional variables (frequency of consumption of) Spicy food 0.8556c 0.1903f Steak 0.7363c 0.6443f Fish 0.0020c 0.5319f Poultry and birds 0.0415c 0.5995f Fresh vegetables 0.2226c 0.6504f Fresh Fruits 0.0587c 0.9235f Tea 0.2864c 0.7307f Coffee 0.8658c 0.2819f Apple infusion 0.5182c 0.0881f Coca leaf infusion 0.3320c 0.5237f Symptoms Pain ,0.0001c 0.7270f Burning ,0.0001c 0.4553f Regurgitation 0.0540c 0.1220f Nausea ,0.0001c 0.0805f Vomit ,0.0001c 0.3120f Heaviness ,0.0001c 0.2794f Factors from multivariate factor analysis Table 1. Socioeconomic, nutritional, and digestive-symptom-related variables and their association with gastric cancer and Native American ancestry. Table 1. Socioeconomic, nutritional, and digestive-symptom-related variables and their association with gastric cancer and Native American ancestry. August 2012 \| Volume 7 \| Issue 8 \| e41200 PLoS ONE \| www.plosone.org Ancestry and Gastric Cancer in Peru Table 1. Cont. Variables P-value of association test with gastric cancer P-value of association test with Native-American Ancestry Factor 1 0.00002 (ORa 0.68, 95%CI: 0.56–0.80)d 0.02017g Factor 2 0.00039 (OR 0.70, 95%CI: 0.58–0.85)d 0.00003g Factor 3 ,0.0001 (OR 1.8895%CI: 1.54–2.29)d 0.0638g Association tests reported in the table are: (a) OR: Odd ratio, (b): x2 test, (c): G-test, (d): logistic regression, (e): Mann-Withney, (f): Kruskal-Wallis, (g): Spearman rank order correlation. doi:10.1371/journal.pone.0041200.t001 Table 1. Cont. Results Peruvian individuals born in the countryside have more Native American ancestry on average than do residents of large urban centers (Table 1). In a case-control study, cases may frequently include individuals with more Native American ancestry because they are referred from small countryside health centers to large urban hospitals to receive better healthcare. This referral pattern has less effect on controls, and therefore may create a spurious association of Native American ancestry with disease. However, we recorded places of birth for all study participants, and thereby controlled this potential confounding factor: when we included the place of birth (Lima vs. countryside) as a covariate in the logistic regression, the association of Native American ancestry with gastric cancer persisted, although at a lower significance (p = 0.05 vs. p = 0.011). We conclude that the association between gastric cancer and ancestry is not an artifact of referrals of countryside individuals with high Native American ancestry to our study hospitals. This result emphasizes the importance of gathering birthplace and residence data for genetic association studies with diseases in Latin America, and other regions where most European colonization and admixture occurred in cities, and more autochthonous individuals predominate in rural areas, such as in Melanesia and South Africa. Ancestry and Gastric Cancer in Peru These urban populations are frequently peopled by immigrants from rural areas. Our results suggest that large cities of Western South America host millions of individuals of predominantly Amerindian genetic background. Contemporary international South-to-North migrations from South American cities from the Andean region are also spreading the genetic background of Native Americans worldwide, and it is expected that the almost one million United States immigrants coming from Andean countries [36] have high levels of Native American background. It would not be surprising if these populations, classified as ‘‘Hispano/Latino’’ in the United States, had more Amerindian ancestry than US individuals classified as Native American. Ancestry informative markers (AIMs) and genotyping To estimate ethnicity for each of the study subjects, we genotyped 106 SNPs that are informative for African, European, and Native American ancestry [26]. The genotyping was performed at the Biomedical Genomic Center of the Children’s Hospital Oakland Research Institute (University of Minnesota, MN, USA), using the Sequenom iPLEX platform (San Diego, CA, USA). Briefly, it is based on an allele-specific primer extension followed by separation of alternative alleles by mass spectrometry. The genotyping involved four multiplexed assays, three containing 26 SNPs and one containing 28 SNPs. Before genotyping, DNA samples underwent a Quality Control (QC) procedure that consisted of: (1) a non-allelic quantitative-PCR analysis that measures the quantity of PCR-amplifiable DNA and (2) an end- point reading from a Taqman SNP genotyping assay (Applied Biosystems, Palo Alto, CA, USA) that, in addition to providing a second assessment of the ability of PCR to amplify each sample, is a sensitive indicator of sample-to-sample cross-contamination. After we removed the SNPs rs30125 and rs888861, which showed a call rate ,95%, the average call rate for the SNPs was 99.7%. Of the 106 markers, 104 robustly generated call rates for at least 95% of samples, but for the SNP rs2592888 there are no genotypes publicly available for the Hapmap populations and the SNP was excluded from further analyses. Thus, we used genotypes for 103 SNPs to estimate admixture (see Table S3 for the complete list, with their allele frequencies in the study populations). In conclusion, we showed that in the urban admixed population from Lima, Native American individual ancestry is associated with gastric cancer, but this is explained by the association of socioeconomic variables with both gastric cancer and Native American ancestry. Ancestry and Gastric Cancer in Peru Ancestry and Gastric Cancer in Peru proved negative for gastric cancer. Individuals with premalignant lesions such as dysplasia (n = 3) or with presence of stomach tumors other than gastric cancer (n = 16) were excluded from the study. Individuals with intestinal metaplasia (n = 46) were included as controls (see Discussion). The final number of subjects considered to test in the association study was 241 gastric cancer cases and 300 controls. individuals attending the same gastroenterology services as the cases, and since they have undergone an endoscopy, we accurately know the status of their gastric mucosa. Also, in Latin America there is a considerable level of population stratification due to socioeconomic level and ancestry, that are correlated in large urban centers in Latin America (in addition to this study, see Avena et al. 2012 [30] for an example in Buenos Aires and Campbell et al. 2012 [27]). By selecting controls among attendants of the same hospital than cases, we mitigate this other potential source of population stratification between cases and controls. We also collected samples from 296 Native American individuals who were used as parental populations to estimate ancestry of the gastric cancer cases and controls. These include 23 farmers from Pichacani (Puno), belonging to the predominant Andean Quechua ethnic group, as well as 87 Shimaa and 186 Ashaninka from the Matsiguenga ethnic group, settled between the Andes and the Amazonian region. For all gastric cancer cases and controls and for the Native Americans, we extracted genomic DNA using the phenol-chlorophorm method described by Sambrook et al. with modifications, or the Gentra Puregene blood kit (Qiagen, USA) [38]. This investigation was approved by IRBs of Asociacio´n Bene´fica PRISMA, Universidad Peruana Cayetano Heredia, Johns Hopkins University, Universidade Federal de Minas Gerais, Hospital Arzobispo Loayza, Hospital Dos de Mayo and the Instituto Nacional de Enfermedades Neopla´sicas. All participants in the study provided written informed consent. In this study we report a surprisingly high Native American ancestry (.74%) both in controls and cases attending public hospitals from the now cosmopolitan city of Lima, the national capital that was also the capital of the Spaniard Viceroyalty of Peru for five centuries and therefore, the center of Spaniard colonial power. Large cities in Ecuador, Peru, Bolivia, and Northern Argentina host populations whose usual cultural identification as mestizo likely ignores their large Amerindian genetic background. Ancestry and Gastric Cancer in Peru Despite the high incidence of gastric cancer in the Peruvian population with a very high Native American ancestry, our result shows that this epidemiological observation does not rely on a genetic basis, suggesting a predominant role for socioeconomic factors and disparities in access to health services. We report a surprisingly high Native American ancestry (.74%) in individuals attending hospitals from the now cosmopolitan city of Lima. Since Native Americans are a neglected group in genomic studies, we suggest that the population from Lima and other large cities in Western South America may be convenient targets of epidemiological studies focused on Native American populations. Pursuit of this avenue of research in subsequently larger studies will begin to close the gap [37] in genetic studies and their potential benefits between European individuals and those from other generally less well served populations. Discussion To test the robustness of our admixture results in respect to the admixture estimation methods, we reanalyzed the data using the alternative maximum-likelihood approaches proposed by Tang et al. [32] and implemented in the software Frappe, and the method by Alexander et al. implemented in the software Admixture v. 1.2 [33]. The three methods produced highly correlated results (Figure S2) and the same pattern of association with gastric cancer (data not shown). In this study, we included as controls individuals with intestinal metaplasia (n = 46), assuming that this does not increase the risk of developing gastric cancer [23,24,25]. However, this assumption is not universally accepted [34]. When we alternatively assume three ordinal categories of disease risk (i.e., individuals without intestinal metaplasia, with intestinal metaplasia, and with gastric cancer), this progression, assessed by an ordinal logistic regression is also associated with Native American ancestry (OR = 2.83, 95%CI of the OR: 1.10–7.29, p = 0.031), but again, this association does not persist when controlled for all covariates (factor 1, 2, and 3 and age) (OR = 1.08, 95%CI of the OR: 0.35–3.34, p = 0.897). Thus, our results do not depend on the inclusion of intestinal metaplasia individuals as controls. An issue in our experimental design is that controls were selected as symptomatic individuals attending a gastroenterology service, undergoing an endoscopy and most of them with a gastric lesion: 6 with histologically normal gastric mucosa, 248 with gastritis, and 46 with metaplasia. It could be argued that the optimal control would be composed only by individuals with normal gastric mucosa. However, only through an endoscopy is it possible to accurately ascertain the absence of gastric lesions, and performing an endoscopy for research purposes only, not motivated by gastric-related symptoms is no longer ethically acceptable. On the other hand, using as controls individuals from the general population who did not undergo endoscopy is not necessarily a better choice since this strategy would have included as controls individuals with undetected gastritis [35] and other similar lesions. Therefore, we believe that our controls are the better operational choice for this study, because they are PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 August 2012 \| Volume 7 \| Issue 8 \| e41200 5 PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org 2. Recavarren-Arce S, Leon-Barua R, Cok J, Berendson R, Gilman RH, et al. (1991) Helicobacter pylori and progressive gastric pathology that predisposes to gastric cancer. Scand J Gastroenterol Suppl 181: 51–57. Table S4 Classification of socioeconomic, nutritional, and digestive-symptom-related variables used in Table 1 and their values. Table S4 Classification of socioeconomic, nutritional, and digestive-symptom-related variables used in Table 1 and their values. g g [ ] To test the association between these non-genetic variables or its representation obtained by multivariate factor analyses with gastric cancer (a binary trait) and ancestry (a continuous trait), we used the following statistical tests: logistic regression for continuous vs. binary traits (or Ordinal Logistic Regression for continuous vs. an ordinal dependent variable), G-test for ordinal vs. binary traits, x2 test for categorical vs. binary traits, Spearman rank order correlation for continuous vs. continuous traits, and Mann-Withney (2 categories) or Kruskal-Wallis (.2 categories) tests for ordinal vs. continuous traits. These analyses were performed in R environment. Ordinal logistic regression was performed using the ‘rms’ R package [45]. Ancestry and Gastric Cancer in Peru Ancestry and Gastric Cancer in Peru Ancestry and Gastric Cancer in Peru 2.3.3 [40,41]. It fits a Bayesian probability model of population structure and admixture using a Markov Chain Monte Carlo (MCMC) procedure, estimating the contribution of K parental populations to the genomes of individuals from the admixed population. We assumed that the HapMap YRI and CEU and our Native American samples were representative of the parental populations and that the gastric cancer cases and controls from Lima were admixed individuals. For this data set, each Structure run had 50,000 burn-in steps followed by 250,000 MCMC steps, and was repeated three times to allow checking for the robustness of the results. This length of the run and the checking procedure exclude the undesirable lack of convergence of the Markov Chains, which happens when the procedure does not properly explore the space of model parameters. All runs were performed assuming three clusters (K = 3), lambda was set to 1.0, and a parameters were estimated for each of the three clusters, GENSBACK = 2, MIGRPRIOR = 0.05 and we did not use a priori information for the individuals from parental populations to assist the clustering (USEPOPINFO = 0). To test the association of ethnicity with gastric cancer we analyzed 443 individuals (245 cases and 198 controls) for whom we have a histopathological diagnosis, collected personal, socio- economic, dietary data, and estimated ancestry. We used the logistic regression (observing the R2 Nagelkerk value – SPSS software) to test the association of ancestry with gastric cancer, which allowed us to control the effect of potential confounding variables (i.e., covariates) that were associated with disease status or ancestry. We included as covariates the original set of non- genetics variables or its factor-analysis synthetic representation. Acknowledgments We thank Jose´ Claudio Rocha, Ricardo Alves Silva, Dulciene Queiroz, Ana Lu´cia Brunialti and Wagner Magalha˜es for discussions on different parts of the project, and Hanaisa Sant’Anna for their logistic and technical help. 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, et al. (2010) Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. Statistical analysis y To represent the genetic structure of our samples in the context of parental population diversity, we performed principal compo- nent analysis (PCA) of individual genotypes (Figure 1), as implemented in the software Adegenet and Ade4 for R environment [42,43]. We also used Ade4 to apply a clustering method that is based on the PCA two-dimensional representation of individuals and their population centroid, designing bi- dimensional ellipses of dispersion. In addition to gender, age and self-reported ethnicity, a wide set of socioeconomic, nutritional, and clinical information was collected including civil status, place of birth, education level, household conditions, eating habits, frequency of consumption of fruits, vegetables, meat and poultry, infusions, as well as gastric-related symptoms. This information was organized in binary, ordinal, or categorical variables, as detailed in Table S4). We excluded variables that had more than 10% missing data, and a final set of 43 variables were included in subsequent analyses. To reduce the dimensionality of this set of personal, socioeconomic, nutritional, and clinical non-genetic variables we performed a multivariate factor analyses using the Statistical Package for the Social Sciences (SPSS) software (SPSS 19 for Windows, SPSS Inc, Chicago, IL, USA). The factor analysis synthesizes the variance of the original set of variables in a specified minor number of transformed variables (in our case 3), called factors. Each factor captures correlated information on the original dataset but the factors are uncorrelated among them [44]. Supporting Information Figure S1 Barplot of individual ancestry estimated with the software Structure for Africans (red), Europeans (green), and Native Americans (blue), as well as gastric cancer cases and controls. (TIF) Figure S2 Scatterplot and Spearman correlation be- tween individual Native American ancestry estimates by Structure versus Frappe (a) and Admixture (b) methods. Frappe was run with 100,000 maximum iteration of EM, K = 3 and 10,000 optional convergence threshold. Variations of these parameters did not show differences in results. Admixture was run using the default parameters with K = 3. (TIF) Author Contributions Conceived and designed the experiments: ETS RHG. Performed the experiments: LP RZ LC CH JMC. Analyzed the data: LP RZ LC CH JC SS GSS FK. Contributed reagents/materials/analysis tools: PH LC CH JMC JC GV WAP SJC DB MRR. Wrote the paper: LP ETS. Table S1 Distribution of cases and controls across hospitals and association with Native American ancestry. (DOC) Table S1 Distribution of cases and controls across hospitals and association with Native American ancestry. (DOC) Table S2 Socioeconomic, nutritional, and digestive- symptom-related variables, their LOD scores with the three first factors of the multivariate factor analysis and significance of Spearman correlation between individual values of the variables and coordinates on each factor. (DOCX) Table S3 Allele frequencies in the populations included in this study for the 103 Ancestry Informative Markers used in the study. (DOC) Table S3 Allele frequencies in the populations included in this study for the 103 Ancestry Informative Markers used in the study. (DOC) Table S4 Classification of socioeconomic, nutritional, and digestive-symptom-related variables used in Table 1 and their values. Subjects We estimated individual ancestry using the following three parental groups composed of unrelated individuals: (1) West African Yoruba from Nigeria (YRI – 118 individuals from the HapMap II/III project); (2) Utah individuals with European ancestry available at the Centre d’Etude du Polymorphisme Humain- CEPH collection (CEU – 60 individuals from the HapMap II Project) and (3) 296 Peruvian Native Americans collected by our group. The genotypes for Africans and Europeans were obtained from the public HapMap database [39], while the Native Americans were genotyped for this study. For this study we initially recruited 576 individuals attending Gastroenterology Divisions of the following three hospitals in Lima, between 2006 and 2009: Arzobispo Loayza, Dos de Mayo (both government-ruled), and the cancer-specialized Instituto Nacional de Enfermedades Neopla´sicas. Each participant an- swered a questionnaire to record age, gender, place of birth and of residence, and other socioeconomic, nutritional, and clinical information (Table 1). Cases were adults referred for endoscopy and whose biopsies were confirmed positive for gastric cancer by histopathological analysis. The control group was also composed of adult individuals referred for endoscopy, but whose biopsies We estimated the individual ancestry and its 90% credibility interval using the method implemented in the program Structure August 2012 \| Volume 7 \| Issue 8 \| e41200 PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 6 Ancestry and Gastric Cancer in Peru Ancestry and Gastric Cancer in Peru 3. Correa P (1992) Human gastric carcinogenesis: a multistep and multifactorial process – First American Cancer Society Award Lecture on Cancer Epidemiology and Prevention. Cancer Res 52: 6735–6740. 22. Freedman ML, Haiman CA, Patterson N, McDonald GJ, Tandon A, et al. (2006) Admixture mapping identifies 8q24 as a prostate cancer risk locus in African-American men. Proc Natl Acad Sci U S A 103: 14068–14073. p gy 4. Asaka M, Dragosics BA (2004) Helicobacter pylori and gastric malignancies. Helicobacter 9 Suppl 1: 35–41. 23. Filipe MI, Munoz N, Matko I, Kato I, Pompe-Kirn V, et al. (1994) Intestinal metaplasia types and the risk of gastric cancer: a cohort study in Slovenia. Int J Cancer 57: 324–329. g Helicobacter 9 Suppl 1: 35–41. 5. Atherton JC (2006) The pathogenesis of Helicobacter pylori-induced gastro- duodenal diseases. Annu Rev Pathol 1: 63–96. 24. Correa P, Piazuelo MB, Wilson KT (2010) Pathology of gastric intestinal metaplasia: clinical implications. Am J Gastroenterol 105: 493–498. 6. Kersulyte D, Mukhopadhyay AK, Velapatino B, Su W, Pan Z, et al. (2000) Differences in genotypes of Helicobacter pylori from different human populations. J Bacteriol 182: 3210–3218. 25. Gonzalez CA, Pardo ML, Liso JM, Alonso P, Bonet C, et al. (2010) Gastric cancer occurrence in preneoplastic lesions: a long-term follow-up in a high-risk area in Spain. Int J Cancer 127: 2654–2660. p p 7. Kersulyte D, Kalia A, Gilman RH, Mendez M, Herrera P, et al. (2010) Helicobacter pylori from Peruvian amerindians: traces of human migrations in strains from remote Amazon, and genome sequence of an Amerind strain. PLoS One 5: e15076. 26. Yaeger R, Avila-Bront A, Abdul K, Nolan PC, Grann VR, et al. (2008) Comparing genetic ancestry and self-described race in african americans born in the United States and in Africa. Cancer Epidemiol Biomarkers Prev 17: 1329– 1338. 8. Suzuki M, Kiga K, Kersulyte D, Cok J, Hooper CC, et al. (2011) Attenuated CagA oncoprotein in Helicobacter pylori from Amerindians in Peruvian Amazon. J Biol Chem 286: 29964–29972. 27. Campbell DD, Parra MV, Duque C, Gallego N, Franco L, et al. (2012) Amerind ancestry, socioeconomic status and the genetics of type 2 diabetes in a Colombian population. PLoS One 7: e33570. 9. Mohebbi M, Wolfe R, Jolley D, Forbes AB, Mahmoodi M, et al. Ancestry and Gastric Cancer in Peru (2011) The spatial distribution of esophageal and gastric cancer in Caspian region of Iran: An ecological analysis of diet and socio-economic influences. Int J Health Geogr 10: 13. p p 28. Mendoza D, Herrera P, Gilman RH, Lanfranco J, Tapia M, et al. (2008) Variation in the prevalence of gastric cancer in Peru. Int J Cancer 123: 414–420. 29. Jemal A, Center MM, DeSantis C, Ward EM (2010) Global patterns of cancer incidence and mortality rates and trends. Cancer Epidemiol Biomarkers Prev 19: 1893–1907. 10. Campbell PT, Sloan M, Kreiger N (2008) Dietary patterns and risk of incident gastric adenocarcinoma. Am J Epidemiol 167: 295–304. 30. Avena S, Via M, Ziv E, Perez-Stable EJ, Gignoux CR, et al. (2012) Heterogeneity in genetic admixture across different regions of Argentina. PLoS One 7: e34695. 11. Yang WG, Chen CB, Wang ZX, Liu YP, Wen XY, et al. (2011) A case-control study on the relationship between salt intake and salty taste and risk of gastric cancer. World J Gastroenterol 17: 2049–2053. 31. Galanter JM, Fernandez-Lopez JC, Gignoux CR, Barnholtz-Sloan J, Fernan- dez-Rozadilla C, et al. (2012) Development of a panel of genome-wide ancestry informative markers to study admixture throughout the Americas. PLoS Genet 8: e1002554. 12. Navarro Silvera SA, Mayne ST, Risch HA, Gammon MD, Vaughan T, et al. (2011) Principal component analysis of dietary and lifestyle patterns in relation to risk of subtypes of esophageal and gastric cancer. Ann Epidemiol 21: 543–550. 13. Steevens J, Schouten LJ, Goldbohm RA, van den Brandt PA (2011) Vegetables and fruits consumption and risk of esophageal and gastric cancer subtypes in the Netherlands cohort study. Int J Cancer. 32. Tang H, Peng J, Wang P, Risch NJ (2005) Estimation of individual admixture: analytical and study design considerations. Genet Epidemiol 28: 289–301. 33. Alexander DH, Novembre J, Lange K (2009) Fast model-based estimation of ancestry in unrelated individuals. Genome Res 19: 1655–1664. 14. Perez-Perez GI, Garza-Gonzalez E, Portal C, Olivares AZ (2005) Role of cytokine polymorphisms in the risk of distal gastric cancer development. Cancer Epidemiol Biomarkers Prev 14: 1869–1873. 34. Genta RM, Rugge M (2001) Review article: pre-neoplastic states of the gastric mucosa – a practical approach for the perplexed clinician. Aliment Pharmacol Ther 15 Suppl 1: 43–50. 15. Abnet CC, Freedman ND, Hu N, Wang Z, Yu K, et al. Ancestry and Gastric Cancer in Peru (2010) A shared susceptibility locus in PLCE1 at 10q23 for gastric adenocarcinoma and esophageal squamous cell carcinoma. Nat Genet 42: 764–767. 35. Dooley CP, Cohen H, Fitzgibbons PL, Bauer M, Appleman MD, et al. (1989) Prevalence of Helicobacter pylori infection and histologic gastritis in asymp- tomatic persons. N Engl J Med 321: 1562–1566. 16. Savage SA, Abnet CC, Mark SD, Qiao YL, Dong ZW, et al. (2004) Variants of the IL8 and IL8RB genes and risk for gastric cardia adenocarcinoma and esophageal squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 13: 2251–2257. p g J 36. U.S.CensusBureau (2009) 2009 American Community Survey 1-Year Estimates Washington, DC:US Census Bureau. g 37. Bustamante CD, Burchard EG, De la Vega FM (2011) Genomics for the world. Nature 475: 163–165. 17. Taguchi A, Ohmiya N, Shirai K, Mabuchi N, Itoh A, et al. (2005) Interleukin-8 promoter polymorphism increases the risk of atrophic gastritis and gastric cancer in Japan. Cancer Epidemiol Biomarkers Prev 14: 2487–2493. 38. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: A Laboratory Manual. New York: Cold Spring Harbor. 39. International HapMap Consortium (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449: 851–861. 18. Ohyauchi M, Imatani A, Yonechi M, Asano N, Miura A, et al. (2005) The polymorphism interleukin 8–251 A/T influences the susceptibility of Helico- bacter pylori related gastric diseases in the Japanese population. Gut 54: 330– 335. 40. Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155: 945–959. 41. Hubisz MJ, Falush D, Stephens M, Pritchard JK (2009) Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources 9: 1322–1332. 19. Rocha GA, Guerra JB, Rocha AM, Saraiva IE, da Silva DA, et al. (2005) IL1RN polymorphic gene and cagA-positive status independently increase the risk of noncardia gastric carcinoma. Int J Cancer 115: 678–683. gy 42. Chessel D, Dufour AB, Thiouluose J (2004) The ade4 package – I: one-table methods. R News 4: 5–10. 20. Ke-Xiang Z, Yu-Min L, Xun L, Wen-Ce Z, Yong S, et al. (2011) Study on the association of COX-2 genetic polymorphisms with risk of gastric cancer in high incidence Hexi area of Gansu province in China. Mol Biol Rep 38: 649–655. 43. Jombart T, Ahmed I (2011) adegenet 1.3–1: new tools for the analysis of genome-wide SNP data. Bioinformatics 27: 3070–3071. 21. References PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 7 Ancestry and Gastric Cancer in Peru Dean B, Levi JM (2003) At the Risk of Being Heard: Identity, Indigenous Rights, and Postcolonial States. Ann Arbor: University of Michigan Press. g 44. Agresti A (2012) Categorical Data Analysis: John Wiley 45. Harrell Jr FE (2012) Regression Modeling Strategies. PLoS ONE \| www.plosone.org August 2012 \| Volume 7 \| Issue 8 \| e41200 8
https://openalex.org/W3208258215	https://www.nature.com/articles/s41525-021-00253-1.pdf	English	null	Underrepresented patient views and perceptions of personalized medication treatment through pharmacogenomics	npj genomic medicine	2,021	cc-by	9,359	INTRODUCTION relationship. One study found that patients felt pharmacoge- nomic testing offered insights to their physicians on dosing12. Other studies demonstrated that patients’ perceptions of receiving personalized care were higher after physicians used pharmacogenomics information to guide medication changes10 and that use of genomic information during prescribing increased patient–provider communication alongside patient recall of medication changes made during clinical visits13. Patients also reported decreases in concerns about medications and experiencing greater conﬁdence about taking medications following pharmacogenomics testing12. Pharmacogenomic testing is increasingly being incorporated into clinical care as a means to improve drug prescribing, reduce drug inefﬁcacy, and avoid adverse drug events (e.g., toxicity)1–5. Pharmacogenomics uses an individual’s unique genetic makeup to assess their potential response to medication treatment, and such tailoring of healthcare delivery is one aspect of modern personalized care. Patient sense of personalized care is a dimension included in existing doctor–patient relationship mea- surement tools6. These and other patient-reported measures have been used to assess quality of care7,8, and they are increasingly becoming the focus of national health insurance programs like Medicare and Medicaid, which serve high-risk and vulnerable populations9. Pharmacogenomics has been shown to have a signiﬁcant positive effect on patient perceptions of personalized care and other dimensions of the doctor–patient relationship10. Evaluating patients’ experiences with pharmacogenomics as a part of their medical care is critical to determining the clinical utility of genomic medicine. Despite these seemingly positive implications, public will- ingness to participate in genetic testing to inform medical care and views about pharmacogenomics use in clinical settings varies across demographic groups1,14,15. African Americans, who are at greater risk for experiencing health disparities16,17 have been largely underrepresented in pharmacogenomic studies18. Though less likely to be included and participate in pharmacogenomic implementation studies, African American, or Black, patients have focused more on the positive beneﬁts of personalized medicine (e.g., fewer side effects and less trial and error prescribing) than White patients19. A study found no differences between Whites and Blacks in their perceived beneﬁts of personalized medicine, but, similar to other research, it showed variation between racial groups in concerns about the use of personalized medicine, with Blacks expressing greater concerns than Whites about privacy and discrimination based on the use of genetic information20–22. Communication is often overlooked when new approaches to personalized care g Patient views while receiving pharmacogenomically guided care demonstrate an understanding that pharmacogenomics can inform prescribing and help distinguish problematic prescriptions11. 1Department of Public Health Sciences, The University of Chicago, Chicago, IL, USA. 2Center for Personalized Therapeutics, The University of Chicago, Chicago, IL, USA. 3Center for Research Informatics, The University of Chicago, Chicago, IL, USA. 4Committee on Clinical Pharmacology and Pharmacogenomics, The University of Chicago, Chicago, IL, USA. 5Department of Medicine, The University of Chicago, Chicago, IL, USA. 6Section of Hospital Medicine, Department of Medicine, The University of Chicago, Chicago, IL, USA. ✉email: lsaulsberry@uchicago.edu ARTICLE OPEN Underrepresented patient views and perceptions of personalized medication treatment through Loren Saulsberry 1✉, Keith Danahey2,3, Brittany A. Borden2, Elizabeth Lipschultz2, Maimouna Traore2, Mark J. Ratain2,4,5, David O. Meltzer6 and Peter H. O’Donnell 2,4,5 Loren Saulsberry 1✉, Keith Danahey2,3, Brittany A. Borden2, Elizabeth Lipschultz2, Maimouna Traore2, Mark J. Ratain2,4,5, David O. Meltzer6 and Peter H. O’Donnell 2,4,5 Loren Saulsberry 1✉, Keith Danahey2,3, Brittany A. Borden2, Elizabeth Lipschultz2, Maimouna Traore2, Mark J. Ratain2,4,5, David O. Meltzer6 and Peter H. O’Donnell 2,4,5 Within an institutional pharmacogenomics implementation program, we surveyed 463 outpatients completing preemptive pharmacogenomic testing whose genetic results were available to providers for guiding medication treatment. We compared views and experiences from self-reported White and Black patients, including education level as a covariate across analyses. Black patients were less conﬁdent about whether their providers made personalized treatment decisions, and overwhelmingly wanted a greater role for their genetic information in clinical care. Both groups similarly reported that providers asked their opinions regarding medication changes, but White patients were more likely (59% vs. 49%, P = 0.005) to discuss the impact of personal/genetic makeup on medication response with providers, and Black patients reported initiating such discussions much less frequently (4% vs. 15%, P = 0.037). Opportunities exist for enhanced communication with underrepresented patients around personalized care. Tailored communication strategies and development of support tools employed in diverse healthcare settings may facilitate pharmacogenomically guided medication treatment that equitably beneﬁts minority patient populations. npj Genomic Medicine (2021) 6:90 ; https://doi.org/10.1038/s41525-021-00253-1 npj Genomic Medicine (2021) 6:90 ; https://doi.org/10.1038/s41525-021-00253-1 www.nature.com/npjgenmed npj Genomic Medicine (2021) 6:90 ; https://doi.org/10.1038/s41525-021-00253-1 INTRODUCTION Similar to other areas of care delivery, patients also expressed a desire for physicians to show personal attention by taking time to listen and discuss pharmacoge- nomic considerations with them11. Pharmacogenomic testing may change patient views about the role of genetics in their medical care, and these shifts in perceptions following genotyping might relate to aspects of the patient–provider Published in partnership with CEGMR, King Abdulaziz University L. Saulsberry et al. 2 Table 1. Patient characteristics by self-reported race and educational attainment (N = 463). Race White Black Educational attainment Total1 HS or less/ some college College graduate/ advanced degree Total2 HS or less/ some college College graduate/ advanced degree P-valuea Total survey respondents [N (%)] 332 (72) 86 (26) 246 (74) 131 (28) 74 (56) 56 (43) Gender [N (%)] Female 138 (42) 44 (51) 94 (38) 94 (72) 51 (69) 43 (77) <0.0001* Age [N (%)] Mean (range) 60 (19–90) 63 (19–90) 59 (20–87) 61 (19–95) 61 (19–89) 61 (26–89) 18–25 years 8 (2) 2 (2) 6 (2) 1 (1) 1 (1) – 26–39 years 26 (8) 3 (3) 23 (9) 9 (7) 6 (8) 3 (5) 40–50 years 33 (10) 5 (6) 28 (11) 21 (16) 10 (14) 11 (20) 0.306 51–64 years 122 (37) 33 (38) 89 (36) 43 (33) 22 (30) 21 (38) 65+ years 143 (43) 43 (50) 100 (41) 57 (44) 35 (47) 21 (38) Total surveys returned evaluating clinical visits (after enrollment) [N (%)] 790 (75) 207 (26) 583 (74) 265 (25) 145 (55) 119 (45) Surveys returned per patient (after enrollment) [N (%)] Mean (range) 2 (1–16) 2 (1–16) 2 (1–14) 2 (1–7) 2 (1–7) 2 (1–6) 1 149 (45) 40 (47) 109 (44) 66 (50) 41 (55) 24 (43) 2 73 (22) 17 (20) 56 (23) 30 (23) 15 (20) 15 (27) 0.512 3–4 75 (23) 18 (21) 57 (23) 26 (20) 12 (16) 14 (25) 5+ 35 (11) 11 (13) 24 (10) 9 (7) 6 (8) 3 (5) Self-reported health [N (%)]b Excellent/very good 479 (61) 101 (50) 378 (65) 105 (40) 42 (30) 63 (53) Good 235 (30) 73 (36) 162 (28) 108 (41) 63 (44) 45 (38) <0.0001* Fair/poor 68 (9) 30 (15) 38 (7) 48 (18) 37 (26) 10 (8) N 782 204 578 261 142 118 Survey responses of patients with “Unknown” educational level are not shown. RESULTS are adopted. Monitoring the impact of emerging technologies on health communication within clinical care may be speciﬁ- cally relevant for improving pharmacogenomics care delivery to minority patients as evidence has shown that minorities, patients of poor health status, patients with less than a high- school education, and older patients rate their visits with physicians as less participatory17. npj Genomic Medicine (2021) 90 Study participants Patient characteristics are summarized in Table 1. Signiﬁcant differences between total groups of White and Black respondents were apparent for gender and education, as more Black patients reported their gender as female and having an educational attainment of high school or less/some college. Black patients were also more likely to report their health as fair/poor. An education gradient was observed whereby higher educational attainment (e.g., college graduate/advanced degree) was asso- ciated with a greater likelihood of reporting excellent/very good health. Additional analyses alternatively stratifying by educational attainment showed signiﬁcant differences remained in gender and self-reported health status when comparing total groups completing high school or less/some college and college graduate/advanced degree (Supplementary Table 1). p y p p y In this study, we aimed to explore views and perceptions of care received among genotyped White and Black patients participating in a large institutional pharmacogenomic imple- mentation program. We hypothesized that patients’ attitudes and perceptions about pharmacogenomics with respect to their medical care would signiﬁcantly differ based on self-reported race. We particularly focused on self-reported race to deﬁne the populations in our study as self-identiﬁed race is used to direct pharmacogenomic clinical guidelines as well as clinical decision-making more generally in healthcare settings. Our study provides the perspectives of African American patients enrolled in a pharmacogenomic implementation program, whose views to date have not been deeply considered within pharmacogenomics implementation studies. Our primary objective was to identify potential gaps in patient perspectives that might guide how to best tailor pharmacogenomics implementation into clinical care for African Americans at greater risk for health disparities. Published in partnership with CEGMR, King Abdulaziz University INTRODUCTION Percent values may not sum to 100% due to rounding effects. aPearson chi- squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. bN values reﬂect responses from all surveys returned not unique patients. P ≤0.05, P ≤0.01, *P ≤0.001. Table 1. Patient characteristics by self-reported race and educational attainment (N = 463). Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. aPearson chi- squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. bN values reﬂect responses from all surveys returned not unique patients. P ≤0.05, P ≤0.01, P ≤0.001. Patient views and experiences with their providers p p White and Black patients reported exceedingly high levels of satisfaction with health care providers participating in the pharmacogenomics clinical study with almost all (99%) patients from both self-reported racial groups indicating being very satisﬁed/somewhat satisﬁed with their clinical visits (Table 2). More than 90% of both self-reported racial groups (White and Black) rated their providers as excellent/very good in three Published in partnership with CEGMR, King Abdulaziz University L. Saulsberry et al. 3 Table 2. Patient experience with providers participating in a pharmacogenomics implementation program by self-reported race and educational attainment. Race White Black Total1 HS or less/ some college College graduate/ advanced degree Total2 HS or less/ some college College graduate/ advanced degree Survey measure/question N (%) N (%) N (%) N (%) N (%) N (%) P-valuea Overall, how satisﬁed were you with your healthcare provider visit today. Would you say you were very satisﬁed, somewhat satisﬁed, somewhat dissatisﬁed, or very dissatisﬁed? Very satisﬁed/ somewhat satisﬁed 774 (99) 203 (99) 571 (99) 259 (99) 139 (99) 119 (100) 0.273 N 783 204 579 260 140 119 Please rate today’s provider visit. How was the provider at…? Patient views and experiences with their providers Being interested in you as a whole person…(Asking/knowing relevant details about your life and your situation; not treating you as “just a number”) Excellent/very good 765 (97) 199 (97) 566 (98) 257 (97) 140 (97) 116 (97) 0.739 N 785 206 579 264 144 119 Explaining things clearly…(Fully answering questions; explaining clearly; giving you adequate information; not being vague) Excellent/very good 766 (98) 202 (98) 564 (98) 260 (98) 142 (99) 117 (98) 0.407 N 784 206 578 264 144 119 Making a plan of action with you…(Discussing the options; involving you in decisions as much as you want to be involved; not ignoring your views) Excellent/very good 737 (94) 197 (96) 540 (93) 247 (95) 131 (93) 115 (97) 0.071 N 785 206 579 261 141 119 Please indicate your agreement or disagreement with each of the following statements: My healthcare provider cares greatly about me and my medical health Agree strongly/ agree somewhat 773 (99) 200 (99) 573 (99) 259 (99) 141 (99) 117 (99) Disagree strongly/ disagree somewhat 1 (<1) 0 (0) 1 (<1) 1 (<1) 1 (1) 0 (0) 0.695 Not sure 4 (1) 2 (1) 2 (<1) 1 (<1) 0 (0) 1 (1) N 778 202 576 261 142 118 My healthcare provider incorporates “personalized medicine” into my treatment decisions Agree strongly/ agree somewhat 720 (94) 181 (92) 539 (95) 231 (90) 125 (90) 105 (91) Disagree strongly/ disagree somewhat 5 (1) 2 (1) 3 (1) 0 (0) 0 (0) 0 (0) 0.010* Not sure 38 (5) 13 (7) 25 (4) 25 (10) 14 (10) 11 (9) N 763 196 567 256 139 116 I want my healthcare provider to make medical decisions based upon the most up-to-date medical and scientiﬁc evidence available Agree strongly/ agree somewhat 752 (97) 194 (97) 558 (97) 249 (96) 132 (95) 116 (97) Disagree strongly/ disagree somewhat 12 (2) 3 (2) 9 (2) 4 (2) 2 (1) 2 (2) 0.305 Not sure 8 (1) 2 (1) 6 (1) 6 (2) 5 (4) 1 (1) N 772 199 573 259 139 119 My healthcare provider follows medical guidelines or standards of practice when making medical decisions about me Agree strongly/ agree somewhat 741 (97) 193 (97) 548 (96) 249 (97) 137 (97) 111 (96) Disagree strongly/ disagree somewhat 1 (<1) 0 (0) 1 (<1) 0 (0) 0 (0) 0 (0) 0.813 Not sure 24 (3) 5 (3) 19 (3) 9 (3) 4 (3) 5 (4) N 766 198 568 258 141 116 N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Patient views and experiences with their providers This ﬁnding of the more highly educated patient population agreeing less frequently that their personal genetic information should play a greater role in their healthcare provider’s treatment decisions was similarly reﬂected in additional comparisons performed between total groups differing by educational attainment alone (high school or less/some college vs. college graduate/advanced degree) instead of race (Supplementary Fig. 1). p g Greater than 80% of all respondents from both racial groups and all educational levels thought healthcare providers had the ability to select medications that might work better for some patients because of knowledge about genetic/DNA factors. Despite this understanding of pharmacogenomics as a type of personalized medicine, White and Black patient perceptions varied on whether their provider incorporated “personalized medicine” into treatment decisions (Table 2). The majority of all respondents, whether White or Black, reported that they agreed strongly/agreed somewhat that providers incorporated “personalized medicine” into their treat- ment decisions, yet Blacks were twice as likely as Whites to report being unsure about receiving personalized care (10% and 5%, respectively, P = 0.01). No signiﬁcant differences between patients’ responses about the use of “personalized medicine” remained when comparisons were performed between aggregate groups differentiated by educational attainment alone (high school or less/ some college vs. college graduate/advanced degree) instead of race (Supplementary Table 2). Patient views and experiences with their providers Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. aPearson chi-squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, *P ≤0.001. Table 2. Patient experience with providers participating in a pharmacogenomics implementation program by self-reported race and educational attainment. ted in you as a whole person…(Asking/knowing relevant details about your life and your situation; not treating yo Being interested in you as a whole person…(Asking/knowing relevant details about your life and your situation; number”) N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. aPearson chi-squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, P ≤0.001. Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 npj Genomic Medicine (2021) 90 L. Saulsberry et al. 4 domains pertaining to patient–provider interactions: (1) being interested in the patient as a whole person, (2) explaining things clearly to the patient, and (3) making a plan of action with the patient (Table 2). Patient perceptions of the type of care approaches providers took during treatment were similar between White and Black respondents. Over ninety percent of survey respondents from both racial groups agreed strongly/agreed somewhat with statements describing provider care as based on up-to-date evidence and adherent to clinical guidelines (Table 2). strongly/agreed somewhat that their personal genetic information should play a greater role in their healthcare provider’s treatment decisions (P < 0.001). Black respondents in the aggregate and across education levels indicated similarly high levels of agree- ment that their personal genetic information should play a greater role in their healthcare provider’s treatment decisions (89% for high school or less/some college and 91% for college graduate/ advanced degree). Such consistency in patient views was not observed across education levels for White respondents. White patients with higher educational attainment (college graduate/ advanced degree) were less likely than those with high school or less/some college to agree that their personal genetic information should play a greater role in their healthcare provider’s treatment decisions for them (73% and 85%, respectively). Patient-reported participation in decision-making for medication changes during clinical visits Before evaluating shared decision-making, we ﬁrst importantly found that providers accessed GPS results at similar rates for evaluable health visits of Black and White study participants (69% vs. 72%, P = 0.317). Survey respondents expressed varying degrees of participation in the decision-making process about medication changes made during clinical visits (Fig. 2). More than half of White and Black respondents reported that their provider y Of all respondents, over 70% of respondents expressed receptivity for a greater role of personal genetic information in their clinical care (Fig. 1). Black patients (90%) were signiﬁcantly more likely than White patients (76%) to report that they agreed Fig. 1 Patient views on the use of pharmacogenomics to guide health care delivery, by self-reported race and educational attainment. N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. Pearson chi-squared tests comparing total sample of self-reported White respondents and total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, P ≤0.001. Fig. 1 Patient views on the use of pharmacogenomics to guide health care delivery, by self-reported race and educational attainment. N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. Pearson chi-squared tests comparing total sample of self-reported White respondents and total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, P ≤0.001. Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 L. Saulsberry et al. Patient-reported participation in decision-making for medication changes during clinical visits 5 Regarding the medicaon change or new medicaon discussed, please choose one of the following: Regarding the medicaon change or new medicaon discussed, please choose one of the following: My healthcare provider recommended, but allowed me to make the ulmate decision My healthcare provider asked my opinion, and we made the decision together My healthcare provider asked my opinion, but made the ulmate decision h l h id d h d i i My healthcare provider made the decision 33% 26% 39% 34% 31% 18% 15% 12% 14% 11% 16% 14% 39% 44% 32% 47% 42% 41% 12% 18% 15% 8% 11% 27% 0 10 20 30 40 50 60 70 80 90 100 e m o S / s s el r o S H ) k c al B ( la t o T ) e tih W ( la t o T college (White) HS or less/Some college (Black) College graduate/Advanced degree (White) College graduate/Advanced degree (Black) Percent(%) (N=82) (N=218) (N=59) (N=38) (N=159) (N=44) Educaonal aainment Fig. 2 Patient-reported roles in decision-making for recalled medication changes made during a clinical visit, by self-reported race and educational attainment. N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. Pearson chi-squared tests comparing total sample of self-reported White respondents and total sample of self- reported Black respondents. Fig. 2 Patient-reported roles in decision-making for recalled medication changes made during a clinical visit, by self-reported race and educational attainment. N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. Pearson chi-squared tests comparing total sample of self-reported White respondents and total sample of self- reported Black respondents. their provider initiated that conversation about individual factors in regard to medication changes, Black patients reported initiating these discussions much less frequently than White patients (4% vs. 15%, P = 0.037). Notably, not a single Black respondent with lower education attainment (high school or less/some college) reported initiating a discussion about personalized medicine with their provider. Patient-reported participation in decision-making for medication changes during clinical visits In subsequent analyses where survey responses were primarily stratiﬁed by educational attainment (high school or less/ some college vs. college graduate/advanced degree) instead of race, racial differences persisted (Supplementary Table 3). That is, even among patients with lower educational attainment (high school or less/some college), White patients reported the highest rates of recalling provider discussions about personalized (genetic) factors, while 23% of Black patients compared to only 4% of White patients with the same lower educational attainment indicated being unsure about whether a personalized medicine discussion occurred with their provider. asked their opinion about a medication change or new medica- tion. White patients with an education level of high school or less/ some college were more likely to report that their provider made the decision about their medication change without patient consultation or input (39%). About 1 in 4 (27%) Black patients with higher educational attainment (college graduate/advanced degree) indicated that their provider made a recommendation but that they were allowed to make the ultimate decision about the medication change. This was almost double that of White patients from all educational backgrounds and nearly three times higher than the reported 8% of Black patients categorized as having lower educational attainment. The results did not differ when comparisons were performed between aggregate groups differentiated by educational attainment alone (high school or less/some college vs. college graduate/advanced degree) instead of race (Supplementary Fig. 2). Published in partnership with CEGMR, King Abdulaziz University DISCUSSION aPearson chi-squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, P ≤0.001. Most genomic testing systems within US healthcare settings use provider-facing results portals (electronic health records; EHRs) to report results, thus (appropriately) requiring that communication about pharmacogenomic test results ﬂows from provider to patient. observed when comparing responses primarily by education; proportional differences in reported uncertainty between Black and White patients within the same educational attainment group further supported this ﬁnding. White patients were more likely to report discussing the impact of personal makeup (genetics) on medication response with their provider, while Black patients reported initiating these discussions far less frequently. In fact, none of the self-reported lower educational attainment Black respondents in this study reported initiating such a discussion about personalized medicine with their provider. p g p p Our implementation model was the same in this study (using the GPS, embedded within the EHR, to release results to providers). This ampliﬁes the provider’s role as a pivotal gatekeeper of health information, especially pharmacogenomic risk information. Provi- ders may be unprepared for the routine use of pharmacogenetic testing for clinical decision-making23 and even less equipped to adjust communication regarding pharmacogenomic information to underrepresented patient populations. As electronic patient portals evolve and patient access to that information technology expands, it is likely that patient demand will increase for using this communication medium to solicit consultation from providers about pharmacogenomics. Our results suggest that Black patients are currently not initiating these discussions with providers during health encounters. Further evaluation is needed to explain this lack of reported initiation and how those factors might bear upon pharmacogenomics implementation within underrepresented patient populations where these electronic tools are rapidly being incorporated into care delivery. p p Together, our results suggest an opportunity for enhanced patient–provider communication, especially for minority patients, around the role of genetic results during prescribing. This recapitulates ﬁndings from our own prior studies, which have demonstrated that the patient–provider relationship is critical to communication about pharmacogenomics10,11,13. The need for— and potential differential impact of—such discussions considering a patient’s race is a previously unappreciated aspect of this current work. DISCUSSION Of all respondents, about 1 in 3 could remember their healthcare provider changing a medication during a clinic visit (Table 3). Recollection of medication changes was comparable between White (31%) and Black (33%) respondents (P = 0.27). Of the respondents that recalled a medication change, White and Black respondents signiﬁcantly differed in reporting a discussion surrounding personalized medicine aspects of the medication decision with their healthcare provider. Black patients (49%) were signiﬁcantly less likely than White patients (59%) to recall this discussion taking place, and Black patients were nearly three times more likely than White patients to be unsure whether they had this discussion with their providers (17% and 6%, respectively, P = 0.005). Although the majority of patients from both racial groups indicated that, if a personalized medication discussion occurred In this study, we compared the views and experiences of self- reported White and Black patients receiving health care within a broad institutional pharmacogenomics implementation program. An overwhelming majority of Black patients wanted a greater role for their genetic information in their clinical care. While we found no disparities in satisfaction with provider care, there were gaps in the perceptions of White and Black patients in whether their providers were incorporating personalized medicine into treat- ment decisions. Compared to White patients, Blacks were less likely to express assurance that they were receiving personalized care from their providers despite having completed the same consent and enrollment process explaining the purpose of the pharmacogenomics implementation study. An association existed between self-reported race and patient uncertainty that was not npj Genomic Medicine (2021) 90 Published in partnership with CEGMR, King Abdulaziz University L. Saulsberry et al. 6 Table 3. Patient recollection of medication changes during clinical visits with providers participating in a pharmacogenomics implementation program by self-reported race and educational attainment. Race White Black Total1 HS or less/ some college College graduate/ advanced degree Total2 HS or less/ some college College graduate/ advanced degree Survey measure/question N (%) N (%) N (%) N (%) N (%) N (%) P-valuea Did your healthcare provider stop or change one of your medications today, or start a new medication? DISCUSSION Yes 237 (31) 61 (31) 176 (31) 87 (33) 41 (28) 46 (40) No 532 (69) 136 (69) 396 (69) 170 (65) 101 (70) 68 (59) 0.273 Unsure 5 (1) 1 (1) 4 (1) 4 (2) 2 (1) 2 (2) N 774 198 576 261 144 116 If yes to Did your healthcare provider stop or change one of your medications today, or start a new medication… Table 3. Patient recollection of medication changes during clinical visits with providers participating in a pharmacogenomics implementation program by self-reported race and educational attainment. id your healthcare provider discuss speciﬁc factors about you or your personal makeup which would suggest that yo han other patients to beneﬁt from the medication change or new medication? Did your healthcare provider discuss speciﬁc factors about you or your personal makeup which would suggest that you were more likely or less likely than other patients to beneﬁt from the medication change or new medication? Yes 128 (59) 43 (75) 85 (53) 43 (49) 18 (46) 25 (52) No 77 (35) 12 (21) 65 (41) 29 (33) 12 (31) 17 (35) 0.005* Unsure 12 (6) 2 (4) 10 (6) 15 (17) 9 (23) 6 (13) N 217 57 160 87 39 48 If yes, who initiated the discussion about individual factors regarding you and your response to the medication change or new medication? I was the one who asked about individual factors 20 (15) 7 (16) 13 (15) 2 (4) 0 (0) 2 (7) My healthcare provider was the one who brought up individual factors 88 (67) 34 (76) 54 (63) 42 (86) 18 (86) 24 (86) 0.037* Unsure 23 (18) 4 (9) 19 (22) 5 (10) 3 (14) 2 (7) N 131 45 86 49 21 28 N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. aPearson chi-squared tests comparing 1total sample of self-reported White respondents and 2total sample of self-reported Black respondents. P ≤0.05, P ≤0.01, *P ≤0.001. N values reﬂect the number of surveys returned from each self-reported racial group regarding clinical visits, not individual/unique patients. Survey responses of patients with “Unknown” educational level are not shown. Percent values may not sum to 100% due to rounding effects. Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 DISCUSSION The 1200 Patients Project incorporated patients receiving care from multiple types of outpatient providers across therapeutic areas, and UCMC’s location on the South Side of Chicago facilitated robust enrollment of Black patients to the clinical study, thus, to our knowledge, compiling one of the largest collections of ﬁrst-hand accounts from Black patients of their views and experiences with pharmacogenomics implementation in the literature. Importantly, rather than utilizing survey measures to assess patient health literacy, we incorporated direct measures of educational status into the current study. The inclusion of direct measures assessing understanding of genetic test results is an area we are now exploring in other studies, before and after direct-to-patient delivery of pharmacogenomic test results27. Separately, though the absence of measures evaluating trust in providers/healthcare system among minority patients may be a limitation of some survey instruments, our survey instrument speciﬁcally measured trust, and our previous analyses of this cohort have reported that trust, deﬁned as a patient’s belief that the provider had his or her best interest in mind when making clinical decisions, was generally high among our genotyped respondents10. Patients were asked to complete each survey immediately following their clinical visit, but in some instances survey responses were returned following a period of time after the visit, increasing the potential for recall bias. Finally, the total number of participating providers in this study was just under 20, so it is possible that some outcomes were driven by provider effects. However, our analyses did not account for physician clustering. Though providers included in the study were actively interested in pharmacogenomics, baseline provider knowledge about pharma- cogenomics was modest, replicating the characteristics of more than 10,000 general US physicians13. Providers’ accessing GPS results was used as a proxy for disclosure. Since providers could elect to view and/or use pharmacogenomic results in their prescribing decisions, survey responses likely reﬂected authentic “patient–provider pair” interactions. The study design permitting inclusion of multiple surveys per patient over time for a more longitudinal view of patient experiences helped normalize any potential effect of prior knowledge. p g p This study is unique in that successful recruitment of Black patients within our institutional PGx implementation program permitted the consideration of both race and education in our analysis, expanding upon the current literature often treating racial groups as monolithic29. DISCUSSION We recognize the differential role that education can serve as a more powerful determinant of health behaviors and outcomes for certain racial/ethnic groups than for others30. Education has also been associated with health literacy and understanding PGx test results31,32. Health literacy is deﬁned by the Institute of Medicine (IOM) as the degree to which individuals can obtain, process, and understand the basic health information and services needed to make appropriate health decisions33. In our analysis, there was a larger representation of White patients with higher educational attainment (college graduate/advanced degree) compared to Black patients with a similar education level, replicating the education gradient frequently reported in the literature. Following stratiﬁcation, we also observed a skewed distribution of gender across different education levels in both Black and White patient groups, illuminating the merit of further inquiry into the intersectionality of race and gender across educational backgrounds in inﬂuencing views and experiences with pharmacogenomics. The factors that make education inﬂuential in shaping health also intersect with race and gender, which inﬂuence social position. The intricacies of independent and/or interdependent contributions of these factors lack clarity and deserve to be actively pursued in future research. lack clarity and deserve to be actively pursued in future research. In terms of thinking about whether personal genetic informa- tion should play a greater role in their provider’s treatment decisions, education appeared to (inversely) impact responses most among White patients. Given that educational attainment, which was associated with self-reported White race, can reﬂect the availability of greater resources and training to process health information, our ﬁnding that the more highly educated patient population agreed less frequently that their personal genetic information should play a greater role in their healthcare could represent more conﬁdent perceptions among the better educated that their genetic information is already being incorporated to maximally beneﬁt them. In contrast, Black respondents across education levels indicated similarly high levels of agreement with this desire, suggesting that receptivity to personalized prescribing may be more broadly distributed across the Black patient population. Numerous mechanisms have been proposed through which education may shape views and perceptions of pharmaco- genomic care. Prior research incorporating both race and education treat educational attainment as a proxy of socio- economic status (SES) because it largely determines occupational status and income30,34. DISCUSSION Utilizing patient and physician “pairs” (where both parties agreed to undergo and receive, respectively, the genetic test results in this study) was an important feature of our pharmaco- genomics model, with the expectation that it would lead to joint patient–provider decision-making, rather than the physician interpreting the information alone. Our results support this expected outcome as more than half of White and Black patients reported being asked their opinion about medication changes during clinical visits, with 40% reporting being asked their opinion and making the decision together with their provider. y Given the signiﬁcance of patient–provider communication in inﬂuencing patient behaviors and outcomes, tailored communica- tion strategies may need to be developed and employed which address Black patients’ perceptions of receiving personalized treatments and help with the interpretation of pharmacogenomic information by providers for minorities and less educated patients. Published in partnership with CEGMR, King Abdulaziz University L. Saulsberry et al. 7 Prior studies evaluating patient engagement found that Blacks experience greater verbal passivity with physicians and lower levels of patient-centered communication compared to Whites and patients with higher educational attainment17,24,25. Our results similarly suggested that Black patients rarely initiated discussions regarding personalized medicine. Prior research also showed that racial minorities and patients with lower education lacked access to online patient portals or displayed limited use when they were available26. Few interventions exist that can simultaneously train patients to engage more fully in the healthcare process and give providers the skills needed to activate and promote patient engagement in the care dialogue17. These ideas illuminate the need for additional attention to developing racially and culturally sensitive tools to educate patients about genetic risks for suboptimal drug responses27,28. pharmacogenomics implementation program at The University of Chicago Medical Center (UCMC) from 2012 to 2017, which may limit generalizability. While some changes to practice patterns may have occurred over time, dramatic changes since 2017 are unlikely as uptake of pharmacogenomics into regular clinical practice remains in the early stages at most academic centers and is not yet available in most community practices. Furthermore, the diffusion of pharmacogenomics into the healthcare delivered to minority patients has been limited. The diversity of our study cohort may strengthen the applicability of ﬁndings to future implementation in outpatient clinical settings and is speciﬁcally relevant to communication and patient experiences as pharma- cogenomics is implemented into clinical care. Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 1200 Patients Project The 1200 Patients Project offered broad, preemptive pharmacogenomic testing to a diverse group of 1200 adults receiving outpatient subspecialty and primary care in order to assess the potential feasibility and utility of personalized medicine based on pharmacogenomics36–38. Patients and physicians were recruited into the study as pairs. Seventeen providers participated in this study, representing a diverse set of specialties across medicine (eight general internists, three oncologists, two cardiologists, one hepatologist, one nephrologist, one gastroenterologist, and one pulmo- nologist). Patients were then recruited to the 1200 Patients Project if they were receiving care from one of the participating physicians. Eligibility criteria were previously described38. The 1200 Patients Project recruited a racial/ethnic patient population that was approximately 60% White, 30% Black, and 10% Other39, reﬂecting the distribution of racial/ethnic populations in the greater Chicago city area40. While recruited patients were not given a formal education session on pharmacogenomics prior to survey completion, all patients received information describing the possible personalized care delivered in the study at the time of consent. These materials described personalized care as personal genetic informa- tion made available (with their permission) to their provider to enable speciﬁc medication treatment decisions for them, which aligns with our conceptualization of personalized care for the current study. Enrolled patients were genotyped across a broad panel of potentially actionable germline markers related to medication prescribing, with the pharmaco- genomic information speciﬁc to each patient then shared with the patient’s physician through an online clinical decision support tool, the Genomic Prescribing System (GPS)36–38. The GPS provides patient-speciﬁc pharmacogenomic information at the point-of-care, and the medications included in the GPS with actionable pharmacogenomic information have been previously reported36. Our prior ﬁndings from the 1200 Patients Project indicated that 34% of all medications on patients’ active drug lists had associated pharmacogenomic results indicating genomically concor- dant medications (low-risk of toxicity and/or high chance for favorable response), genomically cautionary (increased risk of toxicity and/or suboptimal response), or genomically unfavorable (high-risk for toxicity and/or suboptimal drug response). A large majority (61%) of provider decisions to stop medications at clinic visits were inﬂuenced by pharmacogenomic recommendations. Relatedly, for new prescriptions, the decision to prescribe the chosen medication was afﬁrmatively inﬂuenced by favorable pharmacogenomic information in half of all cases39. g g Collectively, our ﬁnal cohort of 463 White and Black patients completed 1055 surveys meeting this criteria (Table 1). DISCUSSION While pharmacogenomic testing was provided as a part of the clinical study free of charge to patients and providers, education might inﬂuence various aspects of access to healthcare, care delivery, and health outcomes that could be reﬂected in perceptions of pharmacogenomics. p p g As pharmacogenomics becomes more integrated into clinical practice and its use more widespread, its adoption is likely to be driven by both patients’ and providers’ understanding and interest in pharmacogenomic applications15,35. Future work should eval- uate how point-of-care resources like the GPS, traditionally used for clinical decision support, might be leveraged to better equip both patients and providers with information and strategies that improve communication on how genetic results might inﬂuence medication treatment and response. Vigilance around the experiences of medically underserved populations will facilitate timely updating of approaches to deliver pharmacogenomically guided care that helps close communication gaps and incorpo- rates patient views and preferences. This study had limitations. Our survey analysis includes responses from patients participating in one institutional Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 L. Saulsberry et al. L. Saulsberry et al. 8 1200 Patients Project Patient surveys were included in the analysis if at least half of the questions were completed (only 5 surveys were excluded from this analysis due to incomplete data). Questions that were left blank or answered inappropriately (e.g., multiple responses when only one was requested) were excluded from analysis of that item. While the survey domains remained the same throughout the study, we expected that patient perceptions of care could change over time because the prospective design of the 1200 patients study meant providers could variably choose to access pharmacogenomic results (or not) at each patient visit. Therefore, since each survey completed probed patient views and perceptions speciﬁc to that clinical visit, all surveys returned meeting inclusion criteria were evaluated, including those instances where the same patient returned multiple surveys over time because of having multiple (longitudinal) clinical visits. A prior evaluation showed an increase in providers’ referencing of pharmacogenomic results over time that inﬂuenced patient–provider communication13, so we predicted a similar impact could be reﬂected in patient perceptions of their numerous visits during the study. The median number of clinic visits was 2 across all patients, and those with more than 1 visit averaged a visit approximately every 8 months. Since provider visits triggered survey administration, these ﬁgures also indicate how far apart surveys were administered. Analysis plan Our primary analysis compared patients in “White” and “Black” self- reported racial groups in the aggregate to learn how attitudes and perceptions might differ between the two groups. The prespeciﬁed analysis plan permitted the inclusion of multiple surveys collected over time for a single individual. This analytical approach involved treating each patient survey as an independent, additional observation evaluating a separate, distinct clinical visit. Since both Black and White patient groups had almost identical survey frequencies and response rates (returning on average 2 surveys/patient over the course of the study [Table 1]), we did not inversely weight the surveys by the number per subject. Survey responses to speciﬁc questions (measures) were compared using chi- square tests with the level of statistical signiﬁcance set at P < 0.05 without adjustment for multiple comparisons. Study participants The study setting consisted of clinical visits to outpatient healthcare providers. Following clinical visits, patients were issued surveys for completion and return to the research team. Indicators for issuing a survey to a patient in the 1200 Patients Project following a clinical visit have been previously described10. This study analyzed survey responses from self-reported White and Black patients enrolled within the 1200 Patients Project who completed genetic testing, were seen in follow-up, and experienced at least one clinical visit with a participating provider with access to the GPS clinical decision support tool during the study period from October 2012 to May 2017. Our ﬁnal sample of survey responses included 463 participating patients: 332 patients who self-identiﬁed their race as White and 131 patients who self-identiﬁed their race as Black. A small number of patients (<12%) who identiﬁed as other races (i.e., not “White” or “Black”) were excluded from the present analysis. j p p While multiple theories have been developed to explain the crystal- lization of views/attitudes about health that relate to individual health behaviors, we undertook this analysis with education as a key covariate because it reﬂects the availability of resources (e.g., income and health insurance), the ability to process various types of information, and multiple socioeconomic indicators, including social and cultural factors41. Frame- works developed to foster health equity incorporate these factors in order to integrate context and promote the equitable diffusion of innovations42. This study included education as a covariate and performed secondary analyses of primary race-based comparisons by educational attainment to further our understanding of the respective roles of race and education. After performing the primary race-comparison analyses, we re-performed the analyses stratiﬁed by educational attainment within each self-reported racial group. For these analyses, educational level was categorized in a binary fashion: (1) “lower educational attainment” included patients reporting having completed high school or less, or some college; and (2) “higher educational attainment” included patients reporting being a college graduate or having an advanced degree. To further delineate the roles of race and education in our study, we also conducted sensitivity analyses of survey measures by also comparing results primarily by educational attainment, and, then, performing secondary analyses stratiﬁed by self-reported race within each educational attainment group (see online Supplement/Supplementary Information). Finally, as the GPS results were disclosed to patients at the discretion of the physician, we also npj Genomic Medicine (2021) 90 Published in partnership with CEGMR, King Abdulaziz University METHODS 1200 Patients Project empathy, medical decision-making, and personalized care, as previously described in detail10. Development and pre-testing of the instrument was performed with the University of Chicago Survey Lab and members of the University of Chicago Center for Health and the Social Sciences, and with input from additional external reviewers. Pre-testing of drafts of the instrument was then performed institutionally by members of the research team, with iteration until the ﬁnal version was achieved. The ﬁnal survey instrument has been previously published and data/ﬁndings from its use have been previously reported for other cohort analyses10,13. For the current study, we speciﬁcally focused on the survey items surrounding patient perceptions of pharmacogenomic care and views regarding the use of genetic information to guide medication treatment. REFERENCES 1. Fargher, E. et al. Current use of pharmacogenetic testing: a national survey of thiopurine methyltransferase testing prior to azathioprine prescription. J. Clin. Pharm. Ther. 32, 187–195 (2007). 28. Truong, T. M. et al. Patient insights on features of an effective pharmacogenomics patient portal. Pharmacogenet. Genomics 30, 191–200 (2020). 29. Witzig, R. The medicalization of race: scientiﬁc legitimization of a ﬂawed social construct. Ann. Intern. Med. 125, 675–679 (1996). 2. Faruki, H., Heine, U., Brown, T., Koester, R. & Lai-Goldman, M. HLA-B5701 clinical 2. Faruki, H., Heine, U., Brown, T., Koester, R. & Lai-Goldman, M. HLA-B*5701 clinical testing: early experience in the United States. Pharmacogenet. Genom. 17, 857–860 (2007). testing: early experience in the United States. Pharmacogenet. Genom. 17, 857–860 (2007). 30. Kimbro, R. T., Bzostek, S., Goldman, N. & Rodríguez, G. Race, ethnicity, and the education gradient in health. Health Aff. (Millwood) 27, 361–372 (2008). 3. Faruki, H. & Lai-Goldman, M. Application of a pharmacogenetic test adoption model to six oncology biomarkers. Personalized Med. 7, 441–450 (2010). 31. Olson, J. E. et al. Participant-perceived understanding and perspectives on pharmacogenomics: the Mayo Clinic RIGHT protocol (Right Drug, Right Dose, Right Time). Genet. Med. 19, 819–825 (2017). 4. Higgs, J., Gambhir, N., Ramsden, S., Poulton, K. & Newman, W. Pharmacogenetic testing in the United Kingdom genetics and immunogenetics laboratories. Genet. Test. Mol. Biomark. 14, 121–125 (2010). 32. Kaphingst, K. A. et al. Relationships between health literacy and genomics-related knowledge, self-efﬁcacy, perceived importance, and communication in a medi- cally underserved population. J. Health Commun. 21, 58–68 (2016) . 5. Hoop, J., Lapid, M., Paulson, R. & Roberts, L. Clinical and ethical considerations in pharmacogenetic testing: views of physicians in 3 “early adopting” departments of psychiatry. J. Clin. Psychiatry 71, 745–753 (2010). 33. Nielsen-Bohlman, L., Panzer, A. M. & Kindig, D. A. (eds) Health Literacy: A Pre- scription to End Confusion (National Academies Press, 2004). 6. Eveleigh, R. M. et al. An overview of 19 instruments assessing the doctor–patient relationship: different models or concepts are used. J. Clin. Epidemiol. 65, 10–15 (2012). 34. Karasek, R. & Theorell, T. Healthy Work: Stress, Productivity, and the Reconstruction of Working Life (Basic Books, 1992). 35. Stanek, E. J. et al. Adoption of pharmacogenomic testing by US physicians: results of a nationwide survey. Clin. Pharm. Ther. 91, 450–458 (2012). 7. Øvretveit, J. et al. Using patient-reported outcome measurement to improve patient care. Int. J. Qual. Reporting summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. 19. De Marco, M. et al. Views on personalized medicine: do the attitudes of African American and white prescription drug consumers differ? Public Health Genomics 13, 276–283 (2010). 20. Diaz, V. A. et al. Racial differences in attitudes toward personalized medicine. Public Health Genomics 17, 1–6 (2014). DATA AVAILABILITY All data generated or analyzed during this study are included in this published article and its supplementary information ﬁles. 21. Bevan, J. L. et al. Informed lay preferences for delivery of racially varied phar- macogenomics. Genet. Med. 5, 393–399 (2003). 22. Condit, C., Templeton, A., Bates, B. R., Bevan, J. L. & Harris, T. M. Attitudinal barriers to delivery of race-targeted pharmacogenomics among informed lay persons. Genet. Med. 5, 385–392 (2003). CODE AVAILABILITY 23. Weitzel, K. W. et al. A stepwise approach to implementing pharmacogenetic testing in the primary care setting. Pharmacogenomics 20, 1103–1112 (2019). Survey analyses were completed in Stata Statistical Software: Release 17 (College Station, TX: StataCorp LLC); the code and/or scripts underlying this article will be shared on reasonable request to the corresponding author. 24. Johnson, RachelL. et al. Patient race/ethnicity and quality of patient-physician communication during medical visits. Am. J. Public Health 94, 2084–2090 (2004). 25. Williams, MarkV. et al. The role of health literacy in patient-physician commu- nication. Fam. Med. 34, 383–389 (2002). Received: 22 April 2021; Accepted: 4 October 2021; 26. Anthony, DeniseL. et al. Who isn’t using patient portals and why? Evidence and implications from a national sample of US adults. Health Aff. (Millwood) 37, 1948–1954 (2018). 27. Truong, T. M. et al. Assessment of patient knowledge and perceptions of phar- macogenomics before and after using a mock results patient web portal. Clin. Transl. Sci. 13, 78–87 (2020). L. Saulsberry et al. 9 assessed whether providers differentially accessed GPS results at health visits for Black and White patients within our ﬁnal study sample. assessed whether providers differentially accessed GPS results at health visits for Black and White patients within our ﬁnal study sample. 16. Institute of Medicine (US) Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (eds. Smedley, B.D., Stith, A.Y. & Nelson, A.R.) (National Academies Press (US), 2003). Ethics declaration 17. Cooper, L. A. et al. A randomized controlled trial of interventions to enhance patient-physician partnership, patient adherence and high blood pressure con- trol among ethnic minorities and poor persons: study protocol NCT00123045. Implement. Sci. 4, 7 (2009). The 1200 Patients Project was an IRB-approved clinical study open at The University of Chicago (clinicaltrials.gov #NCT01280825), and all participants signed written informed consent. 18. Scherr, C. L., Ramesh, S., Marshall-Fricker, C. & Perera, M. A. A review of African Americans’ beliefs and attitudes about genomic studies: opportunities for mes- sage design. Front Genet. 10, 548 (2019). REFERENCES Health Care 29, 874–879 (2017). 36. Hussain, S. et al. Disease-drug database for pharmacogenomic-based prescribing. Clin. Pharm. Ther. 100, 179–190 (2016). 8. Costal Tirado, A. et al. Using patient-reported outcome measures for quality improvement in clinical genetics: an exploratory study. J. Genet. Couns. 26, 1017–1028 (2017). 37. O’Donnell, P. H. et al. The 1200 Patients Project: creating a new medical model system for clinical implementation of pharmacogenomics. Clin. Pharmacol. Therapeutics 92, 446–449 (2012). 9. Centers for Medicare and Medicaid Services (CMS). CMS Measures Management System Blueprint. Version 16.0. https://www.cms.gov/Medicare/Quality-Initiatives- Patient-Assessment-Instruments/MMS/Downloads/Blueprint.pdf (2020). 38. O’Donnell, P. H. et al. Adoption of a clinical pharmacogenomics implementation program during outpatient care–initial results of the University of Chicago “1,200 Patients Project”. Am. J. Med. Genet. C. Semin. Med. Genet. 166C, 68–75 (2014). 10. McKillip, R. P. et al. Patient perceptions of care as inﬂuenced by a large institu- tional pharmacogenomic implementation program. Clin. Pharm. Ther. 102, 106–114 (2017). 39. O’Donnell, P. H. et al. Pharmacogenomics-based point-of-care clinical decision support signiﬁcantly alters drug prescribing. Clin. Pharm. Ther. 102, 859–869 (2017). 11. Lee, Y. M. et al. Assessment of patient perceptions of genomic testing to inform pharmacogenomic implementation. Pharmacogenet. Genomics 27, 179–189 (2017). 40. United States Census Bureau. QuickFacts, Chicago City, Illinois. https://www. census.gov/quickfacts/chicagocityillinois. Accessed 27/09/2021. 41. Cutler, DavidM. & Lleras-Muney, Adriana Understanding differences in health behaviors by education. J. Health Econ. 29, 1–28 (2010). 12. Haga, S. B. et al. Patient experiences with pharmacogenetic testing in a primary care setting. Pharmacogenomics 17, 1629–1636 (2016). 42. Alcaraz, KassandraI. et al. The ConNECT framework: a model for advancing behavioral medicine science and practice to foster health equity. J. Behav. Med. 40, 23–38 (2017). 13. Borden, B. A. et al. Patient–provider communications about pharmacogenomic results increase patient recall of medication changes. Pharmacogenomics 19, 528–537 (2019). 14. Rothstein, M. A. & Hornung, C. A. Public Attitudes About Pharmacogenomics. In Pharmacogenomics: Social, Ethical, and Clinical Dimensions (ed. Rothstein, M.A.) (Wiley-Liss, Inc., 2003). Survey instruments Patients enrolled in the 1200 Patients Project were given anonymous surveys to complete after clinical visits with participating providers. At each clinical visit, providers could access patient pharmacogenomic results via the GPS, but the decision to view and/or use genetic information in their prescribing was at the providers’ discretion. Research staff independent of the providers gave surveys to patients after they saw their provider to complete before leaving clinic. If patients were unable to complete the surveys in the clinic, they were mailed the survey within a week of the patient’s visit. Over 95% of the surveys included in this analysis were completed immediately post visit, in the clinic. Less than 5% of all surveys received were by mail reporting. The patient survey instrument was designed for evaluating ﬁve dimensions of the doctor–patient relationship, including trust, privacy, Published in partnership with CEGMR, King Abdulaziz University Reprints and permission information is available at http://www.nature.com/ reprints Reprints and permission information is available at http://www.nature.com/ reprints 2T35D062719-26) (M.T.), The University of Chicago Comprehensive Cancer Center support grant (P.H.O.), and by NIH K08 HG011505 (L.S.). The authors also wish to thank Susanne B. Haga, Ph.D., Kenneth A. Rasinski, Ph.D., and Linda Patrick-Miller, Ph.D. for their assistance and feedback during development of the survey instrument. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. AUTHOR CONTRIBUTIONS L.S. designed the study, performed the research, and wrote the manuscript. B.A.B, E.L., K.D., and M.T. performed the research and contributed to revisions of the manuscript. M.J.R. and D.O.M. were involved with study design and revised the manuscript. P.H.O. designed the study, performed the research, and participated in writing and revising the manuscript. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. L. Saulsberry et al. 10 Reprints and permission information is available at http://www.nature.com/ reprints COMPETING INTERESTS The authors declare no competing interests. Published in partnership with CEGMR, King Abdulaziz University ACKNOWLEDGEMENTS 15. Haga, S. B., O’Daniel, J. M., Tindall, G. M., Lipkus, I. R. & Agans, R. Survey of US public attitudes toward pharmacogenetic testing. Pharmacogenomics J. 12, 197–204 (2012). Supported originally by NIH K23 GM100288-01A1 (P.H.O.), Bucksbaum Institute Associate Faculty Scholar Pilot Grant (P.H.O.), The William F. O’Connor Foundation (M. J.R.), The University of Chicago Pritzker Summer Research Program (NIH Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 npj Genomic Medicine (2021) 90 L. Saulsberry et al. L. Saulsberry et al. npj Genomic Medicine (2021) 90 Published in partnership with CEGMR, King Abdulaziz University Correspondence and requests for materials should be addressed to Loren Saulsberry ADDITIONAL INFORMATION Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41525-021-00253-1. Correspondence and requests for materials should be addressed to Loren Saulsberry © The Author(s) 2021 Correspondence and requests for materials should be addressed to Loren Saulsberry © The Author(s) 2021 Published in partnership with CEGMR, King Abdulaziz University npj Genomic Medicine (2021) 90 npj Genomic Medicine (2021) 90
https://openalex.org/W4234820943	https://journal.unindra.ac.id/index.php/hortatori/article/download/53/36	Indonesian	null	Influence of larning interest and vocabulary on the writing description of secondary school in Depok	Hortatori	2,019	cc-by-sa	6,346	Jurnal Pendidikan Bahasa dan Sastra Indonesia HORTATORI Jurnal Pendidikan Bahasa dan Sastra Indonesia HORTATORI Volume 1 Number 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) Research Article Research Article Influence of larning interest and vocabulary on the writing description of secondary school in Depok Dwi Septiani ), Universitas Pamulang ) Correspondences author: Jalan Surya Kencana No.1, Pamulang, Kota Tangerang Selatan, Banten 15417; Indonesia e-mail: dosen01401@unpam.ac.id. Abstract: The purpose of this research: (1) to know the influence of the interaction of learning interest and mastery of the standard vocabulary together to the ability to write the description of the students of SMPN in Depok. (2) to know the effect on the ability to write the description of SMPN students in Depok. (3) to know the influence of standard vocabulary mastery to the ability to write the description of SMPN students in Depok. The research method used is survey method with multiple correlation regression analysis. The sample of the research is the students of class VII: SMPN 1 Depok and SMPN 2 Depok as many as 80 students and take a simple random technique. Data collection was done by direct questionnaire to sample. Data analysis uses descriptive statistics, such as searching for mean (mean), median, mode, standard deviation, and inferential statistics, ie to find simple and double correlation coefficients followed by correlation coefficient significance test with t test. The findings of the researcher indicate that: (1) there is a significant influence of learning interest and mastery of the standard vocabulary simultaneously to the ability to write student description with the acquisition value Fo= 21,909 and Sig.= 0,000 < 0.05; (2) there is a significant influence of learning interest on the ability to write student description with thitung (3,564) > t table (1,99) and Sig. 0.001 < 0.05; (3) there is a significant influence of vocabulary mastery of the learning achievement on the subject of writing description. This is evidenced by the value of thitung (5.909) > t table (1.99) and Sig. 0,000 < 0.05. Key Words: Interest in Learning, Vocabulary, Writing Description. Key Words: Interest in Learning, Vocabulary, Writing Description. Article History: Received: 29/10/2017; Revised: 30/11/2017; Accepted: 05/12/2017; Published: 20/12/2017 How to Cite (MLA 7th): Septiani, Dwi. “Influence of larning interest and vocabulary on the writing description of secondary school in Depok.” Hortatori Jurnal Pendidikan Bahasa dan Sastra Indonesia 1.2 (2017): 169–178. Print/Online. Copyrights Holder: Septiani, Dwi. First Publication: Hortatori Jurnal Pendidikan Bahasa dan Sastra Indonesia (2017). This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. Pendahuluan Menurut Keraf (2007:93), karangan deskripsi merupakan sebuah bentuk tulisan yang bertahan dengan usaha para penulis untuk memberikan perincian-perincian dan objek yang sedang dibicarakan. Dari pengertian di atas, dapat disimpulkan bahwa tulisan deskripsi adalah pelukisan objek yang kasatmata. Deskripsi merupakan bentuk pengembangan paragraf yang bertujuan melukiskan sesuatu seolah-olah pembaca dapat mencintrai yang dilukiskan oleh penulisnya. Pembaca pun dituntut untuk dapat berimajinasi seolah-olah mereka melihat, mendengarkan, dan merasakan apa yang penulis lukiskan dalam bentuk tulisan. Dalam proses menguasai keterampilan menulis, siswa atau pelajar haruslah memiliki minat belajar serta kemampuan dalam menggunakan berbagai kosakata baku. Kedua hal tersebut berpengaruh secara langsung dengan kualitas karya tulis yang dihasilkan nantinya. Seorang siswa atau pelajar wajib memiliki minat belajar. Minat (interest) adalah kecenderungan seseorang untuk melakukan sesuatu perbuatan (Mulyasa, 2002: 38). Setiap individu tidak akan melakukan kegiatan apa pun jika ia tidak memiliki minat sama sekali. Akan tetapi, sebaliknya, seseorang akan sanggup menempuh berbagai rintangan untuk melakukan sesuatu demi mewujudkan minat dalam pikirannya jika memiliki minat yang besar. j p y j y g Soekamto (1991: 59—60) mengatakan bahwa minat berhubungan dengan segi-segi psikologis yang lain, seperti perhatian, ketertarikan, motivasi, dan sikap. Namun, dalam kehidupan remaja, bentuk minat dan cita-cita sangat beragam. Sebagaimana pendapat Mappiare (1992: 63) tentang bentuk minat di masa remaja, yaitu beberapa bentuk minat dan cita-cita yang penting dan menonjol dapat dikelompokkan dalam minat pribadi dan minat sosial, minat terhadap rekreasi, minat terhadap agama, serta minat terhadap cita-cita pendidikan dan jabatan. Cita-cita adalah perwujudan dari minat dalam hubungan dengan masa depan yang direncanakan seseorang, terkait dengan menentukan pilihan terhadap pendidikan, pekerjaan teman hidup, dan sebagainya. Minat dan cita-cita pada remaja banyak dipengaruhi oleh adanya faktor dari orang tua maupun dari minat kelompoknya. Dari uraian di atas, penulis menyimpulkan bahwa hakikat minat adalah kekuatan, kecondongan, kegairahan yang menyebabkan individu menaruh perhatian yang besar dan merelakan dirinya terikat terhadap sesuatu yang disukainya untuk mencapai tujuan sehingga menimbulkan rasa puas. Minat dapat muncul dari dalam diri individu karena kebutuhannya. Selain itu, minat dapat juga timbul dari pengaruh-pengaruh eksternal yang berupa respons seseorang yang diwujudkan dengan tindakan. Selain minat belajar, penguasaan kosakata baku pun ikut berperan dalam menentukan kualitas teks deskriptif yang dibuat. Kosakata (lexicon) adalah komponen bahasa yang memuat semua informasi tentang makna dan pemakaian kata. Dengan demikian, semua kata dasar, kata berimbuhan, kata berulang, dan kata majemuk dalam bahasa Indonesia dapat disebut sebagai kosakata bahasa Indonesia. Pendahuluan Semua warga negara Indonesia wajib membina dirinya masing-masing dalam pemakaian bahasa Indonesia agar bahasa itu tumbuh dan berkembang sesuai dengan kaidah yang berlaku. Oleh karena itu, pembina pendidikan, terutama guru bahasa Indonesia harus selalu berupaya menumbuhkan minat belajar siswa pada pelajaran Bahasa dan Sastra Indonesia di sekolah. Dalam Menulis sebagai Suatu Keterampilan Berbahasa, menurut Tarigan (2008: 2), kemampuan berbahasa umumnya memiliki empat komponen, yakni keterampilan menyimak (listening skills), keterampilan berbicara (speaking skills), keterampilan membaca (reading skills), dan keterampilan menulis (writing skills). Jika sudah dapat memahami keempat aspek tersebut, dapat dikatakan bahwa siswa berhasil dalam pembelajaran kebahasaan. Salah satu kemampuan yang masih sangat rendah dikuasai oleh siswa umumnya adalah kemampuan menulis. Akhadiah dkk. (1993: 143) mengatakan bahwa “kemampuan menulis bukanlah kemampuan yang diwariskan secara turun-temurun, tetapi merupakan hasil proses belajar-mengajar dan ketekunan berlatih.” Pendapat lain, Badudu dan Zain (1996: 1547) pengertian menulis jauh lebih sederhana, yakni menurut adalah menggunakan pena, potlot, bolpoin di atas kertas, kain, atau papan, dan sebagainya untuk 169 170 Influence of larning interest and vocabulary on the writing description of secondary school in Depok menghasilkan huruf, kata, atau kalimat. Secara umum, menulis dilaksanakan dengan mengikuti tujuh langkah: (1) pemilihan dan penetapan topik, (2) pengumpulan informasi, (3) penetapan tujuan, (4) perancangan tulisan, (5) penulisan, (6) penyuntingan atau revisi, (7) penulisan naskah jadi (Semi, 1990: 11). Kemampuan menulis dapat dicapai dengan bimbingan yang sistematis serta latihan yang intensif. Berdasarkan dari pengertian di atas, menulis atau mengarang merupakan pengungkapan pikiran, ide, dan perasaan melalui lambang grafis, atau simbol yang sistematis. Dengan menulis, seseorang tidak hanya menggambarkan satu pemikiran, tetapi juga dapat berkomunikasi serta menggunakan kosakata dan struktur bahasa dalam bentuk tulisan yang dapat dipahami oleh orang lain. Salah satu kegiatan menulis yang umumnya dipakai oleh pengajar atau guru untuk tahap awal adalah teks berjenis deskripsi. Kata deskripsi berasal dari kata latin describere yang berarti menulis tentang, atau mebeberkan sesuatu hal. Selain itu, kata deskripsi dapat diterjemahkan menjadi pemerian, yang barasal dari kata peri- memerikan yang berarti “melukiskan sesuatu hal”. Pengertian teks berkaitan erat dengan pengertian wacana. Gillian Brown dan George Yule dalam Okke Zaimar dan Ayo B. Harahap (2011: 14) menyatakan bahwa teks adalah realisasi wacana dan menganggap teks sebagai rekaman verbal suatu tindak komunikasi. Di sisi lain, menurut Suherli (2007: 10), karangan deskripsi menyajikan suatu peristiwa atau objek hasil pengindraan dengan cara melukiskan, menggambarkan, atau memerikan sehingga pembaca seperti menyaksikan, mengindra, atau mengalami sendiri secara langsung. HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi unindra ac id/index php/hortatori/ HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ \| \| ( ) ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ Metode Metode penelitian ini adalah cara yang digunakan dalam mengumpulkan data penelitian adalah dengan menggunakan metode survey (Sugiono, 2008: 7) dengan analisis korelasi regresi ganda. Desain ini digunakan untuk meneliti dua variabel bebas terhadap satu veriabel terikat. Data-data dalam penelitian ini diperoleh berdasarkan angket minat belajar dan penguasaan kosakata baku serta pemberian instrumen pada materi kemampuan menulis teks deskriptif siswa kelas VII SMP Negeri di Depok. Populasi target adalah siswa kelas VII SMPN di Depok sebanyak 785 siswa. Sampel ditetapkan sebesar 10% dari total populasi, yaitu 78,5 atau dibulatkan dengan total 80 siswa. Jadi, sebanyak 40 orang diambil dari kelas VII siswa SMP Negeri 1 Depok dan 40 orang diambil dari kelas VII siswa SMP Negeri 2 Depok. Variabel pertama adalah minat belajar siswa (X1). Yang dimaksud dalam minat belajar adalah kecenderungan hati yang terarah yang dapat timbul dari dalam dan dari luar diri setiap orang terhadap sesuatu objek, mempunyai kesiapan reaktif untuk memberikan perhatian, rela mengorbankan waktu, tenaga dan uang, serta mendorong untuk berbuat dengan rasa senang, karena ada kepentingan dari apa yang diperhatikannya. Nilai yang diperoleh oleh siswa untuk minat belajar diukur dengan tes minat diri sebanyak 30 butir pertanyaan yang mencerminkan ketaatan, pemanfaatan waktu, perhatian, dan keinginan. Skor minat belajar diperoleh dari jumlah skor 30 butir pernyataan yang direspons oleh 80 siswa. Instrumen yang digunakan untuk mengumpulkan data minat belajar berbentuk kuesioner dengan menggunakan rating scale. Model rating scale yang digunakan dalam bentuk kontinum dengan lima kategori. Skor Pertanyaan Sangat setuju Setuju Ragu-ragu Tidak setuju Sangat tidak setuju Positif 5 4 3 2 1 Negatif 1 2 3 4 5 Pengujian validitas dan perhitungan reliabilitas, sebelum digunakan untuk menjaring data penelitian atau digunakan pada subjek penelitian. Instrumen minat belajar siswa perlu dikalibrasi agar diketahui tingkat keandalan instrumen. Maka dari itu, dilakukan uji coba instrumen minat belajar siswa kelas VII SMPN di Depok. Tujuan uji coba instrumen minat belajar ini adalah untuk melihat validitas butir dan reliabilitas instrumen. Pengujian validitas butir instrumen menggunakan rumus Korelasi product moment sesuai dengan Arikunto (2010: 213). Pendahuluan Kosakata pun merupakan kata-kata yang dikuasai oleh seorang atau segolongan orang dalam lingkungan yang sama. Hal ini artinya kosakata yang dikuasai oleh seseorang dalam suatu lingkungan berbeda dengan lingkungan lainnya. Kata baku adalah suatu bentuk pemakaian bahasa yang menjadi model yang dapat dicontoh oleh setiap pemakaian bahasa yang hendak berbahasa secara benar (Moeliono, 1989: 23). Lebih lanjut, Moeliono Septiani, Dwi 171 Septiani, Dwi 171 (1989: 43) juga memaparkan bahwa kata baku standar adalah ragam bahasa yang berkekuatan sanksi sosial dan yang diterima masyarakat bahasa sebagai acuan atau model. Dari beberapa penjelasan di atas, dapat disimpulkan bahwa kosakata baku bahasa Indonesia adalah kosakata yang digunakan baik secara tertulis maupun diucapkan oleh sebagian besar masyarakat (lazim dan berterima) dalam situasi pemakaian bahasa yang bersifat resmi. Selain itu, kosakata baku bahasa Indonesia juga dapat menjadi acuan norma berbahasa Indonesia yang baik dan benar bagi penutur bahasa Indonesia. Dari paparan teori variabel penelitian di atas, peneliti melakukan pengkajian tentang pengaruh minat belajar dan penguasaan kosata baku terhadap kemampuan menulis deskripsi siswa SMPN di Depok. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579 7832 (Print) \| ISSN 2579 7840 (Electronic) \| Universitas Indraprasta PGRI Untuk pengujian reliabilitas instrumen minat belajar digunakan reliabilitas internal consistency dengan rumus Alfa Cronbach (Sugiyono, 2012: 365) Sebagai berikut: Metode Sebagai berikut: rhitung = 𝑁∑𝑋𝑌−(∑𝑋)(∑𝑌) √(𝑁∑𝑋2−(∑𝑋)2)(𝑁∑𝑌2−(∑𝑌)2) Keterangan: rxy = Koefesien korelasi N = Jumlah responden X = Skor butir yang dihitung validitasnya Y = Skor total rxy = Koefesien korelasi N = Jumlah responden X = Skor butir yang dihitung validitasnya Y = Skor total Untuk interpretasi dalam menentukan butir valid atau tidak, selanjutnya nilai rxy atau rhitung tersebut dikonsultasikan dengan nilai rtabel pada α = 0,05 dengan ketentuan rhitung > rtabel maka butir dinyatakan valid dan jika rhitung < rtabel maka butir dinyatakan tidak valid. Hasil pengujian diperoleh 30 butir pertanyaan dinyatakan valid dan 4 butir pertanyaan tidak valid. Peneliti membuang 4 butir pertanyaan yang tidak valid. Data hasil uji validitas instrumen minat belajar sebagai berikut: HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. Metode 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 172 Influence of larning interest and vocabulary on the writing description of secondary school in Depok No Soal rhitung rtabel Validitas No Soal rhitung rtabel Validitas 1 0,397 0,396 V 16 0,436 0,396 V 2 0,424 0,396 TV 17 0,492 0,396 V 3 0,366 0,396 TV 18 0,479 0,396 V 4 0,387 0,396 V 19 0,614 0,396 V 5 0,417 0,396 V 20 0,441 0,396 V 6 0,521 0,396 V 21 0,688 0,396 V 7 0,482 0,396 V 22 0,353 0,396 TV 8 0,521 0,396 V 23 0,335 0,396 TV 9 0,462 0,396 V 24 0,668 0,396 V 10 0,531 0,396 V 25 0,503 0,396 V 11 0,375 0,396 TV 26 0,563 0,396 V 12 0,434 0,396 TV 27 0,398 0,396 V 13 0,426 0,396 V 28 0,584 0,396 V 14 0,426 0,396 V 29 0,413 0,396 V 15 0,450 0,396 V 30 0,514 0,396 V No Soal rhitung rtabel Validitas 1 0,397 0,396 V 2 0,424 0,396 TV 3 0,366 0,396 TV 4 0,387 0,396 V 5 0,417 0,396 V 6 0,521 0,396 V 7 0,482 0,396 V 8 0,521 0,396 V 9 0,462 0,396 V 10 0,531 0,396 V 11 0,375 0,396 TV 12 0,434 0,396 TV 13 0,426 0,396 V 14 0,426 0,396 V 15 0,450 0,396 V No Soal rhitung rtabel Validitas 16 0,436 0,396 V 17 0,492 0,396 V 18 0,479 0,396 V 19 0,614 0,396 V 20 0,441 0,396 V 21 0,688 0,396 V 22 0,353 0,396 TV 23 0,335 0,396 TV 24 0,668 0,396 V 25 0,503 0,396 V 26 0,563 0,396 V 27 0,398 0,396 V 28 0,584 0,396 V 29 0,413 0,396 V 30 0,514 0,396 V Untuk pengujian reliabilitas instrumen minat belajar digunakan reliabilitas internal consistency dengan rumus Alfa Cronbach (Sugiyono, 2012: 365) Sebagai berikut: r11= 𝑘 𝑘−1 [1 − ∑𝑆12 𝑆𝑡2 ] Keterangan g r11 = reliabilitas instrumen secara keseluruhan g r11 = reliabilitas instrumen secara keseluruhan k = jumlah butir ∑S12 = Jumlah varian dari skor tiap butir St2 =varians skor total Untuk menentukan apakah koefesien reliabilitas yang diperoleh memenuhi syarat atau tidak, maka mengacu pada Sudijono (2011: 267) yaitu: g p j ( ) y a) Jika r11 ≥ 0,308 berarti instrumen minat belajar yang diujikan reliabilitasnya tinggi. b) Jika r11 ≤ 0,308 berarti instrumen minat belajar yang diujikan reliabilitasnya rendah. HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI Metode Berdasarkan Tes Reliabilitas dengan Alpha Cronbach Minat Belajar Bahasa Indonesia ( b) Jika r11 ≤ 0,308 berarti instrumen minat belajar yang diujikan reliabilitasnya rendah. b) Jika r11 ≤ 0,308 berarti instrumen minat belajar yang diujikan reliabilitasnya rendah. Berdasarkan Tes Reliabilitas dengan Alpha Cronbach Minat Belajar Bahasa Indonesia (X1) b) Jika r11 ≤ 0,308 berarti instrumen minat belajar yang diujikan reliabilitasnya rendah. Berdasarkan Tes Reliabilitas dengan Alpha Cronbach Minat Belajar Bahasa Indonesia (X1 Reliability Statistics Cronbach's Alpha N of Items .883 30 Note: r-tabel = 0, 396; db = 25; α = 0, 05 Nilai alpha cronbach > r-tabel maka instrumen dapat dikatakan reliable. Reliability Statistics Cronbach's Alpha N of Items .883 30 Note: r-tabel = 0, 396; db = 25; α = 0, 05 Nilai alpha cronbach > r-tabel maka instrumen dapat dikatakan reliable. Oleh karena itu, dapat disimpulkan bahwa instrumen minat belajar memiliki tingkat reliabilitas yang tinggi dan dapat digunakan untuk mengukur dapat penelitian. Selain menguji instrumen minat belajar (X1), peneliti melakukan uji validitas dan reabilitas pada instrumen penguasaan kosakata baku (X2). Hasil pengujian diperoleh 30 butir pertanyaan dinyatakan valid dan 6 butir pertanyaan tidak valid. Peneliti membuang 6 butir pertanyaan yang tidak valid. Data hasil uji validitas instrumen penguasaan kosakata baku sebagai berikut: HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI Metode 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ No Soal rhitung rtabel Validitas No Soal rhitung rtabel Validitas 1 0,478 0,396 V 16 0,471 0,396 V Septiani, Dwi 173 2 0,531 0,396 V 17 0,431 0,396 V 3 0,385 0,396 TV 18 0,447 0,396 V 4 0,471 0,396 V 19 0,438 0,396 V 5 0,502 0,396 V 20 0,398 0,396 V 6 0,375 0,396 TV 21 0,463 0,396 V 7 0,435 0,396 V 22 0,472 0,396 V 8 0,353 0,396 TV 23 0,544 0,396 V 9 0,569 0,396 V 24 0,498 0,396 V 10 0,420 0,396 V 25 0,497 0,396 V 11 0,499 0,396 V 26 0,397 0,396 V 12 0,368 0,396 TV 27 0,457 0,396 V 13 0,432 0,396 V 28 0,547 0,396 V 14 0,424 0,396 V 29 0,384 0,396 TV 15 0,358 0,396 TV 30 0,463 0,396 V Berdasarkan Tes Reliabilitas dengan Alpha Cronbach Penguasaan Kosakata Baku (X2) Berdasarkan Tes Reliabilitas dengan Alpha Cronbach Penguasaan Kosakata Baku (X2) Reliability Statistics Cronbach's Alpha N of Items .896 30 Note: r-tabel = 0,396; db = 25; α = 0,05 Reliability Statistics Cronbach's Alpha N of Items .896 30 Note: r-tabel = 0,396; db = 25; α = 0,05 Dari uraian di atas, jumlah item uji validitas untuk variabel Minat Belajar (X1) = 26 item dan variabel Penguasaan Kosakata (X2) = 24 item. Hasil uji validitas terhadap variabel penelitian terhadap 60 item pernyataan, dinyatakan 50 valid dan 10 drop (tidak valid). Hasil uji reliabilitas yang dilakukan memperlihatkan bahwa reliabilitas kuesioner variabel Minat Belajar (X1) adalah = 0,883 dan reliabilitas kuesioner variabel Penguasaan Kosakata (X2) sebesar = 0,896. Jadi, item yang valid dan reliabel serta layak untuk digunakan dalam menjaring data penelitian menjadi 50 item. Berikut adalah rincian Sebelum dan Sesudah Uji Coba Jenis Instrumen X1/ Reliabilitas X2/ Reliabilitas Jumlah Sebelum Uji Coba / Validasi 30 / 0,883 30 / 0,896 60 soal Item Tidak Valid 4 6 10 soal Sesudah Uji Coba 26 24 50 soal Dalam analisis deskriptif data disajikan dalam bentuk tabel distribusi frekuensi maupun histogram, selanjutnya dihitung nilai rata-rata, median, modus, dan simpangan baku. Adapun langkah-langkah pembuatan tabel distribusi frekuensi dilakukan dengan langkah-langkah sebagai berikut (Supardi, 2012: 35-36): ) a. Mengurutkan data dari terkecil sampai terbesar. ) a. Mengurutkan data dari terkecil sampai terbesar. b. Menghitung jarak atau rentangan (R), yaitu data tertinggi dikurangi data terendah. c. c. Menentukan simpangan baku. c. Menentukan simpangan baku. c. Menentukan simpangan baku. Sebelum melakukan analisis lebih lanjut, terlebih dahulu dilakukan uji persyaratan analisis, yaitu uji normalitas dilakukan untuk mengetahui data yang diperoleh berdistribusi normal atau tidak dengan menggunakan uji Liliofors (Lhitung) dengan bantuan perhitungan program SPSS versi 20. Pengujian dilakukan dengan menggunakan taraf signifikasi 0,05. Data dinyatakan normal jika signifikasi lebih besar dari 0,05. Metode Menghitung jumlah kelas (K) dengan rumus K = 1 + 3,3 log n c. Menghitung jumlah kelas (K) dengan rumus K = 1 + 3,3 log d. Menghitung panjang interval (P), dengan rumus: e. Menentukan ujung data pertama kemudian menghitug kelas interval dengan menjumlahkan ujung bawah kelas sampai pada data akhir. e. Menentukan ujung data pertama kemudian menghitug kelas interval dengan menjumlahkan ujung bawah kelas sampai pada data akhir. f. Membuat tabel sementara dengan cara dihitung satu demi satu yang sesuai dengan urutan interval kelas f. Membuat tabel sementara dengan cara dihitung satu demi satu yang sesuai dengan urutan interval kelas HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 174 Influence of larning interest and vocabulary on the writing description of secondary school in Depok HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169 178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ g. Membuat tabel distribusi frekuensi dengan cara memindahkan semua angka frekuensi. Sedangkan ukuran nilai pusat letak dapat ditentukan dengan langkah-langkah berikut: g. Membuat tabel distribusi frekuensi dengan cara memindahkan semua angka frekuensi. Sedangkan ukuran nilai pusat letak dapat ditentukan dengan langkah-langkah berikut: k ( ) a. Menentukan rerata (mean). b. Menentukan median (Me) c. Menentukan modus (Mo) Dan untuk ukuran simpangan dapat dditentukan dengan langkah-langkah berikut: Dan untuk ukuran simpangan dapat dditentukan dengan langkah-langkah berik a. Menentukan rentangan (R). a. Menentukan rentangan (R). b. Menentukan varian. HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ Hasil dan Diskusi Nilai minat belajar siswa yang diperoleh dari para responden mempunyai rata-rata (mean) = 107,85, median = 107,00, modus (mode) = 102, dengan simpangan baku (standar deviasi) sebesar = 8,104, skor minimum = 86, skor maksimum = 123, serta range skor = 37. Dari deskripisi tersebut, dapat dilihat bahwa antara nilai rata-rata dan median hampir sama, yakni 107,85 dan 107,00. Hal ini menunjukkan bahwa data skor minat pada penelitian ini cukup representatif. Selain itu, skor minat belajar yang lebih tinggi dibanding dengan median menunjukkan bahwa lebih banyak siswa yang memiliki minat belajar lebih tinggi dibandingkan dengan yang memiliki minat belajar rendah. Menurut statistik modus, mayoritas skor minat belajar adalah 102. Skor simpangan baku adalah 8,104 menunjukkan perbedaan minat belajar responden termasuk tinggi. Hal ini menunjukkan bahwa skor minat belajar bahasa Indonesia dari responden termasuk paling beragam di bawah variabel Kemampuan Menulis Deskripsi (Y) 9,331 dan di atas variabel Penguasaan Kosakata Baku (X2) 3,662. Untuk lebih jelasnya, data dilihat dari tabel di bawah ini. j Tabel 5 j y Tabel 5 Statistics Minat Belajar Penguasaan Kosakata Baku Kemampuan Menulis Deskripsi N Valid 80 80 80 Missing 0 0 0 Mean 107.85 17.14 76.46 Median 107.00 17.00 75.00 Mode 102 18 70 Std. Deviation 8.104 3.662 9.331 Skewness .058 -.582 -.184 Std. Error of Skewness .269 .269 .269 Kurtosis -.298 1.112 -.519 Std. Error of Kurtosis .532 .532 .532 Range 37 19 42 Minimum 86 5 50 Maximum 123 24 92 Dilihat dari distribusi, variabel Minat Belajar (X1) mempunyai distribusi menceng ke kanan, yang ditandai dengan angka statistik skewness positif = 0,058 dan jika dilihat keruncingan atau kurtosis menunjukkan bahwa variabel Minat Belajar (X1) cenderung lebih runcing yang ditunjukkan dengan nilai kurtosis yang negatif sebesar = - 0,298. Skor Penguasaan Kosakata Baku (X2) siswa yang diperoleh dari para responden mempunyai rata- rata (mean) = 17, 14, median = 17,00, modus (mode) = 18, dengan simpangan baku (standar deviasi) = 3,662, skor minimum = 5, skor maksimum = 24, serta range skor = 19. Dari deskripisi tersebut, dapat dilihat bahwa antara nilai rata-rata dan median hampir sama, yakni 17,14 dan 17,00. Septiani, Dwi 175 Septiani, Dwi 175 Hal ini menunjukkan bahwa data skor Penguasaan Kosakata Baku (X2) pada penelitian ini cukup representatif. Selain itu, skor minat belajar yang lebih tinggi dibanding dengan median menunjukkan bahwa lebih banyak siswa yang memiliki Penguasaan Kosakata Baku lebih tinggi dibandingkan dengan yang memiliki Penguasaan Kosakata Baku rendah. Hasil dan Diskusi Menurut statistik modus, mayoritas skor Penguasaan Kosakata Baku adalah 18. Skor simpangan baku adalah 3,662 menunjukkan perbedaan Penguasaan Kosakata Baku (X2) responden termasuk tinggi. Hal ini menunjukkan bahwa skor Penguasaan Kosakata Baku (X2) dari responden termasuk beragam di bawah variabel Kemampuan Menulis Teks Deskriptif (Y) = 9,331 dan variabel Minat Belajar (X1) = 8,104. j Dilihat dari distribusi menunjukkan bahwa variabel Penguasaan Kosakata Baku (X2) mempunyai distribusi menceng ke kiri, yang ditandai dengan angka statistik skewness negatif = - 0.582 dan jika dilihat keruncingan atau kurtosis menunjukkan bahwa variabel Penguasaan Kosakata Baku cenderung lebih runcing yang ditunjukkan dengan nilai kurtosis yang positif sebesar = 1,112. Skor Kemampuan Menulis Deskripsi (Y) siswa yang diperoleh dari para responden mempunyai rata-rata (mean) = 76,46, median 75,00, modus (mode) = 70, dengan simpangan baku (standar deviasi) sebesar = 9,331, skor minimum = 50, skor maksimum = 92, serta range skor = 42. Dari deskripisi tersebut, dapat dilihat bahwa antara nilai rata-rata dan median hampir sama, yakni 76,46 dan 75,00. Hal ini menunjukkan bahwa data skor Kemampuan Menulis Teks Deskriptif (Y) pada penelitian ini cukup representatif. Selain itu, skor rata-rata (mean) Kemampuan Menulis Deskripsi (Y) lebih tinggi dibanding dengan median menunjukkan bahwa lebih banyak siswa yang memiliki Kemampuan Menulis Deskripsi (Y) lebih tinggi dibandingkan dengan yang memiliki Kemampuan Menulis Deskripsi (Y) rendah. Menurut statistik modus, mayoritas skor Kemampuan Menulis Deskripsi (Y) adalah 70. Skor simpangan baku adalah 9,331 menunjukkan perbedaan Kemampuan Menulis Deskripsi (Y) responden termasuk paling tinggi. Hal ini menunjukkan bahwa skor Kemampuan Menulis Deskripsi (Y) dari responden termasuk beragam di atas variabel Minat Belajar (X1) 8, 104 dan variabel Penguasaan Kosakata Baku (X2) 3, 662. Dilihat dari distribusi menunjukkan bahwa variabel Kemampuan Menulis Deskripsi (Y) mempunyai distribusi menceng ke kiri, yang ditandai dengan angka statistik skewness negatif = - 0,184 dan jika dilihat keruncingan atau kurtosis menunjukkan bahwa variabel Kemampuan Menulis Deskripsi (Y) lebih runcing yang ditunjukkan dengan nilai kurtosis yang negatif sebesar = - 0,519. g y g j g y g g Pengujian persyaratan analisis data yang dilakukan dalam penelitian ini adalah pengujian normalitas dan linieritas garis regresi parsial antara variabel bebas dan variabel terikat. Ho = sampel berdistribusi normal H1 = sampel tidak berdistribusi normal Pengujian persyaratan analisis data yang dilakukan dalam penelitian ini adalah pengujian normalitas dan linieritas garis regresi parsial antara variabel bebas dan variabel terikat. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 176 Influence of larning interest and vocabulary on the writing description of secondary school in Depok 76 Influence of larning interest and vocabulary on the writing description of secondary school in Depok Pada kolom penguasaan kosakata baku (X2), terdapat nilai Kolmogorov-Smirnov = 0,937 dengan probabilitas 0,344 (Asymp. Sig. (2-tailed)). Persyaratan data berdistribusi normal jika probabilitas atau p > 0,05 pada uji normalitas Kolmogorov Smirnov. Jadi, karena nilai p = 0,344 atau p > 0,05 maka H0 diterima dan H1 ditolak. Artinya, data Penguasaan Kosakata Baku (X2) berasal dari populasi yang berdistribusi normal. Pada kolom kemampuan menulis deskripsi (Y), terdapat nilai Kolmogorov Smirnov = 1,072 dengan probabilitas 0,2 (Asymp. Sig. (2-tailed)). Persyaratan data berdistribusi normal jika probabilitas atau p > 0,05 pada uji normalitas Kolmogorov Smirnov. Jadi, karena nilai p = 0,2 atau p > 0,05 maka H0 diterima dan H1 ditolak. Artinya, data kemampuan menulis deskripsi (Y) berasal dari populasi yang berdistribusi normal. Dengan demikian, disimpulkan bahwa data dalam penelitian ini berasal dari populasi yang berdistribusi normal. Hal ini menunjukkan bahwa salah satu prasyaratan uji F dalam penelitian ini telah terpenuhi. Selanjutnya adalah peneliti melakukan pengujian multikolinieritas. Pengujian ini bertujuan untuk mengetahui apakah model regresi ditemukan adanya korelasi yang sempurna antarvariabel bebas (independen) atau tidak. Model regresi yang baik seharusnya tidak terjadi korelasi yang sempurna di antara variabel bebas. Cara mendeteksi adanya multikolinieritas adalah dengan melihat Tolerance (TOL) atau Varian Inflation Faktor (VIF). Apabila TOL < 0,1 atau VIF > 10 maka terjadi multikolinieritas. f p j Hasil uji multikolinieritas yang dilakukan dengan bantuan komputer program SPSS 20 adalah sebagai berikut. Hasil Uji Multikolonieritas TOL dan VIF Coefficientsa Model Collinearity Statistics 1 (Constant) Tolerance VIF Minat Belajar .990 1.010 Penguasaan Kosakata Baku .990 1.010 a. Dependent Variabel: Kemampuan Menulis Deskripsi a. Dependent Variabel: Kemampuan Menulis Deskripsi Tabel di atas menunjukkan bahwa kedua variabel independent tidak terdapat multikolinieritas karena TOL 0,990 > 0,1 dan nilai VIF 1,010 < 10. Jadi, kedua variabel bebas (Variabel Minat Belajar dan Variabel Penguasaan Kosakata Baku) tidak ada multikolinieritas sehingga analisis regresi dapat dipergunakan lebih lanjut. Hasil Perhitungan Koefisien Korelasi Pengaruh V i b l X d X t h d V i b l Y j Hasil Perhitungan Koefisien Korelasi Pengaruh Variabel X1 dan X2 terhadap Variabel Y Model Summaryb Model R R Square Adjusted R Square Std. Error of the Estimate 1 .602a .363 .346 7.545 a. Predictors: (Constant), Penguasaan Kosakata Baku, Minat Belajar b. HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579 7832 (Print) \| ISSN 2579 7840 (Electronic) \| Url: https://ejournal pbi unindra ac id/index php/hortatori/ Hasil dan Diskusi H sampel berdistribusi normal H1 = sampel tidak berdistribusi normal Dengan kriteria pengujian: Jika amax> Dtabel : Ho ditolak Jika amax< Dtabel : Ho diterima max tabel o Rangkuman dari data hasil uji normalitas menggunakan program olah data SPSS versi 20 for windows Rangkuman dari data hasil uji normalitas menggunakan program olah data SPSS versi 20 for windows Rangkuman dari data hasil uji normalitas menggunakan program olah data SPSS versi 20 for windows One-Sample Kolmogorov-Smirnov Test Minat Belajar Penguasaan Kosakata Baku Kemampuan Menulis Teks Deskriptif N 80 80 80 Normal Parametersa,b Mean 107.85 17.14 76.46 Std. Deviation 8.104 3.662 9.331 Most Extreme Differences Absolute .120 .105 .120 Positive .120 .094 .112 Negative -.062 -.105 -.120 Kolmogorov-Smirnov Z 1.075 .937 1.072 Asymp. Sig. (2-tailed) .198 .344 .200 a. Test distribution is Normal. b. Calculated from data. Dari tabel di atas, pada kolom minat belajar bahasa Indonesia (X1), terdapat nilai Kolmogorov- Smirnov= 1,075 dengan probabilitas 0,198 (Asymp. Sig. (2-tailed)). Persyaratan data berdistribusi normal jika probabilitas atau p > 0,05 pada uji normalitas Kolmogorov Smirnov. Jadi, karena nilai p = 0,198 atau p > 0,05 maka H0 diterima dan H1 ditolak. Artinya, data Minat Belajar (X1) berasal dari populasi yang berdistribusi normal. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 176 Influence of larning interest and vocabulary on the writing description of secondary school in Depok Error Beta 1 (Constant) 12.397 12.439 .997 .322 Minat Belajar .375 .105 .326 3.564 .001 Penguasaan Kosakata Baku 1.377 .233 .540 5.909 .000 Pada data statistik, ditunjukkan bahwa variabel minat belajar (X1) mempunyai nilai thitung = 3,564, sedangkan untuk variabel penguasaan kosakata baku (X2) thitung = 5,909. Nilai Sig. pada variabel minat belajar (X1) adalah 0,001, sedangkan nilai Sig. pada variabel penguasaan kosakata baku (X2) adalah 0,000. Nilai thitung X1 < thitung X2 dan nilai Sig. X1 > Sig. X2 (0,001 > 0,000). Hal ini menunjukkan bahwa pengaruh variabel penguasaan kosakata baku (X2) terhadap prestasi belajar bahasa Indonesia pada materi kemampuan menulis deskripsi (Y) lebih signifikan dibanding dengan pengaruh variabel minat belajar (X1). Kedua variabel bebas, minat belajar (X1) dan penguasaan kosakata baku (X2), secara bersama-sama berpengaruh positif terhadap prestasi belajar bahasa Indonesia pada materi menulis teks deskriptif. Akan tetapi, dari data di atas, dapat disimpulkan bahwa pengaruh penguasaan kosakata baku (X2) lebih besar atau lebih signifikan dibandingkan dengan pengaruh minat belajar (X1) terhadap prestasi belajar bahasa Indonesia pada materi menulis teks deskriptif. Pada data statistik, ditunjukkan bahwa variabel minat belajar (X1) mempunyai nilai thitung = 3,564, sedangkan untuk variabel penguasaan kosakata baku (X2) thitung = 5,909. Nilai Sig. pada variabel minat belajar (X1) adalah 0,001, sedangkan nilai Sig. pada variabel penguasaan kosakata baku (X2) adalah 0,000. Nilai thitung X1 < thitung X2 dan nilai Sig. X1 > Sig. X2 (0,001 > 0,000). Hal ini menunjukkan bahwa pengaruh variabel penguasaan kosakata baku (X2) terhadap prestasi belajar bahasa Indonesia pada materi kemampuan menulis deskripsi (Y) lebih signifikan dibanding dengan pengaruh variabel minat belajar (X1). Kedua variabel bebas, minat belajar (X1) dan penguasaan kosakata baku (X2), secara bersama-sama berpengaruh positif terhadap prestasi belajar bahasa Indonesia pada materi menulis teks deskriptif. Akan tetapi, dari data di atas, dapat disimpulkan bahwa pengaruh penguasaan kosakata baku (X2) lebih besar atau lebih signifikan dibandingkan dengan pengaruh minat belajar (X1) terhadap prestasi belajar bahasa Indonesia pada materi menulis teks deskriptif. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 176 Influence of larning interest and vocabulary on the writing description of secondary school in Depok Dependent Variabel: Kemampuan Menulis Deskripsi Dari tabel di atas, terlihat bahwa koefisien korelasi ganda pengaruh variabel bebas Minat Belajar (X1) dan Penguasaan Kosakata Baku (X2) secara bersama-sama terhadap Prestasi Belajar Bahasa Indonesia pada Materi Kemampuan Menulis Deskripsi (Y) adalah sebesar 0,602. Dari perhitungan tersebut, dapat diketahui bahwa koefisien korelasi tersebut signifikan variabel bebas Minat Belajar (X1) dan Penguasaan Kosakata Baku (X2) secara bersama-sama terhadap Prestasi Belajar Bahasa Indonesia pada Materi Kemampuan Menulis Deskripsi (Y). p p ( ) Koefisien determinasi adalah angka pada kolom R squre pada kolom tabel di atas dikalikan 100% yang menunjukkan besarnya kontribusi pengaruh variabel bebas X1 dan X2 terhadap variabel terikat Y. Berdasarkan hasil tersebut, dalam penelitian ini diperoleh koefisien determinasinya sebesar 36,3% yang menunjukkan bahwa besarnya kontribusi Minat Belajar (X1) dan Penguasaan Kosakata Baku (X2) secara bersama-sama terhadap Prestasi Belajar Bahasa Indonesia pada Materi Kemampuan Menulis Deskripsi (Y) adalah sebesar 36,3%, sedangkan sisanya (63,7%) karena pengaruh faktor lain. Septiani, Dwi 177 Septiani, Dwi 177 Selanjutnya, jika dikaji lebih lanjut berdasarkan koefisien korelasi parsial, ada korelasi antara variabel dependent dengan salah satu variabel independent setelah pengaruh hubungan linier variabel- variabel independen lainnya dibilangkan. Selanjutnya, jika dikaji lebih lanjut berdasarkan koefisien korelasi parsial, ada korelasi antara variabel dependent dengan salah satu variabel independent setelah pengaruh hubungan linier variabel- variabel independen lainnya dibilangkan. ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 2494.477 2 1247.238 21.909 .000b Residual 4383.411 77 56.927 Total 6877.888 79 a. Dependent Variabel: Kemampuan Menulis Teks Deskriptif b. Predictors: (Constant), Penguasaan Kosakata Baku, Minat Belajar Pengujian signifikansi garis regresi yang ada pada tabel ANOVA di atas menunjukkan bahwa nilai Sig. = 0,000, sedangkan nilai Fhitung = 21,909. Nilai Ftabel = 3,115. Karena nilai Sig. < 0,05 dan F hitung > F tabel maka H0 ditolak dan H1 diterima. Hal ini berarti bahwa koefisien regresi tersebut signifikan. Dari hasil pengujian korelasi maupun regresi tersebut, dapat disimpulkan bahwa terdapat pengaruh positif yang signifikan variabel bebas minat belajar (X1) dan penguasaan kosakata baku (X2) secara bersama-sama terhadap variabel terikat prestasi belajar bahasa Indonesia pada materi kemampuan menulis deskripsi siswa (Y). Rekapitulasi Hasil Perhitungan Garis Regresi Pengaruh Rekapitulasi Hasil Perhitungan Garis Regresi Pengaruh Variabel X1 dan X2 terhadap Variabel Y Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Simpulan Terdapat pengaruh yang signifikan variabel minat belajar (X1) dan variabel penguasaan kosakata baku (X2) secara bersama-sama terhadap prestasi belajar bahasa Indonesia pada materi kemampuan menulis teks deskriptif (Y). Hal ini dibuktikan dengan Fhitung = 21,909 dan nilai Sig. = 0,000 < 0,05. Nilai koefisien korelasi atau hubungan positif yang ditunjukkan dengan skor ry1 = 0,602, sedangkan kekuatan sumbangan ditunjukkan dengan koefisien determinasi (r2) = 0,363 yang menunjukkan bahwa variabel minat belajar (X1) dan variabel penguasaan kosakata baku (X2) secara bersama-sama memberikan konstribusi sebesar 36,3% terhadap variabel prestasi belajar bahasa Indonesia pada materi kemampuan menulis teks deskriptif (Y). Terdapat pengaruh yang signifikan variabel minat belajar (X1) terhadap prestasi belajar bahasa Indonesia pada materi kemampuan menulis teks deskriptif (Y). Hal ini dibuktikan dengan thitung = 3,564 dan nilai Sig. = 0,001 < 0,05. Terdapat pengaruh yang signifikan variabel penguasaan kosakata baku (X2) terhadap prestasi belajar bahasa Indonesia pada materi kemampuan menulis teks deskriptif (Y). Hal ini dibuktikan dengan thitung = 5,909 dan nilai Sig. = 0,000 < 0,05. HORATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Universitas Indraprasta PGRI 78 Influence of larning interest and vocabulary on the writing description of secondary school in Depok Dengan demikian, minat belajar dan penguasaan kosakata baku berperan sangat penting dalam menentukan dan meningkatkan prestasi belajar bahasa Indonesia khususnya pada materi kemampuan menulis teks deskriptif siswa. Berdasarkan temuan penelitian pada analisis data penelitian di atas, prestasi belajar bahasa Indonesia pada materi kemampuan menulis teks deskriptif siswa dapat menjadi lebih baik dengan cara meningkatkan minat belajar dan penguasaan kosakata baku. Ucapan Terima Kasih Peneliti mengucapkan terima kasih kepada Kepala SMPN 1 Depok dan SMAN 2 Depok terima kasih atas kemudahannya dalam mengumpulkan data dalam pengerjaan penelitian ini. Selain itu, terima kasih pula kepada semua pihak yang membantu penulis dalam menyelesaikan artikel hasil riset ini sehingga bisa diterbitkan di Jurnal Hortatori. HORTATORI \| Jurnal Pendidikan Bahasa dan Sastra Indonesia \| Vol. 1 No. 2 (2017), 169-178 ISSN 2579-7832 (Print) \| ISSN 2579-7840 (Electronic) \| Url: https://ejournal-pbi.unindra.ac.id/index.php/hortatori/ Daftar Rujukan Akhadiah, Sabarti, et al. Pembinaan Kemampuan Menulis Bahasa Indonesia. Jakarta: Erlangga. 1993. Print. gg Arikunto, Suharsimi, Prosedur Penelitian Suatu Pendekatan Praktek. Jakarta: Rineka Cipta Arikunto, Suharsimi, Prosedur Penelitian Suatu Pendekatan Praktek. Jakarta: Rineka Cipta. 2010. Print. Badudu, J.S. dan Muhammad Zain. Kamus Umum Bahasa Indonesia. Jakarta: Pustaka Sinar Harapan. 1996. Print. Badudu, J.S. dan Muhammad Zain. Kamus Umum Bahasa Indonesia. Jakarta: Pustaka Sinar Harapan. 1996. Print. p Keraf, Gorys. Argumentasi dan Narasi. Jakarta: PT Gramedia Pustaka Utama. 2007. Print. Mapiere, Andi. Psikologi Orang Dewasa. Surabaya: Usaha Nasional. 1992. Print. Moeliono, Anton M., et al. Tata Bahasa Baku Bahasa Indonesia. Jakarta: Balai Pustaka. 1989. Print. Mulyasa, E. Kurikulum Berbasis Kompetensi. Bandung: Remaja Rosda Karya. 2002. Print. y Soekamto, Toeti. Validitas Instrumen Pengukuran di Kawasan Afektif. Jakarta: LPP IKIP. 1991. Print. Semi, M. Atar.. Dasar-Dasar Keterampilan Menulis. Bandung: Angkasa, 2007. Print. Sugiono. Metode Penelitian Pendidikan. Bandung: Alfabeta. 2008. Print. Sugiono.Metode Penelitian Bisnis. Bandung: Alfabeta. 2012. Print. Suherli. Menulis Karangan Ilmiah. Depok: Arya Duta. 2007. Print. Sudijono, Anas. Pengantar Statistik 2Pendidikan. Jakarta: PT. Raja Grafindo Persada. 2010. Pr o, Anas. Pengantar Statistik 2Pendidikan. Jakarta: PT. Raja Grafindo Persada. 2010. Print. Sudijono, Anas. Pengantar Statistik 2Pendidikan. Jakarta: PT. Raja G j , g j Tarigan, Henry Guntur. Menulis sebagai Suatu Keterampilan Berbahasa. Bandung: Angkasa 2008 Print j g j Tarigan, Henry Guntur. Menulis sebagai Suatu Keterampilan Berbahasa. Bandung: Angkasa. 2008. Print. Tarigan, Henry Guntur. Menulis sebagai Suatu Keterampilan Berbahasa. Bandung: Angkasa. 2008. Print. Zaimar, Okke Kusuma Sumantri dan Ayu Basoeki Harahap. Telaah Wacana: Teori dan Penerapannya. Depok: Komodo Book. 20
W4283691435.txt	https://e-journal.unair.ac.id/JDE/article/download/30122/21794	en		Government Debt-Economic Growth Nexus in ASEAN-4 Countries	JDE (Journal of Developing Economies)	2,022	cc-by	10,621	JDE (Journal of Developing Economies) Vol. 7 No. 1 (2022): 45-69 JDE (Journal of Developing Economies) https://e-journal.unair.ac.id/JDE/index GOVERNMENT DEBT-ECONOMIC GROWTH NEXUS IN ASEAN-4 COUNTRIES Hoda Hajian1 Azali Mohamed2 Shivee Ranjanee Kaliappan3 University of Science and Culture, Islamic Republic of Iran Universiti Putra Malaysia, Malaysia 1 2, 3 ABSTRACT Given a background of controversial political and theoretical academic debate and diverse empirical result, as Checherita and Rother (2010) concluded government debt and economic growth relationship is a country specific issue. This paper aims to investigate the causal and dynamic effect of government debt on output growth in the context of developing economies with generally medium debt regime in ASEAN-4 countries. Namely, Indonesia, Malaysia, the Philippines and Thailand during 1985 to 2019 years. A robust multi-variable vector autoregressive (VAR) model at level is employed to capture the long run relations, and causality is addressed using Toda-Yamamoto (1995) approach. As a by-product of the analysis the effect of government debt on two essential factors of sustainable GDP growth, namely, private capital formation and human capital is examined. The findings of this paper which contrast with the general negative effect found in some empirical studies for developing countries, shows debt does not cause output growth in Indonesia, Malaysia and Thailand but the reverse is true. GDP response to debt shock is negative, positive and positive, respectively yet statistically insignificant. In other hand, in the Philippines the result shows the economy is debt-driven as debt positively cause GDP without improving private investment or human capital. Overall, the findings support well debt management. Given current debt regime, improvements on tax collection and government fund allocation in terms of priorities and efficiencies must be continued. ARTICLE INFO Received : September 22th, 2021 Revised : March 11th, 2022 Accepted : April 5th, 2022 Online : June 28th, 2022 Correspondence: Hoda Hajian E-mail: hoda.econ.upm@gmail.com Keywords: Government debt, GDP Growth, ASEAN countries, VAR model, time-series. JEL: O49; C32 To cite this document: Hajian, H., Mohamed, A., & Kaliappan, S.R. (2022). Government Debt-Economic Nexus in ASEAN-4 Countries. JDE (Journal of Developing Economies), 7(1), 45-69 Introduction The issue of government debt has long engaged great economists. Along with the academic and political debate the general increase in the worldwide debt level stimulated new series of empirical and theoretical research. However, they only add to the dimensions of the controversy. Overall, several conclusions has been highlighted. First, debt-growth relationship is an empirical question. Second, this relationship is quite heterogeneous across countries. Therefore, result obtained from large panel research cannot be generalized to individual countries. JDE (Journal of Developing Economies) p-ISSN: 2541-1012; e-ISSN: 2528-2018 DOI: 10.20473/jde.v7i1.30122 Open access under a Creative Commons Attribution 4.0 International (CC BY) Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries From theoretical perspective, recent endogenous growth models argue that fiscal policy and even the method of finance can influence economic growth and private investment. In many theories, private investment is considered as the main channel through which debt affects economic growth (e.g.Modigliani (1961), Diamond (1965)). In view of neoclassical economic school of thought, debt mainly negatively affects growth (Elmendorf and Mankiw, 1999) while in endogenous growth models some cases of crowding-in is considered, that could particularly relate to the case of developing countries. In that, the financially constrained developing governments channel the debt funds into development expenditure which is highly productive and necessary for future economic growth. Finally, the third strand of economic views, the so-called “Ricardians” hold the view that debt-financed expenditures are the same as tax-financed expenditures. Implying that debt does not have a real impact on output, etc. This paper aims to investigate government debt-growth issue in four largest emerging/ developing ASEAN countries namely, Indonesia, Malaysia, the Philippines and Thailand (ASEAN-4). ASEAN-4 countries have experienced high growth rate - to less extent for the Philippines- in the past recent decades and considered as economic miracles. They also have bright economic growth prospect. In the other hand, these countries has experienced continuous budget deficit since 1970, except the period between 1985 until before the 1997 crisis. Furthermore, the government of these countries are known to be successful in using of counter cyclical fiscal policy to stimulate economic growth. They are also known to be conservative regarding maintaining the budget deficit ratio within the self-imposed ceiling rate. Medium level budget deficit caused by expansionary fiscal policy adds to the government debt stock. Theoretical literature contains that this policy of debt accumulation can have its own impact on several macroeconomic factors such as economic growth and private investment which are the focus of this paper. Finally, government debt has the potential to influence human capital, specially, in developing countries. Economic theory does not provide a unanimous prediction for the relationship of government debt and economic growth rather the conclusion is that the issue is an empirical question. Recently, as the consequence of the rise in global debt trend a number of empirical studies emerged to find a global link between debt and growth in developed and developing countries using sample of large number of countries. Reading related literature three conclusions can be made: 1- the results are contradictory. 2- Such relationship is unique for every country and the results obtained for the panel sample cannot be simply interpreted for each individual country (Checherita and Rother, 2010). 3- While all above argument makes the case for investigating this relationship at country level there are lack of evidence for such empirical studies. Considering the various theoretical prediction and inconclusive empirical findings majority based on large panel samples, the question remains whether the government debt accumulation in ASEAN-4 countries contributed to economic growth? This paper attempts to employ time-series models and tools to address this issue in ASEAN-4 countries. The paper is organized as follows. Section 1.1. provides economic background of ASEAN-4 countries. Section two presents literature review. Methodology, model specification and data are discussed in section three. The result of this paper is presented in section four. Finally, section five is to conclude this research. Economic Background This section elaborates on the historical data of government debt and the issue in ASEAN-4 countries. ASEAN-4 experienced budget deficit during 1970-2019 years except between 46 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 1985-1996, in the economic boom. Although budget deficits may not be very large, they have been persistent and worsened after each economic crisis. As a result, the government debt spiked after adverse events, especially in mid-1980s. Next as the aftermath of Asian financial crisis (AFC, hereafter) in 1997-98 debt rose again, another time after global financial crisis 2007-8 and currently due to Covid-pandamic. Government debt accumulation brief history in each country is followed. Indonesia started 1980 with high amount of external debt utilized to finance development. Declining oil prices in the first half of 1980s resulted in the rapid accumulation of debt from 16.9 (%GDP) in 1981 to 52.44 (%GDP) in 1987. In the late 1980s and mid 1990s, during Indonesia’s economic boom, total foreign debt (both government and private) increased due to high foreign investment. AFC hardest hit Indonesia among ASEAN-4. Already having high level of total external debt, the currency depreciation multiplied the burden of government external debt. To restore market confidence Central Bank bailed out private companies including private bank’s external debt. These events explain the massive evolution of government debt in a short time. Government debt spiked from 26.8 (%GDP) in 1996 to 92.91 (%GDP) in 1999. Since 2000, authorities focused on debt reduction to prevent negative consequences of running large debt and to create fiscal space. Meanwhile, Indonesia has undergone fiscal, economic, social and political reform. Fiscal and economic reforms that were part of the debt exit strategy and fiscal sustainability include but not limited to: government decentralization, cutting off food and energy subsidies and privatization. Such that, the burden of debt service declined leaving some resources for development expenditure. Moreover, the composition of government debt has shifted toward domestic debt. The ratio of domestic debt relative to total debt increased from 45% in 2005 to 54% in 2010. In Malaysia, large stock of government debt cumulated from 1980 to 1987 reaching 100 percent of GDP due to increase in government spending on projects in the aftermath of commodity crisis 1985-86. Afterwards the debt reduction episode embarked from 1987 to 1997 in that government debt reduced from 80% to 35%, respectively. The decline was due to some factors such as high economic growth, sound fiscal policy, well debt management and privatization of government agencies leading to debt reduction to the minimum of 31.9% in 1997. Since government of Malaysia dominantly relies on domestic borrowing, its debt is not sensitive to exchange rate fluctuation. After the experience of mid-1980s crisis Malaysia has taken a conservative approach to foreign debt; that was beneficial in AFC considerable currency depreciation. After AFC, government utilized expansionary fiscal policy to alleviate the effect of economic slowdown. During 1998-2002 expansionary monetary policy via lowest interest rate supported private sector activities. Fiscal consolidation effort imposed by the government from 2000 to 2007. But, subsequent to global financial crisis (GFC) government debt increased again as large fiscal stimulus was implemented. Something to add for recent event. In the Philippines, the historical high government debt level in 1986, was accumulated since 1970 as a result of sizable and burdensome borrowing to finance investment in a short time. Beside banking crisis in 1980s, the Philippines defaulted on its external debt in 1983. External debt crisis continued until 1992. during that time, servicing the debt was the main challenge for the new government of Aquino. By improving economic condition from 1987 to 1998 the country could reduce the debt level from 80% to 51% respectively. By the outburst of AFC the debt took rise again until it reached the peak of almost 70% in 2003. In following years, as the economy recovered Philippines managed to reduce its debt to the minimum of 47 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries 39.6% in 2019. The debt reduction accomplished mainly by government significant fiscal consolidation via reducing expenditure, privatization, and major tax reforms in 2005-6 to improve poor government revenue. In Thailand during 1970s, average of government debt to GDP was 22.3%. But the sustained deficits in the 1980s led to a rapid rise in the government debt level leading to a peak of over 36 per cent in 1986. Meanwhile, in this period Thailand suffered from banking crisis from 1980 to 1987. The period of budget surpluses during economic boom from 1987 to 1996 created the resources to pay down the government debt. And the debt ratio fell to a record low of about 11 per cent of GDP in 1996 as implementing government fiscal consolidation including 3% of GDP reduction in government spending from mid 1980s to late 80s. But the AFC in 1997 made a dramatic rise in government debt in following years. Debt reached the climax of near 60% GDP in 2001. The sharp upward trend triggered worrisome and forecast of debt to be exceeding 65% in 2005. But actually Thailand managed to considerably reduce its debt from 2001 to 2007 by about 20% of GDP. Global crisis in 2008, reversed the trend upward, yet remained around 40% (GDP) till 2019. Table 1 shows the average level of government debt of ASEAN-4 in the last five decades. The debt level is far more than the 25% recommended by IMF (Makin, 2005). While the general trend of government debt in Malaysia has been upward since 1997, that has triggered concerns among the policy-makers and the public in the recent years in Indonesia in the late 2010s debt is about and below 30% GDP that is considered relatively low. Some economist criticize too low government debt as it prevents formation of necessary capital investment in the country. Having said that, the debt management in ASEAN-4 is known to be prudent. According to formal statistics the operational budget has been in surplus for almost all of the 1985-2019 years. Prudent governments should only borrow to finance investment to be able to maintain a sustainable budget balance. Much of the change in fiscal positions in ASEAN-4 is explained by discretionary fiscal stimulus packages consisting of development expenditures. Government debt can increase in the future due to factors such as infrastructure spending, stimulating private consumption and expanding social safety nets (Kawai and Morgan, 2013). As a preliminary data analysis Figure 1 shows the scatterplots of the pair of initial government debt-GDP growth data with a fitted regression line for each sample country. The slope of the lines are negative for Indonesia and the Philippines. In contrast, for Malaysia and Thailand the slope is positive. A part of this initial result seems to be in line with the general negative view about the effect of government debt. Whereas the other part is in line with growth-supporting view similar to the result of Ferreira (2016) and Abbas and Christensen (2010). This further highlights the need for a formal analysis of the issue in each of ASEAN-4 countries. Literature Review In economic theory literature, there are different views regarding the effects of government debt. That can be classified into four categories. First is the Keynesian’s view that suggests in the short run government debt-financed fiscal expansion, specially in recession periods, can increase demand and subsequently output. Second view, is the so-called Ricardians, which following the work of Barro (1991). Barro argued that government borrowing which creates deficit in government saving will be compensated by private sector’s saving rise, thus, government borrowing action cannot have any effect on the real sector. Third group refers to the conventional neoclassical view of public debt. They hold negative effect for public debt rise (Mankiw, 1999) as it reduce capital formation and thus output in the long run. The 48 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 literature contains two mechanisms for the reduction in capital accumulation that are called crowding-out effect. One way is that, large government borrowing could compete with private sector in loanable funds market, and reduce available funds for private sector. Another way, is crowding-out through interest rate; as the government borrowing may push the interest rate up, therefore, reduces the private investment. Other adverse effect could occur via expected higher uncertainty about the economy and future taxes when the government debt level is high. Moreover, “debt overhang” hypothesis coined by Krugman (1988) argues that in low income developing countries when the government debt surpasses its future capacity to pay off the debt, economic growth declines. Having high opportunity cost, a high proportion of a countries’ foreign exchange earnings should be used for servicing the foreign debt (Krugman, 1988; Sachs, 1989). As Hofman and Reisen (1990) puts it the requirement to service debt reduces funds available for investment purposes. In addition, a high proportion of government expenditure budget will be absorbed by debt service burden, thus, changes the composition of public spending (Checherita-Westphal and Rother, 2012). Above reasons decline incentive to invest. Table 1: Average government debt level in five decades in ASEAN-4 Government Debt (as percent of GDP) Year Indonesia Malaysia Philippines Thailand 1970s NA 42.13 17.14 22.33 1980s 34.4 79.22 51.6 40.30 1990s 45.84 50.19 60.21 21.33 2000s 53.46 43.00 55.86 46.27 2010s 29.07 53.13 43.20 43.00 2019 30.5 52.7 39.6 41.1 2020 38.5 60.7 53.5 50.5 Fourth group among neoclassicals showed that government debt can contribute to GDP growth, utilizing endogenous growth models that incorporate government debt in the supply side of production function. These theories usually consider conditions for this outcome to happen such as when government spend borrowings into productive investment in the country. In developing countries, especially at the initial stages of development, the return on public investment in infrastructure such as roads, electricity, etc., human capital and health is quite high. Yet, usually there is a threshold for borrowing, exceeding that limit negative consequences gets larger than the positive impacts (Greiner, 2007). Empirical studies on the effect of government debt before the notable study of Reinhart and Rogoff (2010) was a few and mostly about advanced economies specially the U.S. More studies in developing countries had focused on external debt effect motivated by debt overhang hypothesis of Krugman (1988) and Sachs (1989). Pattillo, et. al (2002, 2004) reported negative and significant impact on growth at high debt levels (typically, over 60 percent GDP), but an insignificant impact at low debt levels. In contrast, Cordella, et. al. (2005) found evidence of debt overhang for intermediate debt levels, but an insignificant debt-growth relationship at very low and very high levels of debt. Overall the mainstream view about the effect of high government debt was negative (Elmendorf and Mankiw, 1999). Pattillo, et. al (2004) found that the negative effects of external debt transmit to growth through physical capital accumulation and total factor productivity. 49 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Figure 1: presents the initial debt-GDP growth scatterplot for each country The first left plot shows the pair of data for Indonesia, that has a correlation of -0.109. Next plot refers to the data of Malaysia with a correlation of 0.233. Following plots on the right show data of the Philippines and Thailand with correlation of -0.058, and 0.263, respectively. After 2007-2008 financial crisis followed by the European debt crisis, debt-growth relationship attracted numerous studies, leading by Reinhart and Rogoff (2010). De Rugy and Salmon (2020) reviewed 24 panel studies in the last decade (2010 to 2020) and concluded that all except two studies found negative effect of high government debt on output growth irrespective of finding a threshold or not. They also mentioned that the majority of the thresholds were found to be between 75 to 100 percent of GDP. However, Law et al. (2021) found a much lower threshold of 51.65 % GDP for 71 developing countries. Checherita and Rother (2012) found negative relationship between government debt and growth at high levels of government debt (90 % GDP). The same result was found for the four channels of transmission, namely private saving and investment, public investment, total factor productivity and sovereign long run interest rate for the sample of 12 Euro countries. They mentioned that the relationship bellow threshold of 90% remains a question. Unlike above mentioned, other studies did not find a common threshold. For example, Chudik et al. (2017) using panel ARDL for a sample of 40 developing economies finds negative effect. Negative effect and significant effect was also found by Woo and Kumar (2010) for a 50 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 sample of 38 advanced and emerging economies during 1970-2007. They employed multiple panel estimators and accounting for several econometric issues. Moreover, Schclarek (2004) found linear negative and significant relationship among public external debt and GDP per capita growth and also between public external debt and capital growth for a panel of 59 developing countries during 1970-2002. Using exogenous threshold dummies of total external debt of 20% GDP and 30% GDP alternatively no evidence for nonlinear relationship was obtained. Panizza and Presbitero (2013) in a recent survey of debt-growth nexus in advanced countries emphasize on the point that the debt-growth nexsus is not homogenous across countries, and future research should take account of this fact. Among the rare evidence from the single country studies in developing countries include Bal and Rath (2014) that examined the effect of public debt (divided into domestic and external debt) on GNP per capita. Other explanatory variables were debt service, total factor productivity and export. They found significant adverse effect for both public debt variables and recommended to reduce the debt. Prior to Bal and Rath, Singh (2012) investigated domestic debt and growth relationship in India for the period of 1959-95, using cointegration and Granger causality test and concluded that Ricardian equivalence prevails as debt did not Granger cause growth. Moreover, Swamy (2015) observed a negative relationship from debt to growth. In contrast to above negative results, Thao (2018) is among the few researches that found government debt promotes GDP growth. He used a panel sample of 6 ASEAN countries, namely, Indonesia, Malaysia, Philippines, Singapore, Thailand and Vietnam over 1995-2015. Employing GMM estimation technique he found significant positive effect of government debt on GDP growth. While most of the recent papers as reviewed above, have utilized single equation models and panel estimators such as the generalized method of moments (GMM) and system-GMM, application of vector autoregressive (VAR) modelling and impulse response tool is rather unique. Statistically, this approach is suitable to disentangle the negative debt-growth correlation usually found in empirical papers. In other words, most of the previous findings only showed the existence of negative correlation but did not address the issue of causality. That means, whether high debt negatively affects output growth or low output growth causes government debt to increase. Few recent papers such as Lof and Malinen (2014) applied panel bivariate VAR model and impulse response technique to address this issue. Their robust result indicated debt did not cause economic growth. The impulse response of the total sample in the said study shows an almost insignificant response. However, in low and medium debt regimes, 0-30% and 30-60% respectively, debt shock induced a positive and significant response to economic growth. These studies are outstanding from other literature in terms of the applied methodology. Nonetheless, some shortcomings prevail. First, like most of the relevant literature on this issue, they used large panel samples to derive global stylised facts. Country-specific studies are scant. Second, they used bivariate models. Lof and Manila (2014) admit that although the bivariate model is useful for decomposing the correlation, it does not provide any information about the economic channels though which debt affects growth or vice versa. In a survey study of advanced economies, Panizza and Presbitero (2013) concluded that although empirical studies tend to find negative effects of high debt on output growth there is no paper that makes a strong case for causality of debt to growth. Nonetheless, the results are quite contrasting in a new strand of debt-growth research. For example, a number of research such as Ferreira (2016) for 28 European countries, Ferreira (2009) for 20 OECD countries, Butts (2009) in 27 Latin American and Caribbean countries, Abbas and Christensen (2010b) in 93 low-income countries and emerging markets, found bidirectional relationship. 51 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries In contrast, no causality relation was found by Puente-Ajovín (2015) in 16 OECD countries; Panizza and Presbitero (2014) in a sample of 17 OECD countries and Jayarama and Lau (2009) for six Pacific island countries. Moreover, other findings such as in Jacobs et al. (2020) for 31 OECD countries, Kempa and Khan (2016) in G7 countries, Lof and Malinen (2014) in a sample of 20 advanced countries, indicate uni-directional relationship from growth to debt and not vice versa. In sum, majority of the extant empirical literature on the effect of government debt on GDP growth shows negative effect based on large panel data, whether a threshold is found or not (De Rugy and Salmon , 2020), without making a strong case for causality of debt to growth (Panizza and Presbitero, 2013). Although it has become a fact that debt-GDP growth is a country specific empirical issue both in the matters of causality and sign there is a lack of studies based on single countries including ASEAN-4 economies. Data and Research Method The model in this paper follows the standard neoclassical growth model, which has been the workhorse for examining the effect of government debt on output growth in the empirical papers. More specifically, our model is based on derivation of Mankiw, et al. (1992) of the Solow growth model which shows, output per capita depends on population growth, and investment in physical capital, and human capital. b a a ln Yt = ln A0 + gt + 1 - a ln (sk) - 1 - a ln (n + g + d) + 1 - a ln (h) (1) Where: A0 is level of technology; yt is the output per capita or labour; h is the level of human capital; sk the share of output that is allocated for physical capital accumulation (which could be indexed by investment as percent of GDP). And finally (n+g+δ) is population growth, technological growth and rate of depreciation respectively. This paper extends above output per capita function, to include government debt. Adding debt variable (external or government debt) to the growth equation to investigate the effect of it on growth is a common practice in empirical literature. To name a few examples of this approach in panel framework, are: Pattillo, et. al., (2011), Sen et al., (2007), Clement et al., (2003), Checherita-Westphal and Rother (2012), Schclarek (2004); and in time-series framework: Bal and Rath (2014), Mohd Daud et. al., (2013), Asmaddy and Mohammad (2015). Therefore, present study employs bellow baseline growth model: lYt = F (lpinv, IHK, ID) (2) Where: lyt is the real per capita GDP; lpinv shows private investment ratio which is captured by ratio of private fixed capital formation to GDP. lHK is human capital stock indexed by total average years of schooling of the population above 25 years old. lD is the ratio of government debt stock as percent of GDP which represent the government debt burden of the country. All variables are used in logarithm form indicated by l. Data are collected for the period of 1985 to 2019 from the following sources. The data of government debt is collected from A historical public debt database, Abbas, S. M, et al. (2010), government debt from 2010 to 2019 is obtained from the World bank database. The data of human capital is obtained from Barro-Lee Human capital database (total population over 25 years), v. 2.1, Feb. 2016 . All other data are collected from World Bank database, June 2021. For the years of 2010 to 2019 Barro-Lee data projection was used. In addition, this model incorporates some control variables, which are important determinants of both economic growth and government debt, namely, budget balance, ex52 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 change rate volatility and real interest rate. Budget balance variable captures general economic instability; real interest rate is included as a monetary policy tool. Finally, exchange rate volatility, that is an economic risk factor, could reflect external shocks which are relevant to ASEAN-4 economies. Large currency depreciation during AFC, considerably increased external government debt value in terms of domestic currency. That experience lead to government debt restructuring and relying further on borrowing from domestic market. These control variables are expected to be stationary. This fact led us to specify the control variables in the vector of exogenous variables1 . This approach is beneficial for the need to keep the number of endogenous variables small partly due to small sample size. In addition, employing rather small VAR model will reduce the problem of model identification. This study Utilizes a VAR model with four endogenous variables, namely government debt, GDP per capita, private investment and human capital; and three exogenous variables including budget balance, exchange rate volatility and real interest rate. Since all explanatory variables are endogenous with respect to the dependent variable and we found strong different cointegration relations in each of the sample countries, employing a VAR model is preferred over the single equation modelling (Stock and Watson, 2001). In a VAR model all variables can be endogenous whereas using single equation estimators such as ARDL, FMOLS, DOLS explanatory variables must be exogenous with respect to the dependent variable (Pesaran et. al, 2001), in this case, GDP growth. From theoretical perspective these single-equation models assume a single cointegration relation normalizing on the dependent variable (GDP growth) which is not the case here. Table 2 summarizes the variables of the model. The contribution of this paper is to utilize a small-reduced VAR model to entangle the issue of debt-growth causality beyond bivariate VAR models and investigate the dynamic response of GDP growth to a shock to government debt stock in individual ASEAN-4 economies using historical data from 1985 to 2019. As a by-product of the model, responses of the two major growth factors, namely, private investment and human capital to a debt shock are presented. However, the model is focused on GDP growth. Basically, in this VAR model, GDP per capita equation represents endogenous growth model (equation 4). This model in terms of variable specification resembles the growth model used by Mariotti (2002) who investigated the effect of government spending on GDP growth. The theoretical and empirical support for including government debt in growth model was reviewed in section 2. Private investment equation represents endogenous investment model (equation 5); the prominent empirical paper by Aschauer (1989a, 1989b) is among the early works that emphasized on the role of fiscal policy on private investment. Aschauer (1989a) concludes that government budget deficit can be important in determining private investment decisions and in long term, government debt can be found to crowd-in private investment if it is spent on productive public capital that has spill-over effect to private investment. Bende and Slater (2003) in both individual country estimation but significant only for Malaysia (negative-significant in the Philippines) and a pool-sample of ASEAN-4 countries (1971-1999) found crowding-in effect of external debt. The theoretical and empirical literature consider private investment as an important channel through which debt affects economic growth. Human capital is written as a function of output growth index, private investment, government debt (equation 6). Pattillo, et al. (2004) argued that government debt is a determinant of human capital and investigated the effect of government debt on human capital. And finally, we have government debt as a function of output per capita, private investment, 1 Similar approach is used by Guimaraes and Unteroberdoerster (2006). 53 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries human capital and the control variables, budget balance, interest rate and exchange rate volatility (equation 7). Low or drop in output growth can cause government debt to rise. Also high economic growth can contribute to debt reduction if it is the government policy. Government budget balance, interest rate and exchange rate volatility (regarding external portion of government liabilities) are also important determinants of government debt. Table 2: Variable Description Variable Form-transformation Description Source ly Real GDP per capita Real per capita –local currency Index for growth D Gross debt Gross outstanding central govern- IMF3 new database ment debt pinv Private domestic in- % GDP vestment Hk Human capital stock Linear interpola- Average years of total school- Barro-Lee Human caption with respect ing of people age above 25 years ital database (2016) to time old. BB Budget balance vex olatility of real ex- change rate to USD Subtracting trend (using Ho- WB drick-Prescott filter) from real USD exchange rate r Real interest rate Bank lending interest rate, usually WB refer to short and medium-term needs of borrowers, adjusted for inflation by the GDP deflator government % GDP % GDP - WB2 fixed capital formation-private WB 4 sector minus FDI (%GDP) Final budget balance WB and country’s Central Bank websites Notes: 1. All data are annual and transformed into logarithm form 2. WB indicates World Bank database June 2015 latest version. 3. IMF, indicates International Monetary Fund 4. Private investment for Indonesia is not provided in World Bank (WB) 2015 database. The paper used data from (Bende-Nabende & Slater, 2003) data for 1985-1999 years. Data for 2000-2014 was calculated by reducing government development expenditure plus FDI from total fixed capital formation. Data of development expenditure is taken from Ministry of Finance Indonesia. The rationale behind it is that most of government development expenditure in Indonesia is in infrastructure. Therefore, development expenditure is a good proxy for government investment or government capital formation. lows: Referring to above model specification the reduced form VAR representation is as fol- IDt = lyt - 1 + g + lyt - 1 + lpinvt - 1 + g + lpinvt - k + IHKt - 1 + g + IHKt - k + IDt - 1 + g + IDt - k + lbbt + lrt + lvext + c (3) lyt = lyt - 1 + g + lyt - k + lpinvt - 1 + g + lpinvt - k + IHKt - 1 + g + IHKt - k + IDt - 1 + g + IDt - k + lbbt + lrt + lvext + c (4) lpinvt = lyt - 1 + g + lyt - k + lpinvt - 1 + g + lpinvt - k + IHKt - 1 + g + IHKt - k + IDt - 1 + g + IDt - k + lbbt + lrt + lvext + c (5) 54 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 IHKt = lyt - 1 + g + lyt - k + lpinvt - 1 + g + lpinvt - n + IHKt - 1 + g + IHKt - n + IDt - 1 + g + IDt - n + lbbt + lrt + c (6) The optimal lag for the above VAR model is chosen based on Akaike information criterion (AIC). Since this criterion compared to Schwarz information criterion (SIC) selects higher order of lags, which is more appropriate in order to better capture the behaviour of the data in small samples. Prior to estimation of the model the level of integration of the variables must be determined by unit root test. Augmented Dickey Fuller (ADF) test is a dominant test procedure that is also followed in this research. Usually it is the case that economic time-series are integrated of first order. Furthermore, economic theory suggests the variables in the growth function, namely, output growth, investment rate, government debt and human capital are cointegrated. Applying Johansen cointegration test, verifies this issue. Information regarding unit root and cointegration has implication in VAR model specification. Engle and Granger (1987) suggest that both VAR model at level and vector error correction model (VECM) which regress variables at first difference on their lags, are appropriate to represent the cointegrated relationship among the variables. That means the long run constraint which is imposed on the level variables in VECM are also satisfied asymptotically in a level VAR model. Ramaswamy and Slok (1998) outlined several advantages of using the level VAR as opposed to VECM. The most notable argument in favour of the level VAR is the economic interpretation attached to the impulse-response functions of the model. While the impulse responses from the VECM tend to imply that the impacts of certain shocks are permanent, those from the level VAR allow history to decide on whether the effects of shocks are permanent or not. They further note that if there is no a prior theory to suggest the number of cointegrating vectors and how to interpret them, the VAR model in level for cointegrated series is a reasonable approach. The issue of causality between GDP growth and government debt had been less attended in related empirical studies. To examine this issue this research employs Toda-Yamamoto (1995) method. The merit of using this method compared to Granger causality test by Granger (1969) is that it can be used to test general restrictions on the parameter matrices even if the processes may be integrated or cointegrated. Given that there is cointegration among the variables, it is expected that at least one direction causality will be obtained. In order to make inferences about the effect of government debt on output growth, generalized impulse-response (GIR) method proposed by Pesaran and Shin (1998) is used. The graphical response function provided by this method provides dynamic response of the variable through the specified 10 year-period, including the magnitude and sign of the effect within the bootstrapped generated confidence intervals of ± 2 standard deviation. If both of the generated bootstrapped lines fall either in positive or negative area, then, it can be inferred that the effect is significant, otherwise insignificant. The merit of using GIR is to circumvent the issue of variable ordering 2 meaning that this method is invariant to the order of the variables. Unlike other identification methods such as Cholesky ordering that hold some variables constant at the time of shock, GIR takes variables to be variant. This property of GIR is quite useful in current situation. Usually, ordering of the variables come along with several problems that makes the ordering assumptions unrealistic. The potential simultaneous effect among some variables and country specific economic differences are among the challenges for variables ordering. However, to justify appropriateness of the choice of GIR shock identification, the result of Cholesky impulse-response based on two alternative variables ordering will be presented in 2 This problem is dominant in the popular Sims or Cholesky method 55 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Appendix A. Moreover, Lof and Malinen (2014) found similar result from the VAR model which debt was first placed and the model that debt came second after output growth. Results and Discussion Table 2: ADF Unit Root Test Result of The Variables Included in Growth Model for ASEAN-4 Countries Variable Variable at Level Variable at Frist Difference C C.T C Indonesia ly -1.527 -4.202* lpinv -3.126 -7.584* IDebt -2.438 -3.986* IHK -3.118 -3.604 lBB -4.982* lVEX -3.450* lr -3.636 Malaysia ly -1.719 -5.263* lpinv -3.063 -4.201* lDebt -1.011 -4.131* IHK -3.101 -6.305* ly 0.370 -3.015 lpinv -2.490 -6.746*** lDebt -3.366* -6.030* IHK -2.708 -6.539* ly -3.495* -3.203** lpinv -3.277* -4.003* lDebt -1.884 -4.096* IHK -1.626 -5.232* lBB -3.214 lVEX -3.895* lr -5.885* Philipines lBB -5.110* lVEX -5.901* lr -4.618* Thailand lBB -6.135 lVEX -5.463 lr -4.980 Optimal Lag length selection in ADF unit root testing was automatically selected from an automatic maximum lag based on SC criteria. , , indicate 10%, 5% and 1% level of significance respectively. C and T denote constant and trend respectively. 56 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 The result of ADF unit root test is presented in Table 2. The endogenous variables for all countries are I(1) at 5% significance level and the variables become stationary at first difference. As explained in methodological framework the endogenous variables enter the model at level. As expected based on theory, they form a cointegrating relation together as verified by the result of Johansen cointegration test in Table 3. That implies combination of I(1) variables becomes a stationary process. The control variables that are defined as exogenous in this model setting are shown to be I(0). Table 3 shows that there is at least one cointegrating relation in every country among government debt, output per capita, private investment and human capital. Table 3: Johansen Cointegration Test Result Hypothesized no of CE Trace Statistic Critical Value Prob Max-Eigen statistic Critical Value Prob Indonesia None* 105.203 47.856 0.00 49.547 27.584 0.00 at most 1* 55.655 29.797 0.00 33.243 21.131 0.00 at most 2* 22.411 15.494 0.00 20.376 14.264 0.00 2.035 3.841 0.153 2.035 3.841 0.15 None*** 70.091 47.856 0.00 41.158 27.584 0.00 at most 1* 28.932 29.797 0.062 20.572 21.131 0.059 at most 2 8.359 15.595 0.427 8.345 14.264 0.344 at most 3 0.014 3.841 3.841 0.014 3.841 0.904 146.141 47.856 0.00 74.784 27.584 0.00 at most 1 71.356 29.797 0.00 48.251 21.131 0.00 at most 2 23.104 15.494 0.00 22.802 14.262 0.00 at most 3 0.302 3.841 0.582 0.302 3.841 0.582 152.091 47.856 0.00 85.291 27.584 0.00 at most 1* 66.799 29.797 0.00 52.340 21.131 0.00 at most 2 14.459 15.494 0.071 14.420 14.264 0.047 0.039 3.841 0.842 at most 3 Malaysia Philipines None* Thailand None* at most 3 0.039 3.841 0.842 Note: All specifications include constant and trend in VAR model. - Prob means Probability Having the information regarding stationarity of, and cointegration among the variables, we can proceed to estimation of the model. VAR model with optimal lag of three was chosen for all countries. Then the Toda-Yamamoto Granger causality test was performed, the result of which is presented in Table 4 below. The result of Toda-Yamamoto Granger causality for Indonesia, shows that government debt does not cause output growth. But it is the GDP growth that Granger cause government debt at 1 percent significance level. In case of Malaysia, no causal relationship from debt to GDP growth was found. However, strong evidence was obtained for output growth causing government debt in Malaysia. In the Philippines, government debt and GDP growth are shown to have bidirectional relationship. Government debt causes growth at 1 percent significant but the reverse causal relationship is rather weak only significant at 10 percent. 57 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Table 4: Toda-Yamamoto- Granger causality test result VAR MODEL INDONESIA MALAYSIA PHILIPINES THAILAND Y"B VAR (P=3) 3.276 (0.35) 3.016 (0.399) 20.182 (0.007) 5.087 (0.165) B"Y VAR (P=3) 12.427 (0.00) 17.540 (0.00) 7.031 (0.07) 67.982 (0.00) The result for Thailand, similar to Indonesia and Malaysia supports GDP growth causing government debt (at 1 percent significant level) not vice versa. Overall, result of Table 4 provides evidence for unidirectional causality that is from GDP growth to government debt. Only in the Philippines strong evidence for government debt causing GDP growth was found. In the next step, the result of generalized impulse-response based on VAR (p*=3) for all sample countries are presented in Figures 3 to 6. In Indonesia, the result of impulse response (Figure 3) shows a shock to government debt has negative effect on output growth only in short run while the response becomes insignificant after the first year. This result is compatible with the evidence of no causality from debt to GDP growth (Table 4). Although single equation models which have been examined by the authors of this paper (the result is not reported) tend to render a significant negative coefficient for government debt, this significant result is suspected to be due to the endogeneity problem. That is the case in Handra and Kurniawan (2020) and Cholifihani (2008). Regarding private investment although the response line is in negative area but the confidence bands encompass the zero-line (Figure 3); therefore, the private investment response is not significant. Within this concept, Adiningsih (2009) using monthly data of 2000-2008 and ECM model claims that rising government debt increases interest rate, thus, crowds out private investment. The response of human capital is negligible as well. Finally, response of government debt to itself shows that the increase in government debt fades away in a ten-year period. In Malaysia, GDP growth response to a shock to government debt is increasingly positive but statistically insignificant (Figure 4). This result is compatible with the Granger causality result in that, debt did not Granger cause output growth in Malaysia (refer to Table 4). Similarly, positive and partially significant result was obtained by Burhanudin et Al. (2017) employing ARDL model with one lag and data of 1970 to 2015 of Malaysia. However, as mentioned in this study, choosing single equation models (e.g. ARDL) in this case is not appropriate. Similar to output growth response, private investment response to one standard deviation (S.D) debt shock is initially negative but then rising and positive. Yet, the response remains insignificant. Although their variable is not the same as present study, Bende and Slater (2003) found positive effect of ‘external debt’ to private investment for ASEAN-4 countries during 1971-1999. The government debt could have positive impact on Malaysia’s GDP growth through increasing productivity. Using data from 1970 to 2012, Asmaddy and Abubaker (2015) found positive effect of government debt on growth of total factor productivity in Malaysia. Another reason that prevents crowding-out of private investment in Malaysia is its deep and liquid government bond market that provides the government with reliable and relatively low interest rate financing. This issue plus complementary monetary policy prevent interest rate to increase and reduce private investment. A policy of debt reduction in the long-run can be observed through the response of government debt to a shock to itself (Figure 4). The insignificant response of output growth and private investment is compatible with the Ricardian’s view implication. However, the result of above examination is not enough to conclude Indonesian and Malaysian consumer’s 58 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 behaviour are of Ricardian type. For that to conclude needs further hypothesis testing in other studies. In the Philippines (Figure 5) response of GDP to government debt shock is positive and significant for four and half years, although, both private investment and human capital response are close to zero line; thus, insignificant. The short term to medium run positive output growth shows that Philippines economy is debt-driven. In Thailand, debt shock generates a positive GDP response in short run (Figure 6). The response of output growth is significant in the first and second years. Furthermore, some positive but insignificant response of private investment (except third year that is positive and significant) and human capital are obtained in the short run before the effect tend to zero. Overall, comparing the result of this study to previous literature, some are supportive and complementary to our finding, but some are contrasting. The contrasts, however, are due to different samples, different variables and methodology. For example, Sen et al. (2007) for a panel of six Asian countries including China, India, South Korea, Indonesia, Philippines and Thailand (1982-2002) and using a GMM estimation technique found external debt has moderate negative Effect on GDP growth. A noticeable critique of their paper is the inclusion of two indices of external debt (namely, external debt to GDP and external debt to export) simultaneously in the model - which is not correct. Figure 3: Response of Government Debt, Private Investment, Per Capita Output and Human Capital Stock to a Generalized 1 SD Innovation in Government Debt, Indonesia, 19852019. 59 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Figure 4: Response of Government Debt, Private Investment, Per Capita Output and Human Capital Stock to a Generalized 1 SD Innovation in Government Debt, Malaysia, 19852019 Figure 5: Response of government debt, priate investment, per capita output and human capital stock to a generalized 1 SD innovation in government debt, Philippines, 1985-2019 60 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 Figure 6: Response of Government Debt, Private Investment, Per Capita Output and Human Capital Stock to a Generalized 1 SD Innovation in Government Debt, Thailand, 1985 -2019 Evidence against nonlinear model specification To check the validity of linear modelling assumption, Figure 7 shows the GDP per capita equation’s residual of the VAR model vs initial government debt, for Indonesia, Malaysia, The Philippines and Thailand, respectively from left to right. There is no linear or nonlinear pattern evident in these scatterplots. Moreover, the correlation of the paired data is close to zero, that verifies the linear modelling assumption was appropriate. The correlations from left to right consists of -0.028, 0.000, 0.122, 0.011. If the scatterplots showed a strong linear relationship among debt and the VAR residuals it meant that true model specification had to be nonlinear. Figure 7: Scatterplot of VAR Residuals (GDP per capita equation) Versus Initial Debt for Indonesia, Malaysia, The Philippines and Thailand, Respectively from Left to Right 61 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Further Discussion Overall, the result shows that government debt does not cause GDP in the long run. Along this line, the response of GDP to debt shock was insignificant in Indonesia, Malaysia and Thailand although the response of GDP to debt shock was negative, positive and positive, respectively. Similar to this study’s findings, other studies that support unidirectional causality from GDP to debt not vice versa include Kepma and Khan (2016) and Lof and Malinen (2014). Exceptionally, in The Philippines, debt causes GDP at 1% significance, and the GDP response to debt shock is positive and significant as well. The overall insignificant outcome of this study may suggest that the positive and negative effects of the government debt offset each other in Indonesia, Malaysia and Thailand or in other words, the government debt has been hovering around the optimal level. In addition, Apart from the theories that explain the effect of government debt on economic growth, there could be several economic factors influencing the debt-growth relationship in each individual country such as macroeconomic factors, debt regime and fiscal management, the nature of debt evolution and accordingly the exit strategy. In Indonesia, the government debt burden which is at some episodes huge did not significantly impair output growth, private investment or human capital. Empirical studies suggest the importance of fiscal management and other macroeconomic policies on the outcome of debt-growth relationship. Indonesia is known for using prudent fiscal policy that act as economy anchor (Blondal et al., 2009) and wise macroeconomic management in order to pursue economic growth. More importantly, the insignificant result in case of Indonesia could be attributed to the nature of debt evolution. A large part of the huge debt stock which was accumulated since 1997-1998 crisis was not due to government expenditure but due to government bank bailout. Another cause of it was exchange rate depreciation that increased the external debt obligation in terms of domestic currency. Had to exit from this debt burden Indonesia’s government well managed to do so, by utilizing several major revenue resources other than tax revenue; such as, sale of equity since some government bailout was in exchange of capital share (ref). The debt reduction effort can be observed since 2000. As earlier mentioned, without imposing higher tax burden government managed to considerably reduce its debt over the course of 12 years (2000-2012). After this period government debt is increasing from near 23% in 2012 to 30.5% in 2019. In Malaysia, public debt accumulation is mainly due to government development expenditure. Therefore, it is expected that in the long run the spill-over effect raise the economic growth specially in boom cycles. The response of output growth in this study is increasingly positive but not significant. Beside well macroeconomic and debt management, the insignificant result could be due to the debt regime of Malaysia. Major of the debt observations in Malaysia is within 30-60% range and the average debt of Malaysia in the sample period (19852019) is 55.51% GDP. Meaning that on average the debt is kept near to the self-imposed debt ceiling of the country which is 55% GDP. In that sense, in current range of government debt - ceteris paribus - no statistically significant impact to economic growth was found. In Thailand, although no causality was detected from government debt to GDP, the impulse-response result suggest some economic stimulation in the short run. Knowing that Thailand’s government use countercyclical fiscal policy in face of economic downturn, the positive growth response supports Keynesian prediction and the successful use of fiscal expansion policy in stimulating the economy in the short run. The response of debt to debt shock also shows that debt reduction is made fast. This could in turn, be attributed to fast economic re62 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 covery. Given high investment rate and human capital level in Thailand, The primary purpose of its government in using fiscal deficit could be short term economic stimulation (ref). For the Philippines the results shows government debt shock in a moderate debt regime can contribute to GDP growth. Similar evidence for positive causality (Keynesian effect) was found in Ferreira (2016) in a panel of 28 EU countries especially after the outbreak of global financial crisis; and Abbas and Christensen (2010a) who concluded that a moderate noninflationary debt level has a generally positive impact on economic growth. However, it is necessary to treat this finding with caution because there is concerns beneath the surface of this good result. In previous section evidence was interpreted as the Philippine economy is debt-driven. Meaning that time to time rise in fiscal deficit is used to push up the economic demand and therefore, output growth based on Keynesian fiscal multiplier effect. Although this policy is recommended in certain periods of low demand, it is not advisable to be used frequently in the long term, as it can further economic fluctuation. In addition to that, government debt does not show to improve important growth factors such as private investment and human capital stock despite the need to accelerate physical and human capital accumulation in the Philippines. In sum, it is likely that government debt was not used effectively and productively in the right time and in development areas to support sustainable output growth. Conclusion Although the overall empirical evidence based on panel studies, suggest negative relationship between debt and output growth, or some found evidence of nonlinearity, this study attempts to shed more light on the issue by focusing on single emerging economies with overall moderate debt regime. Using a number of time-series econometrics techniques such as Toda-Yamamoto Granger causality test, this study concludes that there is unidirectional causality from GDP to government debt in three ASEAN countries, namely, Indonesia, Malaysia and Thailand. Only in The Philippines the result support a bidirectional relationship, while debt cause GDP at 1 percent level but GDP cause debt at 10 percent significance level. Furthermore, the result from generalized as well as Cholesky impulse-response analysis of the multivariate VAR model verified above causality conclusions. Given this result, it can be inferred that the fiscal and debt regime implemented since 1985 have been generally well-managed. So that, no negative effect on GDP is evident. Although expectations of statistically significant positive effect on GDP found support only to a limited extent. References Abbas, S. A., & Christensen, J. E. (2010a). The role of domestic debt markets in economic growth: An empirical investigation for low-income countries and emerging markets. IMF Staff Papers, 57(1), 209-255. Abbas, S. M., Belhocine, N., ElGanainy, A. A., & Horton, M. (2010b). A historical public debt database. https://www.imf.org/en/Publications/WP/Issues/2016/12/31/A-Historical-Public-Debt-Database-24332 Adiningsih, S. (2009). The impact of government debt issuance on short-term interest rates in Indonesia. Gadjah Mada International Journal of Business, 11(3), 301-316. Aschauer, D. A. (1989a). Does public capital crowd out private capital? Journal of Monetary Economics, 24(2), 171-188. Aschauer, D. A. (1989b). Is public expenditure productive? Journal of Monetary Economics, 63 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries 23(2), 177-200. Asmaddy, H., & Mohammad, A. K. (2015). The impact of federal government debt levels on productivity growth in malaysia. Paper presented at the Kuala Lumpur International Business, Economics and Law Conference. Bal, D. P., & Rath, B. N. (2014). Public debt and economic growth in India: A reassessment. Economic Analysis and Policy, 44(3), 292-300. doi:http://dx.doi.org/10.1016/j. eap.2014.05.007 Barro, R. J. (1991). Government spending in a simple model of endogenous growth: National Bureau of Economic Research Cambridge, Mass., USA. Barro, R. J., & Lee, J. W. (2013). A new data set of educational attainment in the world, 1950– 2010. Journal of development economics, 104, 184-198. https://barrolee.github.io/ BarroLeeDataSet/Data.html Bende-Nabende, A., & Slater, J. (2003). Private capital formation: Short-and long-run crowding-in (out) effects in ASEAN, 1971-99. Economics Bulletin, 3(28), 1-16. Blöndal, J. R., Hawkesworth, I., & Choi, H. D. (2009). Budgeting in Indonesia. OECD Journal on Budgeting, 9(2), 1-31. Broock, W. A., Scheinkman, J. A., Dechert, W. D., & LeBaron, B. (1996). A test for independence based on the correlation dimension. Econometric reviews, 15(3), 197-235. Burhanudin, M. D. A., Muda, R., Nathan, S. B. S., & Arshad, R. (2017). Real effects of government debt on sustainable economic growth in Malaysia. Journal of International Studies, 10(3). Butts, H. C. (2009). Short term external debt and economic growth—Granger causality: evidence from Latin America and the Caribbean. The Review of Black political economy, 36(2), 93-111. Checherita-Westphal, C., & Rother, P. (2012). The impact of high government debt on economic growth and its channels: An empirical investigation for the euro area. European economic review, 56(7), 1392-1405. doi:http://dx.doi.org/10.1016/j.euroecorev.2012.06.007. Cholifihani, M. (2008). A cointegration analysis of public debt service and GDP in Indonesia. IBT Journal of Business Studies (JBS), 2(2). Chudik, A., Mohaddes, K., Pesaran, M. H., & Raissi, M. (2017). Is there a debt-threshold effect on output growth?. Review of Economics and Statistics, 99(1), 135-150. Clements, B. J., Bhattacharya, R., & Nguyen, T. Q. (2003). External debt, public investment, and growth in low-income countries: International Monetary Fund. Cordella, T., Ricci, L., & Ruiz-Arranz, M. (2005). Debt overhang or debt irrelevance? revisiting the debt-growth link. de Rugy, V., & Salmon, J. (2020). Debt and growth: A decade of studies (No. 10445). Diamond, P. A. (1965). National debt in a neoclassical growth model. The American Economic Review, 55(5), 1126-1150. 64 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 Elmendorf, D. W., & Gregory Mankiw, N. (1999) Chapter 25 Government debt. Vol. 1. Handbook of Macroeconomics (pp. 1615-1669). Engle, R. F., & Granger, C. W. (1987). Co-integration and error correction: representation, estimation, and testing. Econometrica: Journal of the Econometric Society, 251-276. Ferreira, M. C. (2009). Public debt and economic growth: a Granger causality panel data approach. Ferreira, C. (2016). Debt and economic growth in the European Union: a panel Granger causality approach. International Advances in Economic Research, 22(2), 131-149. Greiner, A. (2007). An endogenous growth model with public capital and sustainable government debt. Japanese Economic Review, 58(3), 345-361. Granger, C. (1969). Investigating causal relations by econometric models and cross-spectral methods, Econometrica 37, 424-438. Guimaraes, R. F., & Unteroberdoerster, O. (2006). What’s Driving Private Investment in Malaysia?: Aggregate Trends and Firm-Level Evidence. Handra, H., & Kurniawan, B. (2020). Long-run Relationship Between Government Debt and Growth: the Case of Indonesia. International Journal of Economics and Financial Issues, 10(1), 96. Hofman, B., & Reisen, H. (1990). Debt overhang, liquidity constraints and adjustment incentives. Working Paper No.32 https://doi.org/10.1787/356147450736 Jacobs, J., Ogawa, K., Sterken, E., & Tokutsu, I. (2020). Public debt, economic growth and the real interest rate: A panel VAR Approach to EU and OECD countries. Applied Economics, 52(12), 1377-1394. Jayaraman, T. K., & Lau, E. (2009). Does external debt lead to economic growth in Pacific island countries. Journal of Policy Modeling, 31(2), 272-288. Johansen, S. (1991). Estimation and hypothesis testing of cointegration vectors in Gaussian vector autoregressive models. Econometrica: Journal of the Econometric Society, 1551-1580. Kawai, M., & Morgan, P. J. (2013). Long-term issues for fiscal sustainability in emerging Asia (No. 432). ADBI Working Paper. Kempa, B., & Khan, N. S. (2016). Government debt and economic growth in the G7 countries: are there any causal linkages? Applied Economics Letters, 23(6), 440-443. Krugman, P. (1988). Financing vs. forgiving a debt overhang. Journal of development economics, 29(3), 253-268. Law, S. H., Ng, C. H., Kutan, A. M., & Law, Z. K. (2021). Public debt and economic growth in developing countries: Nonlinearity and threshold analysis. Economic Modelling, 98, 26-40. Lof, M., & Malinen, T. (2014). Does sovereign debt weaken economic growth? A panel VAR analysis. Economics Letters, 122(3), 403-407. Makin, T. (2005). Fiscal risk in ASEAN. Agenda: A Journal of Policy Analysis and Reform, 22765 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries 238. Mankiw, N. G., Romer, D., & Weil, D. N. (1992). A contribution to the empirics of economic growth. The Quarterly Journal of Economics, 107(2), 407-437. Mariotti, M. (2002). An examination of the impact of economic policy on long-run economic growth: An application of a VECM structure to a middle-income context. South African Journal of Economics, 70(4), 688-724. Modigliani, F. (1961). Long-run implications of alternative fiscal policies and the burden of the national debt. The Economic Journal, 71(284), 730-755. Mohd Daud, N., Halim Ahmad, A., & Azman-Sainic, W. N. W. (2013). Does external debt contribute to malaysia economic growth? Ekonomska istraživanja, 26(2), 51-68. Panizza, U., & Presbitero, A. F. (2013). Public debt and economic growth in advanced economies: A survey. Swiss Journal of Economics and Statistics, 149(2), 175-204. Panizza, U., & Presbitero, A. F. (2014). Public debt and economic growth: is there a causal effect?. Journal of Macroeconomics, 41, 21-41. Pattillo, C. A., Poirson, H., & Ricci, L. A. (2002). External debt and growth: International Monetary Fund. Pattillo, C. A., Poirson, H., & Ricci, L. A. (2004). What are the channels through which external debt affects growth? : International Monetary Fund. Pattillo, C., Poirson, H., & Ricci, L. A. (2011). External debt and growth. Review of Economics and Institutions, 2(3), 30. Pesaran, H. H., & Shin, Y. (1998). Generalized impulse response analysis in linear multivariate models. Economics Letters, 58(1), 17-29. Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bounds Testing Approaches to the Analysis of Level Relationships. Journal of applied econometrics, 16(3), 289-326. Puente-Ajovín, M., & Sanso-Navarro, M. (2015). Granger causality between debt and growth: Evidence from OECD countries. International Review of Economics & Finance, 35, 6677. Ramaswamy, R. and T. Slok, 1998, The Real effects of monetary policy in the European Union: What are the differences? IMF Staff Papers, 45, pp. 374–96. Reinhart, C., & Rogoff, K. (2010). Debt and growth revisited. MPRA Paper Sachs, J. D. (1989). Front matter,” Developing Country Debt and the World Economy Developing Country Debt and the World Economy (pp. -14-10): University of Chicago Press. Schclarek, A. (2004). Debt and economic growth in developing and industrial countries. Lund University Department of Economics Working Paper, 2005, 34. Sen, S., Kasibhatla, K. M., & Stewart, D. B. (2007). Debt overhang and economic growth-the Asian and the Latin American experiences. Economic Systems, 31(1), 3-11. Singh, T. (2012). Does public capital crowd-out or crowd-in private capital in India? Journal of Economic Policy Reform, 15(2), 109-133. 66 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 Stock, J. H., & Watson, M. W. (2001). Vector autoregressions. Journal of Economic perspectives, 15(4), 101-115. Swamy, V. (2015). The Dynamics of Government Debt and Economic Growth. Browser Download This Paper. Thao, P. T. P. (2018). Impacts of public debt on economic growth in six ASEAN countries. Ritsumeikan Annual Review of International Studies, 17, 63-88. Toda, H. Y., & Yamamoto, T. (1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of econometrics, 66(1-2), 225-250. Woo, J., & Kumar, M. S. (2010). Public debt and growth: International Monetary Fund. 67 Hajian, H. et al. Government Debt-Economic Nexus in ASEAN-4 Countries Appendix A Response of GDP per capita of ASEAN-4 countries to a Cholesky 1 SD innovation in government debt within two alternative models: VAR model ordering [ld, ly, lpinv, lhk] on the left column vs. [ly, ld, lpinv, lhk] on the right. 68 JDE (Journal of Developing Economies) Vol. 7 No.1 (2022): 45-69 Appendix B Summary statistics of the endogenous variables of the model. Table 1B: Descriptive Statistics - Growth Model Variables- Indonesia GDP Per Capita (Rupiah) Debt (%GDP) Privat Investment (%GDP) Human Capital (Years) Mean 21,267, 411 44.99 15.29 5.03 Median 20,351, 471 40.26 16.47 4.69 Maximum 33,570,451 95.22 22.91 7.5 Minimum 12,603,729 25 0.99 3.23 5,814,605 19.46 5.64 1.49 30 30 30 30 S.D Observations Table 2B: Descriptive Statistics - Growth Model Variables- Malaysia GDP Per Capita (Rupiah) Debt (%GDP) Privat Investment (%GDP) Human Capital (Years) Mean 18,038.5 56.16 11.750 7.982 Median 18,264.8 48.57 10.569 8.035 Maximum 27,661.48 109 24.80 10.3 Minimum 9,713.85 32.3 4.98 5.052 S.D 5,320.75 21.77 5.20 1.43 30 30 30 30 Observations Table 3B: Descriptive Statistics - Growth Model Variables - The Philippines GDP Per Capita (Rupiah) Debt (%GDP) Mean 22,062.69 57.68 Median 22,391.46 Maximum Minimum S.D Observations Privat Investment (%GDP) Human Capital (Years) 16.010 7.32 58.8 15.979 7.5 33493.73 79.2 19.331 8.3 11944.47 41.9 12.754 6.21 6,438.6 10.06 1.54 0.67 30 30 30 30 Table 4B: Descriptive Statistics - Growth Model Variables - Thailand GDP Per Capita (Rupiah) Debt (%GDP) Privat Investment (%GDP) Human Capital (Years) Mean 88,643.59 36.78 19.71 5.14 Median 85,552.16 39.57 17.76 4.65 Maximum 134,938.9 57.16 32.36 7.65 Minimum 40,214.96 10.74 6.19 3.26 27,959.3 13.39 7.42 1.5 30 30 30 30 S.D Observations 69
https://openalex.org/W2133678043	https://www.oncotarget.com/article/1048/pdf/	English	null	CD44variant exon 9 plays an important role in colon cancer initiating cells	Oncotarget	2,013	cc-by	4,137	CD44variant exon 9 plays an important role in colon cancer initiating cells 1 First Department of Surgery, University of Fukui, Fukui, Japan. 2 Department of Biochemistry, Shimane University School of Medicine, Izumo, Japan, Japan. Correspondence to: Takanori GOI, email: tgoi@u-fukui.ac.jp Keywords: colon cancer, CD44 variant exon, stem cells, Cancer initiating cells Received: May 22, 2013 Accepted: June 4, 2013 Published: June 6, 2013 This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT: Cancer stem cells(cancer initiating cells) have become increasingly important in the treatment of malignant tumors. CD44 in particular has been identified as a marker for stem cells in colon cancer, which is a high-morbidity tumor. However, many details remain unknown, including identification of the relevant exon. The elucidation of these details could lead to the development of new therapies and improvements in prognosis. We report our findings on the importance of CD44 variant exon 9(v9) of stem cells in colon cancer. Using the anti-CD44 standard form(s) antibody, as well as antibodies for each of the CD44 variant exons, we studied colon cancer cell lines by examining stained images of stem cells in the crypt of normal colon mucosa. Using the anti-CD44v9 antibody that fits the normal colon mucosa stem cells, we screened cells using flow cytometry to examine colony formation, resistance to anticancer drugs, and tumor mass formation after subcutaneous implantation in mice. The stem cell–containing region in the crypt of normal colon mucosa was negative for anti-Ki67 antibody staining; only the anti-CD44 v9 antibody stain was expressed. As for colony formation, resistance to anticancer drugs, and tumor mass formation, cells positive both for anti-CD44s and anti-CD44v9 antibody stains was significantly more frequent than those positive for anti-CD44s antibody stain and negative for anti-CD44v9 antibody stain and those negative both for anti-CD44s and anti-CD44v9 antibody stains. CD44 variant exon 9 plays an important role in colon cancer stem cells Oncotarget, May, Vol.4, No 5 Oncotarget, May, Vol.4, No 5 www.impactjournals.com/oncotarget/ INTRODUCTION the tissue[4]. A cancer stem cell(cancer initiating cell), likewise, is considered capable of self-replication, self- differentiation, drug resistance, and immune evasion[5-8]. In 1994, Dick and colleagues first identified stem cells in malignant tumors[9]. Using a molecular marker for human acute myeloid leukemia cells, they identified the cell that induces leuemia. In solid cancers, Michael Clarke found breast cancer stem cells in 2004[10]. Peter Dirks discovered cancer stem cells in brain tumors; [11] and O’Brien and Ricci-Vitiani reported the discovery of colon cancer stem cells in 2007[12,13]. All these reports showed that cancer stem cells play an important role in the differentiation and growth of cancer. The incidence of colon cancer is high among all malignant tumors[1,2]. In many cases, colon cancer spreads hematogenously to the liver and lungs. The elucidation of this mechanism may advance the development of new therapies and improve survival rates. To date, molecular biological research has been conducted on metastasis of colon cancer to the liver, and several molecular target drugs have been used in clinical settings[1,3]. In general, a stem cell in a normal tissue can develop into many constituent cells of a tissue and can remain alive continuously by structuring or restructuring CD44, in particular, is a well-known marker Oncotarget 2013; 4: 785-791 www.impactjournals.com/oncotarget 785 for colon cancer stem cells. The gene, located on the short arm of chromosome 11, has a transmembrane structure, allowing at least 10 exons to be inserted into the extracellular domain near the transmembrane domain through an alternative splicing mechanism. A variant exon(v), which differs from the original exon, is inserted on the basis of tissue function[14]. It is involved in adhesion between cells and the extracellular matrix, cellular motion, and growth/invasion/metastasis of cancer cells[15,16]. However, the exon that plays the central role in cancer stem cells remains unknown. According to the recent report, hematogenous metastasis is likely to accompany the expression of the proteins inserted with CD44 variant exon 6 or CD44 variant exon 9(v9) in the primary foci of gastric and colon cancer cells[17-19]. According to a previous publication, breast cancer stem cells with CD44 protein expression are resistant to radiotherapy owing to an uncharacteristic rise of reactive oxygen levels in the cell[20]. INTRODUCTION A recent report has shown that the binding of CD44 variant exons (8, 9, and 10) to xCT proteins that transport cystine/glutamate on the cell membrane accelerates the formation of reduced glutathione and inhibits accumulation of reactive oxygen in cancer cells, thereby suppressing the activation of oxidative stress[21]. in tumors. In this study, we examined the stained images of stem cells from normal colon mucosa to explore the relationship between colon cancer stem cells and the CD44v9 isoform, which is considered important in this type of cancer. Cell separation via flow cytometry using anti- CD44s and anti-CD44v9 fluorescent antibodies Screened colon cancer cell lines and the drug 5-FU were placed in a chamber. ANNEXIN-V staining was conducted to examine cell apoptosis and cell necrosis induction at various concentrations (Fig. 4A). In HT29 colon cancer cells, the non-staining rate of CD44s[+]v9[-] was 41.3% with 10 µg/mL, 17.5% with 30 µg/mL, and 14.7% with 50 µg/mL, whereas in CD44s[+]v9[+], the non-staining rate was 71.6% with 10 µg/mL and 48.7% with 50 µg/ mL(Fig. 4B Left). In HCT116 colon cancer cells, the non- staining rate of CD44s[+]v9[-] was 64.6% with 10 µg/ mL, and 22.5% with 50 µg/mL. The non-staining rate of CD44s[+]v9[+] was 87.5% with 10µg/mL, 81.6% with 30 µg/mL, and 67.8% with 50 µg/mL(Fig. 4B Right). In both colon cancer cells, the non-staining rate with ANEXIN-V in CD44s[+]v9[+] was significantly higher than that in CD44s[+]v9[-]. After the reaction of anti-CD44s and anti-CD44v9 fluorescent antibodies with the colon cancer cell lines, fluorescence-activated cell sorting scanning was used to examine expression state (Fig. 2). The cells were separated into those that were expressed with both anti- CD44s and anti-CD44v9 antibodies(CD44s[+]v9[+]), those that were positive for anti-CD44s antibody but negative for anti-CD44v9 antibody(CD44s[+]v9[-]), and those negative both for anti-CD44s and anti-CD44v9 antibodies(CD44s[-]v9[-]). RESULTS Immunochemical staining of normal colon mucosa with anti-CD44s antibody and antibodies for each variant exon Immunohistochemical staining was conducted using anti-Ki67 antibody, anti-CD44s antibody, and antibodies for each of the CD44 variant exons to examine the stem cells of normal colon mucosa(Fig. 1). Expression by anti- Ki67 antibody was found in transit-amplifying cells and differentiating cells in the normal mucosa, but expression was absent in stem cells in the crypt. Furthermore, in the anti-Ki67 antibody stain–negative region (stem cell), expression was found with anti-CD44s and anti-CD44 v3, v4, v9 and v10 antibodies. Among the anti-CD44 antibody group, only anti-CD44v9 induced expression in the anti-K67 antibody stain–negative region (stem cell). As discussed above, the CD44 molecule family is projected to be relevant at various levels of malignancy Figure 1: Immunohistochemical staining using anti-CD44 antibodies. The expression was found with anti-CD44standard form(s) and anti-CD44 variant exon (3, 4, 5, 6, 7, 8, 9, and 10)antibodies. Among the anti-CD44 antibodies, only anti-CD44 variant exon 9(v9) induced expression in the anti-Ki67 antibody stain–negative region(stem cell). Fi 1 I hi h i l i i i i CD44 ib di Figure 1: Immunohistochemical staining using anti-CD44 antibodies. The expression was found with anti-CD44standard form(s) and anti-CD44 variant exon (3, 4, 5, 6, 7, 8, 9, and 10)antibodies. Among the anti-CD44 antibodies, only anti-CD44 variant exon 9(v9) induced expression in the anti-Ki67 antibody stain–negative region(stem cell). www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 786 Colony formation in colon cancer cells expressed with anti-CD44v9 antibody Formation of tumor mass after subcutaneous implantation of anti-CD44v9 antibody–positive colon cancer cells in mice No tumor mass was formed with CD44s[-]v9[-] or CD44s[+]v9[-] in HT29 colon cancer cells, although a 13×8×7-mm tumor mass was formed with CD44s[+]v9[+] (Fig. 5A). Also, in HCT116 colon cancer cells, no tumor mass was formed with CD44s[-]v9[-] or CD44s[+]v9[-], although a tumor mass was formed with CD44s[+]v9[+]. In both types of colon cancer cells, tumor masses formed only with CD44s[+]v9[+] (Fig. 5A,B). After separation (see Fig 2), the cells were incubated in a 96-well plate (1 cell per well) for the investigation of colony formation. Figure showed a microscopic view of cells after five days(Fig. 3A). In HCT116 colon cancer cells, colony formation was absent in CD44s[-]v9[-], whereas only 2 colonies were formed by CD44s[+]v9[-]. Meanwhile, 14 colonies were formed by CD44s[+]v9[+] (Fig. 3B). In HT29 colon cancer cells, colony formation was absent in CD44s[-]v9[-] and CD44s[+]v9[-]. In anti- CD44s[+]v9[+], 145 colonies were formed. In both types of cancer cells, CD44s[+]v9[+] formed significantly more colonies than CD44s[-]v9[-] or CD44s[+]v9[-] did.(Fig. 3B) DISCUSSION The concept of heterogeneity of cancer cells has recently undergone a major change. At the 2006 American Association for Cancer Research, cancer stem cells were Resistance to 5-FU in anti-CD44v9 antibody–positive coloncancer cells Figure 2: Flow cytometry of colon cancer cells using fluorescence anti-CD44s Ab and anti-CD44v9 Ab. Flow cytometry was performed using colon cancer cell lines, HT29, HCT116 after they were stained for surface markers: CD44s, CD44v9. Vertical axis shows expression of CD44s, and horizontal axis showed expression of CD44v9. Three different sub populations were separated by fluorescence-activated cell sorting (FACS) as indicated: CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-]. Figure 2: Flow cytometry of colon cancer cells using fluorescence anti-CD44s Ab and anti-CD44v9 Ab. Flow cytometry was performed using colon cancer cell lines, HT29, HCT116 after they were stained for surface markers: CD44s, CD44v9. Vertical axis shows expression of CD44s, and horizontal axis showed expression of CD44v9. Three different sub populations were separated by fluorescence-activated cell sorting (FACS) as indicated: CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-]. www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 787 defined as cells that “exist in a tumor, with capability of replicating themselves and generating different types of cancer cells that constitute the tumor tissue[25].” Based on this concept, a hierarchy model was proposed, in which highly oncogenic cancer cells and non-oncogenic cancer cells derived from oncogenic cells might coexist in a tumor tissue[26]. According to this model, the diversity and non-uniformity of cancer cells must be taken into account in planning therapy for full recovery. The origin of this definition of cancer stem cells was the identification of blood stem cells by Till, MuCulloch, Becker, et al. in the 1960s, suggesting the involvement of stem cell-like cells in cancer growth[27]. In 2007, reports were published on stem cells for colorectal cancer, and CD44 was identified as a cancer stem cell marker. CD44, a transmembrane protein, may display extramembrane insertion of variant exons[28]. CD44 molecules with variant exon insertion have been found in blood cells, various normal tissues, and epithelial cancer cells. In the case of malignant tumors in particular, some variant exons are closely related to metastasis and malignancy, which indicates their importance[14-19]. However, no details about the expression of variant exons in CD44 or cancer stem cells in colon cancer are available. defined as cells that “exist in a tumor, with capability of replicating themselves and generating different types of cancer cells that constitute the tumor tissue[25].” Based on this concept, a hierarchy model was proposed, in which highly oncogenic cancer cells and non-oncogenic cancer cells derived from oncogenic cells might coexist in a tumor tissue[26]. According to this model, the diversity and non-uniformity of cancer cells must be taken into account in planning therapy for full recovery. i the normal colon mucosa are found. Next, we examined the expression of individual CD44 exons. According to the results, the expression in the Ki-67 stain–negative region was observed only with the vriant exon 9. The immunohistochemical staining of the colon cancer foci revealed that strong expression of the variant exon 9 was likely to cause hematogenous metastasis, and therefore, this exon was considered important[17]. i By definition, cancer stem cells must have self- propagating or self-differentiating capability, as well as the capability to withstand the human immune system for survival. www.impactjournals.com/oncotarget A cancer stem cell must also have resistance mechanisms against anticancer drugs and the capability to form tumor masses in immunodeficient mice(5-8). In this context, we compared CD44-positive colon cancer cells that did not contain the vriant exon 9 and those that did. CD44-positive colon cancer cells containing the variant exon 9 had significantly stronger colony–forming and tumor mass–forming abilities and resistance against anticancer drugs. Thus, in CD44 molecules in colon stem cancer cells, the variant exon 9 was considered an exceptionally important factor. Future studies on CD44 molecules and variant exon 9 are warranted for elucidating basic concepts of cancer propagation and drug resistance. Accordingly, we examined the expression of variant exons in CD44. First, assuming that stem cells in normal colon mucosa and those of colon cancer use the same exon, we conducted Ki-67 staining to identify the dormant cell at the bottom of the crypt, in which the stem cells of Figure 4: Colon cancer cells stained for ANNEXIN-V using by fluorescence microscope. A.Photograph of cells stained for ANNEXIN-V. Colon cancer cells(1x104) sorted to CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-] incubated in the presence of each concentration of 5-FU for 3 days. After incubating with annexin-V-Alexa568, we detected red cells under a fluorescent microscope. B. Result of living ratio against anticancer drug: 5-FU. After 72 hours of exposure, we detected non-red cells under a fluorescent microscope. Results are presented as the mean±SD (n=4). Left HT29 cells. Right HCT116 cells. Figure 3: Colony formation in colon cancer cells expressed CD44s[+]v9[+], CD44s[+]v9[-], and CD44s[-] v9[-]. A. Microscopic view of cells after 5 days. Colon cancer cells sorted as CD44s[+]v9[+], CD44s[+]v9[-], and CD44s[-] v9[-], separated 96-well plate as 1 cell per well using dilution limited technique. B. The number of colony formations. Results are presented as the mean±SD (n=4). P values were determined using the by Student’s t-test. Values of < 0.05 were considered statistically significant. p<0.05.p<0.01. Figure 4: Colon cancer cells stained for ANNEXIN-V l Figure 4: Colon cancer cells stained for ANNEXIN-V using by fluorescence microscope. A.Photograph of cells stained for ANNEXIN-V. Colon cancer cells(1x104) sorted to CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-] incubated in the presence of each concentration of 5-FU for 3 days. After incubating with annexin-V-Alexa568, we detected red cells under a fluorescent microscope. B. Result of living ratio against anticancer drug: 5-FU. After 72 hours of exposure, we detected non-red cells under a fluorescent microscope. www.impactjournals.com/oncotarget Results are presented as the mean±SD (n=4). Left HT29 cells. Right HCT116 cells. Figure 3: Colony formation in colon cancer cells expressed CD44s[+]v9[+], CD44s[+]v9[-], and CD44s[-] v9[-]. A. Microscopic view of cells after 5 days. Colon cancer cells sorted as CD44s[+]v9[+], CD44s[+]v9[-], and CD44s[-] v9[-], separated 96-well plate as 1 cell per well using dilution limited technique. B. The number of colony formations. Results are presented as the mean±SD (n=4). P values were determined using the by Student’s t-test. Values of < 0.05 were considered statistically significant. p<0.05.*p<0.01. www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 788 serotec), 6(R&D systems, MN, USA), 7/8(AbD serotec), 10(AbD serotec), anti-human CD44 variant exon 9(prepared in our department), and AE1/AE3(Santa Cruz Biotechnology, CA, USA), Ki-67(Invitorogen, CA,USA). serotec), 6(R&D systems, MN, USA), 7/8(AbD serotec), 10(AbD serotec), anti-human CD44 variant exon 9(prepared in our department), and AE1/AE3(Santa Cruz Biotechnology, CA, USA), Ki-67(Invitorogen, CA,USA). Cell culture Normal colon tissues were obtained from a patient with primary colon cancer at the First Department of Surgery, University of Fukui, Japan, in 2005. The human coloncancer cell lines:HT29 and HCT116(Public Health England Culture Collection, UK) were cultured at 37C in 5% CO2 in RPMI 1640 medium(Sigma Chemical Co. MO, USA) containing 10% fetal bovine serum.[22] Surgical specimens of the adjacent normal colon tissues prepared from formalin-fixed, paraffin-embedded tissues were analyzed for protein expression by the ChemMate method using the EnVision system(DAKO, Danmark). MATERIALS AND METHODS serotec), 6(R&D systems, MN, USA), 7/8(AbD serotec), 10(AbD serotec), anti-human CD44 variant exon 9(prepared in our department), and AE1/AE3(Santa Cruz Biotechnology, CA, USA), Ki-67(Invitorogen, CA,USA). Antibody(Ab) Cells were washed twice with phosphate-buffered saline. Dissociated cells were stained with fluorescein isothiocyanate (FITC)-conjugated anti-CD44v9 antibody (prepared in our department)[19] and Allophycocyanin The following antibodies were used: anti-human CD44 standard form(s) (Ab cam, UK), variant exon 3(Neuromics, MN, USA), 4(AbD serotec, UK), 5(AbD igure 5: Evaluation of the tumorigenic potential of isolated CD44s and CD44v9 cells. (A) Photograph of a mouse taken 4 eeks after HT29 cells transplantation. Each 1.0 × 103 cells which sorted as CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-] were injected NOD-SCID mouse subcutaneously. (B) Tumor volume of isolated CD44s and CD44v9 cells. The tumor size was measured with calipers, nd calculated with the formula:(L × W2)/2(L; length of the tumor, W; width of the tumor). Results are presented as the mean±SD (n=4). P alues were determined using the by Student’s t-test. Values of < 0.05 were considered statistically significant.p<0.01 Figure 5: Evaluation of the tumorigenic potential of isolated CD44s and CD44v9 cells. (A) Photograph of a mouse taken 4 weeks after HT29 cells transplantation. Each 1.0 × 103 cells which sorted as CD44s[+]v9[+], CD44s[+]v9[-], CD44s[-]v9[-] were injected to NOD-SCID mouse subcutaneously. (B) Tumor volume of isolated CD44s and CD44v9 cells. The tumor size was measured with calipers, and calculated with the formula:(L × W2)/2(L; length of the tumor, W; width of the tumor). Results are presented as the mean±SD (n=4). P values were determined using the by Student’s t-test. Values of < 0.05 were considered statistically significant.*p<0.01 Tumorigenicity assays in nude mice 9. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, Caceres-Cortes J, , Caceres-Cortes J, Minden M, Paterson B, Caligiuri MA, Dick JE. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994; 367: 645-648. Six-week-old female NOD-SCID mice (Charles river, Japan) were subcutaneously injected in the right armpit region with 1.0 × 103 cells in 0.1 mL of matrix gel(BD Biosciences, NJ, USA ). Three groups of mice were tested. Group 1 was injected with colon cancer cells positive both for anti-CD44s and anti-CD44v9 antibody stains(CD44s[+]v9[+]). Group 2 was injected with colon cancer cells positive for anti-CD44s antibody stain and negative for anti-CD44v9 antibody stain(CD44s[+]v9[-]). Group 3 was injected with colon cancer cells negative both for anti-CD44s and anti-CD44v9 antibody stains(CD44s[-] v9[-]). The tumor size was measured every 3 days with calipers. The tumor volume was calculated with the formula: (L × W2)/2, where L is the length and W is the width of the tumor[24]. 10. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003; 100: 3983- 3988. 11. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman RM, Cusimano MD, Dirks PB. Identification of human brain tumour initiating cells. Nature. 2004; 432: 396-401. 12. O’Brien CA, Pollett A, Gallinger S, Dick JE. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature. 2007; 445: 106-110. 13. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, De Maria R. Identification and expansion of human colon-cancer-initiating cells.Nature. 2007; 445: 111- 115. www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 789 Colony Formation Assay Cells were plated at a density of 1 cell per well in 96 well plates (BD Falcon, USA). We examined the colonies under a microscope every 24 hours. 2. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, and Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. 2010; 127: 2893- 2917. REFERENCES [APC]-conjugated CD44s antibody (BD Pharmingen, NJ, USA) and incubated for 30 minutes at 4°C. After washing with PBS, the cells were sorted by FACScan (Becton- Dickinson, CA, USA) 1. Watanabe T, Itabashi M, Shimada Y, Tanaka S, Ito Y, Ajioka Y, Hamaguchi T, Hyodo I, Igarashi M, Ishida H, Ishiguro M, Kanemitsu Y, Kokudo N, Muro K, Ochiai A, Oguchi M, et al. Japanese Society for Cancer of the Colon and Rectum: Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol. 2012;17: 1-29. Growth inhibitory assay 3. NCCN Clinical Practice Guidelines in Oncology. Antiemesis.V.2.2010. Available from: URL: http://www. nccn.org/professionals/physician-gls/PDF/antiemesis. The cells were seeded in 96 well plates at 1 × 104 cells/well and divided into control and treatment groups [5-Fluorouracl(5-FU) (100 µg/mM)] (Sigma Chemical Co.). After 72 hours of exposure, we detected non-red cells under a fluorescent microscope[23]. 4. Ricci-Vitiani L, Fabrizi E, Palio E, De Maria R. Colon cancer stem cells. J Mol Med. 2009; 87: 1097-1104. 5. Alison MR, Islam S, Wright NA. Stem cells in cancer: instigators and propagators?. J Cell Sci. 2010; 123: 2357- 2368. Cell viability 6. Dean M, Fojo T, Bates S.Tumour stem cells and drug resistance. Nature Reviews. 2005; 5: 275-284. Cell necrosis and apoptosis were detected by cytometry using Annexin-V-FLUOS staining Kit (Roche, Germany). Briefly, cells were incubated with annexin-V- Alexa568 for 15 minutes. After cells were washed thrice in PBS, we detected red and non-red cells under a fluorescent microscope. 7. Ailles LE, Weissman IL.Cancer stem cells in solid tumors. Curr Opin Biotechnol. 2007; 18: 460-466. 8. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB, Dewhirst MW, Bigner DD, Rich JN. Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature. 2006; 444: 756-760. Statistical analyses Differences between two groups were analyzed by Student’s t-test. Values of P<0.05 were considered to indicate statistically significant results. 14. Stamenkovic I, Aruffo A, Amiot M, Seed B. The hematopoietic and epithelial forms of CD44 are distinct polypeptides with different adhesion potentials for hyaluronate-bearing cells. EMBO J. 1991; 10: 343-348. 15. Hofmann M, Rudy W, Zöller M, Tölg C, Ponta H, Herrlich 15. Hofmann M, Rudy W, Zöller M, Tölg C, Ponta H, Herrlich www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 790 27. Becker AJ, Mc culloch EA, Till JE. Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature. 1963; 197: 452-454. P, Günthert U. CD44 splice variants confer metastatic behavior in rats: homologous sequences are expressed in human tumor cell lines.Cancer Res. 1991; 51: 5292-5297. P, Günthert U. CD44 splice variants confer metastatic behavior in rats: homologous sequences are expressed in human tumor cell lines.Cancer Res. 1991; 51: 5292-5297. 16. Guo YJ, Liu G, Wang X, Jin D, Wu M, Ma J, Sy MS. Potential use of soluble CD44 in serum as indicator of tumor burden and metastasis in patients with gastric or colon cancer. Cancer Res 1994; 54: 422-426. 28. Zöller M. CD44: can a cancer-initiating cell profit from an abundantly expressed molecule?” Nature Reviews Cancer. 2011; 11, 254–267. 17. Goi T, Yamaguchi A, Takeuchi K, Nakagawara G, Yamashiro S, Furukawa K, Urano T, Shiku H. CD44 with variant exons 8-10 in colorectal tumors:expression analysis by a variant exon 9-specific monoclonal antibody. Int J Oncol. 1996; 8: 657-662. 18. Yamaguchi A, Goi T, Yu J, Hirono Y, Ishida M, Iida A, Kimura T, Takeuchi K, Katayama K, Hirose K. Expression of CD44v6 in advanced gastric cancer and its relationship to hematogenous metastasis and long-term prognosis. J Surg Oncol. 2002; 79: 230-235. 19. Jang BI, Li Y, Graham DY, Cen P. The role of CD44 in the pathogenesis, diagnosis, and therapy of gastric cancer. Gut Liver. 2011; 3: 397-405. 20. Diehn M, Cho RW, Lobo NA, Kalisky T, Dorie MJ, Kulp AN, Qian D, Lam JS, Ailles LE, Wong M, Joshua B, Kaplan MJ, Wapnir I, Dirbas FM, Somlo G, Garberoglio C, et al. Association of reactive oxygen species levels and radioresistance in cancer stem cells. Nature. 2009; 458: 780-783. 21. Ishimoto T, Nagano O, Yae T, Tamada M, Motohara T, Oshima H, Oshima M, Ikeda T, Asaba R, Yagi H, Masuko T, Shimizu T, Ishikawa T, Kai K, Takahashi E, Imamura Y, et al. CD44 variant regulates redox status in cancer cells by stabilizing the xCT subunit of system xc(-) and thereby promotes tumor growth.Cancer Cell. 2011; 19: 387-400. 22. Goi T, Yamaguchi A, Nakagawara G, Urano T, Shiku H, Furukawa K. Reduced expression of deleted colorectal carcinoma (DCC) protein in established colon cancers. Br J Cancer. 1998; 77: 466-471. 23. www.impactjournals.com/oncotarget Aso K, Goi T, Nakazawa T, Kimura Y, Hirono Y, Katayama K, Yamaguchi A. The expression of integrins is decreased in colon cancer cells treated with polysaccharide K. Int J Oncol. 2013; 42: 1175-1180. 24. Goi T, Fujioka M, Satoh Y, Tabata S, Koneri K, Nagano H, Hirono Y, Katayama K, Hirose K, Yamaguchi A. Angiogenesis and tumor proliferation/metastasis of human colorectal cancer cell line SW620 transfected with endocrine glands-derived-vascular endothelial growth factor, as a new angiogenic factor.Cancer Res. 2004 ; 64: 1906-1910. 25. ClarkMF, Dick JE, Dirks PB, EavesCJ, Jamieson CHM, Jones DL, Visvader J, Weissman IL, Wahl GM. Cancer stem cells-Perspectives on Current Statusand Future Directions:AACR Workshop on cancer stem cells. Cancer Res. 2006; 66: 9339-9344. 26. Dick JE. Stem cell concepts renew cancer research. Blood. 2008; 112: 4793-4807. www.impactjournals.com/oncotarget Oncotarget 2013; 4: 785-791 791
https://openalex.org/W2903998691	https://ecologicalprocesses.springeropen.com/track/pdf/10.1186/s13717-018-0151-7	English	null	Characterizing selected soil attributes of different land-use management to assess reforestation benefits of deforested riparian buffers	Ecological processes	2,018	cc-by	10,830	* Correspondence: velurasiah@outlook.com; s.florentine@federation.edu.au 118 Whispering Ridge Drive, Guelph, Ontario N1G 3Z4, Canada 2Centre for Environmental Management, School of Health and Life Sciences, Federation University Australia, PO Box 663, Mount Helen, VIC 3350, Australia RESEARCH Open Access © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Abstract Introduction: The information available on the sensitivity of soil biotic and abiotic attributes, which can be used to track the impact of reforestation in riparian buffers, is often insufficient to refine management practices and convince stakeholders of the benefits of reforestation. Methods: In this study, conducted in Victoria, Australia, the changes in soil biotic and abiotic attributes, organic carbon (OC), mineral nitrogen (MN), total dissolved solutes (TDS) and pH were characterised to assess the impact of land-use change from bare riparian (BR) to reforested riparian (RR). Additionally, the benefits of revegetating a deforested creek bank with regard to salinity abatement and C-sequestration potentials were assessed. Results: The TDS depletion in the RR strips varied spatiotemporally from 65 to 169 mg/L, the net OC deposition from 16 to 19 g C/kg soil and MN deposition from 1.2 to 2.1 g N/kg soil, respectively. Additionally, the net changes in pH from alkaline to near neutral condition varied by 0.4 to 1.0 pH units. Approximately 30% to 60% of the net OC depletion after deforestation was redeposited under RR over 3 to 6 years. The TDS depletion after land-use changed from BR to RR ranged from 15 to 32% over 3 to 6 years. Conclusion: The soil attributes OC, MN and TDS characteristics under different land-use practices varied spatiotemporally. This information may be useful to convince stakeholders to undertake reforestation of creek banks for salinity abatement, and that change in land-use has the potential to increase C sequestration at a farm scale. Keywords: Creek bank riparian, Deforestation, Reforestation, Soil biotic and abiotic attributes, Site specificity, Temporal changes, Carbon sequestration, Soil salinity Velu Rasiah1* and Singarayer K. Florentine2* Velu Rasiah1* and Singarayer K. Florentine2* (2018) 7:39 Rasiah and Florentine Ecological Processes https://doi.org/10.1186/s13717-018-0151-7 Introduction Before these transects were cleared for cropping in the late nineteenth century, the land cover was native grass- land, with native forest species (Acacia melanoxylon, Eucalyptus camaldulensis, Leptospermum lanigerum, Allocasuarina verticillata and Allocasuarina littoralis) along the riparian strip. y ) Changes in soil functioning depend partially on the dy- namics of soil biotic and abiotic indicators, such as soil organic carbon (OC) (Tran et al. 2015; Rasiah and Kay 1998; Agren and Bosatta 1996), mineral nitrogen (MN) (Rasiah and Kay 1998), total dissolved solutes (TDS) and pH (Agren and Bosatta 1996). These functioning dynam- ics vary spatiotemporally (Rasiah and Kay 1999). Quantifying the dynamics of these indicators requires an appropriate baseline, which is usually obtained from un- disturbed native vegetation in the vicinity of the de- graded riparian strip. However, these are usually unavailable in landscapes used for large-scale commer- cial farming. Under such circumstances, the degraded ri- parian soil conditions may instead serve as the baseline against which to measure improvements in condition (Rasiah et al. 2015). The objectives of this study are to (i) characterise selected soil biotic and abiotic attri- butes, OC, MN, total dissolved solutes (TDS) and pH of soils that have been subjected to three contrasting land-use practices; (ii) assess the impact of reforestation of deforested creek-bank on salinity abatement and C-sequestration potentials; and (iii) explore the poten- tial to convince stakeholders to manage riparian buffers appropriately in-order to sustain agroforestry on large farms in Victoria, Australia. Four transects, ranged in length from 150 to 200 m, and 50 to 100 m in width, and the separation distance between them ranged from 100 to 500 m (Fig. 1, Table 1). During design stage, two out of the four transects were selected as replicates for the 3-year and the remaining two for the 6-year reforestation land-use treatments. The cropland (CP) plot was on the upper aspect of a replicate/transect, and the corresponding bare riparian (BR) and the reforested riparian (RR) plots at the lower aspect, just above the bank of a seasonal creek. Before the study commenced the riparian strip was bare, apart from seasonal weedy vegetation. The RR plots were lo- cated 90 m to 165 m from the corresponding BR plots (Fig. 1, Table 1). The CP, BR and RR plots are considered to represent the different land-use and management practices that were investigated in this study. Introduction thereby retarding solute transport to the soil surface from deeper within the profile (Clarke et al. 2002), pro- viding shade for livestock (Lindenmayer et al. 2008), a potential for the return of native fauna and flora (Jose 2009; Lindenmayer et al. 2008; Hobbs and Norton 1996) and enhancing farm aesthetic values (Petursdottir et al. 2012). Furthermore, perennial reforested stands may serve as windbreaks, and sequester atmospheric CO2 in soil profiles (Jose 2009; Silver et al. 2004; Shepherd and Montagnini 2001). Recently, the farm-level benefits of re- forestation along stream banks has been receiving in- creased attention, due in part to the potential for reduced sediment, nutrient and pesticide transport to streams (Kaleb et al. 2013; Wortley et al. 2013; Nair and Graetz Riparian buffer reforestation benefits include, but are not limited to, improvements in soil biotic and abiotic condition (Rasiah et al. 2015; An et al. 2012; Luo et al. 2010; Lal 2004; Rasiah et al. 2004; Silver et al. 2004), the quality of surface and subsurface water discharged into streams (Collins et al. 2012; Nair and Graetz 2004; Fennessy and Cronk 1997; Lowrance et al. 1984), retard- ing the progression of salinity via improvement in deep drainage, reduction in evaporation from the soil surface, Page 2 of 12 Page 2 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Rasiah and Florentine Ecological Processes 2004). How these are linked to soil biotic and abiotic attri- butes or whether they can be used to discriminate and characterise changes in land-use practices spatiotempo- rally is not well documented. Such information is essential to convince farmers, policy-decision makers and model- lers to upscale their perspective from point measurements at a farm level, to landscape and catchment scales (Chen et al. 1999; Rasiah and Kay 1999). on the age of the parent material. The profiles are often shallow (< 2 m), base-rich clay loams over a strongly sodic heavy clay subsoil. The soils are generally of a low agricultural potential, characterised by high sodicity, erodibility, poor structure, low permeability and low to moderate fertility. Geologically, the soil profile characteristics are similar across the four transects and variability is likely associ- ated with differences in micro-ecosystem functioning. Introduction Native for- est tree seedlings were planted in two of the four RR plots in 2006, and in two remaining plots in 2009. Dur- ing soil sampling in February 2012, the seedlings were 6 and 3 years old, respectively. Methods Study sites l l The 6-year-old RR plots are abbreviated as 6 yr-RR-1 and 6 yr-RR-2, the CP as 6 yr-CP-1 and 6 yr-CP-2 and the BR as 6 yr-BR-1 and 6 yr-BR-2. The 3-year-old RR plots are abbreviated as 3 yr-RR-1 and 3 yr-RR-2, CP as 3 yr-CP-1 and 3 yr-CP-2 and the BR as 3 yr-BR-1 and 3 yr-BR-2 (Table 1). y Soil samples were collected along four transects in the Corangamite Catchment in south-west Victoria, Australia (Fig. 1). The catchment is characterised by a temperate climate, having cool wet winters and warm to hot dry summers (Joyce 2003). The mean annual rainfall at the farm is ~ 600 mm, and land-use is mixed farming, including wheat and oil-seed crops rotated with pasture. The crops are generally sown in autumn (May–June) and harvested in late spring (September– October). No-till and low-till farming practices have re- placed traditional ploughing over the past few decades, along with precision agriculture incorporating spatially variable fertiliser and herbicide applications, depending on specific crop requirements. Soil sampling and laboratory analysis Soil sampling was undertaken in February 2012. In each of the CP plots, a 25-m square was marked out. These were then divided into 25 grid nodes of 5 m × 5 m and each was allocated a number (1 to 25). To select the nodes for soil coring, five numbered cards were ran- domly drawn from a bag. At each of the selected grid nodes, five deep soil cores were extracted to 0.4 m depth using a 0.1-m-diameter hydraulically driven auger. Due to the impracticability of setting up 25-m square plots in the BR and RR plots, five soil cores were extracted at 5-m intervals along a 25 m strip within each plot. The soil types at the farm are Sodosols that developed on weathered Quaternary volcanic rocks (Robinson et al. 2003; Gray et al. 2002; Nicholson 2002). They are tex- ture contrast soils, typically varying in depth, dependent Rasiah and Florentine Ecological Processes (2018) 7:39 Page 3 of 12 (2018) 7:39 Fig. 1 At the top left hand corner is the study site, the Corangamite Catchment, in Victoria, Australia. At the top right hand corner is the detail of catchment authority area (CMA) and bottom picture showing the sampling locations in the farm. The land-use practice treatments at the 6-year- old reforested (RR) transect-1 is abbreviated as 6 yr-rr-1, the corresponding cropland (CP) as 6 yr-c-1, the bare riparian (BR) as 6 yr-br-1. Along the second transect the 6-year-old RR as 6 yr-rr-2, 3 yr-c-2 and 6 yr-br-2, respectively. Similarly, the 3-year-old RR as 3 yr-rr-1, 3 yr-c-1, 3 yr-br-1 and 3 yr-rr-2, 3 yr-c-2 and 3 yr-br-2, respectively, along the other two transects. During design stage, two out the four transects were considered as replicates for the 3-year-old RR and the corresponding CP and RR land-uses, and the other transects as replicates for the 6-year-old RR and the corresponding CP and BR land-uses Fig. 1 At the top left hand corner is the study site, the Corangamite Catchment, in Victoria, Australia. At the top right hand corner is the detail of catchment authority area (CMA) and bottom picture showing the sampling locations in the farm. The land-use practice treatments at the 6-year- old reforested (RR) transect-1 is abbreviated as 6 yr-rr-1, the corresponding cropland (CP) as 6 yr-c-1, the bare riparian (BR) as 6 yr-br-1. Soil sampling and laboratory analysis Along the second transect the 6-year-old RR as 6 yr-rr-2, 3 yr-c-2 and 6 yr-br-2, respectively. Similarly, the 3-year-old RR as 3 yr-rr-1, 3 yr-c-1, 3 yr-br-1 and 3 yr-rr-2, 3 yr-c-2 and 3 yr-br-2, respectively, along the other two transects. During design stage, two out the four transects were considered as replicates for the 3-year-old RR and the corresponding CP and RR land-uses, and the other transects as replicates for the 6-year-old RR and the corresponding CP and BR land-uses Association of Testing Authorities (Australia)) accre- dited laboratory. Three out of the five cores from each location were used for the determination of soil organic carbon (OC), mineral nitrogen (TN), total dissolved solutes (TDS) and pH, while the other two cores were reserved for another study. Each one of the three cores was segmented at 0.1 m depth increments. Large clods were broken down to smaller pieces. The resultant material was air-dried in the laboratory and sieved through a 2 mm sieve. The sieved material was used for the determination of OC, MN, TDS and pH, using the procedures described by Rayment and Higgins (1992) at a NATA (National Data analysis To determine the statistical significance of the data, mean separation analysis and simple linear correlations were performed using the Statgraphics Centrion XVI (2010) package. Initially, we considered that two tran- sects out of the four as replicates for the 6-year-old reforested (RR) land-use treatment and the correspond- ing cropland (CP) and bare riparian (BR) land-uses. The Page 4 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Table 1 Land-use practices at four locations in the farm and their physical characteristics. During the experimental design stage, two out of the four transects were considered as replicates for the 3-year-old reforested riparian (RR) and the corresponding cropland (CP) and bare riparian (BR) land-uses for this time-period. Similarly, the other two transects for the 6-year-old RR, and the corresponding CP, and BR land-uses Table 1 Land-use practices at four locations in the farm and their physical characteristics. During the experimental design stage, two out of the four transects were considered as replicates for the 3-year-old reforested riparian (RR) and the corresponding cropland (CP) and bare riparian (BR) land-uses for this time-period. Data analysis OC or EC) showed that the two replicates belong- ing to a given RR time-frame were statistically not repli- cates; therefore, we consider the replicates as different sites in the farm. This is where spatial variability or site-specificity concept came into play in this paper. In Tables 3 and 5, we show the functional relationship be- tween the soil attributes by defining OC = f (MN or EC or pH), MN = f (EC or pH) and EC = f(pH), to determine whether the land use practice and/or site-specific vari- ability between replicates had any impact on the func- tional variability of the attributes. For the OC correlation, the 16 means from the 4 CP plots (Table 2) were correlated with the corresponding TN or EC or pH (Table 4), and similarly for TN, EC and pH. For these correlations, the depth increment was considered as se- lection variable. Correlations were also conducted to verify whether soil attributes (OC, MN, EC and pH) interacted with each other, and if there was any inter- action among the three land-use practices. Data analysis Similar abbreviations; 3 yr-regorested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for second 3-year-old RR, CP and BR land-uses, and for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other three transects Wheat-pasture rotation, now in wheat phase Well-established 5–10-m-tall stand, thick litter on ground Wheat-pasture rotation, now in pasture phase Not so well-established 5-m-tall stand, thick litter The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-regorested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for second 3-year-old RR, CP and BR land-uses, and for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other three transects other two transects for the 3 years of RR stand, and the corresponding CP and BR land-uses. A mean separation analysis was conducted to determine whether the depth incremented soil attributes organic matter (OC) and mineral nitrogen (MN), electrical conductivity (EC) and pH were sufficiently sensitive to characterise and dis- criminate the land-use practice. The depth incremented means and the corresponding pools for a given attribute (e.g. OC or EC) showed that the two replicates belong- ing to a given RR time-frame were statistically not repli- cates; therefore, we consider the replicates as different sites in the farm. This is where spatial variability or site-specificity concept came into play in this paper. In Tables 3 and 5, we show the functional relationship be- tween the soil attributes by defining OC = f (MN or EC or pH), MN = f (EC or pH) and EC = f(pH), to determine whether the land use practice and/or site-specific vari- ability between replicates had any impact on the func- tional variability of the attributes. For the OC correlation, the 16 means from the 4 CP plots (Table 2) other two transects for the 3 years of RR stand, and the corresponding CP and BR land-uses. A mean separation analysis was conducted to determine whether the depth incremented soil attributes organic matter (OC) and mineral nitrogen (MN), electrical conductivity (EC) and pH were sufficiently sensitive to characterise and dis- criminate the land-use practice. The depth incremented means and the corresponding pools for a given attribute (e.g. Data analysis Similarly, the other two transects for the 6-year-old RR, and the corresponding CP, and BR land-uses Land-uses Slope aspect Proximity to the creek Vegetation character Replicate 1 for the 6-year-old reforest 6 yr-cropland-1 ~ 10–15% upslope, ~ 40 m from 6 yr reforested riprian-1 ~ 50 m from creek Wheat-pasture rotation, now in pasture phase 6 yr-reforested riparian-1 ~ 3% downslope, ~ 230 m from 6 yr-bare riparian-1 < 5 m away from creek bank Well re-established forest, 5–10-m-tall, thick litter on ground 6 yr-bare riparian-1 ~ 3% downslope, ~ 164 m from 6 yr-cropland-1 < 5 m from creek bank Replicate-2 for the 6-year-old reforest 6 yr-cropland-2 ~ 5–10% midslope, ~ 150 m from 6 yr reforested riparian-2 < 10 m away from the creek bank Wheat-pasture rotation, now in wheat phase 6 yr-reforested riparian-2 Relatively flat, ~ 165 m from 6 yr bare riparian-2 < 5 m away from creek bank Well-established 5–10-m-tall stand, thick litter on ground 6 yr-bare riparian-2 Relatively flat < 5 m away from creek bank Replicate 3 for the 3-year-old reforest 3 yr-cropland-1 ~ 10% midslope, ~ 146 m from 3 yr reforested riparian ~ 100 m from creek bank Wheat-pasture rotation, now in pasture phase 3 yr-reforested riparin-1 < 2% downslope < 5 m away from creek bank Not so well established < 5 m tall stand Replicate 4 for the 3-year-old reforest 3 yr-cropland-2 ~ 10% midslope, ~ 550 m from 3 yr-cropland-1 ~ 50 m from creek bank Wheat-pasture rotation, now in pasture phase 3 yr-reforested riparian-2 < 3% downslope, ~ 170 m from 3 yr-cropland-2 < 5 m away from creek bank Not so well-established 5-m-tall stand, thick litter 3 yr-bare riparin-2 < 3% downslope, ~ 90 m from 3 yr reforested riparian-2 < 5 m away from creek bank The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Soil organic carbon The OC content in any given profile section in the CP plot was generally higher than that in the corresponding RR, regardless of the number of years after reforestation, followed by BR. The mean OC-pools between replicates 1 and 2 or that between replicates 3 and 4 were signifi- cantly different from each other. Similar results were also obtained for that between the MN-pools or TDS-pools of replicates 1 and 2 and between replicates 3 and 4 (later in the text). The OC-distributions across all CP plots ranged from 44.0 to 13.40 g C/kg soil, 34.8 Page 5 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Table 2 The summary of mean separation analysis for soil organic matter (OC) and mineral nitrogen (MN) content distributions at 0.10 m depth increments and for their pools (totals) in soil profiles along the four transects to determine whether these were impacted by different land-use practices and/or the physical location of the transects in the farm. Soil organic carbon During the experimental design stage, two out of the four transects were considered as replicates for the 3-year-old reforested riparian (RR) and the corresponding cropland (CP) and bare riparian (BR) land-uses and the other two transects as replicates for the 6-year-old RR, CP and BR land-uses cropland (CP) and bare riparian (BR) land-uses and the other two transects as replicates for the 6-year-old RR, CP and BR land-uses Land-use Mean organic carbon (OC) content in the profile sections g C/kg soil OC-pool Depth increments (m) 0–0.10 0.10–0.20 0.20–0.30 0.30–0.40 0–0.40 Replicate 1 for the 6-year-old reforest 6 yr-cropland-1 35.6A,d 21.4B,d 18.3 C,a 15.7 C,d 91.0e 6 yr-bare riparian-1 24.3A,b 16.9B,b 13.3 C,b 10.3 D,b 64.8b 6 yr-reforested riparian-1 34.4A,d 18.2B,c 15.1 C,d 12.8 D,a 80.5f Replicate 2 for the 6-year-old reforest 6 yr-cropland-2 44.0A,a 36.9B,e 25.0 C,e 25.5 C,e 131.4i 6 yr-bare riparian-2 28.4A,e 18.2B,c 12.8 C,b 11.0 D,b 70.4d 6 yr-reforested riparian-2 34.2A,d 21.7B,d 18.1 C,a 14.9 D,d 88.9e Replicate 3 for the 3-year-old reforest 3 yr-cropland-1 43.1A,a 25.3B,a 17.7 C,a 13.4 D,a 99.5a 3 yr-reforested riparian-1 23.3A,b 15.3B,b 12.2 C,b 11.0 D,b 61.8b Replicate 4 for the 3-year-old reforest 3 yr-cropland-2 40.4A,a 26.1B,a 22.8 C,c 20.2 D,c 109.5c 3 yr-bare riparian-2 26.7A,c 18.2B,c 15.9 C,d 12.9 D,a 73.7d 3 yr-reforested riparian-2 34.8A,d 22.6B,d 16.0 C,d 16.1C,d 88.7e Mean nitrogen (MN) content in the profile sections g N/kg soil MN-pool 0–0.10 0.10–0.20 0.20–0.30 0.30–0.40 0–0.40 Replicate 1 for the 6-year-old reforest 6 yr-cropland-1 3.22 A,e 2.12 B,c 1.69 C,e 1.40 D,f 8.44 c 6 yr-bare riparian-1 2.24 A,b 1.42 B,d 1.21 C,d 0.89 D,b 5.76 b 6 yr-reforested riparian-1 3.02 A,e 1.81 B,e 1.24 C,d 0.88 D,b 6.95f Replicate 2 for the 6-year-old reforest 6 yr-cropland-2 3.61 A,c 3.03 B,f 2.17 C,f 1.81 D,g 10.62 g 6 yr-bare riparian-2 2.50 A,e 1.67 B,g 1.14 C,g 0.87 D,b 6.20d 6 yr-reforested riparian-2 2.84 A,d 2.04 B,c 1.42 C,c 1.12 D,c 7.42f Replicate 3 for the 3-year-old reforest 3 yr-cropland-1 3.96 A,a 2.32 B,a 1.57 C,a 1.61C,a 9.46a 3 yr-reforested riparian-1 2.08 A,b 1.23 B,b 0.94 C,b 0.87 C,b 5.12b Replicate 4 for the 3-year-old reforest 3 yr-cropland-2 3.57 A,c 2.07 B,c 1.44 C,c 1.07 D,c 8.15c 3 yr-bare riparian-2 2.40 A,d 1.50 B,d 1.32 C,d 1.28 C,d 6.50d 3 yr-reforested riparian-2 3.08 A,e 2.23 B,a 1.75 C,e 1.51 D,e 8.57c Same block letter(s) in a given row or small letter(s) in a given column indicate the means are not significantly different at P < 0.05. Same block letter(s) in a given row or small letter(s) in a given column indicate the means are not significantly different at P < 0.05. The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-reforested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for the second 3-year-old RR, the corresponding CP and BR land-uses along the second transect. The abbreviations for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other two transects are similar to what has been used for the 3-year-old RR transects land-uses Mineral nitrogen The mineral nitrogen (MN)-distributions across the CP plots ranged from 1.07 to 3.96 g N/kg soil, 0.70 to 3.08 g N/kg soil in RR and 0.87 to 2.50 g N/kg soil in BR, re- spectively (Table 2). The mean MN content in the CP, RR and BR plots were 2.29 g N/kg soil, 1.75 g N kg/soil and 1.54 g N/kg soil, respectively. The MN-pools across the CP plots ranged from 8.15 to 10.62 g N/kg soil, com- pared with 6.95 to 8.57 g N/kg soil for RR and 5.76 to 6.50 g N/kg soil for BR land-uses (Table 2). The MN content in any given profile section in the CP plot was generally higher than that in the corresponding RR, re- gardless of the number of years after reforestation, followed by BR. The mean MN content in any given profile section of replicate 1 was significantly different from the corresponding section in replicate 2, regard- less of the land-use practice, and similar results were obtained for that between replicates 3 and 4. The OC content in any given profile section of CP plot in replicate 1 was significantly different from the corre- sponding section in RR, which in turn varied from BR (Table 2). This implies the depth incremented OC-distributions among the three land-use practices in replicate 1 were significantly different from each other (Table 2). Similar results were also obtained for repli- cates 2, 3 and 4. These significant differences between OC-distributions of the three land-use practices indicate that mean separation analysis is useful to discriminate the impact of land-use practice on soil OC content along a transect. The OC-pools also showed significant differ- ences between the three land-use practices. The MN-distributions among the three land-use prac- tices on replicates 2, 3 and 4 were 100% significantly dif- ferent from each other, except in replicate 1. The MN-pools among the three land-use practices were 100% significantly different from each other on each one of the replicates 1, 2 and 3, except 66% in replicate 4. Neither the MN-distributions nor the MN-pools in all four replicates showed 100% significant differences among the three land-use practices. The OC contents in three out of the four sections of the soil profile under the 6-year-old RR plot in replicate 1 were significantly different from the corresponding sections in the RR of replicate 2 (Table 2). Soil organic carbon The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-reforested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for the second 3-year-old RR, the corresponding CP and BR land-uses along the second transect. The abbreviations for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other two transects are similar to what has been used for the 3-year-old RR transects land-uses Page 6 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Rasiah and Florentine Ecological Processes (2018) 7:39 Page 6 of 12 Rasiah and Florentine Ecological Processes to 11.0 g C/kg soil in the RR plots and 28.4 to 10.3 g C/kg soil in the BR plots, respectively (Table. 2). The mean OC content in the CP, RR and BR plots were 107.8 g C/kg soil, 80.0 g C kg/soil and 69.6 g C/kg soil, respectively. The OC-pools across the CP plots ranged from 91 to 131 g C/kg soil compared with 61.8 g to 88.9 g C/kg soil for RR and 64.8 to 73.7 g C/kg soil for BR (Table 2). This trend prompted us to claim that a given replicate was independent of the other three with regard to the soil attributes and proceeded to use it as a char- acteristics property for a given land-use along a given transect, implying the means are site-specific property for that land-use (Tables 2 and 4). CP, followed by RR and BR, respectively, suggesting that C turnover dynamics with soil depth depended on the land-use practice. The intercepts followed trends quali- tatively similar to the slopes, in that they decreased with increasing depth and varied between land-use practices. Mineral nitrogen They were carried out using the means listed in Table 2. The correlation for the two 3-year-old reforested riparian (RR) land-use is abbreviated as 3 yr-reforested riparian, the corresponding cropland as 3 yr-cropland (CP), and the bare riparian as 3 yr-bare riparian (BR) in this table. Similar abbreviations are used for the 6-year-old RR and the corresponding CP and BR land-use practices different from each other. The intercepts followed trends similar to the slopes, being highest for CP followed by RR and BR. The EC in a given profile section was generally higher in BR plots, followed by RR and CP, respectively (Table 4). The TDS-pools across the CP plots ranged from 291 to 329 mg/L, compared with 175 to 452 mg/L for RR plots, and 435 to 621 mg/L for the BR plots. After the land-use changed from BR to RR, the average TDS decreased by ~ 36%. The net MN deposition (which is the difference be- tween the MN-pools of RR and BR) in the 3-year-old stands of RR in replicate 4 was 2.07 g N/kg soil, com- pared with 1.19 g N/kg soil and 1.22 g N/kg soil under the 6-year-old RR stands in replicate 1 and replicate 2, respectively. The net MN depletion (MN-pools of CP minus BR) in the deforested BR plot in replicate 4 was 2.02 g N/kg soil, compared with 2.68 g N/kg soil in repli- cate 1 and 4.42 g N/kg soil in replicate 2, respectively. Even though there was a general trend for the EC to decrease with increasing soil depth, the correlations were significant only for the CP plots (Table 3). The TDS depletion that occurred 3 years after reforestation in the 3 yr-RR-2 stand was 169 mg/L, compared with 182 mg/L and 65 mg/L in the 6 yr-RR-1 and 6 yr-RR-2 stands, respectively. Compared with the BR land-use practice, the CP land-use practice retarded the TDS de- pletion by ~ 309 mg/L in the 3 yr-CP-2 plot, compared with 251 mg/L and 106 mg/L in the 6 yr-CP-1 and 6 yr-CP-2 stands, respectively. Electrical conductivity The significant difference between the electrical con- ductivity distributions (EC-distributions) of the 6-year- old RR replicates was 100%, and that between the corresponding BR replicates and CP replicates were 75% and 50%, respectively (Table 4). The significant differ- ence between the 3-year-old RR replicates’ electrical conductivity (EC)-distributions was 100% and 25% for that between the corresponding CP replicates. The TDS-pools between the 6-year-old RR replicates were significantly different from each other, as it was between BR replicates, but not between the CP replicates. The TDS-pools between the 3-year-old RR replicates were significantly different from each other, as it was between the corresponding CP replicates. Neither EC-distributions nor TDS-pools were found to be effective discriminators to characterise the site-specific impact of long-term CP land-use. On the other-hand, both indicators were effective to discrimin- ate the site-specific impact of the change in land-use practice from BR to RR. Mineral nitrogen The slopes were significantly dif- ferent among land-use practices, being highest for the Rasiah and Florentine Ecological Processes (2018) 7:39 Page 7 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Page 7 of 12 Table 3 The summary for the simple linear correlation analysis performed between the mean soil attribute content in a given profile section and the corresponding profile depth to determine whether the distributions are impacted by land-use practices Land-use Intercept Slope Correlation coefficient Correlation between organic carbon content and the profile depth 3 yr-cropland 40.1 −84.6 (4.2) d 0.93 3 yr-bare riparian 26.1 −34.0 (3.1) d 0.86 3 yr-reforested riparian 29.7 −56.2 (4.7) d 0.87 6 yr-cropland 40.8 −65.1 (4.2) d 0.78 6 yr-bare riparian 26.8 −46.9 (2.5) d 0.92 6 yr-reforested riparian 40.4 −72.2 (8.8) d 0.72 Correlation between mineral nitrogen content and profile depth 3 yr-cropland 3.79 −7.97 (0.46) d 0.91 3 yr-bare riparian 2.06 −3.92 (0.28) d 0.91 3 yr-reforested riparian 2.62 −4.56 (0.55) d 0.73 6 yr-cropland 3.61 −6.09 (0.37) d 0.90 6 yr-bare riparian 2.46 −4.87 (0.20) d 0.95 6 yr-reforested riparian 3.07 −6.39 (0.23) d 0.96 Correlation between electrical conductivity and profile depth 3 yr-cropland 0.209 −0.458 (0.03) d 0.88 6 yr-cropland 0.200 −0.359 (0.03) d 0.81 The correlations are significant at P < 0.05. They were carried out using the means listed in Table 2. The correlation for the two 3-year-old reforested riparian (RR) land-use is abbreviated as 3 yr-reforested riparian, the corresponding cropland as 3 yr-cropland (CP), and the bare riparian as 3 yr-bare riparian (BR) in this table. Similar abbreviations are used for the 6-year-old RR and the corresponding CP and BR land-use practices Table 3 The summary for the simple linear correlation analysis performed between the mean soil attribute content in a given profile section and the corresponding profile depth to determine whether the distributions are impacted by land-use practices Table 3 The summary for the simple linear correlation analysis performed between the mean soil attribute content in a given profile section and the corresponding profile depth to determine whether the distributions are impacted by land-use practices Table 3 The summary for the simple linear correlation analysis performed between the mean soil attribute content in a given profile section and the corresponding profile depth to determine whether the distributions are impacted by land-use practices The correlations are significant at P < 0.05. Mineral nitrogen We define this as that 75% of the OC-distributions between the 6-year treatment rep- licates were significantly different from each other. The significant difference between the OC-distributions of the CP plots in replicate 1 and 2 was 100%, and also 100% be- tween the BR plots. The OC-distributions between the RR plots in replicates 3 and 4 were 100% significantly different, and the corresponding significant difference between the CP plots was 50%. The OC-pools among the four RR replicates were 100% significantly differ- ent from each other, and similar results were obtained for the CP and BR replicates. The significant differences between the MN-distribu- tions of the 6-year-old RR replicates, and the corre- sponding BR and CP replicates, were 100%, 75% and 100%, respectively. The MN-distributions between the 3-year-old RR replicates were 100% significantly different and a similar result was obtained for the corresponding CP replicates. The MN-pools between the replicates were 100% significantly different from each other. The MN-pools were generally able to discriminate the im- pacts of land-use practices and site-specific characteris- tics differences between transects/replicates on MN content in soil profiles. The net OC deposition (difference between the OC-pools of RR and BR) 3 years after reforestation in rep- licate 4 was 15.0 g C/kg soil, compared with 15.7 g/kg soil and 18.5 g C/kg soil 6 years after reforestation in replicates 1 and 2, respectively. The net OC depletion (OC-pools of CP minus BR) in the deforested BR plot in replicate 4 was 35.8 g C/kg soil, compared with 26.2 g C/kg soil in repli- cate 1 and 61.0 g C/kg soil in replicate 2, respectively. The magnitudes of the slopes of the correlations indi- cate that MN-distributions decreased with increasing depth, regardless of the land-use practices in the repli- cates/transects (Table 3). The slopes were significantly different among the three land-use practices, being high- est for CP followed by RR and BR, respectively, for the 3-year-old RR replicates. Even though a similar trend was observed for the 6-year-old RR in replicates 1 and 2, the slopes for the CP and RR are not significantly The absolute values of the slopes of the correlations indicate that OC content decreased with increasing depth, regardless of the land-use practices in the repli- cate/transect (Table 3). Soil pH The depth incremented pH distributions between the 6-year-old RR replicates was 100% significantly different, Rasiah and Florentine Ecological Processes (2018) 7:39 Page 8 of 12 Table 4 The summary of mean separation analysis for electrical conductivity (EC) and pH distributions at 0.10 m depth increments and for the total dissolved solutes (TDS-total) in soil profiles along the four transects to determine whether they were impacted by land-use practices and/or the physical locations of the transects in the farm. Soil pH During the experimental design stage, two out of the four transects were considered as replicates for the 3-year-old reforested riparian (RR) and the corresponding cropland (CP) and bare riparian (BR) land-uses and the other two transects as replicates for the 6-year-old RR, CP and BR land-uses riparian (BR) land uses and the other two transects as replicates for the 6 year old RR, CP and BR land uses Land-use Electrical conductivity (EC) in the profile sections dS/m TDS-pool mg/L Depth increments (m) 0.00–0.10 0.10–0.20 0.20–0.30 0.30–0.40 0.00–0.40 Replicate 1 for 6-year-old reforest 6 yr-cropland-1 0.227 A,c 0.098 B,b 0.090 B,b 0.093 B,e 325.1 f 6 yr-bare riparian-1 0.203 A,a 0.233 B,d 0.251 C,e 0.223 D,f 576.0 g 6 yr-reforested riparian-1 0.177 A,f 0.151 B,c 0.140 C,f 0.147 C,g 393.6 h Replicate 2 for 6-year-old reforest 6 yr-cropland-2 0.197 A,a 0.117 B,a 0.100 C,b 0.100 C,e 329.0 f 6 yr-bare riparian-2 0.180 A,f 0.163 B,c 0.170 A,c 0.227 C,f 435.2 i 6 yr-reforested riparian-2 0.107 A,g 0.097 B,b 0.167C,c 0.207 D,h 369.9 j Replicate 3 for 3-year-old reforest 3 yr-cropland-1 0.197 A,a 0.110 B,a 0.077 C,a 0.070 D,a 290.6 a 3 yr-reforested riparian-1 0.083 A,b 0.080 A,b 0.087 B,b 0.023 C,b 174.7 b Replicate 4 for 3-year-old reforest 3 yr-cropland-2 0.230 A,c 0.110 B,a 0.070 C,a 0.077 C,a 311.7 c 3 yr-bare riparian-2 0.347 A,d 0.180 B,c 0.183 B,c 0.260 C,d 620.8 d 3 yr-reforested riparian-2 0.253 A,e 0.167 B,c 0.169 B,d 0.117 C,c 451.8 e Soil pH in the profile sections 0–0.10 0.10–0.20 0.20–0.30 0.30–0.40 Replicate 1 for 6-year-old reforest 6 yr-cropland-1 6.9 A,c 6.5 B,b 6.8 A,b 7.2 C,b 6 yr-bare riparian-1 8.0 A,e 8.4 B,d 8.6 C,d 8.7 C,c 6 yr-reforested riparian-1 7.1 A,d 7.8 B,c 8.0 C,c 8.3 D,c Replicate 2 for 6-year-old reforest 6 yr-cropland-2 6.6 A,c 6.1 B,b 6.3 B,e 6.7 A,a 6 yr-bare riparian-2 7.0 A,d 7.1 A,e 7.8 B,c 8.3 C,c 6 yr-reforested riparian-2 5.9 A,a 6.5 B,b 7.3 C,f 7.8 D,d Replicate3 for 3-year-old reforest 3 yr-cropland-1 5.7 A,a 5.3 B,a 5.8 A,a 6.5 C,a 3 yr-reforested riparian-1 5.7 A,a 6.2 B,b 6.7 C,b 7.1 D,b Replicate 4 for 3-year-old reforest 3 yr-cropland-2 6.1 A,b 5.5 B,a 5.8 C,a 6.2 A,a 3 yr-bare riparian-2 6.5 A,c 6.4 A,b 6.9 A,b 7.3 B,b 3 yr-reforested riparian-2 7.5 A,d 8.0 B,c 8.2 B,c 8.3 B,c Same block letter(s) in a given row or small letter(s) in a given column indicate the means are not significantly different at P < 0.05. Same block letter(s) in a given row or small letter(s) in a given column indicate the means are not significantly different at P < 0.05. The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-reforested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for the second 3-year-old RR, the corresponding CP and BR land-uses along the second transect. The abbreviations for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other two transect are similar to what has been used for the 3-year-old RR transects land-uses Discussion General The CP and BR plots have been under the cropping and bare, respectively, land-use practices for more than 50 years. Under these conditions, we suggest the distri- butions of OC, MN, EC and pH and the pools in the top 0.40 m depth might not have undergone any significant change during the relatively short study period of 3 to 6 years. Because of this, the major emphasis in this paper is on the changes from BR land-use to RR. There- fore, we considered BR as the background or control for this study (Rasiah et al. 2015). This proposal implies that any significant difference between the OC or MN or EC or pH distributions in the BR replicates is primarily due to site-specific impact on these attributes. The signifi- cant difference between the OC-, MN-, EC- and pH dis- tributions in the 6-year-old BR replicates were 50%, 75%, 75% and 75%, respectively (Tables 2 and 4). On the other hand, the significant difference between the pools of OC-, MN- and EC- of the BR replicates were 100% for each of them. Therefore, the pools of the attributes are better than the corresponding distributions to character- ise the appropriateness of design stage replicate assump- tions vs. the statistical validity of the replicates. Comparisons of the OC-distributions between each RR and its corresponding BR profiles indicate that C de- position has occurred down to 0.4 m depth in all of the four RR profiles, regardless of the temporal or site-specific impact (Table 2). This implies that when re- forestation is undertaken using deep-rooted trees, the depth to OC deposition should be given due consider- ation, in order to provide the most reliable information for atmospheric C-trapping benefits of reforestation. In this context, the slopes and intercepts of the correlations can be extrapolated to compute the maximum depth to which OC and MN depositions ceased to occur under RR. The extrapolations indicate the maximum depths to OC and MN depositions to occur under the 3-year-old RR stands are 0.53 m and 0.57 m, respectively, and 0.56 m and 0.48 m under the 6-year RR stands. Soil pH The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-reforested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for the second 3-year-old RR, the corresponding CP and BR land-uses along the second transect. The abbreviations for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other two transect are similar to what has been used for the 3-year-old RR transects land-uses Same block letter(s) in a given row or small letter(s) in a given column indicate the means are not significantly different at P < 0.05. The 3-year-old RR along one out of the four transects is abbreviated as 3 yr-reforested riparian-1, and the corresponding CP as 3 yr-cropland-1 and the BR as 3 yr-bare riparian-1. Similar abbreviations; 3 yr-reforested riparian-2, 3 yr-cropland-2, 3 yr-bare riparian-2 are used for the second 3-year-old RR, the corresponding CP and BR land-uses along the second transect. The abbreviations for the two 6-year-old RR and the corresponding CP and BR land-use practices along the other two transect are similar to what has been used for the 3-year-old RR transects land-uses Rasiah and Florentine Ecological Processes (2018) 7:39 Page 9 of 12 Rasiah and Florentine Ecological Processes Page 9 of 12 and that of the corresponding BR and CP replicates were 75% and 50%, respectively. The significant difference be- tween the 3-year-old RR replicates was 100% and 25% for the corresponding CP replicates. reality in large-scale experimental approach in which the typical experimental designs are difficult to apply. On the other hand, the results are very useful for up-scaling via modelling for situations where site-specific experi- mental soil characteristics approximate the new site/ farm and also may help to convince the stakeholders the impact of site-specific soil attributes variability on spatially varying reforestation issues. The distributions of OC, MN, EC or pH between the RR replicates generally produced 100% significant difference and as did the cor- responding pools. It seems the pools for attributes can be generally used to assess the impact of the land-use change from BR to RR. The distributions indicate that pH was generally higher, i.e. Soil pH in the alkaline range, in the BR plots, in con- trast to the RR, which were near neutral to slightly acidic, and CP, which were generally slightly acidic, ex- cept in the 3 yr-BR-2 plot where it was lower than the corresponding RR plot. The distributions suggest a gen- eral increase in pH with increasing depth, but the correl- ation analysis shows this as insignificant increase. Six years after reforestation, the pH in the RR plots was less than the corresponding BR plots, but this was not ex- plicit 3 years after reforestation. The significant differences (BR > RR > CP) that were observed between the land-use practices at a given site indicate treatment impact. Soil organic matter and mineral nitrogen Significantly higher average net organic matter (OC) de- position (17.1 vs. 15.0 g C/kg soil) was observed in the older RR stands compared to the younger (6 vs. 3 years), as was the case for the net TDS depletion (245.7 vs. 169.0 mg/L, 3 vs. 6 years). This implies that these two soil attributes can be used as indicators to characterise temporal impact of reforestation. However, the same possibility was not evident with regard to net mineral ni- trogen (MN) deposition. The pH distributions were a very effective indicator to discriminate site-specific impact of land-use change from BR to RR. It was also useful to discriminate be- tween BR and CP land-uses. Discussion General Such improvements may help in the return and re-establishment of the fauna and flora that disap- peared after deforestation (Petursdottir et al. 2012). It seems ~ 30% (net OC deposition divided by net OC depletion) of the depletion that occurred in the 6 yr-BR-2 plot had been restored during 6 years under the refor- ested stand compared with ~ 60% in the 6 yr-RR-1. Re- sults from other parts of Australia indicate similar depletions occurred after clearance of native vegetation (Cotching 2012). The OC-depletions that occurred after vegetation clearance have been linked to deteriorations in soil bio-physical-chemical condition (Luo et al. 2010; Lal 1981). Larger OC-pools under RR than BR implies reduction in the CO2 released to the atmosphere (Smith 2008; IPPC 2001). A net-depletion of 169 mg/L in 3 yr-RR-2 compared with 182 mg/L in 6 yr-RR-1 indicates temporal influence on solute wash-down/depletion in the latter. However, a similar comparison between 3 yr-RR-2 and 6 yr-RR-2 (169 vs. 65.3 mg/L) was observed, suggesting a reverse trend due to unfavourable site-specific soil abiotic condi- tions that prevailed at the latter site for salt washout, thereby overriding longer temporal influence there. Larger TDS-pools under BR than CP may be due to bare surface conditions in BR provided a more favourable situation for soil water evaporation than crop/pasture cover, thereby enabling relatively greater upward transport of solutes. The difference between the TDS-pools of 6 yr-BR-1 and 6 yr-CP-1 (251 mg/L) and 6 yr-BR-2 and 6 yr-CP-2 (106 m/L) may be due to differ- ences in site-specific subsurface variability influencing variation of evaporation rates or amounts between the sites. Salt washout after reforestation suggests forest stands along fence lines, riparian buffers and any other unused marginal lands on large farms may help to re- duce salinity risk at a farm level. The slopes of the correlations (Table 3) between OC- and MN-distributions and profile depth indicate the de- crease was most rapid under CP, followed by RR and BR, respectively, suggesting there were three different micro-ecosystem, soil biotic conditions, with regard to OC turnover dynamics along the transect. Discussion General The average net OC deposition (the difference be- tween the OC totals of RR and the corresponding BR) that occurred 6 years after commencement of reforest- ation was 17.1 g C/kg soil and the corresponding net OC depletion (the difference between the OC totals of CP and the corresponding BR) was 43.6 g C/kg soil. These quantities translate to ~ 39% of the depleted OC redeposited by the forest stand. The average MN deple- tion and the corresponding depositions were 3.55 g N/kg soil and 1.21 g N/kg soil, respectively, and these translate to ~ 35% of the depleted MN being redeposited. The average TDS depletion (difference between BR and RR) 6 years after reforestation was 123.8 mg/L and this translates to ~ 12% of what was in the BR profiles. The means separation analysis indicates the design stage replicate assumption is not valid, therefore we de- cided to proceeded with the statistical validity and con- sider the replicates as four independent transects, and the soil attributes characteristics along any given tran- sect is site-specific. Even though the validity of the ex- perimental results without replicates may be an issue, but site-specificity impact on soil attributes is what the Page 10 of 12 Page 10 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Rasiah and Florentine Ecological Processes (2018) 7:39 The net OC depletion that occurred after deforestation (> 50 years ago) in the 3 yr-BR-2 plot was 35.8 g C/kg soil compared with 26.2 g C/kg soil and 61.5 g C/kg soil in the 6 yr-BR-1 and 6 yr-BR-2 plots, respectively, indicating site-specific soil biotic functioning variability impact between the sites can be characterised by net OC deple- tion under BR. The average net OC depletion after forest clearing was ~ 41.2 g C/kg soil. The net OC-depletion of ~ 35.8 g C/kg soil extrapolates to ~ 212.7 Mg C/ha. towards the soil surface after native vegetation clearance (Robinson et al. 2003; Nicholson 2002), leading to a risk of dry-land salinity. The reforestation results from this study indicate that even a short time-frame (3 years) under reforestation was effective in combating salinity risk. The reforested stand could also help to reduce sur- face runoff, thereby reducing pesticides, herbicides and unused fertiliser (N and P) solutes loadings in streams, which may in turn be discharging into sensitive off-site water bodies. Discussion General The net MN deposition of 2.07 g N/kg that occurred 3 years after re- forestation commenced, compared with 1.19 g N/kg soil in 6 yr-RR-1 and 1.24 g N/kg soil in 6 yr-RR-2, suggest the longer temporal influence at the latter two sites was probably masked by the unfavourable site-specific condi- tions overriding the temporal influence. Significant TN deposition seems to have occurred down to 0.4 m depth in 3 yr-RR-2 and 6 yr-RR-2 plots compared with 0.2 m in the 6 yr-RR-1 plot. Even though deposition had occurred down to 0.4 m in 3 yr-RR-2 and 6 yr-RR-2 plots, it was higher in the former, 2.07 g/kg, than the latter, 1.24 g/kg, suggesting the temporal influence in the latter was over- ridden by unfavourable site-specific soil biotic impact. The significant differences between pH distributions of the BR plots indicate site-specific subsurface soil vari- ability, resulting variation of soil acidity/alkalinity be- tween the sites. This suggests the pH in BR can be used as a baseline to assess the impact of reforestation. The decreases in pH, from alkaline to near neutral or mildly acidic, suggests that alkaline solute-washout has oc- curred as a result of reforestation. In Australia, this process is considered a positive change, since it reduces salinity risk following native vegetation clearance (Clarke et al. 2002; van der Moezel and Bell 1992). The change from alkaline to near neutral to slightly acidic condition after reforestation of BR may help to retard, to a limited extent, salinisation. This should encourage farmers to undertake stream bank reforestation as a management option to combat salinity in similar situations. Electrical conductivity and soil acidity/alkalinity (pH) Higher electrical conductivity (EC)-distributions in BR than RR (Table 4) suggests solute washout from 0.4 m depth under RR after reforestation. This was probably due to (i) enhanced deep percolation through the root-chan- nels opened-up by a deep-rooted forest stand, (ii) increases in root-water uptake from deeper sections of profiles instead of being transported upwards, along with solutes, to the soil surface and (iii) decreases in soil evaporation that otherwise was driving upward solute transport. Even though soil acidity is a major issue under dry-land cropping in Australia, the mildly acidic condi- tion under CP are not yet an issue at this farm. Other workers (Kaleb et al. 2013; Wortley et al. 2013; Collins et al. Abbreviations BR B i i BR: Bare riparian; C: Carbon sequestration; CP: Cropland; EC: Electrical conductivity; MN: Mineral nitrogen; OC: Organic carbon; RR: Reforested riparian strips; TDS: Total dissolved solutes Interaction between soil attributes It is not surprising that OC was positively correlated with MN, regardless of the land management practices examined in this study. However, we anticipated similar trends for RR. Nevertheless, the negative correlation be- tween OC and pH under CP is consistent with the ex- pectation that when soil pH changed from alkaline to near neutral, the OC turnover dynamics improved and a similar change was anticipated with EC. The positive correlation between OC and EC implied that when TDS increases, the OC turnover dynamic responds likewise. However, this trend is unrealistic (Table 5). A positive correlation between EC and pH suggests a trend be- tween OC and EC. The interactions involving the abiotic indicators indicate that they can have significant impact on OC and MN accumulation in the reforested land- scape, thereby improving the soil abiotic condition. To differentiate deforestation or reforestation associ- ated changes in soil biotic condition, we therefore rec- ommend the determination of TDS wherever salinity is a major issue, and OC where C-credit at farm scale is considered as an incentive for C-sequestration in soil. The determination of both OC and TDS are recom- mended where C-credit is considered as an incentive and salinity a major issue. In BR land-use, pH distributions provided a sufficiently sensitive baseline to discriminate site-specific subsurface soil conditions between the sites. This suggests that pH in BR can be used as an indicator to assess the impact of re- forestation on pH dynamics in reforestation. The interaction involving OC, EC and TDS indicate the need for further research into how other abiotic in- dicators may interact positively and contribute towards improvement in soil biotic condition. Such information may help us to increase the ability to refine and improve the C-sequestration potential of windbreaks associated with whole farm operations. Acknowledgments This study was funded through the Corangamite Catchment Management Authority (CO0910.06.051). The authors thank the landholders for site access and the Victorian Department of Primary Industries for discussions and analyses. Conclusions Even though the depth-incremented distributions of soil organic carbon (OC), soil mineral nitrogen (MN), total dissolved solutes (TDS) and pH were not sufficiently sen- sitive to discriminate changes in land-use management practices from native forest to cropland (CP) or bare ripar- ian (BR) and then from BR to reforested riparian (RR), the OC-, MN- and TDS-pools were effective as sensitive indi- cators to discriminate changes in land-use. However, the pools were not consistently sensitive enough to differenti- ate the temporal impact of reforestation on the changes in soil condition. The net-deposition of OC and the net-depletion of TDS were found to be sufficiently sensi- tive to differentiate site-specific and temporal impacts of reforestation. However, there was also a potential for the temporal impact of reforestation on the soil attributes to be overridden by site-specific variability. Discussion General The correlations were conducted using the means listed in Table 2 for the four cropland land-use and similarly for the other land-use practices. For any given land-use practice, the correlations were conducted for OC = f (MN or EC or pH), MN = f(EC or pH) and EC = f(pH). Only the significant correlations are reported in this table. Electrical conductivity is abbreviated as EC The correlations are significant at P < 0.05. The correlations were conducted using the means listed in Table 2 for the four cropland land-use and similarly for the other land-use practices. For any given land-use practice, the correlations were conducted for OC = f (MN or EC or pH), MN = f(EC or pH) and EC = f(pH). Only the significant correlations are reported in this table. Electrical conductivity is abbreviated as EC Discussion General 2012; Fennessy and Cronk 1997) have reported that The major proportion of solutes in Victorian sodosols are saline and these are known to have been transported Page 11 of 12 Rasiah and Florentine Ecological Processes (2018) 7:39 Page 11 of 12 Page 11 of 12 Table 5 The summary for the simple linear correlation analysis performed to determine whether the interactions between the different soil attributes under a given land-use practice are significant. The depth increment in the analysis was a selection variable Table 5 The summary for the simple linear correlation analysis performed to determine whether the interactions between the different soil attributes under a given land-use practice are significant. The depth increment in the analysis was a selection variable Intercept Slope Correlation coefficient Cropland OC = f (MN) 3.30 10.36 (3.43) MN 0.94 OC = f (EC) 8.04 154.1 (5.0) EC 0.87 OC = f (pH) 70.99 −71.9 (8.94) pH 0.59 Bare riparian OC = f MN) 2.43 10.35 (1.11) MN 0.98 Reforested riparian OC = f (MN) 0.46 10.89 (2.13) MN 0.97 EC = f(pH) −0.28 0.062 (0.003) pH 0.84 The correlations are significant at P < 0.05. The correlations were conducted using the means listed in Table 2 for the four cropland land-use and similarly for the other land-use practices. For any given land-use practice, the correlations were conducted for OC = f (MN or EC or pH), MN = f(EC or pH) and EC = f(pH). Only the significant correlations are reported in this table. Electrical conductivity is abbreviated as EC reforestation and associated forest litter may help to re- duce agricultural pollutants, such as soluble N and P, sediment and pesticides, in surface runoff from crop- lands, thereby improving the water quality of runoff, and consequently improving water quality downstream. The higher MN concentrations observed in RR than BR might have been due to reduced surface runoff. Vertical transport of solutes, including MN, in soil infiltrating water may have resulted in the observed increase of the MN-pools in the RR profiles. Decreases in bulk density after reforestation and the deep root-channels opened-up under RR might have provided conditions favourable for deep percolation into soil profiles (Rasiah et al. 2015). The anticipated improvements in creek water quality may help to improve the on- and off-site aquatic ecosystem health, in addition to other environ- mental benefits (Montaggnini 2006). The correlations are significant at P < 0.05. References https://doi.org/10.1186/s13021–015–0025–6 Collins KE, Doscher C, Renie HG, Ross JG (2012) The effectiveness of riparian ‘restoration’ on water quality—a case study of lowland stream in Canterbury New Zealand. Restor Ecol 21:40–48 van der Moezel PG, Bell DT (1992) Comparative water use in river red gum and swamp she–oak seedlings stressed with different salinities. Land Water Res News 12:19–22 Cotching WE (2012) Carbon stock in Tasmanian soils. Soil Res 50:83–90 S C k ( ) h ff d l f Cotching WE (2012) Carbon stock in Tasmanian soils. Soil Res 50:83–90 F MS C k JK ( ) Th ff d l f Wortley L, Hero J, Howes M (2013) Evaluating ecological restoration success: a review of the literature. Restor Ecol 21:537–543 Fennessy MS, Cronk JK (1997) The effectiveness and restoration potential of riparian ecotones for the management of nonpoint source pollution, particularly nitrate. Critical Review. Environ Sci Tech 27:285–317 Fennessy MS, Cronk JK (1997) The effectiveness and restoration potential of riparian ecotones for the management of nonpoint source pollution, particularly nitrate. Critical Review. Environ Sci Tech 27:285–317 Gray JM, Murphy BW, Chapman GA (2002) Predicting soil distribution. Joint DLWC and ASSSI Technical Poster, DLWC, Sydney Hobbs RJ, Norton DA (1996) Towards a conceptual framework for restoration ecology. Restor Ecol 4:93–110 IPPC (2001) Intergovernmental panel on climate change, 2001. In: Climate change: the scientific basis. Cambridge Univ. Press, Cambridge Jose S (2009) Agroforestry for ecosystem services and environmental benefits: an overview. Agrofor Syst 76:1–10 Joyce EE (2003) Western volcanic plains, Victoria. CRE LEME, School of Earth Sciences, University of Melbourne, Melbourne Kaleb KH, Matt RW, Charlotte R (2013) Cascading ecological responses to an in– stream restoration project in a Midwestern river. Restor Ecol 22:72–80 Lal R (1981) Deforestation and hydrological problems. In: Lal R, Russel EW (eds) Tropical Agricultural Hydrology. Wiley, UK Lal R (2004) Soil carbon sequestration to mitigate climate change. Geoderma Fennessy MS, Cronk JK (1997) The effectiveness and restoration potential of riparian ecotones for the management of nonpoint source pollution, particularly nitrate. Critical Review. Environ Sci Tech 27:285–317 Gray JM, Murphy BW, Chapman GA (2002) Predicting soil distribution. Joint DLWC and ASSSI Technical Poster, DLWC, Sydney Gray JM, Murphy BW, Chapman GA (2002) Predicting soil distribution. Joint DLWC and ASSSI Technical Poster, DLWC, Sydney Lal R (2004) Soil carbon sequestration to mitigate climate change. Competing interests N in legume–amended soils from a spatially variable landscape. Geoderma 99:239–256 The authors declare that they have no competing interests. Rayment GE, Higgins FR (1992) Australian laboratory handbook of soil and water chemical methods, Australian Soil Survey Handbook Series. Inkata Press, Melbourne Rayment GE, Higgins FR (1992) Australian laboratory handbook of soil and water chemical methods, Australian Soil Survey Handbook Series. Inkata Press, Melbourne Availability of data and materials Availability of data and materials Petursdottir T, Asa AL, Benediktsson K (2012) An evaluation of the short–term progress on restoration combining ecological assessment and public perception. Restor Ecol 21:75–85 All data generated and analysed during this study are included in this article. Rasiah V, Florentine SK, Dahlhaus P (2015) Environmental benefits inferred from impact of reforestation of deforested creek bank on soil conditioning: a case study in Victoria, Australia. Agrofor Syst 89(2):345–355 References Shepherd D, Montagnini F (2001) Above ground carbon sequestration potential in mixed and pure tree plantations in the humid tropics. J Trop Forest Sci 13: 450–459 Shepherd D, Montagnini F (2001) Above ground carbon sequestration potential in mixed and pure tree plantations in the humid tropics. J Trop Forest Sci 13: 450–459 Agren GI, Bosatta E (1996) Quality: a bridge between theory and experiment in soil organic matter studies. Oikos 76:522–528 Agren GI, Bosatta E (1996) Quality: a bridge between theory and experiment in soil organic matter studies. Oikos 76:522–528 Agren GI, Bosatta E (1996) Quality: a bridge between theory and experiment in soil organic matter studies. Oikos 76:522–528 An S, Cheng Y, Huan Y, Liu D (2012) Effects of revegetation on soil microbial biomass, enzyme activities, and nutrient cycling on the loess plateau in China. Restor Ecol 21:600–607 Silver WL, Kueppers LM, Lugo AE, Ostertag R, Matzek V (2004) Carbon sequestration and plant dynamics following reforestation of tropical pasture. Ecol Appl 14:1115–1127 Smith P (2008) Land use change and soil organic carbon dynamics. Nutr Cycl Agroecosyst 81:169–178 Statgraphics Centrion XVI (2010) StatPoint Technologies, Inc, Warrenton, USA Tran DB, Hoang TV, Dargusch P (2015) An assessment of carbon stock and sodicity tolerance of disturbed Melaleuca forests in Southern Vietnam. Carbon Bal Manage 10:15. https://doi.org/10.1186/s13021–015–0025–6 van der Moezel PG, Bell DT (1992) Comparative water use in river red gum and swamp she–oak seedlings stressed with different salinities. Land Water Res News 12:19–22 W l L H J H M (2013) E l i l i l i Silver WL, Kueppers LM, Lugo AE, Ostertag R, Matzek V (2004) Carbon sequestration and plant dynamics following reforestation of tropical pasture. Ecol Appl 14:1115–1127 Chen JM, Liu J, Cihlar J, Goulden ML (1999) Daily canopy photosynthesis model through temporal and spatial scaling for remote sensing applications. Ecol Model 124:99–119 Smith P (2008) Land use change and soil organic carbon dynamics. Nutr Cycl Agroecosyst 81:169–178 Clarke CJ, George RJ, Bell RW, Hatton TJ (2002) Dryland salinity in South–Western Australia: its origins, remedies, and future research directions. Aust J Botany 40:93–113 Clarke CJ, George RJ, Bell RW, Hatton TJ (2002) Dryland salinity in South–Western Australia: its origins, remedies, and future research directions. Aust J Botany 40:93–113 Tran DB, Hoang TV, Dargusch P (2015) An assessment of carbon stock and sodicity tolerance of disturbed Melaleuca forests in Southern Vietnam. Carbon Bal Manage 10:15. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Robinson N, Rees D, Reynard K, MacEwan R, Dahlhaus P, Imhof M, Boyle G, Baxter N (2003) A land resource assessment of the Corangamite region. The State of Victoria, Department of Primary Industries, 2003. Published by the Department of Primary Industries, 2003. Primary Industries Research Victoria, Bendigo, Cnr Midland Highway and Taylor St Epsom Vic 3551. ISBN 1 74106 670 0 Received: 9 May 2018 Accepted: 25 November 2018 Ethics approval and consent to participate Not applicable. Rasiah V, Kay BD (1998) Legume–N mineralization: effects of aeration and size distribution of water filled pores. Soil Biol Biochem 30(1):89–96 Rasiah V, Kay BD (1999) Temporal dynamics of microbial biomass– and mineral– N in legume–amended soils from a spatially variable landscape. Geoderma 99:239–256 Rasiah V, Kay BD (1999) Temporal dynamics of microbial biomass– and mineral– N in legume–amended soils from a spatially variable landscape. Geoderma 99:239–256 Competing interests Authors’ contributions SF received the funding, VR and SF conceived and designed the data collection. VR wrote the MS. VR and SF edited the MS. Both authors have read and approved the final manuscript. Rasiah V, Florentine SK, Williams BL, Westbrooke ME (2004) Soil properties dynamics under abandoned pasture in deforested tropical rainforest in Australia. Geoderma 120:35–45 Funding Fi i l Financial support for this project was provided by the Corangamite Catchment Management Authority (Funding number CO0910.06.051). Page 12 of 12 Page 12 of 12 Page 12 of 12 Rasiah and Florentine Ecological Processes Rasiah and Florentine Ecological Processes (2018) 7:39 (2018) 7:39 Ecological Processes (2018) 7:39 Wortley L, Hero J, Howes M (2013) Evaluating ecological restoration success: a review of the literature. Restor Ecol 21:537–543 References Geoderma 123:1–22 Lindenmayer D, Hobbs RJ, Montague–Drake R (2008) A checklist for ecological management of landscapes for conservation. Ecol Lett 11:78–91 Lowrance RJ, Fail O, Hendrickson R, Leonard LE, Asmussen L (1984) Riparian forests as nutrient filters in agricultural watersheds. BioScience 34:374–377 Luo ZE, Wang G, Sun OJ (2010) Soil carbon change and its response to agricultural practices in Australian agro–ecosystems: a review and synthesis. Geoderma 155:211–223 Montaggnini F (2006) Environmental services of agroforestry systems. Food Products Press, New Haven Nair DV, Graetz DA (2004) Agroforestry as an approach to minimize nutrient loss from heavily fertilized soils: the Florida experience. Agrofor Syst 61:269–279 Nicholson C (2002) Review of the corangamite salinity strategy “Restoring the Balance”. Consulting report by Nicon Rural Services, Corangamite Catchment Management Authority Nicholson C (2002) Review of the corangamite salinity strategy “Restoring the Balance”. Consulting report by Nicon Rural Services, Corangamite Catchment Management Authority
https://openalex.org/W4288456810	http://acta.fapz.uniag.sk/journal/article/download/34/17	English	null	Laying, egg quality and blood profile of native hens	Acta fytotechnica et zootechnica/Acta fytotechnica et zootechnica	2,022	cc-by	7,123	Corresponding Author: Lukáš Zita, Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Animal Science, Kamýcká 129, 165 00 Prague, Czech Republic, e-mail: zita@af.czu.cz https://doi.org/10.15414/afz.2022.25.02.109-116 Licensed under a Creative Commons Attribution 4.0 International License The objective of this study was to assess egg quality parameters for the whole laying period depending on oviposition time and breed of Czech and Slovak native breeds of laying hens. Besides, to determine the differences between selected breeds in laying pattern, related to the oviposition place. Furthermore, biochemical blood parameters were measured at the end of the study. A total of 60 pullets at the age of 20 weeks were divided according to the breed. Each treatment consisted of 3 replications of 10 laying hens. The eggs were collected every day, at 6:00, 10:00 and 14:00 and the number of eggs was recorded for each oviposition time interval (from 14:00 to 5:59, from 6:00 to 9:59 and from 10:00 to 13:59). Moreover, the oviposition place (inside and outside the nest) and the number of eggs in particular place were recorded. In addition, blood samples were collected. Significantly heavier eggs were laid between 10:00 and 13:59 than between 6:00 and 9:59 h (52.44 vs. 51.39 g, resp.). Haugh units were highest in eggs from Czech golden spotted hens that were laid between 6:00 and 9:59 h and in eggs from Oravka hens that were laid between 6:00 and 9:59 h and between 10:00 and 13:59 h. Significantly lower content of yolk cholesterol was found in Czech golden spotted hens compared to Oravka hens (10.64 vs. 11.22 mg g-1, resp.). The Czech golden spotted hens had significantly higher level of glucose in blood serum than Oravka hens (16.47 vs. 14.03 mmol l-1, resp.). The Czech golden spotted hens, gene reserve of the Czech Republic, are not yet sufficiently described in scientific literature, which highlights the importance of this study. Keywords: cholesterol, egg-laying, Czech golden spotted hen, Oravka hen, oviposition acids, including alpha-linolenic acid (omega-3) and linoleic acid (omega-6), are essential for health. One egg contains approximately 70 mg of omega-3 fatty acids. They are formed by metabolization of linoleic acid, arachidonic, alpha-linolenic, eicosapentaenoic (EPA) and docosahexaenoic (DHA). EPA and DHA play an important role in prevention of cardiovascular diseases and furthermore have a positive impact against infections (Sparks, 2006). Cholesterol belongs among the substantial egg constituents. The average amount of cholesterol in one egg, precisely in one egg yolk is 200 mg. It is an important component as it influences the function of steroid hormones, vitamin D and works Laying, egg quality and blood profile of native hens Adam Kraus1, Ondřej Krunt1, Lukáš Zita1, Karolína Machová2, Cyril Hrnčár3, Eva Chmelíková4 1Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Animal Science, Czech Republic 2Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Genetics and Breeding, Czech Republic 3Slovak University of Agriculture in Nitra, Faculty of Agrobiology and Food Resources, Insitute of Animal Husbandry, Slovak Republic 4Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Veterinary Sciences, Czech Republic Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk The Czech golden spotted hens, gene reserve of the Czech Republic, are not yet sufficiently described in scientific literature, which highlights the importance of this study. Original Paper Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk http://www.acta.fapz.uniag.sk Laying, egg quality and blood profile of native hens Adam Kraus1, Ondřej Krunt1, Lukáš Zita1*, Karolína Machová2, Cyril Hrnčár3, Eva Chmelíková4 1Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Animal Science, Czech Republic 2Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Genetics and Breeding, Czech Republic 3Slovak University of Agriculture in Nitra, Faculty of Agrobiology and Food Resources, Insitute of Animal Husbandry, Slovak Republic 4Czech University of Life Sciences Prague, Faculty of Agrobiology, Food and Natural Resources, Department of Veterinary Sciences, Czech Republic Article Details: Received: 2022-01-20 \| Accepted: 2022-03-21 \| Available online: 2022-06-30 https://doi.org/10.15414/afz.2022.25.02.109-116 Licensed under a Creative Commons Attribution 4.0 International License The objective of this study was to assess egg quality parameters for the whole laying period depending on oviposition time and breed of Czech and Slovak native breeds of laying hens. Besides, to determine the differences between selected breeds in laying pattern, related to the oviposition place. Furthermore, biochemical blood parameters were measured at the end of the study. A total of 60 pullets at the age of 20 weeks were divided according to the breed. Each treatment consisted of 3 replications of 10 laying hens. The eggs were collected every day, at 6:00, 10:00 and 14:00 and the number of eggs was recorded for each oviposition time interval (from 14:00 to 5:59, from 6:00 to 9:59 and from 10:00 to 13:59). Moreover, the oviposition place (inside and outside the nest) and the number of eggs in particular place were recorded. In addition, blood samples were collected. Significantly heavier eggs were laid between 10:00 and 13:59 than between 6:00 and 9:59 h (52.44 vs. 51.39 g, resp.). Haugh units were highest in eggs from Czech golden spotted hens that were laid between 6:00 and 9:59 h and in eggs from Oravka hens that were laid between 6:00 and 9:59 h and between 10:00 and 13:59 h. Significantly lower content of yolk cholesterol was found in Czech golden spotted hens compared to Oravka hens (10.64 vs. 11.22 mg g-1, resp.). The Czech golden spotted hens had significantly higher level of glucose in blood serum than Oravka hens (16.47 vs. 14.03 mmol l-1, resp.). 2.1 Animals and management as a precursor for bile to absorb and digest fat. Eggs are also a rich source of vitamins and minerals. Finally yet importantly, eggs are a source of antibodies IgY that are effective against bacterial and virus infections (Zaheer, 2015). Two native breeds of hens were included in present study, the Czech golden spotted (CGS) hen and the Oravka (OR) hen. Floor pens with litter, which met the criteria set by Council Directive 1999/74/EC, were used as housing systems. The housing system design and equipment were made according to Kraus et al. (2021). A total of 120 pullets were obtained from the breeding facility at the age of 20 weeks and immediately divided according to the breed (60 pullets per breed). Each treatment consisted of 3 replications of 20 laying hens. The environmental conditions were controlled and maintained same for all animals. The temperature was kept between 18 °C and 20 °C and humidity between 50 and 60% throughout the whole study. Hens from 20 weeks of age were provided with 14 hours of light, which was regularly extended to 16 h from the 24 weeks of age. The lighting intensity was kept between 5–10 lx. Regarding the feeding, it was provided by commercial feed mixtures. Hens from the age of 20 weeks, feed mixture labelled as N1 was used and contained 16.71% of crude protein (CP) and 11.40 MJ of metabolizable energy (ME) and hens from the age of 42 weeks feed mixture labelled as N2 was used (15.41% of CP, 11.48 MJ of ME). Access to both, feed and water was ad libitum for the duration of the whole study. There is a large number of factors, which have an impact on internal and external egg quality. Scientific studies usually focus their attention on the effect of breed (genotype, resp.), age of animals, nutrition (Tang et al., 2015) housing system (Zita et al., 2018), storage conditions (Vlčková et al., 2019) and microbial contamination (Krunt et al., 2021). However, there are other factors that affect egg quality and one of them is oviposition (Hrnčár et al., 2013; Tůmová et al., 2017; Shaker et al., 2019). According to Tůmová et al. (2017) time of the oviposition is an important factor and influences especially egg weight and eggshell quality parameters. Apart from the oviposition time, attention should be also directed towards oviposition place (Oliveira et al., 2019). 2.2 Egg quality and blood analysis The collection of eggs for the purpose of the study started when hens were 24 weeks old and finished when hens were 64 weeks old. The eggs were collected every day, three times a day, at 6:00, 10:00 and 14:00 and the number of eggs was recorded for each oviposition time interval (from 14:00 to 5:59, from 6:00 to 9:59 and from 10:00 to 13:59). Furthermore, the oviposition place (inside and outside the nest) and the number of eggs in particular place were recorded. The design of the housing system was made according to Kraus et al. (2021). The eggs for egg quality analysis (50 eggs from each breed at each age) and for yolk cholesterol analysis were collected every four weeks and the collection of eggs was performed for 3 consecutive days to reach a required number of eggs for the analysis. After the collection, eggs were stored at 6 °C until the day of the analysis (24 h after the egg collection). The yolks for cholesterol analysis were randomly selected (5 egg yolks from each breed at each age) from the eggs, which were used for quality analysis. Each yolk was evaluated separately as one sample and was evaluated in triplicate. The evaluation of egg quality parameters included egg weight (EW), shape index (SI), eggshell reflectivity (ESR), thickness (EST), strength (ESST), surface (ESS), index (ESI) and proportion (ESP), yolk colour (YC), proportion (YP) and index (YI), cholesterol concentration in yolk (CH_Y), albumen The main objective of this study was to assess egg quality parameters for the whole laying period depending on oviposition time and breed of Czech and Slovak native breeds of laying hens. Besides, to determine the differences between selected breeds in laying pattern, related to the oviposition place. Furthermore, measure and evaluate biochemical blood parameters at the end of the study. 2.1 Animals and management Regarding the effect of genotype, the use of native breeds is still decreasing in favour of commercial hybrids, who achieve higher production. Therefore, breeding of native hens is dependent especially on small and hobby farmers (Krawczyk et al., 2011). The programs for the conservation of animal genetic resources also contribute significantly to preserve native breeds (Belew et al., 2016). Both, the Czech golden spotted and Oravka hens belong into genetic resources of the country of its origin, Czech Republic, and Slovakia (Kraus et al., 2021). Protection of native breeds is important because of the high adaptability and resistance of these animals in local conditions (Begli et al., 2010) and for keeping valuable genes (Belew et al., 2016). Biochemical blood parameters describe the health status and point out any changes in organism, which may have nutritional, physiological, or even pathological character (Koronowicz et al., 2016). These parameters may influence not only health of the animals, but also their production (Pavlík et al., 2007). 1 Introduction Nowadays, eggs belong to the most favourite animal products and the reasons for their popularity are numerous (Lesnierowski & Stangierski, 2018). In terms of nutritional value, eggs represent a great source of all basic nutrients and apart from that dispose of many characteristics, which have a positive effect on human health status (Iannotti et al., 2014). Specifically, eggs contain high- quality proteins that are composed of balanced number of amino acids, such as histidine, isoleucine, leucine, methionine, phenylalanine, threonine, tryptophan, and valine (Zaheer, 2015). Another important nutritional component of eggs are lipids. Polyunsaturated fatty Slovak University of Agriculture in Nitra Faculty of Agrobiology and Food Resources 109 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk 2 Material and methods The experiment was authorized by the Ethical Committee for Animal Experimentation of Czech University of Life Sciences Prague. Faculty of Agrobiology and Food Resources Slovak University of Agriculture in Nitra 110 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk proportion (AP) and index (AI), Haugh units (HU) and yolk to albumen ratio (YAR). All measurements and devices were used according to Kraus et al. (2021). Furthermore, the determination of eggshell index (ESI) was calculated according to Ahmed et al. (2005). The egg quality analysis took place at the laboratory of the Department of Animal Science and blood analysis took place at the Department of Veterinary Sciences of the Faculty of Agrobiology, Food and Natural Resources of the Czech University of Life Sciences Prague. The effect of age was not considered in this study, all evaluated parameters were evaluated for the whole observed period. value of P ≤0.05 was considered as significant for all measurements. Furthermore, the effect of breed on oviposition time and oviposition place was assessed. Although the data were repeatedly measured on two flocks of hens, they were evaluated as independent observations. Pearson’s chi-square tests were used in intergroup comparisons of categorical variables. The relationship between the breed and the choice of nest for laying was tested first. This was followed by testing of the dependence of chosen place for laying and time for each breed separately. P-values lower than 0.05 were considered as statistically significant. The phi coefficient and Cramer’s coefficient were used to estimate the degree of dependence for four-field and the six-field table, respectively. However, based on the nature of our data, there was no difference between the two coefficients. The calculations were performed using a statistical program Statistica 12 (StatSoft, Inc., Tulsa, Oklahoma). Biochemical blood analysis was performed from blood serum and the following parameters were studied: aspartate aminotransferase (AST), total protein (TP), albumin (ALB), glucose (GLU), triacylglycerol (TAG), cholesterol (CHOL), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL), triacylglycerol and high-density lipoprotein cholesterol ratio (TAG_HDL), low-density lipoprotein cholesterol and high-density lipoprotein cholesterol ratio (LDL_HDL), non-high-density lipoprotein cholesterol (Non_HDL) and atherogenic index (ATI). Six birds per replication from each breed (in total, 36 animals) were randomly selected and slaughtered for the purpose of the blood analysis. 2.3 Statistical analysis The computer application SAS (SAS Inst. Inc., Cary, NC, USA) was used for the statistical analysis of the data. The effect of breed and oviposition time on each of egg quality parameters was assessed by the mixed model using the MIXED procedure of SAS: 3 Results and discussion All results are shown in detail in attached tables and figure. 3.1 External and internal egg quality regarding the breed and oviposition time Egg and eggshell quality parameters are presented in Table 1, while yolk and albumen quality parameters are presented in Table 2. The effect of breed, oviposition time and interaction of these two factors are shown in both tables for each parameter. Breed significantly affected ESR, EST, ESI, ESP (Table 1), YC, YP, CH_Y, AP, AI, HU and YAR (Table 2). The effect of oviposition time was calculated as statistically significant in EW, SI, ESR, EST, ESS, ESI, ESP (Table 1), YC, YI, AP, AI, HU and YAR (Table 2). The interaction between breed and oviposition time was significant in ESR, EST, ESST, ESI, ESP (Table 1), YC, YP, AP, HU and YAR (Table 2). 2 Material and methods Blood was collected into two types of tubes (each sample): the empty sterile tubes, which were used for all selected parameters apart from GLU and tubes with sodium chloride, which were used for the determination of GLU. The analysis of AST, TP, ALB, GLU, TAG, CHOL, HDL and LDL was made according to Kraus et al. (2021). Parameters, such as TAG_HDL, LDL_HDL was calculated as a ratio and Non_HDL according to van Deventer et al. (2011) and ATI according to Salma et al. (2007). 3.2 Hens’ oviposition regarding the time, place, and breed Yijkl = µ + Bi + OTj + (B × OT)ij + Aijk + eijkl The percentage of laid eggs regarding the oviposition time and place and the interaction between oviposition time and oviposition place of CGS hens is shown in Table 3 and of OR hens in Table 4. The statistically significant interaction between oviposition time and oviposition place was found only in CGS hens (Table 3). The preference of oviposition place regarding the breed and ratio between total numbers of laid eggs in each place are presented in Figure 1. Statistically significant difference between the oviposition places (inside the nest and outside the nest) was found only in CGS hens. where: Yijk – the value of trait, μ – the overall mean; Bi – the effect of breed (the CGS hens and the OR hens); OTj – the effect of oviposition time (6:00, 10:00 and 14:00); (B × OT)ij – the effect of the interaction between breed and oviposition time; Aijk – independent variable of the age; eijkl – the random residual error The significance of the differences among groups was tested by Duncan‘s multiple range test. 3.2 Hens’ oviposition regarding the time, place, and breed The Slovak University of Agriculture in Nitra Faculty of Agrobiology and Food Resources 111 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk Table 1 Egg and eggshell quality parameters regarding the breed and oviposition time Item Parameter Breed oviposition time EW (g) SI (%) ESR (%) EST (mm) ESST (N cm-2) ESS (cm2) ESI (g 100 cm-2) ESP (%) CGS 52.05 74.68 56.81a 0.316b 38.94 77.19 7.61b 9.53b OR 51.78 74.53 39.89b 0.319a 39.75 76.90 7.72a 9.69a 6 51.92ab 74.90a 47.41b 0.322a 39.57 77.06ab 7.80a 9.78a 10 51.39b 74.43b 49.73a 0.311b 38.97 76.45b 7.51c 9.46b 14 52.44a 74.50b 47.91b 0.319a 39.50 77.62a 7.67b 9.59ab CGS 6 52.07 74.76 55.19b 0.324a 39.33ab 77.23 7.81a 9.78a 10 51.21 74.63 59.32a 0.305c 37.65c 76.25 7.36d 9.28c 14 52.87 74.66 55.93b 0.319b 39.84ab 78.10 7.64c 9.52ab OR 6 51.77 75.04 39.63c 0.320ab 39.81ab 76.89 7.78ab 9.79a 10 51.57 74.23 40.13c 0.317b 40.28a 76.66 7.66c 9.63ab 14 52.01 74.33 39.90c 0.319b 39.15b 77.14 7.71bc 9.66ab P-value B 0.2756 0.3744 0.0001 0.0336 0.0578 0.2828 0.0005 0.0001 OT 0.0018 0.0489 0.0019 0.0001 0.4017 0.0020 0.0001 0.0001 B × OT 0.1202 0.1952 0.0156 0.0001 0.0011 0.1203 0.0003 0.0031 SEM 0.116 0.081 0.310 0.001 0.174 0.130 0.015 0.020 B – breed, OT – oviposition time, EW – egg weight, SI – shape index, ESR – eggshell reflectivity, EST – eggshell thickness, ESST – eggshell strength, ESS – eggshell surface, ESI – eggshell index, ESP – eggshell proportion, CGS – Czech golden spotted hen, OR – Oravka hen; SEM – standard error of the mean; P-value ≤0.05 means significant effect of concrete parameter. Values marked with different superscript letters for each parameter are significantly different Egg and eggshell quality parameters regarding the breed and oviposition time B – breed, OT – oviposition time, EW – egg weight, SI – shape index, ESR – eggshell reflectivity, EST – eggshell thickness, ESST – eggshell strength, ESS – eggshell surface, ESI – eggshell index, ESP – eggshell proportion, CGS – Czech golden spotted hen, OR – Oravka hen; SEM – standard error of the mean; P-value ≤0.05 means significant effect of concrete parameter. B – breed, OT – oviposition time, YC – yolk colour, YP – yolk proportion, YI – yolk index, CH_Y – cholesterol concentration in yolk, AP – albumen proportion, AI – albumen index, HU – Haugh units, YAR – yolk to albumen ratio, CGS – Czech golden spotted hen, OR – Oravka hen; SEM – standard error of the mean; P-value ≤0.05 means significant effect of concrete parameter. Values marked with different superscript letters for each parameter are significantly different 3.3 Biochemical blood parameters regarding the breed Table 5 displays biochemical blood parameters of CGS hens and OR hens. Statistically significant differences between CGS and OR hens in blood serum were found in GLU, TAG, CHOL, HDL, LDL, NonHDL. The rest of the evaluated parameters did not differ significantly. 3.2 Hens’ oviposition regarding the time, place, and breed Values marked with different superscript letters for each parameter are significantly different Table 2 Yolk and albumen quality parameters regarding the breed and oviposition time Item Parameter Breed oviposition time YC (point) YP (%) YI (%) CH_Y (mg g-1) AP (%) AI (%) HU (point) YAR CGS 6.28a 31.20b 45.37 10.64b 59.28a 8.66b 81.64b 0.53b OR 6.13b 31.68a 45.50 11.22a 58.63b 9.34a 83.65a 0.54a 6 6.17b 31.51 44.72c 11.02 58.72b 8.18c 79.65c 0.54a 10 5.97c 31.22 46.03a 10.95 59.33a 9.52a 84.74a 0.53b 14 6.46a 31.59 45.56b 10.82 58.82ab 9.29b 83.56b 0.54a CGS 6 6.31b 31.62abc 44.56 10.57 58.60c 7.76 78.03c 0.54ab 10 5.93c 30.66d 45.97 10.76 60.07a 9.37 84.59a 0.51c 14 6.60a 31.32c 45.57 10.60 59.16b 8.85 82.30b 0.53b OR 6 6.04c 31.40bc 44.88 11.47 58.83bc 8.60 81.26b 0.54ab 10 6.01c 31.77ab 46.09 11.15 58.59c 9.68 84.89a 0.55a 14 6.33b 31.86a 45.54 11.04 58.48c 9.73 84.81a 0.55a P-value B 0.0125 0.0006 0.3939 0.0323 0.0001 0.0001 0.0001 0.0001 OT 0.0001 0.0891 0.0001 0.8188 0.0028 0.0001 0.0001 0.0307 B × OT 0.0382 0.0011 0.6195 0.6746 0.0001 0.0915 0.0236 0.0004 SEM 0.029 0.067 0.076 0.132 0.069 0.055 0.205 0.002 B – breed, OT – oviposition time, YC – yolk colour, YP – yolk proportion, YI – yolk index, CH_Y – cholesterol concentration in yolk, AP – albumen proportion, AI – albumen index, HU – Haugh units, YAR – yolk to albumen ratio, CGS – Czech golden spotted hen, OR – Oravka hen; SEM – standard error of the mean; P-value ≤0.05 means significant effect of concrete parameter. 3.2 Hens’ oviposition regarding the time, place, and breed Values marked with different superscript letters for each parameter are significantly different Table 2 Yolk and albumen quality parameters regarding the breed and oviposition time Item Parameter Faculty of Agrobiology and Food Resources Slovak University of Agriculture in Nitra 112 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk Table 3 Percentage of laid eggs regarding the oviposition time and place in Czech golden spotted hens (%) Oviposition time (OT) Oviposition place (OP) Number of eggs (%) 14:00 – 5:59 inside 34.7ab outside 4.9b 6:00 – 9:59 inside 11b outside 2.8b 10:00 – 13:59 inside 41.9a outside 4.7b P-value OT × OP 0.0042 values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) Figure 1 Preference of oviposition place regarding the breed and percentage of laid eggs values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) 87.6a 32.7 12.4b 67.3 0 10 20 30 40 50 60 70 80 90 100 Czech hen Oravka hen (%) Oviposition place inside outside Figure 1 Preference of oviposition place regarding the breed and percentage of laid eggs values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) Figure 1 Preference of oviposition place regarding the breed and percentage of laid eggs values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) Table 4 Percentage of laid eggs regarding the oviposition time and place in Oravka hens (%) Oviposition time (OT) Oviposition place (OP) Number of eggs (%) 14:00 – 5:59 inside 9.5 outside 21.8 6:00 – 9:59 inside 9.9 outside 18.1 10:00 – 13:59 inside 13.3 outside 27.4 P-value OT × OP 0.4826 values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) Percentage of laid eggs regarding the oviposition time and place in Oravka hens (%) 3.4 External and internal egg quality regarding the breed and oviposition time Similarly, Yang et Slovak University of Agriculture in Nitra Faculty of Agrobiology and Food Resources 113 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk Table 5 Biochemical blood parameters regarding the breed Parameter Breed P-value SEM CGS OR AST (μkat l-1) 2.76 2.83 0.7242 0.093 TP (g/l) 46.45 43.47 0.3218 1.479 ALB (g l-1) 17.48 18.14 0.4671 0.444 GLU (mmol l-1) 16.47a 14.03b 0.0498 0.652 TAG (mg dl-1) 262.85b 274.22a 0.0318 24.356 CHOL (mg dl-1) 114.79a 105.65b 0.0321 4.556 HDL (mg dl-1) 168.18a 135.89b 0.0401 18.853 LDL (mg dl-1) 82.24b 89.08a 0.0461 4.529 TAG_HDL 2.40 2.18 0.7794 0.375 LDL_HDL 0.72 0.72 0.9973 0.062 Non_HDL (mg dl-1) 13.88b 20.72a 0.0313 13.995 ATI 0.90 0.87 0.7886 0.060 CGS – Czech golden spotted hen, OR – Oravka hen; SEM – Standard Error of the Mean; AST – aspartate aminotransferase, TP – total protein, ALB – albumin, GLU – glucose, TAG – triacylglycerol, CHOL – cholesterol, HDL – high-density lipoprotein cholesterol, LDL – low-density lipoprotein cholesterol, TAG_HDL – triacylglycerol high-density lipoprotein cholesterol ratio, LDL_HDL – low-density lipoprotein cholesterol high-density lipoprotein cholesterol ratio, Non_HDL – non high-density lipoprotein cholesterol, ATI – atherogenic index. Values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) CGS – Czech golden spotted hen, OR – Oravka hen; SEM – Standard Error of the Mean; AST – aspartate aminotransferase, TP – total protein, ALB – albumin, GLU – glucose, TAG – triacylglycerol, CHOL – cholesterol, HDL – high-density lipoprotein cholesterol, LDL – low-density lipoprotein cholesterol, TAG_HDL – triacylglycerol high-density lipoprotein cholesterol ratio, LDL_HDL – low-density lipoprotein cholesterol high-density lipoprotein cholesterol ratio, Non_HDL – non high-density lipoprotein cholesterol, ATI – atherogenic index. Values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) CGS – Czech golden spotted hen, OR – Oravka hen; SEM – Standard Error of the Mean; AST – aspartate aminotransferase, TP – total protein, ALB – albumin, GLU – glucose, TAG – triacylglycerol, CHOL – cholesterol, HDL – high-density lipoprotein cholesterol, LDL – low-density lipoprotein cholesterol, TAG_HDL – triacylglycerol high-density lipoprotein cholesterol ratio, LDL_HDL – low-density lipoprotein cholesterol high-density lipoprotein cholesterol ratio, Non_HDL – non high-density lipoprotein cholesterol, ATI – atherogenic index. Values marked with different superscript letters for each parameter are significantly different (P-value ≤0.05) in individuals, which proves a delay of laying eggs, which influences an internal quality (Reynard & Savory, 1999). Tůmová et al. 3.4 External and internal egg quality regarding the breed and oviposition time (2017) also found significant interaction (B × OT) for YP and AP. They observed an increase of the values with the time of oviposition in Bovans and Moravia hybrid strains. These results of HU are in accordance with results from Hrnčár et al. (2013), who found lower values in eggs from Brahma hens that were laid between 6:00 and 9:59 h. Vice versa, Tůmová & Ebeid (2005) discovered higher values of HU in the afternoon eggs. The two-way interaction between breed and oviposition time was found for ESR. In general, breed or hybrid genotype affect the ESR (Kraus & Zita, 2019). Moreover, in terms of oviposition time, Samiullah et al. (2016) observed the trend, where hens laid darker eggs in the morning. Furthermore, statistically significant interactions of breed and oviposition time were calculated for EST, ESST and ESI. These results reflect the real value of the whole eggshell and due to that are valuable (Tyler & Geake, 1961) and important in terms of cracks occurrence, because Kibala et al. (2015) found positive correlation between EST and ESST. Concerning ESI, the highest value means smaller crystals of CaCO3 and higher breaking strength (Ahmed et al., 2005). Similarly to our results, Samiullah et al. (2016) stated the reduction in the EST in eggs, which were laid later in the morning and connected these results with the time that eggs remain in the shell gland and does not necessarily mean of extra calcite. On the other hand, Tůmová & Ebeid (2005) did not observe eggshell parameters as significant in connection with oviposition time. Beside other parameters, Hrnčár et al. (2013) studied the effect oviposition on ESP in Brown Leghorn, Oravka and Brahma hens and did not found any significant effect of oviposition time on ESP with the exception of eggs from Brahma hens, which had significantly lowest value ESP when laid between 14:00 and 5:59. However, the housing was the same, differences in YC could have a link with immune response due to carotenoids, which provide these actions to support immune system (Moller et al., 2000). The oviposition time could be affected by stress 3.4 External and internal egg quality regarding the breed and oviposition time The effect of oviposition time on EW was previously studied by authors such as Samiullah et al. (2016) or Tůmová & Ledvinka (2009). However, the contrary to the results of this study, Samiullah et al. (2016) found out that the heaviest eggs are laid early in the day and Tůmová & Ledvinka (2009) stated that the heaviest eggs were laid between 14:00 at 5:59. The significant effect of oviposition time on SI and ESS was determined, which is in accordance with the study from Tůmová et al. (2017). Furthermore, YI and AI were also significantly affected by oviposition time, where higher values were observed at morning eggs (laid between 6:00 and 9:59). Tůmová & Ebeid (2005) confirmed a significant effect of oviposition time on both parameters and found the same trend for YI, but on the other hand, different for AI, where they found the highest value of AI in eggs that were laid between 10:00 and 13:59 h. al. (2013) confirmed the differences in CH_Y between breeds and added that CH_Y is dependent also on other factors, such as EW, laying intensity or age of hens. Kraus et al. (2021) also compared differences in CH_Y between CGS and OR hens and discovered the lower cholesterol content (11.06 mg g-1) in eggs from CGS hens housed on deep litter compared to eggs from OR hens (12.18 mg g-1), which is quite similar to our results (10.64 vs. 11.22 mg g-1). Consumption of eggs is generally being linked with a higher risk of cardiovascular diseases, especially because of the cholesterol content. Moreover, the problematics around the impact of cholesterol consumption is still controversial. Specifically, Shin et al. (2013), stated that there is no connection between egg consumption and cardiovascular diseases. On the other hand, Zhuang et al. (2021) concluded that intake of cholesterol is associated with higher all-cause, cardiovascular diseases, and even with cancer mortality. In our study, the CH_Y content was affected just by breed, while oviposition time had no significant effect, which is in accordance with Tůmová & Ebeid (2005). Oppositely, Abdalla & Ochi (2018) found lower CH_Y content in the morning eggs. Genotype (or breed) has a direct impact on several egg quality parameters including concentration of cholesterol in egg yolk (Rizzi & Chiericato 2010; Kraus et al., 2021). 3.5 Hens’ oviposition regarding the time, place and breed Results of eggs laid into nests are important thanks to their connection with better hatchability of chickens (Keeling, 2004) or higher status of food safety, because the most of bacteria comes from the litter on the floor (Brandl et al., 2014). Basically, it can be expected that the most of eggs will be laid during the morning. However, some delay in the timing of oviposition can occur, when stress occurs (Reynard & Savory, 1999) or when there is not enough space to lay eggs synchronously. This can also end in adaptation of hens and laying later or choosing a different place to oviposit (Villanueva et al., 2017). These authors also found differences between brown and white laying hybrids, which varied in the live weight as well as our hens (OR hens are heavier than CGS hens). Another factor, which can affect the preference of place to oviposit, is a natural tendency of hens to nest in groups to avoid predators (Riber, 2012). Slovak University of Agriculture in Nitra Slovak University of Agriculture in Nitra Faculty of Agrobiology and Food Resources 114 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk https://doi.org/10.1080/00071668.2010.528750 Belew, A. K. et al. (2016). The state of conservation of animal genetic resources in developing countries: A review. International Journal of Pharma Medicine and Biological Sciences, 5(1), 58–66. Belew, A. K. et al. (2016). The state of conservation of animal genetic resources in developing countries: A review. International Journal of Pharma Medicine and Biological Sciences, 5(1), 58–66. Blaha, M.J. et al. (2008). The importance of non–HDL cholesterol reporting in lipid management. Journal of Clinical Lipidology, 2(4), 267–273. https://doi.org/10.1016/j.jacl.2008.06.013 Brandl, H. B. et al. (2014). Composition of bacterial assemblages in different components of reed warbler nests and a possible role of egg incubation in pathogen regulation. PloS One, 9(12), 114861e. https://doi.org/10.1111/nure.12107 Keeling, L. J. (2004). Welfare of the laying hen. CABI Publishing. g, ( ) y g g Kibala, L. et al. (2015). Ultrasonic eggshell thickness measurement for selection of layers. Poultry Science, 94(10), 2360–2363. https://doi.org/10.3382/ps/pev254 Kibala, L. et al. (2015). Ultrasonic eggshell thickness measurement for selection of layers. Poultry Science, 94(10), 2360–2363. https://doi.org/10.3382/ps/pev254 Koronowicz, A. A. et al. (2016). Dietary conjugated linoleic acid affects blood parameters, liver morphology and expression of selected hepatic genes in laying hens. British Poultry Science, 57(5), 663–673. doi: 10.1080/00071668.2016.1192280 https://doi.org/10.1371/journal.pone.0114861 The effects of breed, oviposition time and their interaction on egg quality were determined as significant in most of the evaluated quality parameters. The preference of CGS hens to lay eggs inside the nest was also confirmed. Based on this finding, we can assume that CGS hens showed less risky nest behaviour regarding the egg quality. It can be also summarised that egg quality varied independently with time of oviposition. The effect of breed on blood serum parameters was calculated as statistically significant especially in cholesterol connected parameters. The blood serum data may help to acquire more complete information about these native breeds. The uniqueness and originality of this study is highlighted using native breeds, the Czech golden spotted and the Oravka hens, which are insufficiently explored from this point of view. Fernandez, M. L., & Webb, D. (2008). The LDL to HDL cholesterol ratio as a valuable tool to evaluate coronary heart disease risk. Journal of the American College of Nutrition, 27(1), 1–5. https://doi.org/10.1080/07315724.2008.10719668 Hrnčár, C. et al. (2013). The effect of oviposition time on egg quality parameters in Brown Leghorn, Oravka and Brahma hens. Animal Science and Biotechnologies, 46(1), 53–57. Hrnčár, C. et al. (2013). The effect of oviposition time on egg quality parameters in Brown Leghorn, Oravka and Brahma hens. Animal Science and Biotechnologies, 46(1), 53–57. Iannotti, L. L. et al. (2014). Eggs: the uncracked potential for improving maternal and young child nutrition among the world‘s poor. Nutrition Reviews, 72(6), 355–368. Iannotti, L. L. et al. (2014). Eggs: the uncracked potential for improving maternal and young child nutrition among the world‘s poor. Nutrition Reviews, 72(6), 355–368. References Abdalla M. A., & Ochi, E. B. (2018). Effect of laying hen’s age and oviposition time on egg cholesterol contents. Science Letters, 6(1), 42–46. GLU and TAG are energy sources, where GLU is considered as the main energy source. Significant differences between CGS and OR hens may be simply caused by different body constitution or different physical activity of particular breed (Kraus et al., 2021). Regarding the cholesterol, Andrews et al. (1968) stated that its origin in eggs is in blood serum. Blood serum cholesterol was influenced by age of hens and housing system in study of Kraus et al. (2021), who proved the negative effect of cage housing of native breeds with an impact on blood and yolk cholesterol. Zita et al. (2018) calculated the correlation between concentration of cholesterol in egg yolk and in blood serum, but the result was non-significant. The role of cholesterol is also important because it is a precursor of steroid hormones (Kraus et al., 2021). Furthermore, the cholesterol fractions (LDL and HDL) can be used for determination of the onset of CVD (Fernandez and Webb, 2008). Non-HDL was previously confirmed to be predictive of CVD as well (Packard & Saito, 2004) and is considered as a superior predictor of CVD compared to LDL (Blaha et al., 2008). Ahmed A. M. H. et al. (2005). Changes in eggshell mechanical properties, crystallographic texture and in matrix proteins induced by moult in hens. British Poultry Science, 46(3), 268–279. https://doi.org/10.1080/00071660500065425 Andrews Jr J. W. et al. (1968). Cholesterol metabolism in the laying fowl. American Journal of Physiology, 214(5), 1078–1083. https://doi.org/10.1152/ajplegacy.1968.214.5.1078 Andrews Jr J. W. et al. (1968). Cholesterol metabolism in the laying fowl. American Journal of Physiology, 214(5), 1078–1083. https://doi.org/10.1152/ajplegacy.1968.214.5.1078 Begli, H. E. et al. (2010). Heritability, genetic and phenotypic correlations of egg quality traits in Iranian native fowl. British Poultry Science, 51(6), 740–744. Acknowledgements Kraus, A., & Zita, L. (2019). The Effect of Age and Genotype on Quality of Eggs in Brown EggLaying Hybrids. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 67(2), 407– 414. https://doi.org/10.11118/actaun201967020407 Kraus, A., & Zita, L. (2019). The Effect of Age and Genotype on Quality of Eggs in Brown EggLaying Hybrids. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 67(2), 407– 414. https://doi.org/10.11118/actaun201967020407 This study was funded by an “S” grant from the Ministry of Education, Youth and Sports of the Czech Republic and grant project no. SV21-6-21320. This study was carried out with the assistance of the Demonstration and Experimental Centre of the Faculty of Agrobiology, Food and Natural Resources, Czech University of Life Sciences Prague. Last, but not least, we would like to thank to Mr. Richard Hardy for the English language corrections. Kraus, A. et al. (2021). Determination of selected biochemical parameters in blood serum and egg quality of Czech and Slovak native hens depending on the housing system and hen age. Poultry Science, 100(2), 1142–1153. https://doi.org/10.1016/j.psj.2020.10.039 Kraus, A. et al. (2021). Determination of selected biochemical parameters in blood serum and egg quality of Czech and Slovak native hens depending on the housing system and hen age. Poultry Science, 100(2), 1142–1153. https://doi org/10 1016/j psj 2020 10 039 https://doi.org/10.1093/ps/86.4.714 Samiullah, S. et al. (2016). Oviposition time, flock age, and egg position in clutch in relation to brown eggshell color in laying hens. Poultry Science, 95(9), 2052–2057. https://doi.org/10.3382/ps/pew197 Samiullah, S. et al. (2016). Oviposition time, flock age, and egg position in clutch in relation to brown eggshell color in laying hens. Poultry Science, 95(9), 2052–2057. https://doi org/10 3382/ps/pew197 https://doi.org/10.1016/j.applanim.2012.01.009 Rizzi, C., & Chiericato, G. M. (2010). Chemical composition of meat and egg yolk of hybrid and Italian breed hens reared using an organic production system. Poultry Science, 89(6), 1239–1251. https://doi.org/10.3382/ps.2008-00045 Zaheer, K. (2015). An updated review on chicken eggs: production, consumption, management aspects and nutritional benefits to human health. Food and Nutrition Sciences, 6(13), 1208–1220. https://doi.org/10.4236/fns.2015.613127 Salma, U. et al. (2007). Effect of dietary Rhodobacter capsulatus on egg-yolk cholesterol and laying hen performance. Poultry Science, 86(4), 714–719. Zhuang, P. et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. https://doi.org/10.1080/00071669986945 Riber, A. B. (2012). Gregarious nesting – An anti-predator response in laying hens. Applied Animal Behaviour Science, 138(1–2), 70–78. Yang, P. K. et al. (2013). Deposition rule of yolk cholesterol in two different breeds of laying hens. Genetics and Molecular Research, 12(4), 5786–5792. https://doi.org/10.4238/2013 https://doi.org/10.1016/j.applanim.2012.01.009 Faculty of Agrobiology and Food Resources https://doi.org/10.1371/journal.pmed.1003508 Zita, L. et al. (2018). Effect of housing system on egg quality and the concentration of cholesterol in egg yolk and blood of hens of native resources of the Czech Republic and Slovakia. Journal of Applied Poultry Research, 27(3), 380–388. https://doi.org/10.3382/japr/pfy009 Zita, L. et al. (2018). Effect of housing system on egg quality and the concentration of cholesterol in egg yolk and blood of hens of native resources of the Czech Republic and Slovakia. Journal of Applied Poultry Research, 27(3), 380–388. htt //d i /10 3382/j / f 009 Shaker, A. S. et al. (2019). Effect of hen oviposition time on some egg characteristics. Journal of Animal and Poultry Production, 10(6), 171–174. 46–51. https://doi.org/10.1016/j.tifs.2017.10.022 Tůmová, E. et al. (2017). Differences in oviposition and egg quality of various genotypes of laying hens. Czech Journal of Animal Science, 62(9), 377–383. Moller, A. P. et al. (2000). Carotenoid-dependent signals: indicators of foraging efficiency, immunocompetence or detoxification ability? Avian and Poultry Biology Reviews, 11(3), 137–160. https://doi.org/10.17221/22/2017-CJAS Tyler, C., & Geake, F. H. (1961). Studies on egg shells. XV. – critical appraisal of various methods of assessing shell thickness. Journal of the Science of Food Agriculture, 12(4), 281–289. Oliveira, J. L. et al. (2019). Effects of litter floor access and inclusion of experienced hens in aviary housing on floor eggs, litter condition, air quality, and hen welfare. Poultry Science, 98(4), 1664–1677. https://doi.org/10.3382/ps/pey525 https://doi.org/10.1002/jsfa.2740120404 Packard, C. J., & Saito, Y. (2004). Non-HDL cholesterol as a measure of atherosclerotic risk. Journal of Atherosclerosis and Packard, C. J., & Saito, Y. (2004). Non-HDL cholesterol as a measure of atherosclerotic risk. Journal of Atherosclerosis and Thrombosis, 11(1), 6–14. https://doi.org/10.5551/jat.11.6 van Deventer, H. E. et al. (2011). Non–HDL cholesterol shows improved accuracy for cardiovascular risk score classification compared to direct or calculated LDL cholesterol in a dyslipidemic population. Clinical Chemistry, 57(3), 490–501. https://doi.org/10.1373/clinchem.2010.154773 Pavlík, A. et al. (2007). Effects of housing systems on biochemical indicators of blood plasma in laying hens. Acta Veterinaria Brno, 76(3), 339–347. Villanueva, S. et al. (2017). Nest use and patterns of egg laying and damage by 4 strains of laying hens in an aviary system. Poultry Science, 96(9), 3011–3020. Reynard, M., & Savory, C. J. (1999). Stress-induced oviposition delays in laying hens: duration and consequences for eggshell quality. British Poultry Science, 40(5), 585–591. https://doi.org/10.3382/ps/pex104 https://doi.org/10.3382/ps/pex104 Vlčková, J. et al. (2019). Changes in the quality of eggs during storage depending on the housing system and the age of hens. Poultry Science, 98(11), 6187–6193. https://doi.org/10.3382/ps/pez401 Slovak University of Agriculture in Nitra https://doi.org/10.1016/j.psj.2020.10.039 Krawczyk, J. et al. (2011). Effect of housing system on cholesterol, vitamin and fatty acid content of yolk and physical Slovak University of Agriculture in Nitra Faculty of Agrobiology and Food Resources 115 Acta fytotechn zootechn, 25, 2022(2): 109–116 http://www.acta.fapz.uniag.sk characteristics of eggs from Polish native hens. Archiv für Geflügelkunde, 75(3), 151–157. characteristics of eggs from Polish native hens. Archiv für Geflügelkunde, 75(3), 151–157. Tang, S. G. H. et al. (2015). Chemical compositions of egg yolks and egg quality of laying hens fed prebiotic, probiotic, and synbiotic diets. Journal of Food Science, 80(8), 1686–1695. https://doi.org/10.1111/1750-3841.12947 Krunt, O. et al. (2021). Guinea fowl (Numida meleagris) eggs and free range housing: A convenient alternative to laying hens’ eggs in terms of food safety? Poultry Science, 100(4), 101006. doi: 10.1016/j.psj.2021.01.029 Tůmová, E., & Ebeid, T. (2005). Effect of time of oviposition on egg quality characteristics in cages and in a litter housing system. Czech Journal of Animal Science, 50(3), 129–134. Lesnierowski, G., & Stangierski, J. (2018). What‘s new in chicken egg research and technology for human health promotion? A review. Trends in Food Science & Technology, 71, 46–51 https://doi org/10 1016/j tifs 2017 10 022 Tůmová, E., & Ledvinka, Z. (2009). The effect of time of oviposition and age on egg weight, egg components weight and eggshell quality. Archiv für Geflügelkunde, 73(2), 110–115. https://doi.org/10.21608/jappmu.2019.48272 Shin, J. Y. et al. (2013). Egg consumption in relation to risk of cardiovascular disease and diabetes: a systematic review and meta-analysis. The American Journal of Clinical Nutrition, 98(1), 146–159. https://doi.org/10.3945/ajcn.112.051318 Sparks, N. H. C. (2006). The hen‘s egg-is its role in human nutrition changing? Worlds Poultry Science Journal, 62(2), 308– 315. https://doi.org/10.1079/WPS200599 Faculty of Agrobiology and Food Resources Slovak University of Agriculture in Nitra 116
https://openalex.org/W2622193041	https://cgspace.cgiar.org/bitstream/10568/82760/1/U17ArtNyineTraitInthomNodev.pdf	English	null	Trait variation and genetic diversity in a banana genomic selection training population	PloS one	2,017	cc-by	13,100	RESEARCH ARTICLE Moses Nyine1,2,3, Brigitte Uwimana2, Rony Swennen2,4,5,6, Michael Batte2, Allan Brown6, Pavla Christelova´3, Eva Hřibova´3, Jim Lorenzen2¤, Jaroslav Dolezˇel3 Moses Nyine1,2,3, Brigitte Uwimana2, Rony Swennen2,4,5,6, Michael Batte2, Allan Brown6, Pavla Christelova´3, Eva Hřibova´3, Jim Lorenzen2¤, Jaroslav Dolezˇel3 1 Faculty of Science, Palacky´ University, Olomouc, Czech Republic, 2 International Institute of Tropical Agriculture, Kampala, Uganda, 3 Institute of Experimental Botany, Centre of the Region Hana´ for Biotechnological and Agricultural Research, Olomouc, Czech Republic, 4 Laboratory of Tropical Crop Improvement, Division of Crop Biotechnics, Katholieke Universiteit Leuven, Leuven, Belgium, 5 Bioversity International, Leuven, Belgium, 6 International Institute of Tropical Agriculture, Arusha, Tanzania a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ¤ Current address: Bill & Melinda Gates Foundation, Seattle, Washington, United States of America * B.Uwimana@cgiar.org (BU); dolezel@ueb.cas.cz (JD) ¤ Current address: Bill & Melinda Gates Foundation, Seattle, Washington, United States of America * B.Uwimana@cgiar.org (BU); dolezel@ueb.cas.cz (JD) Abstract Banana (Musa spp.) is an important crop in the African Great Lakes region in terms of income and food security, with the highest per capita consumption worldwide. Pests, dis- eases and climate change hamper sustainable production of bananas. New breeding tools with increased crossbreeding efficiency are being investigated to breed for resistant, high yielding hybrids of East African Highland banana (EAHB). These include genomic selec- tion (GS), which will benefit breeding through increased genetic gain per unit time. Under- standing trait variation and the correlation among economically important traits is an essential first step in the development and selection of suitable GS models for banana. In this study, we tested the hypothesis that trait variations in bananas are not affected by cross combination, cycle, field management and their interaction with genotype. A training population created using EAHB breeding material and its progeny was phenotyped in two contrasting conditions. A high level of correlation among vegetative and yield related traits was observed. Therefore, genomic selection models could be developed for traits that are easily measured. It is likely that the predictive ability of traits that are difficult to phenotype will be similar to less difficult traits they are highly correlated with. Genotype response to cycle and field management practices varied greatly with respect to traits. Yield related traits accounted for 31–35% of principal component variation under low and high input field management conditions. Resistance to Black Sigatoka was stable across cycles but varied under different field management depending on the genotype. The best cross com- bination was 1201K-1xSH3217 based on selection response (R) of hybrids. Genotyping using simple sequence repeat (SSR) markers revealed that the training population was genetically diverse, reflecting a complex pedigree background, which was mostly influ- enced by the male parents. Introduction (Investment ID: OPP1093845) to the IITA banana breeding program. The genotyping work at the Institute of Experimental Botany was supported by the Ministry of Education, Youth and Sports of the Czech Republic (grant LO1204 from the National Program of Sustainability I). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. East Africa is considered a secondary center of banana genetic diversity. Uganda in particu- lar is a home to over eighty cultivars of East African Highland banana (EAHB) commonly divided into cooking and beer types [1]. The crop greatly contributes to the income and food security of many smallholder farmers in the region. The significance of the crop in the region is reflected in the per capita consumption that ranges between 250kg and 600kg with an aver- age of 400kg in Uganda [2]. Over 85% of the production is consumed locally due to high demand [3, 4]. Sustainable production of bananas is a challenge because of disease, insect and nematode pressure. This is worsened by abiotic stress arising through factors associated with climate change [5]. Yield reductions in EAHB are caused by pests such as root burrow- ing nematodes especially Radopholus similis and banana weevil (Cosmopolites sordidus). Black Leaf Streak (Black Sigatoka), a fungal disease caused by Mycosphaerella fijiensis reduces the photosynthetic area of the plant, which decreases yield. Banana bacterial wilt caused by Xanthomonas campestris pv. musacearum causes 100% yield loss when the banana is attacked [6±8]. Variation in rainfall patterns impacts banana production by causing drought stress because most farmers in the region rely on rain for agricultural production. Although pheno- typic variation is observed in EAHB, their genetic variation is low [9, 10] making them all susceptible to biotic and abiotic stress. Adaptation of cultivated banana varieties to changing environment is limited because while some are capable of sexual reproduction, they are all propagated clonally. Competing interests: The authors have declared that no competing interests exist. Competing interests: The authors have declared that no competing interests exist. In order to meet the food demand for the growing population, breeding for resistance and high yielding varieties is considered to be the most sustainable solution to address banana production constraints [11, 12]. Phenotyping and banana genomic selection PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 OPEN ACCESS Citation: Nyine M, Uwimana B, Swennen R, Batte M, Brown A, Christelova´ P, et al. (2017) Trait variation and genetic diversity in a banana genomic selection training population. PLoS ONE 12(6): e0178734. https://doi.org/10.1371/journal. pone.0178734 Editor: Marco Buenrostro-Nava, Universidad de Colima Facultad de Ciencias Biologicas y Agropecuarias, MEXICO Editor: Marco Buenrostro-Nava, Universidad de Colima Facultad de Ciencias Biologicas y Agropecuarias, MEXICO Editor: Marco Buenrostro-Nava, Universidad de Colima Facultad de Ciencias Biologicas y Agropecuarias, MEXICO Received: October 11, 2016 Accepted: May 18, 2017 Published: June 6, 2017 Copyright: © 2017 Nyine et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: © 2017 Nyine et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its supporting Information files. Funding: This work was supported by several donors through their contributions to the CGIAR Fund (http://www.cgiar.org/who-we-are/cgiar- fund/fund-donors-2/), and in particular to the CGIAR Research Program on Roots, Tubers and Bananas. Additional funding was provided by the Bill & Melinda Gates Foundation (BMGF) 1 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Introduction Unlike other crops, banana breeding is complicated by the polyploid nature of the crop characterized by abnormal meiosis in the cultivated trip- loid varieties that results in reduced fertility or complete sterility [13±15]. Crossing cultivated varieties with resistant wild diploids is possible, but a majority of the generated hybrids are inferior due to linkage drag of unfavorable genes from the wild diploids. However, when tetraploids are obtained, further improvement is possible because they are both male and female fertile (Fig 1). Incorporating resistance while maintaining the unique attributes such as fruit colour, aroma, texture and taste in existing varieties is a big challenge to banana breeders that calls for dedicated effort and careful choice of cross combinations. Crossbreed- ing is labour-intensive, costly and time consuming. In the last two decades, some success has been registered with new hybrids released to farmers while others are in the advanced stages of evaluation [16]. In order to keep up with the pace at which environmental changes occur and the demand for new varieties that are productive and of good quality, new breeding strategies should be employed to increase breeding efficiency and reduce the lengthy selec- tion period [3]. Marker assisted selection (MAS) has been implemented in many animal and crop breed- ing programs. The success of MAS greatly depends on the genetic architecture of traits being improved. To date MAS has not been effectively deployed in banana breeding. The possible reasons are polyploidy, important economic and agronomic traits may be con- trolled by many quantitative trait loci (QTL), each with a small additive effect, and the lack of saturated linkage maps for QTL mapping. It is believed that the application of genomic selection (GS) will improve the efficiency of crossbreeding programs especially for crops with long breeding and selection cycle [17, 18] like banana. GS is a form of MAS where selection is based on the genomic estimated breeding values (GEBV) of superior individuals in the population as determined by a statistical model [19±21]. This technique is well estab- lished in animal breeding [22, 23]. In plants, GS has been tested in maize and wheat [24], PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 2 / 23 Phenotyping and banana genomic selection Fig 1. Conventional banana breeding starts with crossing 3x inferior and parthenocarpic landrace varieties (A) with a wild, seeded 2x accession (B). Introduction 4x resulting from this cross (C) are selected and crossed with improved 2x hybrids (D). The resulting secondary 3x (E) are selected and evaluated as potential improved varieties. This process takes up to 15 years. Fig 1. Conventional banana breeding starts with crossing 3x inferior and parthenocarpic landrace varieties (A) with a wild, seeded 2x accession (B). 4x resulting from this cross (C) are selected and crossed with improved 2x hybrids (D). The resulting secondary 3x (E) are selected and evaluated as potential improved varieties. This process takes up to 15 years. https://doi.org/10.1371/journal.pone.0178734.g001 Fig 1. Conventional banana breeding starts with crossing 3x inferior and parthenocarpic landrace varieties (A) with a wild, seeded 2x accession (B). 4x resulting from this cross (C) are selected and crossed with improved 2x hybrids (D). The resulting secondary 3x (E) are selected and evaluated as potential improved varieties. This process takes up to 15 years. https://doi.org/10.1371/journal.pone.0178734.g001 https://doi.org/10.1371/journal.pone.0178734.g001 white spruce [25], rice [26] and cassava [27]. However, in bananas GS is in its infancy. Given that new varieties are selected based on a combination of traits, a selection index of GEBV in bananas is necessary. GS studies have reported varying accuracies in prediction (predictive ability of GS models) and this has been attributed to differences in trait heritability, number of markers, training population size and genotype x environment interaction [24]. Bananas as perennial plants suf- fer the consequences of nutrient deficiency and soil moisture variation across seasons and locations depending on field management practices. Breeding generates genotypes from many crosses that are genetically different and respond to growth environment differently and this could affect the accuracy of GS. Therefore, understanding trait variation and the correlation between different traits is essential to guide the development and selection of suitable GS mod- els for banana breeding. In this study we tested the hypothesis that trait variations in bananas are not affected by cross combination, cycle, field management and their interaction with genotype. For this, a training population created using EAHB breeding material and its prog- eny was phenotyped in two contrasting conditions. Genetic diversity of the training popula- tion was assessed using simple sequence repeat (SSR) markers. 3 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Phenotyping and banana genomic selection Traits The yield-related traits scored included: days to fruit maturity (DFM) that is, days between flowering and harvesting, bunch weight at full maturity (BWT), number of hands (cluster) (NH) and number of fruit fingers (NF), fruit length (FL), fruit circumference (FC), fruit diam- eter (FRD), pulp diameter (PLD) and peel thickness (PED), where PED = (FRDÐPLD)/2. The vegetative (growth) traits included: number of standing leaves at flowering (NSLF), youngest leaf spotted with Black Sigatoka at flowering (YLSF), index of non-spotted leaves at flowering (INSL), height of tallest sucker at harvesting (HTSH), plant height at flowering (PHF), plant girth at 100 cm from soil surface (PG), height of tallest sucker at flowering (HTSF), total num- ber of suckers at flowering (TS), number of standing leaves at harvesting (NSLH) and youngest leaf spotted with black sigatoka at harvesting (YLSH). g g Total number of suckers (TS) was recorded at flowering in cycle 1 only after which each mat was left with a maximum of three plants and these included the flowered plant, follower sucker and the sucker produced by follower sucker if present. A Vernier caliper was used to measure FRD and PLD. Fruit related traits such as FL, FC, FRD and PLD were recorded from the middle finger of the second hand on the bunch. Measurements for FC, FRD and PLD were recorded midway the length of the finger. However, to measure FRD and PLD, a cross- section of the fruit was made to expose the pulp. The INSL was calculated from the formula, INSL = 100(YLSF-1)/NSLF [30]. This formula should give percentage values ranging from 0±100% to represent completely susceptible (0%) and completely resistant (100%). In order to get 100% INSL for completely resistant genotypes, the YLSF was scored as NSLF +1 thus INSL = 100((NSLF+1)-1)/NSLF or INSL = 100NSLF/NSLF Materials and methods Plant population Data were collected at the International Institute of Tropical Agriculture, Uganda from a banana genomic selection (GS) training population between 2013 and 2016. The institute is located at Namulonge research station, 0.53ÊN 32.58ÊE, 1150 m above sea level with rainfall of about 1200 mm/y split into two rainy seasons, March-June and September-December and an average annual temperature of 22ÊC.The GS population consisted of 307 genotypes that included diploid (11%), triploid (85%) and tetraploid (4%) plants (S1 Table). The ploidy level of the genotypes was determined using flow cytometry [28, 29]. The core breeding lines (parents) accounted for 12% of the entire population. Two fields were established with each genotype replicated three times in a completely randomized design. Suckers were used as planting materials and before planting, 20kg of farmyard manure was applied in each hole. One field (GS1) was managed without mulching, additional manure nor inorganic fertilizer (low input). The second field (GS2) was mulched twice a year. Six months after planting, 480 g of NPK (25:5:5) fertilizer was added and the same amount was added to each mat per year (high input). Phenotyping and banana genomic selection genotype and cross combination (family) in every cycle, and field and these were used in corre- lation analysis and principal component analysis (PCA). Correlation analysis and test of significance for the correlations between traits were done using library (Hmisc) and Student's t-test based on cycle 2 data for cross combinations. Coeffi- cient of determination (R2) was calculated as a square of correlation coefficient between cycle 1 and 2 data. To understand the structure of the population and how different traits influenced that structure, principal component analysis was done using PCA function provided in the library (FactoMineR). Traits (dependent variable), cross combinations and individual geno- types were projected on the first two components (Dim1 and Dim2). Sources of trait variation were assessed using unbalanced analysis of variance (ANOVA) based on cycle 1 and 2 data. Linear models were constructed for each trait in respect to each cycle, field management practice and their interaction with genotype as model_fit = lm(trait response~clone+cycle+field+clone:field+clone:cycle, data = mydata) where lm = linear model function. Type III SS ANOVA tables were generated using Anova function provided in the library(car) as result = Anova(model_fit, singular.ok = TRUE, type = ªIIIº). In cases where no significant interactions were observed between two independent variables and where one explanatory variable was not significant, then type II or type I SS ANOVA was used for further investigation. Selection differential (S) and response to selection (R) were used to compare performance of parental cross combinations [31]. S and R were calculated as, S = PÐG and R = HÐG, where P = average performance of a pair of parents, G is the average performance of all paren- tal lines in the training population and H is the average performance of all hybrid that shared same parental pair. Only cross combinations that had at least five hybrids were compared across all traits using combined data from the two fields. Genetic diversity Genetic diversity of the training population was assessed using simple sequence repeat (SSR) markers. Cigar leaf samples were collected from the training population in Uganda and shipped to the Institute of Experimental Botany, Olomouc, Czech Republic under cold chain. Samples were lyophilized prior to DNA extraction. DNA from lyophilized samples was extracted using NucleoSpin Plant II kit, Macherey-Nagel, Germany, following the manufactur- er's instructions. The concentration and quality of DNA was assessed by NanoDrop ND-1000 spectrophotometer. Nineteen informative Musa SSR primers were used to genotype the GS training population. The list of primers used, polymerase chain reaction (PCR) conditions, and fragment analysis procedure were adopted from ChristelovaÂet al. [32]. Two independent rounds of PCR were performed on each sample. The concordance of alleles from each sample were inspected and scored manually in GeneMarker v1.75 (Softge- netics, State College, PA, USA). A third round of PCR was performed only for samples that showed incongruity with the two reactions. Alleles were scored as dominant markers for pres- ence and absence (1/0). Data were imported in R and squared Euclidean distances were gener- ated using the function dist provided in the library(ape). Clustering was done with function hclust based on ward.D method [33, 34]. Polymorphism information content of each marker was estimated by PowerMarker v3.25 software [35]. Data analysis All analyses were performed in R, open source statistical software from www.r-project.org. A combination of Shapiro-Wilk test, boxplots, standard deviations and histograms were used to check for normality and outliers in the data and where necessary the outliers were removed before further analysis. Total number of suckers and bunch weight were transformed by square root. Using the aggregate function from library (plyr), trait means were calculated for every 4 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Principal component analysis (PCA) Principal component analysis showed that in both fields, the yield (fruit) traits contributed to the first component (Dim 1) while the vegetative (growth) traits contributed to the second component (Dim 2) (Fig 2A and 2B). Among the vegetative traits, PHF and PG contributed to Dim 1. Dim 1 accounted for 31.07% of variation in GS1 and 35.86% in GS2. Dim 2 accounted for 21.89% of variation in GS1 and 15.40% in GS2. The traits with the highest negative loading on Dim 1 included FC, FRD and PLD for GS1 while for GS2 it was FC, FRD, PLD and FL. In both GS1 and GS2, the traits with the highest positive loading on Dim 2 were NSLF, YLSF, INSL and NSLH. Both DFM and TS had the least contribution to the two components with completely different orientation in GS1 and GS2. Generally, in both fields the two compo- nents accounted for 50% of the variation observed in the genotype cross combinations (Fig 3A and 3B). For individual genotypes, a similar trend was observed with Dim 1 and Dim 2 accounting for 31.43% and 19.11% of total trait variation, respectively (Fig 4A). Projection of the individ- ual factors (genotypes) on the two components did not reveal any distinct population structure (Fig 4B). The same trend was observed when individual cross combinations were projected on the two components. However, in GS1 cross combinations C35 (917K-2 x Kokopo), C28 (8817S-1 x 917K-2) and C52 (SH2095 x SH2766) and in GS2 cross combinations C35 (917K-2 x Kokopo), C22 (365K-1 x 660K-1) and C29 (8817S-1 x 917k-2) were distinct and clearly sepa- rated out from the others (Fig 3a and 3b). When the data were re-examined, genotypes from cross C35 had the least average scores on the yield traits while cross C22, C29 and C52 had the highest average scores on the yield traits. All the four planes of the two components were rep- resented in the population. Based on Black Sigatoka resistance and fruit filling (indicated by FRD), four main groups were represented in the population: (i) genotypes with high INSL and good fruit filling, (ii) high INSL with poor fruit filling, (iii) low INSL with good fruit filling and (iv) low INSL with poor fruit filling. Correlation of traits Significant correlations were observed among and between growth and yield traits (Tables 1 and 2). PHF had significant positive correlation with PG followed by HTSF. PG positively cor- related with BWT, NF and HTSF in that respective order. The traits associated with Black Siga- toka resistance (NSLF, YLSF and INSL) also correlated significantly to each other. However, they had significant negative correlations with fruit traits such as FC, FRD and PLD. A positive and significant correlation was observed between BWT and all fruit traits (NH, NF, FL, FC, FRD, PLD), which were similarly significantly and positively correlated to each other. Under conditions of low input field management (GS1), TS, NSLH and NF were not significantly cor- related with other traits while under high input field management (GS2), it was INSL, DFM and HTSH that did not have significant correlation with other traits. In both fields, the highest positive correlations were observed among the yield traits. In this population, absolute apical dominance was not observed as all genotypes had at least one sucker at the time of flowering. However, sucker regulation varied among genotypes with a range of 1±25 suckers per plant. Results During data analysis, some genotypes were excluded for some traits due to missing data or extreme outliers. The outliers were mainly recorded on plants that were infected with banana PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 5 / 23 Phenotyping and banana genomic selection Xanthomonas wilt before full maturity, plants that snapped due to weevil damage and prema- ture breaking of the peduncle due to windstorm. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Principal component analysis (PCA) On average the observed INSL and FRD for the genotypes in the four groups were as follows: (i) 78.1% and 3.0cm, (ii) 80.1% and 1.4cm, (iii) 66.8% and 3.1cm, and (iv) 67.1% and 1.4cm, respectively. Genotypes projected on Dim 2 had high average scores on PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 6 / 23 Phenotyping and banana genomic selection Table 1. Pearson's correlation coefficients of traits under low input field management (GS1). PHF PG NSLF YLSF HTSF TS INSL DFM NSLH BWT NH NF FL FC FRD PLD PED PHF PG 0.807* NSLF 0.108 0.254 YLSF 0.083 0.181 0.926* HTSF 0.373 0.36 0.319 0.364 TS −0.229 −0.329 0.022 0.116 0.3 INSL −0.001 0.021 0.579* 0.834* 0.326 0.24 DFM 0.086 0.133 0.282 0.277 0.109 0.038 0.231 NSLH 0.042 0.078 0.386 0.352 0.363 0.034 0.194 -0.356 BWT 0.346 0.554* −0.083 −0.122 0.213 0.02 −0.133 0.152 −0.22 NH 0.372 0.426 0.19 0.166 0.087 0.041 0.119 0.191 −0.021 0.411 NF 0.412 0.512* 0.226 0.195 0.151 −0.032 0.126 0.221 0.053 0.4 0.878* FL 0.2 0.411 −0.077 −0.113 0.057 −0.042 −0.123 0.173 −0.266 0.855* 0.168 0.169 FC 0.191 0.375 −0.284 −0.338 0.025 −0.097 −0.323 0.005 −0.254 0.807* 0.019 0.008 0.856* FRD 0.206 0.359 −0.357 −0.415 0.022 −0.11 −0.395 −0.055 −0.258 0.782* 0.017 0.02 0.82* 0.987* PLD 0.192 0.333 −0.379 −0.432 −0.004 −0.133 −0.41 −0.099 −0.199 0.717* 0.048 0.057 0.709* 0.9* 0.919* PED −0.26 0.108 0.143 0.024 0.007 −0.078 −0.114 0.108 −0.013 0.225 −0.182 −0.118 0.359 0.293 0.272 0.179 https://doi.org/10.1371/journal.pone.0178734.t001 Table 1. Pearson's correlation coefficients of traits under low input field management (GS1). PHF PG NSLF YLSF HTSF TS INSL DFM NSLH BWT NH NF FL FC FRD PLD PED PHF PG 0.807* NSLF 0.108 0.254 YLSF 0.083 0.181 0.926* HTSF 0.373 0.36 0.319 0.364 TS −0.229 −0.329 0.022 0.116 0.3 INSL −0.001 0.021 0.579* 0.834* 0.326 0.24 DFM 0.086 0.133 0.282 0.277 0.109 0.038 0.231 NSLH 0.042 0.078 0.386 0.352 0.363 0.034 0.194 -0.356 BWT 0.346 0.554* −0.083 −0.122 0.213 0.02 −0.133 0.152 −0.22 NH 0.372 0.426 0.19 0.166 0.087 0.041 0.119 0.191 −0.021 0.411 NF 0.412 0.512* 0.226 0.195 0.151 −0.032 0.126 0.221 0.053 0.4 0.878* FL 0.2 0.411 −0.077 −0.113 0.057 −0.042 −0.123 0.173 −0.266 0.855* 0.168 0.169 FC 0.191 0.375 −0.284 −0.338 0.025 −0.097 −0.323 0.005 −0.254 0.807* 0.019 0.008 0.856* FRD 0.206 0.359 −0.357 −0.415 0.022 −0.11 −0.395 −0.055 −0.258 0.782* 0.017 0.02 0.82* 0.987* PLD 0.192 0.333 −0.379 −0.432 −0.004 −0.133 −0.41 −0.099 −0.199 0.717* 0.048 0.057 0.709* 0.9* 0.919* PED −0.26 0.108 0.143 0.024 0.007 −0.078 −0.114 0.108 −0.013 0.225 −0.182 −0.118 0.359 0.293 0.272 0.179 https://doi org/10 1371/journal pone 0178734 t001 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 7 / 23 Phenotyping and banana genomic selection Table 2. Pearson's correlation coefficients of traits under high input field management (GS2). PHF PG NSLF YLSF HTSF TS INSL DFM NSLH BWT NH NF FL FC FRD PLD PED PHF PG 0.774* NSLF −0.422 −0.257 YLSF −0.197 −0.024 0.75* HTSF 0.702* 0.563* −0.357 −0.251 TS 0.358 0.224 −0.092 0.062 0.45* INSL 0.213 0.272 −0.128 0.548* 0.084 0.197 DFM −0.007 0.006 −0.026 0.063 −0.218 −0.02 0.152 NSLH −0.149 −0.077 0.619* 0.533* −0.222 −0.156 0.002 −0.194 BWT 0.37 0.623* −0.081 −0.14 0.46* 0.165 −0.132 0.019 −0.173 NH 0.218 0.424 0.071 0.119 0.227 0.09 0.095 0.175 −0.068 0.521* NF 0.368 0.582* 0.006 0.11 0.348 0.169 0.194 0.227 −0.007 0.57* 0.843* FL 0.204 0.439 −0.076 −0.145 0.285 0.134 −0.151 −0.065 −0.22 0.826* 0.284 0.27 FC 0.327 0.449 −0.233 −0.255 0.397 0.198 −0.146 −0.151 −0.19 0.807* 0.148 0.153 0.85* FRD 0.39 0.478 −0.254 −0.281 0.42 0.28 −0.156 −0.154 −0.223 0.791* 0.158 0.184 0.803* 0.968* PLD 0.389 0.446 −0.271 −0.3 0.398 0.31 −0.161 −0.176 −0.22 0.741* 0.114 0.135 0.76* 0.945* 0.991* PED 0.005 0.199 0.062 0.022 0.242 −0.171 −0.048 0.022 −0.077 0.513* 0.324 0.34 0.464 0.337 0.217 0.1 Table 2. Pearson's correlation coefficients of traits under high input field management (GS2). PHF PG NSLF YLSF HTSF TS INSL DFM NSLH BWT NH NF FL FC FRD PLD PED PHF PG 0.774* NSLF −0.422 −0.257 YLSF −0.197 −0.024 0.75* HTSF 0.702* 0.563* −0.357 −0.251 TS 0.358 0.224 −0.092 0.062 0.45* INSL 0.213 0.272 −0.128 0.548* 0.084 0.197 DFM −0.007 0.006 −0.026 0.063 −0.218 −0.02 0.152 NSLH −0.149 −0.077 0.619* 0.533* −0.222 −0.156 0.002 −0.194 BWT 0.37 0.623* −0.081 −0.14 0.46* 0.165 −0.132 0.019 −0.173 NH 0.218 0.424 0.071 0.119 0.227 0.09 0.095 0.175 −0.068 0.521* NF 0.368 0.582* 0.006 0.11 0.348 0.169 0.194 0.227 −0.007 0.57* 0.843* FL 0.204 0.439 −0.076 −0.145 0.285 0.134 −0.151 −0.065 −0.22 0.826* 0.284 0.27 FC 0.327 0.449 −0.233 −0.255 0.397 0.198 −0.146 −0.151 −0.19 0.807* 0.148 0.153 0.85* FRD 0.39 0.478 −0.254 −0.281 0.42 0.28 −0.156 −0.154 −0.223 0.791* 0.158 0.184 0.803* 0.968* PLD 0.389 0.446 −0.271 −0.3 0.398 0.31 −0.161 −0.176 −0.22 0.741* 0.114 0.135 0.76* 0.945* 0.991* PED 0.005 0.199 0.062 0.022 0.242 −0.171 −0.048 0.022 −0.077 0.513* 0.324 0.34 0.464 0.337 0.217 0.1 * P-value < 0.001, P-value < 0.05 but > 0.001 https://doi.org/10.1371/journal.pone.0178734.t002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 8 / 23 Phenotyping and banana genomic selection Fig 2. Principal component analysis plots generated in R using package FactoMineR for the traits scored in a banana genomic selection training population. (A) shows the distribution of traits under low input field management (GS1) and (B) shows the distribution of traits under high input field management (GS2) on the first two components. Fig 2. Principal component analysis plots generated in R using package FactoMineR for the traits scored in a banana genomic selection training population. (A) shows the distribution of traits under low input field management (GS1) and (B) shows the distribution of traits under high input field management (GS2) on the first two components. https://doi.org/10.1371/journal.pone.0178734.g002 NSLF, YLSH, INSL, and NSLH and in contrast they had the lowest average scores on BWT, FL, FC, FRD, and PLD and the reverse was true for those projected on Dim 1. NSLF, YLSH, INSL, and NSLH and in contrast they had the lowest average scores on BWT, FL, FC, FRD, and PLD and the reverse was true for those projected on Dim 1. NSLF, YLSH, INSL, and NSLH and in contrast they had the lowest average scores on BWT, FL, FC, FRD, and PLD and the reverse was true for those projected on Dim 1. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Analysis of variance Visual inspection of boxplots for various traits indicated a cycle effect on data distribution of some traits while others were not affected by cycle. For example, Plant height increased at cycle 2 while index of non-spotted leaves did not increase (Fig 5a and 5b) and this was con- firmed by ANOVA results. Fruit traits such as FC, FRD and PLD showed a bimodal distribu- tion with the histogram having two peaks. Based on these parameters, the population could be separated into two main groups, poor fruit filling genotypes with FRD < 2.0 cm and FC < 8.0 cm, and good fruit filling genotypes with FRD 2.0 cm and FC 8.0 cm (S1A±S1D Fig). Fig 3. Principal component analysis plots generated in R using package FactoMineR for the cross combinations in a banana genomic selection training population. (A) shows the distribution of cross combinations under low input field management (GS1) and (B) shows the distribution of cross combinations under high input field management (GS2) on the first two components. https://doi.org/10.1371/journal.pone.0178734.g003 Fig 3. Principal component analysis plots generated in R using package FactoMineR for the cross combinations in a banana genomic selection training population. (A) shows the distribution of cross combinations under low input field management (GS1) and (B) shows the distribution of cross combinations under high input field management (GS2) on the first two components. https://doi org/10 1371/journal pone 0178734 g003 https://doi.org/10.1371/journal.pone.0178734.g003 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 9 / 23 Phenotyping and banana genomic selection Fig 4. Principal component analysis plots generated in R using package FactoMineR for the traits scored in a banana genomic selection training population. (A) shows the distribution of traits for individual genotypes and (B) shows the distribution of individual genotypes on the first two components based on mean of combined data from the two fields. https://doi.org/10.1371/journal.pone.0178734.g004 Fig 4. Principal component analysis plots generated in R using package FactoMineR for the traits scored in a banana genomic selection training population. (A) shows the distribution of traits for individual genotypes and (B) shows the distribution of individual genotypes on the first two components based on mean of combined data from the two fields. https://doi.org/10.1371/journal.pone.0178734.g004 Fig 4. Principal component analysis plots generated in R using package FactoMineR for the traits scored in a banana genomic selection training population. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Analysis of variance (A) shows the distribution of traits for individual genotypes and (B) shows the distribution of individual genotypes on the first two components based on mean of combined data from the two fields. https://doi.org/10.1371/journal.pone.0178734.g004 Coefficients of determination showed that under low input, cycle had less effect on NSLF, YLSF, INSL, TS, HTSF and PED across genotype cross combinations (Table 3). The Student's t-test revealed that both PED and HTSF were the most stable traits across cycles at 95% confi- dence level with P = 0.515 and P = 0.108, respectively. Under high input, cycle accounted for less than 50% of the variation in NSLF, YLSF, INSL, TS, HTSF, DFM, NSLH, NH, NF and PED between cross combinations. Just as in the first field, PED and HTSF were the least affected with P = 0.216 and P = 0.108, respectively. Under high input field management, trait variation due to cycle was more homogenous as compared to low input field management. However, in both cases the effects were statistically significant (P < 0.001) indicating that cycle is a source of variation in genotype performance. When generating ANOVA models, genotype (clone) was assumed to be the main source of variation. In addition to genotype the effect of cycle, field and their interaction with genotype Fig 5. Effect of cycle on trait variation in bananas, where (a) shows an increase in plant height at flowering at cycle 2 while (b) shows no increase in index of non-spotted leaves at cycle 2. https://doi.org/10.1371/journal.pone.0178734.g005 Fig 5. Effect of cycle on trait variation in bananas, where (a) shows an increase in plant height at flowering at cycle 2 while (b) shows no increase in index of non-spotted leaves at cycle 2. https://doi.org/10.1371/journal.pone.0178734.g005 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 10 / 23 Phenotyping and banana genomic selection Table 3. Coefficient of determination and Student's t-test P-values showing the effect of cycle on cross combinations. Analysis of variance GS1 GS2 Traits df R2 P-value df R2 P-value NH 60 0.87 <0.0001 56 0.44 <0.0001 PLD 57 0.78 <0.0001 56 0.65 <0.0001 FRD 59 0.77 <0.0001 56 0.68 <0.0001 PED 58 0.06 0.5150 56 0.03 0.2161 BWT 60 0.79 <0.0001 56 0.74 <0.0001 NF 60 0.54 <0.0001 56 0.37 <0.0001 FL 59 0.77 <0.0001 56 0.64 <0.0001 FC 58 0.79 <0.0001 56 0.73 <0.0001 DFM 59 0.54 <0.0001 56 0.25 <0.0001 NSLH 60 0.63 <0.0001 56 0.38 <0.0001 PHF 66 0.65 <0.0001 63 0.73 <0.0001 PG 66 0.65 <0.0001 63 0.73 <0.0001 NSLF 66 0.25 <0.0001 63 0.28 <0.0001 YLSF 66 0.47 <0.0001 63 0.26 <0.0001 INSL 66 0.14 0.0015 63 0.21 0.0001 TS 68 0.12 0.0032 68 0.12 0.0032 HTSF 68 0.04 0.1084 68 0.04 0.1084 Df d f f d GS1 l i t ® ld GS2 hi h i t ® ld d R2 f® i t f d t i ti Table 3. Coefficient of determination and Student's t-test P-values showing the effect of cycle on cross combinations. was investigated. In all models for all traits, genotype had significant effect on trait variation with P < 0.001 (Table 4, S3 Table). Traits that were not affected by the interaction between genotype and field management practice include PHF and PG whereas traits not affected by interaction between genotype and cycle include NSLF, YLSF, INSL, YLSH, FL, FRD and PED (P > 0.05). Weak interaction between genotype and cycle was observed on NSLH and HTSH with P = 0.0417 and 0.0408, respectively. In some cases, although there were significant inter- actions between genotype and field or cycle, either field or cycle did not show significant effect on the trait when interaction was included in the model. Whereas there were significant interactions between genotype and field management, there was no significant main effect of field on NSLF, YLSF, HTSF, INSL, TS, NSLH, YLSH, HTSH, NH, NF and PED. Similarly, in the presence of significant interaction between genotype and cycle, there was no main effect of cycle on INSL, HTSF, HTSH, FC, PLD and PED (Table 4, S3 Table). When the interactions were removed from the models, all the factors had significant effect on the traits except INSL and PED, for which cycle had no effect. Analysis of variance Analysis was repeated on these two traits using type I and type II ANOVA and both produced similar results as that observed with type III SS. https://doi.org/10.1371/journal.pone.0178734.t003 Phenotyping and banana genomic selection Table 4. Effect of genotype (clone), field management, cycle and their interaction on trait variation. Dep. variable Indep. variable Sum Sq Df F value Pr(>F) PHF Clone 2222889.11 306 3.77 <0.0001 Clone:Field 432297.46 284 0.79 0.9947 Clone:Cycle 332846.71 299 1.05 0.2662 PG Clone 73176.82 306 4.30 <0.0001 Clone:Field 12061.30 284 0.76 0.9981 Clone:Cycle 13057.24 299 1.51 <0.0001 INSL Clone 116602.02 306 2.44 <0.0001 Clone:Field 58583.77 284 1.32 0.0005 Clone:Cycle 51026.49 299 0.95 0.6947 TSsqrt Clone 240.28 305 3.21 <0.0001 Clone:Field 100.88 282 1.46 <0.0001 BWTsqrt Clone 1213.89 303 12.55 <0.0001 Clone:Field 126.77 269 1.48 <0.0001 Clone:Cycle 108.68 276 1.49 <0.0001 FC Clone 9506.06 300 16.11 0.0000 Clone:Field 733.66 269 1.39 0.0001 Clone:Cycle 751.00 272 1.29 0.0021 PLD Clone 865.42 299 17.60 0.0000 Clone:Field 68.27 269 1.54 <0.0001 Clone:Cycle 60.55 271 1.29 0.0022 PED Clone 20.96 299 11.41 <0.0001 Clone:Field 16.61 269 10.05 <0.0001 Clone:Cycle 3.15 271 0.80 0.9913 sqrt Original data transformed by square root https://doi.org/10.1371/journal.pone.0178734.t004 Table 4. Effect of genotype (clone), field management, cycle and their interaction on trait variation. https://doi.org/10.1371/journal.pone.0178734.t004 this work because they had less than five hybrids in the population. However, crosses between different EAHB with Calcutta 4 were treated as one cross because the EAHB represent a clone set with very low genetic diversity [9]. In total sixteen cross combinations were compared and they included one 2x by 2x, one 3x by 2x and fourteen 4x by 2x crosses (Table 5 and S2 Table). this work because they had less than five hybrids in the population. However, crosses between different EAHB with Calcutta 4 were treated as one cross because the EAHB represent a clone set with very low genetic diversity [9]. In total sixteen cross combinations were compared and they included one 2x by 2x, one 3x by 2x and fourteen 4x by 2x crosses (Table 5 and S2 Table). The best cross in terms of yield and fruit size was C10 (1201K-1xSH3217). Many hybrids from this cross had the highest bunch weight (R = 3.8) characterized by longer fruit fingers, big fruit circumference and the highest pulp content. However, the plants were very tall with big girth. Their maturity period was shorter (about 4.5 months on average) and comparable to hybrids from EAHBxCalcutta 4. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Performance of cross combinations (parental pairs) The GS training population consisted of 77 different cross combinations representing about two decades of banana breeding activities by IITA and NARO Uganda. Some of these cross combinations gave rise to the tetraploids and improved diploids that are part of the core breed- ing lines in the program. Tetraploids and triploids were predominantly used as female parents while the diploids provided the pollen source but in some instances 2x by 2x or 4x by 4x crosses were made. The majority of the cross combinations were excluded for this analysis in PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 11 / 23 Phenotyping and banana genomic selection Table 5. Comparison of mean performance of parental cross combinations (S) and hybrids from those crosses (R) against the mean of all parents. CROSS C04 C05 C08 C10 C11 C12 C13 C16 C22 C27 C31 C33 C34 C37 C61 MxC4 S (NSLF) -0.5 -0.2 1.2 0.4 0.2 -0.2 -0.3 1.2 0.7 1.9 0.1 -0.3 -0.1 1.4 0.4 -1.1 R (NSLF) -0.4 0.5 0.8 0.0 1.4 0.9 0.1 1.8 2.1 1.8 0.6 0.3 -0.2 1.4 0.8 0.1 S (YLSF) -0.7 -0.4 1.4 0.3 0.0 -0.2 -0.3 1.7 0.2 1.7 -0.8 -1.1 -0.7 1.1 -0.3 -1.5 R (YLSF) -0.7 0.3 0.7 0.0 0.8 0.4 0.1 1.8 2.2 1.7 0.4 -0.1 -0.1 1.2 0.7 -0.1 S (PHF) 24.1 -33.5 6.6 35.2 35.8 17.2 -37.5 3.8 -21.8 -1.5 -14.0 -11.4 -58.2 -22.5 0.2 25.9 R (PHF) 14.8 -23.8 10.1 33.6 -23.4 -7.4 -39.4 -6.6 -62.3 2.5 0.5 7.9 -31.0 -17.6 -9.5 7.6 S (PG) 9.6 -2.9 5.0 11.1 11.7 2.8 -7.5 -0.1 -5.3 1.3 0.9 1.4 -8.5 -1.6 3.6 3.3 R (PG) 3.6 -3.2 1.2 6.0 -1.4 2.3 -6.8 -1.4 -5.7 -0.6 2.2 4.9 -5.4 -2.0 3.3 2.0 S (HTSF) 11.4 -8.5 30.1 24.2 31.7 -5.5 -18.1 20.5 -46.3 23.0 -27.5 -25.0 -33.7 0.3 -4.7 23.0 R (HTSF) 15.0 -10.3 6.3 23.3 -21.1 -7.3 -26.8 14.3 -32.5 13.4 0.8 -2.5 -14.4 -4.0 -6.4 4.2 S (INSL) -1.8 -1.0 4.9 0.9 0.3 0.2 0.7 7.2 -1.5 3.9 -6.4 -6.5 -4.2 1.1 -4.3 -7.0 R (INSL) -2.9 0.6 1.4 1.2 -1.9 -0.7 1.1 5.7 7.2 4.1 0.1 -1.8 1.8 2.7 2.2 -0.9 S (TS) -1.6 2.8 0.7 -1.0 1.1 -1.1 3.0 1.2 -1.7 0.1 -3.3 -2.9 1.3 -0.7 -0.4 0.0 R (TS) -0.3 1.9 0.6 0.8 -1.0 -1.2 0.8 1.0 -0.4 -0.8 0.3 -1.9 1.2 0.0 0.7 -1.2 S (DFM) 2.4 2.7 15.9 10.0 -1.3 4.9 6.5 31.4 14.2 32.9 8.9 10.9 8.8 28.0 8.2 -21.3 R (DFM) 7.8 6.3 21.1 7.3 -1.9 19.9 1.6 8.3 54.6 32.6 23.9 11.2 13.5 22.3 20.7 7.2 S (NSLH) -0.7 -0.9 0.3 -0.4 -0.5 -0.1 -0.7 1.3 0.5 1.5 0.4 0.1 -0.3 1.4 0.6 -0.7 R (NSLH) -0.9 0.0 0.8 -0.4 1.5 0.6 0.1 2.3 0.3 1.4 0.3 0.1 -0.1 1.1 0.1 0.2 S (YLSH) -0.4 -0.4 0.3 -0.1 -0.2 0.0 -0.3 1.0 0.1 1.1 -0.1 -0.2 -0.3 0.6 0.0 -0.4 R (YLSH) -0.5 0.1 0.5 -0.2 0.9 0.1 0.0 0.8 0.1 0.6 0.2 0.2 0.1 0.8 0.1 0.1 S (HTSH) 27.6 -0.1 25.2 34.0 26.8 5.9 -21.3 10.8 -21.7 28.9 -2.6 4.6 -21.7 1.6 -2.2 7.7 R (HTSH) 23.4 -0.3 45.0 24.0 -18.4 18.4 -23.1 17.3 -15.9 19.1 23.6 9.9 -11.6 15.0 2.9 31.0 S (BWT) 5.6 2.3 4.2 7.2 7.0 1.5 -2.3 -1.5 -0.6 2.1 2.1 1.6 -1.2 -0.2 2.6 -0.7 R (BWT) 3.4 0.7 1.0 3.8 -0.9 1.0 0.4 -4.0 -2.3 -2.3 0.7 2.5 -0.1 -2.8 3.4 1.4 S (NH) 0.7 0.1 0.2 2.6 0.5 0.7 -0.1 0.0 1.1 0.6 0.2 0.3 -0.3 -0.3 -0.1 -0.8 R (NH) 0.4 0.4 1.0 0.9 1.2 1.1 0.3 0.9 1.2 0.4 0.8 1.2 -0.4 0.5 0.7 -0.3 S (NF) 22.1 -1.8 19.7 37.2 17.5 7.0 -19.7 7.7 7.5 15.9 8.8 12.2 -13.4 9.2 3.6 -16.0 R (NF) 15.9 9.0 35.8 12.8 19.9 13.9 1.5 21.7 27.4 10.7 19.6 25.6 -3.1 16.3 13.5 2.0 S (FL) 1.6 -0.2 0.8 2.8 1.9 0.7 -1.1 -1.4 -1.5 0.4 0.5 1.0 -0.9 -0.2 1.2 0.2 R (FL) 2.8 0.3 -0.8 2.5 -1.3 -0.2 -0.2 -3.9 -2.0 -2.6 -0.5 1.6 1.3 -2.6 2.3 0.3 S (FC) 2.2 0.7 2.2 2.1 3.1 1.2 -1.2 -1.1 0.4 0.9 1.2 0.6 -0.7 0.3 1.4 0.9 R (FC) 0.8 0.0 -0.6 1.2 -1.8 -0.7 -0.4 -3.4 -2.8 -2.5 -0.8 0.1 -0.4 -3.0 0.6 0.8 S (FRD) 0.6 0.2 0.6 0.6 0.9 0.4 -0.4 0.0 0.2 0.6 0.5 0.3 -0.2 0.2 0.6 0.1 R (FRD) 0.2 0.0 -0.3 0.3 -0.7 -0.3 -0.2 -1.2 -1.0 -0.8 -0.4 -0.1 -0.2 -1.0 0.1 0.4 S (PLD) 0.6 0.2 0.6 0.6 0.9 0.3 -0.3 0.0 0.1 0.6 0.5 0.3 -0.1 0.2 0.6 0.1 R (PLD) 0.2 0.0 -0.3 0.3 -0.7 -0.3 -0.1 -1.2 -1.0 -0.9 -0.4 -0.1 -0.2 -1.0 0.1 0.4 S (PED) 0.00 0.00 0.02 0.00 0.02 0.03 -0.02 0.01 0.01 0.02 0.00 -0.03 -0.03 -0.02 -0.01 0.01 R (PED) 0.01 0.01 -0.01 0.01 0.00 0.02 -0.01 -0.01 0.01 0.04 0.02 0.01 0.00 0.00 0.01 0.00 S = Selection differential, R = Response to selection, bold values are the highest observations, C04 = 1201K-1x9128-3, C05 = 1201K-1 x cv. Generally, crosses involving SH3217, SH3362 and 9128±3 as male parents produced hybrids that had good fruit filling characteristics although they varied in Black Sigatoka resistance and suckering behavior. For example, crosses involving 9128±3 generated hybrids that had the lowest INSL. Hybrids from a cross between 5610S-1 and 2180K-6 produced the highest number of leaves scored at flowering (R = 2.1). They had the highest resistance to Black Sigatoka as reflected by INSL (R = 7.2%) despite the parents being susceptible. They were the shortest (R = -62.3 cm) with smaller plant girth. Their average maturity period was almost two months more than the average of all parental lines (R = 54.6 days) and the longest of all other hybrids. Due to long maturity period the number of standing leaves at harvest was very low because of normal leaf senescence. Despite producing many fruit fingers and slightly more hands per bunch, their average yield and size of fruits were lower than those of the parents. However, some exceptional lines such as 25031S-7 (diploid) had sizable bunch with relatively big fruits. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 12 / 23 Genetic diversity of GS training population Out of the nineteen SSR markers, eighteen were used to delineate the structure of the study population, because marker mMaCIR164 produced ambiguous allele profiles across samples. From 18 loci, 195 alleles were scored and the number of alleles per locus ranged between 4 and 18 with an average of 10.8. Polymorphism information content (PIC) of the markers was high with an average of 0.87 (0.53±0.95)while the major allele frequency was on average 0.22 (0.1± 0.45). Despite the complex pedigree background of the GS population, SSR markers were infor- mative enough to delineate the structure of the population (Fig 6). Hierarchical clustering based on Ward's criterion revealed ten groups indicating that the genetic diversity of popula- tion was high. The triploid East African highland bananas clearly separated from other trip- loids. They had the lowest within group genetic diversity. The tetraploids that resulted from crossing EAHB by cv. `Calcutta4' and M. acuminata ssp. malaccensis 250 formed their own cluster but were closely linked to that of EAHB, thus supporting the hypothesis that the tetra- ploids were formed after fusion of unreduced gametes from triploid EAHB and haploid gam- etes from diploid cv. `Calcutta4' and M. acuminata ssp. malaccensis 250. The within cluster dispersion was rather homogenous and not highly diverse for the tetraploid hybrids probably due high allele dosage from EAHB. SSR data suggested that the tetraploid presumed to be hybrids of cv. Enzirabahima by M. a malaccensis 250 (29275S-1, 29275S-4 and 29275S-5), were in fact admixtures from pollination of EAHB with cv. `Calcutta4'. These tetraploid inherited 17 alleles specific for cv. `Calcutta4' and none of ssp. malaccensis 250 specific alleles across the 18 SSR markers used. Hierarchical clustering of hybrids was much influenced by male parents used in the cross. The biggest percentage of hybrids was produced from crosses involving tetraploids derived from EAHB and cv. `Calcutta4'. Hybrids from ssp. malaccensis 250 were more distinct from the rest of the population and formed their own cluster. Four hybrids (26998S-1, 27074S-1, 28506S-1 and 27960s-1) presumed to be progeny of 2180K-6, cv. `Calcutta4' and cv. `Rose'as male parents clustered together with ssp. malaccensis 250 hybrids. SSR genotype profiles sug- gested that those four hybrids were misidentified because they had ssp. malaccensis 250 specific alleles. The highest genetic diversity was observed in the diploid parents and between families. Rose, C08 = 1201K-1 x malaccensis, C10 = 1201K-1 x SH3217, C11 = 1201K-1 x SH3362, C12 = 1438K-1 x 5610S-1, C13 = 1438K-1 x cv. Rose, C16 = 1438K-1 S = Selection differential, R = Response to selection, bold values are the highest observations, C04 = 1201K-1x9128-3, C05 = 1201K-1 x cv. Rose, C08 = 1201K-1 x malaccensis, C10 = 1201K-1 x SH3217, C11 = 1201K-1 x SH3362, C12 = 1438K-1 x 5610S-1, C13 = 1438K-1 x cv. Rose, C16 = 1438K-1 x malaccensis, C22 = 5610S-1 x 2180K-6, C27 = 660K-1 x malaccensis, C31 = 917K-2 x 5610S-1, C33 = 917K-2 x 9128±3, C34 = 917K-2 x cv. Rose, C37 = 917K-2 x malaccensis, C61 = 917K-2 x SH3362 and MxC4 = Matooke (EAHB) x Calcutta 4 Crosses involving M. acuminata ssp. malaccensis 250 as male parent produced hybrids that were tall, slender, with bunches that had many fruit fingers poorly filled with pulp but some individual genotype exceptions were observed. The hybrids were resistant to Black Sigatoka and had the highest number of functional leaves at harvesting. Hybrids from cv. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 13 / 23 Phenotyping and banana genomic selection Rose were slender and shorter and were the highest in sucker production while other traits varied considerably. Rose were slender and shorter and were the highest in sucker production while other traits varied considerably. Hybrids from different cross combinations had longer maturity period (128±185 days) than EAHB. On average EAHB mature within 90 days after flowering while the average maturity period for all parental lines was 130 days. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Genetic diversity of GS training population Diploids that were linked by pedigree clustered together but the within cluster differences were high compared to EAHB and tetraploids. Diploids such as cv. `Calcutta4', 861S-1, 5610S- 1, 2180K-1, Kokopo, and cv. `Rose'clustered with their hybrids. Hybrids derived from 5610S-1 x 2180K-1 were all diploids and closely related to cv. `Calcutta4' and 861S-1 and formed a sep- arate cluster. Although the pedigree of 2180K-1 could not be traced, there is a possibility that one of its parents was cv. `Calcutta4'. Hybrids from cv. `LongTavoy' and cv. `Calcutta4' were not easily delineated because of the close resemblance of these genotypes. One cluster (J) com- prising of triploid hybrids showed high within cluster diversity. Majority of advanced hybrids especially NARITA hybrids comprising of potential candidate varieties are found in this clus- ter. The ssp. zebrina accessions included in the analysis clustered within the main clusters sug- gesting their genetic relatedness with other acuminata genotypes. In the population, some genotypes were duplicates. The duplicates identified included 28465S-2 (A&B), 26337S-11 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 14 / 23 Phenotyping and banana genomic selection Fig 6. Dendrogram showing the genetic diversity of the genomic selection training population based on 19 informative SSR markers. The squared Euclidean distances were used to generate the hierarchical clusters based on ward.D criterion. Where cluster A = tetraploids (4x) by M. a. spp. malaccensis 250, * share only female parent, cluster B = matooke (EAHB), cluster C = tetraploids from EAHB (3x) by Calcutta 4 a wild diploid (2x), cluster D = wild and improved diploids, cluster E = Black Sigatoka resistant diploid hybrids, cluster F = hybrids of 5610S-1 as a male parent, * share grandparent Calcutta 4, GC = good for cooking and N = NARITA hybrid, cluster G = cv. Rose was the main male parent, * share genetic background, cluster H = Long Tavoy and Calcutta 4 are the grandparents, cluster I = mostly hybrids of SH3217 as male parent, N = NARITA, @ = released variety as NARITA 7/M9/ Kiwangazi and cluster J = triploid hybrids with complex pedigree, most advanced hybrids such as NARITAs (N) are found in this cluster of which some are promising variety candidates and GC = good for cooking. https://doi.org/10.1371/journal.pone.0178734.g006 Fig 6. Dendrogram showing the genetic diversity of the genomic selection training population based on 19 informative SSR markers. Trait evaluation Bananas express many traits that are used to evaluate hybrids in breeding programs. These traits can be broadly classified as vegetative/agronomic (growth) traits, or yield and consumer appeal (fruit) traits. Growth and yield related traits are used to assess the level of introgression of resistance genes and this is done in the early evaluation trial. The index of non-spotted leaves (INSL) is a measure of resistance to Black Sigatoka, a fungal disease that causes rapid drying of leaves hence reducing the photosynthetic area [7]. Results from ANOVA obtained in this work showed that INSL was not significantly affected by cycle. However, the effect of level of input in field management on INSL depended on genotype. This suggests that resistance to Black Sigatoka might be under strong genetic control and less influenced by cycle. Correlation analysis showed a positive correlation between INSL, NSLF and YLSF. How- ever, these three had low but significant negative correlations with yield-related traits under low input field management conditions. These results suggest that whereas some Black Siga- toka resistant genotypes give good yield, others produce inferior fruits. Reduction in functional leaves and photosynthetic area has been shown to impact banana yield potential [7]. Tushe- mereirwe [36] indicated that Black Sigatoka reduced yield of EAHB by more than 30%. Our results show that under high input field management conditions, the impact of the disease on yield traits was less severe (Tables 1 and 2). This result is in agreement with Mobambo et al. [37] who reported that soil fertility had an effect on host plant response to Black Sigatoka and yield in plantains. The symptoms of Black Sigatoka often increase after flowering probably because at that time the ability of a plant to withstand the fungal attack is lowered as it commits most of the energy and resources to the developing inflorescence. Some genotypes had no functional leaves at harvest, indicating that they were very susceptible to Black Sigatoka after flowering. Selection of hybrids based on the number of functional leaves at harvest as a mea- sure of resistance to Black Sigatoka should be done with caution because of the negative associ- ation between foliar symptoms to Black Sigatoka and fruit filling. Genetic diversity of GS training population The squared Euclidean distances were used to generate the hierarchical clusters based on ward.D criterion. Where cluster A = tetraploids (4x) by M. a. spp. malaccensis 250, * share only female parent, cluster B = matooke (EAHB), cluster C = tetraploids from EAHB (3x) by Calcutta 4 a wild diploid (2x), cluster D = wild and improved diploids, cluster E = Black Sigatoka resistant diploid hybrids, cluster F = hybrids of 5610S-1 as a male parent, * share grandparent Calcutta 4, GC = good for cooking and N = NARITA hybrid, cluster G = cv. Rose was the main male parent, * share genetic background, cluster H = Long Tavoy and Calcutta 4 are the grandparents, cluster I = mostly hybrids of SH3217 as male parent, N = NARITA, @ = released variety as NARITA 7/M9/ Kiwangazi and cluster J = triploid hybrids with complex pedigree, most advanced hybrids such as NARITAs (N) are found in this cluster of which some are promising variety candidates and GC = good for cooking. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 15 / 23 Phenotyping and banana genomic selection (A&B) and 26337S-22 (A&B) while 27524S-12 (A&B) that were assumed to be duplicates were clarified to be genetically different although both were progeny of ssp. malaccensis 250. Other supposed unique genotypes were identified as likely clonal pairs, such as 24948S-9 and 24948S-10, 24948S-22 and 24948S-27, 25623S-11 and 25628S-11, 24948S-12 and 24948S-21, 12479S-1 and 12479S-13, 25737S-1 and 25356S-1, and 25066S-1 and 25066S-2. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Trait evaluation The present study shows that based on yield and growth traits, four groups of bananas existed in the training population that is, genotypes with high INSL and good fruit filling, high INSL with poor fruit filling, low INSL with good fruit filling and low INSL with poor fruit fill- ing representing the four planes of the two components. However, PCA could not resolve the population structure into clear-cut clusters due to complex pedigrees, although Osuji et al. [38] used this approach to distinguish between different Musa triploids. This phenomenon could be attributed to differences in carbon source to sink capacities. Plant physiological studies have shown that the balance between source and sink transloca- tion of photosynthetic assimilates is key to plant productivity [39]. In bananas, Dens et al. [40] demonstrated the effect of manipulating the carbon source (C-source) and carbon sink (C- sink) of mother plant on ratoon crops in cv. `Williams'and cv. `GrandNain' at a mat level. Their results showed genotype and environmental effect on flowering time, plant height and bunch size for the first ratoon crop. They concluded that the bunch was a larger C-sink than the ratoon crop. At individual plant level, it is likely that difference in C-source to C-sink capacity exists in bananas because our results showed that poor fruit filling genotypes were not PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 16 / 23 Phenotyping and banana genomic selection significantly affected by cycle and field inputs. It can be postulated that when plants have a strong C-sink capacity they tend to have high yield with increased leaf senescence, while those with low C-sink capacity maintain many leaves with low yield at harvest. More physiological studies in banana are required to shed light on this aspect. It has been reported that at the time of flowering, the fruits and seeds became major sinks and any factor that reduces translocation of photosynthetic assimilates to fruits reduces the harvest index [41]. The training population consisted of poor and good fruit filling genotypes based on FL, FC, FRD and PLD. This characteristic was consistent across cycles and field management, with two overlapping peaks in a binary pattern (S1A Fig). However, given the consistence of the traits under different field conditions, there is likelihood that fruit filling is under control of one or few major-effect quantitative trait loci (QTL). Trait evaluation Given that the training population was not a classical bi-parental mapping population this argument may not hold, but investigations using genome wide association studies while accounting for pedigree effect [42] may help to unravel the underlying genetic mechanisms using genome-wide markers such as SNPs. This study did not find sufficient evidence to show that absolute apical dominance existed in our training population. Different levels of sucker regulation (1±25 suckers) were observed in different cross combinations. This result is in agreement with the observation made by Ortiz and Vuylsteke [43] that non-apical dominance genes were fixed in AA genotypes of Musa. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Performance of cross combinations The true breeding value of a genotype is determined by the quality of hybrids produced when it is involved in a cross. By comparing the responses to selection (R) and selection differentials (S) of sixteen cross combinations it was concluded that no single cross combination presented all the good qualities targeted by the breeders in hybrids. This further explains the complex trait variation observed within study population. No attempt was made to determine heritabil- ity of the traits because of unbalanced design and the possibility of confounding from heterosis [31]. Some hybrids that had many active leaves at harvest showed variation in fruit filling. P f f h h b id l i fl d b h l i l d i h The true breeding value of a genotype is determined by the quality of hybrids produced when it is involved in a cross. By comparing the responses to selection (R) and selection differentials (S) of sixteen cross combinations it was concluded that no single cross combination presented all the good qualities targeted by the breeders in hybrids. This further explains the complex trait variation observed within study population. No attempt was made to determine heritabil- ity of the traits because of unbalanced design and the possibility of confounding from heterosis [31]. Some hybrids that had many active leaves at harvest showed variation in fruit filling. Performance of the hybrids was greatly influenced by the male parent involved in the cross. Although both diploids and tetraploids had 50% segregation opportunity, the tetraploids were genetically very similar, whereas the diploids were more genetically diverse with the exception of SH3217 and SH3362 that were closely related. Crosses involving diploid SH3217, SH3362 and 9128±3 produced hybrids which were superior in yield compared to other crosses. These diploids are parthenocarpic, with big fruits and many hands (clusters) per bunch. The best cross combination was C10 (120K-1 x SH3217) that produced hybrids that were fairly resistant to Black Sigatoka, high yielding and quick maturing. Despite the susceptibility of 1201K-1 par- ent to Black Sigatoka, segregation was observed and some hybrids that had some acceptable levels of resistance were produced. Tenkouano et al. [49] reported a 4-fold contribution of male parents toward yield traits while Rowe and Rosales [50] highlighted that breeding for improved diploids with pest and disease resistance, parthenocarpy and good yield was the best strategy in banana improvement. GxE interaction The effects of cycle and field input management on the genotype and how the genotype inter- acted with these two aspects of the environment were evaluated. The effect of cross combina- tion was also assessed. Based on coefficients of determination and analysis of variance, genotype, cycle, field and their interactions had different levels of effect on trait variation among cross combinations and individual genotypes. While PHF and PG significantly increased at cycle 2, field management did not have a significant effect on these traits. This could be attributed to the fact that the suckers used were at different physiological maturity. Yield traits were also affected by cycle but the bi-modal distribution was maintained. When bananas are planted in the field they first undergo an establishment phase and build reserves that can accelerate growth of the daughter plants. Therefore, cycle 2 is best to compare geno- types especially with regard to yield traits. Tushemereirwe et al. [16] reported a cycle effect on traits when they analyzed some advanced hybrids, but it was not fully known whether this behavior occurred in different banana genotypes. The effect of cycle alone varied across traits depending on field management except for PED, HTSF and INSL that were most stable. It should however be noted that under optimum field management the cycle explains a small proportion of trait variation in genotypes because most traits had coefficient of determination values below 0.4 in GS2. The present results show that different banana traits may have different genetic architecture with some traits influenced by GxE. In marker assisted selection this can hamper deployment of classical marker technologies that rely on identifying QTLs. Approaches such as genomic selection that utilize genome-wide markers in complex populations such as in this study provide an opportunity to dissect such traits and could be exploited by banana breeders to increase genetic gain per unit time. Genotype by environment interaction has been shown to affect the accuracy of genomic selection models [24, 44]. Therefore, understanding genotype trait variation across different environments is paramount. Many hybrids generated from crossbreeding usually have inferior fruit size irrespective of the ploidy level. Such inferiority has been attributed to linkage drag from wild diploids [45]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 17 / 23 Phenotyping and banana genomic selection Bananas have a long selection cycle, they are labor intensive, costly and require large land area for evaluation. GxE interaction Any technology that can discriminate the inferior genotypes from the good ones at a nursery stage could save a lot of resources and time for the breeders thus increasing the breeding efficiency. With the availability of the Musa reference genome [46, 47] and decreasing costs of next generation sequencing technologies, high density marker technologies such as genotyping by sequencing are available for many plant species [48]. This provides an opportunity to investigate the application of genomic selection in banana breeding. Performance of cross combinations Gene pyramiding has also been suggested so that multiple introgressions of good traits are possible [51]. Most of the improved varieties produced by crossbreeding are triploid and all assumed to be completely sterile but no research has been conducted to evaluate their fertility. Further improvement of these triploids is necessary given that no single hybrid has all traits desired by farmers and consumers. The 2x by 2x hybrids were all diploid and some had sizable bunches compared to other diploids in the core breeding set, i.e. could be interesting as improved 2x parents. Further evaluation of these diploids for pollen viability and partheno- carpy will be necessary before they are incorporated in the core breeding set despite their long maturity period. Hybrids that take four months to mature may be considered quick maturing, given that the majority take more than four months. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Phenotyping and banana genomic selection interpreted. The set of markers used was reported to be informative and has been used on gen- otyping the banana collection from the International Transit Center [32]. The polymorphism information content (PIC) of 0.87 was high enough to resolve even the closest genotypes. Up to ten unique clusters were resolved and results showed that clustering was mostly influenced by the genetic diversity in diploid parents. Triploid EAHB and tetraploids derived from them by crossing with cv. `Calcutta4' formed two distinct but closely related clusters, supporting the hypothesis of production of unreduced 3n and reduced n gametes during meiotic events in the tetraploid progenitors [52]. Despite the high PIC of the markers, the EAHB showed a very low genetic diversity consistent with the hypothesis that this group of bananas is an ancient clone set [9]. Even with a high number of polymorphic SSR markers Kitavi et al. and Karamura et al. [9, 53] failed to separate this group into the corresponding phenotype-based clone sets of Karamura [1]. However, some genetic differences were observed between some individual genotypes that could be attributed to mutations within this ancient clone set. The population was predominated with genetic intro- gression from cv. `Calcutta4'. Hybrids from M. acuminata ssp. malaccensis 250 formed a distinct cluster. Three tetraploids presumed to be arising from a cross of EAHB with ssp. malaccensis 250 grouped together with those derived from EAHB by cv. `Calcutta4'. The pres- ence of Calcutta 4-specific alleles in these tetraploids and the absence of ssp. malaccensis 250 specific alleles suggest that these hybrids are progeny of EAHB by cv. `Calcutta4' hence the high genetic relationship with the rest of the tetraploids. Nevertheless, these tetraploids should be tested as parents to determine their breeding values so that the breeding genetic pool is expanded. The SSR markers proved useful in identifying duplicates and closely related genotypes based on pedigree background. A combination of highly polymorphic SSR markers and the power of Ward's clustering method that minimizes the within-group dispersion [34] in the Euclidean space helped to resolve the structure of the population that was highly interlinked by pedigree background. The high level of genetic complexity observed in this population rep- resents different recombination events that make it suitable as a training population for geno- mic selection. Apart from obtaining important data on the banana GS training population, important les- sons were learned during the course of this work. Dedicated efforts are required to understand the genome organization of bananas through cytological approaches. Ploidy analysis should be routinely employed in breeding programs to differentiate ploidy levels so that different selec- tion criteria are used to select hybrids intended for the breeding pipeline from those eligible for variety release. Despite a majority of the improved hybrids being triploids, their fertility should be tested so that further improvements can be made on them as a way to achieve gene pyramiding while minimizing inbreeding. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Genetic diversity of GS training population Whereas principal component analysis on cross combinations and individual genotypes showed that high genetic diversity existed in the training population, its power to resolve the structure of the population into clear-cut clusters that make biological sense was limited. This was attributed to complex pedigrees in the population with 77 cross combinations represented. The half-sib families were closely related to one another with which they shared a common parent. The population was interconnected due to shared parents in their pedigree. Use of SSR markers proved valuable in delineating the population structure that could be easily PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 18 / 23 Conclusion The response of genotype trait expression to cycle and field management practices varied greatly. The largest proportion of genetic variation was due to the greater genetic diversity of male parents used in crosses since the tetraploids used in the majority of crosses as female parents were genetically related. Yield traits accounted for 31±35% of the total principal com- ponent variation observed in the population and were loaded on the first component while vegetative traits contributed to the second component with 15±22%. A high level of correlation within vegetative- and yield-related traits was observed but correlation between vegetative and yield traits was low and depended on the interaction with field management practices. There- fore, genomic selection models could be developed for traits that are easy to measure. It is PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 19 / 23 Phenotyping and banana genomic selection likely that the predictive ability of traits that are difficult to phenotype will be similar to traits easily measured but highly correlated. The study population was observed to be genetically diverse with complex pedigree structure. Yield-related traits showed a bi-modal distribution, which was not influenced by cycle or field management. Resistance to Black Sigatoka was also stable across cycles but varied under different field management depending on the genotype. Principal component analysis could not delineate this complex population structure but the application of SSR markers in combination with Ward's hierarchical clustering proved power- ful and resolved the structure into biologically meaningful groups. Acknowledgments We thank the field staff of IITA-Uganda banana-breeding program for helping in data collec- tion, Mrs. Marie SeifertovaÂfrom the Institute of Experimental Botany, Olomouc, Czech Republic for the technical service in SSR fragment analysis and NARO, Uganda for providing some of the genotypes used in this study. Supporting information S1 Fig. Variation in fruit characteristics. (A) is a histogram showing the bimodal distribution of fruit circumference (FC), (B) cross sections of poor filling fruits, (C) good filling fruits with fruit diameter (FRD) and pulp diameter (PLD) values in cm, and (D) poor filling and good fill- ing banana fruits. S1 Table. List of genotypes in a banana genomic selection training population. (DOCX) S2 Table. Data used to calculate selection differential and response to selection for the six- teen cross combinations. (XLSX) Author Contributions Data curation: MN MB PC. Data curation: MN MB PC. Formal analysis: MN PC. Formal analysis: MN PC. Funding acquisition: JL RS BU JD. Funding acquisition: JL RS BU JD. Investigation: MN MB PC. Investigation: MN MB PC. Methodology: MN MB PC. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 References 1. Karamura D. Numerical taxonomic studies of the East African highland bananas (Musa AAA- East Africa) in Uganda. Doctor of Philosophy Thesis, The University of Reading. 1998;1±192. 2. Karamura DA, Karamura E, Tinzaara W. editors. Banana cultivar names, synonyms and their usage in East Africa. Bioversity International, Uganda; 2012. 3. Ortiz R, Swennen R. Review: From crossbreeding to biotechnology-facilitated improvement of banana and plantain. Biotechnol Adv. 2014; 32: 158±169. https://doi.org/10.1016/j.biotechadv.2013.09.010 PMID: 24091289 4. Perrier X, De Langhe E, Donohue M, Lentfer C, Vrydaghs L, Bakry F, et al. Multi-disciplinary perspec- tives on banana (Musa spp.) domestication. Proc Natl Acad Sci USA. 2011; 108(28): 11311±11318. https://doi.org/10.1073/pnas.1102001108 PMID: 21730145 5. Van Asten P, Fermont AM, Taulya G. Drought is a major yield loss factor for rainfed East African high- land banana. Elsevier, Agric Water Manag. 2011; 98: 541±552. 6. Lorenzen J, Tenkouano A, Bandyopadhyay R, Vroh B, Coyne D, Tripathi L. Overview of Banana and Plantain (Musa spp.) Improvement in Africa: Past and Future. Acta Hortic. 2010; 879: 595±603. 7. Barekye A. Breeding investigations for black Sigatoka resistance and associated traits in diploids, tetra- ploids and the triploid progenies of bananas in Uganda. Thesis: Doctor of Philosophy (PhD) in Plant Breeding, University of KwaZulu-Natal, Republic of South Africa. 2009:1±230. 8. Gold CS, Kagezi GH, Night G, Ragama P. The effects of banana weevil, Cosmopolites sordidus, dam- age on highland banana growth, yield and stand duration in Uganda. Ann Appl Biol. 2004; 145: 263± 269. 9. Kitavi M, Downing T, Lorenzen J, Karamura D, Onyango M, Nyine M, et al. The triploid East African Highland Banana (EAHB) genepool is genetically uniform arising from a single ancestral clone that underwent population expansion by vegetative propagation. Theor Appl Genet. 2016. https://doi.org/10 1007/s00122-015-2647-1 PMID: 26743524 10. Pillay M, Ogundiwin E, Nwakanma D, Ude G, Tenkouano A. Analysis of genetic diversity and relation- ships in East African banana germplasm. Theor Appl Genet. 2001; 102(6±7): 965±970. 11. Simmonds NW. Bananas, Musa cvs. In: Simmonds NW, editor. Breeding for durable resistance in perennial crops. FAO Technical Papers 70. Food and Agriculture Organization, Rome; 1986; pp. 17± 24. 12. Rowe PR. Breeding bananas and plantains for resistance to fusarial wilt: the track record. In: Ploetz RC, editor. Fusarium wilt of bananas. APS, St. Paul, MN. 1990; pp. 115±119 13. Pillay M, Tenkouano A. Genome, cytogenetics, and flow cytometry of Musa. In: Pillay M, Tenkouano A, editors. Writing ±review & editing: BU JD AB PC RS MB EH JL. Writing ±review & editing: BU JD AB PC RS MB EH JL. Project administration: JD RS. Project administration: JD RS. Resources: JD EH. Software: PC. Software: PC. Supervision: JL JD RS. Validation: JD BU RS EH. 20 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Phenotyping and banana genomic selection Writing ±original draft: MN. Writing ±review & editing: BU JD AB PC RS MB EH JL. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 References Banana breeding: progress and challenges. CRC Press, Boca Raton. 2011; pp. 53±70. 14. Ssebuliba R, Talengera D, Makumbi D, Namanya P, Tenkouano A, Tushemereirwe W, et al. Reproduc- tive efficiency and breeding potential of East African highland (Musa AAA-EA) bananas. Field Crops Res. 2006; 95: 250±255. 15. Vuylsteke D, Swennen R, Ortiz R. Development and performance of black sigatoka resistant tetraploid hybrids of plantain (Musa spp., AAB group). Euphytica. 1993; 65: 33±42. 16. Tushemereirwe W, Batte M, Nyine M, Tumuhimbise R, Barekye A, Ssali T, et al. Performance of NAR- ITA hybrids in the preliminary yield trial for three cycles in Uganda. IITA, NARO, Uganda; 2015. 17. Desta ZA, Ortiz R. Genomic selection: genome-wide prediction in plant improvement. Trends Plant Sci. 2014; 19(9): 592±601. https://doi.org/10.1016/j.tplants.2014.05.006 PMID: 24970707 18. Meuwissen THE, Hayes BJ, Goddard ME. Prediction of total genetic value using genome-wide dense marker maps. Genetics. 2001; 157: 1819±1829. PMID: 11290733 19. PeÂrez P, de los Campos G. Genome-wide regression and prediction with the BGLR statistical package. Genetics. 2014; 198: 483±495. https://doi.org/10.1534/genetics.114.164442 PMID: 25009151 20. Lorenz AJ, Chao S, Asoro FG, Heffner EL, Hayashi T, Iwata H, et al. Genomic selection in plant breed- ing: knowledge and prospects. Adv Agron. 2011; 110: 77±123. 21. Nakaya A, Isobe SN. Will genomic selection be a practical method for plant breeding? Ann Bot. 2012; 110: 1303±1316. https://doi.org/10.1093/aob/mcs109 PMID: 22645117 22. Togashi K, Lin CY, Yamazaki T. The efficiency of genome-wide selection for genetic improvement of net merit. J Anim Sci. 2011; 89: 2972±2980. https://doi.org/10.2527/jas.2009-2606 PMID: 21512116 21 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Phenotyping and banana genomic selection 23. Goddard ME, Hayes BJ. Review article: Genomic selection. J Anim Breed Genet. 2007; 124: 323±330. https://doi.org/10.1111/j.1439-0388.2007.00702.x PMID: 18076469 24. Crossa J, PeÂrez P, Hickey J, Burgueño J, Ornella L, CeroÂn-Rojas J. et al. Genomic prediction in CIM- MYT maize and wheat breeding programs. Heredity. 2014; 112: 48±60. https://doi.org/10.1038/hdy. 2013.16 PMID: 23572121 25. Beaulieu J, Doerksen T, CleÂment S, MacKay J, Bousquet J. Accuracy of genomic selection models in a large population of open-pollinated families in white spruce. Heredity. 2014; 113: 343±352. https://doi. org/10.1038/hdy.2014.36 PMID: 24781808 26. Onogi A, Watanabe M, Mochizuki T, Hayashi T, Nakagawa H, Hasegawa T, et al. Toward integration of genomic selection with crop modelling: the development of an integrated approach to predicting rice heading dates. Theor Appl Genet. 2016; 129: 805±817. https://doi.org/10.1007/s00122-016-2667-5 PMID: 26791836 27. References De Oliveira EJ, de Resende MDV, Santos VS, Ferreira CF, Oliveira GAF, da Silva MS, et al. Genome- wide selection in cassava. Euphytica 2012. 28. DolezÏel J, LysaÂk MA, Van den Houwe I, DolezÏelovaÂM, Roux N. Use of flow cytometry for rapid ploidy determination in Musa species. New Methods, INFOMUSA. 1997; 6(1): 6±9. 29. DolezÏel J. Flow cytometric analysis of nuclear DNA content in higher plants. Phytochem Anal. 1991; 2: 143±154. 30. Craenen K. Black Sigatoka disease of banana and plantain: A reference manual, International Institute of Tropical Agriculture, Nigeria, Ibadan. Host plant response to black sigatoka. 1998; pp. 41±45. 31. Piepho H-P, MoÈhring J. Computing Heritability and Selection Response from Unbalanced Plant Breed- ing Trials. Genetics. 2007; 177: 1881±1888. https://doi.org/10.1534/genetics.107.074229 PMID: 18039886 32. ChristelovaÂP, ValaÂrik M, HřibovaÂ, Van den houwe I, Channelière S, Roux N, et al. A platform for efficient genotyping in Musa using microsatellite markers. AoB Plants. 2011. https://doi.org/10.1093/aobpla/ plr024 PMID: 22476494 33. Ward JH. Hierarchical grouping to optimize an objective function. J Am Stat Assoc. 1963; 58: 236±244. 34. Murtagh F, Legendre P. Ward's Hierarchical Agglomerative Clustering Method: Which Algorithms Implement Ward's Criterion? J Classif. 2014; 31: 274±295. 35. Liu K, Muse SV. PowerMarker: integrated analysis environment for genetic marker data. Bioinformatics. 2005; 21(9): 2128±2129. https://doi.org/10.1093/bioinformatics/bti282 PMID: 15705655 36. Tushemereirwe WK. Factors influencing the expression of leaf spot diseases of highland bananas in Uganda. PhD thesis. University of Reading, U.K; 1996. 37. Mobambo K, Zoufa K, Gauhl F, Adeniji M, Pasberg-Gauhl C. Effect of soil fertility on host response to black leaf streak of plantain (Musa spp., AAB group) under traditional farming systems in southeastern Nigeria. Int J Pest Manag. 1994; 40: 75±80 38. Osuji JO, Okol BE, Vuylsteke D, Ortiz R. Multivariate pattern of quantitative trait variation in triploid banana and plantain cultivars. Sci Hortic. 1997; 71: 197±202. 39. Lemoine R, La Camera S, Atanassova R, DeÂdaldeÂchamp F, Allario T, Pourtau N, et al. Source-to-sink transport of sugar and regulation by environmental factors. Front Plant Sci. 2013; 4: 1±21. 40. Dens KR, Romero RA, Swennen R, Turner DW. Removal of bunch, leaves, or pseudostem alone, or in combination, influences growth and bunch weight of ratoon crops in two banana cultivars. J Hortic Sci Biotechnol. 2008; 83(1): 113±119. 41. Wardlaw IF. Tansley Review No. 27ÐThe control of carbon partitioning in plants. New Phytol. 1990; 116: 341±381. 42. Korte A, Farlow A. PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 References The advantages and limitations of trait analysis with GWAS: a review. Plant Methods. 2013; 9: 29. https://doi.org/10.1186/1746-4811-9-29 PMID: 23876160 43. Ortiz R, Vuylsteke DR. Genetics of Apical Dominance in Plantain (Musa spp., AAB Group) and Improve- ment of Suckering Behavior. J Am Soc Hortic Sci. 1994; 119(5): 1050±1053. 44. Burgueño J, de los Campos G, Weigel K, Crossa J. Genomic prediction of breeding values when model- ing genotype x environment interaction using pedigree and dense molecular markers. Crop Sci. 2012; 52: 707±719. 45. Lorenzen J, Tenkouano A, Bandyopadhyay R, Vroh I, Coyne D, Tripathi L. The role of crop improve- ment in pest and disease management. Acta Hortic. 2009; 828: 305±314. 46. Martin G, Baurens FC, Droc G, Rouard M, Cenci A, Kilian A. et al. Improvement of the banana ªMusa acuminataº reference sequence using NGS data and semi-automated bioinformatics methods. BMC Genomics. 2016; 17: 243. https://doi.org/10.1186/s12864-016-2579-4 PMID: 26984673 22 / 23 PLOS ONE \| https://doi.org/10.1371/journal.pone.0178734 June 6, 2017 Phenotyping and banana genomic selection 47. D'Hont A, Denoeud F, Aury JM, Baurens FC, Carreel F, Garsmeur O, et al. The banana (Musa acumi- nata) genome and the evolution of monocotyledonous plants. Nature. 2012; 488: 213±219. https://doi. org/10.1038/nature11241 PMID: 22801500 48. Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, et al. A robust, simple Genotyp- ing-by-Sequencing (GBS) approach for high diversity species. PLoS ONE. 2012; 6(5): e19379. https:// doi.org/10.1371/journal.pone.0019379 PMID: 21573248 49. Tenkouano A, Ortiz R, Vuylsteke D. Combining ability for yield and plant phenology in plantain derived populations. Euphytica. 1998; 104: 151±158. 50. Rowe P, Rosales F. Diploid breeding at FHIA and the development of Goldfinger. InfoMusa. 1993; 2: 9± 11. 51. Pillay M, Ude G, Kole C, editors. Genetics, Genomics, and Breeding of Bananas. Science Publishers; 2012. pp. 116±123. 52. Ssali RT, Nawankunda K, Erima RB, Batte M, Tushemereirwe WK. On-Farm Participatory Evaluation of East African Highland Banana `Matooke'Hybrids (Musa spp.). Acta Hortic. 2010; 879: 585±591. 53. Karamura D, Kitavi M, Nyine M, Ochola D, Muhangi S, Talengera D, et al. Genotyping the local banana landrace groups of East Africa. Acta Hortic. 2016; 1114: 67±73. 23 / 23
https://openalex.org/W2303232007	https://cancerci.biomedcentral.com/track/pdf/10.1186/s12935-016-0297-2	English	null	Analysing the mutational status of adenomatous polyposis coli (APC) gene in breast cancer	Cancer cell international	2,016	cc-by	4,908	Cancer Cell International Cancer Cell International Chang et al. Cancer Cell Int (2016) 16:23 DOI 10.1186/s12935-016-0297-2 © 2016 Chang et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Breast cancer is a heterogeneous disorder for which the underlying genetic basis remains unclear. We developed a method for identifying adenomatous polyposis coli (APC) mutations and we evaluated the possible asso‑ ciation between APC genetic variants and breast cancer susceptibility. Methods: Genomic DNA was extracted from tumor and matched peripheral blood samples collected from 89 breast cancer patients and from peripheral blood samples collected from 50 controls. All samples were tested for mutations in exons 1–14 and the mutation cluster region of exon 15 by HRM analysis. All mutations were confirmed by direct DNA sequencing. Results: We identified a new single nucleotide polymorphism (SNP), c.465A>G (K155K), in exon 4 and seven known SNPs: c.573T>C (Y191Y) in exon 5, c.1005A>G (L335L) in exon 9, c.1458T>C (Y486Y) and c.1488A>T (T496T) in exon 11, c.1635G>A (A545A) in exon 13, and c.4479G>A (T1493T) and c.5465T>A (V1822D) in exon 15. The following altera‑ tions were found in 2, 1, 2, and 1 patients, respectively: c.465A>G, c.573T>C, c.1005A>G, and c.1488A>T. There was no observed association between breast cancer risk and any of these APC SNPs. Conclusions: APC mutations occur at a low frequency in Taiwanese breast cancer cases. HRM analysis is a powerful method for the detection of APC mutations in breast. Keywords: HRM, APC, Breast cancer, Direct DNA sequencing 2 (HER2). Combinations of these markers can be used to further subtype the tumor and guide treatment plan- ning, including luminal A (ER+ and/or PR+; HER2−), luminal B (ER+ and/or PR+; HER2+), basal-like (ER−, PR−, and HER2−), and HER2-enriched (ER−, PR−, and HER2+) markers [4]. Although breast cancer survival has improved significantly within the last few decades, the need for new therapeutic strategies aimed at specific tumorigenic cells and their key oncogenic pathways has become obvious. Analysing the mutational status of adenomatous polyposis coli (APC) gene in breast cancer ‑Sian Chang1,2, Chien‑Yu Lin2, Shu‑Fen Yang2, Cheng‑Mao Ho2 and Jan‑Gowth Chang1,2,3* Background In addition, new cancer genes that had not previously been implicated in the development of breast cancer have been found to be mutated, including ARID1B, CASP8, MAP3K1, MAP3K13, NCOR1, SMARCD1, and CDKN1B [7, 8]. APC, which was cloned in 1991, is located at chromo- some 5q21-22 and contains 15 exons. Its tumor-sup- pressing activity is thought to be based on regulation of the intracellular level of beta-catenin within the Wing- less/Wnt signal transduction pathway [9]. In the absence of Wnts, casein kinase 1 and glycogen synthase kinase- 3-bata phosphorylate beta-catenin, causing the ubiqui- tination and subsequent degradation of beta-catenin by the 26S proteasome. Conversely, when Wnt proteins are secreted from cells, the phosphorylation and degrada- tion of beta-catenin is blocked, leading to its accumula- tion. The stabilized beta-catenin then translocates to the nucleus, where it binds with T-cell factor/lymphoid enhancer-binding factor-1 to induce the expression of downstream target genes. The mutation of APC may cause constitutive stimulation of the Wingless/Wnt sig- nal transduction pathway, promoting the accumulation of beta-catenin in the cytoplasm and leading to aberrant cellular proliferation. Therefore, APC is a negative regu- lator of the Wingless/Wnt signal transduction pathway. The Wingless/Wnt signal transduction pathway may represent a novel therapeutic target for the treatment of triple-negative breast cancer (ER, PR, and HER 2-nega- tive breast cancer) [10]. In this regard, mutations in APC in patients with breast cancer are of particular interest. a TNM stage based on the sixth edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (2002) including c.465A>G in exon 4 (Fig. 1), c.573T>C in exon 5, c.1005A>G in exon 9, c.1458T>C and c.1488A>T in exon 11, c.1635G>A in exon 13, and c.4479G>A and c.5465T>A in exon 15 (Additional file 1: Figure S4, S8, S10, S12 and S14). HRM analysis of exons 1–3, 6–8, 10, 12, and 14 failed to identify mutations or polymorphisms in these regions (Additional file 1: Figure S1-S3, S5-S7, S9, S11 and S13). The frequencies of the SNPs in the controls were 0 % for c.465A>G, c.573T>C, and c.1488A>T. The frequencies for c.1005A>G, c.1458C, c.1635A, c.4479A, and c.5465A were 1, 67, 85, 85, and 94 %, respectively, and therefore similar to those in the malignant tissues. An evaluation of the relationship between each APC SNP and breast cancer risk (Table 2) did not show an association with any of the SNPs at any of the loci. Background Breast cancer is a common malignancy and the leading cause of death from cancer among females in economi- cally developing countries, accounting for 23 % (1.38 million) of all new cancer cases and 14 % (458,400) of all cancer deaths in 2008 [1]. The incidence of breast can- cer has been increasing in Asia [2, 3]; moreover, breast cancer in Asians occurs at a younger age than in Cau- casians. To date, treatment planning for breast cancer patients has been based on a histologic analysis of the primary tumor and the expression of molecular mark- ers such as estrogen receptor (ER), progesterone recep- tor (PR), and human epidermal growth factor receptor Like most cancers, the development of breast can- cer is the result of accumulated mutations in oncogenes and tumor suppressor genes. These genes are involved in cell proliferation, growth, differentiation, survival, apop- tosis, and DNA repair [5, 6]. A large number of well- established cancer genes have been implicated in breast cancer development, including BRCAl, RB1, TP53, PTEN, Correspondence: d6781@mail.cmuh.org.tw 1 Epigenome Research Center, China Medical University Hospital, 2 Yuh‑Der Road, Taichung 404, Taiwan Full list of author information is available at the end of the article Chang et al. Cancer Cell Int (2016) 16:23 Page 2 of 6 Table 1 Clinical characteristics of the breast cancer patients a TNM stage based on the sixth edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (2002) Characteristics No Percentage Total patients 89 Age mean 53 range 33–85 TNM stagea 0 0 0 I 8 11 IIa/IIb 32 44 III/IV 32 44 ER-positive 51 58 PR-positive 71 81 HER2-positive 29 33 AKT1, CDH1, GATA3, and PIK3CA. Other groups of genes responsible for signal transduction including APC, ARID1A, ARID2, ASXL1, BAP1, KRAS, MAP2K4, MLL2, MLL3, NF1, SETD2, SF3B1, SMAD4, and STK11, have also been found to contain somatic mutations in tumors. In addition, new cancer genes that had not previously been implicated in the development of breast cancer have been found to be mutated, including ARID1B, CASP8, MAP3K1, MAP3K13, NCOR1, SMARCD1, and CDKN1B [7, 8]. Table 1 Clinical characteristics of the breast cancer patients AKT1, CDH1, GATA3, and PIK3CA. Other groups of genes responsible for signal transduction including APC, ARID1A, ARID2, ASXL1, BAP1, KRAS, MAP2K4, MLL2, MLL3, NF1, SETD2, SF3B1, SMAD4, and STK11, have also been found to contain somatic mutations in tumors. HRM analysis of APCh The clinical characteristics of the breast cancer patients enrolled in this study, including mean age, age range, and receptor status, are presented in Table 1. Background APC mutations are a major contributing factor to colo- rectal cancer. In breast cancer patients, the overall APC mutational rate ranges from 0.4 to 18 % [11, 12]. We used high-resolution melting (HRM) analysis and direct sequencing to screen exons 1–14 and part of exon 15, including the mutation cluster region (MCR) (codons 1286–1513), of APC in breast cancer tissues. The aim of this study was to understand the APC mutation status in a Taiwanese cohort of breast cancer patients. Concordance of SNPs in paired tumor and blood samples To determine whether c.465A>G, c.573T>C, c.1005A>G, and c.1488A>T were somatic or germline abnormali- ties, we investigated the concordance between geno- types from all 89 paired tumor and blood samples. The data indicated a 97.75 % (87/89) concordance between tumor and blood samples, except for a discrepancy in two samples for c.573T>C and c.1488A>T, in which the mutations were somatic and were detected in the tumor tissues but not in the blood sample. Discussion Mutations in APC have been found in several cancers, including colorectal [13], pancreatic [14], and hepa- tocellular [15], but they are reported to be much rarer in breast cancer. The first report of APC mutations in breast cancer was published by Kashiwaba et al. [16], The APC gene mutations identified in the 89 patients and 50 controls by HRM analysis are summarized in Table 2. Eight single-nucleotide alterations were detected in the cancerous tissues of the patients, Page 3 of 6 Chang et al. Cancer Cell Int (2016) 16:23 Table 2 Single nucleotide alterations in APC identified in breast cancers (n = 89) and controls (n = 50) using high resolu- tion melting analysis P value by Fisher’s exact test when the cell expectation was less than five Position in gene Nucleotide loca- tion, cDNA (from start codon) Amino acid loca- tion dbSNP identifier Han Chinese in Bejing, China (1000 genomes project data) (%) Cases n = 178 alleles (%) Controls n = 100 alleles (%) P value Exon 4 c.465A>G K155K Allele A 176 (98.88) 100 (100) 0.5377* Allele G 2 (1.12) 0 (0) Exon 5 c.573T>C Y191Y rs185154886 Allele T 100 177 (99.44) 100 (100) 1.000* Allele C 0 1 (0.56) 0 (0) Exon 9 c.1005A>G L335L rs3797704 Allele A 100 176 (98.88) 99 (99) 1.000* Allele G 0 2 (1.12) 1 (1) Exon 11 c.1458T>C Y486Y rs2229992 Allele T 32.52 52 (29.2) 33 (33) 0.511 Allele C 67.48 126 (70.8) 67 (67) Exon 11 c.1488A>T T496T rs9282599 Allele A 99.51 177 (99.44) 100 (100) 1.000* Allele T 0.49 1 (0.56) 0 (0) Exon 13 c.1635G>A A545A rs351771 Allele G 18.93 29 (16.3) 15 (15) 0.777 Allele A 81.07 149 (83.7) 85 (85) Exon 15 c.4479G>A T1493T rs41115 Allele G 18.93 24 (13.5) 15 (15) 0.727 Allele A 81.07 154 (86.5) 85 (85) Exon 15 c.5465T>A V1822D rs459552 Allele T 9.22 14 (7.9) 6 (6.0) 0.564 Allele A 90.78 164 (92.1) 94 (94.0) Table 2 Single nucleotide alterations in APC identified in breast cancers (n = 89) and controls (n = 50) using high resolu- tion melting analysis fi * P value by Fisher’s exact test when the cell expectation was less than five used whole exome sequencing to detect somatic altera- tions in 100 patients. They found two mutations in two breast samples (2 %). The above-mentioned mutations were unique. Discussion In the present study, among the eight APC SNPs detected, seven were silent in the 89 breast cancer samples. Of those, five were also found in the control population, one was present in both blood and cancerous tissue, and two (c.573T>C and c.1488A>T) were somatic mutations detected in cancerous tissues only and not in patients’ blood cells. These results suggest that APC does not play an important role in the development of breast cancer in the Taiwanese population, although the APC mutation rate was higher in breast cancer patients than in controls. However, a limitation of our study was that we did not comprehensively evaluate all of the exons in the control samples.h who detected a mutation rate of 6 % (2 of 31) in primary breast carcinoma samples using PCR followed by restric- tion fragment length polymorphism (RFLP) and single- strand conformation polymorphism (SSCP) analyses. Sorlie et al. [11] studied mutations in exon 15 using a protein truncation test (PTT). They reported a frameshift mutation at nucleotide 4678 in one patient (1 of 227). Furuuchi et al. [12] screened almost the entire coding region using a yeast-based assay. They reported a muta- tion rate of 18 % (13 of 70) in primary breast cancers; the mutations included five nonsense mutations, five single base pair deletions, and five single base pair insertions. Until now, their study had the highest reported mutation rate in breast cancer. Hayes et al. [17] analyzed exon 15 in 27 metaplastic carcinomas by direct DNA sequenc- ing and found one mutation in two samples (7.4 %). Kan et al. [18] used mismatch repair detection (MRD) tech- nology to identify somatic mutations in 441 tumors, including 183 breast cancers. They found three mutations in three breast tissue samples (1.6 %). Stephens et al. [8] who detected a mutation rate of 6 % (2 of 31) in primary breast carcinoma samples using PCR followed by restric- tion fragment length polymorphism (RFLP) and single- strand conformation polymorphism (SSCP) analyses. Sorlie et al. [11] studied mutations in exon 15 using a protein truncation test (PTT). They reported a frameshift mutation at nucleotide 4678 in one patient (1 of 227). Furuuchi et al. [12] screened almost the entire coding region using a yeast-based assay. Discussion They reported a muta- tion rate of 18 % (13 of 70) in primary breast cancers; the mutations included five nonsense mutations, five single base pair deletions, and five single base pair insertions. Until now, their study had the highest reported mutation rate in breast cancer. Hayes et al. [17] analyzed exon 15 in 27 metaplastic carcinomas by direct DNA sequenc- ing and found one mutation in two samples (7.4 %). Kan et al. [18] used mismatch repair detection (MRD) tech- nology to identify somatic mutations in 441 tumors, including 183 breast cancers. They found three mutations in three breast tissue samples (1.6 %). Stephens et al. [8] The methods typically used to detect APC mutations have some weakness. PCR-RFLP and PCR-SSCP are sim- ple to perform but require gel electrophoresis to separate the PCR amplicons; they are therefore not suitable for Chang et al. Cancer Cell Int (2016) 16:23 Page 4 of 6 Fig. 1 HRM assays and sequence traces for APC exon 4. a Difference plot showed two different melting profiles, wild-type (WT) samples was blue, mutation in other colors. Direct DNA sequencing confirmed the b WT and the presence of the APC exon 4 mutation: c c.465A>G Fig. 1 HRM assays and sequence traces for APC exon 4. a Difference plot showed two different melting profiles, wild-type (WT) samples was blue, mutation in other colors. Direct DNA sequencing confirmed the b WT and the presence of the APC exon 4 mutation: c c.465A>G Fig. 1 HRM assays and sequence traces for APC exon 4. a Difference plot showed two different melting profiles, wild-type (WT) samples was blue, mutation in other colors. Direct DNA sequencing confirmed the b WT and the presence of the APC exon 4 mutation: c c.465A>G may produce only small differences in melting tempera- ture (TM). This will affect the accuracy of genotyping which depends on the temperature resolution capability of the instruction. Instrument resolution is of particular concern in the case of thermal block instruments with 96 or 384 wells. In this study, we failed to identify the sub- stitution of a single nucleotide in exons 11 (c.1458T>C), 13 (1635G>A), and 15 (c.4479G>A and c.5465T>A) by HRM analysis because homozygotes could not be dis- tinguished. To solve this problem, homozygous samples were mixed with a known genotype before HRM analy- sis. Discussion When an equal amount of wild-type DNA was added, the wild-type samples remained the same in the normal- ized and temperature-shifted difference plots, while the homozygous mutant samples exhibited heteroduplex for- mation; however, this approach doubled the amount of work. investigating large genes with many exons, such as APC, or a large number of samples. In addition, PCR-RFLP can only detect known mutations. PTT requires the use of radioactive labeling for protein detection, after which the DNA from samples with positive PTT results is sequenced. The yeast colorimetric assay consists of four steps: PCR, homologous recombination, fusion protein expression, and a colorimetric assay. The MRD method is a robust and cost-effective, but it is not efficient in detect- ing insertions/deletions. Exome- or APC-specific target sequencing using next-generation sequencing requires expensive equipment and bioinformatic analysis; its clini- cal application is therefore limited. Compared with the above-mentioned molecular diag- nostic methods, HRM analysis is relatively cost-effective and technically feasible. Although HRM analysis repre- sents the next generation of mutation scanning technol- ogy, it also has several limitations [19, 20]. For example, many homozygous variants and small insertions/dele- tions are more difficult to detect by HRM analysis and Although HRM analysis could be used to identify het- erozygotes, different heterozygotes may produce simi- lar melting curves to each other. In this study, the exon Page 5 of 6 Chang et al. Cancer Cell Int (2016) 16:23 Chang et al. Cancer Cell Int (2016) 16:23 11 amplicon was found to contain one SNP, c.1458T>C (p.Y486Y; rs2229992) and one mutation, c.1488A>T (p.T496T; rs9282599). Our HRM analysis was successful in distinguishing variant-specific heterozygosity. Tindall et al. [21] also demonstrated that different heterozy- gotes were distinguishable by HRM analysis. Unfortu- nately, HRM analysis is unable to identify which codon is mutated; therefore, direct DNA sequencing should be performed to confirm the position of the mutation when different heterozygotes produce highly similar melting curves. In addition, HRM analysis cannot detect muta- tions surrounding an entire exon if the exon is too large or if entire exons and genes have been deleted. This would require the design of primer pairs that surround the entire exon. Amplicon length influences the sensi- tivity of HRM analysis, with amplicons of 100–300 base pairs considered to be optimal. The shorter the ampli- con, the greater the Tm differences among genotypes and therefore the better the differentiation between mutant and non-mutant samples. HRM analysis PCR d PCR and a melting analysis of APC mutations were per- formed using the LightCycler® 480, a real-time PCR machine with HRM and 96/384 well capacity. The primers used in this study were previously described [26] and were of standard molecular biology quality (Protech Technol- ogy Enterprise Co., Ltd., Taipen City, Taiwan). Controls samples were tested for mutations in exons 4, 5, 9, 11, and 13, and the MCR of exon 15 using HRM analysis. PCR was performed with ~10 ng of DNA and 0.25 μM each of the relevant forward and reverse primers in a total reac- tion volume of 10 μl. Each reaction consisted of 5 μl of LightCycler® 480 high-resolution melting master reagent (Reference 04909631001; Roche Diagnostics, Basel, Swit- zerland) and 2.5 mM MgCl2. PCR was performed under the following conditions: 95 °C for 10 min, followed by 45 cycles of 95 °C for 15 s, 60 °C for 15 s, and 72 °C for 15 s, with a melting program at 95 °C for 1 min, 40 °C for 1 min, and 60–90 °C (25 acquisitions/°C). y The four APC SNPs (c.1458T>C, c.1635G>A, c.4479G>A, and c.5465T>A) had a minor allele fre- quency of 29.21, 16.29, 13.48, and 7.87 %, respectively. The allele frequencies were similar to those seen in the phase 3 data of the 1000 Genomes Project (Table 2). This is the first investigation of the association between SNPs in APC and breast cancer. Previous studies have addressed the association between SNPs in APC and colorectal cancer risk [22, 23]. c.3920T>A is the most common APC variant described in breast cancer. Pre- vious studies investigated the association between this variant and susceptibility to breast cancer among Ashke- nazi Jews [24, 25]; however, no significant contribution to breast cancer susceptibility was found for the four SNPs. However, large-scale studies are needed to validate these findings. Discrimination of two homozygotes was achieved by generating artificial heterozygotes. Before the HRM anal- ysis, the two homozygous genotypes (1458T vs. 1458C, 1635G vs. 1635A, 4479G vs. 4479A and 5465T vs. 5465A) were mixed with a known genotype (homozygous CC DNA, homozygous AA DNA, homozygous AA DNA and homozygous AA DNA, respectively). Sample preparation and DNA isolation Sample preparation and DNA isolation Breast tissue samples and matched peripheral blood sam- ples from 89 breast cancer patients who had undergone surgery at China Medical University Hospital (Taichung City, Taiwan) from 2003 to 2009 were included in this study. The tissues were frozen immediately after surgi- cal resection and stored in liquid nitrogen until DNA extraction. As controls, 50 healthy individuals were also recruited. Genomic DNA was extracted using a commer- cially available kit (GE Healthcare, Little Chalfont, UK). The quality of the isolated genomic DNA was assessed using agarose gel electrophoresis, and the concentration was determined using a Nano-Drop-1000 spectropho- tometer (Nano-Drop Technologies Inc., Wilmington, DE, USA). This study was approved by the Institutional Review Board of the China Medical University. Discussion Moreover, long ampli- cons may contain more melting domains, leading to rather complex melting profiles. Factors that affect the melting behavior of double-stranded (ds) DNA include the MgCl2 concentration and the quality and quantity of the DNA. The amounts of DNA in the samples should not differ significantly, in order to achieve similar thresh- old cycles. Statistical analysis All statistical analyses were carried out using the statisti- cal software program SPSS 17.0 for Windows (SPSS, Inc., Chicago, IL, USA). Differences in the genotypes of breast cancer patients vs. controls were evaluated using a χ2 test. Fisher’s exact test was used when the expected number in any cell was <5. A P value of <0.05 was considered to indicate statistical significance. 11. Sorlie T, Bukholm I, Borresen-Dale AL. Truncating somatic mutation in exon 15 of the APC gene is a rare event in human breast carcinomas. Mutations in brief no. 179. Online. Hum Mutat. 1998;12(3):215. 11. Sorlie T, Bukholm I, Borresen-Dale AL. Truncating somatic mutation in exon 15 of the APC gene is a rare event in human breast carcinomas. Mutations in brief no. 179. Online. Hum Mutat. 1998;12(3):215. 12. Furuuchi K, Tada M, Yamada H, Kataoka A, Furuuchi N, Hamada J, Takahashi M, Todo S, Moriuchi T. Somatic mutations of the APC gene in primary breast cancers. Am J Pathol. 2000;156(6):1997–2005. 13. Palacio-Rua KA, Isaza-Jimenez LF, Ahumada-Rodriguez E, Muneton-Pena CM. Genetic analysis in APC, KRAS, and TP53 in patients with stomach and colon cancer. Rev Gastroenterol Mex. 2014;79(2):79–89. Competing interests g The authors declare that they have no competing interests. 23. Feng M, Fang X, Yang Q, Ouyang G, Chen D, Ma X, Li H, Xie W. Association between the APC gene D1822V variant and the genetic susceptibility of colorectal cancer. Oncol Lett. 2014;8(1):139–44. Received: 5 December 2014 Accepted: 10 March 2016 Received: 5 December 2014 Accepted: 10 March 2016 Received: 5 December 2014 Accepted: 10 March 2016 24. Petrukhin L, Dangel J, Vanderveer L, Costalas J, Bellacosa A, Grana G, Daly M, Godwin AK. The I1307K APC mutation does not predispose to colorec‑ tal cancer in Jewish Ashkenazi breast and breast-ovarian cancer kindreds. Cancer Res. 1997;57(24):5480–4. 25. Redston M, Nathanson KL, Yuan ZQ, Neuhausen SL, Satagopan J, Wong N, Yang D, Nafa D, Abrahamson J, Ozcelik H, et al. The APCI1307K allele and breast cancer risk. Nat Genet. 1998;20(1):13–4. Authors’ contributions 16. Kashiwaba M, Tamura G, Ishida M. Aberrations of the APC gene in primary breast carcinoma. J Cancer Res Clin Oncol. 1994;120(12):727–31. YSC performed the experiments and wrote the manuscript. CYL contributed to the technical support. SFY participated in the statistical analysis. CMH par‑ ticipated in the coordination of the study. JGC designed the study. All authors read and approved the final manuscript. 17. Hayes MJ, Thomas D, Emmons A, Giordano TJ, Kleer CG. Genetic changes of Wnt pathway genes are common events in metaplastic carcinomas of the breast. Clin Cancer Res. 2008;14(13):4038–44. 18. Kan Z, Jaiswal BS, Stinson J, Janakiraman V, Bhatt D, Stern HM, Yue P, Hav‑ erty PM, Bourgon R, Zheng J, et al. Diverse somatic mutation patterns and pathway alterations in human cancers. Nature. 2010;466(7308):869–73. Direct DNA sequencingi 5. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW. Cancer genome landscapes. Science. 2013;339(6127):1546–58. 5. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW. Cancer genome landscapes. Science. 2013;339(6127):1546–58. To confirm the results of the HRM analysis, the same amplicons were sequenced. The PCR products were cleaned with a PCR-M™ clean-up system (Viogen, Sun- nyvale, CA, USA). Sequencing reactions were set up using a BigDye Terminator, version 3.1, cycle sequencing kit (Applied Biosystems, Foster City, CA, USA). BigDye® XTerminator™ (Applied Biosystems) was used to purify the reaction products. Sequencing analyses were carried out on an ABI Prism 3130 genetic analyzer (Applied Biosystems). g 6. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–74. 6. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–74. 7. Byler S, Goldgar S, Heerboth S, Leary M, Housman G, Moulton K, Sarkar S. Genetic and epigenetic aspects of breast cancer progression and therapy. Anticancer Res. 2014;34(3):1071–7. 7. Byler S, Goldgar S, Heerboth S, Leary M, Housman G, Moulton K, Sarkar S. Genetic and epigenetic aspects of breast cancer progression and therapy. Anticancer Res. 2014;34(3):1071–7. 8. Stephens PJ, Tarpey PS, Davies H, Van Loo P, Greenman C, Wedge DC, Nik-Zainal S, Martin S, Varela I, Bignell GR, et al. The landscape of cancer genes and mutational processes in breast cancer. Nature. 2012;486(7403):400–4. 8. Stephens PJ, Tarpey PS, Davies H, Van Loo P, Greenman C, Wedge DC, Nik-Zainal S, Martin S, Varela I, Bignell GR, et al. The landscape of cancer genes and mutational processes in breast cancer. Nature. 2012;486(7403):400–4. 9. Fodde R, Smits R, Clevers H. APC, signal transduction and genetic instabil‑ ity in colorectal cancer. Nat Rev Cancer. 2001;1(1):55–67. 9. Fodde R, Smits R, Clevers H. APC, signal transduction and genetic instabil‑ ity in colorectal cancer. Nat Rev Cancer. 2001;1(1):55–67. 10. King TD, Suto MJ, Li Y. The Wnt/beta-catenin signaling pathway: a poten‑ tial therapeutic target in the treatment of triple negative breast cancer. J Cell Biochem. 2012;113(1):13–8. 10. King TD, Suto MJ, Li Y. The Wnt/beta-catenin signaling pathway: a poten‑ tial therapeutic target in the treatment of triple negative breast cancer. J Cell Biochem. 2012;113(1):13–8. Additional file 14. Jiao Y, Yonescu R, Offerhaus GJ, Klimstra DS, Maitra A, Eshleman JR, Her‑ man JG, Poh W, Pelosof L, Wolfgang CL, et al. Whole-exome sequenc‑ ing of pancreatic neoplasms with acinar differentiation. J Pathol. 2014;232(4):428–35. Additional file 1. HRM assays for APC exons 1, 2, 3 and 5-15. 15. Kan Z, Zheng H, Liu X, Li S, Barber TD, Gong Z, Gao H, Hao K, Willard MD, Xu J, et al. Whole-genome sequencing identifies recurrent mutations in hepatocellular carcinoma. Genome Res. 2013;23(9):1422–33. Author details 1 1 Epigenome Research Center, China Medical University Hospital, 2 Yuh‑Der Road, Taichung 404, Taiwan. 2 Department of Laboratory Medicine, China Medical University Hospital, Taichung, Taiwan. 3 School of Medicine, China Medical University, Taichung, Taiwan. 19. Wittwer CT. High-resolution DNA melting analysis: advancements and limitations. Hum Mutat. 2009;30(6):857–9. 20. Er TK, Chang JG. High-resolution melting: applications in genetic disor‑ ders. Clin Chim Acta. 2012;414:197–201. Conclusions Upon completion of the run, the samples were ana- lyzed using the gene scanning software (version 1.5, Roche Diagnostics), supplied with the LightCycler® 480. The melting curve analysis included: the normalization of melting curves equal to 100 % of the initial fluorescence and to 0 % of the remaining fluorescence. The tempera- ture axis of the normalized melting curves was shifted such that the entire dsDNA was completely denatured. Difference plots were obtained by subtracting the curves Our results indicate that the prevalence of APC muta- tions in Taiwanese breast cancer patients is rare, simi- lar to other populations. It can thus be concluded that, at least in these patients, the APC gene does not play an important role in the oncogenesis of breast cancer. We also identified two somatic APC mutations. HRM analy- sis is a rapid and cost-effective method for the mutational analysis of APC. Page 6 of 6 Chang et al. Cancer Cell Int (2016) 16:23 of the wild-type and mutant DNAs and were used to cluster the samples into groups. 3. Jung YS, Na KY, Kim KS, Ahn SH, Lee SJ, Park HK, Cho YU. Nation-wide Korean breast cancer data from 2008 using the breast cancer registration program. J Breast Cancer. 2011;14(3):229–36. 4. Viale G. The current state of breast cancer classification. Ann Oncol. 2012;23(Suppl 10):x207–10. 4. Viale G. The current state of breast cancer classification. Ann Oncol. 2012;23(Suppl 10):x207–10. Acknowledgements 21. Tindall EA, Petersen DC, Woodbridge P, Schipany K, Hayes VM. Assessing high-resolution melt curve analysis for accurate detection of gene vari‑ ants in complex DNA fragments. Hum Mutat. 2009;30(6):876–83. This study was supported in part by Taiwan Ministry of Health and Wel‑ fare Clinical Trial and Research Center of Excellence (MOHW105-TDU- B-212-133019), Ministry of Science and Technology of Taiwan (MOST104- 2314-B-039-001, and the China Medical University Hospital (DMR-100-129). 22. Wong HL, Peters U, Hayes RB, Huang WY, Schatzkin A, Bresalier RS, Velie EM, Brody LC. Polymorphisms in the adenomatous polyposis coli (APC) gene and advanced colorectal adenoma risk. Eur J Cancer. 2010;46(13):2457–66. 2. Yang L, Parkin DM, Ferlay J, Li L, Chen Y. Estimates of cancer incidence in China for 2000 and projections for 2005. Cancer Epidemiol Biomarkers Prev. 2005;14(1):243–50. 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin. 2011;61(2):69–90. References 26. Chang YS, Lin CY, Yang SF, Ho CM, Chang JG. High-resolution melting analysis for gene scanning of adenomatous polyposis coli (APC) gene with oral squamous cell carcinoma samples. Appl Immunohistochem Mol Morphol. 2014.
https://openalex.org/W4391225853	https://emergent-scientist.edp-open.org/articles/emsci/pdf/2024/01/emsci230001.pdf	English	null	Fluorescence emission modification of triphenylamine derivatives by aggregate formation in solution or by mechanical stress in solid state	Emergent scientist	2,024	cc-by	8,585	1 Introduction luminogens and mechanoluminochroms change color quite drastically, researchers have been investigating whether a gradual change in color is possible [11–13]. Organic molecules able to change their interaction with light according to external stimuli have aroused interest for a long time due to their possible and diverse applica- tions in sensors, data storage tools and data protection [1,2]. Mechanoluminochroms are part of these types of organic molecules whose color of luminescence changes when pressure is applied on the solid state [3]. As quench- ing, and more precisely self-quenching, allows other path- ways of deexcitation for molecules with large delocaliza- tion, a phenomenon known as ACQ (aggregation-caused quenching), the solid state or the thin film state of such molecules generally have little to no luminescent proper- ties [4]. However, a new category of molecules was firstly reported in 2001 to display aggregation-induced emis- sion (AIE) and aggregation-induced enhanced emission (AIEE) [5,6]. In the case of AIEE, the compound displays a shifted, higher intensity fluorescence in the aggregated state than in solution. AIE luminogens have been used as a basis for potential mechanofluorochroms [7,8]. A corre- lation between AIE properties and piezofluorochromism has been indeed identified (PAIE: piezofluorochromic aggregation-induced emission) as the two properties are often coupled [9,10]. Since PAIE luminogens as well as AIE g g p [ ] The derivatives of pyrene or triphenylamine are the most commonly studied molecules in this context because of their strong AIE properties. Due to their aggregation, the geometries of the electronic ground state and excited states are modified and the restriction of intramolecu- lar rotations between some groups leads to a decrease in the nonradiative decay rate. The modification of the luminescence of organic molecules such as triphenylamine 1 is due to changes in the electronic properties of the ground state and the excited state and this effect is enhanced with the delocalisation of the electronic density over the π system [14]. It could therefore be interesting to add specific functions to triphenylamine 1 to inves- tigate their effect on the luminescence (AIE, AIEE and mechanoluminochromism) of the new compound. The two molecules, 4-(diphenylamino)benzaldehyde 2 and 2-(4- (diphenylamino)benzylidene)malononitrile 3 (Scheme 1), chosen for their acceptor mesomeric groups, were stud- ied by Cao et al. [10], who investigated modifica- tions in their luminescence under grinding, anneal- ing or solvent fuming. Received: 4 May 2023 / Accepted: 25 November 2023 Abstract. Two derivatives of triphenylamine, 4-(diphenylamino)benzaldehyde and 2-(4- (dipheny- lamino)benzylidene)malononitrile, with acceptor mesomeric groups, an aldehyde moiety and a dicyanovinyl group respectively, were synthesized and found to have aggregation-induced emission (AIE) or aggregation- induced enhanced emission (AIEE) properties in several DMF/H2O mixtures. The size of the formed aggregates was studied with Dynamic Light Scattering (DLS) and a significant change of approximately 200 nm was observed according to the water fraction in DMF. Mechanoluminochromism characteristics in the solid state were also observed. A shift in the emission bands was noticed when comparing the solid before and after grinding, as well as when comparing the crystalline and amorphous form. Keywords: AIE, AIEE, Fluorescence, Mechanoluminochromism, Triphenylamine Keywords: AIE, AIEE, Fluorescence, Mechanoluminochromism, Triphenylamine This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Emergent Scientist 8, 1 (2024) © L. Estournet et al., published by EDP Sciences, 2024 https://doi.org/10.1051/emsci/2023004 Available online at: Emergent Scientist 8, 1 (2024) © L. Estournet et al., published by EDP Sciences, 2024 https://doi.org/10.1051/emsci/2023004 Emergent Scientist 8, 1 (2024) © L. Estournet et al., published by EDP Sciences, 2024 https://doi.org/10.1051/emsci/2023004 * e-mail: ines.mezghani@ens-paris-saclay.fr n Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly ci Lise Estournet1, Mathieu Loriot1, In`es Mezghani1,2,, Audrey Pollien1,2 , and Nina Villette1,2 1 Universit´e Paris-Saclay, 91400, Orsay, France 2 Universit´e Paris-Saclay, ENS Paris-Saclay, DER Chimie, 91190, Gif-sur-Yvette, France Available online at: www.emergent-scientist.edp-open.org RESEARCH ARTICLE 1 Introduction AIE or AIEE properties were studied with mixtures made with different water frac- tions in dimethylformamide (DMF), but no correlation was established between the fluorescence properties and L. Estournet et al.: Emergent Scientist 8, 1 (2024) 2 Scheme 1. Chemical structures of compounds 1–3. values: ηH2O = 0.890 mPa.s and ηDMF = 0.801 mPa.s, at 25 ◦C. xH2O and xDMF correspond to the molar fractions of H2O and DMF respectively (with xDMF + xH2O = 1). D, kB, T and d are, respectively, the diffu- sion coefficient of the particle (m².s−1), the Boltzmann constant (J.K−1), the temperature (K) and the hydrody- namic diameter of the particle (m). The hydrodynamic diameters plotted on the spectra, dspectra, were therefore obtained by combining equations (1) and (2) to con- vert dDLS, the hydrodynamic diameters obtained as DLS output, according to equation (3): values: ηH2O = 0.890 mPa.s and ηDMF = 0.801 mPa.s, at 25 ◦C. xH2O and xDMF correspond to the molar fractions of H2O and DMF respectively (with xDMF + xH2O = 1). D, kB, T and d are, respectively, the diffu- sion coefficient of the particle (m².s−1), the Boltzmann constant (J.K−1), the temperature (K) and the hydrody- namic diameter of the particle (m). The hydrodynamic diameters plotted on the spectra, dspectra, were therefore obtained by combining equations (1) and (2) to con- vert dDLS, the hydrodynamic diameters obtained as DLS output, according to equation (3): Scheme 1. Chemical structures of compounds 1–3. the size of aggregates formed in solution. This study aims to compare the luminescence properties in solution and in the solid state (crystalline and amorphous states) of the two derivatives. It also endeavors to find a correlation between the size of aggregates formed in solution and the emission intensity. dspectra = dDLS ηH2O η (3) (3) 2 Material and method For each compound, commercial 2 (Tokyo Chemical Industry, Zwijndrecht Belgium) and synthesized 3, a range of H2O/DMF solutions (100 mL) were prepared using distilled water and purchased DMF (Carlo Erba, Val de Reuil, France), at a concentration of C2,0 = 22.9 µmol.L−1 and C3,0 = 19.0 µmol.L−1, respectively. The commercial compound 2 was used because these experiments were car- ried out in parallel with the synthesis of compound 2. As AIE phenomena are particularly sensitive to their environ- ment, the commercial product made it possible to reduce the apparition of impurities in the spectra. The follow- ing water volume fractions were tested in the H2O/DMF mixtures: 0%, 25%, 50%, 75%, 90%, 95% and 99%. For each 2 or 3 solution, emission and excitation spectra were recorded. DLS measurements were carried out for the fol- lowing water volume fractions: 75% – 99% and 75% – 95%, for 2 and 3, respectively. In the following study, the acronyms cr, gr and am stand respectively for crystalline, grinded and amorphous, refer- ring to the state of the solid compounds. The studied 1 and 2 solids are supposed to be initially crystalline. 2.1 General method Absorbance spectra were recorded with a Libra S12 biochrom spectrophotometer. The emission and excita- tion spectra of the solutions were recorded in a 1 cm path length quartz cuvette with a Cary Eclipse Varian fluorimeter (Appendices 1 and 2). The emission and exci-tation spectra of the solids deposited on a glass slide were recorded with a Xenius Safas Monaco spectropho- tometer. NMR spectra were recorded with a Spinsolve Carbon benchtop NMR spectrometer (80 MHz) from Magritek. IR spectra were recorded with a Nicolet Avatar 330 FT-IR spectrometer from Thermo Fischer Scientific. To obtain amorphous phase solids 1am, 2am and 3am, a sample of each compound, 1, 2 and 3, was deposited on a glass slide, melted on a Kofler bench, and quickly solidi- fied or not, after removing the glass slide from the bench. Light scattering (DLS) measurements were recorded on the DMF/H2O solutions of 2 and 3 with a Zetasizer nano series Malvern apparatus in a 1 cm length quartz cuvette. The parameters of the apparatus were the follow- ing: water as dispersant; temperature of 25 ◦C; automatic mode. Water was chosen as the dispersant since it is the major solvent in the mixture. The spectra plotted in this study were corrected using the Arrhenius viscosity model (1) for solvents (DMF/H2O) and the Stokes-Einstein law (2). The formulas used were the following: 3.1 AIE and AIEE study The emission spectra of 2 and 3 in several DMF/H2O mix- tures with different water volume fractions are reported in Figures 1a and 2a, respectively. Scheme 2. Synthesis pathways of the derivatives of tripheny- lamine 1. Regardless of the composition of the DMF/H2O mix- ture, the emission spectrum of compound 2 are at lower wavelengths, and therefore higher energies, than for com- pound 3. This red shift is probably due to the increase in the conjugation system from 2 to 3, but also because the dicyanovinyl group is more electron-withdrawing than the aldehyde [17,18]. To determine the influence of the acceptor mesomeric groups of 2 and 3 on the relative positions of their emission bands, quantum chemical cal- culations were performed to examine the shape of the frontier molecular orbitals. where ω is the frequency of the applied electromagnetic field and ⟨ψf \| r \| ψi⟩is the matrix element of the transi- tion dipole moment between the electronic states \|ψi⟩and \|ψf⟩. The latter represents the degree of spatial overlap between the wave functions of these states. 2.5 4-diphenylamino-benzaldehyde (2) Triphenylamine 1 (4.9625 g, 20.23 mmol) was dissolved in DMF (8.5 mL, 109 mmol). Phosphorus oxychloride (3.4 mL, 18.3 mmol) was added dropwise, and the mixture was stirred and heated at 100 ◦C for 1 h. The reaction was monitored by TLC (Thin Layer Chromatography) (eluent: petroleum ether/AcOEt 99:1). After cooling, the mixture was dissolved in CH2Cl2 (50 mL) and washed with a dilute sodium hydrogen bicarbonate solution (2 × 50 mL), and then with distilled H2O (25 mL). The organic fraction was dried over MgSO4, filtered, and concentrated in vacuo. Column chromatography eluting with CH2Cl2 provided the title compound 2 (1.1153 g, yield = 20%) as light-yellow crystals, m.p. (melting point) 135–136 ◦C. 1H NMR (80 MHz, CDCl3) δ 9.93 (s, 1H, CHO), 7.02 – 7.91 (m, 14H, Ph). IR (solid state): 1690 cm−1 (CHO), 3040 cm−1 (Ph). Additionally, it is noteworthy that the implicit con- sideration of solvation in water or DMF reduced the HOMO-LUMO energy gap by approximately 0.10 eV for 2, and 0.12 eV for 3 (Appendix 9). Taking into account the solvation also reduced the energy of the first excited state for molecule 3 by approximately 0.2 eV, while it had no effect on the second excited state. It is difficult however to confirm a similar trend for molecule 2, since the first excited state calculated in the gas phase did not appear to be due to the same transition as in the solvent (Appendix 11). In fact, the first excited state for 2 in the gas phase was mainly due to the HOMO-2→LUMO tran- sition, while it was the HOMO →LUMO transition in the solvent. Despite this particularity, if the HOMO→LUMO transition is tracked for 2, a behavior like that of the molecule 3 is visible: a red shift of more than 0.1 eV for the intense transition (with the higher oscillator strength) was 2.3 Theoretical calculations The molecular structures of 1, 2 and 3 were optimized using the density functional theory (DFT) with the Coulomb Attenuating Method Becke’s three-parameter exchange, Lee, Yang, and Parr correlation hybrid func- tional (CAM-B3LYP), commonly used for the prediction of excited states of molecules with intramolecular charge transfer, in combination with the 6-31G(d,p) basis set, using Gaussian 16. Then, the oscillator strengths for the two first electronic excited states were calculated for all derivatives employing time-dependent DFT (TD-DFT) at the CAM-B3LYP/6-31G(d,p) level of theory. Solvent effects were taken into account through the polarizable continuum model (PCM) using either water (dielectric constant: εH2O = 78.355) or N,N-dimethylformamide (εDMF = 37.219). We recall that the oscillator strength, fif, is a dimensionless quantity which expresses the prob- ability of absorption or emission of an electromagnetic radiation between the initial electronic state \|ψi⟩and the final excited electronic state \|ψf⟩. It can be defined as follows: ln(η) = xH2Oln(ηH2O) + xDMFln(ηDMF) (1) ln(η) = xH2Oln(ηH2O) + xDMFln(ηDMF) (1) (1) D = kBT 3πdη (2) D = kBT 3πdη (2) with η, ηH2O, ηDMF the dynamic viscosity of the mixture, H2O and DMF, respectively (in mPa.s). The viscosities of H2O and DMF were obtained from the lit- erature [15,16] and were assumed to have the following fif ∝2 3ω\|⟨ψf \| r \| ψi⟩\|2 (4) (4) L. Estournet et al..: Emergent Scientist 8, 1 (2024) L. Estournet et al..: Emergent Scientist 8, 1 (2024) 3 Scheme 2. Synthesis pathways of the derivatives of tripheny- lamine 1. 2.4 Syntheses A decrease in the HOMO-LUMO energy gap by increas- ing the attractive character of the substituents was obtained with the theoretical calculations (Appendix 9). The lower this gap, the more the spectra will shift at high wavelengths, therefore towards the red. The observed red shift was also rationalized in the ground state with the increase in the electric dipole moment from derivative 1 to 3 (Appendix 10). Considering the HOMO-LUMO tran- sition, a notable redistribution of the electronic density in the LUMO shape compared to that of the HOMO was observed over the aldehyde group of compound 2, and even more pronouncedly over the dicyanovinyl group of 3, which further supports the higher electron-attractivity of the latter (Appendix 9). Moreover, the energy of the first excited state clearly exhibited a greater red shift for 3 owing to the more pronounced electron-withdrawing nature of the dicyanovinyl group compared to the alde- hyde. However, the hypothesis asserting that a more pronounced red shift takes place with the more electron- attracting groups remains valid solely when the geometry of the excited state closely resembles that of the ground state, and when the observed transition corresponds to the HOMO-LUMO transition. Hazard classes, category codes and suppliers of the used chemicals are mentioned in Appendix 3. Scheme 2 shows the synthesis pathways described below. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile Commercial 4-diphenylamino-benzaldehyde 2 (1.0 g, 3.6 mmol) was added to the ethanol solution (60 mL) of mal- ononitrile (0.3 g, 5.4 mmol). The mixture was stirred at 60 ◦C and the reaction was monitored by TLC (eluent: CH2Cl2). After 2 h, the reaction was complete and the mixture was cooled down to room temperature. After a step of trituration in petroleum ether, an orange powder was obtained through vacuum filtration (0.646 g, yield = 55%), m.p. 143 ◦C. 1H NMR (80 MHz, CDCl3) (Cf. Dead End). IR (solid state): 1591 cm−1 (C=C), 2220 cm−1 (CN), 3053 cm−1 (Ph). The NMR and IR spectra of the products are reported in Appendices 4–8. L. Estournet et al.: Emergent Scientist 8, 1 (2024) 4 Fig. 1. Study of 2 in DMF/H2O mixtures. a) Normalized emission spectra (λexc = 355 nm) with different water fractions (0% – 99%). Concentration of 2: C2,0 = 22.9 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 2 for different water fractions (75% – 99%), measured with DLS. Concentration of 2: C2,0 = 22.9 µmol.L−1. Fig. 2. Study of 3 in DMF/H2O mixtures. a) Normalized emission spectra (λexc= 435 nm) with different water fractions (0% – 95%). Concentration of 3: C3,0 = 19.0 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 3 with different water fractions (75% – 95%), measured with DLS. Concentration of 3: C3,0 = 19.0 µmol.L−1. Fig. 1. Study of 2 in DMF/H2O mixtures. a) Normalized emission spectra (λexc = 355 nm) with different water fractions (0% – 99%). Concentration of 2: C2,0 = 22.9 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 2 for different water fractions (75% – 99%), measured with DLS. Concentration of 2: C2,0 = 22.9 µmol.L−1. Fig. 1. Study of 2 in DMF/H2O mixtures. a) Normalized emission spectra (λexc = 355 nm) with different water fractions (0% – 99%). Concentration of 2: C2,0 = 22.9 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 2 for different water fractions (75% – 99%), measured with DLS. Concentration of 2: C2,0 = 22.9 µmol.L−1. Fig. 2. Study of 3 in DMF/H2O mixtures. a) Normalized emission spectra (λexc= 435 nm) with different water fractions (0% – 95%). Concentration of 3: C3,0 = 19.0 µmol.L−1. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile Indeed, if the nucleation rate is slow compared to the growth rate, the firstly formed nuclei grow while new nuclei are formed in parallel. The firstly formed nuclei lead therefore to the biggest aggregates, while the nuclei formed afterwards lead to smaller ones because there are less and less precursors (molecules 2 or 3) in the solution. The diameter distribution is therefore very broad. Regard- ing the different diameter distribution for a water fraction of 75% and since the results obtained with the four differ- ent mixtures are limited, it remains arduous to establish an overall trend. To determine a clear trend, a standard range of solutions with different DMF/H2O mixtures is required. was observed, whereas an increase by 32% was recorded for 2 aggregates (Tab. 1). It could be suggested that the slight increase in the emission intensity of 3 (+4%) was due to the relatively small modification in the aggregate diameters between 90% and 95% of water fraction. On the contrary, the emission intensity increase was more pronounced for 2 (+63%), following the important modi- fication in diameters between 90% to 95%. It seems that there is also a correlation between the size of the aggre- gates and the red shift, which would support the trend observed between 90% and 95%. Although the red shift for both compounds was of the same order of magnitude in terms of wavelength (2-3 nm), the red shift for compound 2 was 134 cm−1, while it was 58 cm−1 for compound 3. Fig. 3. AIEE and AIE phenomena for the triphenylamine derivatives 2 and 3 as a function of the water volume fraction in the DMF/H2O mixture, under UV light (365 nm). visible to the naked eye (Fig. 3). As expected, a nonlinear progressive rise in their emission intensity was noticed. The obtained results were consistent with those reported in the literature [10]. The arrows in Figures 1a and 2a point out the evolution of the maximum wavelength for the fluorescence emission upon water volume fraction increase. Distinct behaviors for this evolution were observed for compounds 2 and 3. For 2, a slight shift was observed between 75% and 90% water fraction (from 468 nm/ 21 368 cm−1 to 473 nm/ 21 142 cm−1), but no distinguishable shift was visible from 90% to 99% water fraction. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile For 3, the more the water fraction increased, the more the maximum emission wavelength was red shifted (passing from 551 nm/ 18 149 cm−1 to 589 nm/ 16 978 cm−1). , p However, for a water fraction of 75%, an abrupt change occurred with a much more heterogenous and asymmet- ric diameter distribution, with a maximum at 363 nm for 2 and 3. For both compounds, the major hydrody- namic diameter had a high value of standard deviation compared to the other water fractions: 291 and 351 nm, for 2 and 3 respectively. The heterogeneity of the dis- tribution could be due to a slower nucleation rate [19] because of a relative stabilization of the compounds by DMF molecules present at a non-negligible fraction of 25%. Indeed, if the nucleation rate is slow compared to the growth rate, the firstly formed nuclei grow while new nuclei are formed in parallel. The firstly formed nuclei lead therefore to the biggest aggregates, while the nuclei formed afterwards lead to smaller ones because there are less and less precursors (molecules 2 or 3) in the solution. The diameter distribution is therefore very broad. Regard- ing the different diameter distribution for a water fraction of 75% and since the results obtained with the four differ- ent mixtures are limited, it remains arduous to establish an overall trend. To determine a clear trend, a standard range of solutions with different DMF/H2O mixtures is required. The AIE and AIEE phenomena, and thus the presence of aggregates, were confirmed through DLS analysis after the viscosity correction. According to Figures 1b and 2b, aggregates with diameters of the order of 100 nm formed in the water-rich DMF/H2O mixtures. In the case of 2 (Fig. 1b and Table 1), there was an increase in the major hydrodynamic diameter of the aggregates from 146 nm to 398 nm for water fractions between 90% and 99%, with an increase in the width of the gaussian distribution. This result suggests that the larger the aggregates, the higher the emission intensity: between 90% and 99% of water fraction, the maximal diameter increased by 173% and the maximum emission intensity by 141%. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile This trend may be due to a limited solubility of 2 in water: the more water there is, the bigger the aggregates are, since it might be energetically more favourable for 2 molecules to aggregate, rather than being in solution in the water- rich DMF/H2O mixture. Therefore, the higher the water fraction is, the larger the aggregates are, and the less the molecules in solution are free. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile b) Distribution of hydrodynamic diameters of the formed aggregates of 3 with different water fractions (75% – 95%), measured with DLS. Concentration of 3: C3,0 = 19.0 µmol.L−1. Fig. 2. Study of 3 in DMF/H2O mixtures. a) Normalized emission spectra (λexc= 435 nm) with different water fractions (0% – 95%). Concentration of 3: C3,0 = 19.0 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 3 with ( % %) 1 Fig. 2. Study of 3 in DMF/H2O mixtures. a) Normalized emission spectra (λexc= 435 nm) with different water fractions (0% – 95%). Concentration of 3: C3,0 = 19.0 µmol.L−1. b) Distribution of hydrodynamic diameters of the formed aggregates of 3 with different water fractions (75% – 95%), measured with DLS. Concentration of 3: C3,0 = 19.0 µmol.L−1. observed for 2 when solvents are taken into account. On the other hand, for the least intense transition a blue-shift of almost 0.1 eV took place. then drastically increased for water fractions between 75% and 99% to reach an intensity 6 times greater than that of the 75% water DMF/H2O mixture. / A relatively comparable phenomenon was observed for 3, but it was only AIE because 3 was not fluorescent in pure DMF (0% water). Starting from a water fraction of 50%, its emission intensity increased with a progres- sive red shift of the fluorescence maximum until the water fraction reached 95%. The effect of the increase in water fraction was greater for 2. The evolution of the fluorescence of the solutions of 2 and 3 was also The AIEE property of 2 and the AIE property of 3 were studied. A form of AIEE was observed for 2 in the pure DMF solution (0% water): 2 was fluorescent with a maximum emission peak at 530 nm. When the water fraction was between 25% and 75%, the fluorescence of 2 disappeared. However, when the water fraction reached 75%, the fluorescence of 2 was recovered, blue shifted by approximately 60 nm (2274 cm−1). The emission intensity 5 L. Estournet et al..: Emergent Scientist 8, 1 (2024) Fig. 3. AIEE and AIE phenomena for the triphenylamine derivatives 2 and 3 as a function of the water volume fraction in the DMF/H2O mixture, under UV light (365 nm). Table 1. Distribution of hydrodynamic diameters mea- sured with DLS for 2 and 3 aggregates in DMF/H2O mixtures Table 1. 2.6 2-(4-(diphenylamino)benzylidene)malononitrile Distribution of hydrodynamic diameters m sured with DLS for 2 and 3 aggregates in DMF/ mixtures Sample (water volume fraction) Major hydrodynamic diameter in nm (standard deviation) / Percentage of associated aggregates 2 (75%) 363 (± 291) / 16.9% 2 (90%) 146 (± 44) / 33.3% 2 (95%) 193 (± 112) / 24.3% 2 (99%) 398 (± 161) / 32.1% 3 (75%) 363 (± 351) / 15.7% 3 (90%) 263 (± 132) / 22.6% 3 (95%) 224 (± 112) / 23.6% Sample (water volume fraction) Major hydrodynamic diameter in nm (standard deviation) / Percentage of associated aggregates 2 (75%) 363 (± 291) / 16.9% 2 (90%) 146 (± 44) / 33.3% 2 (95%) 193 (± 112) / 24.3% 2 (99%) 398 (± 161) / 32.1% 3 (75%) 363 (± 351) / 15.7% 3 (90%) 263 (± 132) / 22.6% 3 (95%) 224 (± 112) / 23.6% was observed, whereas an increase by 32% was recorded for 2 aggregates (Tab. 1). It could be suggested that the slight increase in the emission intensity of 3 (+4%) was due to the relatively small modification in the aggregate diameters between 90% and 95% of water fraction. On the contrary, the emission intensity increase was more pronounced for 2 (+63%), following the important modi- fication in diameters between 90% to 95%. It seems that there is also a correlation between the size of the aggre- gates and the red shift, which would support the trend observed between 90% and 95%. Although the red shift for both compounds was of the same order of magnitude in terms of wavelength (2-3 nm), the red shift for compound 2 was 134 cm−1, while it was 58 cm−1 for compound 3. However, for a water fraction of 75%, an abrupt change occurred with a much more heterogenous and asymmet- ric diameter distribution, with a maximum at 363 nm for 2 and 3. For both compounds, the major hydrody- namic diameter had a high value of standard deviation compared to the other water fractions: 291 and 351 nm, for 2 and 3 respectively. The heterogeneity of the dis- tribution could be due to a slower nucleation rate [19] because of a relative stabilization of the compounds by DMF molecules present at a non-negligible fraction of 25%. 3.2 Mechanoluminochromism In the next stage, the fluorescent properties of the deriva- tives 2 and 3 in the solid state were determined: in their crystalline form, 2cr and 3cr, in the form obtained after grinding a sample with a pestle and a mortar, 2gr In the case of 3 (Fig. 1b and Table 1), the trend was the opposite of that of 2 between 90% and 95% water fractions: a decrease in the maximal diameter by 15% L. Estournet et al.: Emergent Scientist 8, 1 (2024) 6 le 2. Wavelengths and FWHM (Full Width at Half Maximum) of maximum emission bands for compoun he solid state (crystalline form, amorphous form, and after grinding). Compound 1 2 2’ 3 3’ λexc (nm) 355 365 365 365 365 λmax (nm) cr 385 446 462 584 581 FWHM (cm−1) / FWHM (nm) cr 1320 / 19 2831 / 59 2011 / 70 λmax (nm) gr - 442 478 589 602* FWHM (cm−1) / FWHM (nm) gr 2644 / 53 2046 /73 λmax (nm) am 467 - 611 - FWHM (cm−1) / FWHM (nm) am 3376 / 76 2248 / 90 Identified as amorphous after grinding [10]. *Obtention of a crystalline form. - Not performed (due to fast recrystallization) cr: crystalline; gr: grinded; am: amorphous. 2’ and 3’ are references from literature [10]. Table 2. Wavelengths and FWHM (Full Width at Half Maximum) of maximum emission bands for compounds 1–3 in the solid state (crystalline form, amorphous form, and after grinding). WHM (Full Width at Half Maximum) of maximum emission bands for compounds 1–3 orm, amorphous form, and after grinding). Under UV lamp (365 nm) Under natural light 2 (crystalline form) 3 (crystalline form) 3 (amorphous form) 2 (amorphous form) Fig. 4. Crystalline and amorphous forms of 2 and 3 under natural light and under UV light (365 nm). difference in the direction of the shift may be due to the form observed after grinding. It may be suggested that the powder obtained in this experiment after grinding could be another crystalline form or could contain more defects than the crystalline one obtained after the synthe- sis. X-ray diffraction measurements may be carried out to verify whether the form obtained after grinding is still crystalline. 3.2 Mechanoluminochromism Although the shape and the width of the emission bands were quite similar before and after grinding in both cases (Table 2), the compounds 2 and 3 were found to have slight mechanoluminochromism properties and no mechanochemical ones. Mechanochemistry involves a chemical modification of the molecule under mechanical stress, leading to a modifi- cation of the absorption spectra, which was not observed in this case: the compounds obtained after grinding had the same absorbance spectrum in solution as the com- pounds before grinding Appendices 12 and 13. In the case of 3, the spectrum of grinded solid (3gr) was located between the crystalline form (3cr) spectrum and that of the amorphous form (3am): it could be inferred that a part of the crystalline sample was transformed to the amorphous state [10]. Fig. 4. Crystalline and amorphous forms of 2 and 3 under natural light and under UV light (365 nm). and 3gr, and in the amorphous form, 2am and 3am. The wavelengths of the maxima in the emission spec- tra (Appendices 12, 13 and 14) are summarized in Table 2. Emitted colors were cyan blue for 2cr and 2am and orange for 3cr and 3am (Fig. 4). On the one hand, a trend can be observed. The emis- sion spectra of the amorphous forms were significantly red shifted for both derivatives 2am and 3am (a shift of 21 nm/1 008 cm−1 and 27 nm/757 cm−1 respectively) compared to those of the crystalline forms. 5 Conclusion 5. F. Song, B. Liu, B.Z. Tang, Celebration of the 20th anniversary of aggregation-induced emission with research highlights from royal society of chemistry journals, J. Mater. Chem. C 8, 7972–7974 (2020) To conclude, two derivatives of triphenylamine 2 and 3 substituted with acceptor mesomeric groups were synthe- sized to investigate their photophysical properties. Both were found to have aggregated-induced or aggregated- induced enhanced emission properties in a DMF/H2O mixture. A notable red shift was observed between the emission spectra of 2 and 3 in solution and in solid state, which was larger for 3 due to the greater electron attrac- tivity of the dicyanovinyl group compared to the aldehyde, as well as the increased conjugation. Solid crystalline forms 2cr and 3cr also displayed a slight mechano- luminochromic response to manual grinding. Moreover, their amorphous forms 2am and 3am were found to have a luminescence distinct from that of their crystalline forms 2cr and 3cr. The grinded sample 3gr might have partially become amorphous. 6. J. Luo, Z. Xie, J.W.Y. Lam, L. Cheng, B.Z. Tang, H. Chen, C. Qiu, H.S. Kwok, X. Zhan, Y. Liu, D. Zhu, Aggregation- induced emission of 1-methyl-1,2,3,4,5-pentaphenylsilole, Chem. Commun. 18, 1740–1741 (2001) 7. Z.-F. Chang, L.-M. Jing, C. Wei, Y.-P. Dong, Y.- C. Ye, Y.S. Zhao, J.-L. Wang, Hexaphenylbenzene- based, π-conjugated snowflake-shaped luminophores: tun- able aggregation-induced emission effect and piezofluo- rochromism, Chem. – Eur. J. 21, 8504–8510 (2015) ( ) 8. B. Prusti, M. Chakravarty, Disparity in piezofluo- rochromism for twisted mono-carbazole-based AIEgens resulting from interchanging electron-rich substituents: effect of coplanarity on twisted π-conjugates., Mater. Adv. 2, 1752–1759 (2021) ( ) 9. B. Xu, Z. Chi, J. Zhang, X. Zhang, H. Li, X. Li, S. Liu, Y. Zhang, J. Xu, Piezofluoro-chromic and aggregation- induced-emission compounds containing triphenylethylene and tetraphenylethylene moieties, Chem. – Asian J. 6, 1470–1478 (2011) Since the emission spectra of 2 and 3 were close in their crystalline and grinded forms, X-Ray diffraction should be carried out on the solid samples to verify whether another solid form was obtained after grinding - either another crystalline form or an amorphous form - which would be responsible for mechanoluminochromism. Additional investigation is also required to determine the minimum solute concentration allowing the observation of the AIEE phenomenon. Besides, further studies could be carried out with a focus on the solvatochromism of the two syn- thesized derivatives 2 and 3. 4.2 Characterisations Acknowledgements. We would like to warmly thank Cl´emence Allain, Claire Lambard, Rachel M´eallet, and Jonathan Piard for their valuable help and advice. The quantum chemistry calcula- tions were made possible thanks to the help of Federica Agostini. We would like to thank the Chemistry Research Department of the ENS Paris-Saclay as well as its technical staff, Cl´ement Dor´e, Sandrine and Pushpa Mourougayane for welcoming us and allowing us to use their equipment. The AIEE study of 2 was performed at C2,0 = 22.9 µmol.L−1. At this concentration, it was possi- ble to observe aggregation-induced enhanced emission for small fractions of DMF in water (1% to 25%). No aggregation-induced enhanced emission was observed at a lower concentration of C2,0/2, which means the concentration was not sufficient for the formation of aggregates (Appendix 15). The emission was quenched for a water fraction of 90%, which suggests that AIEE is concentration-dependent: a threshold concentration should be reached that 2 molecules could aggregate. References 1. M. Qayyum, T. Bushra, Z.A. Khan, H. Gul, S. Majeed, C. Yu, U. Farooq, A.J. Shaikh, S.A. Shahzad, Synthesis and tetraphenylethylene-based aggregation-induced emis- sion probe for rapid detection of nitroaromatic compounds in aqueous media, ACS Omega 6, 25447–25460 (2021) gg g A correlation between the emission of 2 and 3 and the size of the aggregates could not be established because the results were limited. It would be necessary to per- form stirring-controlled DLS measurements and to record emission spectra on a full range of mixtures with dif- ferent water fractions. Stirring is crucial to prevent any coalescence. The fluorescence emission study with triph- enylamine 1 did not provide any significant results in the solid state or in a DMF solution. For molecule 1, the first excited state is also predominantly due to the HOMO- LUMO transition but has a very low oscillator strength, which may explain the absence of emission. The AIE study of 1 was therefore not performed. 2. S. Majeed, T.A. Khan, M.T. Waseem, H.M. Junaid, A.M. Khan, S.A. Shahzad, A ratiometric fluorescent, colorimet- ric, and paper sensor for sequential detection of Cu2+ and Glutathione in food: AIEE and reversible piezo fluo- rochromic activity, J. Photochem. Photobiol. Chem. 431, 114062 (2022) 3. L. Wilbraham, M. Louis, D. Alberga, A. Brosseau, R. Guil- lot, F. Ito, F. Labat, R. M´etivier, C. Allain, I. Ciofini, Revealing the origins of mechanically induced fluorescence changes in organic molecular crystals, Adv. Mater. 30, 1800817 (2018) 4. S.A. Jenekhe, J.A. Osaheni, Excimers and exciplexes of conjugated polymers, Science 265, 765–768 (1994) L. Estournet et al..: Emergent Scientist 8, 1 (2024) 7 4.1 Syntheses For one reason or another, a sample of molecule 3, which was not studied, was not as pure as the other. To facil- itate crystallization, the reaction medium was separated: one part was concentrated by evaporation of the solvent under reduced pressure and the other part was dissolved in petroleum ether to help recrystallization. The solid obtained after trituration in petroleum ether did not con- tain traces of the derivative with aldehyde while the one obtained after the evaporation of the solvent did. A slight shift in wavelength was also noticed between the emission spectra of the crystalline form before and after grinding for each derivative (a 4 nm/203 cm−1 and 5 nm/145 cm−1 shift for 2 and 3 compounds respectively). Nonetheless, the two compounds had different behaviors when the crystalline form was grinded: the emission spec- trum of 2gr was blue shifted (from 446 nm/22 422 cm−1 to 442 nm/22 624 cm−1), whereas that of 3gr was red shifted (from 584 nm/17 123 cm−1 to 589 nm/16 978 cm−1). The NMR spectrum of the obtained molecule 3 did not present the expected peaks. It may be possible to record the NMR spectrum in DMSO-D6 [10] or with a higher Larmor frequency, such as 300 or 400 MHz. However, the IR spectrum and the aspect of the product obtained were quite encouraging. Besides, a slight shift can be observed compared to the reference values. The emission spectrum of 2gr displaying a blue shift was in contradiction with a pre- vious study [10] in which grinded 2gr led to a red shift because an amorphous phase was obtained (Tab. 2). The L. Estournet et al..: Emergent Scientist 8, 1 (2024) Cite this article as: Lise Estournet, Mathieu Loriot, In`es Mezghani, Audrey Pollien, Nina Villette. Fluorescence emission modification of triphenylamine derivatives by aggregate formation in solution or by mechanical stress in solid state, Emergent Scientist 8, 1 (2024) 5 Conclusion Finally, another possibility would be to synthesize a triphenylamine derivative with a donor group to observe its potential effects on the AIE mechanoluminochromic properties. 10. Y. Cao, W. Xi, L. Wang, H. Wang, L. Kong, H. Zhou, J. Wu, Y. Tian, Reversible piezofluorochromic nature and mech- anism of aggregation-induced emission-active compounds based on simple modification, RSC Adv. 4, 24649–24652 (2014) 11. Z.-Q. Yao, J. Xu, B. Zou, Z. Hu, K. Wang, Y.-J. Yuan, Y.-P. Chen, R. Feng, J.-B. Xiong, J. Hao, X.-H. Bu, A dual- stimuli-responsive coordination network featuring reversible wide-range luminescence-tuning behavior , Angew. Chem. Int. Ed. 58, 5614–5618 (2019) L. Estournet et al.: Emergent Scientist 8, 1 (2024) 8 16. NIST. Chemical Substance Database, 2023. Retrieved July 8, 2023, from https://webbook.nist.gov/chemistry/fluid/ 12. J. Sturalan, M.K. Etherington, A.N. Bismillah, H.F. Hig- ginbotham, W. Trewby, J.A. Aguilar, E.H.C. Bromley, A.-J. Avestro, A.P. Monkman, P.R. McGonigal, Excited-state aro- matic interactions in the aggregation-induced emission of molecular rotors, J. Am. Chem. Soc. 139, 17882–17889 (2017) 17. E. Ramachandran, R. Dhamodharan, Rational design of phenothiazine (PTz) and ethylenedioxythiophene (EDOT) based donor–acceptor compounds with a molecular aggregation breaker for solid state emission in red and NIR regions, J. Mater. Chem. C 3, 8642—8648 (2015) ( ) 13. P. Das, A. Kumar, A. Chowdhury, P.S. Mukherjee, Aggregation-induced emission and white luminescence from a combination of π- conjugated donor–acceptor organic Luminogens, ACS Omega 3, 13757–13771 (2018) ( ) 18. S.S.M. Fernandes, M. Belsley, A.I. Pereira, D. Ivanou, A. Mendes, L.L.G. Justino, H.D. Burrows, M.M.M. Raposo, Push–pull N,N-diphenylhydrazones bearing bithiophene or thienothiophene spacers as nonlinear optical second har- monic generators and as photosensitizers for nanocrystalline TiO2 dye-sensitized solar cells, ACS Omega 3, 12893—12904 (2018) 14. D.W. Cho, D.W. Cho, Excimer and exciple emissions of 1,8- naphthalimides caused by aggregation in extremely polar or nonpolar solvents, New J. Chem. 38, 2233–36 (2014) 15. J.G. Baragi, M.I. Aralaguppi, T.M. Aminabhavi, M.Y. Kariduraganavar, A.S. Kittur, Density, viscosity, refrac- tive index, and speed of sound for binary mixtures of anisole with 2-chloroethanol, 1,4-dioxane, tetrachloroethy- lene, tetrachloroethane, DMF, DMSO, and diethyl oxalate at (298.15, 303.15, and 308.15) K , J. Chem. Eng. Data 50, 910–916 (2005) 19. Conor Parks et al., Solubility curves and nucleation rates from molecular dynamics for polymorph prediction – mov- ing beyond lattice energy minimization, Phys. Chem. Chem. Phys. 19, 5285–5295 (2017) L. Estournet et al..: Emergent Scientist 8, 1 (2024) 9 Appendices Appendix 1: Emission (Em) and excitation (Exc) spectra of 2 at C2,0 = 22.9 µmol.L−1 for different water fractions (0% – 99%) in DMF. Excitation and emission wavelengths (in nm) are specified in brackets. Appendix 1: Emission (Em) and excitation (Exc) spectra of 2 at C2,0 = 22.9 µmol.L−1 for different water fractions (0% – 99%) in DMF. Excitation and emission wavelengths (in nm) are specified in brackets. Appendix 2: Emission (Em) and excitation (Exc) spectra of 3 at C3,0=19.0 µmol.L−1 for water fractions in DMF (0% – 95%). Excitation and emission wavelengths (in nm) are specified in brackets. Appendix 2: Emission (Em) and excitation (Exc) spectra of 3 at C3,0=19.0 µmol.L−1 for water fractions in DMF (0% – 95%). Excitation and emission wavelengths (in nm) are specified in brackets. L. Estournet et al.: Emergent Scientist 8, 1 (2024) 10 L.Estournet et al. Names CAS Hazard classes and category codes Safety pictograms Supplier Dichloromethane 75-09-2 Carc. 2 Eye Irr. 2 Carlo Erba, Val de Reuil, France Dimethylformamide 68-12-2 Repr. 1B Acute Tox. 4 Eye Irr. 2 Carlo Erba, Val de Reuil, France Ethanol 64-17-5 Flam. Liq. 2 Eye Irr. 2 Carlo Erba, Val de Reuil, France Malonitrile 109-77-3 Acute Tox. 3 Aquatic Acute 1 Aquatic Chronic 1 Merck, Munich Phosphorus oxychloride 10025-87- 3 Skin Corr. 1A Acute Tox. 4 Acute Tox. 2 Sigma Aldrich, Darmstadt, Germany Triphenylamine 603-34-9 Skin Sens. 1 Eye Irr. 1 Sigma aldrich, Saint- Louis, USA 4- (Diphenylamino)be nzaldehyde 4181-05-9 Eye Irr. 2 Skin Irr. 2 STOT SE 3 Tokyo Chemical Industry, Zwijndrecht, Belgium Appendix 3: Hazard classes, category codes and suppliers of the used chemicals. Appendix 3: Hazard classes, category codes and suppliers of the used chemicals. L. Estournet et al..: Emergent Scientist 8, 1 (2024) 11 Appendix 4: NMR spectrum of the commercial 4-diphenylamino-benzaldehyde 2 in CDCl3. Appendix 4: NMR spectrum of the commercial 4-diphenylamino-benzaldehyde 2 in CDCl3. Appendix 4: NMR spectrum of the commercial 4-diphenylamino-benzaldehyde 2 in CDCl3. L. Estournet et al.: Emergent Scientist 8, 1 (2024) 12 12 L. Estournet et al.: Emergent Scientist 8, 1 (2024) Appendix 5: NMR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2 in CDCl3. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 1 650 1 150 650 Wavenumber (cm-1) Appendix 6: IR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2. Appendices Estournet et al.: Emergent Scientist 8, 1 (2024) Solvent=Water : CAM-B3LYP/6-31G(d,p) 1 2 3 Δ(HOMO-LUMO) = 7.241 eV Δ(HOMO-LUMO) = 6.297 eV Δ(HOMO-LUMO) = 5.374 eV LUMO (-0.348 eV) HOMO (-6.645 eV) HOMO (-6.725 eV) LUMO (-1.351 eV) LUMO (0.847 eV) HOMO (-6.394 eV) Solvent=DMF : CAM-B3LYP/6-31G(d,p) LUMO (0.851 eV) LUMO (-0.344 eV) LUMO (-1.351eV) Gas phase : CAM-B3LYP/6-31G(d,p) 1 2 3 Δ(HOMO-LUMO) = 7.233 eV Δ(HOMO-LUMO) = 6.400 eV Δ(HOMO-LUMO) = 5.490 eV HOMO (-6.241 eV) LUMO (0.992 eV) LUMO (-0.189 eV) HOMO (-6.589 eV) HOMO (-6.830 eV) LUMO (-1.340 eV) Gas phase : CAM-B3LYP/6-31G(d,p) 1 2 3 Δ(HOMO-LUMO) = 7.233 eV Δ(HOMO-LUMO) = 6.400 eV Δ(HOMO-LUMO) = 5.490 HOMO (-6.241 eV) LUMO (0.992 eV) LUMO (-0.189 eV) HOMO (-6.589 eV) HOMO (-6.830 eV) LUMO (-1.340 eV) Gas phase : CAM-B3LYP/6-31G(d,p) 1 Δ(HOMO-LUMO) = 7.233 eV HOMO (-6.241 eV) LUMO (0.992 eV) Gas phase : CAM-B3LYP/6-31G(d,p) LUMO (-0.189 eV) LUMO (-1.340 eV) se : CAM-B3LYP/6-31G(d,p) LUMO (0.992 eV) LUMO (-0.189 eV) LUMO (-1.340 eV) 2 Δ(HOMO-LUMO) = 6.400 eV HOMO (-6.589 eV) 3 Δ(HOMO-LUMO) = 5.490 eV HOMO (-6.830 eV) 1 Δ(HOMO-LUMO) = 7.233 eV HOMO (-6.241 eV) Solvent=Water : CAM-B3LYP/6-31G(d,p) 1 2 3 Δ(HOMO-LUMO) = 7.241 eV Δ(HOMO-LUMO) = 6.297 eV Δ(HOMO-LUMO) = 5.374 eV LUMO (-0.348 eV) HOMO (-6.645 eV) HOMO (-6.725 eV) LUMO (-1.351 eV) LUMO (0.847 eV) HOMO (-6.394 eV) Solvent=DMF : CAM-B3LYP/6-31G(d,p) 1 2 3 Δ(HOMO-LUMO) = 7.241 eV Δ(HOMO-LUMO) = 6.298 eV Δ(HOMO-LUMO) = 5.375 eV HOMO (-6.390 eV) LUMO (0.851 eV) LUMO (-0.344 eV) HOMO (-6.642 eV) HOMO (-6.726 eV) LUMO (-1.351eV) 1 2 3 Δ(HOMO-LUMO) = 7.233 eV Δ(HOMO-LUMO) = 6.400 eV Δ(HOMO-LUMO) = 5.490 eV HOMO (-6.241 eV) HOMO (-6.589 eV) HOMO (-6.830 eV) cular orbitals of triphenylamine 1 and its derivatives 2 and 3, calculated respectively in gas olvation with the TDDFT method, using the CAM-B3LYP functional and the basis set 6 geometry of the molecules in each environment at CAM-B3LYP/6-31G(d,p) level. Appendices Appendix 5: NMR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2 in CDCl3. Appendix 5: NMR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2 in CDCl3 Appendix 5: NMR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2 in CDCl3. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 1 650 1 150 650 Wavenumber (cm-1) Appendix 6: IR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 1 650 1 150 650 Wavenumber (cm-1) Appendix 6: IR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2. Appendix 6: IR spectrum of the synthetised 4-diphenylamino-benzaldehyde 2. L. Estournet et al..: Emergent Scientist 8, 1 (2024) Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 1 650 1 150 650 Wavenumber (cm-1) Appendix 7: IR spectrum of the commercial 4-diphenylamino-benzaldehyde 2. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 Wavenumber (cm-1) 1 650 1 150 650 Appendix 8: IR spectrum of the synthetised 2-(4-(diphenylamino)benzylidene)malononitrile 3. L. Estournet et al..: Emergent Scientist 8, 1 (2024) 13 Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 1 650 1 150 650 Wavenumber (cm-1) Appendix 7: IR spectrum of the commercial 4-diphenylamino-benzaldehyde 2. Appendix 7: IR spectrum of the commercial 4-diphenylamino-benzaldehyde 2. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 Wavenumber (cm-1) 1 650 1 150 650 Appendix 8: IR spectrum of the synthetised 2-(4-(diphenylamino)benzylidene)malononitrile 3. Transmittance 100 90 80 70 60 50 40 30 20 10 0 3 650 3 150 2 650 2 150 Wavenumber (cm-1) 1 650 1 150 650 Appendix 8: IR spectrum of the synthetised 2-(4-(diphenylamino)benzylidene)malononitrile 3. Appendix 8: IR spectrum of the synthetised 2-(4-(diphenylamino)benzylidene)malononitrile 3. 14 L. Appendices HOMO (-6.830 eV) HOMO (-6.589 eV) 1 Δ(HOMO-LUMO) = 7.233 eV HOMO (-6.241 eV) 2 Δ(HOMO-LUMO) = 6.400 eV 3 Δ(HOMO-LUMO) = 5.490 eV Solvent=Water : CAM-B3LYP/6-31G(d,p) 1 Δ(HOMO-LUMO) = 7.241 eV LUMO (0.847 eV) HOMO (-6.394 eV) Solvent=Water : CAM-B3LYP/6-31G(d,p) LUMO (-1.351 eV) LUMO (-0.348 eV) LUMO (0.847 eV) LUMO (0.847 eV) LUMO (-0.348 eV) LUMO (-1.351 eV) 3 Δ(HOMO-LUMO) = 5.374 eV HOMO (-6.725 eV) 3 Δ(HOMO-LUMO) = 5.374 eV HOMO (-6.725 eV) 1 Δ(HOMO-LUMO) = 7.241 eV HOMO (-6.394 eV) 2 Δ(HOMO-LUMO) = 6.297 eV HOMO (-6.645 eV) 2 Δ(HOMO-LUMO) = 6.297 eV HOMO (-6.645 eV) 1 Δ(HOMO-LUMO) = 7.241 eV HOMO (-6.394 eV) 3 Δ(HOMO-LUMO) = 5.374 eV HOMO (-6.725 eV) 2 Solvent=DMF : CAM-B3LYP/6-31G(d,p) Solvent=DMF : CAM-B3LYP/6-31G(d,p) LUMO (-1.351eV) LUMO (-0.344 eV) nt=DMF : CAM-B3LYP/6-31G(d,p) 1 Δ(HOMO-LUMO) = 7.241 eV HOMO (-6.390 eV) LUMO (0.851 eV) LUMO (0.851 eV) LUMO (0.851 eV) LUMO (-0.344 eV) LUMO (-1.351eV) 3 Δ(HOMO-LUMO) = 5.375 eV HOMO (-6.726 eV) 2 Δ(HOMO-LUMO) = 6.298 eV HOMO (-6.642 eV) 1 Δ(HOMO-LUMO) = 7.241 eV HOMO (-6.390 eV) 1 Δ(HOMO-LUMO) = 7.241 eV HOMO (-6.390 eV) 2 Δ(HOMO-LUMO) = 6.298 eV HOMO (-6.642 eV) 3 Δ(HOMO-LUMO) = 5.375 eV HOMO (-6.726 eV) Appendix 9: Molecular orbitals of triphenylamine 1 and its derivatives 2 and 3, calculated respectively in gas phase, with water and DMF implicit solvation with the TDDFT method, using the CAM-B3LYP functional and the basis set 6-31G(d,p) after an optimization of the geometry of the molecules in each environment at CAM-B3LYP/6-31G(d,p) level. Appendix 9: Molecular orbitals of triphenylamine 1 and its derivatives 2 and 3, calculated respectively in gas phase, with water and DMF implicit solvation with the TDDFT method, using the CAM-B3LYP functional and the basis set 6-31G(d,p) after an optimization of the geometry of the molecules in each environment at CAM-B3LYP/6-31G(d,p) level. L. Appendices Estournet et al..: Emergent Scientist 8, 1 (2024) 15 Gas phase (CAM-B3LYP/6,31-G(d,p)) 1 2 3 Ground State μ = 0.0010 D μ = 4.6750 D μ = 9.5678 D Excited State 1 4.5078 eV f = 0.0216 3.9653 eV f = 0.0011 3.4639 eV f = 0.9979 Excited State 2 4.6118 eV f = 0.2851 4.0996 eV f = 0.5698 4.5046 eV f = 0.0166 Solvent = Water (CAM-B3LYP/6,31-G(d,p)) Ground State μ = 0.0049 D μ = 6.2868 D μ = 12.6292 D Excited State 1 4.5086 eV f = 0.0275 3.9720 eV f = 0.7249 3.2634 eV f = 1.1760 Excited State 2 4.5697 eV f = 0.3522 4.0684 eV f = 0.0026 4.5177 eV f = 0.0120 Solvent = DMF (CAM-B3LYP/6,31-G(d,p)) Ground State μ = 0.0040 D μ = 6.2503 D μ = 12.5654 D Excited State 1 4.5065 eV f = 0.0292 3.9568 eV f = 0.7448 3.2433 eV f = 1.1973 Excited State 2 4.5595 eV f = 0.3668 4.0654 eV f = 0.0021 4.4950 eV f = 0.0129 Mostly due to HOMO-LUMO transition endix 10: Energy values, oscillator strength f, and electric dipole moment µ of the ground state obtained from TDD etical calculations for molecules 1, 2, and 3 in gas phase or considering the solvent implicitly. Mostly due to HOMO-LUMO transition Appendix 10: Energy values, oscillator strength f, and electric dipole moment µ of the ground state obtained from TDDFT theoretical calculations for molecules 1, 2, and 3 in gas phase or considering the solvent implicitly. Appendix 10: Energy values, oscillator strength f, and electric dipole moment µ of the ground state obtained from TDDFT theoretical calculations for molecules 1, 2, and 3 in gas phase or considering the solvent implicitly. L. Estournet et al.: Emergent Scientist 8, 1 (2024) 16 g , ( ) ppendix 11: Molecular orbitals involved in the transitions to the first and second excited states in a mono-excitation appro tion. ppendix 11: Molecular orbitals involved in the transitions to the first and second excited states in a mono-excitation approx ation. Appendix 11: Molecular orbitals involved in the transitions to the first and second excited states in a mono-excitation approxi- mation. L. Estournet et al..: Emergent Scientist 8, 1 (2024) 17 Appendix 12: Crystalline phase emission (Em) and excitation (Exc) spectra of 1. Appendices Solid phase emission spectrum of 1 after melting on a Kofler bench and rapid recrystallization (Em BK). Appendix 12: Crystalline phase emission (Em) and excitation (Exc) spectra of 1. Solid phase emission spectrum of 1 after melting on a Kofler bench and rapid recrystallization (Em BK). Appendix 13: Crystalline and amorphous (KB) phase emission and excitation spectra of 2. Com: commercial 2; BK: Kofler b melted synthesized 2 which gave the amorphous phase after rapid solidification; grinded: grinded synthesized 2. Excitation i i l th ifi d i b k t Appendix 13: Crystalline and amorphous (KB) phase emission and excitation spectra of 2. Com: commercial 2; BK: Kofler bench melted synthesized 2 which gave the amorphous phase after rapid solidification; grinded: grinded synthesized 2. Excitation and emission wavelengths are specified in brackets. Appendix 13: Crystalline and amorphous (KB) phase emission and excitation spectra of 2. Com: commercial 2; BK: Kofler bench melted synthesized 2 which gave the amorphous phase after rapid solidification; grinded: grinded synthesized 2. Excitation and emission wavelengths are specified in brackets. L. Estournet et al.: Emergent Scientist 8, 1 (2024) 18 Appendix 14: Crystalline and amorphous (KB) phase emission and excitation spectrum of 3. KB: Kofler bench melted 3 which gave the amorphous phase after rapid solidification; grinded: grinded 3. Excitation and emission wavelengths are specified in brackets. Appendix 14: Crystalline and amorphous (KB) phase emission and excitation spectrum of 3. KB: Kofler bench melted 3 which gave the amorphous phase after rapid solidification; grinded: grinded 3. Excitation and emission wavelengths are specified in brackets. Appendix 15: Excitation (Exc) and emission (Em) spectra of 2 at C2,0/2 = 11.5 µmol.L−1 for different water fractions (1% – 90%) in DMF. Excitation and emission wavelengths (in nm) are specified in brackets. Appendix 15: Excitation (Exc) and emission (Em) spectra of 2 at C2,0/2 = 11.5 µmol.L−1 for different water fractions (1 90%) in DMF. Excitation and emission wavelengths (in nm) are specified in brackets. Appendix 15: Excitation (Exc) and emission (Em) spectra of 2 at C2,0/2 = 11.5 µmol.L−1 for different water fractions (1% – 90%) in DMF. Excitation and emission wavelengths (in nm) are specified in brackets.
W4389303809.txt	https://sciendo.com/pdf/10.2478/amns.2023.2.01234	en		A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning	Applied mathematics and nonlinear sciences	2,023	cc-by	6,534	Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 Applied Mathematics and Nonlinear Sciences https://www.sciendo.com A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning Han Ding1,† 1. Jilin Provincial Institute of Education College of Ethnic Education, Changchun, Jilin, 130022, China. Submission Info Communicated by Z. Sabir Received January 14, 2023 Accepted June 5, 2023 Available online November 25, 2023 Abstract Based on the flipped classroom in the context of deep learning, this paper combines the teaching design of university mathematics in colleges and universities with the learning characteristics of college students and designs and builds a flipped classroom teaching model of university mathematics in colleges and universities in the context of deep learning. To analyze the application effect of the university mathematics flipped classroom teaching model in colleges and universities, we first analyze the differences in the deep learning level of college students and propose the evaluation model of the flipped classroom teaching model, evaluate the application of the model and analyze the evaluation results, then analyze the independent learning ability of college students in a pre and post-test, and finally analyze the participation of college students. There are significant differences in the four dimensions of students’ deep learning awareness, learning motivation, learning engagement, and learning strategies, i.e., p<0.05. The final scores of college mathematics in classes A and B are 40.11 and 36.98. It shows that the effect of flipped classroom teaching in college mathematics is better in class A than in class B. The total value of the independent learning ability of the students in this class before and after the practice increased by 6.8468 points. In the flipped classroom, 37.2 % of the students thought that they completed the task poorly, which was related to the content of practicing the flipped classroom. University math teachers should guide students at the right time to improve their ability to divergence and analogical reasoning. Keywords: Flipped classroom; Deep learning; Pedagogical model; University mathematics; Independent learning ability. AMS 2010 codes: 97C70 †Corresponding author. Email address: weixdkz@163.com ISSN 2444-8656 https://doi.org/10.2478/amns.2023.2.01234 © 2023 Han Ding, published by Sciendo. This work is licensed under the Creative Commons Attribution alone 4.0 License. Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 2 1 Introduction The flipped classroom course is designed to record a short video by the teacher for certain knowledge points or topics. The length of the video is usually very short. After the recording is successful, the teacher will share the small video on the Internet. After class, students can either pre-study or check the shortcomings and fill in the gaps. The classroom becomes a place for teachers and students to ask questions and exchange ideas with each other so as to improve students’ learning efficiency and achieve better teaching results [1-3]. In the flipped classroom teaching model, the roles of teachers and students have changed greatly; students occupy an absolutely dominant position in learning, and teachers only need to guide them appropriately [4-5]. The flipped classroom is a new type of teaching mode, which has a strong scientific and advanced nature [6-7]. For this reason, university mathematics teaching should develop and utilize microclasses as a new type of teaching resource. In order to make up for the various shortcomings of the traditional university mathematics classroom teaching mode, provide students with richer learning methods, stimulate students’ learning interest, give full play to students’ subjective initiative in mathematics learning, and then enhance students’ mathematical learning ability, and provide assistance for students’ future development [8-9]. The education and teaching system’s deepening changes have led to an increase in people’s requirements for university mathematics teaching work [10]. The comprehensive promotion and application of microteaching mode under the flipped classroom model help to cultivate students’ summarizing and analyzing ability, logical thinking ability, spatial imagination, and arithmetic ability so as to improve students’ core professional qualities, enhance the level of classroom education and teaching, and satisfy the talent cultivation needs of the new era [11]. The convenience brought by the network in the new era has led to a new trend in college mathematics classroom teaching with the flipped classroom teaching mode [12]. As an emerging classroom teaching method, it has advantages but also inevitably has shortcomings. University teachers should actively learn new technologies and make full use of the convenience and efficiency brought by flipped classroom teaching so that the teaching of university mathematics can move forward in a more scientific and modern direction [13]. Spotts, J. D. et al. focused their research on examining two pre-calculus math classrooms randomly assigned to either a flipped classroom instructional model or one using a traditional classroom model. The results showed that students exposed to the flipped classroom model outperformed traditional classroom students in the control room [14]. Karimah, W. argued that teachers engage students in learning by asking them to conduct interviews and observation activities in the field. The teachers believe that these activities will help students become familiar with solving math problems. However, when students are given different problems, they are unable to solve them. The flipped classroom is an alternative learning model that can overcome these problems [15]. Leo, C. The study aimed to establish the link between classroom pedagogy and student achievement. The participant researcher aimed to employ classroom pedagogical models, such as videos, to guide the student participants [16]. Albalawi, A. S. aimed to explore the role of flipped classrooms in the Saudi Arabian pre-college students of Tabuk University Teaching Mathematics 2 course in effectiveness. The results revealed a significant difference in student performance in favor of the treatment group using flipped classroom instruction, but it was not statistically significant [17]. Vang et al. examined the effect of flipped classrooms on mathematics achievement and self-efficacy. The study compared four independent methods, i.e., three unit exams. In addition, a survey was conducted to collect selfdata. The results of the study showed no statistically significant differences between students in the treatment and control groups on any outcomes [18]. Lesage A. et al. concluded that the flipped classroom model is a promising approach to addressing at-risk higher education math students A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 3 because it allows for increased interaction and support within the classroom after students have watched instructional videos at home. However, there has been some limited research on classroom pedagogy in improvised classrooms [19]. This paper combines the flipped classroom teaching mode under the perspective of deep learning with college math teaching and constructs a college math flipped classroom teaching mode based on a deep learning background. The specific teaching content of college mathematics and the learning characteristics of college students mainly determines the design of the teaching mode. The analysis of the application effect of the university mathematics flipped classroom teaching model mainly includes the analysis of the differences in the level of deep learning of college students, the evaluation of the teaching effect of the university mathematics flipped classroom, the analysis of the independent learning ability of college students, and the analysis of the participation of college students in the flipped classroom. Finally, recommendations are made based on the analysis results and the uniqueness of university mathematics teaching. 2 2.1 Construction of a flipped classroom teaching model for university mathematics in colleges and universities under the perspective of deep learning Instructional Model of Flipped Classroom On the basis of extensive reading of the literature, combined with the teaching objectives and teaching concepts of college mathematics, based on the characteristics of the students in the sample experimental school. Figure 1 shows the teaching model of flipped classrooms constructed in this study. Before the class, teachers should fully prepare relevant teaching resources, which are divided into two kinds: basic resources and extended resources. The classroom learning process is summarized as cooperative inquiry, personalized instruction, consolidation practice, summary pointers, and feedback evaluation. Students should reflect and summarize carefully according to the requirements of the postclass learning list, fill in the list form and submit it on time, and develop the good learning habit of being responsible for their learning. In summary, the flipped classroom is to utilize the existing information technology means to build an information-based teaching environment and to re-plan the pre-course, in-course, and post-course. By reversing the three teaching processes of knowledge transfer, knowledge internalization and knowledge consolidation, it realizes the flip of the roles of teachers and students in traditional teaching and achieves the innovation of the traditional classroom teaching mode teaching design and method. Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 4 Before Class Teacher In Class Student Prepare Lessons Personalized Guidance Cooperate Obtain Knowledge Feedback Evaluation Prepare Resources Assign Task Sum Up Sum Up Consolidation Exercise After Class Study List Reflect On The Summary Figure 1. Teaching model of the flipped classroom 2.2 Design of Flipped Classroom Teaching Mode for College Mathematics under Deep Learning Perspective Through the concept of instructional design and deep learning theory-related literature, combined with domestic and international scholars for a deep learning perspective of the flipped classroom teaching practice research. Based on the specific teaching content and the characteristics of college students, this paper analyzes the four aspects of the flipped classroom teaching design for deep learning in combination with specific teaching cases and constructs a flipped classroom teaching model for college mathematics under the perspective of deep learning. Figure 2 displays the flipped classroom teaching design for college university mathematics under the perspective of deep learning. Before school During class After school Results Showcase Video Exercise Introduction Teacher Student Feedback evaluation Reflection Consolidation Evaluation Panel Discussion TeacherStudent Exchange Knowledge transfer Knowledge priming Personalized instruction Defining the problem Deep Learning Builds Teaching objectives Independent Thinking Internalization of knowledge Consolidation Reflection Migration and Applications Evaluation and reflection Teaching content Teaching activities Teaching and Learning evaluation Deep Learning Theory Figure 2. Flipped classroom teaching design from the perspective of deep learning A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 3 3.1 5 Effectiveness of the application of the flipped classroom teaching model of university mathematics in higher education institutions Analysis of variability in the level of deep learning The Deep Learning Level Evaluation Scale is mainly used to judge the effectiveness of the model based on the comparison of the variability of the deep learning level of the respondents before and after the implementation of the flipped classroom teaching model. Due to the limited space of this paper, it will not be summarized here. Before the beginning and after the end, this study investigated the relationship between the differential comparison of the deep learning levels of 20 students before and after the implementation of the flipped classroom teaching model in terms of the four dimensions of deep learning awareness, motivation to learn, commitment to learning, and learning strategies. The validity of the model was verified after determining if there was a significant difference in each dimension in the students’ deep learning level. The results were statistically analyzed using SPSS software to conclude that the data on the dimensions of the deep learning level of the students in this class were approximately normally distributed, and a paired-sample t-test was conducted. Table 1 displays the pre-and post-test variance analysis of the deep learning level. The paired samples t-test on the deep learning level of 20 students before and after the implementation of the flipped classroom teaching mode revealed that in terms of the deep learning state, there were significant differences in the four dimensions of students’ deep learning awareness, motivation to learn, commitment to learning, and learning strategies, i.e., p<0.05. This indicates that the flipped classroom teaching mode has a significant enhancement of the students’ deep learning state. Further comparing the effect sizes using the value of Cohen ' s d , it is found that the students’ improvement in deep learning motivation and learning commitment is large, i.e., d>0.8. The improvement in deep learning awareness is average, i.e., 0.5<d<0.8, and the improvement in deep learning strategies is insignificant, i.e., 0.2<d<0.5, which is a small effect size, which indicates that the students are able to clearly recognize that the linkage between teaching activities before and after class is very close, and each linkage requires certain knowledge transfer strategies and evaluation and reflection strategies. The linkage between various teaching activities during and after the lesson is very close, and each link has a clear purpose and task requirements, which requires certain knowledge transfer strategies, evaluation and reflection strategies, and so on. However, because the deep learning strategies are more implicit, not easy to be revealed, and need to be learned and cultivated for a long time, the deep learning strategies reached a level of significant difference after the experiment but still need to be strengthened. To summarize the above analysis, the deep learning status of all 20 students under the flipped classroom teaching mode has been effectively improved, reaching extremely significant differences in deep learning awareness, learning motivation and learning engagement, and significant differences in deep learning strategies, which need to be continued to be strengthened. It can be seen that this flipped teaching practice is beneficial. Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 6 Table 1. Difference analysis before and after deep learning level Dimension N before measurement M  SD After test M  SD T P Cohen ' s d value Deep learning awareness 20 3.86  0.63 4.35  0.61 -3.052 0.006 0.523 Deep learning motivation 20 3.52  0.72 4.56  0.68 -4.355 0.001 0.633 Deep learning investment 20 3.54  0.68 4.76  0.61 -4.450 0.000 1.036 3.02  0.75 3.53  0.53 -2.242 0.028 0.225 Deep learning strategy 3.2 3.2.1 20 Evaluation model of flipped classroom teaching mode under the perspective of deep learning Evaluation methodology model According to the research, it is found that it is not sufficient to evaluate teaching effectiveness only by relying on the surface information of students’ college math performance. This is because this paper proposes a three-dimensional indicator measure of students’ college math performance, including usual grades, grade fluctuation, and progress. Average grade ( Avg ) is the average score of students. Achievement fluctuation ( Std ) is the standard deviation of the student’s achievement. Progress ( Asc) , which is an indicator that responds to the change of students’ performance in their studies [20]. A student i ‘s grade can be represented as Si =  Avgi , Stdi , Asci  , and a student i ‘s grade on all n quizzes as Sti = gi1 , gi 2 ,, gin  . Assuming that the number of students in the course taught by the instructor is m , all the students’ grades can be constructed as a matrix of m rows and n columns, with each row representing a student’s grade, and each column representing the specific composition of the student’s grades.  g11 g G =  21    g m1 g12 g 22 g m1 g m1 g1n  g 2 n    g mn  (1) Assuming a weight of  j on the j st quiz, the average grade for the course is:  m     i  gij  j =1 i =1  Avg =  m n n (2) The essence of fluctuating grades is the standard deviation, which describes fluctuations in the data. m Std =  i =1  1 n  1 n    gij −  gij  n  j =1  n j =1   2     (3) A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 7 Finally, the student’s degree of progress is calculated. Since each student has a different foundation, even for students with the same final grade, the magnitude of their grade improvement is not the same. The Hidden Markov Transfer Matrix happens to be suitable for the current scenario. Therefore, the Markov Transfer Matrix will be used next to calculate the degree of progress of the students. A brief explanation of the mathematical interpretation of the Markov transfer matrix is necessary: For a random variable n , there is a random sequence E , n  N ( N +) * n which is its state space, denoted E , and E is a listable set. m, n  N * , and if i, j, ik  E (1, 2,3,, n − 1) , has P (m+n = j1n = i, n−1 = in−1 =,, = i = i1 ) = P (m+n = j \| n = i ) , then E , n  N ( N +) * n is said to be a Markov chain. The specific ideas are as follows: 1) Divide the grades into K rating, the The state vector can be expressed as: G j = x1 ( j ), x2 ( j ),, xK ( j ) (4) Where j denotes the j nd grade and xt ( j ) denotes the proportion of people whose j th grade is K in the t th class, obviously,  x ( j ) = 1 . Assuming that the number of people who scored t =1 t grades in the first grade is u1 , u2 ,, uK and K u t =1 t K = S , respectively, where S is the total number of students who participated in the assessment, the state vector is: u  u u R (1) =  1 , 2 , , K  S  S S (5) 2) Assuming that u1 corresponds to a student whose achievement on the next test is recorded as u11 , u12 ,, u1K , respectively, there are: K u d =1 1d = u1 (6) Then the transfer probability of u1 corresponding to a rated student is denoted as: u u u  P(1) =  11 , 12 ,., 1K  u1   u1 u1 (7) The transfer probability is calculated in the same way for the other rated students, and the final transfer matrix can be obtained as follows: 8 Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15  u11 u  1  u21  P =  u2    uS1  uS u12 u1 u22 u2 uS 2 uS u1S  u1   u2 S  u2  = ( pij ) S S   uSS  uS  (8) After obtaining the transfer matrix, further discussion on the change in student ratings is needed, we know that pij represents the current rating of i and the next rating of j . If i  j , represents that the student has progressed after a period of time, and i = j , represents that after a period of time, there is no change in the student’s rating. If i  j , the student has regressed, then \| i − j \| can represent how much the student has progressed and regressed. Progress should be rewarded and regression should be considered as a punishment, which can then be expressed as (i − j )3 . As a value, when a student progresses, a positive value is obtained, and when a student regresses, a negative value is obtained. At the same time 3 times can ensure that there is a large reward or punishment for progress or regression when the magnitude is large, such as i − j = 3 , indicating that the student has progressed three grades, in this case the coefficient is 33 , the reward is very large. If \| i − j \|= 1 , the student has progressed or regressed by one grade, the reward or penalty is small. If \| i − j \|= 0 , the student has not progressed or regressed, then his coefficient is 0. From this, we can obtain the progress matrix corresponding to P :  (i − j )3 u11  u1   (i − j )3 u 21  A= u2    (i − j )3 uS 1  uS  (i − j )3 u12 u1 (i − j )3 u22 u2 (i − j )3 uS 2 uS (i − j )3 u1S   u1  (i − j )3 u2 S   u2  = ( aij ) S S   (i − j )3 uSS   uS  (9) Then the degree of progress is calculated as follows: Asc = S  aij = i , j =1 S  (i − j )3 pij = i , j =1 S (i − j )3 uij i , j =1 ui  (10) S = { Avg , Std , Asc} indicates the effectiveness of teacher instruction in this area of student achievement. 3.2.2 Application of evaluation of teaching effectiveness of university mathematics in higher education and analysis of results In order to analyze the validity of the teaching effectiveness evaluation model proposed above, this paper conducts the following analytical study. A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 9 Take the results of three consecutive tests of 29 students and 26 students in two classes of university mathematics course as an example. The raw scores of the three tests of the students in the two classes are shown in Fig. 3, where (a) is the three test scores of the students in Class A and (b) is the three test scores of the students in Class B. Among the first test scores of Classes A and B, nine students scored in the range of 90 to 100 points, and two of them scored 100 points. There were some ups and downs in the three test scores of both classes, with the lowest score in the 20-30 range. (a) Results of the three tests for the students in Class A (b) Results of students in class B Figure 3. Original scores of the three tests of the students in the two classes Based on Figure 3, the number of students with each rating for each of the three scores in Class A can be determined. Table 2 displays the number of students with each rating for each of the three quiz scores in Class A. Table 2. Number of students who rated each test for the three tests in class A Student ID First grade Second grade Third grade A(90~100) 9 5 12 B(80~90) 9 6 5 C(70~80) 4 7 6 D(60~70) 1 6 2 E(0~60) 6 5 4 There are 9 students who rated their first grade as A. These 9 students rated 2 A’s, 1 B, 3 C’s, 2 D’s, and 1 E on the second test. Then there are vectors: 2 1 1 2 1 P12 (1) =  , , , ,  9 9 9 9 9 (11) 2 1 2 1 P12 ( 2 ) =  , , 0, ,  9 9 9 9 (12) Ditto: 10 Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 1 1 1 P12 ( 3) = ( 0, , , , 0 ) 4 2 4 (13) P12 ( 4) = ( 0,1,1,0,0 ) (14) 1 1 1 1 1 P12 ( 5 ) =  , , , ,  6 6 3 6 3 (15) Define the three test scores to be denoted as Test1 , Test2 , and Test3 , then the change matrix for Test1 → Test2 is: 2 1 1 2 1 9 , 9 , 9 , 9 , 9    2 , 1 , 0, 2 , 1  9 9 9 9    1 1 1   P12 = 0, , , , 0  4 2 4  0,1,1, 0, 0    1 1 1 1 1  6 , 6 , 3 , 6 , 3      (16) In the same way the change matrix for Test2 → Test3 can be calculated: 3 1 1   5 , 5 , 5 ,0,0     5 , 1 , 5 , 0, 0  7 7 7    2 1 1 2 P23 =  , , , 0,  7 7 7 7  1 1 1 1   , , , , 0 6 6 2 6  1 1 1   , 0, , , 0  4 4 4  (17) Then the change matrix for Test1 → Test3 is: 0.3728 0.1922 0.2466 0.0647 0.1232  0.4641 0.1295 0.2588 0.0925 0.0555    P13 = P12 • P23 = 0.3630 0.2203 0.2322 0.0418 0.1431    0 0.7145 0.1429 0.1428 0  0.3835 0.1803 0.2299 0.0696 0.1371  (18) A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 11 Then the change matrix corresponding to P13 is: , −0.1933, −1.9745, −1.7496, −7.8733  0 0.4655, 0 , −0.2586, −0.7409, −1.5023   A13 =  2.9045, 0.2206, 0 , − 0.0452, −1.1436    ,0 19.2863,1.1433, 0.1436, 0   24.5668, 4.8665,1.8369, 0.0698, 0  (19) Ultimately, Teaching Effect E ( A) is calculated as follows: s s i , j =1 i , j =1 s E ( A) =  aij =  ( i − j ) pij =  3 i , j =1 (i − j ) 3 ni nij = 40.11 The progress of class A is calculated to be 40.11, and the progress of class B is 36.98. To sum up, the final grades of college mathematics in classes A and B can be calculated as 40.11 and 36.98. The teaching effect of a college mathematics flipped classroom in class A is better than that in class B. The final grades of college mathematics flipped classroom in class A can be calculated as 40.11 and 36.98. At the same time, it also shows that the evaluation model established in this paper for college math flipped classroom teaching is effective. 3.3 Pre- and post-test analysis of self-directed learning skills In order to analyze the specific impact of the flipped classroom teaching mode on the independent learning ability of the students in the practice class, the level of the independent learning ability of the students in the practice class was analyzed as a whole by using the SPSS 20.0 paired samples ttest, and the pre and post-test differences were analyzed. That is, the five dimensions of the independent learning ability of sophomore freshmen were analyzed one by one to determine whether the flipped classroom teaching mode had an impact on the independent learning ability of students in the practice class. The pre-and post-test analysis of independent learning ability is shown in Table 3. In the pre-practice period, the mean value of the independent learning ability of the students in the class was 125.3867 points, and in the post-practice period, it was 132.2335 points. The total value of students’ independent learning ability in the class before and after the practice increased by 6.8468 points and had an extremely significant difference with the level total value of t=-3.807, P=0 and P<0.05. Therefore, the independent learning ability of the students in the class was significantly improved through this practice of flipped classroom teaching in college mathematics. In particular, the student’s independent learning ability in terms of motivation, self-monitoring, selfregulation and learning environment and the overall mean value increased significantly and significantly after the practice. Learning motivation t=-3.651, self-monitoring t=-3.045, selfregulation t=-2.536, and learning environment t=-2.823, which suggests that through the teaching practice of implementing the flipped classroom model in the university mathematics courses in colleges and universities, college students’ independent learning ability has been improved significantly, and the flipped classroom teaching model is able to cultivate the independent learning ability of college students, especially in the areas of improving the learning ability, self-monitoring, self-regulation, and learning environment. The flipped classroom teaching mode can cultivate college Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 12 students’ independent learning ability, especially in improving their motivation, self-monitoring, selfregulation and learning environment. Table 3. Difference data before and after autonomous learning ability Dimension content Academic motivation Learning strategy Self-monitoring Self-adjustment Academic environment Total horizontal value 3.4 Before and after the test Pre-test Number of people 55 27.0933 Standard error 3.19256 -3.651 post-test 55 28.9311 3.95425 - Pre-test 55 24.7856 4.62875 -1.815 0.075 post-test 55 25.8252 3.85300 - - Pre-test 55 25.4748 3.24331 -3.045 0.003 post-test 55 26.8685 3.73912 - - Pre-test 55 27.6845 3.99401 -2.536 0.0015 post-test 55 28.9233 3.93856 - - Pre-test 55 20.3527 2.80654 -2.823 0.006 post-test 55 20.7215 3.26313 - - Pre-test 55 125.3867 14.05996 -3.807 0 post-test 55 132.2335 14.86653 - - Mean T Sig. (bilateral) 0 Analysis of student engagement To further understand the impact of the flipped classroom teaching model in college mathematics on college students, engagement data statistics were conducted on 55 students in two classes. The statistics of college students’ engagement data are shown in Table 4. According to the table, 81.4% of the students seldom communicate with other students in the pre-course learning stage, which is related to their realistic learning environment. Students do not have time for pre-course video learning during school due to heavy school work, and most of them do video learning at home after school, which lacks the condition to communicate with other students, so most of them tend to study alone. 70.2% of the students would solve the problem by going back to the study material when they encountered points they did not quite understand during the independent study stage. More than 20% of the students would solve the problem by consulting the teacher, communicating with other students, or searching for information. 71% of the students can basically complete the independent study tasks assigned by the teacher before class, and nearly half of the students can complete 100% of the tasks, which shows that this teaching practice has successfully aroused the curiosity of the students and thus fully mobilized the students’ learning enthusiasm. 37.2% of students in the flipped classroom believe that they completed the task poorly, which is related to the flipped classroom’s content. The flipped classroom of college mathematics requires college students to have high thinking dispersion ability and analogical reasoning ability, and teachers’ timely guidance, so most of the students think that they do not complete the tasks in the classroom satisfactorily. A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 13 Table 4. Statistics of college students’ participation data Question In the pre-class preview stage, interact with other students When you do not understand the knowledge point in the preview stage before class, will you? Can you complete the study tasks assigned by the teacher during the preview stage before class? How did you complete your classroom tasks in a flipped classroom? 4 Option Proportion a great many 1.6% more 4.5% same as 12.5% less 52.1% a fat lot 29.3% Ask the teacher 6.7% Look back at the learning material 70.2% Communicate with classmates 10.6% Find data to solve 3.2% Ask the teacher later in class 9.3% accomplish 45.6% Complete 80% 25.4% Complete 50% 9.1% Complete 30% 8.9% None of them were done 11% Be able to complete quickly 13.8% Proficient to complete 18.6% Basically complete 30.4% Part of the completion 24.3% Incomplete 12.9% Conclusion This paper applies the flipped classroom in the context of deep learning to the university mathematics classroom, constructs a university mathematics flipped classroom teaching model, and carries out the analysis of differences in the level of deep learning, the analysis of teaching effect, the analysis of pre and post-tests of independent learning ability, and the analysis of students’ participation. The analysis of the difference in deep learning level of some students who participated in the university mathematics flipped classroom teaching model found that the deep learning status of 20 students had been effectively improved. Extremely significant differences were reached in deep learning awareness, motivation, and learning engagement. However, there are still significant differences in deep learning strategies that need to be strengthened. It can be discerned that this flipped teaching approach is successful. The independent learning ability of the students in the practice class was analyzed as a whole in terms of pre-and post-test variability. In the pre-practice period, the students in the class had a mean independent learning ability of 125.3867 points, while in the post-practice period, it was 132.2335 points. The total value of the independent learning ability of the students in the class before and after the practice increased by 6.8468 points, and there is an extremely significant difference, the level of the total value of t=-3.807, P=0 and P<0.05. This shows that through the teaching practice of this college mathematics course, the implementation of the flipped classroom model, the independent learning ability of university students has significantly improved, and the flipped classroom teaching Han Ding. Applied Mathematics and Nonlinear Sciences, 9(1) (2024) 1-15 14 model can cultivate the independent learning ability of university students, especially in the improvement of learning motivation, self-learning ability, self-learning ability, and self-learning ability. The flipped classroom teaching mode can cultivate college students’ independent learning ability, especially in improving their learning motivation, self-monitoring, self-regulation, and learning environment. The analysis of student’s participation in the university mathematics flipped classrooms shows that 37.2% of the students think that they are poor at completing the tasks in the flipped classroom, which is related to the content of practicing the flipped classroom. The flipped classroom of college mathematics requires college students to have a high level of divergent thinking and analogical reasoning ability and requires timely guidance from teachers, so most students think they complete tasks poorly in the classroom. 5 Discussion Through the implementation and analysis of this teaching case, it is concluded that the application of university mathematics flipped classroom teaching mode in college mathematics classrooms is feasible and can improve the teaching effect of the classroom, but there are also some problems and some suggestions for teachers and students. 1) Teachers need to change the traditional teaching ideology and reposition their roles, no longer as the transmitter of knowledge but as the guide in teaching activities. To meet the demands of the university mathematics flipped classroom, teachers must continuously improve their professional quality. In the process of teaching practice, it has been found that there are some problems with students’ independent learning, group discussion, and cooperative inquiry. Students are more casual in the process of surfing the Internet, chatting online, surfing the Internet at will, chatting about issues that have nothing to do with the discussion in the classroom, and not actively participating in the discussion. Therefore, these factors need to be taken into consideration when providing open Internet resources, eliminating interference, and strengthening the construction and management of groups. 2) Students are the main body, and they need to actively participate in the whole teaching activities of college mathematics. Students need to improve their self-discipline, selfregulation, and self-control. In order to study independently before class, students must have a plan, monitor themselves reasonably, and take the initiative to complete the task. In group cooperative inquiry, as a member of the group, students should actively speak, discuss, think, show themselves, and develop good learning habits. References [1] Pierce, R., Fox, J., & Bernard. (2017). Instructional design and assessment vodcasts and active-learning exercises in a flipped classroom model of a renal pharmacotherapy module. american journal of pharmaceutical education. [2] Rui, Z., Lian-Rui, X., Rong-Zheng, Y., Jing, Z., & Wan Xue-hong…. (2017). Friend or foe? flipped classroom for undergraduate electrocardiogram learning: a randomized controlled study. Bmc Medical Education, 17(1). [3] Lin, Y., Zhu, Y., Chen, C., Wang, W., & Liu, Y. (2017). Facing the challenges in ophthalmology clerkship teaching: is flipped classroom the answer?. Plos One, 12(4), e0174829. A Flipped Classroom Teaching Model of College Mathematics in Colleges and Universities Based on the Background of Deep Learning 15 [4] Xin, Cheng, Kenneth, Ka, Ho, & Lee, et al. (2017). The “flipped classroom” approach: stimulating positive learning attitudes and improving mastery of histology among medical students. Anatomical Sciences Education. [5] Kwan, C., Lie, Wai, C., Hew, & Foon, K. (2018). Applying “first principles of instruction” as a design theory of the flipped classroom: findings from a collective study of four secondary school subjects. Computers & education. [6] Lo, C., Lie, C., & Hew, K. (2018). Applying “first principles of instruction” as a design theory of the flipped classroom: findings from a collective study of four secondary school subjects. Computers & Education, 118, 150-165. [7] Francisco, H. L., Mingorance-Estrada ángel, Juan, T. T., Aznar-Díaz Inmaculada, & Cáceres Reche María. (2018). Incidence of the flipped classroom in the physical education students’ academic performance in university contexts. Sustainability, 10(5), 1334. [8] Dorie, A., Hurst, J., & Loranger, D. (2017). It All Adds Up: Flipped Classroom Approaches in Retail Math Instruction. [9] Ko, I., & Herbst, P. (2020). Subject matter knowledge of geometry needed in tasks of teaching: relationship to prior geometry teaching experience. Journal for Research in Mathematics Education, 51(5), 600-630. [10] Jansen, A., Berk, D., & Meikle, E. (2017). Investigating alignment between elementary mathematics teacher education and graduates’ teaching of mathematics for conceptual understanding. Harvard Educational Review, 87(2), 225-250. [11] Ficano, C. K. C. (2019). Identifying differential benefits from a flipped-group pedagogy in introductory microeconomics. International Review of Economics Education, 30. [12] Desmaniati, E., & Fitriza, R. (2019). Praktikalitas perangkat pembelajaran transformasi geometri berbasis flipped classroom pada peserta didik kelas xi sman 1 padang. Math Educa Journal(1). [13] Ricardo-Adán Salas-Rueda. (2020). Flipped classroom: pedagogical model necessary to improve the participation of the students during the learning process. LUMAT International Journal on Math Science and Technology Education(1). [14] Spotts, J. D., & Blume, A. P. G. D. (2020). A pilot study on the effect of the flipped classroom model on pre-calculus performance. SAGE Open, 10. [15] Karimah, W. (2019). Penerapan model flipped classroom berbantuan video pembelajaran terhadap kemampuan pemecahan masalah matematika siswa. Delta Jurnal Ilmiah Pendidikan Matematika. [16] Leo, C. (2017). Flipped Classroom Pedagogical Model and Middle-Level Mathematics Achievment: An Action Research Study. [17] Albalawi, A. S. (2018). The effect of using flipped classroom in teaching calculus on students’ achievements at university of tabuk. International Journal of Research in Education & Science, 4, 198207. [18] Vang, & Veng, Y. (2017). The impact of the flipped classroom on high school mathematics students’ academic performance and self-efficacy. [19] Lesage, A., Kay, R., & Tepylo, D. (2019). A Flipped Classroom Approach to Supporting At-Risk University Mathematics Students: Shifting the Focus to Pedagogy. iCERi 2019 - 12th International Conferences of Education, Research & Innovation. [20] Yang, Y. (2021). Quality evaluation method of a mathematics teaching model reform based on an improved genetic algorithm. Scientific Programming.
https://openalex.org/W4231713801	https://www.qeios.com/read/UB5703/pdf	English	null	Chemoprotection	Definitions	2,020	cc-by	49	Qeios · Definition, February 2, 2020 Open Peer Review on Qeios Chemoprotection National Cancer Institute Qeios ID: UB5703 · https://doi.org/10.32388/UB5703 Source National Cancer Institute. Chemoprotection. NCI Thesaurus. Code C15533. The use of drugs to protect normal tissue from damage by cancer therapy agents. Qeios ID: UB5703 · https://doi.org/10.32388/UB5703 1/1
https://openalex.org/W3036463195	https://ieeexplore.ieee.org/ielx7/10/9329183/09120232.pdf	English	null	The Impact of Torso Signal Processing on Noninvasive Electrocardiographic Imaging Reconstructions	IEEE transactions on bio-medical engineering/IEEE transactions on biomedical engineering	2,021	cc-by	10,143	The Impact of Torso Signal Processing on Noninvasive Electrocardiographic Imaging Reconstructions Laura R. Bear , Yesim Serinagaoglu Dogrusoz, Wilson Good , Jana Svehlikova , Jaume Coll-Font , Eelco van Dam, and Rob MacLeod , Member, IEEE Laura R. Bear , Yesim Serinagaoglu Dogrusoz, Wilson Good , Jana Svehlikova , Jaume Coll-Font , Eelco van Dam, and Rob MacLeod , Member, IEEE HFR does not impact electrogram accuracy, but does im- pact post-processing to extract features such as activation times. Removal of line noise is insufﬁcient to see these changes. HFR should be applied post-reconstruction to en- sure over-ﬁltering does not occur. Abstract—Goal: To evaluate state-of-the-art signal pro- cessing methods for epicardial potential-based nonin- vasive electrocardiographic imaging reconstructions of single-site pacing data. Methods: Experimental data were obtained from two torso-tank setups in which Langendorff- perfused hearts (n = 4) were suspended and potentials recorded simultaneously from torso and epicardial sur- faces. 49 different signal processing methods were ap- plied to torso potentials, grouped as i) high-frequency noise removal (HFR) methods ii) baseline drift removal (BDR) methods and iii) combined HFR+BDR. The in- verse problem was solved and reconstructed electrograms and activation maps compared to those directly recorded. Results: HFR showed no difference compared to not ﬁlter- ing in terms of absolute differences in reconstructed elec- trogram amplitudes nor median correlation in QRS wave- forms (p > 0.05). However, correlation and mean abso- lute error of activation times and pacing site localization were improved with all methods except a notch ﬁlter. HFR applied post-reconstruction produced no differences com- pared to pre-reconstruction. BDR and BDR+HFR signiﬁ- cantly improved absolute and relative difference, and cor- relation in electrograms (p < 0.05). While BDR+HFR com- bined improved activation time and pacing site detection, BDR alone produced signiﬁcantly lower correlation and higher localization errors (p < 0.05). Conclusion: BDR im- proves reconstructed electrogram morphologies and am- plitudes due to a reduction in lambda value selected for the inverse problem. The simplest method (resetting the isoelectric point) is sufﬁcient to see these improvements. Index Terms—ECGI, inverse problem, PVC, Signal pro- cessing, Torso Tank. I. INTRODUCTION N di N ONINVASIVE electrocardiographic imaging (ECGI) has been developed to provide high-resolution images of car- diac electrical activity. ECGI is increasingly being used to guide ablation therapy, such as in the identiﬁcation of the origin of premature ventricular contractions (PVCs) or epicardial exit sites of ventricular arrhythmias [1]–[4]. The Impact of Torso Signal Processing on Noninvasive Electrocardiographic Imaging Reconstructions Despite the increase in clinical adoption, previous validation studies of epicardial potential-based methods have shown vary- ing results with respect to accuracy. Mean localization errors for known pacing sites in studies with in-vivo animal and human subjects have ranged from 6–50 mm [5]–[8]. Recent clinical validation studies have also shown large variability in activation map reconstruction accuracy (correlation from −0.68 to 0.82 in one study [9], 0.29 to 0.80 in another [10]). The variability in accuracy seen between different centers may be linked to the different inverseor post-processingmethods usedbyeachgroup. However, given that large variability exists even in single center studies using the same inverse method pipeline, it is unlikely the inverse or post-processing methods are the sole source of this variability. Manuscript received March 23, 2020; revised May 18, 2020; ac- cepted June 7, 2020. Date of publication June 18, 2020; date of cur- rent version January 20, 2021. This work was supported in part by the French National Research Agency (ANR-10-IAHU04-LIRYC), NIH (R24GM136986 and P41GM103545), in part by the Slovak VEGA Grant Agency (2/0125/19), and in part by the Nora Eccles Treadwell Foun- dation and the Research and Development Agency (APVV-14-0875). (Corresponding author: Laura R. Bear.) ECGI is an ill-posed problem meaning any error in the model can have a drastic inﬂuence on the solution. This error can come from the presence of noise in the body surface potentials not related to the underlying cardiac electrical activity (i.e. power line interference, channel noise etc.), or the deﬁnition of the forward model itself. Signal processing can be used to minimize the impact of noise in the ECG including ﬁltering, baseline drift removal and/or signal averaging. However, over-processing of signals may introduce further error by removing or distorting po- tentially important information relating to the cardiac electrical activity that would impact the ECGI reconstruction. ( p g ) Laura R. Bear is with the IHU-LIRYC, Fondation Bordeaux Université, Université de Bordeaux and INSERM, U1045 CRCTB, 33000 Bordeaux, France (e-mail: laura.bear@ihu-liryc.fr). ( y ) Yesim Serinagaoglu Dogrusoz is with the Electrical and Electronics Engineering Department. Wilson Good and Rob MacLeod are with the Department of Biomedi- cal Engineering and SCI Institute, University of Utah. g g y Jana Svehlikova is with the Institute of Measurement Science, Slovak Academy of Sciences. Jaume Coll-Font is with the Computational Radiology Laboratory, Children’s Hospital. Jaume Coll-Font is with the Computational Radiology Laboratory, Children’s Hospital. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 436 Manuscript received March 23, 2020; revised May 18, 2020; ac- cepted June 7, 2020. Date of publication June 18, 2020; date of cur- rent version January 20, 2021. This work was supported in part by the French National Research Agency (ANR-10-IAHU04-LIRYC), NIH (R24GM136986 and P41GM103545), in part by the Slovak VEGA Grant Agency (2/0125/19), and in part by the Nora Eccles Treadwell Foun- dation and the Research and Development Agency (APVV-14-0875). (Corresponding author: Laura R. Bear.) reative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0/ (Corresponding author: Laura R. Bear.) Laura R. Bear is with the IHU-LIRYC, Fondation Bordeaux Université, Université de Bordeaux and INSERM, U1045 CRCTB, 33000 Bordeaux, France (e-mail: laura.bear@ihu-liryc.fr). Yesim Serinagaoglu Dogrusoz is with the Electrical and Electronics Engineering Department. Wilson Good and Rob MacLeod are with the Department of Biomedi- cal Engineering and SCI Institute, University of Utah. Jana Svehlikova is with the Institute of Measurement Science, Slovak Academy of Sciences. Jaume Coll-Font is with the Computational Radiology Laboratory, Children’s Hospital. Eelco van Dam is with the Peacs BV, Nieuwerbrug aan den Rijn. Digital Object Identiﬁer 10.1109/TBME.2020.3003465 ( y ) Yesim Serinagaoglu Dogrusoz is with the Electrical and Electronics Engineering Department. ( p g ) Laura R. Bear is with the IHU-LIRYC, Fondation Bordeaux Université, Université de Bordeaux and INSERM, U1045 CRCTB, 33000 Bordeaux, France (e-mail: laura.bear@ihu-liryc.fr). Wilson Good and Rob MacLeod are with the Department of Biomedi- cal Engineering and SCI Institute, University of Utah. Jana Svehlikova is with the Institute of Measurement Science, Slovak Academy of Sciences. his work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativec y Jaume Coll-Font is with the Computational Radiology Laboratory, Children’s Hospital. A. Bordeaux Data The experimental protocol used to obtain two of the four data sets has previously been described in [16] and is summarized below. All experimental procedures were approved by the Di- rective2010/63/EUoftheEuropeanParliamentontheprotection of animals used for scientiﬁc purposes and the local ethical committee. Excised pig hearts were perfused in Langendorff mode with 100% oxygenated Tyrode’s solution (pH 7.4, 37 °C). An epicar- dial electrode sock (108 electrodes with inter-electrode spacing 9.9 ± 2.2 mm) was attached to the ventricles and bipolar pacing leads to the RV apex. The hearts were transferred to a human- shaped torso tank with 128 electrodes (inter-electrode spacing 66 ± 24 mm) embedded in the surface (Fig. 1A). Tank and sock signals were referenced to an electrode at the bottom of the tank and recorded simultaneously (BioSemi, the Netherlands) at 2048 Hz for approximately 30 s during RV pacing. Afterwards, a 3D ﬂuoroscopy (Artis, Siemens) was used to obtain the position of the epicardium and electrodes with respect to the tank. Fig. 1. Torso Tank experimental setups used to obtain (A) two Bor- deaux and (B) two Utah data sets. (C) The four sets were selected: one including “Noisy” torso signals being highly contaminated by high frequency noise and base line drift (left) and three with “Clean” torso signals with minimal high frequency noise or baseline wander (right). for use with ECGI. Indeed, nearly every group implementing ECGI uses different signal processing methods. This is partially also due to the non-standardized acquisition environments, with differing hardware, electrode conﬁgurations and noise levels potentially requiring different processing methods. The effect of noise on ECGI has previously been assessed by adding noise (typically Gaussian and high frequency) to forward simulated ECGs, then solving the inverse problem to determine their effects. These computational studies have been used to demonstrate that there is an increase in error in inverse reconstructions with increasing noise levels, as would be expected [12], [13]. Furthermore, the increase in inverse reconstruction error seen with signal noise is greater in more complex forward models incorporating inhomogeneous struc- tures, meaning their accuracy becomes as good as a more simply deﬁned homogeneous model [14], [15]. The effect of removing noiseusingdifferentﬁlteringmethodsoninversereconstructions in an experimental setting has yet to be assessed. BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 437 Fig. 1. Torso Tank experimental setups used to obtain (A) two Bor- deaux and (B) two Utah data sets. (C) The four sets were selected: one including “Noisy” torso signals being highly contaminated by high frequency noise and base line drift (left) and three with “Clean” torso signals with minimal high frequency noise or baseline wander (right). data set being highly contaminated by high frequency noise and base line drift (Fig. 1C left). This higher level of noise was likely due to environmental factors (i.e. a moving reference cable or additional noise from other nearby apparatus) and was included as this is a common problem seen in clinical recordings. B. Utah Data The experimental protocol used to obtain the other two data sets has previously been described in [17], and is summarized below. The experiment described was performed under deep anesthesia using procedures approved by the Institutional An- imal Care and Use Committee of the University of Utah and conformed to the Guide for the Care and Use of Laboratory Animals. An excised canine heart was Langendorff-perfused using a mixture of whole blood and Tyrode’s solution. Arterial blood was supplied from a second canine under deep anesthesia. A right ventricular cannula extracted venous blood from the isolated heart and returned it to the support dog via a blood pump and a cannula in the jugular vein. The isolated, perfused heart was suspended in a human torso shaped tank with 192 electrodes (with inter-electrode spacing 40.2 ± 16.8mm) and ﬁlled with an electrolytic solution (500 Ω-cm). The heart was instrumented with33intramuralplungeelectrodeneedles.Inaddition,ventric- ular signals were sampled using a 247-electrode epicardial sock (inter-electrode spacing 6.5 ± 1.3 mm). Intramural needles were used to stimulate the heart using bipolar stimulation from the RV. The epicardial, intramural, and torso tank electrodes were referenced to a Wilson’s Central Terminal and were sampled at 1000 Hz simultaneously for ﬁve seconds during RV pacing. At the end of the experiment, a three-dimensional mechanical digitizer (Microscribe from Immersion Corp.) was used to locate landmarks marked on the tank and electrode array, which then provided the reference points for proper orientation of the sock geometry within the torso tank. The objective of this study was to collect the signal processing methods used by different ECGI centers and test their effects on a standard ECGI approach in the reconstruction of single-site pacing data, mimicking PVC’s seen in clinics. We have used data from two independent experimental setups to ensure the results are not biased to a single acquisition environment. The Impact of Torso Signal Processing on Noninvasive Electrocardiographic Imaging Reconstructions While guidelines exist for the processing of the standard 12- lead ECG signals to ensure appropriate interpretation [11], there is currently no consensus on the best signal processing methods Eelco van Dam is with the Peacs BV, Nieuwerbrug aan den Rijn. Digital Object Identiﬁer 10.1109/TBME.2020.3003465 II. EXPERIMENTAL DATA SETS In this study, experimental data came from two different torso tank experimental setups (Fig. 1A and B), from IHU-Liryc (Bordeaux, France) and the CRVTI/SCI Institutes (Salt Lake City, Utah), available at http://edgar.sci.utah.edu/. Four data sets were selected, three with comparable levels of noise that were relatively “clean” with little to no high frequency noise or baseline wander (example in Fig. 1C right), and one “Noisy” IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 438 Fig. 2. Filters applied to each signal included those for high frequency removal (HFR), baseline drift removal (BDR) and all combinations of the two types of ﬁlters. In addition, signal averaging was used as a special HFR + BDR ﬁlter. frequencies, 30 Hz (HFR6) and 60 Hz (HFR7) to be below or at/above the line frequency noise present in each dataset [20]. frequencies, 30 Hz (HFR6) and 60 Hz (HFR7) to be below or at/above the line frequency noise present in each dataset [20]. C. High Frequency Noise and Baseline Drift Removal The methods for HFR and BDR were combined, by ﬁrst performing each of the seven HFR on the raw signal, then applying the ﬁve BDR to each of these seven signals. This resulted in a total of 35 different HFR + BDR combinations. III. ECG SIGNAL PROCESSING METHODS 4) Cubic-Spline (BDR4): Predeﬁned expected isoelectric points of the measured signal were used for computation of the zero line by ﬁtting the isoelectric line with a cubic function. Torso signal processing methods and their associated parame- ters were collected from different international research centers working with ECGI. These were divided into three distinct cate- gories (Fig. 2). The ﬁrst were methods for high-frequency noise removal (HFR) including the 50 or 60 Hz power line interfer- ence. The second were methods for low frequency baseline drift removal (BDR) alone. The ﬁnal was for methods that combined HFR and BDR. For this ﬁnal category, all HFR and BDR meth- odswerecombined.Furthermore,asignalaveragingmethodwas included in the study. The different ﬁltering categories resulted in a total of 49 different processed signals including the original raw potentials. BDR1 was applied to directly recorded signals for all data sets. The different ﬁlters used are outlined below. 5) Butterworth High Pass Filter (BDR5): IIR high pass ﬁlters with ﬁlter order 5 at cut-off frequency of 0.5 Hz. B. Baseline Drift Removal (BDR) Five methods for BDR were applied on the original signal: 1) Simple (BDR1): A naive baseline removal approach in which the mean over a 20 or 17 ms time window during the isoelectric period prior to the QRS was subtracted from each signal. 2) Wavelet-Based (BDR2): A wavelet bandpass ﬁlter (0.5- 150 Hz) was applied with 20 levels of decomposition performed using Coiﬂet wavelets with four vanishing moments. Fig. 2. Filters applied to each signal included those for high frequency removal (HFR), baseline drift removal (BDR) and all combinations of the two types of ﬁlters. In addition, signal averaging was used as a special HFR + BDR ﬁlter. g g 3) Savitzy-Golay (BDR3): A Savitzky-Golay FIR smooth- ing ﬁlter was applied to the data using the ‘sgolayﬁlt’ function in MATLAB with a polynomial order of 3 and a frame length of 3000 ms. IV. INVERSE MAPPING METHODS Tank signals which were absent or of poor quality were excluded from the computations (3 ± 2 channels removed). The ECGI approach assessed in this study was chosen to reﬂect the most common approach used by the different research centers. 4) Savitzky-Golay Filter (HFR4): A Savitzky-Golay FIR smoothing ﬁlter was applied to the data using the ‘sgolayﬁlt’ function in MATLAB R2018a with an order of 3 and a frame length of 20 or 17 ms. D. Signal Averaging (SA) In addition to simple HFR and BDR methods, a signal aver- aging method was implemented [21]. Brieﬂy, baseline wander was ﬁrst removed using the wavelet BDR method (BDR2). The time window containing one heart beat was deﬁned by hand as the QRST interval. The tank signals were then decomposed using a principal component analysis (PCA). The ﬁrst principal component was deﬁned as the virtual lead and the pre-deﬁned heart beat in this virtual lead as the virtual template. The virtual template was compared with each beat of the virtual lead, by cross correlation. The position for the alignment was determined as the position where the cross correlation was maximal. Finally beat averaging was performed over all recorded and aligned beats for each lead. Seven different HFR methods were applied: pp 1) Moving Average Filter (HFR1): A simple moving aver- age ﬁlter computed over the 20 or 17 ms time window, corre- sponding to one cycle of the line noise present in each dataset (50 Hz for Bordeaux, 60 Hz for Utah). The weight function is constant normalized by the number of samples in window. 2) Pipberger’s Filter (HFR2): Also a moving average ﬁlter, suggestedin[18],wheretheaveragingwindowcorrespondswith the length of two cycles of 50 or 60 Hz (40 or 33 ms), and the summed samples are weighted by a cosine function. 3) Notch Filter (HFR3): Signals were transformed into the Fourier space, and a notch ﬁlter was applied around the line frequency and its harmonics before applying the inverse Fourier transform. C. Inverse Problem Solution 1) Activation Wavefront: was compared using the Pearson’s correlation and the mean absolute error (MAE). Inverse solutions were found by applying the Tikhonov regu- larization method [22] to the problem in (1) at each time instant separately. Thus, time index was removed from the following de- scription of the method. This method aims to achieve a trade-off between a good ﬁt to the measured data and an a priori constraint on the solution, thus minimizing the cost function: 2) Pacing site Localization Error (LE) - The deﬁnition of the pacing site was automated and deﬁned as the site with the earliest activation time where the median of all neigh- boring points activated within 30 ms (to prevent selecting a site within a misplaced activation marker). If multiple points demonstrated the same activation time, the mean of these points was taken as the earliest activation. The LE was then computed as the Euclidian distance between the ECGI and recorded pacing sites. J (x) = ∥Ax −y∥2 + λ2∥Rx∥2 (2) (2) with respect to x at each time instant, where ∥.∥is the L2-norm, R is a regularization matrix representing the constraint on the solution and λ is a regularization parameter controlling the trade-off between the two components of the cost function. In this study, R was chosen as the identity matrix (zero-order reg- ularization). Per C. Hansen’s L-curve method [23] was applied λ(t) at each time instant. The median over time was computed and deﬁned as the ﬁnal λ value that was used to solve the inverse problem at each time instant. Statistical analysis was conducted using GraphPad Prism 7.04. For each metric the signiﬁcance of differences was tested using 1-way ANOVA with p < 0.05 deﬁned as signiﬁcant. A repeated measures method was used for comparisons within a data set and unpaired for comparisons between the two data sets. Data are expressed as mean ± SD unless otherwise stated. Figures contain representative data, with Tables containing the complete results available as an online supplement. B. Forward Problem Solution The boundary element method (BEM) was used to deﬁne the forward matrix, employing a homogeneous conductivity be- tween a reﬁned epicardial mesh (internode spacing 5.1 ± 2.1 mm for Bordeaux and 7.4 ± 2.0 mm for Utah data) and a reﬁned tank mesh (internode spacing of 14.4 ± 5.9 mm for Bordeaux and 24.2 ± 5.1 mm for Utah data). After the forward matrix is obtained for these reﬁned meshes, the rows corresponding to torso measurement electrodes were sampled to form a reduced forward matrix, which we denote as A in (1), relating the electro- grams (in the reﬁned epicardial meshes) to torso measurements. 2) Electrogram Morphology: The morphology of ECGI reconstructed electrograms were compared to those recorded over the QRS using a Pearson’s correlation). Finally, activation times markers were deﬁned from sock electrograms as the time of minimum derivative (dV/dt). A spatio-temporal algorithm was used to deﬁne activation times from ECGI electrograms [24]. Activation maps were compared using the following metrics BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 439 A ∈RM×N is the forward matrix, and n(t) ∈RM×T is the vector representing noise in the measurements. The following features of the recorded and ECGI recon- structed electrograms were quantiﬁed and compared: 1) Electrogram Amplitude: Electrogram amplitude was measured as the mean of the peak-to-peak amplitudes from each lead. Comparison of ECGI to recorded electro- gram amplitudes were made using the absolute difference and the relative difference using the maximum recorded amplitude for normalization. VI. RESULTS First, the lambda value used for each inverse solution and the processed torso signals were evaluated using two metrics: A. Problem Deﬁnition 5) Rational Transfer Function (HFR5): A rational transfer function (RTF) implemented using MATLAB, the default ﬁlter implemented in the open-source PFEIFER toolbox [19]. This ﬁlter serves as a weighted running average with an 11-element kernel size. The electrograms on the epicardial nodes are linearly related to torso measurements: y (t) = Ax (t) + n (t) (1) (1) where x(t) ∈RN×T and y(t) ∈RM×T are the epicardial po- tential and the torso measurement vectors at time t, respectively, 6) Butterworth Low Pass Filters (HFR6 and HFR7): Two different IIR low pass ﬁlters with ﬁlter order 7 at cut-off A. Torso Surface Potentials and Lambda Values Results presented for selected HFR, BDR and their combination. In addition, SA was used as a special HFR + BDR ﬁlter. (HFR3+BDR2) improved SNR-HF by 3-10 dB for the different data sets due to the added high-pass ﬁltering effects of the wavelet ﬁlter (p < 0.001). There was also no further substantial reduction in baseline shift with the addition of HFR to BDR compared to BDR alone. The lambda value was signiﬁcantly d d i th j it f H ith th B li Fig. 3. Body Surface ECG noise properties and Lambda values for each data set. Results presented for selected high frequency removal (HFR), baseline drift removal (BDR) and combinations of the two meth- ods. In addition, signal averaging (SA) was used as a special HFR + BDR ﬁlter. Fig. 4. Comparison between absolute difference and correla BDR and their combination. In + BDR ﬁlter. (HFR3+BDR2) improved data sets due to the adde wavelet ﬁlter (p < 0.001). T reduction in baseline shift d t BDR l Fig. 3. Body Surface ECG noise properties and Lambda values for each data set. Results presented for selected high frequency removal (HFR), baseline drift removal (BDR) and combinations of the two meth- ods. In addition, signal averaging (SA) was used as a special HFR + BDR ﬁlter. Fig. 4. Comparison between recorded and ECGI electrograms using absolute difference and correlation. Results presented for selected HFR, BDR and their combination. In addition, SA was used as a special HFR + BDR ﬁlter. Fig. 4. Comparison between recorded and ECGI electrograms using absolute difference and correlation. Results presented for selected HFR, BDR and their combination. In addition, SA was used as a special HFR + BDR ﬁlter. Fig. 3. Body Surface ECG noise properties and Lambda values for each data set. Results presented for selected high frequency removal (HFR), baseline drift removal (BDR) and combinations of the two meth- ods. In addition, signal averaging (SA) was used as a special HFR + BDR ﬁlter. (HFR3+BDR2) improved SNR-HF by 3-10 dB for the different data sets due to the added high-pass ﬁltering effects of the wavelet ﬁlter (p < 0.001). There was also no further substantial reduction in baseline shift with the addition of HFR to BDR compared to BDR alone. The lambda value was signiﬁcantly reduced in the majority of cases. A. Torso Surface Potentials and Lambda Values However, as with the Baseline shift and SNR-HF the relative change was minimal with a mean difference ranging from 0.00001 to 0.0008. For all data sets, no BDR methods signiﬁcantly improved the SNR-HF of the torso signals except the wavelet method (BDR2) as this includes a high frequency cut-off of 150 Hz. Though signiﬁcant, the improvement was small. Most BDR methods demonstrated a marked improvement in the baseline shift.However,fortheUtah1dataset(blue),thewavelet(BDR2) and the high pass ﬁlter (BDR5) did not signiﬁcantly change the baseline offset. Furthermore, the spline method (BDR4) signiﬁcantly increased the baseline shift. Here, while the spline ﬁt was accurate, it created a constant offset across the entire signal due to the automated method for ﬁnding the isoelectric points deﬁning a point at the start of the QRS. For all data sets, there was a signiﬁcant reduction in lambda with BDR, with no difference between the BDR methods. SA improved the SNR-HF and baseline shift compared to no ﬁltering for all data, as well as reducing lambda values (p < 0.05). Globally, SA showed similar SNR-HF and baseline shift values to many HFR + BDR combinations, though the SNR-HF was greater than any method for Bordeaux 1 data. A. Torso Surface Potentials and Lambda Values 1) Signal-to-Noise Ratio for High Frequencies (SNR-HF): The RMS voltage (RMSV) was computed across all leads after a simple baseline correction (BDR1). The SNR-HF was then taken as the mean ratio between the QRS and noise amplitudes during an isoelectric period 40 or 34 ms prior to the QRS for Bordeaux and Utah data respectively. SNR-HF, Baseline Shift and lambda values for selected pro- cessing methods for all data sets are presented in Fig. 3. As expected, the SNR-HF of the raw potentials for the three clean data sets were signiﬁcantly larger than for the noisy data (Bor- deaux 1; red), but also had signiﬁcantly less baseline shift and 5x smaller lambda values. 2) Baseline Shift: was deﬁned as the deviation of the iso- electric point prior to the QRS. To account for noise, the average was taken over 20 or 17 ms. The mean absolute deviation over all leads was used. All HFR methods except the notch ﬁlter (HFR3) signiﬁcantly improved the SNR-HF compared to no ﬁltering for all data sets (p < 0.0001) with the moving average (HFR1), Pipberger (HFR2) and the 30 Hz Lowpass ﬁlter (HFR6) producing the best results. The improvements seen with the RTF (HFR5) and a low-pass ﬁlter above the line noise (HFR7) were small. The Notch ﬁlter signiﬁcantly improved SNR-HF for Bordeaux 1 where lines noise was prominent. HFR did not signiﬁcantly change the baseline shift for any data sets. While there was a signiﬁcant difference in lambda values between HFR methods, the absolute differences were minimal. The inverse solution was found for each beat in the signal (14- 31 beats) and compared to the same beat measured by the sock. As SA produced a single beat, reconstructed electrograms were compared to all un-averaged sock beats. The inverse solutions were obtained for reﬁned epicardial meshes and comparisons were carried out at a subset of the nodes corresponding to sock measurement locations (108 and 247 leads for Bordeaux and Utah data respectively). IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 440 Fig. 3. Body Surface ECG noise properties and Lambda values for each data set. Results presented for selected high frequency removal (HFR), baseline drift removal (BDR) and combinations of the two meth- ods. In addition, signal averaging (SA) was used as a special HFR + BDR ﬁlter. Fig. 4. Comparison between recorded and ECGI electrograms using absolute difference and correlation. B. Electrogram Reconstruction Within the HFR+BDR plots, reconstructions after a low pass ﬁlter (HFR6; blue) and the RTF ﬁlter (HFR5; orange) combined with the spline ﬁlter (BDR4) are presented alongside the signal averaged reconstruction (yellow). reﬂection. By using a higher 60 Hz cut-off for the low-pass ﬁlter (Fig. 6C magenta dashed); the second down stroke is preserved. Interestingly, by performing HFR post-reconstruction (Fig. 6B red and C yellow lines), we produce the same signal as when ﬁltering was performed pre-reconstruction. This was also seen for reconstructions after BDR (Fig. 6D and E), where HFR7 and 8 performed pre-reconstruction (green and magenta dotted re- spectively) match the electrograms with HFR7 and 9 performed post-reconstruction (red and yellow dotted respectively). HFR did not signiﬁcantly improve electrogram amplitudes compared to no ﬁltering for the Noisy data. For the Clean data sets all methods signiﬁcantly decreased amplitudes except the notch ﬁlter (HFR3). The moving average (HFR1) showed the largest difference. This can be seen in Fig. 5, which presents representative recorded (black) and ECGI electrograms for the Bordeaux 1 (top) and Utah 1 (bottom) data sets at two electrode locations marked on recorded activation maps (left). Here, the ECGI electrograms using the moving average (HFR1) alone (yellow) are slightly smaller than those without ﬁltering (blue) or using a notch ﬁlter (HFR3; orange) in the clean data. BDR improved potential amplitudes for all data compared to no ﬁltering with a signiﬁcant reduction in absolute differ- ences (Fig. 4). The spline method (BDR4) provided the greatest improvement, and simple/wavelet methods (BDR1/BDR2) the least. While amplitudes were improved, the noise level in the electrograms was also dramatically increased reducing the SNR, as seen in Fig. 5 BDR electrograms for both data sets, with no ﬁltering in blue, wavelet (BDR2) in orange and simple (BDR1) in yellow. The wavelet method provided some ﬁltering of high frequencynoiseasexpectedfrombodysurfacesignals.BDRsig- niﬁcantly improved the median correlation between ECGI and recorded potentials compared to no ﬁltering for both Bordeaux and the Utah 1 data. However, for the Utah 2 data the correla- tion was overall reduced. There were no signiﬁcant differences between the BDR methods for all data sets (p > 0.99). While certain HFR methods did signiﬁcantly improve the correlation for both Bordeaux 1 and Utah 1 data compared to no ﬁltering, the improvements were minimal (0.2–2.3% maxi- mum). B. Electrogram Reconstruction Fig. 4 presents the absolute amplitude difference and corre- lation between ECGI and recorded electrogram over the QRS interval for selected ECG processing techniques. With no ﬁltering of the ECG, ECGI electrogram amplitudes were smaller than recorded for all data sets. With no ﬁltering, the Noisy data set showed signiﬁcantly worse median correla- tion in electrograms QRS waveform between reconstructed and recorded potentials than any Clean data set (52 ± 3% vs. 67 ± 4 to 76 ± 1%). For all data sets, there was no signiﬁcant improvement or reduction in SNR-HF in most cases with the addition of BDR to HFR compared to using HFR methods alone (Hence only HFR1+BDR1 is represented). The exceptions to this are with the Notch ﬁlter which in combination with the Wavelet BDR BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 441 Fig. 5. Representative recorded and reconstructed electrograms using selected signal processing methods. Representative recorded EGMs for both data sets at two electrode locations are marked on recorded activation maps (left). HFR, BDR and HFR+BDR depict the selected reconstructed EGMs for HFR only, BDR and the combination of HFR+BDR respectively. Within the HFR plots, the blue line represents reconstructions with no ﬁltering, the yellow line with a moving average ﬁlter (HFR1) and the orange line with a notch ﬁlter (HFR3). Within the BDR plots, the blue line represents reconstructions with no ﬁltering, the orange line with a wavelet ﬁlter (BDR2) and the yellow line with a simple BDR (BDR1). Within the HFR+BDR plots, reconstructions after a low pass ﬁlter (HFR6; blue) and the RTF ﬁlter (HFR5; orange) combined with the spline ﬁlter (BDR4) are presented alongside the signal averaged reconstruction (yellow). Fig. 5. Representative recorded and reconstructed electrograms using selected signal processing methods. Representative recorded EGMs for both data sets at two electrode locations are marked on recorded activation maps (left). HFR, BDR and HFR+BDR depict the selected reconstructed EGMs for HFR only, BDR and the combination of HFR+BDR respectively. Within the HFR plots, the blue line represents reconstructions with no ﬁltering, the yellow line with a moving average ﬁlter (HFR1) and the orange line with a notch ﬁlter (HFR3). Within the BDR plots, the blue line represents reconstructions with no ﬁltering, the orange line with a wavelet ﬁlter (BDR2) and the yellow line with a simple BDR (BDR1). C. Activation Maps 6B and C), but is present in all reconstructions with BDR (Fig. 6D and E). In a few electrograms, detrimental changes were found after BDR, most prominently in the Utah 2, e.g., in Fig. 5 electrode 2 an initial downstroke is reconstructed after BDR that does not exist in the recording nor in unﬁltered or HFR-alone reconstructions. g For the Bordeaux 1 data, HFR signiﬁcantly improved correlation of activation maps with those recorded, except with the Notch ﬁlter. For the other data sets, no methods changed the correlation except the moving average (HFR1) and 30 Hz low pass (HFR6), which reduced it. Closer inspection of reconstructed maps from the Bordeaux 1 data set revealed that ﬁltering removed isolated activation marker errors (HFR3; Figs. 7 right wall and 8 posterior wall). For the majority of the beats, the Notch ﬁlter showed similar but smoother activation maps than no ﬁltering. However, in approximately a ¼ of the beats the activation map was changed drastically in the Noisy data, as seen in Fig. 10 beat 7. Inspection of the electrograms in altered regions showed the global QRS morphology was unchanged (as was noted in the previous section), but large amplitude high frequency noise was still present after the notch ﬁlter that has likely altered activation marker placements. Other HFR methods also altered the activation sequences compared to no ﬁltering (example Fig. 7 with the moving average ﬁlter) but unlike with the notch they showed little beat-to-beat variability. On the other hand, for the other data sets there were no substantial visual differences between the HFR ﬁltered and Unlike with HFR, changes in electrogram morphology with BDR were only produced when ﬁltering was applied prior to inverse reconstruction, and could not be reproduced through ﬁltering post-reconstruction. The combination of HFR and BDR on ECG signals provided a signiﬁcant improvement in the signal amplitudes for all datasets by reducing the absolute amplitude difference when compared to HFR alone (p < 0.05), but were increased slightly compared to BDR alone. For both Bordeaux and Utah 1 data sets, most combinations of HFR methods with BDR also signiﬁcantly improved the correlation of QRS EGMs compared to BDR alone though these absolute changes were minimal (<0.04 change in the median correlation). Only in combination with the RTF ﬁlter (HFR5) were correlation values reduced compared to BDR alone (p < 0.05). Representative re- constructed electrograms in Fig. B. Electrogram Reconstruction Visual inspection demonstrated no obvious changes in QRS morphology when correlation was improved other than a reduction in high frequency noise content (Fig. 5 - HFR). For the Utah 2 data set, the moving average ﬁlter (HFR1) and the 30 Hz low pass ﬁlter (HFR6) signiﬁcantly reduced the correlation values. For this data set, visual inspection demonstrated that these ﬁlters have resulted in an over-smoothed reconstruction on a small area of the heart. This is demonstrated in Fig. 6A and B with a recorded electrogram (black) and the equivalent inverse reconstruction without ﬁltering (dark blue) and with a 30 Hz low pass ﬁlter (HFR6; green dashed). As can be seen, HFR6 has over-smoothed the inverse reconstruction, remov- ing the second downstroke that corresponds to the intrinsic In all data sets, when improvements in QRS morphology occurred, these were substantial. As seen in Fig. 5 BDR plots key electrogram features were better reproduced in ECGI IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 442 Fig. 6. Example (A) recorded and (B,C,D,E) reconstructed electro- grams with (right) and without (left) simple BDR (BDR1), and either no HFR, a 30 Hz (HFR6) or a 60 Hz (HFR7) low pass ﬁlter used either pre- or post- reconstruction. (BDR4) presented alongside the signal averaged reconstruction (yellow). Signal averaging, like BDR and BDR + HFR, improved signal amplitudes compared to no ﬁltering. While for the Utah 1 and Bordeaux 2 data sets, correlation of QRS EGMs were signiﬁcantly better than for any other method, for the noisy Bor- deaux 1 data set and the Utah 2 data set they were signiﬁcantly worse. This can be seen in Fig. 5 electrode 2, where the signal averaged reconstruction (yellow plot HFR + BDR) has inverted the electrograms completely now showing no similarity to the recorded plot. For all data sets, signal averaged reconstructions still contain high frequency noise likely because not enough beats are used in the averaging (14 to 31 beats used). C. Activation Maps Figs. 7 and 8 present representative recorded (A) and ECGI activation maps for the Bordeaux 1 and Utah 1 data sets, respec- tively with (B) no ﬁltering, (C and D) HFR alone, (E and F) BDR alone, (G) BDR + HFR and (H) signal averaging. Fig. 9 presents the correlation and mean absolute error (MAE) between ECGI and recorded activation times as well as the localization error (LE) for selected processing methods. Activation maps without ﬁltering were well correlated with those recorded and had a low MAE for all data, though results were best for the clean data sets. Pacing site LE was also signiﬁcantly higher for the noisy Bordeaux 1 data set (p < 0.05). Variation between the activation maps on a beat-to-beat basis was also greater for the noisy Bordeaux 1 data as demonstrated by the larger standard deviation in correlation and MAE com- pared to the other data. These beat-to-beat changes in the ECGI activation maps are presented in Fig. 10for the Bordeaux 1 data. Beat-to-beat variability in the recorded sock data was minimal, with each activation maps showing no discernible difference, il- lustrating this variability arises when signiﬁcant noise is present in the ECG signals. Fig. 6. Example (A) recorded and (B,C,D,E) reconstructed electro- grams with (right) and without (left) simple BDR (BDR1), and either no HFR, a 30 Hz (HFR6) or a 60 Hz (HFR7) low pass ﬁlter used either pre- or post- reconstruction. electrograms using the wavelet (orange) and simple (yellow) BDR compared to no ﬁltering (blue); including the initial R- wave in electrode 1 and 3, and the S-wave in electrode 4. This is also seen in Fig. 6 for the Utah 2, where the R-wave is not present in reconstructions without BDR (Fig. 6B and C), but is present in all reconstructions with BDR (Fig. 6D and E). In a few electrograms, detrimental changes were found after BDR, most prominently in the Utah 2, e.g., in Fig. 5 electrode 2 an initial downstroke is reconstructed after BDR that does not exist in the recording nor in unﬁltered or HFR-alone reconstructions. electrograms using the wavelet (orange) and simple (yellow) BDR compared to no ﬁltering (blue); including the initial R- wave in electrode 1 and 3, and the S-wave in electrode 4. This is also seen in Fig. 6 for the Utah 2, where the R-wave is not present in reconstructions without BDR (Fig. C. Activation Maps 5 HFR+BDR demonstrate that these changes are limited to a reduction in high frequency noise content,withreconstructionusingalowpassﬁlter(HFR6)(blue) and the RTF ﬁlter (HFR5) (orange) combined with spline BDR BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 443 Fig. 7. Representative recorded and reconstructed activation maps using selected signal processing methods for the Bordeaux 1 data set. Activation maps show right (left) and left (right) ventricular views of the heart. Lightning bolt represents sites of earliest activation as deﬁned by each map. Fig. 7. Representative recorded and reconstructed activation maps using selected signal processing methods for the Bordeaux 1 data set. Activation maps show right (left) and left (right) ventricular views of the heart. Lightning bolt represents sites of earliest activation as deﬁned by each map. Fig. 8. Representative recorded and reconstructed activation maps using selected signal processing methods for the Utah 1 data set. Activations show and anterior (right) and posterior (left) view of the heart. Lightning bolt represents sites of earliest activation as deﬁned by each map. Fig. 8. Representative recorded and reconstructed activation maps using selected signal processing methods for the Utah 1 data set. Activations show and anterior (right) and posterior (left) view of the heart. Lightning bolt represents sites of earliest activation as deﬁned by each map. After BDR, the increased noise amplitude, seen in Fig. 5, neg- atively impacted activation marker placement, as seen in Figs. 7 and 8 where activation maps became patchy after BDR. Overall there was a decrease in activation accuracy with BDR despite the improvements seen in electrogram accuracy. The wavelet method (BDR2) was an exception to this, improving activation map accuracy in terms of correlation and MAE as it provided some HFR to electrograms. However, for the Bordeaux 1 data set, an artefactual second region of early activation appeared on theposteriorwallimpactingLEforallactivationmapsafterBDR (Fig. 7E). Inspecting the electrograms in this region revealed an artefactual early downslope that is reconstructed only after the unﬁltered reconstructed activation maps other than shifts in the early activated region (Fig. 8B, C and D). Like with the Bordeaux 1 data, this was not reﬂected in the electrograms which had very similar morphologies in the early activated region. The already minimal beat-to-beat variability also did not change with ﬁltering. For both Bordeaux and Utah 1 data sets, while there were HFR methods that improved the mean LE, or reduced the standard deviation, none were signiﬁcant. VII. DISCUSSION Thisstudyhasdemonstratedtheimpactofdifferentsignalpro- cessing methods on an epicardial potential based inverse method using four data sets from two distinct experimental setups. As there was no one optimal processing method for all data, we recommendsignalprocessingbeappliedonacase-by-casebasis. Several trends were applicable to all data sets that can help guide the choice of processing method used and ensure the best inverse solution is found: 1) HFR does not impact electrogram reconstruction accu- racy, but does impact post-processing to extract features such as activation times. HFR should be applied post- reconstruction to ensure over-ﬁltering does not occur. The HFR methods and parameters will depend on the features one wishes to extract. 2) BDR improves reconstructed electrogram due to a reduc- tion in lambda value selected. We recommend using the simple method (BDR1) as this is sufﬁcient to see these improvements. 3) Signal averaging may be a useful processing tool but care should be taken in aligning the beats. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 444 Fig. 9. Comparison between recorded and reconstructed activation maps using correlation, mean absolute error (MAE) and localization error (LE). Results presented for selected HFR, BDR and combinations of the two methods. In addition, SA was used. SA produced activation maps with similar correlation and MAE values to no ﬁltering for the Utah and Bordeaux 2 data but lower correlation and higher MAE for the Bordeaux 1 data. While activation maps for the Utah data looked very similar to thosewithoutﬁltering,fortheBordeaux1datatherewereseveral artefacts including a second site of activation on the right surface (like with all BDR or BDR+HFR ﬁlters). For both all data, the SA found the earliest site of activation <15 mm, in the closer range compared to any other processing method. C. Activation Maps For Utah 2, LEs were larger but not signiﬁcantly. Interestingly there were methods that improved LE for one data set but deteriorated results for another. The only consistent change was for the moving average ﬁlter (HFR1) that produced larger LE than using no ﬁlter (p < 0.05). the unﬁltered reconstructed activation maps other than shifts in the early activated region (Fig. 8B, C and D). Like with the Bordeaux 1 data, this was not reﬂected in the electrograms which had very similar morphologies in the early activated region. The already minimal beat-to-beat variability also did not change with ﬁltering. For both Bordeaux and Utah 1 data sets, while there were HFR methods that improved the mean LE, or reduced the standard deviation, none were signiﬁcant. For Utah 2, LEs were larger but not signiﬁcantly. Interestingly there were methods that improved LE for one data set but deteriorated results for another. The only consistent change was for the moving average ﬁlter (HFR1) that produced larger LE than using no ﬁlter (p < 0.05). MPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 445 Fig. 10. ECGI activation maps for Bordeaux 1 data demonstrating the beat-to-beat variability (left-right) using selected processing methods (top-bottom). Fig. 10. ECGI activation maps for Bordeaux 1 data demonstrating the beat-to-beat variability (left-right) using selected processing methods (top-bottom). no ﬁltering. As positive changes are also seen with the simple method, we would advise this approach as there appears to be no beneﬁt from using more complex BDR approaches. In order to limit detrimental changes, the method for computing the lambda value needs to be optimized. the LE of the pacing site was larger than without ﬁltering. We suspect this is due to over-smoothing the early QRS at the body surface, that resulted in an increase in the presence of isoelectric- line clustering or apparent line of conduction slowing and shifted the early activation site to these borders, as seen in Figs. 7 and 8 (earliest activation denoted with yellow lightning bolt). By using a smaller time window for averaging this defect may not occur. While BDR signiﬁcantly improved speciﬁc morphological features, the presence of high frequency noise had the biggest impact on activation maps, and therefore HFR methods were required to improve their accuracy. If the reconstructed signals had little high frequency noise, HFR had little impact (i.e. the three clean data sets after only HFR). Of the HFR methods eval- uated in this study, the notch ﬁlter was the only method deemed insufﬁcient to improve activation time mapping given the high frequency noise was still present in the signal. Even a low pass ﬁlter above the line noise improved activation reconstruction accuracy, indicating the line noise does not impact activation marker placement. Further investigation into ﬁltering artefacts at the body sur- face demonstrated that if these are present they will remain in reconstructions. For example, the low pass ﬁlter of 30 Hz created an obvious ringing on either side of QRS in the Clean data set. Whilst this does not affect the correlation values or timing of the intrinsic deﬂection, the ringing was still present in the reconstructed electrograms and could be mistaken as late potentials after the QRS. For a purely spatial (static) inverse approach, as used in this study, the regularized pseudo-inverse matrix is a purely spatial operator, treating every timestep of the torso signals individually. y(t)filtered = My (t) A. High Frequency and Baseline Drift Removal Torso signals with baseline drift present (i.e. no ﬁltering or HFR) resulted in a higher regularization parameter choice using the L-curve method (Fig. 3). The more baseline drift present, the more regularization the signals received. With Tikhonov zero-order regularization, the presence of baseline drift resulted in very low amplitude and smooth reconstructed signals. Con- versely, reconstructions after BDR improved electrogram mor- phology and amplitudes. Fig. 9. Comparison between recorded and reconstructed activation maps using correlation, mean absolute error (MAE) and localization error (LE). Results presented for selected HFR, BDR and combinations of the two methods. In addition, SA was used. Improvement in electrograms after BDR included the ap- pearance of key features like the initial R-wave (Fig. 5 and 6). Though less common, detrimental changes could also be found, such as the development of a double intrinsic deﬂection (Fig. 5). Both positive and negative changes were present with all BDR methods including the simple method (BDR1), meaning these changes are likely due to the selected lambda value rather than distortion in the torso surface QRS from ﬁltering. To ver- ify this, reconstructions were compared before and after BDR using a ﬁxed lambda value and demonstrated no perceivable differenceinQRSmorphology.Weconcludethechangesinelec- trogram morphology after BDR are due to the large reduction in lambda value selected. This conclusion is further supported by HFR-only reconstructions where electrogram morphologies and lambda values are not signiﬁcantly different from those with BDR (as demonstrated in Fig. 5 electrogram 2). For all except the Bordeaux 2 data, all of the BDR methods except the wavelet method increased the mean and standard deviation of the LE for the earliest site of activation. For the Utah 1 data set this increase was signiﬁcant (p < 0.05). The combination of HFR and BDR provided no substantial further improvement in correlation or MAE over HFR alone. Furthermore, the pattern of activation was not substantially changed. Inspection of the maps themselves demonstrated that the artefactual region of early activation was present in all HFR+BDR reconstructions for the Bordeaux 1 data set (ex- ample Fig. 7G). No such artefacts were seen in the other data sets. Though not signiﬁcant, the mean and SD in LE improved substantially for most combinations, particularly for cases that had increased with only HFR (HFR1, 6). B. Signal Averaging The application of averaging on an ECG makes three as- sumptions, i) the ECG pattern repeats cyclically, ii) the high frequency noise superimposed on the signal is a random noise with zero mean value, and iii) ECG can be correctly aligned to perform averaging. If all three assumptions are correct, then signal averaging will prevent any distortion of the ECG and yield a considerable decrease of the noise dispersion. Here we have found signal averaging to be beneﬁcial for the clean data and detrimental for the noisier Bordeaux 1 data set. This is despite the SNR-HF being improved and QRS waveforms at the tank surface appearing normal. We assume the difference in results comes down to the alignment of the beats; good for the clean data sets with no QRS deformation at the body surface and bad for the noisy data set resulting in an unperceivable QRS deformation in the ECG. It is clear that if alignment is not good, signal averaging can be more detrimental than any other ﬁltering approach to the inverse solution. This is conﬁrmed with the large deformations seen in the reconstructed electrograms for the Bordeaux 1 data with signal averaging that are not present for the other data sets. We hypothesize that if the optimal alignment approach can be determined, it would provide the best ﬁltering tool to use for stable rhythms as there would be no QRS deformation. As such we have commenced an investigation into different signal averaging methods in relation to the inverse problem. The signals on the heart and torso were recorded simultane- ously for all data sets to avoid alignment issues and beat-to-beat variability. This means the noise present on the torso is also present in the epicardial signals and may bias the results. How- ever, because the amplitude of the epicardial signals is much greater than on the torso, this bias does not impact the results of this study. To test this, we have compared reconstructions without ﬁltering using only the 1st and 18th epicardial beats as the ground truth. The was nearly no difference in any metrics, e.g. the difference in absolute error was 0.03–0.07 mV higher than doing a beat-to-beat comparison, a change of less than 1.5%, and the correlation change was 0.2–1.4% reduced. There was no change for activation maps. B. Signal Averaging The study was conducted in an experimental setting, and it cannot fully represent the recording environment of a clinical setting. Here, only high-frequency noise from powerline in- terference and the baseline drift were considered. In a clinical environment, patient/electrode movement, breathing, and noise from muscles would also affect the signals. Despite this, we expect our results are directly transferable to the clinic. The channel white Gaussian noise present in the tank data sets are of a similar frequency to electromyogram noise. Likewise, the level of baseline drift present in the noisy tank data are of a similar level to that seen during breathing or patient movement, although morphologically different. MPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS A linear temporal ﬁlter can be expressed as a matrix M, likewise treating every timestep of the torso signals individually The moving average (HFR2) and 30 Hz low pass (HFR6) ﬁlter methods were detrimental to deﬁning activation times for the Utah 2 data set by over-smoothing the reconstructed electrogram (Fig. 6). Interestingly, while the global activation pattern was improved with the moving average ﬁlter for the other data sets, y(t)filtered = My (t) IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 68, NO. 2, FEBRUARY 2021 446 ﬁrst-order Tikhonov regularization would result in smoother re- constructions, while with BDR they would be more fractionated due to the smaller lambda values used. To test these hypotheses, we are currently running a variety of different methods on these data sets, which are also available for those who wish to test their own methods. The method to deﬁne activation times was based on a speciﬁc spatio-temporal approach combining the estimated time delays between neighboring electrograms and the standard maximum negative derivative approach [24]. Analysis using the standard derivative method to deﬁne activation produced the same trends as the spatio-temporal method, but the overall ac- curacy was worse due to the increase in activation time artefacts from high frequency noise. While we suspect the same trends would also be seen with other activation methods, these have not been assessed. Therefore, the order of temporal ﬁltering and purely spatial (static) reconstruction can be also formally be exchanged, as long as the same lambda is used. We therefore recommend applying HFR after reconstruction when using a purely spatial (static) inverse approach to ensure there are no deformations in morphology and accuracy in activation map computation. However, it should be kept in mind that for spatio-temporal inverse approaches the order does make a difference. REFERENCES [17] S. Shome and R. Macleod, “Simultaneous high-resolution electrical imag- ingofendocardial,epicardialandtorso-tanksurfacesundervaryingcardiac metabolic load and coronary ﬂow,” in FIMH, Berlin Heidelberg: Springer, 2007, pp. 320–329. [1] Y. Wang et al., “Noninvasive electroanatomic mapping of human ventric- ular arrhythmias with electrocardiographic imaging,” Sci. Transl. Med., vol. 3, no. 98, 2011, Art. no. 98ra84. [1] Y. Wang et al., “Noninvasive electroanatomic mapping of human ventric- ular arrhythmias with electrocardiographic imaging,” Sci. Transl. Med., vol. 3, no. 98, 2011, Art. no. 98ra84. pp [18] F. Stallmann and H. Pipberger, “Automatic Recognition of Electrocardio- graphic Waves by Digital Computer,” Circ Res, vol. IX, pp. 1138–1143, 1961. [2] M. Hocini et al., “Focal arrhythmia ablation determined by high reso- lution non- invasive maps: Multicenter feasibility study,” J. Cardiovasc. Electrophysiol., vol. 26, no. 7, pp. 754–760, 2015. [2] M. Hocini et al., “Focal arrhythmia ablation determined by high reso- lution non- invasive maps: Multicenter feasibility study,” J. Cardiovasc. Electrophysiol., vol. 26, no. 7, pp. 754–760, 2015. [19] A. Rodenhauser et al., “PFEIFER: Preprocessing framework for electro- grams intermittently ﬁducialized from experimental recordings,” J. Open Source Softw., vol. 3, no. 21, p. 472, 2018. [3] R. Dubois et al., “Non-invasive cardiac mapping in clinical practice: Application to the ablation of cardiac arrhythmias,” J. Electrocardiol., vol. 48, no. 6, pp. 966–974, 2015. [3] R. Dubois et al., “Non-invasive cardiac mapping in clinical practice: Application to the ablation of cardiac arrhythmias,” J. Electrocardiol., vol. 48, no. 6, pp. 966–974, 2015. [20] R. M. Rangayyan, Biomedical Signal Analysis. A Case-Study Approach. Washington, DC, USA: Wiley, 2002. [4] T. Berger et al., “Single-beat noninvasive imaging of ventricular endocar- dial and epicardial activation in patients undergoing CRT,” PLOS ONE, vol. 6, no. 1, 2011, Art. no. e16255. [4] T. Berger et al., “Single-beat noninvasive imaging of ventricular endocar- dial and epicardial activation in patients undergoing CRT,” PLOS ONE, vol. 6, no. 1, 2011, Art. no. e16255. g y [21] N. Tan et al., “Analysis of signal-averaged electrocardiogram performance for body surface recordings,” Comput. Cardiol., Sep. 2019. [5] M. Cluitmans et al., “In-vivo validation of electrocardiographic imaging,” JACC Clin. Electrophysiol., vol. 3, no. 3, pp. 1–11, 2017. [22] A. Tikhonov and V. Arsenin, Solution of ill-posed problems. Washington, DC, USA: Wiley, 1977. [6] L. R. Bear et al., “How accurate is inverse electrocardiographic mapping?” Circ. Arrhythmia Electrophysiol., vol. 11, no. 5, 2018, Art. no. e006108. [23] P. C. Hansen, D. P. O’Leary, and D. VIII. CONCLUSION VIII. CONCLUSION [9] A. Graham et al., “Simultaneous comparison of electrocardiographic imaging and epicardial contact mapping in structural heart disease,” Circ. Arrhythm. Electrophysiol., vol. 12, pp. 1–10, 2019. ECG signal processing has a fairly low impact on ECGI reconstruction accuracy. Removal of baseline drift improves electrogram reconstruction due to an improvement in the lambda selected for regularization, with a simple method of reset- ting the isoelectric point sufﬁcient to see these improvements. High frequency noise removal does not impact electrogram reconstruction accuracy, but can improve post-reconstruction feature extraction. High frequency removal should be applied post-reconstruction with care to ensure over-ﬁltering does not occur. [10] J. Duchateau et al., “Performance and limitations of noninvasive cardiac activation mapping,” Hear. Rhythm, vol. 16, no. 3, pp. 435–442, 2018. pp g pp [11] P. Kligﬁeld et al., “Recommendations for the standardization and in- [11] P. Kligﬁeld et al., “Recommendations for the standardization and in- terpretation of the electrocardiogram,” Circulation, vol. 115, no. 10, pp. 1306–1324, 2007. g terpretation of the electrocardiogram,” Circulation, vol. 115, no. 10, pp. 1306–1324, 2007. [12] L. K. Cheng et al., “Effects of experimental and modeling errors on electrocardiographic inverse formulations,” IEEE Trans. Biomed. Eng., vol. 50, no. 1, pp. 23–32, Jan. 2003. [13] J. P. Barnes and P. R. Johnston, “Application of robust Generalised Cross- Validation to the inverse problem of electrocardiology,” Comput. Biol. Med., vol. 69, pp. 213–225, 2016. [14] C. Ramanathan and Y. Rudy, “Electrocardiographic imaging: II. Effect of torso inhomogeneities on noninvasive reconstruction of epicardial po- tentials, electrograms, and isochrones,” J. Card. Electrophysiol., vol. 12, no. 2, pp. 241–252, 2001. ACKNOWLEDGMENT ACKNOWLEDGMENT [15] C. P. Bradley, A. J. Pullan, and P. J. Hunter, “Effects of material properties and geometry on electrocardiographic forward simulations,” Ann. Biomed. Eng, vol. 28, no. 7, pp. 721–741, 2000. The authors are grateful for the support of the Consortium for ECG Imaging (CEI). We would like to thank Steffen Schuler for his commentary on spatio-temporal approaches. [16] L. R. Bear et al., “Cardiac electrical dyssynchrony is accurately detected by noninvasive electrocardiographic imaging,” Heart. Rhythm, vol. 15, no. 7, pp. 1058–1069, 2018. BEAR et al.: IMPACT OF TORSO SIGNAL PROCESSING ON NONINVASIVE ELECTROCARDIOGRAPHIC IMAGING RECONSTRUCTIONS 447 C. Limitations The results presented should be considered in light of limita- tions. First, while a large variety of signal processing methods have been used, the study has only investigated one inverse problem “pipeline”. That is, one forward model (BEM), one regularization technique (zero-order Tikhonov), one method to deﬁne lambda (L-curve), and one method to compute activation times. Previous computational studies have demonstrated that the relationship between signal noise and error in inverse re- constructions is dependent on the regularization and parameter selection methods used [12], [13]. We have demonstrated that signal processing (and therefore noise) does not dramatically change the inverse reconstruction, but rather has a major impact on the lambda value chosen (using an L-curve method) and therefore the accuracy of the inverse reconstruction. Different regularization parameter selection methods may be more or less sensitive to noise. Furthermore, the impact of different lambda values would change with different regularization methods, e.g. a higher lambda from the presence of baseline drift with The sites of initial activation were determined automatically from activation maps. It is possible that an expert observer may have been able to identify sites of initial activation more precisely and that LE are over-estimated as a result. On the other hand, with our approach it was possible to analyze this large dataset efﬁciently and investigator bias was removed. Finally, these methods were only evaluated in the presence of single-site pacing data. Given the low impact of signal process- ing on the inverse reconstruction, we expect the results of this study are applicable not only to stable “single” beat rhythms of clinical interest such as premature ventricular contractions or sinus rhythm, but also to the reconstruction of non-stable or multi-beat rhythms such as ﬁbrillation or tachycardia, and for the reconstruction of repolarization. The question with these later cases is how to deﬁne the isoelectric point to ensure that the optimal lambda value is chosen. [8] J. Sapp et al., “Inverse solution mapping of epicardial potentials,” Circ Arrhythm Electrophysiol, vol. 5, no. 5, pp. 1001–1009, 2012. REFERENCES P. O’Leary, “The use of the L-curve in the regularization of discrete ill-posed problems,” SIAM J. Sci. Comput., vol. 14, pp. 1487–1503, 1993. [7] A. S. Revishvili et al., “Validation of the mapping accuracy of a novel non- invasive epicardial and endocardial electrophysiology system,” Europace, vol 17, pp. 1–7, 2015. [24] J. Duchateau et al., “Spatially coherent activation maps for electrocar- diographic imaging,” IEEE Trans. Biomed. Eng., vol. 64, no. 5, pp. 1–8, 2016. [8] J. Sapp et al., “Inverse solution mapping of epicardial potentials,” Circ Arrhythm Electrophysiol, vol. 5, no. 5, pp. 1001–1009, 2012.
https://openalex.org/W3029144977	https://repository.publisso.de/resource/frl:6421165/data	English	null	Mental health and psychosocial well-being during the COVID-19 pandemic: the invisible elephant in the room	International journal of mental health systems	2,020	cc-by	4,652	© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativeco mmons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. International Journal of Mental Health Systems International Journal of Mental Health Systems Open Access Mental health and psychosocial well‑being during the COVID‑19 pandemic: the invisible elephant in the room Akaninyene Otu1,2, Carlo Handy Charles3,4 and Sanni Yaya5,6* Abstract The novel SARS-CoV-2 coronavirus pandemic has emerged as a truly formidable threat to humankind’s existence. In the wake of the massively volatile global situation created by COVID-19, it is vital to recognize that the trauma it causes can affect people in different ways, at the individual and collective levels, resulting in mental health challenges for many. Although mental health problems account for about one-third of the world’s disability among adults, these issues tend to be under-addressed and overlooked in society and are closely associated with deadly disease outbreaks. In large scale outbreaks, the mental health problems experienced are not limited to infected persons but also extend to involve frontline health workers and community members alike. While it is crucial to limit the spread of infections during an outbreak, previous experience suggests that mental and behavioural health interventions should be fully included in public health response strategies. Keywords: Mental Health, Psychosocial well-being, Covid-19, Pandemic, Outbreak, Health inequalitie [3]. Socioeconomic disparities resulting from job losses and other systemic barriers can exacerbate mental health issues among the general population amid COVID-19. In this regard, primary considerations should be given to how we guarantee service access and continuity for persons with existing mental health conditions whilst protecting the mental wellbeing of frontline workers, persons with COVID-19, and the general population [1]. [3]. Socioeconomic disparities resulting from job losses and other systemic barriers can exacerbate mental health issues among the general population amid COVID-19. In this regard, primary considerations should be given to how we guarantee service access and continuity for persons with existing mental health conditions whilst protecting the mental wellbeing of frontline workers, persons with COVID-19, and the general population [1]. Background As the novel SARS-CoV-2 coronavirus pandemic sweeps rapidly across the globe, it is undoubtedly hav- ing immense psychological impacts on communities. Some of these impacts can be linked to elevated levels of stress or anxiety [1]. There are legitimate concerns that an epidemic of mental illness could actually occur in the midst of the current COVID-19 pandemic and affect all generations and all majority and minority groups–albeit differently [2]. Measures such as quarantine and physi- cal distancing, while vital to stemming transmission of COVID-19, may contribute to a rise in depression, self- harm, harmful alcohol and drug use with further negative psychological consequences. Infectious disease outbreaks have also been associated with stigma and xenophobia In a recent survey carried out among 775 adults in the United States, an alarming 55% of them admitted that COVID-19 has had deleterious effects on their mental wellbeing and 71% were worried about the potential neg- ative impact of isolation on the mental health of Ameri- cans [1]. Substance abuse and mental health hotline services across the United States have registered unprec- edented increases in call volumes since February 2020 [4]. Although mental health problems account for about one- third of the world’s disability among adults, these issues tend to be under-addressed and overlooked in society [5]. A peculiarity of mental health management is that UK Full list of author information is available at the end of the article Otu et al. Int J Ment Health Syst (2020) 14:38 https://doi.org/10.1186/s13033-020-00371-w Otu et al. Int J Ment Health Syst (2020) 14:38 https://doi.org/10.1186/s13033-020-00371-w International Journal of Mental Health Systems *Correspondence: sanni.yaya@uOttawa.ca 6 The George Institute for Global Health, The University of Oxford, Oxford, UK Full list of author information is available at the end of the article These tensions are further magnified by fear of being quarantined or having to deal with inadequa- cies on the system such as a lack of PPE and ventilators. Some health workers have described feeling coerced and betrayed by their employers and the government in outbreak situations [10]. There are also the added anxi- eties involved with the assumption of new or unfamiliar clinical roles in outbreak situations. When interviewed, health workers involved in pandemics described feeling afraid, lonely, ostracized and unhappy about restric- tions on behaviours that enhanced coping, such as attending social events and physical contact [11].hi Although the actual scale of the mental health burden of COVID-19 is only beginning to emerge, the effects are likely to remain long after the pandemic has been dealt with. In many ways, the threat to people’s mental health appears to be invisible and ambiguous, one which requires urgent attention nonetheless [4]. The description of the mental health problems associated with COVID-19 as a “slow-motion disaster” is very apt as the psychologi- cal fallouts are likely to be widespread and persistent [4]. With lockdowns and physical distancing strategies being imposed in most nations, people have turned to differ- ent media sources to get news updates on the pandemic. Accessing these information sources for prolonged peri- ods of time is very likely to increase anxiety and potential for hysteria thus impacting one’s mental wellbeing. The term “vicarious trauma” was first described in 1996 [12] and was initially applied to situations where psychotherapists became affected by long-term con- tact with patients with mental diseases and manifested mental symptoms similar to psychological trauma [13]. The term has since been extended to include sympathy for survivors of a trauma resulting in physical symp- toms such as loss of appetite, irritability, fatigue and sleep disorders. Given the devastating nature of the current COVID-19 pandemic, it is very likely that vicar- ious trauma is occurring in healthcare settings dealing with the COVID-19 pandemic [14]. A study conducted among over 1200 health care workers providing care to COVID-19 patients in China reported higher levels of severe mental health symptoms in the frontline work- ers when compared to those in secondary roles [15]. The COVID-19 response has been characterized by an unprecedented global market failure in the supply of PPE to health workers, which has been linked to the unacceptably high infection rates among health work- ers [16]. © The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativeco mmons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Otu et al. Int J Ment Health Syst (2020) 14:38 Page 2 of 5 solutions need to be consistently applied to be successful and there is little room for “quick fixes” with these issues. Therefore, the institution of accessible and sustainable community-wide support systems would be the preferred approach to addressing mental health challenges in this pandemic. However, the global focus is currently on implementing measures of physical distancing whilst car- ing for unwell persons with COVID-19 in hospitals and ensuring that ventilators are available to support the sick in intensive care settings. Issues such as mental health of the general population, frontline health workers and per- sons with COVID-19 are being overlooked. were associated with a spectrum of mental health prob- lems among both the general population and health workers such as depression, anxiety, fear, post-trau- matic stress disorder and frustration [7–9]. Health workers are characteristically resilient and constitute the first line of defense in the face of dis- ease outbreaks. They carry out their life-saving duties at great personal risk and this sometimes entails isolat- ing themselves from family members to limit the risk of spreading a contagion. Underneath the façade of calm these health workers typically project lies a vul- nerability characterized by anxiety, fear of contracting an infectious disease and anticipation of impending doom [10]. There have also been reports of overwork by nurses caring for COVID-19 patients without adequate provision being made for their recuperation and mental wellbeing [16]. Patients infected with similar infections do not appear to be spared mental problems as depres- sion, suicidality, anxiety, panic attacks and delirium have been reported among such cohorts [7, 17]. Addressing Socio‑economic disparities in mental health: The need to protect underserved and vulnerable populations Existing evidence has clearly established a link between socio-economic characteristics (ethnicity, gender, social position, educational and wealth status and standard of living, etc.) and rates mental health service utilization [24]. There is a pressing need to address unequal access to healthcare as it is well known that social stratification produces all types of inequality around the world [25]. For instance, in Western countries, the most vulnerable people tend to be those without permanent legal status, such as asylum seekers, temporary foreign nationals [26], and other poor, marginalized citizens [27] with limited or no access to healthcare [28, 29]. During pandemics, these vulnerable populations and the frontline health workers with whom they interact face a considerable challenge to help these foreign nationals who may not master the local language or may not be familiar with how to access avail- able resources. Social research needs to be conducted to shed light on how socioeconomic disparities affect men- tal health amid the COVID-19 pandemic in comparison to past pandemics [27, 28]. p Specific legislations are needed, at national and local levels, to mitigate the mental health fallouts of COVID- 19 and strengthen existing community mental health programmes. There is a need for both horizontal and ver- tical integration of these policies. The horizontal aspect should involve the establishment of comprehensive men- tal health services while vertical integration should pro- mote the alignment of these services with relevant social, economic, and political policy domains [20]. Governance frameworks should be put in place to address mental health literacy and achieve stigma reduction [21]. Prior- itization of mental health and deliberate appropriation of financial resources to tackle mental health issues by policy makers are crucial measures. Access to publicly funded mental health therapy should be widely pro- moted. In this vein, the Australian government recently released a $1.1 billion package to boost mental health services to the Australian people battling the devastat- ing impacts of COVID-19 [22]. This represents a positive step in the right direction. Since the COVID-19 outbreak in the United States, emerging data have suggested that black Americans are disproportionately impacted by the virus [30]. This is mirrored by high mortality rates in states such as Louisi- ana, New Orleans, Chicago Detroit, where black Ameri- cans constitute a majority of the population [30]. Similar imbalances are also being seen in Midwestern cities such as Detroit and Chicago. The COVID‑19 pandemic: an opportunity to strengthening mental health systems and policies The COVID‑19 pandemic: an opportunity to strengthening mental health systems and policies health services at primary health care level during and after the COVID-19 pandemic. There may also be a need to build in a robust health management information sys- tem (HMIS) integrated with mental health indicators to facilitate the monitoring of mental healthcare within the communities [23]. Those with serious mental health ill- nesses are particularly vulnerable at this time and safe- guards should be put in place to protect them [21]. y In response to the threat of death from COVID-19, the world has experienced massive shifts and re-organiza- tion at a collective level with potential for even greater social transformation [18]. There are arguments that the COVID-19 pandemic might be leveraged to merge public health with mental health, foster togetherness and reduce prejudice and stigma [18]. In a bid to curb community transmission of COVID-19, many national governments have resorted to activating emergency measures, such as imposition of lockdowns, closure of schools, self-isolation, limiting the people in public places and at ceremonies, such as weddings and burials to ensure physical distancing. The family and friends of persons with COVID-19 who get admitted in hospitals are not being allowed to visit their sick loved ones out of fear of transmitting the virus. These strategies, which limit normal human interaction–combined with fear of the consequences of infection and social media misinfor- mation–increase the levels of chaos, stress and tensions within communities. The additional stressors such as loss of loved ones due to COVID-19, economic anxiety from job losses and financial instability all constitute clear and present dangers. Isolation and physical seclusion are reportedly associated with increase in issues such as child abuse, intimate-partner violence and suicide [7, 19]. When hard‑pressed frontline health workers are putting their patients’ needs before their own, it’s time to care for their mental health Frontline health workers are faced with the daunting task of caring for large numbers of very sick persons in the midst of uncertainty and upheaval. With every passing day, the personal protective equipment (PPE), laboratory testing and other resources needed to pro- tect healthcare workers from COVID-19 infection are becoming increasingly scarce. The anxiety of working in such pressured environments and lack of rest can indi- rectly increase the probability of acquiring COVID-19 from working in health facilities. The fear of contract- ing a contagion can negatively impact the willingness of health workers to provide service in outbreak situa- tions. In a survey carried out among 308 public health workers in Maryland USA, nearly half of them were not likely to report to duty during a pandemic due to the perceived risks of infection [6]. Such perceptions are likely to negatively impact a health system’s readiness to deal with pandemics such as COVID-19. Previous disease outbreaks such as severe acute respiratory syn- drome (SARS) in 2003 and Ebola disease in 2014–2016 Otu et al. Int J Ment Health Syst (2020) 14:38 Page 3 of 5 Addressing Socio‑economic disparities in mental health: The need to protect underserved and vulnerable populations Even in states such Illinois and Milwaukee where black people represent 14.6% and 26 percent of the population respectively, black people made up almost half and 42% percent of fatalities respectively [31]. Some of the reasons attributed to this disparity include the high prevalence of chronic diseases, which predispose to death with COVID-19 disease among back Americans, health inequity and racial disparity [31]. The human resources that provide mental health services should also be augmented and this could be achieved by task-sharing [23]. Specific short and medium-term training programmes targeting general health workers could be rolled out. This would bolster capacity to provide sustainable and integrated mental Research intersecting socio-economic inequalities and health can be used to understand why black Americans are disproportionately being affected by the COVID- 19 pandemic as opposed to other ethno-racial groups. Indeed, overwhelming evidence suggests that socioeco- nomic inequalities–meaning inequalities that relate to Otu et al. Int J Ment Health Syst (2020) 14:38 Page 4 of 5 role they place in saving lives and keeping our loved ones safe [33]. differences in income, social class, occupational back- ground, educational achievement, and neighbourhood deprivation can significantly impact people’s health out- comes [32]. In the United States, socioeconomic inequal- ities leave many poor black Americans with no choice but to use public transportation and take up roles such as cleaning in hospitals, for instance, which puts them at greater risk of acquiring COVID-19 [31]. Rigorous quan- titative and qualitative social research needs to be carried out in order to shine light on the multidimensional fac- tors that may account for this increased trend of COVID- 19-related mortality among black Americans. In this highly critical time, it is important that every- body does their part to slow the spread of COVID-19 by following governments’ public health measures of physi- cal distancing as well as the above-mentioned WHO recommendations. In doing so, social media and other communication technologies can be our allies. Social media, for example, can be a useful tool for information dissemination, which can be leveraged to share strate- gies for dealing with psychological stress in the society. General measures such as leading active lifestyles, eat- ing healthily and maintaining social links via a number of communication platforms should be encouraged. Addressing Socio‑economic disparities in mental health: The need to protect underserved and vulnerable populations The establishment of community coalitions and networks to support the elderly and infirm who may be unable to cope with the isolation measures should be given priority. Expanding mental and behavioural health measures in a time of pandemic matters more than everh The World Health Organization has recently put out guidance on mental health and psychosocial considera- tions during the COVID-19 pandemic, which specifi- cally targets healthcare workers, the general population, those in isolation and people with co-morbidities [33]. For health workers, education on both appropriate use of PPE and key principles underpinning the management of suspected/confirmed COVID-19 patients is advocated. Health workers should be trained to identify early signs of undue stress/burnout at work and seek help imme- diately. Other practical measures include incorporat- ing rotations from higher-to lower stress tasks and team “huddles” to ease tensions within the workplace. Regular clinical screening for anxiety, depression and other men- tal health issues might be useful for both health workers and patients infected with COVID-19. Multidisciplinary mental health and crisis teams will need to be constituted and made accessible by health workers, patients and community members. Provision of psychological coun- selling hotlines and online channels may help to boost access to such teams by the above groups. Conclusionh The COVID-19 pandemic is putting an enormous strain on the health care systems. It is not only a medical cri- sis. Given the apocalyptic speed with which COVID-19 is sweeping across the globe, the mental health care of patients, health professionals and communities is very likely to be under-addressed thus giving rise to major medium and long-term consequences. The establishment of mitigating strategies to preserve the mental wellbeing of all is a complicated, but crucial imperative. A proac- tive and multipronged approach comprising longer term strategies rather than short-term crisis responses is desir- able. Inaction in the face of this under-recognized threat is likely to have grave consequences for humankind. Acknowledgements None. Authors’ contributions All authors conceived, wrote and edited the commentary. All authors read and approved the final manuscript.All authors read and approved the final manuscript. As for the general population, the World Health Organization recommends that we treat people who are affected by COVID-19 with compassion and kindness; and that we separate a person from having an identity defined by COVID-10-in order to reduce stigma. The WHO advises that we curb our exposure to COVID- 19-related news that can cause anxiety and psychological distress. Instead, the WHO suggests that we get informed by trusted persons while focusing on facts rather than misinformation and rumours. In addition to protect- ing ourselves, we should also be supportive to others by amplifying positive and hopeful stories and positive images of local people who have experienced COVID-19. Finally, the WHO advises the general population to hon- our carers and healthcare workers by acknowledging the Funding CHC receives doctoral funding from the Pierre Elliott Trudeau Foundation and SSHRC-Vanier Canada Graduate Scholarships for his joint Ph.D. in sociology and geography. He also works as a Principal Investigator on his project “Black Refuge” funded by the France-Canada Research Fund. Availability of data and materials y Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Ethics approval and consent to participate Not applicable. Ethics approval and consent to participate Not applicable. References 7. Barbisch D, Koenig KL, Shih FY. Is there a case for quarantine? Perspectives from SARS to Ebola. Disaster Med Public Health Prep. 2015;9:547–53. 28. Wickramage K, Gostin LO, Friedman E, Prakongsai P, Suphanchaimat R, Huiet C, et al. Missing: where are the migrants in pandemic influenza preparedness plans? Health Hum. Rights. 2018;20(1):251. 8. Liu TB, Chen XY, Miao GD, Zhang L, Zhang Q, Cheung T, et al. Recom- mendations on diagnostic criteria and prevention of SARS-related mental disorders. J Clin Psychol Med. 2003;13:188–91. 29. Spiegel PB, Nankoe A. UNHCR, HIV/AIDS, and refugees: lessons learned. Forced Migr Rev. 2004;19:21–3. 9. Wei YL, Han B, Liu W, Liu G, Huang Y. Psychosomatic discomfort and related factors among 1,411 first-line SARS staff in Beijing. Manual of the 7th national experimental medicine symposium of Chinese Society of Integrated Traditional Chinese and Western Medicine. China; 2004: 6–12. 30. The Guardian. ‘It’s a racial justice issue’: Black Americans are dying in greater numbers from Covid-19. https://www.theguardian.com/world /2020/apr/08/its-a-racial-justice-issue-black-americans-are-dying-in-great er-numbers-from-covid-19. Accessed 10 Apr 2020. 10. Gold J. The COVID-19 crisis too few are talking about: health care workers’ mental health. Stat Reports. https://www.statnews.com/2020/04/03/ the-covid-19-crisis-too-few-are-talking-about-health-care-workers-menta l-health/. Accessed 10 Apr 2020. 31. Vox. COVID-19 is disproportionately taking black lives. https://www. vox.com/identities/2020/4/7/21211849/coronavirus-black-americans. Accessed 10 Apr 2020. 11. Diamond MB, Woskie L. COVID-19: Protecting frontline healthcare work- ers—what lessons can we learn from Ebola? The BMJ Opinion. https:// blogs.bmj.com/bmj/2020/03/25/healthcare-workforce-safety-and-ebola -in-the-context-of-covid-19/. Accessed 1 Apr 2020. 32. Centre for Ageing Research and Development in Ireland. 2014. Under- standing Socio-economic inequalities affecting older people. https:// www.cardi.ie/sites/default/files/publications/cardi-execsummary-inequ alities-FINAL-lores.pdf. Accessed 10 Apr 2020. 12. McMahon SA, Ho LS, Brown H, Miller L, Ansumana R, Kennedy CE. Health- care providers on the frontlines: a qualitative investigation of the social and emotional impact of delivering health services during Sierra Leone’s Ebola epidemic. Health Policy Plan. 2016;31(9):1232–9. 33. World Health Organization 2020. Mental health and psychosocial con- siderations during the COVID-19 outbreak. https://www.who.int/publi cations-detail/mental-health-and-psychosocial-considerations-durin g-the-covid-19-outbreak. Accessed 10 Apr 2020. 13. Close DH, Patel-Kerai G, Rumsey N, Naqvi H, White P. The psychosocial experiences of breast cancer amongst Black South Asian and White survivors do differences exist between ethnic groups. Psycho-Oncol. 2017;26(4):515–22. References 1. World Health Organization. Mental health and COVID-19. http://www. euro.who.int/en/health-topics/health-emergencies/coronavirus-covid -19/novel-coronavirus-2019-ncov-technical-guidance/coronavirus-disea se-covid-19-outbreak-technical-guidance-europe/mental-health-and- covid-19. Accessed 10 Apr 2020. 1. World Health Organization. Mental health and COVID-19. http://www. euro.who.int/en/health-topics/health-emergencies/coronavirus-covid -19/novel-coronavirus-2019-ncov-technical-guidance/coronavirus-disea se-covid-19-outbreak-technical-guidance-europe/mental-health-and- covid-19. Accessed 10 Apr 2020. 22. Prime Minister of Australia. $1.1 billion to support more mental health, medicare and domestic violence services. https://www.pm.gov.au/media /11-billion-support-more-mental-health-medicare-and-domestic-viole nce-services. Accessed 10 Apr 2020. 2. Miller BF. Mental illness is epidemic within the coronavirus pandemic. USA Today. https://eu.usatoday.com/story/opinion/2020/04/08/menta l-health-our-epidemic-within-coronavirus-pandemic-column/29395 11001/. Accessed 10 Apr 2020. 2. Miller BF. Mental illness is epidemic within the coronavirus pandemic. USA Today. https://eu.usatoday.com/story/opinion/2020/04/08/menta l-health-our-epidemic-within-coronavirus-pandemic-column/29395 11001/. Accessed 10 Apr 2020. 23. Semrau M, Alem A, Ayuso-Mateos JL, Chisholm D, Gureje O, Hanlon C, et al. Strengthening mental health systems in low- and middle-income countries: recommendations from the Emerald programme. BJPsych Open. 2019;5(5):e73. https://doi.org/10.1192/bjo.2018.90. 3. Moukaddam N, Shah A. Psychiatrists Beware! The impact of COVID-19 and pandemics on mental health. Psychiatric times. https://www.psych iatrictimes.com/psychiatrists-beware-impact-coronavirus-pandemics- mental-health. Accessed 10 April 2020. 3. Moukaddam N, Shah A. Psychiatrists Beware! The impact of COVID-19 and pandemics on mental health. Psychiatric times. https://www.psych iatrictimes.com/psychiatrists-beware-impact-coronavirus-pandemics- mental-health. Accessed 10 April 2020. 24. Amaddeo F, Jones J. What is the impact of socio-economic inequalities on the use of mental health services? Epidemiologia e Psichiatria Sociale. 2007;16:16–9.il 25. Kerbo HR. Social Stratification and inequality: class conflict in historical, comparative, and global perspective. 8th ed. McGraw-Hill Education 2011; Chapter 16. https://www.researchgate.net/publication/49309 282_Social_Stratification_and_Inequality. Accessed 10 Apr 2020. 4. Hopkins JS, Russell D. The mental health effects of Coronavirus are a “slow-motion disaster”. Mother Jones. https://www.motherjones.com/ coronavirus-updates/2020/04/the-mental-health-effects-of-coronaviru s-are-a-slow-motion-disaster/. Accessed 10 Apr 2020. 4. Hopkins JS, Russell D. The mental health effects of Coronavirus are a “slow-motion disaster”. Mother Jones. https://www.motherjones.com/ coronavirus-updates/2020/04/the-mental-health-effects-of-coronaviru s-are-a-slow-motion-disaster/. Accessed 10 Apr 2020. 26. Truman BI, Tinker T, Vaughan E, Kapella BK, Brenden M, Woznica CV, et al. Pandemic influenza preparedness and response among immigrants and refugees. Am J Public Health. 2009;99(S2):S278–86. 5. Lake J, Turner MS. Urgent need for improved mental health care and a more collaborative model of care. Perm J. 2017;21:17–024. 5. Lake J, Turner MS. Urgent need for improved mental healt more collaborative model of care. Perm J. 2017;21:17–024. 6. Balicer RD, Omer SB, Barnett DJ, Everly GS Jr. Local public health workers’ perceptions toward responding to an influenza pandemic. BMC Public Health. 2006;6:99. 27. Tolbert J. What issues will uninsured people face with testing and treat- ment for COVID-19? KFF. https://www.kff.org/uninsured/fact-sheet/what- issues-will-uninsured-people-face-with-testing-and-treatment-for-covid -19/. Accessed 10 Apr 2020. Received: 11 April 2020 Accepted: 16 May 2020 19. Urback R. COVID-19 could spark a mental-health tsunami. The Globe and Mail. https://www.theglobeandmail.com/opinion/article-covid-19-could -spark-a-mental-health-tsunami/ Accessed 10 Apr 2020. 20. Plagerson S. Integrating mental health and social development in theory and practice. Health Policy Plan. 2015;30(2):163–70. 21. Druss BG. Addressing the COVID-19 pandemic in populations with seri- ous mental illness. JAMA Psychiatry. 2020. https://doi.org/10.1001/jamap sychiatry.2020.0894. Competing interests h h d l h Page 5 of 5 Otu et al. Int J Ment Health Syst (2020) 14:38 Otu et al. Int J Ment Health Syst (2020) 14:38 Otu et al. Int J Ment Health Syst Author details 1 15. Lai J, Ma S, Wang Y, Cai Z, Hu J, Wei N, et al. Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019. JAMA Netw Open. 2020;3(3):e203976. 1 Department of Infection and Travel Medicine, Leeds Teaching Hospitals NHS Trust, Leeds, UK. 2 Department of Internal Medicine, College of Medical Sciences, University of Calabar, Calabar, Cross River State, Nigeria. 3 Depart- ment of Sociology, McMaster University, Hamilton, ON, Canada. 4 Geography at the Laboratoire Caribéen de Sciences Sociales du Centre National de la Recherche Scientifique, Université des Antilles, Schoelcher, Martinique, France. 5 School of International Development and Global Studies, University of Ottawa, Ottawa, ON, Canada. 6 The George Institute for Global Health, The University of Oxford, Oxford, UK. 16. UN News. COVID-19 highlights nurses’ vulnerability as backbone to health services worldwide. https://news.un.org/en/story/2020/04/10612 32. Accessed 10 Apr 2020. 17. Sinclair HA, Hamill C. Does vicarious traumatisation affect oncology nurses? A literature review. Eur J Oncol Nurs. 2007;11:348–56. 18. Ahmed A. Rapid Response: Covid-19 pandemic: a public and global mental health opportunity for social transformation? BMJ 2020. https:// www.bmj.com/content/368/bmj.m800/rr-13. Accessed 10 Apr 2020. Received: 11 April 2020 Accepted: 16 May 2020 Received: 11 April 2020 Accepted: 16 May 2020 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 14. Collins S, Long A. Working with the psychological effects of trauma: con- sequences for mental health-care workers–a literature review. J Psychiatr Ment Health Nurs. 2003;10:417–24. 14. Collins S, Long A. Working with the psychological effects of trauma: con- sequences for mental health-care workers–a literature review. J Psychiatr Ment Health Nurs. 2003;10:417–24.
https://openalex.org/W4312913006	https://www.academicoa.com/ILCPA.30.1.pdf	English	null	Synthesis and Characterization of some Thiazolidinone Derivatives Possessing Benzimidazole Nucleus	International letters of chemistry, physics and astronomy	2,014	cc-by	2,772	Online: 2014-03-12 Online: 2014-03-12 International Letters of Chemistry, Physics and Astronomy ISSN: 2299-3843, Vol. 30, pp 1-8 doi:10.56431/p-2061f9 CC BY 4.0. Published by Academic Open Access LTD, 2014 ABSTRACT Synthesis of new thiazolidinone derivatives possessing benzimidazole nucleus by condensation reaction of various substituted mercapto acids in presence of anhydrous ZnCl2. In our present study we have also synthesized Schiff bases of benzimidazole derivatives. All synthesized compounds were characterized by IR, 1H NMR and Mass spectroscopy. Keywords: 4-Thiazolidinone; Benzimidazole; Schiff-base Synthesis and characterization of some thiazolidinone derivatives possessing benzimidazole nucleus Prakash Mehta1, Prakash Dawedra2, Vaishali Goswami2, Hitendra S. Joshi2,* 1Home Science Department, Dr. Subhash Mahila Arts, Commerce & Home Science College, Junagadh - 362 001, Gujarat, India 2Department of Chemistry, Saurashtra University, Rajkot - 360 005, Gujarat, India *E-mail address: drhsjoshi49@gmail.com This paper is an open access paper published under the terms and conditions of the Creative Commons Attribution license (CC BY) (https://creativecommons.org/licenses/by/4.0) 1. INTRODUCTION Benzimidazole derivatives are of wide interest because of their diverse biological activity and clinical applications, they are remarkably effective compounds both with respect to their inhibitory activity and their favorable selectivity ratio [1-3]. The benzimidazole ring is an important pharmacophor in modern drug discovery. In recent years, attention has increasingly been given to the synthesis of benzimidazole derivatives. The synthesis of novel benzimidazole derivatives remains a main focus of medicinal research. Recent observations suggest that substituted benzimidazoles and heterocyclic, show easy interactions with the biopolymers, possess potential activity with lower toxicities in the chemotherapeutic approach in man [4]. There is a growing interest over the past years for the synthesis of benzimidazole-based heterocycles due to the crucial role of benzimidazole unit in the functions of biologically important molecules [5-7]. There are numerous biologically active molecules which contain various heteroatoms such as nitrogen, sulphur and oxygen, always drawn the attention of chemist over the years mainly because of their biological importance. Thiazolidinones are thiazolidine derivatives and have an atom of sulfur at position 1, an atom of nitrogen at position 3 and a carbonyl group at position 2, 4, or 5. However, its ILCPA Volume 30 2 derivatives belong to the most frequently studied moieties and its presence in penicillin was the first recognition of its occurrence in nature [8]. derivatives belong to the most frequently studied moieties and its presence in penicillin was the first recognition of its occurrence in nature [8]. Similarly 1,3-thiazolidin-4-ones are heterocyclic nucleus that have an atom of sulfur and nitrogen at position 1 and 3, respectively and a carbonyl group at position 4 have been subjected to extensive study in the recent years. The 4-thiazolidinone scaffold is very versatile and has featured in a number of clinically used drugs. They have found uses as antitubercular, antimicrobial, anti- inflammatory and as antiviral agents, especially as anti-HIV agents [9]. y g p y g [ ] The therapeutic importance of the compounds inspired us to synthesize some potential thiazolidinones containing benzimidazole as scaffold [10-14]. 2. EXPERIMENTAL All chemicals and solvents were purchased from Spectrochem Pvt Ltd., Mumbai of AR grade and were used without further purification. Melting points were taken in open capillary method and are uncorrected. IR spectra were recorded on FTIR-8400 spectrophotometer (Shimadzu, Kyoto, Japan), using DRS probe KBr pallet. 1H-NMR spectra of the synthesized compounds were recorded on a Bruker-Avance-II (400 MHz) DMSO-d6 solvent. Chemical shifts are expressed in δ ppm downfield from TMS as an internal standard. Mass spectra were determined using direct inlet probe on a GCMS- QP 2010 mass spectrometer (Shimadzu, Kyoto, Japan). 2. 1. Procedure for synthesis of 2-(1H-Benzimidazol-2-yl)-N-(4-hydrazinocarbonyl- phenyl) benzamide (1) Step 1: An equimolar amount of pthalic anhydride and o-phenylene diamine were taken in an RBF and the reaction mass was heated directly at 140-150 °C to obtained o-Benzoylene 2-1-benzimidazole [10]. Step 1: An equimolar amount of pthalic anhydride and o-phenylene diamine were taken in an RBF and the reaction mass was heated directly at 140-150 °C to obtained o-Benzoylene 2-1-benzimidazole [10]. Step 2: A mixture of o-benzoylene 2-1-benzimidazole (1 mmol) and benzocaine (1 mmol) were refluxed for 4-5 hours in DMF at 150 °C. Completion of reaction was monitored by TLC. The reaction mass was poured on ice-cold water and the crude product was filtered out and dried in vacuo. Crude product was crystallized from DMSO to obtained analytical grade pure 2-o-(4'-carbethoxyphenyl amino carbonyl phenyl) benzimidazole. Yield 85%. Step 3: In ethanolic solution of 2-o(4'-carbethoxyphenyl amino carbonyl phenyl)- benzimidazole (1 mmol), hydrazine hydrate (10 mmol) was added and refluxed overnight. The reaction mass was cooled and the obtained product was filtered. The product was washed with chilled ethanol to collect the analytical pure grade 2-(1H-Benzimidazol-2-yl)-N-(4-hydrazinocarbonyl-phenyl) benzamide. Yield 80 % Step 3: In ethanolic solution of 2-o(4'-carbethoxyphenyl amino carbonyl phenyl)- benzimidazole (1 mmol), hydrazine hydrate (10 mmol) was added and refluxed overnight. The reaction mass was cooled and the obtained product was filtered. The product was washed with chilled ethanol to collect the analytical pure grade 2-(1H-Benzimidazol-2-yl)-N-(4-hydrazinocarbonyl-phenyl) benzamide. Yield 80 % International Letters of Chemistry, Physics and Astronomy Vol. 30 3 NH2 NH2 O O O N N O H2N COOEt NH N O N H COOEt NH N O N H CONHNH2 NH2-NH2 Reflux Reflux 140-150 C Reflux 150 C (1) Step-1 Step- 2 Step- 3 Scheme I COOEt COOEt NH2-NH2 Reflux Step- 3 CONHNH2 (1) 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) A mixture of (1) (1 mmol) and substituted benzaldehyde (1 mmol) in presence of catalytic amount of acetic acid was refluxed with stirring until the reaction got complete (reaction progress was monitored by TLC) (Scheme II). The mixture was then cooled down and poured on crushed ice. The solid product was separated by filtration, dried and purified by crystallization. N NH NH O NH O N R RCHO (1) 2(a-j) N NH NH O NH O NH2 CH3COOH N NH NH O NH O N R RCHO (1) 2(a-j) N NH NH O NH O NH2 CH3COOH (1) 2. 3. General procedure for the synthesis of 2-aryl-o-benzimidazol-2'-yl-benzamido-p'- benzamido-5-H/Methyl/ Carboxymethyl-4-Thiazolidinones 2. 3. General procedure for the synthesis of 2-aryl-o-benzimidazol-2'-yl-benzamido-p'- benzamido-5-H/Methyl/ Carboxymethyl-4-Thiazolidinones A mixture of compound 2(a-j) (1 mmol) and substituted thioacids (1 mmol) was taken in a round bottom flask and anhydrous ZnCl2 was added to it. The reaction mixture was refluxed in oil bath (Scheme III). The progress of the reaction was checked by TLC. The reaction mass was poured on crushed ice and the product obtained was dissolved in sodium bicarbonate and reprecipitated from con. HCl. The product was dried and recrystallized from DMSO. ILCPA Volume 30 ILCPA Volume 30 4 4 N NH NH O NH O N R 2(a-j) N NH NH O NH O N S O X R SH-CHX-COOH X= H, 3(a-j) X=CH3, 4(a-j) X=CH2COOH, 5(a-j) R- sub. aldehyde Scheme III NH N NH N SH-CHX-COOH Table 1. Physical data table for 2-(1H-benzo[d]imidazol-2-yl)-N-(4-((4-oxo-2-substituted thiazolidin- 3-yl)carbamoyl)phenyl)benzamide: 3(a-j). Table 1. Physical data table for 2-(1H-benzo[d]imidazol-2-yl)-N-(4-((4-oxo-2-substituted thiazolidin- 3-yl)carbamoyl)phenyl)benzamide: 3(a-j). Sr. No. R = Substituted Aldehyde Molecular Formula M.P. °C Yeild (%) 3a Phenyl C30H23N5O3S 197 62 3b 4-Aminophenyl C30H24N6O3S 268 58 3c 2-Chlorophenyl C30H22N5O3ClS 246 49 3d 4-Chlorophenyl C30H22N5O3ClS 238 55 3e 3,4-dimethylphenyl C32H26N5O3S 279 48 3f 3,4-dimethoxy-5-nitrophenyl C32H26N6O7S 305 54 3g 2-Furyl C28H21N5O4S 215 55 3h 2-Hydroxyphenyl C30H23N5O4S 237 54 3i 4-Methoxyphenyl C31H25N5O4S 217 60 3j 4-N,N-dimethylaminophenyl C32H28N6O3S 278 64 Table 2. Physical data table for 2-(1H-benzo[d]imidazol-2-yl)-N-(4-((2-substituted-5-methyl-4- oxothiazolidin-3-yl)carbamoyl)phenyl)benzamide: 4(a-j). Sr. No. R = Substituted Aldehyde Molecular Formula M.P. °C Yeild (%) 4a Phenyl C31H25N5O3S 185 58 4b 4-Aminophenyl C31H26N6O3S 197 60 4c 2-Chlorophenyl C31H24N5O3ClS 245 59 able 2. Physical data table for 2-(1H-benzo[d]imidazol-2-yl)-N-(4-((2-substituted-5-methyl-4- oxothiazolidin-3-yl)carbamoyl)phenyl)benzamide: 4(a-j). International Letters of Chemistry, Physics and Astronomy Vol. 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) 30 International Letters of Chemistry, Physics and Astronomy Vol. 30 5 5 4d 4-Chlorophenyl C31H24N5O3ClS 203 63 4e 3,4-dimethylphenyl C33H29N5O3S 155 53 4f 3,4-dimethoxy-5-nitrophenyl C33H28N6O7S 167 61 4g 2-Furyl C29H23N5O4S 289 65 4h 2-Hydroxyphenyl C31H25N5O4S 217 60 4i 4-Methoxyphenyl C32H27N5O4S 186 59 4j 4-N,N-dimethylaminophenyl C33H30N6O3S 198 63 Table 3. Physical data table for 2-(1H-benzo[d]imidazol-2-yl)benzamido) benzamido)2-substituted-4- oxothiazolidin-5-yl)acetic acid: 5(a-j). Sr. No. R = Substituted Aldehyde Molecular Formula M.P. °C Yeild (%) 5a Phenyl C32H25N5O5S 191 42 5b 4-Aminophenyl C32H26N6O5S 269 48 5c 2-Chlorophenyl C32H24N5O5ClS 215 50 5d 4-Chlorophenyl C32H24N5O5ClS 245 43 5e 3,4-dimethylphenyl C34H29N5O5S 198 45 5f 3,4-dimethoxy-5-nitrophenyl C34H28N6O9S 167 49 5g 2-Furyl C30H23N5O6S 235 52 5h 2-Hydroxyphenyl C32H25N5O6S 249 57 5i 4-Methoxyphenyl C33H27N5O6S 287 43 5j 4-N,N-dimethylaminophenyl C34H30N6O5S 173 56 . Physical data table for 2-(1H-benzo[d]imidazol-2-yl)benzamido) benzamido)2-substituted-4- oxothiazolidin-5-yl)acetic acid: 5(a-j). 3. SPECTRAL ANALYSIS OF SOME 4-THIAZOLIDINONE DERIVATIVES Compound 3a : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H . SPECTRAL ANALYSIS OF SOME 4-THIAZOLIDINONE DERIVATIVES ILCPA Volume 30 6 def. ), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 8.91 (s, 1H), 8.06 (dd, 1H), 7.82 – 7.48 (m, 9H), 7.47 – 7.39 (m, 2H), 7.33 – 7.18 (m, 5H), 6.16 (d, 1H), 5.92 (s, 1H), 3.67 (d, 1H), M+ (m/z) = 533. def. ), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 8.91 (s, 1H), 8.06 (dd, 1H), 7.82 – 7.48 (m, 9H), 7.47 – 7.39 (m, 2H), 7.33 – 7.18 (m, 5H), 6.16 (d, 1H), 5.92 (s, 1H), 3.67 (d, 1H), M+ (m/z) = 533. Compound 3d : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 8.92 (s, 1H), 8.06 (dd, 1H), 7.83 – 7.71 (m, 4H), 7.71 – 7.39 (m, 9H), 7.24 (dd, 2H), 5.92 (s,1H), 5.68 (d, 1H), 3.67 (d, 2H), M+ (m/z ) = 567. 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) Compound 3i : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 8.91 (s, 1H), 8.06 (dd, 1H), 7.83 – 7.71 (m, 4H), 7.70 – 7.49 (m, 5H), 7.47 – 7.39 (m, 2H), 7.24 (dd, 2H), 6.97-6.89 (m, 2H), 5.92 (s, 1H), 5.66 (t, 1H), 3.80 (s, 3H), 3.71 – 3.57 (m, 2H), M+ (m/z) = 563. Compound 4a : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.86 (s, 1H), 9.03 (s, 1H), 8.06 (dd, 1H), 7.83 – 7.39 (m, 11H), 7.33 – 7.18 (m, 5H), 6.15 (t,1H), 5.92 (s,1H), 3.71 – 3.61 (m, 1H), 1.58 (d, 3H), M+ (m/z) = 547. Compound 4d : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.97 (s, 1H), 8.85 (s, 1H), 8.03 (dd, 1H), 7.89 – 7.71 (m, 4H), 7.71 – 7.53 (m, 3H), 7.48 – 7.39 (m, 6H), 7.24 (dd, 2H), 6.23 (s, 1H), 5.92 (s,1H), 3.71 – 3.63 (m, 1H), 1.60 (d, 3H), M+ (m/z) = 581. Compound 4i : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1720 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 9.09 (s, 1H), 8.06 (dd, 1H), 7.83 – 7.71 (m, 4H), 7.71 – 7.53 (m, 3H), 7.47 – 7.36 (m, 4H), 7.24 (dd, 2H), 6.95 – 6.86 (m, 2H), 6.15 (t, 1H), 5.92 (s,1H), 3.78 (s,3H), 3.66 (q, 1H), 1.59 (d, 3H), M+ (m/z) = 577. 2. 2. General process for synthesis of 2(Benzimidazol-2'-yl)-N-(p'-substituted benzal Hydrazino Carbonylphenyl): 2(a-j) Compound 5a : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.85 (s, 1H), 9.27 (s, 1H), 8.26 (s, 1H), 8.06 (dd, 1H), 7.82 – 7.39 (m, 11H), 7.33 – 7.18 (m, 5H), 6.20 (d, 1H), 5.92 (s,1H), 4.22 (t, 1H), 3.03 (dd, 1H), 2.75 (dd, 1H), M+ (m/z) = 591. Compound 5d : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.99 (s, 1H), 9.00 (s, 1H), 8.26 (s, 1H), 8.03 (dd, J = 7.5, 2.0 Hz, 1H), International Letters of Chemistry, Physics and Astronomy Vol. 30 7 7.85 – 7.53 (m, 7H), 7.47 – 7.38 (m, 4H), 7.24 (dd, 2H), 6.98 – 6.90 (m, 2H), 6.18 (t, 1H), 5.92 (s,1H), 4.25 (t, 1H), 3.79 (s, 3H), 3.02 (dd, 1H), 2.74 (dd, 1H), M+ (m/z) = 625. Compound 5i : IR (KBr) ν(cm-1): 3290 (N-H str.), 3040 (C-H str.), 2970 (C-H str.), 2873 (C-H str.), 1620 (C=C str.), 1455 (C-H bend.), 1375 (C-H bend.), 1310 (C-N str.), 1220 (C-H def.), 1140 (C-N-C str.), 1020 (C-O-C str.), 1060 (C=O str.), 810 (C-H def. bend.), 1H NMR (DMSO-d6) δ (ppm) 10.99 (s, 1H), 9.00 (s, 1H), 8.26 (s, 1H), 8.03 (dd, 1H), 7.85 – 7.53 (m, 7H), 7.47 – 7.38 (m, 4H), 7.24 (dd, 2H), 6.98 – 6.90 (m, 2H), 6.18 (t, 1H), 5.92 (s,1H), 4.25 (t, 1H), 3.79 (s, 3H), 3.02 (dd, 1H), 2.74 (dd, 1H), M+ (m/z) = 621. 4. CONCLUSION In present report, we synthesized new carbohydrazide derivative of benzimidazole from very cheap starting material. Synthesis of some new thiazolidinones derivatives 3(a-j), 4(a-j), 5(a-j) were carried out by reaction with substituted arylidene derivatives (2a-2j). All synthesized compounds were obtained in good to moderate yield. The synthesized compounds were characterized by 1H NMR, Mass and IR spectroscopy and the obtained results are showing good agreement with the synthesized structures. ACKNOWLEDGEMENT The author are thankful for facilities and grants given under UGC – SAP (DRS - Department Research Support) - 540/6/DRS/2004 (SAP-1) Dt. 26-03-2004 and Department of Science and Technology, New Delhi, for fund for improvement of Science and Technology (FIST-sanction no. SR/FIST/CSI/072/2003 date. 24-12-2003). [10]. Bistrzycki and A. Lecco, Helv. Chim. Acta 4 (1921) 427. ( Received 27 February 2014; accepted 04 March 2014 ) References [1]. K. Ansari, C. Lal, Eur. J. Med. Chem. 44 (2009) 4028-4033. [2]. S. Budow, M. Kozlowska, A. Gorska, Z. Kazimierczuk, H. Eickmeier, P. Colla, G. Gosselin, F. Seela, ARKIVOC iii (2009) 225-250. [3]. H. Grocer, C. Kus, D. Boykin, S. Yildiz, N. Altanlar, Bioorg. Med. Chem. 10 (2002) 2589-2596. [4]. D. Haugwitz, J. Med. Chem. 25 (1982) 969-974. [5]. D. Lednicer, John Wiley & Sons: NewYork, (1998). [6]. R. Silverman, The Organic Chemistry of Drug Design and Drug Action, 2nd Edn. Elsevier Academic Press: Oxford 2004. [7]. M. Demeunynck, Bailly, C. D. Wilson, D.N.A. and R.N.A. Binders, Wiley-VCH: Weinheim, 2002. [8]. P. Singh, S. Parmar, K. Raman, I. Stenberg, Chem. Rev. 81 (1981) 175. [9]. A. Jain, A. Vaidya, V. Ravichandran, S. Kashaw, R. Agrawal, Bioorganic & Medicinal Chemistry 20(11) (2012) 3378-3395 [10]. Bistrzycki and A. Lecco, Helv. Chim. Acta 4 (1921) 427. ILCPA Volume 30 8 [11]. G. Thirunarayanan, International Letters of Chemistry, Physics and Astronomy 5 (2014) 89-98. [12]. Nirav M. Shah, Hitendra S. Joshi, International Letters of Chemistry, Physics and Astronomy 6 (2014) 56-60. [13]. Mita D. Khunt, Vipul C. Kotadiya, Denish J. Viradiya, Bharat H. Baria, Umed C. Bhoya, International Letters of Chemistry, Physics and Astronomy 6 (2014) 61-68. [14]. Prakash Mehta, Prakash Davadra, Nirav Shah, Hitendra Joshi, International Letters of Chemistry, Physics and Astronomy 10 (2014) 74-80. ( Received 27 February 2014; accepted 04 March 2014 )
https://openalex.org/W2955990578	https://www.scielo.br/j/pab/a/ZGzJhKmL456MBkbxDtdGFPS/?lang=en&format=pdf	English	null	Feasibility of wood production according to productivity class and rotation criteria	Pesquisa Agropecuária Brasileira	2,019	cc-by	6,192	Forestry/ Original Article Forestry/ Original Article ISSN 1678-3921 Journal homepage: www.embrapa.br/pab For manuscript submission and journal contents, access: www.scielo.br/pab Feasibility of wood production according to productivity class and rotation criteria E-mail: castrorvo@yahoo.com.br, glauciana@ufsj.edu.br, anaflaviacastro@ufsj.edu.br , Campus Sete Lagoas, Rua Sétimo Moreira Martins, s/no, Itapoã II, CEP 35702-031 Sete Lagoas, MG, Brazil. E-mail: castrorvo@yahoo.com.br, glauciana@ufsj.edu.br, anaflaviacastro@ufsj.edu.br (2) Universidade Federal de Minas Gerais, Instituto de Ciências Agrárias, Avenida Universitária, no 1.000, Bairro Universitário, CEP 39404-547 Montes Claros, MG, Brazil. E-mail: araujocaj@gmail.com (3) Universidade Federal do Rio Grande do Norte, Unidade Especializada em Ciências Agrárias, Rodovia RN-160, Km 03, Caixa Postal 07, Distrito de Jundiaí, CEP 59280-000 Macaíba, RN, Brazil. E-mail: meire_caico@yahoo.com.br (4) Universidade Federal de Uberlândia, Campus Araras, Rodovia LMG 746, Km 01, s/no, CEP 38500-000 Monte Carmelo, MG, Brazil. E-mail: lidiomar.ef@gmail.com Corresponding author Received May 9, 2018 Accepted January 23, 2019 How to cite CASTRO, R.V.O.; ATAÍDE, G. da M.; CASTRO, A.F.N.M.; ARAÚJO JÚNIOR, C.A.; SANTOS, R.C. dos; COSTA, L.S. da. Feasibility of wood production according to productivity class and rotation criteria. Pesquisa Agropecuária Brasileira, v.54, e00778, 2019. DOI: https://doi. org/10.1590/S1678-3921.pab2019.v54.00778. Index terms: forest economics, forest planning, harvest age. (2) Universidade Federal de Minas Gerais, Instituto de Ciências Agrárias, Avenida Universitária, no 1.000, Bairro Universitário, CEP 39404-547 Montes Claros, MG, Brazil. E-mail: araujocaj@gmail.com (2) Universidade Federal de Minas Gerais, Instituto de Ciências Agrárias, Avenida Universitária, no 1.000, Bairro Universitário, CEP 39404-547 Montes Claros, MG, Brazil. E-mail: araujocaj@gmail.com (2) Universidade Federal de Minas Gerais, Instituto de Ciências Agrárias, Avenida Universitária, no 1.000, Bairro Universitário, CEP 39404-547 Montes Claros, MG, Brazil. E-mail: araujocaj@gmail.com (3) Universidade Federal do Rio Grande do Norte, Unidade Especializada em Ciências Agrárias, Rodovia RN-160, Km 03, Caixa Postal 07, Distrito de Jundiaí, CEP 59280-000 Macaíba, RN, Brazil. E-mail: meire_caico@yahoo.com.br (4) Universidade Federal de Uberlândia, Campus Araras, Rodovia LMG 746, Km 01, s/no, CEP 38500-000 Monte Carmelo, MG, Brazil. E-mail: lidiomar.ef@gmail.com Corresponding author Received May 9, 2018 Accepted January 23, 2019 How to cite CASTRO, R.V.O.; ATAÍDE, G. da M.; CASTRO, A.F.N.M.; ARAÚJO JÚNIOR, C.A.; SANTOS, R.C. dos; COSTA, L.S. da. Feasibility of wood production according to productivity class and rotation criteria. Pesquisa Agropecuária Brasileira, v.54, e00778, 2019. DOI: https://doi. org/10.1590/S1678-3921.pab2019.v54.00778. Renato Vinícius Oliveira Castro(1) , Glauciana da Mata Ataíde(1) , Ana Flávia Neves Mendes Castro(1) , Carlos Alberto Araújo Júnior(2) , Rosimeire Cavalcante dos Santos(3) and Lidiomar Soares da Costa(4 ) Feasibility of wood production according to productivity class and rotation criteria Abstract – The objective of this work was to evaluate the effect of forest rotation on the technical and economic feasibility of eucalyptus wood production for charcoal, in different productivity classes. Data came from Eucalyptus stands whose area was stratified into three classes of productivity: high, medium, and low. To each class, a different criterion of forest rotation was applied, as follows: age of maximum productivity, economic rotation age, single harvest age, and no technical parameter. Analyses of economic feasibility and production were performed for a 21-year planning horizon. For rotation without a technical parameter, a simulation by the Monte Carlo method was performed. For all rotation criteria, feasibility was observed for high- and medium-productivity classes, and unfeasibility for the low-productivity class. For rotation without a technical parameter, there is no chance for the project to be unfeasible. Wood production viability for charcoal under the studied conditions depends on the productivity class, for any rotation criterion adopted. Rotation criterion influences profitability, whose definition, according to productivity class, contributes to forest production maximization and to the economic return of the project, under the studied conditions. This is an open-access article Renato Vinícius Oliveira Castro(1) , Glauciana da Mata Ataíde(1) , Ana Flávia Neves Mendes Castro(1) , Carlos Alberto Araújo Júnior(2) , Rosimeire Cavalcante dos Santos(3) and Lidiomar Soares da Costa(4 ) (1) Universidade Federal de São João Del-Rei, Campus Sete Lagoas, Rua Sétimo Moreira Martins, s/no, Itapoã II, CEP 35702-031 Sete Lagoas, MG, Brazil. E-mail: castrorvo@yahoo.com.br, glauciana@ufsj.edu.br, anaflaviacastro@ufsj.edu.br (2) Universidade Federal de Minas Gerais, Instituto de Ciências Agrárias, Avenida Universitária, no 1.000, Bairro Universitário, CEP 39404-547 Montes Claros, MG, Brazil. E-mail: araujocaj@gmail.com (3) Universidade Federal do Rio Grande do Norte, Unidade Especializada em Ciências Agrárias, Rodovia RN-160, Km 03, Caixa Postal 07, Distrito de Jundiaí, CEP 59280-000 Macaíba, RN, Brazil. E-mail: meire_caico@yahoo.com.br (4) Universidade Federal de Uberlândia, Campus Araras, Rodovia LMG 746, Km 01, s/no, CEP 38500-000 Monte Carmelo, MG, Brazil. E-mail: lidiomar.ef@gmail.com Corresponding author Received May 9, 2018 Accepted January 23, 2019 How to cite CASTRO, R.V.O.; ATAÍDE, G. da M.; CASTRO, A.F.N.M.; ARAÚJO JÚNIOR, C.A.; SANTOS, R.C. dos; COSTA, L.S. da. Feasibility of wood production according to productivity class and rotation criteria. Pesquisa Agropecuária Brasileira, v.54, e00778, 2019. DOI: https://doi. org/10.1590/S1678-3921.pab2019.v54.00778. (1) Universidade Federal de São João Del-Rei, Campus Sete Lagoas, Rua Sétimo Moreira Martins, s/no, Itapoã II, CEP 35702-031 Sete Lagoas, MG, Brazil. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Introduction Data used in the present work refer to nonthinned stands of clones and hybrids of Eucalyptus spp., belonging to a forestry company located in the northwest of Minas Gerais state (17°36'S and 46°42'W), conducted for charcoal production in a high-forest system. The climate of the region, according to the Köppen-Geiger’s classification, is Aw – tropical climate with dry winters. The average annual temperature is 22.6°C, and annual average precipitation is 1.450 mm. In 2016, Brazil had a total of 7.84 million hectares of planted forests, out of which 5.80 million were Eucalyptus spp. (IBÁ, 2017). One of the biggest advantages of eucalyptus plantations is its high-growth rate, known to be the highest in the world among the hardwood forests (Stape et al., 2010; Myburg et al., 2014). This high-growth rate, combined with a correct planning, allows of the efficient supply of the demand for wood for different purposes, such as cellulose and paper, charcoal, energy, among others, since the settlements may have smaller rotation periods. The production area had 4,052 hectares, with stands established in 3.0 x 2.0 m spacing, and it was initially stratified in three productivity classes: high, medium, and low. The high-productivity class (HP) was in a 1,398 ha area, the medium one (MP) in a 1,230 ha area, and the low-productivity class (LP) in a 1,424 ha area. The volume growth curves in each productivity class were described by the logistic model (Figure 1). In the process of forest planning, one of the main factors is the correct definition of rotation age, as the productive structure depends on the wood flow, which is planned based on the age previously established for harvest (Nautiyal, 2011). The forest stands rotation can be influenced by several factors, which may be technical and economical. Among the technical factors the productive capacity of the site can be highlighted, and can be defined as the potential to produce wood or other type of product, under the existing environmental conditions and silvicultural techniques employed, in a given area, for a particular species or clone (Leite et al., 2011). Forest area classification as to its productivity potential is of great importance for the stratification of regions, which allows of the decision making on the type of forest management possible, according to the productive capacity of these areas (Bila et al., 2012). Introduction The forest rotation criteria were defined by the age of maximum productivity (AMP), that is, the forest rotations set by the ages of maximum productivity were 51 (HP), 56 (MP), and 61 (LP) months; the economic rotation age (ERA), in which forest rotation is defined by the economic harvest age – 60 months for HP, and 72 months for MP and LP classes; single harvest age (SHA), which is performed by cutting the forest at 84 months, for all productivity classes; and rotation with no technical parameter (NTP), which is the rotation without technical or economic criteria. The forest is harvest in the range of age between 51 to 102 months. The rotation set by AMP, in the different productivity classes, is obtained when the mean annual increment (MAI) is maximum and equal to the current annual increment (CAI) (Campos & Leite, 2017). The net present value of an infinite series is used for the definition of the ERA, considering an interest rate of 8%. The SHA alternative is established as an average age of 84 months, regardless of the productivity class. NTP was established based on the frequency of distribution of rotation ages, observed in the company over the last 10 years, in the conduction of the stands. This variation range is often a consequence of the company’s logistical factors, due to the need to cut newer or older forests than that in the established age because of the distance of the processing units at some time of the year, such as in the rainy season, or because of the availability of labor close to certain sites. The rotation set by AMP, in the different productivity classes, is obtained when the mean annual increment (MAI) is maximum and equal to the current annual increment (CAI) (Campos & Leite, 2017). The net present value of an infinite series is used for the definition of the ERA, considering an interest rate of 8%. The SHA alternative is established as an average age of 84 months, regardless of the productivity class. Many studies have been published involving the economic analyses (Cordeiro et al., 2014; Virgens et al., 2016; Timofeiczyk Junior et al., 2017), effects and classifications of productive capacity (Retslaff et al., 2015; Castro et al., 2016; Silva et al., 2018), and rotation of forest stands (Rodriguez et al., 1997; Resende et al., 2004). Viabilidade da produção de madeira em função da classe de produtividade e do critério de rotação Resumo – O objetivo deste trabalho foi avaliar o efeito da rotação florestal sobre a viabilidade técnica e econômica da produção de madeira de eucalipto para carvão, em diferentes classes de produtividade. Utilizaram-se dados de povoamentos de Eucalyptus cuja área foi estratificada em três classes de produtividade: alta, média e baixa. A cada classe, aplicaram-se diferentes critérios de rotação florestal, conforme: idade de máxima produtividade, idade econômica de corte, única idade de corte e rotação sem parâmetro técnico. Realizaram-se análises de viabilidade econômica e de produção para o horizonte de planejamento de 21 anos. Para a rotação sem parâmetro técnico, realizou-se uma simulação pelo método de Monte Carlo. Para todos os critérios de rotação, observou-se viabilidade para as classes de alta e média produtividade, e inviabilidade para a classe de baixa produtividade. Para a rotação sem parâmetro técnico, não há chance de o projeto ser inviável. A viabilidade da produção de madeira para carvão nas condições estudadas depende da capacidade produtiva do local, para qualquer critério de rotação adotado. O critério de rotação influencia a rentabilidade, cuja definição, em razão da capacidade produtiva do local, contribui para a maximização da produção e para o retorno econômico do projeto florestal nas condições estudadas. (4) Universidade Federal de Uberlândia, Campus Araras, Rodovia LMG 746, Km 01, s/no, CEP 38500-000 Monte Carmelo, MG, Brazil. E-mail: lidiomar.ef@gmail.com How to cite CASTRO, R.V.O.; ATAÍDE, G. da M.; CASTRO, A.F.N.M.; ARAÚJO JÚNIOR, C.A.; SANTOS, R.C. dos; COSTA, L.S. da. Feasibility of wood production according to productivity class and rotation criteria. Pesquisa Agropecuária Brasileira, v.54, e00778, 2019. DOI: https://doi. org/10.1590/S1678-3921.pab2019.v54.00778. Termos para indexação: economia florestal, planejamento florestal, idade de corte. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 This is an open-access article distributed under the Creative Commons Attribution 4.0 International License R.V.O. Castro et al. 2 Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Feasibility of wood production according to productivity 3 The AMP, ERA, and SHA criteria are deterministic in nature, which presuppose certainty on the values of NPV, IRR, and production obtained at established harvest ages, since these ages were fixed in each productivity class. The SPT criterion, however, is classified as probabilistic because it covers a wide range of ages. Thus, Monte Carlo simulation was performed, using the software @RISK (Ithaca, NY, USA), Operations costs considered in the accomplishment of the present work were provided by technical managers for the implantation and conduction of the stands (Table 1). The interest rate of 8% per year was considered, as well as the cost of harvesting and extraction at R$ 19.00 m-3, and the price of wood, cut, and stack at R$ 60.00 m-3. The land factor cost was considered by the method of interest on the value of the land, as, according to Silva et al. (2008), it provides more consistent results with market values. For the analysis of economic and technical feasibility of wood production a 21-year planning horizon was considered. The economic feasibility evaluation under the four alternatives of rotation forest was performed using two methods: net present value (NPV) and internal rate of return (IRR) (Silva & Fontes, 2005; Rezende & Oliveira, 2008). The IRR is considered an average rate of project growth, which is considered feasible if its value is greater than or equal to the rate of return on capital, usually known as the minimum acceptable rate of return (rate used in financial investments, such as savings etc.). In the presents work, the minimum acceptable rate of return (MARR) was considered as 8.0% (Silva et al., 2013). Table 1. Data of the implementation and maintenance costs, used in the economic viability analysis of wood production for charcoal from Eucalyptus spp. in the northwest of Minas Gerais. Introduction However, there are no studies evaluating the profitability and economic feasibility of eucalyptus forest projects for charcoal production, taking into account the variation of forest stands rotation, and economic feasibility. NTP was established based on the frequency of distribution of rotation ages, observed in the company over the last 10 years, in the conduction of the stands. This variation range is often a consequence of the company’s logistical factors, due to the need to cut newer or older forests than that in the established age because of the distance of the processing units at some time of the year, such as in the rainy season, or because of the availability of labor close to certain sites. The objective of this work was to evaluate the forest rotation effect on the technical and economic feasibility of eucalyptus wood production, in different classes of productivity, destined for charcoal making. Feasibility of wood production according to productivity Results and discussion In the analyzed harvest ages, the alternative showing the lower-economic viability for wood production was the one that adopted the harvest without a defined technical or economic parameter, while the technical viability was reduced when a single harvest age was adopted, in the three productivity classes (Table 2). Spathelf & Seling (2000) studied the economic effects of different thinning programs of Pinus elliottii stands, and observed that the IRR decreased as the rotation age increased. However, for the present work, this situation could not apply completely, and both IRR Rodriguez et al. (1997) argue that a harvest age defined regardless of economic issues – as the values of the products obtained with the forest harvest, the opportunity cost of the capital asset, and the time of land occupation in each forest rotation – may compromise the feasibility of a forestry project. However, Resende et al. (2004) suggest that choosing the best cycle cannot be made without considering technical issues, among which these authors include yields obtained, wood demand, and operational characteristics of the area. Table 2. Economic indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands, for different productivity classes and forest rotation criteria, in the northwest of the state of Minas Gerais, Brazil. Indicators Rotation criterion(1) Productivity class HP MP LP NPV (R$ ha-1) AMP 5,603.0 1,825.0 -1,853.0 ERA 8,165.0 2,648.0 -1,431.0 SHA 6,272.0 2,082.0 -1,484.0 NTP 5,584.0 1,449.0 -2,106.0 IRR (%) AMP 17.2 11.8 2.8 ERA 22.0 12.9 4.6 SHA 17.0 11.5 5.1 NTP 16.1 10.5 3.6 Wood production (m³ ha-1)(2) AMP 1,451.0 1,061.0 760.0 ERA 1,443.0 1,021.0 727.0 SHA 1,254.0 947.0 683.0 NTP 1,300.0 975.0 697.0 (1)AMP, age of maximum productivity; ERA, economic rotation age; SHA, single harvest age; and NTP, no technical parameter. (2)For the planning horizon. The economic criteria NPV and IRR indicated feasibility for high- and medium-productivity classes, and unfeasibility for the low-productivity class, for all harvest ages. Forestry projects implemented in areas classified as low-productivity would only be viable under interest rates lower than those of the IRR obtained for the used harvesting criteria, which usually does not happen in the forestry sector. An alternative to be considered for these areas is the diversification of wood products, and the uses that provide greater economic return. Feasibility of wood production according to productivity Year Activity/input Cost (R$ ha-1) 0 Ant control 110.00 0 Mechanical mowing in total area 50.00 0 Limestone application in total area 35.00 0 Herbicide application in total area 35.00 0 Harrowing on the subsoiling line 45.00 0 Subsoiling with phosphating 150.00 0 Planting 90.00 0 Replanting 30.00 0 Manual application of fertilizer (NPK) 80.00 0 Chemical control of intra-row weed 100.00 0 Seedlings transportation to the field 15.00 0 Intra-row sprout thinning with hoe 90.00 0 Inter-row mechanical mowing 45.00 0 Inter-row herbicide application 45.00 0 Intra-row herbicide application (costal sprayer) 100.00 0 Maintenance of seedling deposit 10.00 0 Top-dressing fertilization (90 days, manual) 80.00 0 Irrigation (1x) 120.00 0 NPK fertilizer 06-30-06 + micros 150.00 0 Reactive phosphate 301.0 0 Termiticide 35.00 0 Formicidal bait 65.00 0 Powdered formicide 3.00 0 Glyphosate herbicide 72.00 0 Pre-emergence herbicide 50.00 0 Soil conditioner (gel) 30.00 0 NPK fertilizer 20-00-20 + micros 216.00 0 Clonal seedlings 437.50 1 Herbicide + fertilizer + formicide + labor + firebreaks conservation 550.00 2 Herbicide + fertilizer + formicide + labor + firebreaks conservation 350.00 3 to 8 Formicide + labor + firebreaks conservation 100.00 1 to n Land 224.00 1 to n Administration 80.00 0 and n Licenses, rates 100.00 n Pre-harvest mowing 90.00 n: harvest age. In the technical feasibility analysis, the wood volume obtained from each rotation criterion, during the determined planning horizon (21 years), was calculated from the respective production equations for the productive capacity classes. Figure 1. Wood production curves of Eucalyptus spp. stands, in areas of high- (HP), medium- (MP), and low- productive capacity (LP), in the northwest region of the state of Minas Gerais, Brazil. V, volume (m³); I, age (months); r, correlation coefficient; and SEE, standard error of the estimate. Figure 1. Wood production curves of Eucalyptus spp. stands, in areas of high- (HP), medium- (MP), and low- productive capacity (LP), in the northwest region of the state of Minas Gerais, Brazil. V, volume (m³); I, age (months); r, correlation coefficient; and SEE, standard error of the estimate. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 4 R.V.O. Castro et al. to evaluate this scenario. Two thousand interactions were simulated, by alternating the rotation age in each productivity class according to the real frequency of harvest ages, which was the same for the three productivity classes. Feasibility of wood production according to productivity The stands stratification by their productive capacity is important for harvest forest planning (Oliveira et al., 2008), given that in sites with low-productive capacity, the tendency is that the growth rates of forest species are smaller. In other words, more time are required to reach the volumetric production potential as to the technical harvest age, which would result in low profitability of the project. The Monte Carlo method for the simulation was applied in the following sequence: the probability distribution of the harvest ages (rotation) was identified; the value of the harvest age was selected at random from its probability distribution; the value of the dependent variable of interest (NPV, IRR, and production) was calculated; the process was repeated until the complete frequency distribution of the dependent variables was obtained. For the high-productivity class, the NPV and IRR values were higher than R$ 5,500.00 ha-1 and 16.0%, respectively, among the four rotation age alternatives, showing the high viability of these projects. The ERA criterion showed itself more advantageous among the proposed alternatives. Therefore, considering a high-productivity site, the decision to cut the forest in different ages of EHA could result in a reduction of profitability between 23.2 and 31.6% by the NPV criterion, and between 21.8 and 26.8% by the IRR criterion. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Results and discussion The productive capacity of the site, by directly affecting the growth and production of forest stands, acts as an intensifier of economic results, with effects on both the optimal economic harvest age and the profitability. Feasibility of wood production according to productivity 5 and NPV were highly influenced by the productive capacity of the site, which in turn influenced the forest rotation age. that option with the highest-economic return. It can be observed that this alternative offers the biggest investment risk, showing lower IRR. However, the financial return obtained by the ERA was dependent on the flow of costs and revenues adopted by the company, which are variables for the various regions of the country, as well as interest rates adopted (Vitale & Miranda, 2010; Folmann et al., 2014; Chichorro et al., 2017). Among the criteria of rotation, ERA and AMP are those defined in more detail; it should be noted that AMP shows economic indicators very similar to those of the SHA and NTP criteria. In the high- and medium -productivity classes, all forest rotation options are feasible; however, considering the production of wood for own consumption, as in the cases of companies producing cellulose and charcoal, AMP rotation is the best option, as the production reaches rates from 8.1 to 13.6%, higher than the SHA and NTP criteria. This result also indicates that the definition of the criterion of forest rotation, consistent with the planning for the destination of the produced wood, contributes to the fact that the activity is financially and technically more advantageous. Forest planning, which encompasses, among other things, the implantation and age of forest rotation, is a process in which the planner needs to know about the many variables, constraints, and assumptions that range from forest formation to timber delivery, so that wood costs suit with reality. The recommendation of a single harvest age for species of the genus Eucalyptus disregards the rotation, the type of species growth, the use of different interest rates, among others (Rodriguez et al., 1997). Such generalization impairs the productivity gains of the forest, since the species and clones currently planted have high-physiological and nutritional efficiency, which results in higher rates of stand growth. Growth tends to stabilize earlier, with a consequent lower- technical harvesting age, which justifies the harvest anticipation (Ferreira et al., 2017; Gonçalves et al., 2017). Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Results and discussion The AMP criterion showed a global production of up to 11.9%, higher than that of the other rotation criteria, but still with a lower NPV and IRR than those of the ERA and SHA options. This was due to the rotation that was lower for the AMP alternative, in comparison to that used in other options. This results in higher costs of deployment throughout the analyzed period to allow of a continuous flow of wood, and such a situation corroborates the report of Guerra-Bugueño et al. (2014), who performed an economic evaluation of Eucalyptus globulus plantations, and observed that the cost of deployment could reduce profitability. The global economic and technical indicators, considering the entire forest, without stratification by productivity class, indicate feasibility for wood production under the analyzed conditions for all the alternatives of rotation age (Table 3). The lowest economic return was obtained by the NTP alternative, whose NPV was 47.9% lower than The practice of forest rotation without technical parameters led to the harvesting of forest stands at higher ages more frequently than those with maximum productivity and optimum economic situation (Figure 2). Therefore, for this alternative, a lower viability was observed. Table 3. Global indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands, for different productivity classes and forest rotation criteria, in the northwest of the state of Minas Gerais, Brazil. In the analysis of the minimum values obtained for the economic indicators, the chance that a global investment be economically unviable was observed as null, and NPV and IRR were observed as very close to the normal distribution, which means that about 50% probability shows a result below average, as well as 50% of it, above the average (Figure 3). Timofeiczyk Junior et al. (2017) observed a similar behavior for these economic indicators, in a risk assessment for Pinus taeda stands. These authors also mention that Rotation criterion(1) NPV (R$ ha-1) IRR (%) Wood production(2) (m³ ha-1)* AMP 1,836.0 10.5 1,090.0 ERA 3,118.0 13.1 1,063.0 SHA 2,274.0 11.2 960.0 NTP 1,626.0 10.0 989.0 (1)AMP, age of maximum productivity; ERA, economic rotation age; SHA, single harvest age; and NTP, no technical parameter. (2)For the planning horizon. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 R.V.O. Castro et al. Results and discussion 6 6 the use of the Monte Carlo method is more secure for projects and investment recommendations. or market oscillations. However, it can be affirmed that the establishment of the harvest on the basis of a concise and appropriate forest planning is essential for maximization of results, while rotations without pre- set parameters can affect the economic sustainability of the project. For the production obtained by the NTP alternative, the distribution of probability shows that volumetric increments smaller than 1000 m3 ha-1 have at least 75% chance of occurrence. This production is lower than the most advantageous criteria, ERA and AMP; in addition, the NTP has a longer rotation period. Rotation age of forests can be greatly influenced by the demand for larger trees, which justifies the adoption of higher- rotation ages (Castro et al., 2011; Weimann et al., 2017). But, when it comes to the wood production for charcoal, the tree individual volume is not considered. the most advantageous criteria, ERA and AMP; in addition, the NTP has a longer rotation period. Rotation age of forests can be greatly influenced by the demand for larger trees, which justifies the adoption of higher- rotation ages (Castro et al., 2011; Weimann et al., 2017). But, when it comes to the wood production for charcoal, the tree individual volume is not considered. The economic rotation is cited as the most convenient, since it considers interest rates, production costs, and tree dimensions (Silva et al., 2012). However, Nautiyal (2011) states that a rigid criterion should not be applied in the definition of harvest age, in order to preserve the interests of the company. In case of any changes in economic conditions, or the company’s objectives, the criteria used should follow such changes. The present work corroborates such statement because it shows that more than one option is feasible. For situations that prioritizes the maximum productivity, the production can be destined to self-supply; and for situations in which maximum profitability is sought, wood is produced to be marketed. Not always, will the optimal rotation age (AMP or ERA) be applied for all the management units due to restrictions imposed by the regulation model, Figure 2. Age frequency distribution at which Eucalyptus spp. stands were harvested under no technical parameter (NTP) in a 10-year period. Figure 3. References BILA, J.M.; SANQUETTA, C.R.; MACHADO, S. do A. Classificação de sítios com base em fatores edáficos para Pinus caribaea var. hondurensis na região de Prata, Minas Gerais. Floresta, v.42, p.465-474, 2012. DOI: https://doi.org/10.5380/ rf.v42i3.19014. MYBURG, A.A.; GRATTAPAGLIA, D.; TUSKAN, G.A.; HELLSTEN, U.; HAYES, R.D.; GRIMWOOD, J.; JENKINS, J.; LINDQUIST, E.; TICE, H.; BAUER, D.; GOODSTEIN, D.M.; DUBCHAK, I.; POLIAKOV, A.; MIZRACHI, E.; KULLAN, A.R.K.; HUSSEY, S.G.; PINARD, D.; VAN DER MERWE, K.; SINGH, P.; VAN JAARSVELD, I.; SILVA-JUNIOR, O.B.; TOGAWA, R.C.; PAPPAS, M.R.; FARIA, D.A.; SANSALONI, C.P.; PETROLI, C.D.; YANG, X.; RANJAN, P.; TSCHAPLINSKI, T.J.; YE, C.-Y.; LI, T.; STERCK, L.; VANNESTE, K.; MURAT, F.; SOLER, M.; SAN CLEMENTE, H.; SAIDI, N.; CASSAN- WANG, H.; DUNAND, C.; HEFER, C.A.; BORNBERG-BAUER, E.; KERSTING, A.R.; VINING, K.; AMARASINGHE, V.; RANIK, M.; NAITHANI, S.; ELSER, J.; BOYD, A.E.; LISTON, A.; SPATAFORA, J.W.; DHARMWARDHANA, P.; RAJA, R.; SULLIVAN, C.; ROMANEL, E.; ALVES-FERREIRA, M.; KÜLHEIM, C.; FOLEY, W.; CAROCHA, V.; PAIVA, J.; KUDRNA, D.; BROMMONSCHENKEL, S.H.; PASQUALI, G.; BYRNE, M.; RIGAULT, P.; TIBBITS, J.; SPOKEVICIUS, A.; JONES, R.C.; STEANE, D.A.; VAILLANCOURT, R.E.; POTTS, B.M.; JOUBERT, F.; BARRY, K.; PAPPAS JR, G.J.; STRAUSS, S.H.; JAISWAL, P.; GRIMA-PETTENATI, J.; SALSE, J.; VAN DE PEER, Y.; ROKHSAR, D.S.; SCHMUTZ, J. The genome of Eucalyptus grandis. Nature, v.510, p.356-362, 2014. DOI: https://doi.org/10.1038/nature13308. CAMPOS, J.C.C.; LEITE, H.G. Mensuração florestal: perguntas e respostas. 5.ed. atual. e ampl. Viçosa: UFV, 2017. 636p. CASTRO, R.V.O.; ARAÚJO, R.A.A.; LEITE, H.G.; CASTRO, A.F.N.M.; SILVA, A.; PEREIRA, R.S.; LEAL, F.A. Modelagem do crescimento e da produção de povoamentos de Eucalyptus em nível de distribuição diamétrica utilizando índice de local. Revista Árvore, v.40, p.107-116, 2016. DOI: https://doi.org/10.1590/0100- 67622016000100012. CASTRO, R.V.O.; LEITE, H.G.; NOGUEIRA, G.S.; SOARES, C.P.B.; ARAÚJO JÚNIOR, C.A.; CASTRO, A.F.N.M.; CRUZ, J.P. da; SANTOS, F.L. dos; SOUZA, C.C. Avaliação econômica de um povoamento de eucalipto desbastado e destinado a multiprodutos da madeira. Scientia Forestalis, v.39, p.351-357, 2011. CHICHORRO, J.F.; SILVA, A.L.P. da; ANDRADE, W.S. de P.; HEGEDUS, C.E.N.; KUBOYAMA, F.A.Q. Custos e índices econômicos de povoamentos de eucalipto do Programa Produtor Florestal no Espírito Santo. Pesquisa Florestal Brasileira, v.37, p.447-456, 2017. DOI: https://doi.org/10.4336/2017.pfb.37.92.1232. B.M.; JOUBERT, F.; BARRY, K.; PAPPAS JR, G.J.; STRAUSS, S.H.; JAISWAL, P.; GRIMA-PETTENATI, J.; SALSE, J.; VAN DE PEER, Y.; ROKHSAR, D.S.; SCHMUTZ, J. The genome of Eucalyptus grandis. Nature, v.510, p.356-362, 2014. DOI: https://doi.org/10.1038/nature13308. NAUTIYAL, J.C. Forest economics: principles and applications. Toronto: Natraj, 2011. 580p. CORDEIRO, S.A.; SILVA, M.L. Conclusions 1. The viability of wood production for charcoal, in the studied conditions, depends on the productivity class for any rotation criterion adopted. GUERRA-BUGUEÑO, E.; CÉLIS-MOSQUEIRA, F.; MORENO-GARCÍA, N. Efecto de la densidad de plantación en la rentabilidad de plantaciones de Eucalyptus globulus. Revista Chapingo Serie Ciencias Forestales y del Ambiente, v.20, p.21- 31, 2014. DOI: https://doi.org/10.5154/r.rchscfa.2012.08.051. 2. The rotation criterion influences profitability, and the definition of profitability according to the productivity class, contributes to the forest production maximization, and to the economic return of the project. 2. The rotation criterion influences profitability, and the definition of profitability according to the productivity class, contributes to the forest production maximization, and to the economic return of the project. IBÁ. Indústria Brasileira de Árvores. Relatório IBÁ 2017. Brasília, 2017. 80p. LEITE, H.; CASTRO, R.; SILVA, A.; JÚNIOR, C.; BINOTI, D.; CASTRO, A.F.; BINOTI, M. Classificação da capacidade produtiva de povoamentos de eucalipto utilizando diâmetro dominante. Silva Lusitana, v.19, p.181-195, 2011. Feasibility of wood production according to productivity Feasibility of wood production according to productivity 7 florestal de povoamentos de eucalipto utilizando a simulação de Monte Carlo. Ciência Florestal, v.27, p.1339-1347, 2017. DOI: https://doi.org/10.5902/1980509830215. Results and discussion Probability distribution of global economic indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands that were harvested under no technical parameter (NTP), in the northwest of the state of Minas Gerais, Brazil. Figure 3. Probability distribution of global economic indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands that were harvested under no technical parameter (NTP), in the northwest of the state of Minas Gerais, Brazil. The economic rotation is cited as the most convenient, since it considers interest rates, production costs, and tree dimensions (Silva et al., 2012). However, Nautiyal (2011) states that a rigid criterion should not be applied in the definition of harvest age, in order to preserve the interests of the company. In case of any changes in economic conditions, or the company’s objectives, the criteria used should follow such changes. The present work corroborates such statement because it shows that more than one option is feasible. For situations that prioritizes the maximum productivity, the production can be destined to self-supply; and for situations in which maximum profitability is sought, wood is produced to be marketed. Not always, will the optimal rotation age (AMP or ERA) be applied for all the management units due to restrictions imposed by the regulation model, Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 Figure 2. Age frequency distribution at which Eucalyptus spp. stands were harvested under no technical parameter (NTP) in a 10-year period. Figure 3. Probability distribution of global economic indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands that were harvested under no technical parameter (NTP), in the northwest of the state of Minas Gerais, Brazil. Figure 3. Probability distribution of global economic indicators, net present value (NPV), internal rate of return (IRR), and wood production of Eucalyptus spp. stands that were harvested under no technical parameter (NTP), in the northwest of the state of Minas Gerais, Brazil. Figure 2. Age frequency distribution at which Eucalyptus spp. stands were harvested under no technical parameter (NTP) in a 10-year period. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 References da; OLIVEIRA NETO, S.N. de; OLIVEIRA, T.M. de; NERY, K.C.M. da S. Análise de custos e rendimentos de sistemas agroﬂorestais na Zona da Mata (MG). Revista Agrogeoambiental, v.6, p.59-70, 2014. DOI: https://doi.org/10.18406/2316-1817v6n22014600. OLIVEIRA, M.L.R. de; LEITE, H.G.; NOGUEIRA, G.S.; GARCIA, S.L.R.; SOUZA, A.L. de. Classificação da capacidade produtiva de povoamentos não desbastados de clones de eucalipto. FERREIRA, D.H.A.A.; LELES, P.S. dos S.; OLIVEIRA NETO, S.N. de; PAULA, T.R.; COUTINHO, R.P.; SILVA, R.L. da. Crescimento e produção de eucalipto na região do Médio Paraíba do Sul, RJ. Floresta e Ambiente, v.24, e00131315, 2017. DOI: https://doi.org/10.1590/2179-8087.131315. Pesquisa Agropecuária Brasileira, v.43, p.1559-1567, 2008. DOI: https://doi.org/10.1590/S0100-204X2008001100015. RESENDE, R.R.; VALE, A.B. do; SOARES, T.S.; SILVA, M.L. da; COUTO, L.; VALE, R.S. do. Emprego de um modelo de crescimento e produção para determinação da rotação em povoamentos de eucalipto. Revista Árvore, v.28, p.219-225, 2004. DOI: https://doi.org/10.1590/S0100-67622004000200008. FOLMANN, W.T.; MIRANDA, G. de M.; DIAS, A.N.; MORO, F. de C.; FERNANDEZ, M.L.Q. Viabilidade de projetos florestais em três regimes de manejo na Mesorregião Centro- Oriental do Paraná. Floresta, v.44, p.153-160, 2014. DOI: https://doi.org/10.5380/rf.v44i1.30096. RETSLAFF, F.A. de S.; FIGUEIREDO FILHO, A.; DIAS, A.N.; BERNETT, L.G.; FIGURA, M.A. Curvas de sítio e relações hipsométricas para Eucalyptus grandis na região dos Campos Gerais, Paraná. Cerne, v.21, p.219-225, 2015. DOI: https://doi.or g/10.1590/01047760201521021349. GONÇALVES, J.C.; OLIVEIRA, A.D. de; CARVALHO, S. de P.C. e; GOMIDE, L.R. Análise econômica da rotação Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 R.V.O. Castro et al. 8 Pinus elliottii. Ciência Florestal, v.10, p.21-44, 2000. DOI: https://doi.org/10.5902/19805098393. Pinus elliottii. Ciência Florestal, v.10, p.21-44, 2000. DOI: https://doi.org/10.5902/19805098393. REZENDE, J.L.P.; OLIVEIRA, A.D. Análise econômica e social de projetos florestais. 2.ed. Viçosa: UFV, 2008. 386p. STAPE, J.L.; BINKLEY, D.; RYAN, M.G.; FONSECA, S.; LOOS, R.A.; TAKAHASHI, E.N.; SILVA, C.R.; SILVA, S.; HAKAMADA, R.E.; FERREIRA, J.M. de A.; LIMA, A.M.N.; GAVA, J.L.; LEITE, F.P.; ANDRADE, H.B.; ALVES, J.M.; SILVA, G.G.C.; AZEVEDO, M.R. The Brazil Eucalyptus Potential Productivity Project: influence of water, nutrients and stand uniformity on wood production. Forest Ecology and Management, v.259, p.1686-1694, 2010. DOI: https://doi.org/10.1016/j.foreco.2010.01.012. RODRIGUEZ, L.C.E.; BUENO, A.R.S.; RODRIGUES, F. Rotações de eucaliptos mais longas: análise volumétrica e econômica. Scientia Forestalis, n.51, p.15-28, 1997. SILVA, C.S.J. e; OLIVEIRA, A.D. de; REZENDE, J.L.P. de; MELLO, J.M. de; SCOLFORO, J.R.S. Economic feasibility and rotation age for stands of candeia (Eremanthus erythropappus). Cerne, v.18, p.695-706, 2012. DOI: https://doi.org/10.1590/S0104- 77602012000400020. SILVA, G.C.C.; CALEGARIO, N.; SILVA, A.A.L. da; CRUZ, J.P. da; LEITE, H.G. Pesq. agropec. bras., Brasília, v.54, e00778, 2019 DOI: 10.1590/S1678-3921.pab2019.v54.00778 References Site index curves in thinned and non-thinned eucalyptus stands. Forest Ecology and Management, v.408, p.36-44, 2018. DOI: https://doi.org/10.1016/j.foreco.2017.10.036. TIMOFEICZYK JUNIOR, R.; BOUCHARDET, D. de A.; FOLMANN, W.T.; HOEFLICH, V.A.; FERNANDEZ, M.L.Q. Simulação econômica de Monte Carlo aplicada à análise de risco florestal. Revista Espacios, v.38, p.5, 2017. SILVA, M.L. da; FONTES, A.A. Discussão sobre os critérios de avaliação econômica: Valor Presente Líquido (VPL), Valor Anual Equivalente (VAE) e Valor Esperado da Terra (VET). Revista Árvore, v.29, p.931-936, 2005. DOI: https://doi.org/10.1590/ S0100-67622005000600012. VIRGENS, A.P. das; FREITAS, L.C. de; LEITE, A.M.P. Análise econômica e de sensibilidade em um povoamento implantado no sudoeste da Bahia. Floresta e Ambiente, v.23, p.211-219, 2016. DOI: https://doi.org/10.1590/2179-8087.104914. SILVA, M.L. da; FONTES, A.A. Discussão sobre os critérios de avaliação econômica: Valor Presente Líquido (VPL), Valor Anual Equivalente (VAE) e Valor Esperado da Terra (VET). Revista Árvore, v.29, p.931-936, 2005. DOI: https://doi.org/10.1590/ S0100-67622005000600012. VITALE, V.; MIRANDA, G. de M. Análise comparativa da viabilidade econômica de plantios de Pinus taeda e Eucalyptus dunnii na região Centro-Sul do Paraná. Floresta, v.40, p.469-476, 2010. DOI: https://doi.org/10.5380/rf.v40i3.18908. SILVA, M.L.; JACOVINE, L.A.G.; VALVERDE, S.R. Economia florestal. Viçosa: UFV, 2013. 178p. SILVA, M.L.; RESENDE, J.L.P. de; LIMA JÚNIOR, V.B.; CORDEIRO, S.A.; COELHO JÚNIOR, L.M. Métodos do cálculo de custo da terra na atividade florestal. Cerne, v.14, p.75-81, 2008. WEIMANN, C.; FARIAS, J.A. de; DEPONTI, G. Viabilidade econômica do componente arbóreo de sistema agrossilvipastoril comparado ao de plantio ﬂorestal na pequena propriedade rural. SPATHELF, P.; SELING, I. Efeitos econômicos de diferentes programas de desbaste em povoamentos de Pesquisa Florestal Brasileira, v.37, p.429-436, 2017. DOI: https://doi.org/10.4336/2017.pfb.37.92.1147.
W3046185302.txt	https://link.springer.com/content/pdf/10.1007/s00506-020-00697-2.pdf	de		Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben	Österreichische Wasser- und Abfallwirtschaft	2,020	cc-by	4,604	Originalarbeit Österr Wasser- und Abfallw 2020 · 72:403–409 https://doi.org/10.1007/s00506-020-00697-2 Eine neue Methode für die Bestimmung des MikroplastikMassenanteils in Umweltproben Stefan Spacek · Ole Mallow · Therese Schwarzböck · Johann Fellner · Helmut Rechberger Online publiziert: 27. Juli 2020 © Der/die Autor(en) 2020 Zusammenfassung Die zunehmende Verschmutzung von aquatischen und terrestrischen Ökosystemen mit Mikroplastik stellt ein ernstzunehmendes Umweltproblem dar, wobei insbesondere die hohe Widerstandfähigkeit von Kunststoffen gegenüber Abbauprozessen zu einer stetigen Anreicherung von Mikroplastik in der Umwelt führt. Für die Bewertung der bereits vorhandenen Mikroplastikbelastung in der Umwelt sowie der Evaluierung von Haupteintragsquellen ist die Entwicklung geeigneter Methoden für die akkurate und einfache Bestimmung von Mikroplastik ein entscheidender Faktor. Die im gegenständlichen Artikel vorgestellte Analysenmethode stellt eine derartige Möglichkeit für die massenbezogene Analyse von Mikroplastik dar. Die an der TU Wien entwickelte EA-OEM (Elemental Analysis Overdetermined Equation Method) kann als thermoanalytische Methode eingestuft werden und nutzt die ausgeprägten Unterschiede in der Elementarzusammensetzung zwischen Kunststoffen, biogenen und anorganischen Materialien. Daraus wird der Gehalt an Mikroplastik in Umweltproben quantitativ bestimmt. Die EA-OEM wurde in den letzten Jahren anhand verschiedener Probentypen getestet und durch zusätzliche Wiederfindungsversuche mit gespikten Matrixproben und der Teilnahme an einem Ringversuch bereits einer eingehenden Validierung unterzogen. Die gegenständliche Arbeit fasst einige dieser Untersuchungen zusammen. Diese ResulDI S. Spacek, Bakk.techn. () · Dr. O. Mallow, MSc · Dr. T. Schwarzböck, MSc · Assoc. Prof. DI Dr. J. Fellner · Univ.-Prof. DI Dr. H. Rechberger Institut für Wassergüte und Ressourcenmanagement, Forschungsbereich Abfallwirtschaft und Ressourcenmanagement, Technische Universität Wien, Karlsplatz 13, 1040 Wien, Österreich stefan.spacek@tuwien.ac.at tate untermauern einerseits die hohe Wiederfindungsrate von Kunststoffen bei gespikten Mikroplastikproben (im Bereich von 93 bis 110 %) und andererseits eine ungefähre Nachweisgrenze der Methode, die im Bereich von 1 Massen-% Mikroplastikgehalt liegt. Die EAOEM positioniert sich als verhältnismäßig einfaches und kostengünstiges Messverfahren, um Mikroplastik in Umweltproben zu bestimmen. tection was estimated with 1 wt% microplastics. The EA-OEM poses a comparatively simple and cost-effective method to determine microplastics in environmental samples. Schlüsselwörter Mikroplastik · EAOEM · Thermoanalytische Verfahren · Monitoring Die Verschmutzung von aquatischen und terrestrischen Ökosystemen durch den kontinuierlichen Eintrag von Kunststoffabfällen sowie Mikroplastik (MP) stellt unsere Gesellschaft aktuell vor große Herausforderungen (Derraik 2002; Law und Thompson 2014). Der möglichst akkuraten Quantifizierung von Mikroplastik-Eintragsquellen sowie der Bestimmung von bereits bestehenden Hintergrundbelastungen in der Umwelt wird dabei eine wichtige Rolle zuteil. Dies kann nur durch fortlaufende Bestrebungen in Richtung Entwicklung und Optimierung von geeigneten Analysenmethoden sowie der Standardisierung und Normung dieser Methoden erreicht werden. Methodisch betrachtet, kann der Großteil der sich derzeit in Entwicklung befindlichen Analysenmethoden in zwei Hauptgruppen eingeteilt werden (La Nasa et al. 2020): i) Mikroskopische & spektroskopische Methoden und ii) Thermoanalytische Methoden. Innerhalb der ersten Hauptgruppe können MP-Analysen vielfach bereits mit einfachen Lichtmikroskopen durchgeführt werden. Eine eindeutige optische Differenzierung zwischen Partikeln biogenen Ursprungs und Plastiktfragmenten ist in diesem Fall jedoch nicht immer gewährleistet. Aus diesem Grund hat sich in den letzten Jahren die Kopplung von Mikroskopiertechniken mit spektroskopischen Detektionsmethoden etabliert (Prata et al. 2019). Mithilfe von Infrarotspektroskopie (FTIR) oder Raman-unterstützten mikroskopischen Untersuchungen, kann das FTIRoder Raman-Spektrum jedes einzel- A new method for determining microplastic content in environmental samples Abstract The increasing pollution of aquatic and terrestrial ecosystems with microplastics is a severe problem for the environment. Especially the high resistance of plastics to degradation processes causes a constant accumulation of microplastics in the environment. The development of suitable methods for the accurate and simple determination of microplastics is a vital factor to evaluate existing microplastic accumulations in the environment. The method presented here describes a thermoanalytical method for massbased microplastic determination. The EA-OEM (Elemental Analysis Overdetermined Equation Method) was developed at TU Wien and makes use of the distinct differences in the elemental composition of plastic, biogenic and inorganic materials to quantitatively measure the content of microplastics in environmental samples. The EA-OEM has been optimized in past years using polymer spiked samples and additionally evaluated in an inter-laboratory round robin test. The aim of this paper is to summarize recent investigations. The results demonstrate good recovery rates for microplastics in polymer spiked samples (in the range of 93 to 110%). Furthermore, the limit of de- Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Keywords Microplastics · EA-OEM · Thermoanalytical Methods · Monitoring 1 Einleitung 403 Originalarbeit nen Partikels aufgenommen werden und dadurch die Materialzusammensetzung in den meisten Fällen eindeutig bestimmt werden. Um die Präzision der Zuordnung von Partikeln noch weiter zu erhöhen, gehen aktuelle methodische Weiterentwicklungen dieses Ansatzes in Richtung Kombination von FTIR- und Raman-Spektren (Brandt et al. 2020). Bildgebende Analysen liefern in erster Linie Informationen zu Partikelanzahl, Größenverteilung, vorhandenen Partikelformen sowie Polymertypen. Diese Parameter sind wichtig für die Beantwortung ökotoxikologischer Fragestellungen. Die Partikelanzahl kann zudem als Indikator für die Grundbelastung von MP in der Umwelt herangezogen werden. Eine umfassende und vergleichbare Charakterisierung der MP-Belastung ist jedoch nur mit der Bestimmung von massebezogenen Parametern möglich, da nur damit der mengenmäßige Eintrag von MP in die Umwelt beschrieben werden kann. Die Umrechnung von der Partikelanzahl und Partikelgröße auf die Gesamtmasse ist zwar möglich, jedoch sind diese Berechnungen häufig mit sehr hohen Unsicherheiten verbunden. Um zukünftige Analysenergebnisse zu MPBelastungen eindeutig vergleichbar zu machen, bedarf es daher zusätzlicher Methoden, welche den Massenanteil von MP in Umweltproben analysieren können. Der Großteil der zurzeit in der Literatur beschriebenen massebezogenen Methoden gehört der zweiten Hauptgruppe an. Diese wird derzeit am besten unter dem Begriff thermoanalytische Methoden zusammengefasst. Thermoanalytische Methoden basieren auf der Verbrennung oder Pyrolyse des Probenmaterials und anschließender Auftrennung, Detektion sowie Quantifizierung der dabei entstehenden Gase (La Nasa et al. 2020). Diese Gase, beziehungsweise deren Verhältnisse zueinander, sind jeweils stoffspezifisch und ermöglichen dadurch einen Rückschluss auf die ursprüngliche Zusammensetzung der Feststoffprobe. Dies ermöglicht die Bestimmung von Kunststoffanteilen und zumeist auch die Detektion der Kunststofftypen. Im Gegensatz zu mikroskopischen Untersuchungen besteht bei thermoanalytischen Methoden keine Einschränkung bezüglich der Größe der erfassten Partikel, da durch die vollständige Verbrennung oder Pyrolyse der Gesamtkunststoffgehalt einer Probe erfasst wird. Als 404 Hauptvertreter, bezogen auf die Veröffentlichungen in den vergangenen Jahren, sind die Methoden TED-GCMS (Thermoextraktion DesorptionGaschromatographie-Massenspektrometrie, Dümichen et al. 2017) und Pyrolyse-GC-MS (Pyrolyse-Gaschromatographie-Massenspektrometrie, Fischer und Scholz-Bottcher 2017) zu nennen. Eine weitere Methode, um MP massebezogen zu bestimmen, wurde jüngst entwickelt und ist Gegenstand dieses Artikels. Die sogenannte Elemental Analysis combined with Overdetermined Equation Method (EA-OEM) basiert in erster Linie auf der Analyse der makro-elementaren Zusammensetzung einer Probe (CHNSO-Bestimmung). Die EA-OEM wird daher thematisch zu den thermoanalytischen Methoden gezählt. Sie erlaubt eine kostengünstige sowie im Routinebetrieb einfach durchzuführende Bestimmung des massebezogenen Summenparameters „Gesamtkunststoffanteil“ für die MP-Belastung in Umweltproben. Details zu der Methodik und erste Ergebnisse wurden von Mallow et al. (2020) veröffentlicht. Ziel des gegenständlichen Artikels ist es, weiterführende Ergebnisse zu Untersuchungen und Weiterentwicklung der EA-OEM darzulegen und damit die Anwendbarkeit und auch Einschränkungen der Methodik zu diskutieren. Insbesondere werden Untersuchungen zu Messbereich, Einfluss von Matrixeffekten und der Nachweisgrenze beschrieben. 2 EA-OEM und die adaptierte Bilanzenmethode Das Messprinzip der EA-OEM für die Bestimmung von MP basiert grundsätzlich auf der adaptierten Bilanzenmethode (aBM), eine an der TU Wien am Institut für Wassergüte und Ressourcenmanagement entwickelte Methode für die Bestimmung des Kunststoffanteils in Ersatzbrennstoffen (Fellner et al. 2011; Schwarzböck et al. 2018). Die Methode wurde in den vergangenen Jahren konsequent weiterentwickelt und auf die Quantifizierung von Mikroplastik in Umweltproben angepasst. Die Bestimmung des unbekannten MP-Anteils erfolgt dabei durch die Unterscheidung von Kunststoffen und biogenen Materialien anhand deren unterschiedlicher makro-elementarer Zusammensetzung (Gehalt an Kohlenstoff C, Wasserstoff H, Stickstoff N, Schwefel S, Sauerstoff O). So bestehen biogene Materialen zu- meist zu 40 bis 50 % aus Sauerstoff. Kunststoffe hingegen enthalten entweder keinen Sauerstoff (PE, PP, PVC, PS) oder einen geringeren Anteil (PUR, PET). Der Kohlenstoffgehalt von Kunststoffen ist deutlich höher als von biogenen Materialien (PE/PP: 86 %, PS: 92 %, PET: 65 %). Als Ausnahme muss an dieser Stelle die Stoffklasse der Fette genannt werden (Abb. 1). Diese weist eine CHNSO-Zusammensetzung auf, welche näher an der Gruppe der Kunststoffe liegt. Da Fette in durchaus höheren Mengen in Umweltproben vorhanden sein können, müssen diese im Zuge der Probenaufbereitung für die EAOEM durch Extraktion oder Verseifung der Probe vollständig entfernt werden, damit sie nicht fälschlicherweise als Kunststoffe detektiert werden. 2.1 Instrumentelle Elementaranalyse (CHNSO-Bestimmung) Für die Anwendung der EA-OEM ist die Bestimmung der makro-elementaren Zusammensetzung (CHNSO-Zusammensetzung) in der Probe nötig. Durch die unterschiedliche CHNSO-Zusammensetzung von Kunststoffen und biogenen Materialien ermöglicht diese Bestimmung eine quantitative Aussage über den MP-Anteil in Umweltproben. Die Messung der Elemente C, H, N, und S basiert im Grunde auf der vollständigen Oxidierung der zu analysierenden Probe bei hohen Verbrennungstemperaturen von über 1000 °C. Die entstehenden Verbrennungsgase werden durch spezifische Adsorptionssowie Desorptionsvorgänge voneinander separiert. Durch die anschließende Messung der Gaskonzentrationen kann auf die ursprüngliche CHNS-Zusammensetzung rückgeschlossen werden. Dieses Messprinzip ist automatisiert und im Routinebetrieb einsetzbar. Die Bestimmung von Sauerstoff verläuft ähnlich, jedoch wird die Probe unter Sauerstoffausschluss pyrolysiert anstatt oxidiert. 2.2 Berechnung des MP Anteils Die aus der Elementaranalyse erhaltenen CHNSO-Messwerte (T x ) beziehen sich auf die trockene Gesamtprobe inklusive des Aschegehalts ( A ). Um die für die Methode benötigten wasserund aschefreien TOx (x = CHNSO)-Werte zu erhalten, müssen anschließend noch der Aschegehalt und die dazugehörige CHNSO-Zusammensetzung der Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Originalarbeit 2.3 CHNSO-Zusammensetzung der Reinstofffraktionen Abb. 1 Massenanteile an C, H, und O (auf wasser- und aschefreier Basis, angegeben als TOC, TOH und TOO) in Materialien, die in Umweltproben häufig vorkommen (Algen, Holz, Lignin, Cellulose, Fett) sowie eine Auswahl an Hauptkunststoffen, welche häufig bei Mikroplastikuntersuchungen detektiert werden. Die Werte für Algen, Fett und Holz basieren auf eigenen Messungen. Die Werte für die Polymertypen wurden anhand ihrer chemischen Summenformel berechnet. Die Angaben für Lignin beziehen sich auf extrahiertes Lignin aus einer Mischung von verschiedenen Harthölzern und stammen aus Jahan et al. (2012) Asche (T I x ) bestimmt werden. Anhand Gl. 1 können im Anschluss daran die wasser- und aschefreien T Ox -Gehalte berechnet werden. T Ox = (T x − T I x · A)/(1 − A) (1) Die T Ox -Werte der Probe (TOxProbe ) repräsentieren die chemische Zusammensetzung der in der Probe vorhandenen, vermischten biogenen Fraktion und Kunststofffraktion. Diese Zusammensetzung muss der T Ox -Zusammensetzung der beiden Reinstofffraktionen biogen (T Ox B) und Kunststoff (T Ox K) multipliziert mit den dazugehörigen Massenanteilen der biogenen Fraktion (x B ) und der Kunststofffraktion (x K ) entsprechen (Gl. 2). Die Massenanteile (x B und x K ) in dieser Bilanz sind je- weils die zwei unbekannten Variablen, die es zu bestimmen gilt. Diese können mithilfe einer nichtlinearen Ausgleichsrechnung bestimmt werden. Der dafür verwendete Algorithmus wird in Fellner et al. (2011) im Detail beschrieben. Das Ergebnis der Ausgleichsrechnung liefert den statistisch abgesicherten Mittelwert der Massenanteile und die dazugehörigen Unsicherheiten. TOCProbe = x B ∗TOCB + x K ∗TOCK TOHProbe = x B ∗TOHB + x K ∗TOHK TOOProbe = x B ∗TOOB + x K ∗TOOK (2) TONProbe = x B ∗TONB + x K ∗TONK TOSProbe = x B ∗TOSB + x K ∗TOSK Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Die von den AutorInnen in den letzten Jahren gesammelten Erfahrungen zeigen, dass insbesondere die Bestimmung der CHNSO-Zusammensetzung der biogenen Reinfraktion sowie der Kunststoff-Reinfraktion (T Ox B und T Ox K) herausfordernd sein kann. Da diese Werte wichtige Parameter für das Analysemodell sind, sollten sie jedoch möglichst genau bestimmt werden. Anders als bei der Anwendung der aBM für Ersatzbrennstoffe, kann die CHNSO-Zusammensetzung der beiden Reinfraktionen nicht durch händisches Sortieren eindeutig abgeleitet werden. Für eine schnelle Erstabschätzung der Mikroplastikbelastung in Umweltproben ist die Elementaranalyse daher, aus derzeitiger Sicht, nur eingeschränkt geeignet. In den vergangenen Jahren durchgeführte Untersuchungen zeigen jedoch, dass die MP-Polymer-Zusammensetzung in den unterschiedlichen Umweltmatrizen häufig sehr ähnlich ist und zumeist von einem hohen Anteil an PE dominiert wird. Dadurch können für Erstabschätzungen des Mikroplastikgehalts Näherungswerte zur chemischen Zusammensetzung der vorhandenen Kunststoffe herangezogen werden. Abseits davon besteht für die Entwicklung des Analysemodells auch die Möglichkeit, dass für eine Erstbestimmung alternative thermoanalytische Methoden wie zum Beispiel TED-GC-MS oder mikroskopische Untersuchungen zum Einsatz kommen. 2.4 Kurzbeschreibung der EA-OEM Die EA-OEM kann anhand der folgenden Punkte zusammengefasst werden: • Unterteilung der Gesamtprobe in die 3 Fraktionen biogener Anteil, Kunststoffanteil sowie Mineralik- und Inertanteil, wobei die biogene Fraktion und die Kunststofffraktion eine signifikant unterschiedliche makroelementare Zusammensetzung aufweisen müssen. • Bestimmung der makro-elementaren Zusammensetzung (CHNSO-Bestimmungen) der wasserfreien Gesamtprobe (T x ). • Bestimmung des Aschegehalts aus der Gesamtprobe ( A ) und anschließende Messung der dazugehörigen CHNSO-Zusammensetzung der Asche (T I x ). 405 Originalarbeit Abb. 2 Zusammenfassung der Ergebnisse des Wiederfindungsversuchs 1. Grafik a Gegenüberstellung der theoretischen PE/PPAnteile in matrixgespikten Proben und der anschließenden Wiederfindung (Mallow et al. 2020). Grafik b Veränderung der Verteilung der TOx WerteinAbhängigkeitvonderdazugespiktenPE/PP-Menge.DadieSummedergemessenen TOx -WerteaufgrundvonMessunsicherheiten nicht genau 100 % ergibt, wurden die Ergebnisse für die Abbildung auf 100 % normiert • Berechnung der asche- und wasserfreien TOx -Werte (x = CHNSO) anhand Gl. 1. • Einmalige Bestimmung der probenspezifischen T Ox -Werte der biogenen Reinstofffraktion und der Kunststoffreinfraktion. • Erstellung eines Gleichungssystems, bei dem die bestimmten T Ox -Werte einer Probe den T Ox -Werten der Reinfraktionen, multipliziert mit den jeweiligen Massenanteilen, gegenübergestellt werden (Gl. 2). • Berechnung der beiden unbekannten Massenanteile von biogener Fraktion und Kunststofffraktion mithilfe einer nichtlinearen Ausgleichsrechnung. 3 Ergebnisse 3.1 Wiederfindungsversuch 1 Aufgrund des derzeitigen Mangels an für die MP-Analytik geeigneten Referenzmaterialien, wurde die Analysenperformance der EA-OEM anhand von Wiederfindungsversuchen überprüft. Für den Wiederfindungsversuch 1 wurden zwei verschiedene Feststoffproben herangezogen, die aus Abwasserproben eines Kunststoffherstellers abgetrennt wurden. Diese Feststoffproben enthielten bereits Mikroplastik in Form von PE + PP und wurden mit zusätzlichem 406 PE/PP (im Verhältnis 1:1, Korngröße <500 µm) in aufsteigenden Konzentrationen von 0,5 bis 50 % MP gespikt. Anhand der Analysenergebnisse sowie der Wiederfindungsrate wurden einerseits der Messbereich der Methode überprüft und andererseits das Auftreten von Matrixeffekten bei der Analyse bestimmt. Abb. 2 zeigt die Ergebnisse des Versuchs 1. Die Wiederfindungsraten liegen zwischen 93 und 110 % und liefern damit eine erste Bestätigung für die Anwendbarkeit der EA-OEM zur Quantifizierung des Gehalts an MP. 3.2 Wiederfindungsversuch 2 Da die beim Wiederfindungsversuch 1 verwendeten Feststoffproben bereits eine Grundkontamination mit MP aufwiesen, konnte anhand der zuvor beschriebenen Ergebnisse keine absolute Nachweisgrenze für die EA-OEM bestimmt werden. Im Zuge von weiterführenden Versuchen wurde dies daher nachgeholt. Für den Wiederfindungsversuch 2 wurde Probenmaterial (biogene Probenmatrix), welches möglichst stark von den TOx -Werten von PE/PP abweicht, ausgewählt und mit PE/PP gemischt. Der zweite Versuch beschreibt daher die Nachweisgrenze der EA-OEM in einem optimalen Szenario. Es wurde ein im Zuge des Projekts BASEMAN WP2 (JPI Oceans) zur Verfügung gestelltes Probenmaterial herangezogen. Die Probe bestand aus gezüchteten, zerkleinerten Miesmuscheln und wurde vor dem Versand mit Wasser vermengt. Die Probe wurde im Labor gefriergetrocknet, homogenisiert und anschließend mit einer PE/PPMischung (Verhältnis 1:1, Korngröße <500 µm) in den Konzentrationen 1 %, 3 % sowie 5 % gespikt. Abb. 3 zeigt die erhaltenen Wiederfindungsraten der drei mittels EA-OEM gemessen Proben. Unter den oben beschriebenen optimalen Bedingungen, d. h. die TOx -Werte der biogenen Probenmatrix als auch die TOx -Werte für die Kunststofffraktion sind mit hoher Genauigkeit (geringer Unsicherheit) bekannt, konnten sehr gute Wiederfindungsraten für den Bereich 1 bis 5 % MP erzielt werden. Eine Nachweisgrenze von etwas über 1 % konnte anhand dieser Ergebnisse bestimmt werden. Veränderte Bedingungen können jedoch die Nachweisgrenze nach oben hin verschieben. Dies ist vorwiegend davon abhängig, wie akkurat die TOx -Werte der beiden Reinstofffraktionen bekannt sind bzw. bestimmt werden können. Zusätzlich ist auch die Unterschiedlichkeit der makro-elementaren Zusammensetzung zwischen der biogenen Matrix und den enthaltenen Kunststoffen entscheidend. Je unterschiedlicher die CHNSO-Gehalte ausfallen, desto genauer sind auch Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Originalarbeit 3.3 Ringversuch Abb. 3 Ergebnisse des Wiederfindungsversuchs 2: Der obere Teil der Grafik zeigt den theoretischen Wert sowie den durch EA-OEM ermittelten MP-Anteil von mit PE/PP gespiktem biogenen Probenmaterial. Die verhältnismäßig geringen MPAnteile wurden deshalb gewählt, um eine initiale Abschätzung der Nachweisgrenze für die EA-OEM zu erhalten. Die unteren Balkendiagramme zeigen die Verteilung der TOx -Werte der drei gespikten Proben. Da die Summe der gemessenen TOx Werte aufgrund von Messunsicherheiten nicht genau 100 % ergibt, wurden die Ergebnisse für die Abbildung auf 100 % normiert die Ergebnisse der EA-OEM. Die Analysengenauigkeit beeinflusst in weiterer Folge auch die Nachweisgrenze. Eine allgemeingültige Aussage über die Sensitivität der Methode lässt sich daher zum aktuellen Zeitpunkt nicht treffen. Sicher ist jedoch, dass diese von den jeweiligen Spezifikationen des Probenmaterials abhängt. Die Abhängigkeit der Bestimmungsgenauigkeit des MPGehalts vom enthaltenen Aschegehalt kann vom derzeitigen Erfahrungsstand aus als gering beurteilt werden. Weitere Erkenntnisse zur Analysenperformance der EA-OEM konnten im Zuge eines Ringversuchs gewonnen werden. Dieser wurde von der deutschen Bundesanstalt für Materialforschung und -prüfung (BAM) durchgeführt. Für diese Studie wurde eine synthetische Umweltprobe mit definiertem, aber für die Teilnehmer unbekanntem Analytgehalt an 11 verschiedene Testlabors versendet (Becker et al. 2020). Ziel war es, erste Erfahrungen über die allgemeine Eignung von thermoanalytischen Methoden hinsichtlich der Bestimmung des MP-Gehalts zu sammeln. Die Probe bestand aus einer kunststofffreien Matrix und wurde mit den vier Polymeren PE, PP, PS und PET gespikt. Die vier eingesetzten Polymertypen waren bekannt, ihre jeweiligen Anteile zueinander jedoch nicht. Zusätzlich wurde die kunststofffreie Matrix für den Ringversuch zur Verfügung gestellt. Vor Beginn der Analysen für den Ringversuch wurden intern folgende Forschungsfragen formuliert: • Ist es möglich, die unbekannte TOx Zusammensetzung der reinen Kunststofffraktion in der Probe, die einen essenziellen Eingangsparameter für die EA-OEM darstellt, anhand alternativer Strategien abzuschätzen? • Standardmäßig wird mittels EA-OEM nur der Summenparameter MP erfasst. Lassen die gemessenen TOx Werte der Kunststoffreinfraktion Rückschlüsse auf die Massenverteilung von PE/PP, PS und PET zu? Abb. 4 zeigt das Ergebnis der mittels der EA-OEM durchgeführten Bestimmungen. Der Sollwert von 45,9 µg/mg Gesamtkunststoff in der zur Verfügung gestellten Probe wurde mit dem gemessenen Wert von 44,7 µg/mg sehr gut wiedergefunden. Der Gesamtkunststoffgehalt entspricht einem Anteil von 4,5 % in der Gesamtprobe und liegt somit über der zuvor beschriebenen, theoretischen Nachweisgrenze der Me- thodik. Vor der Bestimmung des Gesamtkunststoffgehalts wurde die unbekannte Verteilung der eingesetzten Kunststoffe (Anteile an PE, PP, PS und PET) anhand einer Extraktion des vorhandenen Gesamtkunststoffs aus der gespikten Matrixprobe ermittelt. Hierfür wurde eine Dichteseparation mit gesättigtem NaI und eine anschließende oxidative Aufreinigung (H2O2) durchgeführt, um den Kunststoffanteil möglichst vollständig von der restlichen Matrix abzutrennen. Von dieser Reinfraktion wurde anschließend die elementare Zusammensetzung bestimmt und als Inputparameter für T Ox K für Gl. 2 herangezogen. Die sehr gute Annäherung an den tatsächlichen Gesamtkunststoffgehalt zeigt dabei, dass die beschriebene Aufreinigung eine Möglichkeit darstellt um eine unbekannte Kunststoffzusammensetzung bestimmen zu können. Da mittels EA-OEM standardmäßig nur der Gesamtkunststoffgehalt ermittelt wird und die Anteile der einzelnen Polymere nicht erfasst werden, wurde die Verteilung der Polymertypen in der Probe abgeschätzt. Dazu wurden die durch Extraktion bestimmten T Ox KWerte erneut in ein Gleichungssystem eingefügt (angelehnt an die Gl. 2, wobei T Ox K anstatt T Ox Probe eingesetzt wurde und die Gleichungen um 3 Fraktionen erweitert wurden: PE/PP, PS und PET), mit dem Ziel, die Massenanteile der einzelnen Polymere in der Kunststoffmischung abzuschätzen. Die Berechnung ergab Anteile von 60,1 % PE/PP, 0,3 % PS und 39,6 % PET. Diese Bestimmung der Polymeranteile in der Kunststoffmischung wurde jedoch nicht genauer überprüft und kann daher nur als grobe Abschätzung verstanden werden. Tab. 1 stellt die Sollwerte mit der ermittelten Kunststoffverteilung gegenüber. Die PE/PP-Anteile sowie die PET-Anteile konnten sehr gut bestimmt werden (Wiederfindungsraten von 104 und 99 %). Einzig der PS-Anteil konnte nicht korrekt bestimmt werden (Wiederfindung bei nur knapp 6 %). Ein Tab. 1 Ermittelte Wiederfindungsraten des Ringversuchs. Zusätzlich zu der Bestimmung des Summenparameters „Gesamtkunststoff“ wurde im Zuge des Ringversuchs erstmals auch die Verteilung der Kunststoffe abgeschätzt. Diese Modellrechnung anhand der TOx -Werte der Kunststoffreinfraktion ergab sehr gute Wiederfindungsraten für PE/PP und PET. PS wurde in dem Modell zu gering eingeschätzt und vermutlich eher der PE/PP-Fraktion zugeordnet Polymer Gesamtkunststoff PE+PP PS PET Theoretischer Wert [µg/mg] 45,9 25,7 2,2 18,0 Gemessener Wert [µg/mg] 44,72 26,74 0,13 17,85 Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Wiederfindungsrate [%] 97,4 104,0 5,9 99,2 407 Originalarbeit Abb. 4 Ergebnisse der EA-OEM-Messungen für die Proben des von der deutschen Bundesanstalt für Materialforschung und prüfung (BAM) organisierten Ringversuchs (Becker et al. 2020). Grafik a Gegenüberstellung der theoretischen Werte und der mittleren EA-OEM wiedergefundenen Kunststoffanteile. Grafik b TOx -Zusammensetzung (x = CHNSO) der MP-freien Matrix, der unbekannten Testprobe (Matrix + Mikroplastik) und der im Labor durch Dichteseparation und Aufreinigung gewonnenen Kunststoffmischung aus der Testprobe (Kunststoffmix). Diese dargestellten Werte wurden für die Berechnung des MP-Anteils in der Probe mittels EA-OEM herangezogen möglicher Grund hierfür ist die vergleichsweise niedrige Konzentration an PS in der Kunststoffmischung und die doch recht ähnliche makro-elementare Zusammensetzung von PS und PE/PP. Dadurch werden vermutlich PS-Anteile teilweise der PE/PP-Fraktion zugeordnet. Es kann davon ausgegangen werden, dass PS grundsätzlich im Zuge der EA-OEM mitbestimmt wird, da dieses Polymer im Verbrennungsofen genauso vollständig aufgeschlossen wird wie die anderen Polymere. 4 Zusammenfassung und Ausblick Wie bereits Schwarzböck et al. (2016, 2018) in vorangegangenen Studien zur adaptierten Bilanzenmethode demonstrieren konnten, lässt sich der Kunststoffanteil in Abfallströmen sehr gut mithilfe der Elementaranalyse bestimmen. Die Anpassung der Methode für die MP-Bestimmung in Umweltproben war dabei methodisch gesehen nur noch ein kleiner Schritt. Hauptsächlich die deutlich geringeren Probemengen (oftmals weniger als 1 g Gesamtprobe) machten einige Anpassungen in der praktischen Durchführung der EA-OEM im Vergleich zur Anwendung auf Abfallproben notwendig. Die Ergebnisse der Wiederfindungsversuche sowie des Ringversuchs bestätigen dabei die Zuverlässigkeit der Methodik über einen weiten Messbereich. 408 Die Elementaranalyse (CHNSO-Bestimmung) als Kern der Methode stellt eine kosteneffiziente und mit verhältnismäßig geringem Aufwand zu betreibende Analysemethode dar, welche auch in kleineren Betriebslaboren für Routineuntersuchungen etabliert werden kann. Der gesamte organische Kohlenstoff (TOC Wert), als Summenparameter, ist beispielsweise eine häufig analysierte Messgröße in der Umweltanalytik. Die Erweiterung von TOC auf TOx (x = CHNSO) ist zwar mit zusätzlichem Aufwand verbunden, jedoch bleibt die Methode zeit- und kosteneffizient. Zusätzlich ist das instrumentelle Messprinzip bereits seit Jahrzehnten im Einsatz und hat sich als robuste Methode erwiesen, die mit einem verhältnismäßig geringen Aufwand und überschaubarer Einschulung betrieben werden kann. Der kritischste Punkt in der aktuellen Entwicklung liegt in der Bestimmung der Zusammensetzung der Reinstofffraktionen (CHNSO-Zusammensetzung der Kunststoffe und der biogenen Fraktion in der Probe), da diese, anders als bei Ersatzbrennstoffen, nicht durch händisches Sortieren bestimmt werden können. Vielmehr muss bei MP-Umweltproben meist mit Dichteseparation, chemischer Vorbehandlung oder einer Kombination mit anderen analytischen Methoden (z. B. TED-GC-MS, PyrolyseGC-MS oder FTIR-Mikroskopie) gearbeitet werden, um der geringen Proben- menge und den kleinen Partikelgrößen gerecht zu werden. Der dadurch entstehende Mehraufwand ist jedoch nur einmalig zu Beginn der Entwicklung eines Messmodells für eine spezifische Probenzusammensetzung notwendig. Laufende bzw. wiederholende Messungen des MP-Gehalts an denselben Probenahmeorten werden dadurch in Bezug auf Zeit und Kosten nicht beeinflusst. Eine geeignete Erstabschätzung der CHNSO-Zusammensetzung der Reinstofffraktionen (biogen und Kunststoffe in der Probe) ist oftmalig schon aufgrund des bereits beschriebenen engen Wertebereiches der CHNSOZusammensetzung (auf wasser- und aschefreier Basis) von jeweils biogenen und Kunststoffmaterialien möglich. Ergebnisse vergangener Untersuchungen deuten darauf hin, dass oftmals nur wenige Hauptkunststoffsorten (primär PE, PP, PA, PET) den Großteil der Mikroplastikbelastung in Umweltproben ausmachen, was die Charakterisierung der Kunststoffreinfraktion erleichtert. Weiters kann mithilfe alternativer Methoden wie TED-GC-MS, Pyrolyse-GCMS aber auch mikroskopischer Verfahren eine Erstbestimmung der Probenzusammensetzung hinsichtlich Polymertypen durchgeführt werden. Diese Methoden sind deutlich aufwendiger im Betrieb und mit höheren Kosten in der Geräteanschaffung verbunden. Für die einmalige Bestimmung notwendiger Eingangsdaten für die EA-OEM ist Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben Originalarbeit die Kombination aus verschiedenen Analysenmethoden jedoch vielversprechend. Ein weiterer wichtiger Schritt in Richtung einer möglichst genauen Bestimmung von MP in Umweltproben ist die Entwicklung von standardisierten Probenaufbereitungsprotokollen. Dieser Aspekt ist nicht nur besonders wichtig für die verhältnismäßig wenig sensitive EA-OEM (Nachweisgrenze von MP beträgt im Optimalfall rund 1 %), sondern auch für den Rest der in der Literatur beschriebenen mikroskopischen und auch thermoanalytischen Methoden. Eine standardisierte Probenaufbereitung ist ein wesentlicher Faktor für akkurate und vergleichbare Analysenergebnisse. Für eine ganzheitliche Betrachtung der Problematik Analytik von Mikroplastik in der Umwelt wird in Zukunft die Kombination von thermoanalytischen Methoden und mikroskopischen Untersuchungen, welche sich ebenfalls laufend weiterentwickeln, unumgänglich sein, um sowohl ökotoxikologisch wichtige Parameter wie Partikelgrößen und Partikelformen zu bestimmen und andererseits den massebezogenen MPGehalt als vergleichenden Parameter zu erhalten. Die Bestimmung der MPMassen in der Umwelt ist eine Größe, die dabei für nationale und internationale Vergleiche und Auswertungen zwingend notwendig ist. Mittels der EA-OEM kann der MP-Gehalt in Umweltproben massenbezogen bestimmt werden. Durch die zusätzlich auch kostengünstige Anwendung und die vielversprechenden Ergebnisse positioniert sich die vorgestellte Methodik daher innerhalb der bisher vorhandenen Methoden als potenzielles Messverfahren für ein routinemäßiges Monitoring von Mikroplastik in Umweltkompartimenten (z. B. zur Grenzwertüberwachung). Funding Open access funding provided by TU Wien (TUW). Open Access Dieser Artikel wird unter der Creative Commons Namensnennung 4.0 International Lizenz veröffentlicht, welche die Nutzung, Vervielfältigung, Bearbeitung, Verbreitung und Wiedergabe in jeglichem Medium und Format erlaubt, sofern Sie den/die ursprünglichen Autor(en) und die Quelle ordnungsgemäß nennen, einen Link zur Creative Commons Lizenz beifügen und angeben, ob Änderungen vorgenommen wurden. Die in diesem Artikel enthaltenen Bilder und sonstiges Drittmaterial unterliegen ebenfalls der genannten Creative Commons Lizenz, sofern sich aus der Abbildungslegende nichts anderes ergibt. Sofern das betreffende Material nicht unter der genannten Creative Commons Lizenz steht und die betreffende Handlung nicht nach gesetzlichen Vorschriften erlaubt ist, ist für die oben aufgeführten Weiterverwendungen des Materials die Einwilligung des jeweiligen Rechteinhabers einzuholen. Weitere Details zur Lizenz entnehmen Sie bitte der Lizenzinformation auf http://creativecommons.org/licenses/ by/4.0/deed.de. Literatur Becker, R., Altmann, K., Sommerfeld, T., Braun, U. (2020): Quantification of microplastics in a freshwater suspended organic matter using different thermoanalytical methods—outcome of an interlaboratory comparison. Journal of Analytical and Applied Pyrolysis 148:104829. https://doi.org/10.1016/j.jaap.2020.104829 Brandt, J., Bittrich, L., Fischer, F., Kanaki, E., Tagg, A., Lenz, R., Labrenz, M., Brandes, E., Fischer, D., Eichhorn, K.J. (2020): High-Throughput Analyses of Microplastic Samples Using Fourier Transform Infrared and Raman Spectrometry. Appl Spectrosc. https://doi.org/10.1177/ 0003702820932926 Derraik, J.G.B. (2002): The pollution of the marine environment by plastic debris: a review. Mar Pollut Bull 44 (9):842–852. Doi https://doi.org/ 10.1016/S0025-326x(02)00220-5 Dümichen, E., Eisentraut, P., Bannick, C.G., Barthel, A.K., Senz, R., Braun, U. (2017): Fast identification of microplastics in complex environmental samples by a thermal degradation method. Chemosphere 174:572–584. https:// doi.org/10.1016/j.chemosphere.2017.02.010 Fellner, J., Aschenbrenner, P., Cencic, O., Rechberger, H. (2011): Determination of the biogenic and fossil organic matter content of refuse-derived fuels based on elementary analyses. Fuel 90 (11):3164–3171. https://doi.org/10.1016/j.fuel. 2011.06.043 Fischer, M., Scholz-Bottcher, B.M. (2017): Simultaneous Trace Identification and Quantification of Common Types of Microplastics in Environmental Samples by Pyrolysis-Gas Chromatography-Mass Spectrometry. Environ Sci Technol 51 (9):5052–5060. https://doi.org/10. 1021/acs.est.6b06362 Jahan, M.S., Liu, Z., Wang, H., Saeed, A., Ni, Y. (2012): Isolation and Characterization of Lignin from Prehydrolysis Liquor of Kraft-Based Dissolving Pulp Production. Cell Chem Technol 46 (3–4):261–267 La Nasa, J., Biale, G., Fabbri, D., Modugno, F. (2020): A review on challenges and developments of analytical pyrolysis and other thermoanalytical techniques for the quali-quantitative determination of microplastics. Journal of Analytical and Applied Pyrolysis 149:104841. https:// doi.org/10.1016/j.jaap.2020.104841 Law, K.L., Thompson, R.C. (2014): Microplastics in the seas. Science 345 (6193):144–145. https:// doi.org/10.1126/science.1254065 Mallow, O., Spacek, S., Schwarzböck, T., Fellner, J., Rechberger, H. (2020): A new thermoanalytical method for the quantification of micropla- Eine neue Methode für die Bestimmung des Mikroplastik-Massenanteils in Umweltproben stics in industrial wastewater. Environ Pollut 259. doi:ARTN 11386210.1016/j.envpol.2019.113862 Prata, J.C., da Costa, J.P., Duarte, A.C., Rocha-Santos, T. (2019): Methods for sampling and detection of microplastics in water and sediment: A critical review. Trac-Trend Anal Chem 110:150–159. https://doi.org/10.1016/j. trac.2018.10.029 Schwarzböck, T., Aschenbrenner, P., Spacek, S., Szidat, S., Rechberger, H., Fellner, J. (2018): An alternative method to determine the share of fossil carbon in solid refuse-derived fuels—Validation and comparison with three standardized methods. Fuel 220:916–930. https://doi.org/10.1016/ j.fuel.2017.12.076 Schwarzböck, T., Eygen, E., Rechberger, H., Fellner, J. (2016): Determining the amount of waste plastics in the feed of Austrian waste-toenergy facilities. Waste Manage Res 35. https:// doi.org/10.1177/0734242X16660372 Hinweis des Verlags Der Verlag bleibt in Hinblick auf geografische Zuordnungen und Gebietsbezeichnungen in veröffentlichten Karten und Institutsadressen neutral. 409
https://openalex.org/W3109570958	https://vbn.aau.dk/files/402004989/Frischknecht_2020_IOP_Conf._Ser._Earth_Environ._Sci._588_022029.pdf	English	null	Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches – IEA EBC Annex 72	IOP conference series. Earth and environmental science	2,020	cc-by	5,576	Aalborg Universitet Citation for published version (APA): Frischknecht, R., Ramseier, L., Yang, W., Birgisdottir, H., Chae, C. U., Lützkendorf, T., Passer, A., Balouktsi, M., Berg, B., Bragança, L., Butler, J., Cellura, M., Dixit, M., Dowdell, D., Francart, N., García Martínez, A., Gomes, V., Gomes da Silva, M., Guimaraes, G., ... Zara, O. (2020). Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches - IEA EBC Annex 72. In H. Walbaum, A. Hollberg, L. Thuvander, P. Femenias, I. Kurkowska, K. Mjörnell, & C. Fudge (Eds.), Conference Proceedings: World Sustainable Built Environment online conference BEYOND 2020: 2- 4 November 2020 (pp. 1-9). Article 022029 IOP Publishing. https://doi.org/10.1088/1755-1315/588/2/022029 Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches - IEA EBC Annex 72 Frischknecht, R.; Ramseier, L.; Yang, W.; Birgisdottir, H.; Chae, Ch U.; Lützkendorf, T.; Passer, A.; Balouktsi, M.; Berg, B.; Bragança, L.; Butler, J.; Cellura, M.; Dixit, M.; Dowdell, D.; Francart, N.; García Martínez, A.; Gomes, V.; Gomes da Silva, M.; Guimaraes, G.; Hoxha, E.; Kjendseth Wiik, M.; König, H.; Llatas, C.; Longo, S.; Lupíšek, A.; Martel, J.; Mateus, R.; Nygaard Rasmussen, F.; Ouellet-Plamondon, C.; Peuportier, B.; Pomponi, F.; Pulgrossi, L.; Röck, M.; Satola, D.; Soust Verdaguer, B.; Szalay, Z.; Truong Nhu, A.; Veselka, J.; Volf, M.; Zara, O. Published in: Published in: Conference Proceedings DOI (link to publication from Publisher): 10.1088/1755-1315/588/2/022029 Creative Commons License CC BY 3.0 Document Version Publisher's PDF, also known as Version of record Link to publication from Aalborg University Citation for published version (APA): Frischknecht, R., Ramseier, L., Yang, W., Birgisdottir, H., Chae, C. U., Lützkendorf, T., Passer, A., Balouktsi, M., Berg, B., Bragança, L., Butler, J., Cellura, M., Dixit, M., Dowdell, D., Francart, N., García Martínez, A., Gomes, V., Gomes da Silva, M., Guimaraes, G., ... Zara, O. (2020). Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches - IEA EBC Annex 72. In H. Walbaum, A. Hollberg, L. Thuvander, P. Femenias, I. Kurkowska, K. Mjörnell, & C. Fudge (Eds.), Conference Proceedings: World Sustainable Built Environment online conference BEYOND 2020: 2- 4 November 2020 (pp. 1-9). Article 022029 IOP Publishing. https://doi.org/10.1088/1755-1315/588/2/022029 IOP Conference Series: Earth and Environmental Science PAPER • OPEN ACCESS PAPER • OPEN ACCESS This content was downloaded from IP address 130.225.247.88 on 22/01/2021 at 08:33 Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches – IEA EBC Annex 72 To cite this article: R Frischknecht et al 2020 IOP Conf. Ser.: Earth Environ. Sci. 588 022029 View the article online for updates and enhancements. BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 Comparison of the greenhouse gas emissions of a high-rise residential building assessed with different national LCA approaches – IEA EBC Annex 72 R Frischknecht1, L Ramseier1, W Yang2, H Birgisdottir3, Ch U Chae4, T Lützkendorf5, A Passer6, M Balouktsi5, B Berg7, L Bragança8, J Butler7, M Cellura9, M Dixit10, D Dowdell7, N Francart11, A García Martínez12, V Gomes13, M Gomes da Silva14, G Guimaraes13,E Hoxha6, M Kjendseth Wiik15, H König16, C Llatas12, S Longo9, A Lupíšek17, J Martel18, R Mateus8, F Nygaard Rasmussen3, C Ouellet-Plamondon19, B Peuportier20, F Pomponi21, L Pulgrossi13, M Röck6, D Satola22, B Soust Verdaguer12, Z Szalay23, A Truong Nhu6, J Veselka17,M Volf17 and O Zara13 1 treeze Ltd., Uster, Switzerland; 1 treeze Ltd., Uster, Switzerland; 2 Tianjin University, School of Architecture, Tianjin, China 3 3 Aalborg University, Danish Building Research Institute, Copenhagen, Denmark; g y, g , p g , ; 4 Korea Institute of Civil Engineering and Building Technology, Gyeonggi-do Republic of Korea; 5 Karlsruhe Institute of Technology, Karlsruhe, Germany; 6 Graz University of Technology, Graz, Austria; 7 BRANZ, Porirua, New Zealand; 8 University of Minho, Civil Engineering, Guimarães, Portugal; 9 9 University of Palermo, Palermo, Italy; 10 Texas A&M University, Construction Science College, Station, USA 11 KTH Royal Institute of Technology, Stockholm, Sweden; yal Institute of Technology, Stockholm, Sweden 12 Universidad de Sevilla, Construcciones Arquitectónicas I., Seville, Spain; 13 University of Campinas GBLab, Campinas, Brazil; 14 Federal University of Espírito Santo, Vitoria, Brazil 15 SINTEF Building and Infrastructure, Oslo, Norway 16 Ascona, Gröbenzell, Germany; 17 Czech Technical University in Prague, University Centre for Energy Efficient 17 Czech Technical University in Prague, University Centre for Energy Efficient Buildings, Prague, Czech Republic; 18 Groupe Ageco, Montreal, Canada É 19 École de technologie supérieure, Génie de la construction, Montreal, Canada; 20 20 MINES ParisTech, Centre Efficacité énergétique des Systèmes, Paris, France 21 Edinburgh Napier University, Resource Efficient Built Environment Lab, Edinburgh Scotland; 22 NTNU – Norwegian University of Science and Technology, Trondheim, Norway NTNU – Norwegian University of Science and Technology, Trondheim, Norway Budapest University of Technology and Economics, Budapest, Hungary 23 Budapest University of Technology and Economics, Budapest, Hungary Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by IOP Publishing Ltd 1 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 Abstract. Introduction: The international research project IEA EBC Annex 72 investigates the life cycle related environmental impacts caused by buildings. The project aims inter alia to harmonise LCA approaches on buildings. Methods: To identify major commonalities and discrepancies among national LCA approaches, reference buildings were defined to present and compare the national approaches. A residential high-rise building located in Tianjin, China, was selected as one of the reference buildings. The main construction elements are reinforced concrete shear walls, beams and floor slabs. The building has an energy reference area of 4566 m2 and an operational heating energy demand of 250 MJ/m2a. An expert team provided information on the quantities of building materials and elements required for the construction, established a BIM model and quantified the operational energy demand. Results: The greenhouse gas emissions and environmental impacts of the building were quantified using 17 country-specific national assessment methods and LCA databases. Comparisons of the results are shown on the level of building elements as well as the complete life cycle of the building. Conclusions: The results of these assessments show that the main differences lie in the LCA background data used, the scope of the assessment and the reference study period applied. Despite the variability in the greenhouse gas emissions determined with the 17 national methods, the individual results are relevant in the respective national context of the method, data, tool and benchmark used. It is important that environmental benchmarks correspond to the particular LCA approach and database of a country in which the benchmark is applied. Furthermore, the results imply to include building technologies as their contribution to the overall environmental impacts is not negligible. Grant support: The authors thank the IEA for its organizational support and the funding organizations in the participating countries for their financial support. frischknecht@treeze.ch Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by IOP Publishing Ltd 1 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 view. Hence, it is no cause for a major concern if the environmental impacts of the same building differ between the different national approaches. However, it is recommended to use LCA databases representative for the countries’ relevant economic sectors. view. Hence, it is no cause for a major concern if the environmental impacts of the same building differ between the different national approaches. However, it is recommended to use LCA databases representative for the countries’ relevant economic sectors. p A residential high-rise building (TJ-CSY-11) located in Tianjin, China, was selected as one of three reference buildings within the IEA EBC Annex 72 project. Compared to the “be2226” office reference building [5], this residential building is more complex regarding materialisation and building technology. An expert team from the Tianjin University in China provided information on the quantities of building materials and elements required for the construction, established a BIM model and quantified the operational energy demand. The building has 12 floors and the main construction elements are reinforced concrete shear walls, beams and floor slabs. The building has an energy reference area of 4566 m2. The operational energy demand is 250 MJ per m2 per year. The operational energy demand includes space heating provided by a waste-heat-source heat pump operated with natural gas. The electricity demands for generating hot water, ventilation and cooling, elevators, lighting and other operational facilities were quantified separately. 1. Introduction The construction and operation of buildings are a major cause for climate change and other environmental impacts [1-3]. Environmental life cycle assessment (LCA) is widely used to quantify greenhouse gas (GHG) emissions and other environmental impacts of buildings and to highlight optimization and improvement potentials over their whole life cycle (production, construction, use - including repair and replacement - and end of life). LCA results support decision making in favour of a more climate and environmental friendly production and consumption and therefore help to achieve the UN Sustainable Development Goals (SDG) number 11 (sustainable cities and communities) number 12 (responsible consumption and production) and 13 (climate action). The international research project IEA EBC Annex 72 investigates the life cycle related environmental impacts caused by buildings and aims inter alia to discuss and harmonise LCA approaches on buildings [4]. To present existing national approaches and identify commonalities and discrepancies three reference buildings were defined within the IEA EBC Annex 72 project. For each reference building an expert team provided the bill of materials and operational energy demands in local context. National experts assessed the environmental impacts using the provided information on quantities of building materials and operational energy demands, but applying their national or regional LCA approach and LCA database, whenever available. The first analysed reference building was the “be2226” office building located in Austria. The building is a massive construction with thick exterior walls with a high thermal capacity. Therefore, no active heating and air-conditioning is required. 22 different institutions assessed the “be2226” office building according to their national or regional LCA approach and LCA database. Depending on the assessment the GHG emissions of the be2226 building were between 10 and 71 kg CO2-eq per m2 per year. Most of the GHG emissions were either caused in the product stage or during the operational energy use. The differences in GHG emissions were due to variances in GHG emissions per kg building material, differences in the applied reference study periods and the different GHG intensities of the national electricity mixes [5]. National or regional LCA databases reflect the production conditions and the energy mix in a specific country and are therefore important for assessing the environmental impacts in a true and fair 2 2. Methods and databases 2.1. Used national methods including reference study period and databases 2.1. Used national methods including reference study period and databases 2.1. Used national methods including reference study period and databases The environmental impacts of the high-rise building were assessed by 17 institutions. The authors of the study used the same material amounts and energy demand but applied their regional LCA methods for evaluating the primary energy demand (non-renewable and renewable) and the GHG emissions. The LCA methods applied use different reference study periods, apply a different scope (i.e. life cycle stages included) and apply different background databases. 10 methods use a reference study period of 50 years and 4 methods use 60 years. New Zealand uses 90 years, France 100 years and Denmark 120 years as reference study period for the residential building. The ecoinvent database (different versions) was mostly used as background data source, but some country-specific databases (e.g. Ökobau.dat [6]) and EPDs were also applied (see Table 1). A comparison of the environmental impacts of different materials and electricity mixes from different databases is presented in [5]. Table 1: Overview of the reference study periods and databases used within the LCA methods applied to assess the environmental impacts of the TJ-CSY-11 reference building. Reference study period [years] Database Field of application AT 50 ecoinvent 3.5 [7] Research BR 50 ecoinvent 3.4 [8] /ecoinvent 3.5 [7] adapted to Brazilian context and EPD Research CA 60 ecoinvent 3.5 [7] adapted to Canadian context and EPDs Building certification schemes, EPDs CN 50 ecoinvent 3.5[7]; CLCD-China-ECER 0.8.1, Oekobau.dat [6, 9] Building certification scheme CZ 50 ecoinvent 3.3 [10], boundary condition from SBToolCZ methodology [11] Decision-making tool, voluntary certification DE 50 Ökobau.dat 2018 [6] BNB and DGNB DK 120 Ökobau.dat 2016 [9] DGNB Denmark ES 50 ecoinvent 2.0 [12] research FR 100 ecoinvent 2.2 [13] EQUER HU 50 ecoinvent 3.5 [7] adapted to Hungarian context Education and research IT 50 Ecoinvent 3.4 [8], EPDs Research NO 60 Ecoinvent 3.0 [14], EPDs Research, decision-making tool NZ 90 NZ whole building whole of life framework - materials data developed from EPDs for materials and modelling in ecoinvent 3.1 [15] (specific process data with NZ Grid electricity) Certification, research e 1: Overview of the reference study periods and databases used within the LCA methods applie to assess the environmental impacts of the TJ-CSY-11 reference building. 3 3 BEYOND 2020 – World Sustainable Built Environment conference BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. 2.1. Used national methods including reference study period and databases Series: Earth and Environmental Science 588 (2020) 022029 g doi:10.1088/1755-1315/588/2/022029 Reference study period [years] Database Field of application PT 50 LCIA Database for Portuguese Building Technologies [16], based on generic data from Ecoinvent 2.1 [17], Ecoinvent version 3.3 [10] Research SE 50 Swedish Building Sector Environmental Calculation Tool (BM) [18] Building certification schemes UK 50 Database embedded in OneClickLCAa Building certification schemes US 50 Database embedded in ATHENA Impact Estimatorb Building certification schemes and research a https://www.oneclicklca.com/support/faq-and-guidance/documentation/database/, last visited on: 8.01.2020 b https://calculatelca.com/software/impact-estimator/lca-database-reports/, last visited on: 8.01.2020 a https://www.oneclicklca.com/support/faq-and-guidance/documentation/database/, last visited on: 8.01.2020 b https://calculatelca.com/software/impact-estimator/lca-database-reports/, last visited on: 8.01.2020 The life cycle stages included in the approaches are shown in Table 2. The life cycle stage B1 is not considered by any approach. The life cycle stages included in the approaches are shown in Table 2. The life cycle stage B1 is not considered by any approach. Table 2: Overview of the life cycle stages included in the applied approaches. 2.1. Used national methods including reference study period and databases Life cycle stage A1- A3 A4- A5 B2 B3 B4 B5 B6- SH B6- HW B6- VC B6- LO B6- E B7 C1 C2 C3 C4 D AT X X X X X X X X X X X BR X X X X X X X X X X CA X X X X X X X X X X X X CN X X X X X X X X CZ X X X X X X X DE X X X X X X X X X (X) DK X X X X X X X X X ES X X X X X X X X X X X X X X X X FR X X X X X X X X X X X X HU X X X X X X X X X X X X IT X X X X X X NO X X X X X X X X X NZ X X X X X X X X X X X X X X X PT X X X X SE X X UK X X X X X X X X X X X X X US X X X X X X X X X X X X X B6-SH: space heating B6-HW: hot water B6-VC: ventilation and cooling B6-LO: lighting, operational facilities (electric doors, shadowing equipment), auxiliaries B6-E: elevators Table 2: Overview of the life cycle stages included in the applied approaches. 2.1. Used national methods including reference study period and databases Life cycle stage A1- A3 A4- A5 B2 B3 B4 B5 B6- SH B6- HW B6- VC B6- LO B6- E B7 C1 C2 C3 C4 D AT X X X X X X X X X X X BR X X X X X X X X X X CA X X X X X X X X X X X X CN X X X X X X X X CZ X X X X X X X DE X X X X X X X X X (X) DK X X X X X X X X X ES X X X X X X X X X X X X X X X X FR X X X X X X X X X X X X HU X X X X X X X X X X X X IT X X X X X X NO X X X X X X X X X NZ X X X X X X X X X X X X X X X PT X X X X SE X X UK X X X X X X X X X X X X X US X X X X X X X X X X X X X B6-SH: space heating B6-HW: hot water B6-VC: ventilation and cooling B6-LO: lighting, operational facilities (electric doors, shadowing equipment), auxiliaries B6-E: elevators 3. Results: GHG emissions of TJ-CSY-11 building 3. Results: GHG emissions of TJ-CSY-11 building 3.1. Results – life cycle stages The national experts reported the results according to the life cycle stages defined in EN-15804:2012 [19] and EN-15978:2011 [20]. Figure 1 presents the GHG emissions of the TJ-CSY-11 reference building over the different life cycle stages. Depending on the national approach used, the GHG emissions range between 15 and 67 kg CO2-eq per m2 per year. 3.1. Results – life cycle stages The GHG emissions of the product stage were reported by all countries and differ between 4 and 16 kg CO2-eq per m2 per year. The construction process stages (A4 and A5) were assessed by 10 approaches and vary between 0.4 and 6.7 kg CO2-eq per m2 per year. 4 4 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 BEYOND 2020 – World Sustainable Built Environment conference BEYOND 2020 World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 doi:10.1088/1755-1315/588/ IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 -10 0 10 20 30 40 50 60 70 80 AT BR CN CZ DE ES HU IT PT SE CA NO* NO** UK US NZ FR DK 50 a 60 a 90 a 100 a 120 a kg CO2-eq/m2a A1-A3 A4-A5 B1 B2-B5 B6 - total B7 C1-C4 D Figure 1. GHG emissions in kg CO2-eq per m2 per year of the reference building “TJ-CSY-11” assessed according to the national/regional approaches of the countries listed. Results variations reflect the different national methods and databases used. NO: Norwegian electricity grid1; NO: open electricity grid (EU 28+NO)2 -10 0 10 20 30 40 50 60 70 80 AT BR CN CZ DE ES HU IT PT SE CA NO NO** UK US NZ FR DK 50 a 60 a 90 a 100 a 120 a kg CO2-eq/m2a A1-A3 A4-A5 B1 B2-B5 B6 - total B7 C1-C4 D Figure 1. GHG emissions in kg CO2-eq per m2 per year of the reference building “TJ-CSY-11” assessed according to the national/regional approaches of the countries listed. Results variations reflect the different national methods and databases used. 1 Consideration of isolated energy system in Norway based on the dominant share of hydropower (based on the data from the Statistic Norway, www.ssb.no) 2 Average value that is representative of a 60-year building lifetime, taking into consideration future evolutions in the European electricity generation towards 2050. 3 New Zealand reported B4 together with B2. 3 New Zealand reported B4 together with B2. 3.2. Results – elements level Figure 2 presents the GHG emission of the different building elements in the TJ-CSY-11 reference building (reported by 14 out of 17 countries) in kg CO2-eq over the whole life cycle and reference study period. In 12 assessments most of the emissions caused by the building materials (i.e. without GHG emissions caused by operational energy use) are either caused by the interior walls or the exterior walls. In the Italian assessment the floor structure plus finishes and in the French assessment the building service systems cause most of the GHG emissions. Figure 2: GHG emissions of building elements of TJ-CSY-11 in kg CO2-eq over the whole life cycle and reference study period. Life cycle stages included, reference study period and applied databases differ between the national approaches. 0.E+00 1.E+06 2.E+06 3.E+06 4.E+06 5.E+06 6.E+06 AT BR CN CZ DE ES HU IT PT SE CA NZ FR DK 50a 60a 90a 100 a 120 a kg CO2-eq Foundation Floor structure + finishes Interior wall + finishes Exterior wall + finishes Roofing Openings others building elements Building service systems others (A5/C1) Figure 2: GHG emissions of building elements of TJ-CSY-11 in kg CO2-eq over the whole life cycle and reference study period. Life cycle stages included, reference study period and applied databases differ between the national approaches. The distribution of the shares of the GHG emissions of the foundation, floor structure, interior and exterior walls of the GHG emissions caused by the building elements is similar across the assessments from the different countries. The distribution of the shares of the GHG emissions of the foundation, floor structure, interior and exterior walls of the GHG emissions caused by the building elements is similar across the assessments from the different countries. Within the approaches applying a 50 year reference period, the GHG emissions of the building elements vary between factor 2.5 and 3.4. Exceptions are the building element roofing and the openings, for which the GHG emissions vary by a factor of 6.1 and 11.8, respectively. An overview of the variance for each building element is shown in Table 3. Table 3. Overview of the variance of the absolute GHG emissions of the building elements for the approaches applying a 50 years reference study period. 3. Results: GHG emissions of TJ-CSY-11 building NO: Norwegian electricity grid1; NO*: open electricity grid (EU 28+NO)2 The GHG emissions of replacements (B4) are between 0.33 and 11.9 kg CO2-eq per m2 per year, and were considered by all countries except Italy, Portugal and Sweden. The operational energy use (B6) was split into space heating, hot water, ventilation and cooling, lighting and operational facilities and elevators. 14 approaches considered all operational energy uses. China and Portugal did not consider the operational energy use from elevators. Furthermore Portugal did neither consider operational energy use from lighting and operational facilities. The approach applied by Sweden is not considering the operational energy use at all. Overall, the GHG emissions caused by the operational energy use vary between 7.9 and 45 kg CO2-eq per m2 per year. The difference is mainly due to the variations in the GHG emissions of the electricity mixes used in the different countries. The GHG emissions of the operational water use were assessed in 3 approaches and are of minor importance. Different modules of the end-of-life stage were taken into account. The deconstruction stage (C1) was assessed by 6 countries, the transport to the disposal (C2) by 9 countries and waste processing (C3) and disposal (C4) each by 11 countries. The GHG emissions of the end-of-life stage are between 0.14 and 2.6 kg CO2-eq per m2 per year. 5 countries reported potential loads and benefits beyond the system boundaries (module D), which range between -3.9 and -0.6 kg CO2-eq per m2 per year. 5 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 4. Discussion In all assessments, except in the Swedish assessment, most of the GHG emissions are caused by the operational energy demand. In the Swedish approach the operational energy demand is not assessed. The differences in the GHG emissions of the operational energy demand are mainly caused by existing differences in the national electricity mixes. As for the “be2226” reference building [5] the difference in the GHG emissions in the product stage are due to diversities in the reference study period applied and the differences in GHG emissions, caused by the production of the construction materials. The GHG emissions caused by the production of materials used for the building service system tend to cause a more significant share of the total absolute GHG emissions with longer reference study periods. This is mainly due to the often shorter service life of the building service system compared to the building life time and thus leading to several replacements causing additional GHG emissions. For example in the Danish assessment the service life of the heating system is 30 years. A reference study period of 120 years leads to 3 replacements of the heating system. This effect is also observed for other building elements with a shorter service life (i.e. windows and doors), but they are of minor importance with respect to the total absolute GHG emissions of the building elements. However, the GHG emissions caused by the building service system of the assessments with a reference study period of 50 years show a similar variance like other building elements. As in the “be2226” reference building case [5], the GHG emissions of the TJ-CSY-11 reference building differ substantially depending on the applied approach and are mainly due to the different GHG emissions of the energy carriers and construction materials used, differences in scope (inclusion or not of building service systems) and reference study period. In most cases, methodological differences are of minor importance. As concluded in [5], it is important that environmental benchmarks correspond to the particular LCA approach and database of a country in which the benchmark is applied. pp Despite the variability in the GHG emissions determined with the 17 national methods, the individual results are relevant in their respective national contexts. Many nations run their labelling or certification systems applying a national method, national or nationally adapted LCA data and national building benchmarks. 4. Discussion It is more important that a national assessment of the TJ-CSY-11 is in accordance with the national system of assessing environmental impacts of buildings than of reaching identical results in assessments done according to national approaches. BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 The building service systems include the systems for water, sewage, electricity, heating, cooling, ventilation, elevator and hot water. The share of the GHG emissions of building service systems of the GHG emissions caused by the building elements is between 5 % and 21 % depending on the national approach applied. 3 countries out of 14 countries did not assess the GHG emissions caused by building service systems. 3.2. Results – elements level Building element Min GHG emissions [kg CO2-eq] Max GHG emissions [kg CO2-eq] Variation factor Max/Min Foundation 1.4E+05 4.3E+05 3.2 Floor structure + finishes 4.3E+05 1.1E+06 2.5 Interior wall + finishes 5.6E+05 1.7E+06 3.1 Exterior wall + finishes 5.3E+05 1.5E+06 2.8 Roofing 5.2E+04 3.2E+05 6.1 Openings 6.3E+04 7.4E+05 11.8 others building elements 1.5E+05 3.7E+05 2.5 Building service systems 1.3E+05 4.3E+05 3.4 Module D -8.9E+05 -2.3E+05 0.3 others (A5/C1) 0.0E+00 3.6E+04 Table 3. Overview of the variance of the absolute GHG emissions of the building elements f approaches applying a 50 years reference study period ble 3. Overview of the variance of the absolute GHG emissions of the building elements for the approaches applying a 50 years reference study period. 6 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 5. Outlook A Canadian pre-fabricated wood building, will be the third reference building and assessed with a special focus on methodological choices of the modelling of biogenic carbon and biogenic CO2 and CH4 emissions. The lessons learned from the assessment of reference buildings by different national experts help to point out commonalities and discrepancies among the approaches applied. This knowledge is used along with other results to develop a harmonized methodological guideline for the assessment of GHG emissions and other environmental impacts in the full life cycle within the international research project IEA EBC Annex 72. 7 References [1] EEA 2013 Environmental pressures from European Consumption and Production: A study in integrated environmental and economic analysis EEA Technical Report No 2/2013) [2] UNEP 2009 Buildings and climate change: a summary for decision-makers (Paris, France: UNEP SBCI (Sustainable Buildings & Climate Initiative)) [3] Röck M, Saade M R M, Balouktsi M, Rasmussen F N, Birgisdottir H, Frischknecht R, Habert G, Lützkendorf T and Passer A 2019 Embodied GHG emissions of buildings – The hidden challenge for effective climate change mitigation. Applied Energy 114107 [4] Frischknecht R, Birgisdottir H, Chae C U, Lützkendorf T and Passer A 2019 IEA EBC Annex 72 - Assessing life cycle related environmental impacts caused by buildings – targets and tasks. In: Sustainable built environment D-A-CH conference 2019, (Graz, Austria: IOP Conference Series: Earth and Environmental Science 323 012042) [5] Frischknecht R, Birgisdottir H, Chae C U, Lützkendorf T, Passer A, Alsema E, Balouktsi M, Berg B, Dowdell D, García Martínez A, Habert G, Hollberg A, König H, Lasvaux S, Llatas Olive C, Nygaard Rasmussen F, Peuportier B, Ramseier L, Röck M, Soust Verdaguer B, Szalay Z, Bohne R A, Bragança L, Cellura M, Chau C K, Dixit M, Francart N, Gomes V, Huang L, Longo S, Lupíšek A, Martel J, Mateus R, Ouellet-Plamondon C, Pomponi F, Ryklová P, Trigaux D and Yang W 2019 Comparison of the environmental assessment of an identical office building with national methods. In: Sustainable built environment D-A-CH conference 2019, (Graz, Austria: IOP Conference Series: Earth and Environmental Science 323 012037) [6] Deutsche Bundesministerium des Innern, für Bau und Heimat, 2018 Ökobau data 20 Germany: Deutsche Bundesministerium des Innern, für Bau und Heimat) [7] ecoinvent Centre 2018 ecoinvent data v3.5 (Zürich, Switzerland: ecoinvent Association) [8] ecoinvent Centre 2017 ecoinvent data v3.4 (Zürich, Switzerland: ecoinvent Association) [9] Deutsche Bundesministerium des Innern, für Bau und Heimat, 2016 Ökobau data 2016 (Berlin, Germany: Deutsche Bundesministerium des Innern, für Bau und Heimat) [10] ecoinvent Centre 2016 ecoinvent data v3.3 (Zürich, Switzerland: ecoinvent Association) [11] Vonka M et al. 2011 Metodika SBToolCZ: manuál hodnocení administrativních budov ve fázi návrhu [SBToolCZ Methodology: Manual of Administrative Buildings at Design Phase] (Prague: CIDEAS - Centrum integrovaného navrhování progresivních stavebních konstrukcí) ( g g p g ) [12] ecoinvent Centre 2007 ecoinvent data v2.01, ecoinvent reports No. 1-25 (Duebendorf, Switzerland: Swiss Centre for Life Cycle Inventories) [13] ecoinvent Centre 2010 ecoinvent data v2.2, ecoinvent reports No. Acknowledgments g The authors would like to thank the IEA for its organizational support and the colleagues involved in the project (Members: Australia, Austria, Belgium, Canada, Czech Republic, P.R. China, Denmark, Finland, France, Germany, Italy, R. Korea, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, United States of America, Observers: Brazil, Hungary, India, Slovenia) for their cooperation. Additionally, they would like to thank the organizations and institutions in the participating countries for their financial support. 8 8 BEYOND 2020 – World Sustainable Built Environment conference BEYOND 2020 – World Sustainable Built Environment conference IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 IOP Publishing doi:10.1088/1755-1315/588/2/022029 IOP Conf. Series: Earth and Environmental Science 588 (2020) 022029 g doi:10.1088/1755-1315/588/2/022029 References 1-25 (Duebendorf, Switzerland: Swiss Centre for Life Cycle Inventories) ecoinvent Centre 2013 ecoinvent data v3.0 (Zürich, Switzerland: ecoinvent association) i C i d ( i h S i l d i i i ) [15] ecoinvent Centre 2014 ecoinvent data v3.1 (Zürich, Switzerland: ecoinvent association) [16] Bragança L and Mateus R 2012 Life-cycle analysis of buildings: environmental impact of building elements (iiSBE Portugal: ISBN 978-989-96543-3-4. [http://hdl.handle.net/1822/20481]) [ p ]) [17] ecoinvent Centre 2009 ecoinvent data v2.1, ecoinvent reports No. 1-25 (Duebendorf, Switzerland: Swiss Centre for Life Cycle Inventories) [18] IVL Swedish Environmental Institute 2018 Byggsektorns Miljöberäkningsverktyg (Swedish Building Sector Environmental Calculation Tool). Retrieved May 15, 2019, from https://www.ivl.se/sidor/vara-omraden/miljodata/byggsektorns-miljoberakningsverktyg.html) p j ygg j g yg ) [19] EN 15804 2013 EN 15804:2012+A1:2013 - Sustainability of construction works - Environmental product declarations - Core rules for the product category of construction products (Brussels: European Committee for Standardisation (CEN)) [20] EN 15978 2011 EN 15978:2011 - Sustainability of construction works - Assessment of environmental performance of buildings - Calculation method (Brussels: European Committee for Standardisation (CEN)) 9
https://openalex.org/W3022269645	https://europepmc.org/articles/pmc7205235?pdf=render	English	null	Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation on metabolic and inflammatory adverse effects induced by high-fat diet in a mouse model of obesity	PloS one	2,020	cc-by	9,678	Editor: Michele Vacca, University of Cambridge, UNITED KINGDOM Editor: Michele Vacca, University of Cambridge, UNITED KINGDOM Received: October 4, 2019 Accepted: April 18, 2020 Published: May 7, 2020 Copyright: © 2020 Ryyti et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Editor: Michele Vacca, University of Cambridge, UNITED KINGDOM Received: October 4, 2019 Accepted: April 18, 2020 Published: May 7, 2020 Copyright: © 2020 Ryyti et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: © 2020 Ryyti et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper. PLOS ONE PLOS ONE RESEARCH ARTICLE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation on metabolic and inflammatory adverse effects induced by high-fat diet in a mouse model of obesity Riitta RyytiID1, Mari Ha¨ma¨la¨inen1, Rainer Peltola2, Eeva MoilanenID1* 1 The Immunopharmacology Research Group, Faculty of Medicine and Health Technology, Tampere University and Tampere University Hospital, Tampere, Finland, 2 Natural Resources Institute Finland, Bioeconomy and environment, Rovaniemi, Finland a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * eeva.moilanen@tuni.fi * eeva.moilanen@tuni.fi * eeva.moilanen@tuni.fi OPEN ACCESS Citation: Ryyti R, Ha¨ma¨la¨inen M, Peltola R, Moilanen E (2020) Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation on metabolic and inflammatory adverse effects induced by high-fat diet in a mouse model of obesity. PLoS ONE 15(5): e0232605. https://doi. org/10.1371/journal.pone.0232605 Abstract Obesity is a constantly increasing health problem worldwide. It is associated with a sys- temic low-grade inflammation, which contributes to the development of metabolic disor- ders and comorbidities such as type 2 diabetes. Diet has an important role in the prevention of obesity and its adverse health effects; as a part of healthy diet, polyphenol- rich berries, such as lingonberry (Vaccinium vitis-idaea L.) have been proposed to have health-promoting effects. In the present study, we investigated the effects of lingonberry supplementation on high-fat diet induced metabolic and inflammatory changes in a mouse model of obesity. Thirty male C57BL/6N mice were divided into three groups (n = 10/ group) to receive low-fat (LF), high-fat (HF) and lingonberry-supplemented high-fat (HF +LGB) diet for six weeks. Low-fat and high-fat diet contained 10% and 46% of energy from fat, respectively. Lingonberry supplementation prevented the high-fat diet induced adverse changes in blood cholesterol and glucose levels and had a moderate effect on the weight and visceral fat gain, which were 26% and 25% lower, respectively, in the lingonberry group than in the high-fat diet control group. Interestingly, lingonberry supplementation also restrained the high-fat diet induced increases in the circulating levels of the proinflam- matory adipocytokine leptin (by 36%) and the inflammatory acute phase reactant serum amyloid A (SAA; by 85%). Similar beneficial effects were discovered in the hepatic expres- sion of the inflammatory factors CXCL-14, S100A10 and SAA by lingonberry supplementa- tion. In conclusion, the present results indicate that lingonberry supplementation significantly prevents high-fat diet induced metabolic and inflammatory changes in a murine model of obesity. The results encourage evaluation of lingonberries as a part of healthy diet against obesity and its comorbidities. 1. Introduction The prevalence of obesity, metabolic syndrome and type 2 diabetes has increased rapidly worldwide during the last decades. Reasons can be found in changing lifestyles, which lead to reduced physical activity and obesogenic diet. It seems that obesity is continuing to be an increasing global health burden; according to the WHO statistics, 13% of adults aged 18 and over (corresponding to over 650 million people) were obese (body mass index BMI > 30 kg/ m2) and 39% overweight in 2016 [1]. Obesity is a complex chronic disorder with a multifacto- rial etiology, involving genetics, hormones, diet and environment [2], and it has a major impact on various metabolic and (patho)physiological functions in the human body. Obesity is a major risk factor and underlying condition in the progression of many metabolic disor- ders, particularly type 2 diabetes, cardiovascular diseases and cancer, through its effects on the development of, for instance hypertension, insulin resistance, nonalcoholic fatty liver disease and inflammation [3,4]. Adipose tissue is an active tissue regulating various physiological and pathological pro- cesses, including immunity and inflammation. It is therefore no longer considered only as a passive energy storage. Adipose tissue produces and releases several hormone-like factors called adipokines, and many of them have pro- or anti-inflammatory properties [2,5]. In obese adipose tissue, immune cells secreting pro-inflammatory substances increase in number while those producing anti-inflammatory substances have been shown to decrease. This imbalance is responsible for the obesity induced low-grade inflammation and insulin resistance in the body [3]. Diet has an important role in the prevention and treatment of obesity, type 2 diabetes and other obesity-related diseases. Diets containing plenty of polyphenol-rich vegetables have been shown to lower the risk of obesity-related comorbidities [6–8]. Berries are specifically rich in various polyphenols [9,10]. Diets containing berries are associated with lowered risk of type 2 diabetes, probably due to the flavonoids [11], anthocyanidins [8,12,13] or other polyphenols [14] present in berries. Several intervention studies have shown beneficial effects of berries also on inflammation and cardiovascular diseases [15]. Lingonberry (Vaccinium vitis-idaea L.) has been reported to have promising health-benefi- cial effects and anti-inflammatory properties in experimental models [16–19]. Lingonberries / lingonberry extracts were found to exhibit antidiabetic potential in various in vitro tests [20] and beneficial metabolic effects in mice exposed to high-fat diet [21–25]. Competing interests: RR is an employee of Kiantama Ltd; she confirms, that her position has not altered her adherence to the PLOS ONE policies. She and the other authors declare no other competing interests. This does not alter our adherence to PLOS ONE policies on sharing data and materials. 1. Introduction Lingonberries are rich in dietary polyphenols with high antioxidant activities [16,26–28] and contain also essen- tial omega-3 fatty acids [29] and plant sterols [30,31], which may contribute to their health- promoting effects. Lingonberries are commonly consumed in the Nordic countries and com- mercially available in many different forms. They are also the most generally collected and commercially utilized wild berries in Finland [32]. In the present study, we aimed to investigate the effects of lingonberry supplementation on metabolic and inflammatory changes in high-fat diet induced experimental obesity in mice to extend the current understanding on the health benefits of lingonberries. As the test material, we used commercially available air-dried lingonberry powder made from Finnish lingonberries. Data Availability Statement: All relevant data are within the paper. Funding: The study was financially supported by a grant from the European Regional Development Fund (ERDF). Kiantama Ltd, Suomussalmi, Finland provided the lingonberry powder. The funders had no role in the study design, data collection and analysis, decision to publish or preparation of the manuscript. 1 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Nutrient content/100 g lingonberry powder: fat 0.8 g, carbohydrates 61 g (of which sugars 38 g), fiber 24 g, protein 3 g LF = low-fat diet, HF = high-fat diet, HF+LGB = lingonberry-supplemented high-fat diet 2. Materials and methods Animals and study design Male C57BL/6N mice (Scanbur Research A/S, Karlslunde, Denmark), 8 weeks of age and 24.3 ±0.2g of weight at the beginning of the experiment, were divided into three groups of 10 mice, 2 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity and housed two mice per cage in the animal facility of the Tampere University under standard conditions (12/12h light/dark cycle, 22±1 ˚C temperature, and 50–60% humidity) with food and water provided ad libitum. The mice were fed with normal low-fat diet (LF, 10 kcal% fat), with high-fat diet (HF, 46 kcal% fat) or with high-fat diet supplemented with air-dried lingonberry (Vaccinium vitis- idaea L.) powder (HF + LGB, 20% w/w) for 6 weeks. Both high-fat diets contained 46% of energy from fat and 36% from carbohydrate, while the low-fat diet had 10% of energy from fat and 72% from carbohydrate (Table 1). Otherwise the custom-made pelleted diets (Research Diets, Inc, New Brunswick, NJ, USA) were matched for protein (18% of energy from protein), fiber, vitamin and trace element contents considering the composition of the lingonberry pow- der. Air-dried lingonberry powder (100 g powder corresponds to ca 900 g fresh berries) was produced from Finnish lingonberries by Kiantama Ltd (Suomussalmi, Finland). Table 1. Composition of the experimental diets. LF HF HF+LGB Calculated energy (kcal) Protein 716 716 716 Carbohydrate 2840 1422 1422 Starch 2110 691 691 Sugar 730 731 731 Fat 405 1823 1823 Total energy 3961 3961 3961 Calculated energy per gram diet (kcal/g) 3.60 4.39 4.30 Calculated Energy (kcal%) Protein 18 18 18 Carbohydrate 72 36 36 Fat 10 46 46 Fiber (g%) 9 10 10 Lingonberry powder (g) 0 0 184 Ingredients (g) (+ from LGB powder) Casein 200 200 194 (+ 6) total: 200 L-Cystine 3 3 3 Corn Starch 452 73 31 (+ 42) total: 73 Maltodextrin 10 75 100 100 Sucrose 173 173 103 (+ 70) total: 173 Cellulose 94 94 50 (+ 44) total: 94 Soybean Oil 25 25 24 (+ 1) total: 25 Lard 20 178 178 Mineral Mix S10026 10 10 10 DiCalcium Phosphate 13 13 13 Calcium Carbonate 6 6 6 Potassium Citrate 17 17 17 Vitamin Mix V10001 10 10 10 Choline Bitartrate 2 2 2 Table 1. Composition of the experimental diets. Table 1. Composition of the experimental diets. https://doi.org/10.1371/journal.pone.0232605.t001 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 Blood samples and analyses Six-hour fasting (morning fast) blood glucose levels in mice were measured from the tip of the tail with Contour Next One (Oy Diabet Ab, Lemu, Finland). Blood collected by cardiac punc- ture was centrifuged for 15 minutes at 1500 x g after 30 minutes incubation at room tempera- ture, and obtained serum was immediately storaged at -80 ˚C. Serum triglyceride and total cholesterol levels, and alanine aminotransferase (ALT) activity were measured by fluorometric assays (Abcam, Cambridge, UK). Enzyme-linked immunoassays were used to measure the concentrations of leptin, resistin and adiponectin (R&D Systems Europe Ltd., Abingdon, UK), insulin (Mercodia Ltd., Uppsala, Sweden) and serum amyloid A (Tridelta Development Ltd., Maynooth, Ireland) in serum samples. Detection limits were 7.8 pg/mL for leptin and resistin, 15.6 pg/mL for adiponectin, 33 pmol/L for insulin and 16 ng/mL for SAA. 2. Materials and methods Animals and study design 3 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Body weight of the mice and food consumption was monitored weekly. At the end of the study, 6h-fasted mice were anesthetized with isoflurane (Oriola Corp., Espoo, Finland), blood glucose was measured and blood was collected by cardiac puncture. Tissue samples were col- lected for further analyses. The study was approved by the National Animal Experimental Board (permission number ESAVI- 984/04.10.07/2018) and the experiments were carried out in accordance with the EU legislation for the protection of animals used for scientific purposes (Directive 2010/63/EU). RNA extraction and qRT-PCR Total RNA was extracted from liver using RNeasy Mini Kit (Qiagen Inc., Hilden, Germany). Briefly, samples stored immediately after collection in RNA Later1 (Ambion, Thermo Fisher Scientific, Waltham, MA, USA) were weighed and maximum of 30 mg of tissue was cut into smaller pieces and homogenized with Qiashredder (Qiagen). RNA was extracted with RNeasy Mini Kit with on-column DNase digestion. RNA was transcribed to cDNA by using Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) in 10 ul reaction volume and diluted 1:5 with RNase-free water. Quantitative PCR was performed using TaqMan Universal Master Mix and ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, CA, USA). The PCR cycling parameters were incubation at 50 ˚C for 2 minutes, incubation at 95 ˚C for 10 minutes, and thereafter 40 cycles of denaturation at 95 ˚C for 15 s and annealing and extension at 60 ˚C for 1 minute. Primers and probe for the housekeeping gene glyceralde- hyde 3-phosphate dehydrogenase GAPDH) were GCATGGCCTTCCGTGTTC (forward, 300 nM), GATGTCATCATACTTGGCAGGTTT (reverse, 300nM) and TCGTGGATCTGACGTG CCGCC (probe, 150 nM); and TCGGAGGCTTAATTACACATGTTC (forward, 900 nM), CAAGTGCATCATCGTTGTTCATAC (reverse, 300 nM) and CAGAATTGCCATTGCACAACT CTTTTCTCA (probe, 200 nM) for interleukin 6 (IL-6). The sequences and concentrations were optimized according to the manufacturer’s guidelines in TaqMan Universal PCR Master Mix Protocol part number 4304449 revision C (Applied Biosystems). TaqMan Gene Expres- sion assays for interleukin 1β (IL-1β, Mm00434228_m1), monocyte chemoattractant protein 1 (MCP-1, Mm00441242_m1), tumor necrosis alpha (TNF-α, Mm00443260_g1), glucose trans- porter 2 (GLUT2, Mm00446229_m1), serum amyloid A2 (SAA2, Mm04208126_mH), C-X-C motif chemokine ligand 14 (CXCL-14, Mm00444699_m1), S100 calcium-binding protein A10 (S100A10, Mm00501458_g1), and insulin receptor (Insr, Mm01211875_m1) were used (Thermo Fisher Scientific) and expression levels were calculated using the 2(−ΔΔCT) method. When calculating results, all of the mRNA expression levels were first normalized against GAPDH mRNA levels. 4 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Statistics Results are expressed as mean + standard error of mean (SEM). One-way and two-way ANOVA with Bonferroni’s post-test, and the analysis of covariance were performed using GraphPad InStat version 3.10 and GraphPad Prism 8 (GraphPad Software, San Diego, USA), and IBM SPSS Statistics version 25.0 (IBM Corporation, Armonk, NY, USA). Asterisks , , and indicate p values smaller than 0.05, 0.01 and 0.001, respectively. 3.1. Weight gain The weight of mice in the high-fat (HF) group increased considerably during the study as com- pared to the mice in the low-fat (LF) control group. Interestingly, lingonberry supplementa- tion prevented significantly the high-fat diet induced weight gain (p < 0.001 between HF and HF+LGB groups). After 6 weeks, the average weight of the low-fat group was 28.0±0.4 g, 37.4 ±0.6 g in the high-fat group and 34.1±0.6 g in the lingonberry supplemented high-fat group, respectively (p < 0.001 between groups). The development of weight of the mice in the test groups is presented in the Fig 1. The amount of epididymal fat increased in the high-fat diet group, when compared to the low-fat control group (p < 0.001). Lingonberry supplementation prevented significantly the accumulation of epididymal fat when compared to the high-fat control group (p < 0.001). At the end of the study, the amount of epididymal fat was 2.4±0.1 g in the high-fat group and 1.8 ±0.1 g in the lingonberry supplemented high-fat group. Both high-fat groups had significantly Fig 1. Body weight gain of the mice during the study. Animals received low-fat diet (LF diet, 10% of energy from fat, dark grey line), high-fat diet (HF diet, 46% of energy from fat, black line) or high-fat diet supplemented with lingonberry (HF + LGB diet, light grey line). Weight was measured once a week. The results are expressed as grams (g). Values represent mean + SEM, n = 10 mice per group. Repeated measures two-way ANOVA with Bonferroni post-test was used in the statistical analysis. Mean values significantly different from the high-fat group (HF diet) are marked with p < 0.05, p < 0.01 and p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g001 Fig 1. Body weight gain of the mice during the study. Animals received low-fat diet (LF diet, 10% of energy from fat, dark grey line), high-fat diet (HF diet, 46% of energy from fat, black line) or high-fat diet supplemented with lingonberry (HF + LGB diet, light grey line). Weight was measured once a week. The results are expressed as grams (g). Values represent mean + SEM, n = 10 mice per group. Repeated measures two-way ANOVA with Bonferroni post-test was used in the statistical analysis. Mean values significantly different from the high-fat group (HF diet) are marked with p < 0.05, p < 0.01 and p < 0.001. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 3.1. Weight gain htt //d i /10 1371/j l 0232605 001 https://doi.org/10.1371/journal.pone.0232605.g001 5 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 2. The amount of epididymal fat of the mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). The amount of epididymal fat was measured at the end of the study. The results are expressed as grams (g). Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g002 Fig 2. The amount of epididymal fat of the mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). The amount of epididymal fat was measured at the end of the study. The results are expressed as grams (g). Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g002 (p < 0.001) higher amount of epididymal fat than the low-fat group (0.9±0.0 g; Fig 2). When the ratio of the epididymal fat to the whole-body mass was calculated, it was lower in the HF +LGB group than in the HF group (p < 0.01) suggesting that lingonberry supplementation prevents particularly the accumulation of the metabolically highly detrimental visceral adipose tissue. That was also supported by the fact that in the analysis of covariance when the body weight was set as a covariate, the epididymal fat mass was lower in the LF and the HF+LGB groups than in the HF group (p < 0.01). Food consumption (kcal/g body weight) was measured weekly and it did not differ between the high-fat and the lingonberry supplemented high-fat diet groups although energy intake in the low-fat diet group was lower, particularly during the first half of the study (Fig 3A). 3.1. Weight gain We also calculated the cumulative food consumption (kcal/g body weight) during the six weeks’ study: no difference was found between the high-fat and the lingonberry supplemented high- fat groups while the value in the low-fat group was lower (p < 0.01; Fig 3B). This result was reproduced in repeated measures analysis of covariance with body weight as a covariate. 3.3. Cholesterol and triglycerides Cholesterol level was significantly increased in the high-fat group when compared to the low- fat group (p < 0.001) being 2.6±0.1 mmol/L in the high-fat group and 1.7±0.1 mmol/L in the low-fat group at the end of the study. Importantly, the cholesterol level in the lingonberry sup- plemented high-fat group (2.0±0.2 mmol/L) was significantly lower when compared to the high-fat group (p < 0.01). There was no statistically significant difference in the cholesterol levels between the low-fat and the lingonberry supplemented high-fat groups indicating that lingonberry supplementation prevented the high-fat diet induced increase in cholesterol levels (Fig 6A). There were no differences in the triglyceride levels between the low-fat and high-fat groups whereas the triglyceride level in the lingonberry supplemented high-fat group was lower than that in the two other groups (p < 0.05; Fig 6B). 3.2. Glucose and insulin As expected, fasting insulin level was significantly higher in the high-fat group (156.5 pmol/ L) as compared to the low-fat group (65.5 pmol/L; p < 0.001). In the lingonberry supple- mented high-fat group, the insulin level (113.7 pmol/L) was lower than in the high-fat group, although the effect did not reach statistical significance (Fig 5). high-fat control group (11.2±0.4 mmol/L; p < 0.05), and there was no statistically significant difference between the low-fat and the lingonberry supplemented high-fat groups (Fig 4). As expected, fasting insulin level was significantly higher in the high-fat group (156.5 pmol/ L) as compared to the low-fat group (65 5 pmol/L; p < 0 001) In the lingonberry supple- g y pp g g p ( g ) As expected, fasting insulin level was significantly higher in the high-fat group (156.5 pmol/ L) as compared to the low-fat group (65.5 pmol/L; p < 0.001). In the lingonberry supple- mented high-fat group, the insulin level (113.7 pmol/L) was lower than in the high-fat group, although the effect did not reach statistical significance (Fig 5). 3.2. Glucose and insulin Fasting blood glucose level at the end of the study was increased in the high-fat diet group, as compared to the low-fat control group (p < 0.01). Interestingly, in the lingonberry supple- mented high-fat diet group, the glucose level (10.0±0.5 mmol/L) was lower than that in the PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 6 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 3. Food consumption during the study. Fig 3A shows the weekly and Fig 3B the cumulative food consumption during the six weeks’ study. Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). Food consumption was measured once a week. The results are expressed as kcal/body weight in grams. Values represent mean + SEM, n = 10 mice per group; as the mice were housed two mice per cage, n = 5 was used in the statistical calculations. Repeated measures two-way ANOVA (Fig 3A) and one-way ANOVA (Fig 3B) with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p<0.01 and p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g003 Fig 3. Food consumption during the study. Fig 3A shows the weekly and Fig 3B the cumulative food consumption during the six weeks’ study. Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). Food consumption was measured once a week. The results are expressed as kcal/body weight in grams. Values represent mean + SEM, n = 10 mice per group; as the mice were housed two mice per cage, n = 5 was used in the statistical calculations. Repeated measures two-way ANOVA (Fig 3A) and one-way ANOVA (Fig 3B) with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p<0.01 and p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g003 high-fat control group (11.2±0.4 mmol/L; p < 0.05), and there was no statistically significant difference between the low-fat and the lingonberry supplemented high-fat groups (Fig 4). high-fat control group (11.2±0.4 mmol/L; p < 0.05), and there was no statistically significant difference between the low-fat and the lingonberry supplemented high-fat groups (Fig 4). PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 3.4. Adipokines and inflammatory factors As expected, leptin levels were significantly higher in the high-fat group as compared to the low-fat group (p < 0.001). Leptin levels were lower in the lingonberry supplemented high-fat group (27.1±3.5 ng/mL) than in the high-fat control group (39.7±2.8 ng/mL; p < 0.01, PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 7 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 4. The fasting blood glucose levels of mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). At the end of the study, blood samples for glucose measurements were collected after 6 h fasting. The results are expressed as mmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.05, p < 0.01 and ns = not significant. https://doi.org/10.1371/journal.pone.0232605.g004 Fig 4. The fasting blood glucose levels of mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). At the end of the study, blood samples for glucose measurements were collected after 6 h fasting. The results are expressed as mmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.05, p < 0.01 and ns = not significant. https://doi.org/10.1371/journal.pone.0232605.g004 https://doi.org/10.1371/journal.pone.0232605.g004 Table 2). Interestingly, the statistically significant difference was also observed in weight- related values (1.1±0.1 ng/mL/body weight (g) in HF group vs. 0.8±0.1 ng/mL/body weight (g) in HF+LGB group, p < 0.01). In adiponectin, there was a trend towards decreased levels in the high-fat group when compared to low-fat control group (p = 0.277). Adiponectin level was maintained at normal levels with lingonberry supplementation. There were no significant dif- ferences in the resistin levels between the groups (Table 2). PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 Fig 5. The insulin levels of mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). At the end of the study, blood samples were collected after 6 h fasting and serum insulin concentrations were analyzed with ELISA. The results are expressed as pmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g005 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 3.4. Adipokines and inflammatory factors The levels of the inflammatory acute phase reactant serum amyloid A (SAA) were increased (p < 0.001) in the high-fat diet group (11.7±0.7 μg/mL) as compared to the low-fat diet group (6.6±0.6 μg/mL). Importantly, the SAA levels in the lingonberry supplemented high-fat group (7.4±0.4 μg/mL) were significantly lower (p < 0.001) than those in the high-fat control group. Similarly, alanine aminotransferase activity was increased in the high-fat diet group as com- pared to the low-fat diet group (p<0.001) which was prevented by the lingonberry supplemen- tation (p < 0.001) (Table 2). High-fat diet is known to induce metabolic and inflammatory changes in the liver. There- fore, we analyzed the expression of insulin receptor and glucose transporter GLUT2 as well as the inflammatory factors TNF-α, IL-1β, IL-6, MCP-1, SAA2, CXCL-14 and S100A10 in the liver by quantitative RT-PCR. As shown in the Fig 7, the hepatic expression of CXCL-14 (p < 0.001), S100A10 (p < 0.05), and SAA2 (p < 0.01) were significantly lower in the lingon- berry supplemented high-fat diet group than in the high-fat control group, while IL-6 was under detection limit and no statistically significant differences in the expression of TNF-α, IL-1β, MCP-1, GLUT2 or insulin receptor were observed between the groups. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 8 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 5. The insulin levels of mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey column), high-fat diet (HF diet, 46% energy from fat, black column) or high-fat diet supplemented with lingonberry (HF + LGB diet, white column). At the end of the study, blood samples were collected after 6 h fasting and serum insulin concentrations were analyzed with ELISA. The results are expressed as pmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.001. Discussion In the present study, we found that lingonberry supplementation prevented high-fat diet induced adverse effects on blood cholesterol, glucose and insulin levels as well as visceral fat gain in a murine model of obesity. In addition, the circulating levels of the pro-inflammatory adipocytokine leptin and the inflammatory acute phase reactant and biomarker serum amy- loid A (SAA), as well as the alanine aminotransferase (ALT) activity were maintained at lower level by lingonberry supplementation, and the same was detected in hepatic expression of the inflammatory factors CXCL-14, S100A10 and SAA. To our knowledge, this is the first study in which such significant results have been obtained with an air-dried lingonberry powder. The results are remarkable also considering the rather short duration (6 weeks) of the study. Obesity is a risk factor for glucose intolerance leading to diabetes, which is first detected as increased circulating insulin concentrations followed by increased fasting blood glucose levels [33,34]. In the present study, increased insulin and glucose levels were found already after six weeks on high-fat diet. Lingonberry supplementation prevented these effects. These findings are supported by previous studies with freeze-dried lingonberries and lingonberry extract on a longer follow-up [21–23,25,35] while bilberry supplementation did not have any effect on PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 9 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 6. Cholesterol (A) and triglyceride (B) levels of mice at the end of the study. Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). At the end of the study, blood samples were collected after 6 h fasting and serum cholesterol and triglyceride concentrations were analyzed with fluorometric assays. The values are presented as mmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.05, p < 0.01, p < 0.001 and ns = not significant. Fig 6. Cholesterol (A) and triglyceride (B) levels of mice at the end of the study. Adiponectin, leptin, resistin and SAA concentrations were measured by ELISA, and ALT activity by fluorometric assay. The results are presented as mean ± SEM; n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Discussion Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). At the end of the study, blood samples were collected after 6 h fasting and serum cholesterol and triglyceride concentrations were analyzed with fluorometric assays. The values are presented as mmol/L. Values represent mean + SEM, n = 10 mice per group. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.05, p < 0.01, p < 0.001 and ns = not significant. https://doi.org/10.1371/journal.pone.0232605.g006 blood glucose levels [36]. Anthocyanins in the lingonberry powder could contribute to the beneficial effects on glucose tolerance as anthocyanin-rich supplements from Montmorency tart cherries have been reported to have insulin lowering effects in patients with metabolic syn- drome [37] and purified blueberry anthocyanins reduced blood glucose levels in a murine model of obesity [38]. In the present study, lingonberry supplementation did not alter hepatic expression of the glucose transporter GLUT2 or insulin receptor. Further studies are needed to understand the mechanisms which mediate the positive effects of lingonberries on glucose tolerance in obesity. Hyperlipidemia is another characteristic feature of obesity associated metabolic syndrome [4]. In the present study, lingonberry supplementation normalized the detrimental effects of the high-fat diet on the cholesterol levels, and triglycerides were also reduced. Proanthocyani- dins present in lingonberries may explain, at least partly, the favorable effects on cholesterol levels [39–41]. In addition, the purified lingonberry anthocyanidin compound cyanidin-3-O- Table 2. The circulating adipokine and serum amyloid A (SAA) levels, and alanine aminotransferase (ALT) activity of the mice at the end of the study. Low-fat (LF) High-fat (HF) High-fat + lingonberry (HF+LGB) p-value between LF and HF p-value between HF and HF+LGB Adiponectin (μg/mL) 6.9 ± 0.2 6.4 ± 0.2 7.1 ± 0.3 0.277 0.100 Leptin (ng/mL) 4.6 ± 0.6 39.7 ± 2.8 27.1 ± 3.5 < 0.001 < 0.01 Resistin (ng/mL) 17.8 ± 1.4 16.8 ± 0.8 17.6 ± 0.9 > 0.999 > 0.999 SAA (μg/mL) 6.6 ± 0.6 11.7 ± 0.7 7.4 ± 0.4 < 0.001 < 0.001 ALT (U/L) 8.1 ± 0.6 14.1 ± 0.8 7.2 ± 0.2 < 0.001 < 0.001 ine and serum amyloid A (SAA) levels, and alanine aminotransferase (ALT) activity of the mice at the end of the study. Discussion The lipid lowering effect seems to be characteristic for lingonberry among berries since freeze-dried blueberry and black raspberry [44] as well blackcurrant [21] were reported to enhance serum lipids in high-fat diet fed mice, while our findings are sup- ported by previous studies with lingonberry containing diets [21,35,45]. High-fat treatment normally results in increased circulating triglycerides in humans [46], but in mice their levels may stay unchanged or decrease [21,38,45,47–49], and the former was found also in the pres- ent study. These effects on triglycerides may result from increased clearance of triglycerides and/or reduced production of triglycerides in the liver [47–49]. Increased ALT activity in the serum, as was also shown in the current study, supports the former assumption likewise associ- ated with the development of nonalcoholic fatty liver disorder [50]. Adipokines are cytokine-like hormones originally found to be produced by adipose tissue and to regulate energy metabolism and appetite [51,52]. Leptin is a prototype adipokine, the concentrations of which are closely related to the amount of adipose tissue and body mass index [53,54]. This is the first study showing reduced leptin levels during lingonberry supple- mentation of high-fat diet but the results are supported by studies using blueberry anthocya- nins [38,44]. It is not clear if this effect is simply a consequence of the moderate weight gain restraining effect of lingonberry supplementation or whether it has mechanistic relevancy. Anyhow, reduced leptin level may be considered as a beneficial effect because leptin has pro- inflammatory effects and contributes to the development of co-morbidities of obesity [55]. In addition, an increasing trend in adiponectin levels was seen in mice on lingonberry supple- mented high-fat diet. Since adiponectin has insulin sensitizing effects [4], this is a positive sig- nal and motivates further studies. Obesity is associated with low-grade inflammation, which contributes to the obesity-related metabolic diseases. Serum amyloid A (SAA) is an example of the biomarkers and effectors of obesity-related inflammation, and it is involved in the mechanisms of lipid metabolism, ath- erosclerosis and acute phase response besides inflammation [56,57]. In the present study, cir- culating SAA concentrations, as well as its expression in the liver, were substantially increased during high-fat diet while lingonberry supplementation prevented these changes in a statisti- cally significant manner. In addition, lingonberry supplementation reduced the hepatic expression of two other pro-inflammatory factors, i.e. the chemokine CXCL-14 and the S100 calcium-binding protein A10. Discussion The circulating adipokine and serum amyloid A (SAA) levels, and alanine aminotransferase (ALT) activity of the mice a PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 10 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity Fig 7. The expression of C-X-C motif chemokine ligand 14 (CXCL-14, A), S100 calcium-binding protein A10 (S100A10, B), serum amyloid A2 (SAA2, C), tumor necrosis alpha (TNF-α, D), interleukin 1β (IL-1β, E), monocyte chemoattractant protein 1 (MCP-1, F), glucose transporter 2 (GLUT2, G), and insulin receptor (H) i the liver. Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). A Fig 7. The expression of C-X-C motif chemokine ligand 14 (CXCL-14, A), S100 calcium-binding protein A10 (S100A10, B), serum amyloid A2 (SAA2, C), tumor necrosis alpha (TNF-α, D), interleukin 1β (IL-1β, E), monocyte chemoattractant protein 1 (MCP-1, F), glucose transporter 2 (GLUT2, G), and insulin receptor (H) in the liver. Animals received low-fat diet (LF diet, 10% energy from fat, grey columns), high-fat diet (HF diet, 46% energy from fat, black columns) or high-fat diet supplemented with lingonberry (HF + LGB diet, white columns). At PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 11 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity the end of the study, liver samples were collected after 6 h fasting and the expression of the genes of interest were measured with RT-PCR. After normalizing to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH), the expression level in each sample was compared to the mean expression level in the LF-group, which was set as 1. One-way ANOVA with Bonferroni post-test was used in the statistical analysis. Differences between the groups are marked with p < 0.05, p < 0.01 and p < 0.001. https://doi.org/10.1371/journal.pone.0232605.g007 β-glucoside was found effective in decreasing total cholesterol levels in hypercholesterolemic ApoE deficient mice [42]. Lingonberries are also known to contain plant sterols at moderate levels of 20–30 mg/100 g of fresh weight [30,31] and they may contribute to the lipid lowering effects of lingonberries [43]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 Discussion These data indicate that lingonberry supplementation indeed suppresses the obesity-related inflammatory response. Based on the present data, it remains unclear, which factor(s) present in the lingonberry powder may have caused the beneficial effects found in the current study, and further studies are needed to reveal the active constituents. The effects of lingonberry supplementation on the weight gain were moderate while it significantly prevented many of the metabolic and inflam- matory adverse effects induced by the high-fat diet. Therefore, it is possible that lingonberry supplementation in the high-fat containing diet has anti-inflammatory consequences and shifts endogenous responses towards metabolically healthier obesity. Lingonberries contain PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 12 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity several polyphenolic compounds, which have anti-inflammatory potential such as anthocya- nins, ellagitannins, flavonols, phenolic acids and proanthocyanidins [58], flavonols and proanthocyanidins as well as anthocyanins forming the largest groups among the lingonberry polyphenols [58–62]. Quercetin is abundantly present in lingonberries [63]. Quercetin has been reported to have anti-inflammatory effects in in vitro and in vivo models [64–66]. Moreover, quercetin has also been suggested to have anti-adiposity activity in vitro by activating the AMPK signal pathway in adipocytes [67] and in vivo by preventing high-fat diet induced obesity in mice [68]. Lingon- berries are also a rich source of proanthocyanidins [60,69] and anthocyanins, particularly cya- nidin-3-galactoside, cyanidin-3-arabinoside, and cyanidin-3-glucoside [70,71]. Proanthocyanidins have been reported to contribute to the reduced weight gain and hypolipi- demic effects of berry-derived preparations in animal models of obesity [40,72,73] while puri- fied anthocyanins from mulberry (M. australis P.) and cherry (Prunus avium L.) prevented obesogenic diet induced weight gain, oxidative stress and inflammation in mice [74,75]. Ling- onberry contains also the stilbenoid compound resveratrol (together with its glycosylated derivative piceid), which is a strong antioxidant with anti-inflammatory properties [76–78]. We have recently shown that resveratrol inhibits the production of nitric oxide and pro- inflammatory cytokines in activated macrophages and suppresses inflammatory responses in vivo [79]. Resveratrol has also been beneficial in animal models of obesity [80–82], and in a clinical study resveratrol supplementation improved glycemic control in patients with type 2 diabetes [83]. In conclusion, the present results show that lingonberry supplementation significantly pre- vents high-fat diet induced metabolic and inflammatory changes in a murine model of obesity. Acknowledgments We wish to acknowledge Meiju Kukkonen and Salla Hietakangas for excellent technical assistance. Discussion Further studies are needed to reveal the detailed molecular mechanisms and the active constit- uents responsible for the observed beneficial effects. The results encourage evaluation of ling- onberries as a part of healthy diet against obesity and its comorbidities. Conceptualization: Eeva Moilanen. Funding acquisition: Eeva Moilanen. Investigation: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Rainer Peltola, Eeva Moilanen. Methodology: Mari Ha¨ma¨la¨inen, Eeva Moilanen. Methodology: Mari Ha¨ma¨la¨inen, Eeva Moilanen. Project administration: Eeva Moilanen. Supervision: Mari Ha¨ma¨la¨inen, Eeva Moilanen. Visualization: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Eeva Moilanen. Visualization: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Eeva Moilanen. Writing – original draft: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Rainer Peltola, Eeva Moilanen. Writing – original draft: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Rainer Peltola, Eeva Moilanen. Writing – review & editing: Riitta Ryyti, Mari Ha¨ma¨la¨inen, Rainer Peltola, Eeva Moilanen. Author Contributions Conceptualization: Eeva Moilanen. References 1. World Health Organization. Obesity and overweight. 2018. 13 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity 2. Lee H, Lee IS, Choue R. Obesity, Inflammation and Diet. Pediatric Gastroenterology, Hepatology & Nutrition. 2013; 16: 143–152. 2. Lee H, Lee IS, Choue R. Obesity, Inflammation and Diet. Pediatric Gastroenterology, Hepatology & Nutrition. 2013; 16: 143–152. 3. Asghar A, Sheikh N. Role of immune cells in obesity induced low grade inflammation and insulin resis- tance. Cell Immunol. 2016; 315: 18–26. 4. Jung UJ, Choi M. Obesity and its metabolic complications: The role of adipokines and the relationship between obesity, inflammation, insulin resistance, dyslipidemia and nonalcoholic fatty liver disease. International Journal of Molecular Sciences. 2014; 15: 6184–6223. https://doi.org/10.3390/ ijms15046184 PMID: 24733068 5. Mraz M, Haluzik M. The role of adipose tissue immune cells in obesity and low-grade inflammation. J Endocrinol. 2014; 222: 113–127. 6. Dembinska-Kiec A, Mykka¨nen O, Kiec-Wilk B, Mykka¨nen H. Antioxidant phytochemicals against type 2 diabetes. Br J Nutr. 2008; 99: ES109–ES117. https://doi.org/10.1017/S000711450896579X PMID: 18503731 7. Pan M, Lai C, Ho C. Anti-inflammatory activity of natural dietary flavonoids. Food and Function. 2010; 1: 15–31. https://doi.org/10.1039/c0fo00103a PMID: 21776454 8. Williamson G. The role of polyphenols in modern nutrition. Nutr Bull. 2017; 42: 226–235. https://doi.org/ 10.1111/nbu.12278 PMID: 28983192 9. Szajdek A, Borowska EJ. Bioactive Compounds and Health-Promoting Properties of Berry Fruits: A Review. Plant Foods for Human Nutrition. 2008; 63: 147–156. https://doi.org/10.1007/s11130-008- 0097-5 PMID: 18931913 10. Ha¨kkinen SH, Ka¨renlampi SO, Heinonen IM, Mykka¨nen HM, To¨rronen AR. Content of the flavonols quercetin, myricetin, and kaempferol in 25 edible berries. J Agric Food Chem. 1999; 47: 2274–2279. https://doi.org/10.1021/jf9811065 PMID: 10794622 11. Knekt P, Kumpulainen J, Ja¨rvinen R, Rissanen H, Helio¨vaara M, Reunanen A, et al. Flavonoid intake and risk of chronic diseases. Am J Clin Nutr. 2002; 76: 560–568. https://doi.org/10.1093/ajcn/76.3.560 PMID: 12198000 12. Nile SH, Park SW. Edible berries: Bioactive components and their effect on human health. Nutrition. 2014; 30: 134–144. https://doi.org/10.1016/j.nut.2013.04.007 PMID: 24012283 13. Kolehmainen M, Mykka¨nen O, Kirjavainen PV, Leppa¨nen T, Moilanen E, Adriaens M, et al. Bilberries reduce low-grade inflammation in individuals with features of metabolic syndrome. Molecular Nutrition and Food Research. 2012; 56: 1501–1510. https://doi.org/10.1002/mnfr.201200195 PMID: 22961907 14. Mursu J, Virtanen JK, Tuomainen TP, Nurmi T, Voutilainen S. References Arctic berry extracts target the gut– liver axis to alleviate metabolic endotoxaemia, insulin resistance and hepatic steatosis in diet-induced obese mice. Diabetologia. 2018; 61: 919–931. https://doi.org/10.1007/s00125-017-4520-z PMID: 29270816 25. Marungruang N, Kovalenko T, Osadchenko I, Voss U, Huang F, Burleigh S, et al. Lingonberries and their two separated fractions differently alter the gut microbiota, improve metabolic functions, reduce gut inflammatory properties, and improve brain function in ApoE−/−mice fed high-fat diet. Nutr Neu- rosci. 2018: 1–13. 26. Bhullar KS, Rupasinghe HPV. Antioxidant and cytoprotective properties of partridgeberry polyphenols. Food Chem. 2015; 168: 595–605. https://doi.org/10.1016/j.foodchem.2014.07.103 PMID: 25172753 27. Zheng W, Wang SY. Oxygen radical absorbing capacity of phenolics in blueberries, cranberries, choke- berries, and lingonberries. J Agric Food Chem. 2003; 51: 502–509. https://doi.org/10.1021/jf020728u PMID: 12517117 28. Ehala S, Vaher M, Kaljurand M. Characterization of phenolic profiles of Northern European berries by capillary electrophoresis and determination of their antioxidant activity. J Agric Food Chem. 2005; 53: 6484–6490. https://doi.org/10.1021/jf050397w PMID: 16076138 29. Bere E. Wild berries: a good source of omega-3. Eur J Clin Nutr. 2007; 61: 431–433. https://doi.org/10. 1038/sj.ejcn.1602512 PMID: 16900081 30. Piironen V, Toivo J, Lampi A-M. Natural Sources of Dietary Plant Sterols. Journal of Food Composition and Analysis. 2000; 13: 619–624. 31. Szakiel A, Pączkowski C, Koivuniemi H, Huttunen S. Comparison of the triterpenoid content of berries and leaves of lingonberry Vaccinium vitis-idaea from Finland and Poland. J Agric Food Chem. 2012; 60: 4994–5002. https://doi.org/10.1021/jf300375b PMID: 22490120 32. Turtiainen M, Salo K, Saastamoinen O. Variations of yield and utilisation of bilberries (Vaccinium myrtil- lus L.) and cowberries (V. vitis-idaea L.) in Finland. Silva Fenn. 2011; 45: 237–251. 33. Blaak EE, Antoine J, Benton D, Bjo¨rck I, Bozzetto L, Brouns F, et al. Impact of postprandial glycaemia on health and prevention of disease: Postprandial glycaemia and health. Obesity Reviews. 2012; 13: 923–984. https://doi.org/10.1111/j.1467-789X.2012.01011.x PMID: 22780564 34. Boucher J, Kleinridders A, Kahn CR. Insulin Receptor Signaling in Normal and Insulin-Resistant States. Cold Spring Harbor perspectives in biology. 2014; 6: a009191. https://doi.org/10.1101/cshperspect. a009191 PMID: 24384568 35. Anhê FF, Varin TV, Le Barz M, Pilon G, Dudonne´ S, Trottier J, et al. Arctic berry extracts target the gut– liver axis to alleviate metabolic endotoxaemia, insulin resistance and hepatic steatosis in diet-induced obese mice. Diabetologia. 2018; 61: 919–931. https://doi.org/10.1007/s00125-017-4520-z PMID: 29270816 36. Mykka¨nen OT, Huotari A, Herzig K, Dunlop TW, Mykka¨nen H, Kirjavainen PV. References Intake of fruit, berries, and vegetables and risk of type 2 diabetes in Finnish men: the Kuopio Ischaemic Heart Disease Risk Factor Study. Am J Clin Nutr. 2014; 99: 328–333. https://doi.org/10.3945/ajcn.113.069641 PMID: 24257723 15. Basu A, Rhone M, Lyons TJ. Berries: Emerging impact on cardiovascular health. Nutr Rev. 2010; 68: 168–177. https://doi.org/10.1111/j.1753-4887.2010.00273.x PMID: 20384847 16. Drozdz P, Seziene V, Wojcik J, Pyrzynska K. Evaluation of Bioactive Compounds, Minerals and Antioxi- dant Activity of Lingonberry (Vaccinium vitis-idaea L.) Fruits. Molecules. 2017; 23: 53. 17. Kowalska K, Olejnik A. Current evidence on the health-beneficial effects of berry fruits in the prevention and treatment of metabolic syndrome. Curr Opin Clin Nutr Metab Care. 2016; 19: 446–452. https://doi. org/10.1097/MCO.0000000000000322 PMID: 27583706 18. Mane C, Loonis M, Juhel C, Dufour C, Malien-Aubert C. Food grade lingonberry extract: Polyphenolic composition and in vivo protective effect against oxidative stress. J Agric Food Chem. 2011; 59: 3330– 3339. https://doi.org/10.1021/jf103965b PMID: 21375302 19. Kivima¨ki AS, Siltari A, Ehlers PI, Korpela R, Vapaatalo H. Lingonberry juice negates the effects of a high salt diet on vascular function and low-grade inflammation. Journal of Functional Foods. 2014; 7: 238–245. 20. Harbilas D, Martineau LC, Harris CS, Adeyiwola-Spoor DCA, Saleem A, Lambert J, et al. Evaluation of the antidiabetic potential of selected medicinal plant extracts from the Canadian boreal forest used to treat symptoms of diabetes: part II. Can J Physiol Pharmacol. 2009; 87: 479–492. https://doi.org/10. 1139/y09-029 PMID: 19526043 21. Heyman L, Axling U, Blanco N, Sterner O, Holm C, Berger K, et al. Evaluation of beneficial metabolic effects of berries in high-fat fed C57BL/6J mice. Journal of Nutrition and Metabolism. 2014; 2014: 1–12. 22. Heyman-Linde´n L, Kotowska D, Sand E, Bjursell M, Plaza M, Turner C, et al. Lingonberries alter the gut microbiota and prevent low-grade inflammation in high-fat diet fed mice. Food & nutrition research. 2016; 60: 29993–14. 23. Zhang Z, Zhou Q, Huangfu G, Wu Y, Zhang J. Anthocyanin extracts of lingonberry (Vaccinium vitis- idaea L.) attenuate serum lipids and cholesterol metabolism in HCD-induced hypercholesterolaemic male mice. International Journal of Food Science and Technology. 2018. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 14 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity 24. Anhê FF, Varin TV, Le Barz M, Pilon G, Dudonne´ S, Trottier J, et al. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 References Wild blueberries (Vacci- nium myrtillus) alleviate inflammation and hypertension associated with developing obesity in mice fed with a high-fat diet. PLoS ONE. 2014; 9: e114790. https://doi.org/10.1371/journal.pone.0114790 PMID: 25501421 37. Desai T, Roberts M, Bottoms L. Effects of Montmorency tart cherry supplementation on cardio-meta- bolic markers in metabolic syndrome participants: A pilot study. Journal of Functional Foods. 2019; 57: 286–298. 38. Prior RL, E. Wilkes S, R. Rogers T, Khanal RC, Wu X, Howard LR. Purified blueberry anthocyanins and blueberry juice alter development of obesity in mice fed an obesogenic high-fat diet. J Agric Food Chem. 2010; 58: 3970–3976. https://doi.org/10.1021/jf902852d PMID: 20148514 39. Rasmussen SE, Frederiksen H, Krogholm KS, Poulsen L. Dietary proanthocyanidins: Occurrence, die- tary intake, bioavailability, and protection against cardiovascular disease. Molecular Nutrition and Food Research. 2005; 49: 159–174. https://doi.org/10.1002/mnfr.200400082 PMID: 15635686 40. Blade´ C, Arola L, Salvado´ M. Hypolipidemic effects of proanthocyanidins and their underlying biochemi- cal and molecular mechanisms. Molecular Nutrition and Food Research. 2010; 54: 37–59. https://doi. org/10.1002/mnfr.200900476 PMID: 19960459 41. Razavi S, Gholamin S, Eskandari A, Mohsenian N, Ghorbanihaghjo A, Delazar A, et al. Red grape seed extract improves lipid profiles and decreases oxidized low-density lipoprotein in patients with mild hyper- lipidemia. Journal of Medicinal Food. 2013; 16: 255–258. https://doi.org/10.1089/jmf.2012.2408 PMID: 23437789 42. Wang DL, Xia M, Gao S, Li D, Zhang Y, Jin TR, et al. Cyanidin-3-O-ss-glucoside upregulates hepatic cholesterol 7a-hydroxylase expression and reduces hypercholesterolemia in mice. Molecular Nutrition and Food Research. 2012; 56: 610–621. https://doi.org/10.1002/mnfr.201100659 PMID: 22495986 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 15 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity 43. AbuMweis SS, Jones PJH. Cholesterol-lowering effect of plant sterols. Curr Atheroscler Rep. 2008; 10: 467–472. https://doi.org/10.1007/s11883-008-0073-4 PMID: 18937893 44. Prior RL, Wu X, Gu L, Hager T, Hager A, Wilkes S, et al. Purified berry anthocyanins but not whole ber- ries normalize lipid parameters in mice fed an obesogenic high fat diet. Molecular Nutrition and Food Research. 2009; 53: 1406–1418. https://doi.org/10.1002/mnfr.200900026 PMID: 19743407 45. Eid HM, Ouchfoun M, Brault A, Vallerand D, Musallam L, Arnason JT, et al. Lingonberry (Vaccinium vitis-idaea L.) exhibits antidiabetic activities in a mouse model of diet-induced obesity. Evidence-based Complementary and Alternative Medicine. 2014; 2014: 645812–10. https://doi.org/10.1155/2014/ 645812 PMID: 25013446 46. Tonstad S, Despre´s J. Treatment of lipid disorders in obesity. Expert Review of Cardiovascular Ther- apy. 2011; 9: 1069–1080. https://doi.org/10.1586/erc.11.83 PMID: 21878051 47. References Biddinger SB, Almind K, Miyazaki M, Kokkotou E, Ntambi JM, Kahn CR. Effects of Diet and Genetic Background on Sterol Regulatory Element-Binding Protein-1c, Stearoyl-CoA Desaturase 1, and the Development of the Metabolic Syndrome. Diabetes. 2005; 54: 1314–1323. https://doi.org/10.2337/ diabetes.54.5.1314 PMID: 15855315 48. Guo J, Jou W, Gavrilova O, Hall KD. Persistent diet-induced obesity in male C57BL/6 mice resulting from temporary obesigenic diets. PLoS ONE. 2009; 4: e5370. https://doi.org/10.1371/journal.pone. 0005370 PMID: 19401758 49. Podrini C, Cambridge EL, Lelliott CJ, Carragher DM, Estabel J, Gerdin A, et al. High-fat feeding rapidly induces obesity and lipid derangements in C57BL/6N mice. Mammalian Genome. 2013; 24: 240–251. https://doi.org/10.1007/s00335-013-9456-0 PMID: 23712496 50. Smith BW, Adams LA. Nonalcoholic fatty liver disease and diabetes mellitus: Pathogenesis and treat- ment. Nature Reviews Endocrinology. 2011; 7: 456–465. https://doi.org/10.1038/nrendo.2011.72 PMID: 21556019 51. Shibata R, Ouchi N, Ohashi K, Murohara T. The role of adipokines in cardiovascular disease. J Cardiol. 2017; 70: 329–334. https://doi.org/10.1016/j.jjcc.2017.02.006 PMID: 28325521 52. Scotece M, Conde J, Vuolteenaho K, Koskinen A, Lo´pez V, Go´mez-Reino J, et al. Adipokines as drug targets in joint and bone disease. Drug Discov Today. 2014; 19: 241–258. https://doi.org/10.1016/j. drudis.2013.07.012 PMID: 23906693 53. Pan H, Guo J, Su Z. Advances in understanding the interrelations between leptin resistance and obe- sity. Physiol Behav. 2014; 130: 157–169. https://doi.org/10.1016/j.physbeh.2014.04.003 PMID: 24726399 54. Zhou Y, Rui L. Leptin signaling and leptin resistance. Frontiers of Medicine. 2013; 7: 207–222. https:// doi.org/10.1007/s11684-013-0263-5 PMID: 23580174 55. Paz-Filho G, Mastronardi C, Franco CB, Wang KB, Wong M, Licinio J. Leptin: molecular mechanisms, systemic pro-inflammatory effects, and clinical implications. Arquivos Brasileiros de Endocrinologia & Metabologia. 2012; 56: 597–607. 56. Sack GH. Serum amyloid A—a review. Mol Med. 2018; 24: 46. https://doi.org/10.1186/s10020-018- 0047-0 PMID: 30165816 57. Sun L, Ye RD. Serum amyloid A1: Structure, function and gene polymorphism. Gene. 2016; 583: 48– 57. https://doi.org/10.1016/j.gene.2016.02.044 PMID: 26945629 58. Joseph SV, Edirisinghe I, Burton-Freeman BM. Berries: Anti-inflammatory effects in humans. J Agric Food Chem. 2014; 62: 3886–3903. https://doi.org/10.1021/jf4044056 PMID: 24512603 59. Bujor O, Ginies C, Popa VI, Dufour C. Phenolic compounds and antioxidant activity of lingonberry (Vac- cinium vitis-idaea L.) leaf, stem and fruit at different harvest periods. Food Chemistry. 2018; 252: 356– 365. https://doi.org/10.1016/j.foodchem.2018.01.052 PMID: 29478554 60. Heinonen M. Antioxidant activity and antimicrobial effect of berry phenolics—A Finnish perspective. Molecular Nutrition and Food Research. 2007; 51: 684–691. https://doi.org/10.1002/mnfr.200700006 PMID: 17492800 61. Ek S, Kartimo H, Mattila S, Tolonen A. References Zheng S, Huang K, Zhao C, Xu W, Sheng Y, Luo Y, et al. Procyanidin attenuates weight gain and modifies the gut microbiota in high fat diet induced obese mice. Journal of Functional Foods. 2018; 49: 362–368. 73. Garbacki N, Kinet M, Nusgens B, Desmecht D, Damas J. Proanthocyanidins, from Ribes nigrum leaves, reduce endothelial adhesion molecules ICAM-1 and VCAM-1. J Inflamm. 2005; 2: 9. 74. Wu T, Yin J, Zhang G, Long H, Zheng X. Mulberry and cherry anthocyanin consumption prevents oxida- tive stress and inflammation in diet-induced obese mice. Molecular Nutrition & Food Research. 2016; 60: 687–694. 75. Wu T, Qi X, Liu Y, Guo J, Zhu R, Chen W, et al. Dietary supplementation with purified mulberry (Morus australis Poir) anthocyanins suppresses body weight gain in high-fat diet fed C57BL/6 mice. Food Chemistry. 2013; 141: 482–487. https://doi.org/10.1016/j.foodchem.2013.03.046 PMID: 23768383 76. Rimando AM, Kalt W, Magee JB, Dewey J, Ballington JR. Resveratrol, pterostilbene, and piceatannol in Vaccinium berries. J Agric Food Chem. 2004; 52: 4713–4719. https://doi.org/10.1021/jf040095e PMID: 15264904 77. Wahab A, Gao K, Jia C, Zhang F, Tian G, Murtaza G, et al. Significance of resveratrol in clinical man- agement of chronic diseases. Molecules. 2017; 22: 1329. 78. Era¨salo H, Ha¨ma¨la¨inen M, Leppa¨nen T, Ma¨ki-Opas I, Laavola M, Haavikko R, et al. Natural Stilbenoids Have Anti-Inflammatory Properties in Vivo and Down-Regulate the Production of Inflammatory Media- tors NO, IL6, and MCP1 Possibly in a PI3K/Akt-Dependent Manner. J Nat Prod. 2018; 81: 1131–1142. https://doi.org/10.1021/acs.jnatprod.7b00384 PMID: 29726680 79. Laavola M, Nieminen R, Leppa¨nen T, Eckerman C, Holmbom B, Moilanen E. Pinosylvin and Mono- methylpinosylvin, Constituents of an Extract from the Knot of Pinus sylvestris, Reduce Inflammatory Gene Expression and Inflammatory Responses in Vivo. J Agric Food Chem. 2015; 63: 3445–3453. https://doi.org/10.1021/jf504606m PMID: 25763469 80. Ji G, Wang Y, Deng Y, Li X, Jiang Z. Resveratrol ameliorates hepatic steatosis and inflammation in methionine/choline-deficient diet-induced steatohepatitis through regulating autophagy. Lipids in Health and Disease. 2015; 14: 134. https://doi.org/10.1186/s12944-015-0139-6 PMID: 26498332 81. Ohara K, Kusano K, Kitao S, Yanai T, Takata R, Kanauchi O. ε-Viniferin, a resveratrol dimer, prevents diet-induced obesity in mice. Biochem Biophys Res Commun. 2015; 468: 877–882. https://doi.org/10. 1016/j.bbrc.2015.11.047 PMID: 26596701 82. Chang C, Lin K, Peng K, Day Y, Hung L. Resveratrol exerts anti-obesity effects in high-fat diet obese mice and displays differential dosage effects on cytotoxicity, differentiation, and lipolysis in 3T3-L1 cells. Endocr J. 2016; 63: 169–178. References Characterization of phenolic compounds from lingonberry (Vacci- nium vitis-idaea). J Agric Food Chem. 2006; 54: 9834–9842. https://doi.org/10.1021/jf0623687 PMID: 17177509 62. Ka¨hko¨nen MP, Hopia AI, Heinonen M. Berry phenolics and their antioxidant activity. J Agric Food Chem. 2001; 49: 4076–4082. https://doi.org/10.1021/jf010152t PMID: 11513713 63. Hajazimi E, Landberg R, Zamaratskaia G. Simultaneous determination of flavonols and phenolic acids by HPLC-CoulArray in berries common in the Nordic diet. LWT—Food Science and Technology. 2016; 74: 128–134. 64. Nardi GM, De Farias Janua´rio Adriana Graziele, Freire CG, Megiolaro F, Schneider K, Perazzoli MRA, et al. Anti-inflammatory activity of berry fruits in mice model of inflammation is based on oxidative stress PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 16 / 17 PLOS ONE Beneficial effects of lingonberry (Vaccinium vitis-idaea L.) supplementation in a mouse model of obesity modulation. Pharmacognosy Research. 2016; 8: S42–S49. https://doi.org/10.4103/0974-8490.178642 PMID: 27114691 modulation. Pharmacognosy Research. 2016; 8: S42–S49. https://doi.org/10.4103/0974-8490.178642 PMID: 27114691 65. 65. Ha¨ma¨la¨inen M, Nieminen R, Vuorela P, Heinonen M, Moilanen E. Anti-inflammatory effects of flavo- noids: Genistein, kaempferol, quercetin, and daidzein inhibit STAT-1 and NF-κB activations, whereas flavone, isorhamnetin, naringenin, and pelargonidin inhibit only NF-κB activation along with their inhibi- tory effect on iNOS expression and NO production in activated macrophages. Mediators Inflamm. 2007 2007: 1–10. 66. Yoon JS, Chae MK, Lee SY, Lee EJ. Anti-inflammatory effect of quercetin in a whole orbital tissue cul- ture of Graves’ orbitopathy. Br J Ophthalmol. 2012; 96: 1117–1121. https://doi.org/10.1136/ bjophthalmol-2012-301537 PMID: 22661653 67. Ahn J, Lee H, Kim S, Park J, Ha T. The anti-obesity effect of quercetin is mediated by the AMPK and MAPK signaling pathways. Biochem Biophys Res Commun. 2008; 373: 545–549. https://doi.org/10. 1016/j.bbrc.2008.06.077 PMID: 18586010 68. Jung CH, Cho I, Ahn J, Jeon T, Ha T. Quercetin Reduces High-Fat Diet-Induced Fat Accumulation in the Liver by Regulating Lipid Metabolism Genes. Phytotherapy Research. 2013; 27: 139–143. https:// doi.org/10.1002/ptr.4687 PMID: 22447684 69. Grace MH, Esposito D, Dunlap KL, Lila MA. Comparative analysis of phenolic content and profile, anti- oxidant capacity, and anti-inflammatory bioactivity in wild Alaskan and commercial Vaccinium berries. J Agric Food Chem. 2014; 62: 4007–4017. https://doi.org/10.1021/jf403810y PMID: 24219831 70. Bakowska-Barczak AM, Marianchuk M, Kolodziejczyk P. Survey of bioactive components in Western Canadian berries. Can J Physiol Pharmacol. 2007; 85: 1139–1152. https://doi.org/10.1139/Y07-102 PMID: 18066116 71. Debnath SC, Sion M. Genetic Diversity, Antioxidant Activities, and Anthocyanin Contents in Lingon- berry. International Journal of Fruit Science. 2009; 9: 185–199. 72. PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 References https://doi.org/10.1507/endocrj.EJ15-0545 PMID: 26698690 83. Bhatt JK, Thomas S, Nanjan MJ. Resveratrol supplementation improves glycemic control in type 2 dia- betes mellitus. Nutr Res. 2012; 32: 537–541. https://doi.org/10.1016/j.nutres.2012.06.003 PMID: 22901562 PLOS ONE \| https://doi.org/10.1371/journal.pone.0232605 May 7, 2020 17 / 17
https://openalex.org/W3152606308	https://www.mdpi.com/2071-1050/13/8/4434/pdf?version=1618552754	English	null	A Managerial and Behavioral Approach in Aligning Stakeholder Goals in Sustainable Last Mile Logistics: A Case Study in the Netherlands	Sustainability	2,021	cc-by	11,622	Citation: Tolentino-Zondervan, F.; Bogers, E.; van de Sande, L. A Managerial and Behavioral Approach in Aligning Stakeholder Goals in Sustainable Last Mile Logistics: A Case Study in the Netherlands. Sustainability 2021, 13, 4434. https:// doi.org/10.3390/su13084434 Academic Editor: António Abreu Received: 23 March 2021 Accepted: 13 April 2021 Published: 15 April 2021 Publisher’s Note: MDPI stays neutral Citation: Tolentino-Zondervan, F.; Bogers, E.; van de Sande, L. A Managerial and Behavioral Approach in Aligning Stakeholder Goals in Sustainable Last Mile Logistics: A Case Study in the Netherlands. Sustainability 2021, 13, 4434. https:// doi.org/10.3390/su13084434 Keywords: city logistics; living lab; Policy Deployment; stakeholder analysis; sustainable last mile logistics; Theory of Planned Behavior Academic Editor: António Abreu Received: 23 March 2021 Accepted: 13 April 2021 Published: 15 April 2021 sustainability sustainability sustainability sustainability Frazen Tolentino-Zondervan 1 Research Group Logistics & Alliances, HAN University of Applied Sciences, 31 6826 CC Arnhem, The Netherlands; Enide.Bogers@han.nl 2 International School of Business, HAN University of Applied Sciences, 31 6826 CC Arnhem, The Netherlands 3 Institute for Management Research, Radboud University, 141 6525 AJ Nijmegen, The Netherlands; l.vandesande@fm.ru.nl * Correspondence: Frazen.TolentinoZondervan@han.nl 1 Research Group Logistics & Alliances, HAN University of Applied Sciences, 31 6826 CC Arnhem, The Netherlands; Enide.Bogers@han.nl 2 International School of Business, HAN University of Applied Sciences, 31 6826 CC Arnhem, The Netherlands 3 Institute for Management Research, Radboud University, 141 6525 AJ Nijmegen, The Netherlands; l.vandesande@fm.ru.nl * Correspondence: Frazen.TolentinoZondervan@han.nl Abstract: Stakeholders play vital roles in the implementation of sustainable last mile logistics solutions. Therefore, the ﬁrst step in setting up successful sustainable last mile logistics is to conduct stakeholder analysis. This paper analyzes the goals of the stakeholders in the Heijendaal living lab, a city logistics project that uses two hubs for bundling goods to be delivered to the Heijendaal campus in The Netherlands. We use the Theory of Planned Behavior and Policy Deployment to present a qualitative case study, which examines the goals of stakeholders in relation to their roles in the supply chain and in the organization, and if these goals lead to their expected participation behavior. Results show that stakeholders have economic, social, and environmental goals and that some of these goals are prominent within certain groups of stakeholders along the supply chain and within the organization. In addition, the set goals do not always lead to participation behavior of stakeholders due to identiﬁed disruptions and habits. This study identiﬁes the importance of information sharing and collaboration within the supply chain, the leading role of middle-level managers in translating strategic to operational goals, and the stimulation of behavioral factors to increase participation of stakeholders in the living lab. Citation: Tolentino-Zondervan, F.; Bogers, E.; van de Sande, L. A Managerial and Behavioral Approach in Aligning Stakeholder Goals in Sustainable Last Mile Logistics: A Case Study in the Netherlands. Sustainability 2021, 13, 4434. https:// doi.org/10.3390/su13084434 A Managerial and Behavioral Approach in Aligning Stakeholder Goals in Sustainable Last Mile Logistics: A Case Study in the Netherlands Frazen Tolentino-Zondervan 1,2,* , Enide Bogers 1 and Luc van de Sande 3 Frazen Tolentino-Zondervan 1,2,* , Enide Bogers 1 and Luc van de Sande 3 1. Introduction The increased urbanization in European cities combined with increased consumption has led to sustainability-related problems in city centers and their logistics. Despite its contribution to approximately 5% of the European Union (EU) GDP and employment of 10 million people [1], the EU transport sector is one of the largest contributors of CO2 emission, congestions, and accidents in the cities, affecting the social well-being and health of city inhabitants [2–4]. In order to improve the livability of the cities and to achieve a 90% cut in transport emissions by 2050, the EU made the Green Deal on sustainable and smart mobility strategies for EU transports [1]. Part of the Green Deal includes city logistics strategies that focus on zero-emission vehicles, smart delivery designs, and planning and management of the logistics of goods [5–7]. One popular solution to decrease the movements in cities is the use of city hubs. A city hub enables incoming trucks to unload goods outside the city, bundles these goods, and performs efﬁcient and emission-free last mile logistics. City hubs were shown to be cost-efﬁcient, saved time and fuel, and contributed to a reduction of CO2 [8,9]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Most city logistics projects involve various stakeholders such as supply chain actors, government, and knowledge institutes (e.g., [10,11]). The presence of multiple stakeholders https://www.mdpi.com/journal/sustainability Sustainability 2021, 13, 4434. https://doi.org/10.3390/su13084434 Sustainability 2021, 13, 4434 2 of 17 in city logistics is complicated. Stakeholders have different interests and views, which are often conﬂicting, in terms of their goals and the management and decision-making of the project [12]. For instance, the sustainability goals from a consumer point of view could require investments from suppliers, thereby complicating the economic goals of suppliers. Therefore, managing stakeholders’ interests and preferences at the beginning of a project is crucial to gain wide support for any city logistics project [13]. Failure to take into account the goals and views of the stakeholders has resulted in the termination of many city logistics projects. 1. Introduction This results, for instance, in inadequate usage of hubs, misallocation of cost and beneﬁts, the reluctance of the government in providing subsidies, wrong location and type of vehicle choices, and/or poor organization of the city logistics projects (see for example [5,14,15]). A traditional way of doing stakeholder analysis in a project is by looking at the alignment of goals of different actors in the supply chain (e.g., [16]). However, we found out prior to this research that within a stakeholder group, goals may vary at different management levels. In the end, it is the individuals who represent organizations and interact with other stakeholders, both within the supply chain and their own organizations. Therefore, this paper argues that in order to analyze the goals of stakeholders, it is important to look at both the alignment of goals within the supply chain and between the various management levels of an organization. g g This research focuses on the Heijendaal living lab, a city logistics project that aims to improve the sustainability of delivering goods in Nijmegen, the Netherlands, using the City and Campus hubs. This living lab is an outcome of the signed agreement among the three organizations in the Heijendaal campus, namely the university, hospital, and university of applied sciences, to test sustainable last mile delivery. By deﬁnition, a living lab promotes a user-centered, open-innovation ecosystem, integrating transdisciplinary research within a public-private-people partnership (4Ps) [17]. The concept of living labs has been applied in city logistics as a method for action-driven partnerships that follows a solution-driven approach [18,19] and as an approach to foster stakeholders’ engagement in co-creating policies [20]. This is translating to the Heijendaal living lab through its setup, which looks at a new way of working through the collaboration of supply chain actors, academia, and governments, and ﬁnding an optimal solution for cost-efﬁcient, consolidated and emission-free “last mile” logistics of goods in the Heijendaal campus through learning-by-doing. Since the Heijendaal living lab has a Business-to-Business structure and involves large organizations such as suppliers and receivers, the participation of each actor is inﬂuenced at various management levels. g The research questions of this study are: (1) What are the goals of each stakeholder in relation to the role in the supply chain and the management level within an organization? and (2) To what extent do these speciﬁed goals lead to successful participation in sustainable last mile logistics? 2.1. Stakeholder Analysis City logistics involved multiple stakeholders, which are deﬁned as “any group of people, organized or not organized, who share a common interest or stake in a particular issue or system” [34]. City logistics stakeholders can be classiﬁed into three: (1) industrial, represented by suppliers, hubs, and freight carriers; (2) institutional policymakers; and (3) consumers, the receivers of goods [35,36]. In any project, like city logistics, stakeholders have different goals and interests [11]. Industrial stakeholders have an interest in the ﬁnancials such as costs and revenues and the process performances that include delivery time, volume, and responsiveness. Institutional stakeholders are interested in aspects such as reduction in congestion, accidents, pollutions, and noise in the cities. Consumers’ interests focus on service levels that involve timeliness, costs, and traceability. Conduct- ing stakeholder analysis helps to understand the interests and expectations of multiple stakeholders, which can be used to shape stakeholders’ decisions [37]. 1. Introduction To answer these questions, we use a framework of organizational and behavioral mode choice processes developed by Bogers [21] as part of a multimodal research project [22]. Boger’s framework is based on the Theory of Planned Behavior (TPB), which explains that the intentions of individuals serve as the basis of their behavior, and the Policy Deployment, which assesses the alignment of the goals of the top, middle, and lower level managements. Since this research is exploratory in nature and is the ﬁrst step in understanding the setup of the living lab, we used a qualitative research method. By understanding the goals that each stakeholder ﬁnds important across the supply chain and within the organization, possible improvements can be recommended to enhance the participation of stakeholders in the living lab. Ultimately, these improvements could promote the continuous use of the sustainable last mile logistics concept by supply chain actors even after the living lab ends. g The next section starts with explaining the research framework. This is followed by an explanation of the qualitative case study method of this research. Next, the results on different stakeholder goals and whether or not these goals lead to participation in the Heijendaal living lab are presented. This is followed by a discussion of the importance of Sustainability 2021, 13, 4434 3 of 17 3 of 17 goal alignment in order to gain a high probability of success in last mile logistics and possi- bilities to enhance goal alignment. Finally, this paper concludes with the learnings from stakeholder analysis and recommendations on how to improve stakeholders’ participation in the living lab. 2. Theoretical Framework Multi-Criteria Decision Making methodologies, such as Social Cost-Beneﬁt Analysis and MAMCA, have been applied to solve multi-stakeholder city logistics challenges (see for example [23–27]). These methodologies involve multiple stakeholders in the imple- mentation process and assess quantitatively the different criteria of stakeholders based on a set of alternatives, to determine the best solution to achieve consensus among stake- holders. However, in reality, sustainable city logistics projects are complex in nature, with stakeholders having various degrees of power, inﬂuence, and organizational backgrounds, making the actual implementation of these methods often challenging. Gammelgaard [6] argues to instead use organizational theory on change management, to gain insights on the changes that are needed in the process to improve the implementation of city logistics. An important element of the organizational change process is understanding individual behavior in an organization, as individuals eventually are the agents driving change [28]. TPB is a behavioral approach and has been proven to be useful in sustainable logistics deci- sion making, such as on the choice of transportation [29,30], implementing green practices in logistics [31], and route choices [32]. The behavioral approach in city logistics is still under-researched and this study is a contribution to this ﬁeld. In combination with the Policy Deployment, this study gives a deeper understanding on how decision happens at organization and supply chain levels in city logistics. The theoretical framework of this study combines the Stakeholder Analysis and the framework developed by Bogers [21] of organizational and behavioral mode choice processes. By involving key stakeholders in an open-innovation ecosystem like a living lab, it creates new knowledge and leads to a change in practice [33]. Bogers’ framework emphasizes the importance of understanding the decision-making at organizational and individual levels through the lens of TPB and Policy Deployment. TPB highlights individu- als as decision-makers while Policy Deployment highlights that individuals are also part of an organization comprised of multiple levels. Therefore, using TPB combined with Policy Deployment can help understand how the decision-making of an individual at various management levels takes place. The next section explains the Stakeholder Analysis, Policy Deployment, and TPB, the link between these theories, and the complete research model. 2.3. Theory of Planned Behavior Individuals are decision-makers at different levels in the organization. The TPB predicts and explains individual behavior based on the combination of three behavior- speciﬁc factors [42,43]. These three factors are (1) attitude towards a behavior, (2) subjective norms, and (3) perceived behavioral control. The attitude towards behavior is determined by one’s belief that the behavior leads to either favorable or unfavorable outcomes. The subjective norms are inﬂuenced by external environments such as family, education, or peers, and result in perceived societal standards on how a person should (or should not) perform the behavior. The perceived behavioral control refers to the resources or opportunities present in performing the desired behavior. Together, the combination of these three factors determines the intention, which ultimately becomes the basis of an individual’s behavior. The intention is an indicator of how strongly a person is willing to try a certain behavior or decision. The behavior is the ﬁnal outcome of whether the intention is carried out or not. As a rule of thumb, the stronger the three behavior-speciﬁc factors to an individual, the stronger the intention to perform the behavior in question. In some cases, intentions do not automatically lead to the desired behavior because habits and disruptions can overrule intentions [32,44,45]. Habits include automatic and repetitive behavior, like frequently used routes and modes of transports [46], while disruptions refer to changes that interrupt the planned action, such as a system failure [47]. 2.2. Policy Deployment To understand how strategies are deployed across multiple levels of an organization, we use the theory of Policy Deployment (also called Hoshin Kanri). The Policy Deployment is a tool that ensures that the goals of a company are in alignment with different man- Sustainability 2021, 13, 4434 4 of 17 agement levels, namely strategic, tactical, and operational levels [38]. At a strategic level, a company or individual sets overall long-term goals in alignment with the company’s strategies, which are based on the characteristics of the company and the demand of its network. These long-term goals are made measurable at the tactical level (mid-term goal) using Key Performance Indicators (KPIs) [39]. The measurement of KPIs is based on the set metrics (short-term goals) at an operational level. Policy Deployment is useful to understand the sustainability goals of organizations at various management levels in a supply chain. An example of this is GreenSCOR, a dedicated tool that identiﬁes measurable sustainability goals, such as zero emissions, environmental certiﬁcations, reduction in waste, use of green materials, collaboration with suppliers, and so on [40,41]. 2.4. The Research Framework The research framework used in this study links Policy Deployment and TPB. The framework in Figure 1 presents processes in terms of how the set goals play a role in determining behavior (process A) and how behavior-speciﬁc factors inﬂuence set goals, which affect the intention and behavior of individual stakeholders (process B). In process A (refer to the ﬁrst box), the set sustainability goals of an organization are formulated at a strategic level, and are then translated to the department and operational goals. These goals can be related to zero-emissions, ﬁnancial and social goals. Some of these sustainability goals can be prominent to a speciﬁc organization based on their stakeholder roles (industrial, institutional, and consumers) in city logistics. The sustainability goals inﬂuence the individual behavior-speciﬁc factors such as attitudes, beliefs, and perceived behavioral control. These factors determine the intentions that ultimately lead to behavior, which refers to the participation in the living lab of individuals at various organizational levels. In support of this process, Montani et al. [48] argue that a goal may inﬂuence the cognitive process and decision-making of an individual based on a goal regulation perspective. Individuals envision the future based on the set goals and make a plan of action to achieve the desired goals. Even though the intention to pursue a goal is present, it is also possible that the behavior of an individual is not aligned with his intention due to unforeseen barriers that include habits [32,44] or disruptions [21]. In process B (refer to the second box), the behavioral factors of an individual have an influence on how the goals are set within the various levels of organizations. As argued by Dholakia and Bagozzi [49], the set goals of an individual could also be an outcome Sustainability 2021, 13, 4434 5 of 17 5 of 17 of the pre-decisional psychological processes, such as investments in resources to obtain information, norms and beliefs, and confidence. For example, some studies have shown that the sustainability attitude and values of an individual have an influence on the sustainability goals setting of a company at both strategic and operational levels and/or individual deci- sions to adopt sustainable practices (e.g., [21,50,51]). Selecting a goal leads to the formation of intention, which serves as the basis of decision-making [52]. These two processes are called goal setting and goal implementation, respectively [53]. Figure 1. The theoretical framework adapted from the framework of organizational and behavioral mode choice processes as developed by Bogers (2017). 3. Materials and Methods The research design of this study is an exploratory qualitative case study design. The next section explains the case study background, followed by the case study analysis and the data collection. 2.4. The Research Framework Goal setting is the process in which goals are set to maximize the realization of intentions. In general, a strong intention to pursue the goal is based on the belief that a given goal is both highly desirable and feasible [43]. While intention is merely a desired end state, the goal implementation is the behavior to achieve the goal. The implementation helps individuals (or companies) to close the gap between the initial goal setting and meeting the goals. In this research, we will focus on the process B in explaining the result. Figure 1. The theoretical framework adapted from the framework of organizational and behavioral mode choice processes as developed by Bogers (2017). Heijendaal Living Lab Stakeholders Heijendaal Living Lab Stakeholders Heijendaal Living Lab Stakeholders The Heijendaal living lab follows a Triple Helix approach, in which the innovation is the outcome of a collaboration between three parties—government, private sector, and knowledge institutes. In this research, we focused on the following stakeholders in the supply chain: suppliers, City and Campus hubs, and receivers (the three Heijendaal organizations). According to the supply chain structure of the Heijendaal living lab (Figure 2), the suppliers ﬁrst deliver their goods to the City hub. From the City hub, the goods are delivered either directly to the three Heijendaal organizations or to the Campus hub, which will deliver the goods to individual Heijendaal organizations. Figure 2. The supply chain structure that aims to deliver the goods to receivers in the Heijendaal living lab. Figure 2. The supply chain structure that aims to deliver the goods to receivers in the Heijendaal living lab. Most suppliers are large organizations and therefore involve decision-makers at strategic, tactical, and operational levels. The same applies to the three organizations at the Heijendaal campus. At the strategic level, we ﬁnd the boards of the Heijendaal organizations and the steering group of the living lab. The tactical level managers consist of mainly procurement managers. The operational level includes the employees of logistics, procurement, Information and Technology departments, campus stores, and facility goods users. The Campus hub is located within the goods reception of the academic hospital and plans have been made to use this hub for the last mile distribution to the three academic organizations. The operations at the Campus hub are under the supply chain department of the hospital, whose head is also part of the steering group committee of the living lab. The City hub is located at the periphery of the city and is used for bundling and transporting goods using zero emission transport to the campus. Unlike the Campus hub and receivers, which have approximately 4000–11,000 employees, the City hub is a small private company with few people (between 11–50). This means that its organizational process is simpler when compared to bigger supplier companies and academic organizations. 3.1. Case Study Background: Heijendaal Living Lab The Heijendaal campus hosts three organizations: a hospital, a university, and a university of applied sciences. The Heijendaal campus is characterized by sizable ﬂows of students, patients, visitors, and employees, and supplies needed for research, treatment, and teaching to each of these organizations. This situation leads to trafﬁc congestion and consequently to negative impacts on the environment and quality of life. In order to reduce these negative consequences, the three organizations have decided to collaborate in a long-term living lab project, with the aim to gradually learn by doing. As an initial solution for the campus, a new concept has been proposed consisting of newly established City and Campus hubs, where shipments are bundled and delivered with zero emission vehicles to the organizations. Sustainability 2021, 13, 4434 6 of 17 3.2. Case Analysis The case of Heijendaal living lab is a new phenomenon, because - unlike most other last mile logistics solutions - this living lab is initiated by three large receivers. This results Sustainability 2021, 13, 4434 7 of 17 in a market driven (instead of subsidy driven) learning-by-doing approach. Therefore, this research follows the qualitative case study method of Yin [54], which addresses the hows and whys of a research in a contemporary event. A qualitative case study draws conclusions based on the depth and quality of information provided by the respondents, rather than the size of the samples [55]. The collection of data is guided by the analytical concepts that will help to understand the goals and behaviors of various stakeholders in Heijendaal living lab. Table 1 summarizes these analytical concepts including their deﬁnitions and how they are useful in the practical research. This case is further analyzed by iteratively cross checking and comparing theory and empirical data [54,56]. First, the results were presented using the analytical concepts. Second, the patterns, similarities and differences in terms of how the goals of each stakeholders lead (or do not lead) to participation in Heijendaal living lab were identiﬁed. And third, we compared our ﬁndings to our framework in terms of how the individual behavioral factors inﬂuence goals, which ultimately lead to actual individual behavior. Table 1. The summary of analytical concepts and how they are applied in the case of Heijendaal Living Table 1. The summary of analytical concepts and how they are applied in the case of Heijendaal Living Lab. Analytical Concepts Deﬁnition Key Questions Expected Outcome Goals A goal is “a cognitive representation of a desired endpoint that impact evaluations, emotions, and behaviors” [57] Which goals do you have in mind for your organization [and/or department] and for the living lab project? Lists of economic, social, and environmental goals of individual stakeholders Behavioral factors Attitude towards participation Determined by one’s belief that the behavior leads to either favorable or unfavorable outcomes [43] How will this project beneﬁt your organization? Individual feeling (positive or negative) in relation to the living lab What do you expect will be the outcomes of the Heijendaal living lab? Subjective norms Refers to perceived social pressures, that are inﬂuenced by the standards set by the society on how a person should (or should not) perform the behavior [43] How important is sustainability in your organization? 3.2. Case Analysis Answer to whether there is an urgency to act sustainably (yes or no)? What does sustainability mean to you? Perceived behavioral control Resources or opportunities present in performing the desired behavior [43] Would it be feasible for your organization to participate in the Heijendaal living lab? (Why or why not?) Factors that could permit or limit participation Behavior An act of carrying out or not, the individual action when the opportunity appears [43] Is your organization (Are you) currently participating in the Heijendaal living lab? Why or why not? Individual decision to participate (or not) in the living lab Unforeseen barriers These are the unpredicted hurdles that hinder the execution of intended behaviors. Examples include habits, which are repeated actions, routines, or past behaviors, and disruptions, which refers to sudden changes such as system failure [47] What are the main challenges, or bottlenecks in the Heijendaal living lab project? Set of constraints that overrule the intention of a stakeholder to participate in the Heijendaal living lab. Do you think that the goals of your organization will be the same as the goals of other organizations? Is there an alternative solution in addition to the suggested logistics structure that could improve the sourcing (or supplying) of goods to the Heijendaal campus? Sustainability 2021, 13, 4434 8 of 17 3.3. Data Collection 3.3. Data Collection Multiple methods were employed in collecting data. First, a total of sixteen in-depth interviews were conducted from December 2019 to June 2020 which include the executives (six) of which three are procurement department heads, employees of the organizations in Heijendaal campus (four), suppliers (two), hubs representatives (two), municipal public servant (one), and internship student (one). To ensure the representativeness of the samples, the choice of respondents are based on three criteria, (1) knowledge of the living lab, (2) involvement in this project, and (3) their positions in the organization and supply chain of the living lab. Interviewees provided information concerning the goals of the stakeholders in the project, the behavioral factors that contribute to their choice of behavior in the living lab, and their current behavior. To ensure the validity of this research, the open-ended questions in the interviews were based on a set of analytical concepts and were adapted to the different groups of stakeholders (see Table 1). Cross-checking of answers for the goals of other stakeholders was done during interviews to increase the reliability of the research. Second, literature and organizational reports were reviewed to further support the identiﬁed goals of each stakeholder in the interviews. And third, observations of the situations in the campus are used to validate the ﬁndings of the interviews. 4. Results This section presents the results of the stakeholder analysis. The ﬁrst part answers the question: What are the goals of each stakeholder in relation to the role in the supply chain and the management level within the organization? The second part answers the question: To what extent do these speciﬁed goals lead to successful participation in sustainable last mile logistics? 4.1. Stakeholder Goals The goals of the following stakeholders are presented in Table 2: suppliers, the City and Campus hubs, and receivers in Heijendaal living lab. These goals are the performance criteria that each stakeholder ﬁnds important for participating in Heijendaal living lab. From a sustainability point of view, the stakeholder goals can be classiﬁed as environmental, economic, and social goals. dual stakeholder goals in relation to the role in the supply chain and in the management level within Table 2. Individual stakeholder goals in relation to the role in the supply chain and in the managem the organization. Table 2. Individual stakeholder goals in relation to the role in the supply chain and in the management level within the organization. Stakeholders Management Levels Speciﬁc Goals Sustainability Dimension Suppliers Strategic - Circularity of products - More efﬁcient process - Reduction of environmental footprints - People’s development Environmental, Social, Economic Tactical - Improved contracts with customers Economic Operational - Supply more volume - Affordability of using the city hub Economic The City hub Strategic - Reduced environmental pollution - Revenue - Upscaling the operation (More volumes and expanding) Environmental, Social, Economic The Campus Hub Tactical - Efﬁciency in receiving of goods Economic Operational - Job security Economic 9 of 17 Sustainability 2021, 13, 4434 Table 2. Cont. Table 2. Cont. Stakeholders Management Levels Speciﬁc Goals Sustainability Dimension Receivers (Heijendaal Organizations) Strategic - Zero emission - Less accidents and improved health - Centralizing purchasing behavior of employees - Sustainable innovations - Upscaling Environmental, Social, Economic Tactical - Collaboration for joint tendering Economic Operational - 24/7 delivery - Traceability and improved delivery in locations Economic Sources: Interviews. Sustainability Dimension 4.1.1. Goals between Different Supply Chain Actors 4.1.1. Goals between Different Supply Chain Actors 4.1.1. Goals between Different Supply Chain Actors In this section, we looked at the goals of different supply chain actors that are in different management levels and compared them with other actors. Looking at stakeholder goals of different actors such as suppliers, City hub, and Heijendaal organizations at a strategic level, all goals were focused on the overall sustainability criteria and therefore were aligned. According to one of the board members of a receiving organization, “[Our] number one goal is zero emission in the campus in 2025. There is much trafﬁc, such as trucks, but also cars in general and bikes in this campus. People are walking all over the campus. 4.1. Stakeholder Goals [Our] number two goal is reducing the accidents in the campus”. Both the City hub and the Heijendaal organizations aimed to improve sustainability impact by upscaling the operation of the City hub through an increase in the supplier’s use of the City hub and an increase in the number of customers. For suppliers, participating in the Heijendaal living lab was a strategy to engage in sustainable operations. gy g g p Examining the goals of the supply chain actors at tactical and operational levels, the economic goals appeared to be more prominent. Some goals were aligned at a tactical level, such as the collaboration goals of tactical managers of suppliers and receivers. In a few cases, the collaboration resulted in the use of the City hub because the suppliers and receivers shared costs. There were also goals that were not aligned at an operational level. For example, the standard rate for using the City hub did not work well with some suppliers. As mentioned by one of the interviewees, “A challenge is that for every supplier, you have to ﬁnd a business case”. In addition, the Campus hub aimed to improve efﬁciency by bundling the orders and the receiving of goods for employees of Heijendaal organizations. However, a number of unnecessary priority deliveries were still made by employees, as observed in the Campus hub. 4.1.2. Goals between Management Levels of One Supply Chain Actor In this section, we analyzed the goals within one organization by comparing the goals in two levels: strategic vs tactical and tactical vs operational. We did this for both the cases of suppliers and receivers. Looking at the alignment of goals between strategic and tactical levels, these two levels had a different focus but were mostly aligned. For the suppliers, establishing relationships and contracts with receivers at the tactical level led to the use of the City hub, which con- tributed to achieving the sustainability goals of the supplier company at a strategic level. In another case, some middle-level managers of suppliers chose to join sustainable initiatives for the company, such as engaging in sustainable projects, even though sustainability is not promoted at a strategic level. For the receivers, the joint collaboration of procurement department heads is expected to generate more volume of supplies, which will enable them to demand suppliers to use the City hub in delivering goods to the Heijendaal campus. As a consequence, there would be less movement of goods at the campus and there would be more centralized ordering for the three organizations. The establishment of the Campus Sustainability 2021, 13, 4434 10 of 17 10 of 17 hub also aligned with centralizing the purchasing behavior of employees, which was also part of the strategic goals of Heijendaal organizations. By having uniﬁed ordering and addressing systems, the ﬂow of orders and incoming goods can be centralized. Similar to the case of suppliers, other middle-level managers of the receiving organizations also chose to incorporate sustainable decisions in their work regardless of whether sustainability is part of the strategic goals or not. Looking at the goals at tactical and operational levels, the goals for suppliers are aligned while the alignment for the Campus hub and receivers was still unclear. For suppliers, the collaboration established by key account managers with receivers could result in more affordable use of the City hub, which was an important indicator at the operational level. For the Campus hub, the tactical goal was to improve the efﬁciency of the ﬂow of goods through combining the ordering and receiving processes of the Heijendaal organizations into one standard process. However, for employees, it was unclear whether the standardizing of the logistics process through the Campus hub could lead to fewer working hours and possible elimination of other receiving departments or a shift in their tasks. 4.2. Goals and Participation in Sustainable Last Mile Logistics This section presents the individual behavioral factors, unforeseen barriers, and the actual behavior of stakeholders in the living lab (refer to Table 3). Using these elements of the theoretical framework (refer back to Figure 1), we explain the extent to which the speciﬁed goals of stakeholders lead to successful participation in sustainable last mile logistics. 4.1.2. Goals between Management Levels of One Supply Chain Actor In addition, the training to comply with the standard process was perceived as additional work by employees of the university. For the receivers, employees perceived the 24/7 delivery as a form of higher service level, while the plan at both strategic and tactical levels was to bundle the purchasing of goods within the organization, which could lead to fewer ordering moments and longer lead times. 4.2.1. Individual Behavioral Factors The attitude towards participation in the Heijendaal living lab is mixed. On the one hand, the City hub and the Heijendaal institutions at strategic and tactical levels were very positive towards the living lab, because they believed that Heijendaal living lab can bring about a sustainable impact. On the other hand, the suppliers, the campus hubs, and Heijendaal organizations at operational levels were not always positive towards the initiative. For suppliers, some had a positive attitude, especially when the living lab was aligned with the sustainability goals of the company. These were the participating suppliers in the living lab. However, most had a mixed attitude due to the perceived lack of ﬁnancial beneﬁts in the living lab. For employees of the campus hub, there was less enthusiasm due to the fear of additional work and fear of losing their job. In addition, some employees perceived uncertainty due to missing information on the planning and progress of the project. For other employees, the centralization of purchasing was seen as a bureaucratic and less ﬂexible system. In terms of the subjective norms, all stakeholders still found sustainability important, either on a personal or organizational level. The sustainability norms for some stakeholders were prominent and had clear effects on the goal setting in the organization. These stake- holders were also the ones who currently participated in the living lab. For instance, some suppliers at strategic and tactical levels were convinced that in order to be ‘future-proof’, meaning being able to adapt to all sustainability requirements in the future, it was best to include sustainability in the goals of the company as early as possible. The Green Deal in the Netherlands (see [58]), which push transport companies towards zero emission goals, was also seen as inevitable by some suppliers. Therefore these suppliers directed their operations towards sustainability. The executive of the city hub also had a strong orientation towards sustainability, both at personal and organizational levels. Similar to supplier views, the representative of the City hub mentioned: “We are stepping on an early stage of this whole Sustainability 2021, 13, 4434 11 of 17 11 of 17 zero emission future. This will be the future, let us be honest”. This value or belief contributes to how the sustainability elements such as efficient bundling, track and trace systems, and zero emission delivery, were set as part of the sustainability performance of the City hub. 4.2.1. Individual Behavioral Factors For the executives of the Heijendaal living lab, their belief in making a sustainable Heijendaal campus was the reason why the living lab existed in the first place. Table 3. Summary of behavioral factors, actual behaviors, and unforeseen barriers for the stakeholders of Heijendaal living. Stakeholders Management Levels Individual Behavioral Factors Unforeseen Barriers Actual Behavior Attitude towards participation Subjective norms (Urgency to act sustainably?) Perceived behavioral control (Factors that control participation) Participating or not participating? Suppliers Strategic Positive and negative Yes Some are participat- ing.Majority do not participate (yet). Tactical Positive and negative Yes Nature of products, own logistics struc- ture/department Operational Positive and negative Yes Costs of participation, Small volume of supplies The City Hub Strategic Positive Yes Volume of goods from suppliers Participating The Campus Hub Tactical Positive and negative Yes Differences in IT system Participating with limited support Operational Positive and negative Yes Job uncertainty Lack of time Participating with limited support Heijendaal Organizations Strategic Positive Yes Lack of time Participating Tactical Positive Yes Better tender contracts with other parties Participating Operational Positive/Negative Yes More work Time Uncertainty 24/7 delivery culture IT system not supported Participating with limited support Sources: Interviews and organizational websites. Sources: Interviews and organizational websites. The perceived behavioral controls were also factors that limited the participation of many stakeholders in the Heijendaal living lab. For some suppliers at a tactical level, the nature of the company products could complicate the participation. For instance, some suppliers found it difﬁcult to combine services while at the same time storing the product at the City hub. Others found it difﬁcult to store hardware products of high value because it would result in high storage costs. In addition, others could not participate (yet) because they had their own transport facilities and own logistics employees. Switching from having their own logistics structure to delivering their goods to a city hub could be an expensive decision. p At the operational level, some suppliers saw the high operational costs of using the City hub. For example, some suppliers supplied only limited volume to Heijendaal organi- Sustainability 2021, 13, 4434 12 of 17 12 of 17 zations, thus using the City hub could lead to unproﬁtable business cases. Other suppliers that used couriers still found the rate of their current couriers cheaper as compared to using the City hub. For the City hub, it had in general no problem with participation. 5. Discussion Using the theoretical framework in Figure 1, we reﬂect on how the alignment of stakeholder goals at supply chain and organizational levels, using policy deployment, can inﬂuence the success of a city logistics project. We then extend the discussion on the importance of inﬂuencing behavioral factors of TPB to encourage decision-makers to participate in sustainable last mile logistics. Finally, we discuss how to cope with the unforeseen barriers in order to lead stakeholders towards sustainable decisions. 4.2.2. Unforeseen Barriers There were also a number of unforeseen barriers such as habits and disruption fac- tors that limited the efﬁcient participation of currently participating stakeholders in the Heijendaal living lab. Habits include the ordering culture of employees, such as last- minute, unbundled, small order size, and next-day delivery. Disruption factors include the interruption in the participation process despite the positive intention. There were three disruption factors that were identiﬁed in the ﬁeld. First was an opportunity for the procurement department to also ﬁnd collaboration in sourcing goods outside the living lab. For instance, universities could join tendering processes with other universities while hospitals could also join with other hospitals, particularly if the collaboration could save costs for the organization. As a consequence, the procurement department may engage in a collaboration that is either sustainable or non-sustainable. The second was the time issue. It required time and information sharing to determine which goods were suited for joint tendering within and outside the living lab. In addition, the executive and employees lacked the time to work on the living lab. Third, the IT ordering system did not support the change in buying behavior of employees (e.g., ordering 24/7 or on priority). 4.2.1. Individual Behavioral Factors However, if it continues to receive little volume from suppliers, it could encounter issues in covering its operating costs. At the operational level of the Campus hub and Heijendaal organizations, the perceived high workload of employees came from the differences in the IT systems and the ordering and receiving procedures of goods of the three Heijendaal organizations. Aligning the systems and procedures could take a lot of time in addition to the normal work of employees. The job uncertainty issue arose based on the required restructuring of the ordering and receiving procedures. 5.1. Goal Alignment Goal alignment in sustainable last mile logistics that involve multiple stakeholders is complex and challenging, yet it can improve the chance of success of the last mile logistics project. Literature has shown so far the isolated analysis in terms of the alignment of stakeholder goals at project [59], supply chain [11,13], and management [60] levels. The unique contribution of this research to literature is that it provides overall insights into the alignment of goals at the supply chain and organizational levels to determine stakeholders’ participation in sustainable last mile logistics. In general, if the sustainability ambitions are shared by stakeholders in the project, along the supply chain, and between the management levels, the more likely it is that they will participate in sustainable last mile logistics. Organizations that incorporate sustainability in their goals, could serve as a starting point to determine suitable partners for sustainable last mile logistics. We will discuss below how to align goals at supply chain and organizational levels, in order to achieve effective participation in sustainable last mile logistics. 5.1.2. Alignment of Goals at Management Levels 5.1.2. Alignment of Goals at Management Levels Policy deployment has been useful in understanding how the sustainability goals of an organization are translated at different management levels. This study shows that the traditional top-down approach in translating organizational goals from strategic to operational level does not always result in the expected alignment of goals within an organization. Based on the results, on one hand, alignment of goals exists between strategic and tactical levels for both suppliers and receivers. However, the alignment of goals between the tactical and operational levels for receivers is still unclear. Due to this, the realization of the strategic goal of an organization is currently sub-optimal. On the other hand, some cases of suppliers and receivers show that the sustainable behavior of an organization exists at middle-level managers despite the lack of promoting sustainability goals at a strategic level. In order to improve the alignment between tactical and operational goals, it may help to strengthen the role of the middle-level managers as “boundary spanners”. Boundary spanners have both a high level of contact with the external environment of an organization (or department) and with the internal of an organization that they work with [63]. As boundary spanners, middle-level management serves as a hub for information and knowl- edge on various topics, like problem deﬁnition, project assessment, resource identiﬁcation, and matters related to decision making [64]. In addition, middle-level management has an important role in integrating the strategic goals with the values, which are determined by societal, cultural, and religious experiences of employees, in the workplace [65]. This could work well in the living lab since the procurement department heads are part of the steering group committee and interact with suppliers as well as employees. The middle-level managers could be the medium in communicating plans and updates on the project at strategic and operational levels, in facilitating the interactions with the employees, and in identifying the bottlenecks at lower levels. 5.1.1. Alignment of Goals at Supply Chain Level Aligning stakeholder goals in sustainable last mile logistics can help to optimize individual supply chain actor’s resources and can strengthen each other’s position in Sustainability 2021, 13, 4434 13 of 17 13 of 17 the industry [61]. If a supply chain actor experiences goal alignment, the more likely it will see the beneﬁts leading to participation in last mile logistics. Based on the result, supply chain actors experienced both alignment of goals (e.g., the case of the executives of Heijendaal organizations and the City hub; tactical managers of suppliers and receivers) and misalignment of goals (e.g., suppliers and the City hub; operational employees of Heijendaal organizations and the Campus hub). Supply chain actors whose goals align with other actors are currently participating in the living lab, while those whose goals do not align with other actors are limited in their participation. To align the goals within the supply chain, coordination and collaboration must be further strengthened. Information sharing as part of collaboration can help solve bottlenecks in sustainable last mile logistics [62]. For instance, knowing the information about the customers of suppliers outside the Heijendaal organizations can help to increase the volume of goods to be delivered to the City hub and can further increase the bundling of goods to consumers. For coordination, the procurement departments can demand suppliers to use the City hub by incorporating this as a standard requirement of the tendering process. Finally, intervention from the receiver could help other suppliers (e.g., sharing of costs in the use of hub) to clearly realize the beneﬁt of participating in the living lab. 5.2. Behavioral Factors The insights of this study also conform to the literature that emphasizes the importance of inﬂuencing individual behavioral factors to help an organization set sustainability goals and engage in sustainable projects (see for example [51,66]). As seen in the results, a decision-maker who has a sustainability background or an afﬁnity to sustainability incorporates sustainability goals and KPIs within the organization, and therefore is more likely to perform sustainable actions (e.g., participating in the living lab). As a consequence, stimulating sustainable personal norms, attitudes, and control is necessary. As shown in the studies of Sparkman and Walton [67] and She and MacDonald [68], an increase in an Sustainability 2021, 13, 4434 14 of 17 14 of 17 individual’s exposure to normative and sustainability information improves the individual attitude and behavior towards sustainability. Alternatively, incorporating “green human resource practices”, such as hiring someone who already has a sustainability mindset and training, especially young employees, can help an organization achieve its sustainability goals [69]. 5.3. Unforeseen Barriers Consistent with our framework in Figure 1 and the literature [45,70,71], this study has identiﬁed multiple unforeseen barriers, which mostly exist at a personal level and hinder achieving the desired behavior in sustainable last mile logistics. These include (1) habits, such as last-minute, unbundled, small order size, or next day deliveries and (2) disruption factors such as possible collaborations of procurement department outside the project, the lack of time of employees for project tasks on top of their normal job and the lack of IT system that supports employees to improve their ordering behavior. Addressing these unforeseen barriers is needed towards improved participation of stakeholders in sustainable last mile logistics. Raising awareness can play a key role in addressing habits to move towards sustainable behavior [72,73]. For example, increased awareness could be achieved by adding a function in the IT system that informs employees of the consequences of 24/7 delivery such as added costs and more CO2 emission. In this way, employees become more aware of their actions and the disruption factor related to the lack of an IT system to support sustainable behavior can also be addressed. Addressing the disruption factor related to lack of time requires the organization’s support (time and ﬁnancial) to allow its employees to implement change, such as making the participation in the project part of their normal work. 6. Conclusions This study combines managerial and behavioral approaches in analyzing the goals of each stakeholder in relation to their role in the supply chain and within the organizations and the extent to which these goals lead (or do not lead) to successful participation in sustainable last mile logistics. Individual stakeholders in the living lab have economic, social, and environmental goals and some of these goals are prominent with certain groups of stakeholders. In general, some stakeholders share the same goals while others do not among supply chain actors and within the different management levels. Based on the results, there are three main conclusions derived from this study. First, if the sustainability goals of stakeholders are aligned along the supply chain and between the management levels, the more likely the organization will participate in sustainable last mile logistics. On a supply chain level, alignment could be achieved via information sharing and collaboration, requiring the use of the City hub on a tendering process, and intervention of receivers to help suppliers. On an organizational level, middle-level managers could play an important role as boundary spanners by translating strategic goals to operational levels, by inﬂuencing cultures at lower levels, and by bringing information and insights to both lower and upper levels. Second, inﬂuencing individual behavioral factors such as attitude, norms, and control, could lead individuals and organizations to set sustainability goals and to participate in sustainable last mile logistics. Strengthening the behavioral factors may include individual exposure to sustainability, training, and hiring personnel based on their afﬁnity to sustainability. Third, coping with unforeseen barriers such as habits and disruption factors could improve the sustainable behavior of an individual and eventually the organization as a whole. This could be done by increasing awareness, such as by adding an IT function that will inform employees of the consequences of their buying behavior, and by making the participation in sustainable last mile logistics part of an individual’s work. The limitations of this study include the use of only one case that focuses on one coun- try using solely qualitative analysis. For further research, the behavioral and managerial approaches done in this study can be expanded to other sustainable last mile logistics and Sustainability 2021, 13, 4434 15 of 17 15 of 17 supply chain projects in multiple areas inside and outside the Netherlands. Data Availability Statement: Not Applicable. Data Availability Statement: Not Applicable. Data Availability Statement: Not Applicable. Acknowledgments: The authors would like to thank the participants of the Heijendaal living lab and other organizations for their contribution to this research. Acknowledgments: The authors would like to thank the participants of the Heijendaal living lab and other organizations for their contribution to this research. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. European Commission. Sustainable & Smart Mobility Strategy; European Commission: Brussels, Belgium, 2020. 1. European Commission. Sustainable & Smart Mobility Strategy; European Commission: Brussels, Belgium, 2020. 2 Li dh l M U b F i h T f L l A h i P i A Li R i E T 2. Lindholm, M. Urban Freight Transport from a Local Authority Perspective—A Literature Review. Eu online: http://hdl.handle.net/10077/8869 (accessed on 10 April 2021). p p 3. Letnik, T.; Marksel, M.; Luppino, G.; Bardi, A.; Božiˇcnik, S. Review of Policies and Measures for Sustainable and Energy Efﬁcient Urban Transport. Energy 2018, 163, 245–257. [CrossRef] p gy 4. Nathanail, E.; Adamos, G.; Gogas, M. A Novel Approach for Assessing Sustainable City Logistics. Transp. Res. Procedia 2017, 25, 1036–1045. [CrossRef] 5. Van Duin, J.H.R.; Quak, H.; Muñuzuri, J. New Challenges for Urban Consolidation Centres: A Case Study in The Hague. Procedia-Soc. Behav. Sci. 2010, 2, 6177–6188. [CrossRef] 6. Gammelgaard, B. The Emergence of City Logistics: The Case of Copenhagen’s Citylogistik-Kbh. Int. J. Phys. Distrib. Logist. Manag. 2015, 45, 333–351. [CrossRef] D. A Theoretical Research Framework of Future Sustainable Urban Freight Transport for Smart Cities 2, 1975. [CrossRef] g 7. He, Z.; Haasis, H.-D. A Theoretical Research Framework of Future Sustainable Urban Freight Tra Sustainability 2020, 12, 1975. [CrossRef] 8. Letnik, T.; Farina, A.; Mencinger, M.; Lupi, M.; Božiˇcnik, S. Dynamic Management of Loading Bays for Energy Efﬁcient Urban Freight Deliveries. Energy 2018, 159, 916–928. [CrossRef] g gy 9. Dupas, R.; Taniguchi, E.; Deschamps, J.-C.; Qureshi, A.G. A Multi-Commodity Network Flow Model for Sustainable Performance Evaluation in City Logistics: Application to the Distribution of Multi-Tenant Buildings in Tokyo. Sustainability 2020, 12, 2180. [CrossRef] 10. Macharis, C.; Turcksin, L.; Lebeau, K. Multi Actor Multi Criteria Analysis (MAMCA) as a Tool to Support Sustainable Decisions: State of Use. Decis. Support Syst. 2012, 54, 610–620. [CrossRef] pp y 11. Harrington, T.S.; Singh Srai, J.; Kumar, M.; Wohlrab, J. Identifying Design Criteria for Urban System ‘Last-Mile’ Solutions—A Multi-Stakeholder Perspective. Prod. Plan. Control 2016, 27, 456–476. [CrossRef] p 12. Anand, N.; Quak, H.; van Duin, R.; Tavasszy, L. City Logistics Modeling Efforts: Trends and Gaps—A Review. Procedia-Soc. Behav. Sci. 2012, 39, 101–115. [CrossRef] 13. Van Duin, R.; Slabbekoorn, M.; Tavasszy, L.; Quak, H. Identifying Dominant Stakeholder Perspectives on Urban Freight Policies: A Q-Analysis on Urban Consolidation Centers in the Netherlands. Transport 2018, 33, 867–880. [CrossRef] 14. Marcucci, E.; Danielis, R. 6. Conclusions The qualitative analysis can also be extended to quantitative research, such as measuring the effects of behavioral factors on goals, which inﬂuences intention and behavior, and determining the weights of stakeholders’ criteria for participation using any Multi-Criteria Decision-Making tool. Further research may also include how TPB and Policy deployment could serve as instruments to align goals and inﬂuence decision-making in the organizations and supply chains. Others could also focus on how to overcome disruptions and habits in making decisions. Finally, as a consequence of this outcome to the living lab project, future studies could explore how to facilitate collaboration and information sharing within the living lab supply chain in order to improve the participation of various supply chain actors in the project. Author Contributions: Conceptualization, F.T.-Z., E.B. and L.v.d.S.; methodology, F.T.-Z. and E.B.; investigation, F.T.-Z. and L.v.d.S.; formal analysis, F.T.-Z.; writing—original draft preparation, F.T.-Z.; writing—review and editing, F.T.-Z., E.B. and L.v.d.S.; funding acquisition, E.B. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Netherlands Organization for Scientiﬁc Research (NWO) under the Living Lab Sustainable Supply Chain Management in Healthcare project (project number: 439.18.457). 15. Olsson, J.; Woxenius, J. Localisation of Freight Consolidation Centres Serving Small Road Hauliers in a Wider Urban Area: Barriers for More Efﬁcient Freight Deliveries in Gothenburg. J. Transp. Geogr. 2014, 34, 25–33. [CrossRef] References The Potential Demand for a Urban Freight Consolidation Centre. Transportation 2008, 35, 269–284. [CrossRef] 15. Olsson, J.; Woxenius, J. Localisation of Freight Consolidation Centres Serving Small Road Hauliers in a Wider Urban Area: Barriers for More Efﬁcient Freight Deliveries in Gothenburg. J. Transp. Geogr. 2014, 34, 25–33. [CrossRef] Sustainability 2021, 13, 4434 16 of 17 16 of 17 16. Buldeo Rai, H.; Verlinde, S.; Macharis, C. Shipping Outside the Box. Environmental Impact and Stakeholder Analysis of a Crowd Logistics Platform in Belgium. J. Clean. Prod. 2018, 202, 806–816. [CrossRef] g g 17. Westerlund, M.; Leminen, S. The Multiplicity of Research on Innovation through Living Labs. In Proceedings of the XXV ISPIM Conference—Innovation for Sustainable Economy & Society, Dublin, Ireland, 8–11 June 2014. g g 17. Westerlund, M.; Leminen, S. The Multiplicity of Research on Innovation through Living Labs. In Proceedings of the XXV ISPIM C f I ti f S t i bl E & S i t D bli I l d 8 11 J 2014 17. Westerlund, M.; Leminen, S. The Multiplicity of Research on Innovation through Living Labs. In Pr Conference—Innovation for Sustainable Economy & Society, Dublin, Ireland, 8–11 June 2014. y y 18. Nesterova, N.; Quak, H. A City Logistics Living Lab: A Methodological Approach. Transp. Res. Procedia 2016, 16, 403–417. [CrossRef] 18. Nesterova, N.; Quak, H. A City Logistics Living Lab: A Methodological Approach. Transp. Res [CrossRef] 19. Quak, H.; Lindholm, M.; Tavasszy, L.; Browne, M. From Freight Partnerships to City Logistics Living Labs—Giving Meaning to the Elusive Concept of Living Labs. Transp. Res. Procedia 2016, 12, 461–473. [CrossRef] 20. Gatta, V.; Marcucci, E.; Le Pira, M. Smart Urban Freight Planning Process: Integrating Desk, Living Lab and Modelling Approaches in Decision-Making. Eur. Transp. Res. Rev. 2017, 9, 32. [CrossRef] 21. Bogers, E. A Framework of Organizational and Behavioural Mode Choice Processes. 2017. Available online: https://www. researchgate.net/publication/346426231_A_framework_of_organizational_and_behavioural_mode_choice_processes. (accessed on 4 December 2020). 22. Vierth, I.; Lindgren, S.; Lobig, A.; Matteis, T.; Liedtke, G.; Burgschweiger, S.; Bogers, E.; Davydenko, I.; Burgess, A.; van de Ree, S. FALCON Handbook Understanding What Inﬂuences Modal Choice; CEDR: Brussels, Belgium, 2017. 23. Huang, H.; De Smet, Y.; Macharis, C.; Doan, N.A.V. Collaborative Decision-Making in Sustainable Mobility: Identifying Possible Consensuses in the Multi-Actor Multi-Criteria Analysis Based on Inverse Mixed-Integer Linear Optimization. Int. J. Sustain. Dev. World Ecol. 2021, 28, 64–74. [CrossRef] 24. Semanjski, I.; Gautama, S. References A Collaborative Stakeholder Decision-Making Approach for Sustainable Urban Logistics. Sustainability 2019, 11, 234. [CrossRef] 25. Buldeo Rai, H.; van Lier, T.; Meers, D.; Macharis, C. Improving Urban Freight Transport Sustainab Framework and Case Study. Res. Transp. Econ. 2017, 64, 26–35. [CrossRef] y p 26. Macharis, C.; Milan, L.; Verlinde, S. A Stakeholder-Based Multicriteria Evaluation Framework for City Distribution. Res. Transp. Bus. Manag. 2014, 11, 75–84. [CrossRef] g 27. Verlinde, S.; Macharis, C.; Milan, L.; Kin, B. Does a Mobile Depot Make Urban Deliveries Faster, More Sustainable and More Economically Viable: Results of a Pilot Test in Brussels. Transp. Res. Procedia 2014, 4, 361–373. [CrossRef] 28. Lauche, K. Insider Activists Pursuing an Agenda for Change: Selling the Need for Collaboration. In Managing Inter-Organizational Collaborations: Process Views; Research in the Sociology of Organizations; Emerald Publishing Limited: Bingley, UK, 2019; pp. 119–138. [CrossRef] pp [ ] 29. Kaplan, S.; Gruber, J.; Reinthaler, M.; Klauenberg, J. Intentions to Introduce Electric Vehicles in the Commercial Sector: A Model Based on the Theory of Planned Behaviour. Res. Transp. Econ. 2016, 55, 12–19. [CrossRef] y p 30. Lois, D.; Moriano, J.A.; Rondinella, G. Cycle Commuting Intention: A Model Based on Theory of Planned Behaviour and Social Identity. Transp. Res. Part F Trafﬁc Psychol. Behav. 2015, 32, 101–113. [CrossRef] 30. Lois, D.; Moriano, J.A.; Rondinella, G. Cycle Commuting Intention: A Model Ba Identity. Transp. Res. Part F Trafﬁc Psychol. Behav. 2015, 32, 101–113. [CrossRef] 31. Arslan, M.; ¸Sar, S. Examination of Environmentally Friendly “Green” Logistics Behavior of Managers in the Pharmaceutical Sector Using the Theory of Planned Behavior. Res. Soc. Adm. Pharm. 2018, 14, 1007–1014. [CrossRef] 32. Bogers, E. Trafﬁc Information and Learning in Day to Day Route Choice. Ph.D. Thesis, Delft University of Technology, Delft, The Netherlands, 2009. ke, H.; Christinck, A. From Assessing Knowledge to Joint Learning. In Cultivate Diversity! A Handbook on 33. Kaufmann, B.; Arpke, H.; Christinck, A. From Assessing Knowledge to Joint Learning. In Cultivate Transdisciplinary Approaches to Agrobiodiversity Research; Margraf Publishers: Weikersheim, Germany, 20 B.; Arpke, H.; Christinck, A. From Assessing Knowledge to Joint Learning. In Cultivate Diversity! A Hand ary Approaches to Agrobiodiversity Research; Margraf Publishers: Weikersheim, Germany, 2013; pp. 115–141. , ; , g g J g y proaches to Agrobiodiversity Research; Margraf Publishers: Weikersheim, Germany, 2013; pp. 115–141. g y pp 34. Grimble, R.; Wellard, K. References Stakeholder Methodologies in Natural Resource Management: A Review of Principles, Contexts, Experiences and Opportunities. Agric. Syst. 1997, 55, 173–193. [CrossRef] p pp g y 35. Taniguchi, E.; Tamagawa, D. Evaluating City Logistics Measures Considering the Behavior of Several Stakeholders. J. East. Asia Soc. Transp. Stud. 2005, 6, 3062–3076. [CrossRef] p 36. Russo, F.; Comi, A. City Characteristics and Urban Goods Movements: A Way to Environmental Transportation System in a Sustainable City. Procedia-Soc. Behav. Sci. 2012, 39, 61–73. [CrossRef] usso, F.; Comi, A. City Characteristics and Urban Goods Movements: A Way to Environmental Transport ; Comi, A. City Characteristics and Urban Goods Movements: A Way to Environmental Transportation S ble City Procedia-Soc Behav Sci 2012 39 61–73 [CrossRef] o, F.; Comi, A. City Characteristics and Urban Goods Movements: A Way to Environmental Transportatio inable City. Procedia-Soc. Behav. Sci. 2012, 39, 61–73. [CrossRef] nable City. Procedia-Soc. Behav. Sci. 2012, 39, 61–73. [Cr 37. Lelea, M.A. “All Relevant Stakeholders”: A Literature Review of Stakeholder Analysis to Support Inclusivity of Innovation Processes in Farming and Food Systems. In Proceedings of the 12th European IFSA Symposium, Newport, UK, 12–15 July 2015; pp. 1–10. pp 38. Lee, R.G.; Dale, B.G. Policy Deployment: An Examination of the Theory. Int. J. Qual. Reliab. Manag. 1998, 15, 520–540. [CrossRef] 39 K l R N t D Th B l d S d M Th t D i P f H B R 1992 70 71 79 [P bM d] pp 38. Lee, R.G.; Dale, B.G. Policy Deployment: An Examination of the Theory. Int. J. Qual. Reliab. Manag. 1998, 15, 520–540. [CrossRef] 39. Kaplan, R.; Norton, D. The Balanced Scorecard—Measures That Drive Performance. Harv. Bus. Rev. 1992, 70, 71–79. [PubMed] pp 38. Lee, R.G.; Dale, B.G. Policy Deployment: An Examination of the Theory. Int. J. Qual. Reliab. Manag. 199 R.G.; Dale, B.G. Policy Deployment: An Examination of the Theory. Int. J. Qual. Reliab. Manag. 1998, 15, 520–5 n, R.; Norton, D. The Balanced Scorecard—Measures That Drive Performance. Harv. Bus. Rev. 1992, 70, 71– 39. Kaplan, R.; Norton, D. The Balanced Scorecard—Measures That Drive Performance. Harv. Bus. Rev. 19 40. Chardine-Baumann, E.; Botta-Genoulaz, V. A Framework for Sustainable Performance Assessment of Supply Chain Management Practices. Comput. Ind. Eng. 2014, 76, 138–147. [CrossRef] 41. Cash, R.; Wilkerson, T. GreenSCOR: Developing a Green Supply Chain Analytical Tool; Logistics Management Institute: Washington, DC, USA, 2003. ntions to Actions: A Theory of Planned Behavior. References [CrossRef] 50. Tölkes, C.; Butzmann, E. Motivating Pro-Sustainable Behavior: The Potential of Green Events—A C Streetlife Festival. Sustainability 2018, 10, 3731. [CrossRef] n, E. Motivating Pro-Sustainable Behavior: The Potential of Green Events—A Case-Study from the Munich ustainability 2018, 10, 3731. [CrossRef] y 51. Cantor, D.E.; Morrow, P.C.; McElroy, J.C.; Montabon, F. The Role of Individual and Organizational Factors in Promoting Firm Environmental Practices. Int. J. Phys. Distrib. Logist. Manag. 2013, 43, 407–427. [CrossRef] 52. Gollwitzer, P.M. Goal Achievement: The Role of Intentions. Eur. Rev. Soc. Psychol. 1993, 4, 141–185. [CrossRef] ettingen, G.; Gollwitzer, P.M. Strategies of Setting and Implementing Goals: Mental Contrasting and Impleme Social Psychological Foundations of Clinical Psychology; Maddux, J.J.E., Ed.; Guilford: New York, NY, USA, 20 R K C St d R h D i d M th d 5th d S Ri id CA USA 2009 ial Psychological Foundations of Clinical Psychology; Maddux, J.J.E., Ed.; Guilford: New York, NY, USA, 2010; p R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. In Social Psychological Foundations of Clinical Psychology; Maddux, J.J.E., Ed.; Guilford: New York, NY, USA, 2010; pp. 114–135. 54. Yin, R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. In Social Psychological Foundations of Clinical Psychology; Maddux, J.J.E., Ed.; Guilford: New York, NY, USA, 2010; pp. 114–135. 54. Yin, R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. y g f y gy; , J J , ; , , , ; p R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. y g f y gy pp 54. Yin, R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. 54. Yin, R.K. Case Study Research: Design and Methods, 5th ed.; Sage: Riverside, CA, USA, 2009. y g g 55. Njie, B.; Asimiran, S. Case Study as a Choice in Qualitative Methodology. IOSR J. Res. Method Educ. 2014, 4, 35–40. [CrossRef] 56. Eisenhardt, K.M. Building Theories from Case Study Research. Acad. Manag. Rev. 1989, 14, 532–550. [CrossRef] j y gy 56. Eisenhardt, K.M. Building Theories from Case Study Research. Acad. Manag. Rev. 1989, 14, 532–550. [CrossRef] 57. Fishbach, A.; Ferguson, M.J. The Goal Construct in Social Psychology. In Social Psychology: Handboo Guilford Press: New York, NY, USA, 2007; pp. 490–515. 58. Ganzevles, J.; Potting, J.; Hanemaaijer, A. References In Action-Control: From Cognition to Behavior; Springer ny, 1985; pp. 11–39. 42. Ajzen, I. From Intentions to Actions: A Theory of Planned Behavior. In Action-Control: From Cognit Heidelberg, Germany, 1985; pp. 11–39. y pp Theory of Planned Behavior. Organ. Behav. Hum. Decis. Process. 1991, 50, 179–211. [CrossRef] 43. Ajzen, I. The Theory of Planned Behavior. Organ. Behav. Hum. Decis. Process. 1991, 50, 179–211. [CrossRef] j , y g , , [ ] 44. Aarts, H.; Dijksterhuis, A. The Automatic Activitation of Goal-Directed Behaviour: The Case of Travel Habit. J. Environ. Psychol. 2000, 20, 75–82. [CrossRef] 17 of 17 17 of 17 Sustainability 2021, 13, 4434 45. Bogers, E.A.I.; Viti, F.; Hoogendoorn, S.P. Joint Modeling of Advanced Travel Information Service, Habit, and Learning Impacts on Route Choice by Laboratory Simulator Experiments. Transp. Res. Rec. J. Transp. Res. Board 2005, 1926, 189–197. [CrossRef] 46. Schwanen, T.; Banister, D.; Anable, J. Rethinking Habits and Their Role in Behaviour Change: The Case of Low-Carbon Mobility. J. Transp. Geogr. 2012, 24, 522–532. [CrossRef] 45. Bogers, E.A.I.; Viti, F.; Hoogendoorn, S.P. Joint Modeling of Advanced Travel Information Service, Habit, and Learning Impacts on Route Choice by Laboratory Simulator Experiments. Transp. Res. Rec. J. Transp. Res. Board 2005, 1926, 189–197. [CrossRef] y y p p J p , , [ ] 46. Schwanen, T.; Banister, D.; Anable, J. Rethinking Habits and Their Role in Behaviour Change: The Case of Low-Carbon Mobility. J. Transp. Geogr. 2012, 24, 522–532. [CrossRef] J. Transp. Geogr. 2012, 24, 522–532. [CrossRef] J p g [ ] 47. Klöckner, C.A.; Matthies, E. How Habits Interfere with Norm-Directed Behaviour: A Normative Decision-Making Model for Travel Mode Choice. J. Environ. Psychol. 2004, 24, 319–327. [CrossRef] p g 47. Klöckner, C.A.; Matthies, E. How Habits Interfere with Norm-Directed Behaviour: A Normative Dec Travel Mode Choice. J. Environ. Psychol. 2004, 24, 319–327. [CrossRef] Klöckner, C.A.; Matthies, E. How Habits Interfere with Norm-Directed Behaviour: A Normative Decisio Travel Mode Choice. J. Environ. Psychol. 2004, 24, 319–327. [CrossRef] y 48. Montani, F.; Odoardi, C.; Battistelli, A. Individual and Contextual Determinants of Innovative Work Behaviour: Proactive Goal Generation Matters. J. Occup. Organ. Psychol. 2014, 87, 645–670. [CrossRef] 49. Dholakia, U.M.; Bagozzi, R.P. Mustering Motivation to Enact D Realization. J. Behav. Decis. Mak. 2002, 15, 167–188. [CrossRef] 49. Dholakia, U.M.; Bagozzi, R.P. Mustering Motivation to Enact Decisions: How Decision Process Characteristics Inﬂuence Goal Realization. J. Behav. Decis. Mak. 2002, 15, 167–188. References Evaluation of Green Deals for a Circular Economy; PBL Publishers: The Hague, The Netherlands, 2017. 59. Kaiser, M.G.; El Arbi, F.; Ahlemann, F. Successful Project Portfolio Management beyond Project Selection Techniques: Under- standing the Role of Structural Alignment. Int. J. Proj. Manag. 2015, 33, 126–139. [CrossRef] g g j g 60. Mikalef, P.; Pateli, A.; Batenburg, R.; van de Wetering, R. Investigating the Impact of Procurement Alignment on Supply Chain Management Performance. Procedia Technol. 2013, 9, 310–319. [CrossRef] 61. De Souza, R.; Goh, M.; Lau, H.-C.; Ng, W.-S.; Tan, P.-S. Collaborative Urban Logistics—Synchronizing the Last Mile a Singapore Research Perspective. Procedia-Soc. Behav. Sci. 2014, 125, 422–431. [CrossRef] 62. Wu, I.-L.; Chuang, C.-H.; Hsu, C.-H. Information Sharing and Collaborative Behaviors in Enabling Supply Chain Performance: A Social Exchange Perspective. Int. J. Prod. Econ. 2014, 148, 122–132. [CrossRef] g p 63. Zhao, Z.J.; Anand, J. Beyond Boundary Spanners: The ‘Collective Bridge’ as an Efﬁcient Interunit S Collective Knowledge. Strateg. Manag. J. 2013, 34, 1513–1530. [CrossRef] 64. Tushman, M.L.; Scanlan, T.J. Boundary Spanning Individuals: Their Role in Information Transfer and Their Antecedents. Acad. Manag. J. 1981, 24, 289–305. [CrossRef] g 65. Paarlberg, L.E.; Perry, J.L. Values Management: Aligning Employee Values and Organization Goals. Am. Rev. Public Adm. 2007, 37, 387–408. [CrossRef] 66. Gattiker, T.F.; Carter, C.R. Understanding Project Champions’ Ability to Gain Intra-Organizational Commitment for Environmental Projects. J. Oper. Manag. 2010, 28, 72–85. [CrossRef] 67. Sparkman, G.; Walton, G.M. Dynamic Norms Promote Sustainable Behavior, Even If It Is Counternormative. Psychol. Sci. 2017, 28, 1663–1674. [CrossRef] [PubMed] E.F. Exploring the Effects of a Product’s Sustainability Triggers on pro-Environmental Decision-Making. J 0, 011102. [CrossRef] 68. She, J.; MacDonald, E.F. Exploring the Effects of a Product’s Sustainability Triggers on pro-Environme Mech. Des. 2018, 140, 011102. [CrossRef] Mech. Des. 2018, 140, 011102. [CrossRef] 69. Jerónimo, H.M.; Henriques, P.L.; de Lacerda, T.C.; da Silva, F.P.; Vieira, P.R. Going Green and Sustainable: The Inﬂuence of Green HR Practices on the Organizational Rationale for Sustainability J Bus Res 2020 112 413 421 [CrossRef] , , [ ] 69. Jerónimo, H.M.; Henriques, P.L.; de Lacerda, T.C.; da Silva, F.P.; Vieira, P.R. Going Green and Sustainable: The Inﬂuence of Green HR Practices on the Organizational Rationale for Sustainability. J. Bus. Res. 2020, 112, 413–421. [CrossRef] 69. Jerónimo, H.M.; Henriques, P.L.; de Lacerda, T.C.; da Silva, F.P.; Vieira, P.R. Going Green and Sustainabl HR Practices on the Organizational Rationale for Sustainability. J. Bus. References Res. 2020, 112, 413–421. [CrossR nriques, P.L.; de Lacerda, T.C.; da Silva, F.P.; Vieira, P.R. Going Green and Sustainable: The Inﬂuence of Green Organizational Rationale for Sustainability. J. Bus. Res. 2020, 112, 413–421. [CrossRef] 70. Marsden, G.; Docherty, I. Insights on Disruptions as Opportunities for Transport Policy Change. Transp. Res. Part A Policy Pract. 2013, 51, 46–55. [CrossRef] alker, I.; Thomas, G.O.; Verplanken, B. Old Habits Die Hard. Environ. Behav. 2015, 47, 1089–1106. [CrossRef] 71. Walker, I.; Thomas, G.O.; Verplanken, B. Old Habits Die Hard. Environ. Behav. 2015, 47, 1089–1106. [CrossRef] C Y h ﬂ Ch h d G d d h h f C 72. Lin, P.-C.; Huang, Y.-H. The Inﬂuence Factors on Choice Behavior Regarding Green Products Based on Values. J. Clean. Prod. 2012, 22, 11–18. [CrossRef] 73. Buerke, A.; Straatmann, T.; Lin-Hi, N.; Müller, K. Consumer Awareness and Sustainability-Focused Value Orientation as Motivating Factors of Responsible Consumer Behavior. Rev. Manag. Sci. 2017, 11, 959–991. [CrossRef]
https://openalex.org/W4293688479	https://pubs.rsc.org/en/content/articlepdf/2022/ta/d2ta90172b	English	null	Correction: Effect of structure on oxygen diffusivity in layered oxides: a combined theoretical and experimental study	Journal of materials chemistry. A	2,022	cc-by	375	aDepartment of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. E-mail: tuller@mit.edu bJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA cMaterials Architecturing Research Center, Korea Institute of Science and Technology, 5 Hwarang-ro 14-gil Seongbuk-gu, Seoul 02792, Republic of Korea dUniversit´e de Nantes, CNRS, Institut des Mat´eriaux Jean Rouxel, IMN, Nantes, France 17424 \| J. Mater. Chem. A, 2022, 10, 17424 CORRECTION Open Access Article. Published on 19 August 2022. Downloaded on 10/24/2024 5:37:25 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Correction: Eﬀect of structure on oxygen diﬀusivity in layered oxides: a combined theoretical and experimental study Cite this: J. Mater. Chem. A, 2022, 10, 17424 Cite this: J. Mater. Chem. A, 2022, 10, 17424 ChangSub Kim,ab Kyoung-Won Park,c Dmitri Kalaev,a Clement Nicolletad and Harry L Tuller*a Correction for ‘Eﬀect of structure on oxygen diﬀusivity in layered oxides: a combined theoretical and experimental study’ by ChangSub Kim et al., J. Mater. Chem. A, 2022, 10, 15402–15414, https://doi.org/ 10.1039/D2TA02580A. The authors regret that the values of d in the last paragraph of Section 3.2 in the published article are incorrect. Specically, the sentence “To describe the defects in experimentally prepared T- and T0-La2CuO4,31,32 we chose single and paired oxygen defects in eight unit cells of T- and T0-La2CuO4d (i.e., d ¼ 1/64 and 1/32).” in Section 3.2 should instead read as follows: “To describe the defects in experimentally prepared T- and T0-La2CuO4,31,32 we chose single and paired oxygen defects in eight unit cells of T- and T0-La2CuO4d (i.e., d ¼ 1/16 and 1/8).” The authors conrm that the analysis and discussion were based on the correct values of d and that these errors do not aﬀect the overall conclusions of the article. The authors conrm that the analysis and discussion were based on the correct values of d and that these errors do not aﬀect the overall conclusions of the article. The Royal Society of Chemistry apologises for these errors and any consequent inconvenience to authors and readers. The Royal Society of Chemistry apologises for these errors and any consequent inconvenience to authors and readers. This journal is © The Royal Society of Chemistry 2022 17424 \| J. Mater. Chem. A, 2022, 10, 17424
https://openalex.org/W4283805330	https://www.scielo.br/j/ce/a/hyWZ5M4pwy6rTZ9J7VCfTKt/?lang=en&format=pdf	English	null	Synthesis and characterization of α-SnWO4 powders obtained by microwave-assisted hydrothermal method	Cerâmica	2,022	cc-by	4,405	INTRODUCTION times are usually required. y q The use of the MAH method for the synthesis of oxides was introduced in 1992 by Komarneni et al. [13]. This method provides a fast and efficient heating approach, which reduces processing time, increases yield, and improves product purity [14]. According to the literature, only one work was found on the preparation of α-SnWO4 using this method [15]. In this work, Su et al. [15] synthesized α-Sn1-xZnxWO4 (x=0, 0.02, 0.05, 0.07, 0.10, and 0.12) nanocrystals using a MAH approach at 180 °C for 10 min. This work had the main objective of studying the effects of Zn2+ doping on the structural, electronic, and photocatalytic properties of α-SnWO4. However, we have found no research literature detailing the influence of the MAH method on the photoluminescent properties and formation of pure α-SnWO4. Therefore, in our work, the main objective was to study the influence of different exposure times to microwave radiation on the structural, morphological, and optical properties of the α-SnWO4. The α-SnWO4 was obtained by the MAH method at a low temperature (150 °C) for 16, 32, and 64 min. Then, the powders were characterized by X-ray diffraction (XRD), Rietveld refinements, Raman microscopy, field emission gun-scanning electron microscopy (FE- SEM), UV/vis diffuse reflectance spectroscopy (DRS), and photoluminescence spectroscopy (PL). Tin tungstate (SnWO4) has been highlighted in recent decades due to its unique structural and optical properties, showing potential for applications such as gas-sensing [1, 2], luminescent devices [3, 4], and visible-light-driven photocatalyst [4, 5]. It can be obtained in two different crystalline structures, including the cubic (β-SnWO4) and orthorhombic (α-SnWO4) phases. Both have a low band gap (Egap), with values around 2.68 eV (β-SnWO4) and 1.64 eV (α-SnWO4). The α-SnWO4 is the most studied due to being a low-temperature phase, which was first synthesized by a solid-state reaction in 1972 by Jeitschko and Sleight [6], and, in 1974, its properties were investigated by the same group [7]. The solid-state reaction is a traditional method used in the preparation of α-SnWO4 [8-10], which involves mixing solid reactants and subsequent calcination at high temperatures. Despite its simplicity and low cost, this method has disadvantages, such as non-uniformity of morphology, high agglomeration, large particle size, formation of secondary phases, and lack of reproducibility [11]. To overcome these disadvantages, other processing approaches, such as hydrothermal and microwave-assisted hydrothermal (MAH) methods have been employed to obtain α-SnWO4. Abstract The structural, morphological, and optical properties of α-SnWO4 synthesized by microwave-assisted hydrothermal method at 150 °C for different times were investigated. XRD revealed the formation of the orthorhombic phase, without the presence of a secondary phase. The sample processed for 64 min showed a reduction in unit cell volume, suggesting a decrease in the structural defects present in the sample. The Raman spectroscopy confirmed the formation of α-SnWO4, with the presence of vibrational modes attributed to the [WO6] octahedron. FE-SEM images revealed the formation of agglomerate nanosheets in all samples. The band gap (Egap) of the samples presented values in the range from 1.58 to 1.70 eV. The increase in Egap value of the sample processed for 64 min corroborated the decrease in the structural defects suggested by the Rietveld refinement. Finally, the photoluminescence was measured, with maximum emission in the blue region of the electromagnetic spectrum around 454 nm, which was attributed to the [WO6] octahedron.i 6 Keywords: α-SnWO4, photoluminescence, synthesis, Rietveld refinement. Synthesis and characterization of α-SnWO4 powders obtained by microwave-assisted hydrothermal method M. M. P. Barros1, K. C. Almeida1, S. A. Silva1, G. Botelho1* 1Federal University of Tocantins, Department of Environmental Chemistry, 77402-970, Gurupi, TO, Brazil M. M. P. Barros1, K. C. Almeida1, S. A. Silva1, G. Botelho1* y of Tocantins, Department of Environmental Chemistry, 77402-970, Gurupi, TO, Brazil INTRODUCTION For example, nanoplates of α-SnWO4 were prepared with good dispersity by hydrothermal method at 200 °C for 24 h, without any template [12]. Meanwhile, Liu et al. [5] prepared α-SnWO4 nanoparticles with different morphologies (nanostrip to nanosheet) also by hydrothermal method, changing the water/ethylene glycol ratio. Despite the efficiency of the hydrothermal method, long synthesizing http://dx.doi.org/10.1590/0366-69132022683863259 http://dx.doi.org/10.1590/0366-69132022683863259 Cerâmica 68 (2022) 236-241 236 *gleice.lorena@uft.edu.br https://orcid.org/0000-0002-7591-1381 EXPERIMENTAL Synthesis of the powders: SnWO4 powders were obtained by the MAH method using different times. Initially, 1x10-3 mol of tin chloride dihydrate (SnCl2.2H2O, 98.0% purity, Sigma-Aldrich) and 1x10-3 mol of sodium tungstate dihydrate (Na2WO4.2H2O, 99.0% purity, Sigma-Aldrich) were separately dissolved in 1 and 5 mL of distilled water, respectively. Both solutions were mixed, resulting in a dark- M. M. P. Barros et al. / Cerâmica 68 (2022) 236-241 237 red color precipitate. This resulting suspension was diluted to a final volume of 100 mL and then transferred into a Teflon autoclave, that was sealed and placed inside a modified domestic microwave oven of 2.45 GHz of radiation with a maximum output power of 800 W [16]. In the sequence, the suspension was processed at 150 °C for 16, 32, and 60 min, respectively. After microwave processing, the suspension obtained was washed with distilled water and dried in an oven, resulting in a brown-colored precipitate. red color precipitate. This resulting suspension was diluted to a final volume of 100 mL and then transferred into a Teflon autoclave, that was sealed and placed inside a modified domestic microwave oven of 2.45 GHz of radiation with a maximum output power of 800 W [16]. In the sequence, the suspension was processed at 150 °C for 16, 32, and 60 min, respectively. After microwave processing, the suspension obtained was washed with distilled water and dried in an oven, resulting in a brown-colored precipitate. profile was described by Thompson-Cox-Hastings pseudo- Voigt function. In the refinements, the Inorganic Crystal Structure Database (ICSD) No. 2147 [7] was used. Micro- Raman spectroscopy was performed from 50 to 1300 cm-1 using a spectrometer (iHR550, Horiba Jobin-Yvon) coupled to a CCD detector and an argon-ion laser operating at 514 nm. Micrographs of the samples were obtained in a field-emission scanning electron microscope (FE-SEM, Supra 35-VP, Carl Zeiss). The UV/vis diffuse reflectance spectroscopy (DRS) was obtained using a spectrophotometer (Cary 5G, Varian) from 200 to 800 nm. Photoluminescence (PL) measurements were performed with a monochromator (Monospec 27, Thermal Jarrell Ash) coupled to a photomultiplier (R446, Hamamatsu Photonics), with a krypton ion laser (Innova 200 K, Coherent) with emission at λ=350 nm. Characterization: the samples were characterized by XRD, using a diffractometer (D/Max-2500PC, Rigaku) with CuKα radiation in a scan from 15° to 105° and a scanning speed of 0.02 °/s. EXPERIMENTAL The Rietveld refinements were employed, using the General Structure Analysis System (GSAS) software and EXPGUI graphical interface. The following parameters have been refined: scale factor, background modeled using the Chebyshev polynomial of the first kind, shift lattice constants, profile half-width parameters (u, v, w), isotropic thermal parameters, lattice parameters, strain anisotropy factor, and preferential orientation. The peak RESULTS AND DISCUSSION Fig. 1a shows the XRD patterns of the SnWO4 powders processed in a MAH for 16, 32, and 60 min. All samples Figure 1: XRD patterns of α-SnWO4 samples (a) and Rietveld refinements plots of α-SnWO4 processed in MAH for 16 min (b), 32 min (c and 64 min (d). Inset in (d) represents the schematic representation of the orthorhombic structure of α-SnWO4. 2q (degree) 2q (degree) 2q (degree) 2q (degree) Intensity Intensity Intensity Intensity 15 15 15 15 45 45 45 45 85 85 85 85 25 25 25 25 55 55 55 55 95 95 95 95 35 35 35 35 75 75 75 75 65 65 65 65 105 105 105 105 30 30 30 30 70 70 70 70 60 60 60 60 100 100 100 100 20 20 20 20 50 50 50 50 90 90 90 90 40 40 40 40 80 80 80 80 a) c) d) b) 2q (degree) Intensity 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 b) 2q (degree) Intensity 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 a) b) Intensity 2q (degree) 2q (degree) 2q (degree) Intensity 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 c) 2q (degree) Intensity 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 d) d) c) Intensity 2q (degree) 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 2q (degree) 15 45 85 25 55 95 35 75 65 105 30 70 60 100 20 50 90 40 80 Figure 1: XRD patterns of α-SnWO4 samples (a) and Rietveld refinements plots of α-SnWO4 processed in MAH for 16 min (b), 32 min (c), and 64 min (d). Inset in (d) represents the schematic representation of the orthorhombic structure of α-SnWO4. M. M. P. Barros et al. RESULTS AND DISCUSSION / Cerâmica 68 (2022) 236-241 238 Sample Lattice parameter (Å) Unit cell volume (Å3) Rp (%) Rwp (%) χ2 a b c α-SnWO4 - 16 min 5.5872(1) 11.7865(0) 4.9376(6) 325.16(4) 6.57 8.29 1.48 α-SnWO4 - 32 min 5.5939(3) 11.7555(7) 4.9530(6) 325.71(3) 6.54 8.29 1.51 α-SnWO4 - 64 min 5.5905(1) 11.7185(0) 4.9530(1) 324.48(4) 9.44 11.80 1.18 α-SnWO4 - ICSD 2147 5.6270(3) 11.6486(7) 4.9973(3) 327.56(0) - - - Table I - Rietveld refinement results obtained for α-SnWO4 samples. Sample W-O Sn-O α-SnWO4 - 16 min 2.1249(2) 1.7888(2) 1.9032(4) 2.8394(5) 2.3860(3) 2.1755(2) α-SnWO4 - 32 min 2.1282(2) 1.7918(1) 1.9001(3) 2.8365(4) 2.3867(2) 2.1777(2) α-SnWO4 - 64 min 2.1270(2) 1.7908(2) 1.8951(4) 2.8302(5) 2.3837(3) 2.1753(2) α-SnWO4 - ICSD 2147 2.141 1.802 1.889 2.826 2.392 2.184 Table II - Bond lengths results (Å) obtained for α-SnWO4 samples. Table I - Rietveld refinement results obtained for α-SnWO4 samples. Sample W-O Sn-O α-SnWO4 - 16 min 2.1249(2) 1.7888(2) 1.9032(4) 2.8394(5) 2.3860(3) 2.1755(2) α-SnWO4 - 32 min 2.1282(2) 1.7918(1) 1.9001(3) 2.8365(4) 2.3867(2) 2.1777(2) α-SnWO4 - 64 min 2.1270(2) 1.7908(2) 1.8951(4) 2.8302(5) 2.3837(3) 2.1753(2) α-SnWO4 - ICSD 2147 2.141 1.802 1.889 2.826 2.392 2.184 Table II - Bond lengths results (Å) obtained for α-SnWO4 samples. Table II - Bond lengths results (Å) obtained for α-SnWO4 samples. agreement with the ICSD Nº 2147 [7] and with the previous reports [4, 8-10]. were well indexed with the orthorhombic structure of α-SnWO4 with space group Pnna, according to the ICSD No. 2147 [7]. The diffractograms have well-defined peaks, indicating a high degree of long-range periodicity; however, the XRD patterns presented broad peaks, suggesting the presence of structural defects arising from the synthesis method and/or the experimental conditions used [17]. The Rietveld refinement method was employed to obtain crystallographic information from the samples. The refinements results (Figs. 1b to 1d) confirmed the formation of the orthorhombic structure of α-SnWO4 without the formation of secondary phases in the samples. The profile acquired by Rietveld refinements (calculated) was in agreement with the experimental profiles (observed), as showed by the residual line (Obs-Calc). This information was supported by the statistical fitting parameters illustrated in Table I (Rp<9.44%, Rwp<11.80%, and χ2 <1.51), which indicated that the quality of refinements was acceptable. In all samples, the values of the lattice parameters and unit cell volume were in accordance with the ICSD Nº 2147 [7] and with the previous reports [5, 9]. RESULTS AND DISCUSSION The unit cell volume and lattice parameters’ values decreased for the sample processed for 64 min, suggesting that the time of exposure to the microwave radiation exerted an effect on the structural parameters of α-SnWO4, decreasing the structural defects present in the sample. Results obtained from the Rietveld refinements were used as input data in the Diamond, a crystal and molecular structure visualization, software to model the schematic representation of the orthorhombic structure of α-SnWO4. As represented in the inset of Fig. 1d, the α-SnWO4 is composed of W coordinated to the six oxygen, arranged in the form of a two-dimensional sheet separated by layers of Sn2+ ions, that are also coordinated to six oxygens. The WO6 clusters within a sheet are connected by four corners. The WO6 and SnO6 clusters are highly distorted on the lattice of the α-SnWO4, since there are three different W-O bond lengths and three different Sn-O bond lengths, as illustrated in Table II. The obtained results are in were well indexed with the orthorhombic structure of α-SnWO4 with space group Pnna, according to the ICSD No. 2147 [7]. The diffractograms have well-defined peaks, indicating a high degree of long-range periodicity; however, the XRD patterns presented broad peaks, suggesting the presence of structural defects arising from the synthesis method and/or the experimental conditions used [17]. The Rietveld refinement method was employed to obtain crystallographic information from the samples. The refinements results (Figs. 1b to 1d) confirmed the formation of the orthorhombic structure of α-SnWO4 without the formation of secondary phases in the samples. The profile acquired by Rietveld refinements (calculated) was in agreement with the experimental profiles (observed), as showed by the residual line (Obs-Calc). This information was supported by the statistical fitting parameters illustrated in Table I (Rp<9.44%, Rwp<11.80%, and χ2 <1.51), which indicated that the quality of refinements was acceptable. In all samples, the values of the lattice parameters and unit cell volume were in accordance with the ICSD Nº 2147 [7] and with the previous reports [5, 9]. The unit cell volume and lattice parameters’ values decreased for the sample processed for 64 min, suggesting that the time of exposure to the microwave radiation exerted an effect on the structural parameters of α-SnWO4, decreasing the structural defects present in the sample. RESULTS AND DISCUSSION Results obtained from the Rietveld refinements were used as input data in the Diamond, a crystal and molecular structure visualization, software to model the schematic representation of the orthorhombic structure of α-SnWO4. As represented in the inset of Fig. 1d, the α-SnWO4 is composed of W coordinated to the six oxygen, arranged in the form of a two-dimensional sheet separated by layers of Sn2+ ions, that are also coordinated to six oxygens. The WO6 clusters within a sheet are connected by four corners. The WO6 and SnO6 clusters are highly distorted on the lattice of the α-SnWO4, since there are three different W-O bond lengths and three different Sn-O bond lengths, as illustrated in Table II. The obtained results are in Raman spectroscopy was used to complement the structural characterization of the α-SnWO4 samples (Fig. 2). The presence of Raman modes in all samples revealed that the α-SnWO4 presents short-range periodicity, regardless of the processing time. The spectra were very close to previous reports of α-SnWO4, showing a strong Raman mode at approximately 778 cm-1, and three Raman modes of low intensity at approximately 405, 288, and 198 cm-1 [8, 18, 19]. Kuzmin et al. [18] attributed the band at 778 cm-1 to the stretching O-W-O vibrations of the WO6 clusters. Moreover, all Raman spectra consisted of wide bands, typically associated with the presence of structural defects. The morphologies of α-SnWO4 samples were investigated by FE-SEM microscopy (Fig. 3). All samples of α-SnWO4 showed similar morphology, with the formation of highly agglomerated nanosheets with irregular shapes. It is possible to observe that these nanosheets aggregated, forming agglomerates. Wang et al. [20] investigated the Figure 2: Raman spectra of α-SnWO4 samples. Intensity 200 600 1000 400 800 1200 300 700 1100 500 900 1300 Figure 2: Raman spectra of α-SnWO4 samples. 200 600 1000 400 800 1200 300 700 1100 500 900 1300 M. M. P. Barros et al. / Cerâmica 68 (2022) 236-241 239 Figure 3: FE-SEM images of α-SnWO4 samples processed for 16 min (a,b), 32 min (c,d), and 64 min (e,f). Figure 3: FE-SEM images of α-SnWO4 samples processed for 16 min (a,b), 32 min (c,d), and 64 min (e,f). RESULTS AND DISCUSSION To estimate the band gap (Egap) of the samples, the Tauc equation [αhν=A(hν-Egap)n] [21] was employed, where α is the absorption coefficient, h the Planck constant, ν the frequency, A is a proportionality constant, and the exponent n a constant that depends on the type of electronic transition present in the semiconductor. The absorption coefficient is proportional to the Kubelka-Munk function [F(R∞)], resulting in the equation [F(R∞)hν]1/n=A(hν-Egap) [22]. According to the previous reports, the SnWO4 powders present an indirect allowed transition with n=2 [5, 9, 20, 23]. Then, the Egap of the samples was obtained by extrapolating the plots of morphologies of α-SnWO4 prepared in a different solvent by the solvothermal method. The morphologies showed the formation of nanosheets and agglomerates, suggesting that the solvent is an important factor in tuning the morphology. The α-SnWO4 obtained using only water as solvent presented well-defined nanosheets. The high agglomeration of particles observed in our study may be related to the processing of the sample, under microwave radiation. The microwave radiation accelerates the nucleation and growth process, favoring the aggregation of the particles [17]. Similar morphology of α-SnWO4 was observed in previous studies [4, 15]. morphologies of α-SnWO4 prepared in a different solvent by the solvothermal method. The morphologies showed the formation of nanosheets and agglomerates, suggesting that the solvent is an important factor in tuning the morphology. The α-SnWO4 obtained using only water as solvent presented well-defined nanosheets. The high agglomeration of particles observed in our study may be related to the processing of the sample, under microwave radiation. The microwave radiation accelerates the nucleation and growth process, favoring the aggregation of the particles [17]. Similar morphology of α-SnWO4 was observed in previous studies [4, 15]. M. M. P. Barros et al. / Cerâmica 68 (2022) 236-241 240 [F(R∞)hν]1/2 versus hν (photon energy). Fig. 4 shows the Egap of the samples processed at different times. The Egap value increased for the sample processed for 64 min, indicating a reduction in the structural defects with increasing processing time, which was in accordance with the results observed in the Rietveld refinements (Table I). Moreover, these band gap values were in accordance with the previous reports of α-SnWO4, which typically have values in the range of 1.53- 1.76 eV [5, 9, 20, 23]. presence of deep levels within the band gap, favoring the red emission. CONCLUSIONS The α-SnWO4 was successfully synthesized by the microwave-assisted hydrothermal (MAH) method, even using low temperature (150 °C) and short processing times (16, 32, and 64 min), showing the efficiency of microwave heating. The XRD and Rietveld refinements confirmed the formation of the orthorhombic phase, composed of highly distorted WO6 and SnO6 clusters. The increase in microwave processing time led to slight lattice contraction, suggesting a decrease in the structural defects. This observation was corroborated by the increase in Egap values observed with the increase in processing time. The morphology composed of agglomerated nanosheets was maintained in all samples. The photoluminescence (PL) emission was observed in the blue region in all samples, which was attributed to the WO6 cluster. These results showed that synthesis methods and processing conditions directly influence the structural, optical, and morphological properties of the α-SnWO4. Photon energy (eV) 3.2 2.4 1.8 1.6 6 4 2 0 2.8 2.0 3.4 2.6 3.0 2.2 Figure 4: Graph showing the determination of band gap energy of α-SnWO4 samples. [F(R∞)hν]1/2 (a.u.) Figure 5: PL spectra of α-SnWO4 samples. Wavelength (nm) Intensity 400 550 350 500 650 450 600 700 750 Photon energy (eV) 3.2 2.4 1.8 1.6 6 4 2 0 2.8 2.0 3.4 2.6 3.0 2.2 Fi 4 G h h i h d i i f b d f [F(R∞)hν]1/2 (a.u.) ACKNOWLEDGMENTS The authors acknowledge the CNPq, and Liec-CDMF- UFSCar. Photon energy (eV) Photon energy (eV) gy ( ) Figure 4: Graph showing the determination of band gap energy of α-SnWO4 samples. gy ( ) Figure 4: Graph showing the determination of band gap energy of α-SnWO4 samples. RESULTS AND DISCUSSION Therefore, we attributed the emission at around 454 nm present in all samples to the distorted WO6 clusters of the α-SnWO4, as observed in other studies on the PL properties of α-SnWO4 [3, 4]. And the red emission observed in the sample processed for 16 min can be associated with a high concentration of defects resulting from the short processing time. It can also be seen in Fig. 5 that the PL emission of the sample processed for 32 min presented the lowest intensity, which can be associated with several factors, such as the formation of different structural defects resulting from the MAH method, aggregation and orientation of the particles, morphology, and surface defects [17]. [F(R∞)hν]1/2 versus hν (photon energy). Fig. 4 shows the Egap of the samples processed at different times. The Egap value increased for the sample processed for 64 min, indicating a reduction in the structural defects with increasing processing time, which was in accordance with the results observed in the Rietveld refinements (Table I). Moreover, these band gap values were in accordance with the previous reports of α-SnWO4, which typically have values in the range of 1.53- 1.76 eV [5, 9, 20, 23]. Fig. 5 shows the photoluminescence (PL) spectra of the α-SnWO4 processed for different durations of time. All samples showed a broad emission band, associated with the multi-photon process, with a maximum at around 454 nm (blue region of the electromagnetic spectrum); however, the sample processed for 16 min presented an additional weak emission from 590 to 732 nm. Other metal tungstates, such as Ag2WO4, showed a similar emission in the blue and red regions [24, 25]. For example, Longo et al. [24] associated the emission centered at 449 nm with the distorted [WO6] octahedra, while emission from 621 to 640 nm was associated with the lattice modifier with different coordination and the REFERENCES Wavelength (nm) Intensity 400 550 350 500 650 450 600 700 750 [1] J. Solis, V. Lantto, Phys. Scr. T69 (1997) 281. [2] H. Dong, Z. Li, Z. Ding, H. Pan, X. Wang, X. Fu, Sens. Actuators B Chem. 140, 2 (2009) 623. [3] A.P. Chowdhury, B.H. Shambharkar, Adv. Chem. Eng. 4 (2020) 100040. [4] F.A. Alharthi, M. Shashank, J. Shashikanth, R. Viswantha, A.A. Alghamdi, J. Algethami, M.A. Alsaiari, M.S. Jalalah, N. Ganganagappa, Ceram. Int. 47, 7 (2021) 10242. [5] X. Liu, B. Liang, J. Yang, W. Li, J. Taiwan Inst. Chem. Eng. 95 (2019) 575. [6] W. Jeitschko, A.W. Sleight, Acta Crystallogr. B Struct. Sci. Cryst. Eng. Mater. 28, 11 (1972) 3174. [7] W. Jeitschko, A.W. Sleight, Acta Crystallogr. B Struct. Sci. Cryst. Eng. Mater. 30, 9 (1974) 2088. [8] J.L. Solis, J. Frantti, V. Lantto, L. Haggstrom, M. Wikner, Phys Rev B 57 21 (1998) 13491 [1] J. Solis, V. Lantto, Phys. Scr. T69 (1997) 281. [1] J. Solis, V. Lantto, Phys. Scr. T69 (1997) 281. [2] H. Dong, Z. Li, Z. Ding, H. Pan, X. Wang, X. Fu, Sens. Actuators B Chem. 140, 2 (2009) 623. [2] H. Dong, Z. Li, Z. Ding, H. Pan, X. Wang, X. Fu, Sens. Actuators B Chem. 140, 2 (2009) 623. ( ) [3] A.P. Chowdhury, B.H. Shambharkar, Adv. Chem. Eng. 4 (2020) 100040. [4] F.A. Alharthi, M. Shashank, J. Shashikanth, R. Viswantha, A.A. Alghamdi, J. Algethami, M.A. Alsaiari, M.S. Jalalah, N. Ganganagappa, Ceram. Int. 47, 7 (2021) 10242. [5] X. Liu, B. Liang, J. Yang, W. Li, J. Taiwan Inst. Chem. Eng. 95 (2019) 575. [5] X. Liu, B. Liang, J. Yang, W. Li, J. Taiwan Inst. Chem. Eng. 95 (2019) 575. [6] W. Jeitschko, A.W. Sleight, Acta Crystallogr. B Struct. Sci. Cryst. Eng. Mater. 28, 11 (1972) 3174. [7] W. Jeitschko, A.W. Sleight, Acta Crystallogr. B Struct. Sci. Cryst. Eng. Mater. 30, 9 (1974) 2088. Wavelength (nm) 550 500 6 0 600 [8] J.L. Solis, J. Frantti, V. Lantto, L. Haggstrom, M. Wikner, Phys. Rev. B 57, 21 (1998) 13491. Figure 5: PL spectra of α-SnWO4 samples. Figure 5: PL spectra of α-SnWO4 samples. M. M. P. Barros et al. / Cerâmica 68 (2022) 236-241 241 BR200815393-A2 (2010). [17] G. Botelho, J.C. Sczancoski, J. Andres, L. Gracia, E. Longo, J. Phys. Chem. C 119, 11 (2015) 6293. [9] I.-S. Cho, C.H. Kwak, D.W. Kim, S. Lee, K.S. Hong, J. REFERENCES Phys. Chem. C 113, 24 (2009) 10647. BR200815393-A2 (2010). [17] G. Botelho, J.C. Sczancoski, J. Andres, L. Gracia, Longo, J. Phys. Chem. C 119, 11 (2015) 6293. [10] A. Kuzmin, A. Anspoks, A. Kalinko, J. Timoshenko, R. Kalendarev, Phys. Scri. 89, 4 (2014) 44005. [18] A. Kuzmin, M. Zubkins, R. Kalendarev, Ferroelectrics 484, 1 (2015) 49. [11] S. Ivanov, in “Science and technology of atomic, molecular, condensed matter & biological systems”, 2, T.P. [19] R. Huang, H. Ge, X. Lin, Y. Guo, R. Yuan, X. Fu, Z. Li, RSC Adv. 3, 4 (2013) 1235. g y Das, B. Sanyal, O. Eriksson (Eds.), Elsevier (2012) 163. [12] Z. Zhu, H. Tian, M. Zhang, B. Liang, W. Li, Ceram. Int. 42, 13 (2016) 14743. [20] Q. Wang, H. Li, H. Jiang, S. Ding, Z. Song, J. Shi, Mater. Technol. 30, 5 (2015) 288. [13] S. Komarneni, R. Roy, Q.H. Li, Mater. Res. Bull. 27, 12 (1992) 1393. [21] D. Wood, J. Tauc, Phys. Rev. B 5, 8 (1972) 3144. [21] D. Wood, J. Tauc, Phys. Rev. B 5, 8 (1972) 3144. [22] L. Tolvaj, K. Mitsui, D. Varga, Wood Sci. Technol. 4 [21] D. Wood, J. Tauc, Phys. Rev. B 5, 8 (1972) 3144. [22] L. Tolvaj, K. Mitsui, D. Varga, Wood Sci. Technol. 45, 1 (2011) 135. [22] L. Tolvaj, K. Mitsui, D. Varga, Wood Sci. Technol. 45, 1 (2011) 135. [14] M. Baghbanzadeh, L. Carbone, P.D. Cozzoli, C.O. Kappe, Angew. Chem. Int. Ed. 50, 48 (2011) 11312. [23] S. Yao, M. Zhang, J. Di, Z. Wang, Y. Longa, W. Li, Appl. Surf. Sci. 357 (2015) 1528. [15] Y. Su, L. Hou, C. Du, L. Peng, K. Guan, X. Wang, RSC Adv. 2, 15 (2012) 6266. [24] E. Longo, D.P. Volanti, V.M. Longo, L. Gracia, I.C. [24] E. Longo, D.P. Volanti, V.M. Longo, L. Gracia, I.C. Nogueira, M.A.P. Almeida, A.N. Pinheiro, M.M. Ferre, L.S. [ ] g , , g , , Nogueira, M.A.P. Almeida, A.N. Pinheiro, M.M. Ferre, L.S. [16] E.L. da Silva, J.A. Varela, D.K.A. Almeida, D.P. Volanti, “Microwave aided device for hydrothermal synthesis of nanostructured oxides, particularly obtaining particles of metal oxides, comprises container, in which hydrothermal reaction takes place, and lid for container”, Brazil. Patent, Cavalcante, J. Andres, J. Phys. Chem. C. 118, 2 (2014) 1229. [25] M. Mondego, R.C. de Oliveira, M. Penha, M.S. Li, E. Longo, Ceram. Int. 43, 7 (2017) 5759. (Rec. 24/09/2021, Rev. REFERENCES 05/01/2022, 15/02/2022, Ac. 21/02/2022)
https://openalex.org/W2780543116	https://www.spinesurgery.ru/jour/article/download/447/315	Russian	null	The influence of complete vertebra reduction on the development of neurological complications after surgical treatment for the L5 spondylolitic spondylolisthesis with high degree of dislocation	Hirurgiâ pozvonočnika	2,017	cc-by	5,770	Влияние полной редукции позвонка на развитие неврологических осложнений при хирургическом лечении спондилолизных спондилолистезов L5 позвонка с высокими степенями смещения Ш.Ш. Магомедов, Ю.Ю. Поляков, С.П. Роминский Российский НИИ травматологии и ортопедии им. Р.Р. Вредена, Санкт-Петербург, Россия The influence of complete vertebra reduction on the development of neurological complications after surgical treatment for the L5 spondylolitic spondylolisthesis with high degree of dislocation S.S. Magomedov, Yu.Yu. Polyakov, S.P. Rominskiy Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Petersburg, Russia Цель исследования. Ретроспективный анализ динамики не- врологической симптоматики при полной редукции L5 позвон- ка в группах с низкостепенными (I–II ст.) и высокостепенны- ми (III–V ст., птоз) спондилолизными листезами L5 позвонка. Материал и методы. В исследование включены 158 пациентов, которым проведено оперативное лечение спондилолизного спондилолистеза L5 позвонка. Результаты. При одинаковой хирургической тактике с низко- и высокостепенными спондилолистезами отмечается выражен- ная положительная динамика в послеоперационном периоде. При истмическом спондилолистезе болевой синдром и невро- логический дефицит регрессируют дольше у пациентов с высо- костепенным спондилолистезом в раннем послеоперационном периоде. Через 6 мес. и 1 год достоверной разницы между низ- ко- и высокостепенными спондилолистезами не прослеживается. Заключение. Полная редукция L5 позвонка достоверно не ухуд- шает неврологического статуса пациента. Полная редукция сме- щенного позвонка и коррекция поясничного отдела позвоноч- ника позволяют прогнозировать восстановление сагиттального баланса относительно угла поясничного лордоза, величины на- клона таза, что приводит к восстановлению трудоспособности и ранней социальной адаптации пациента. Objective. To perform retrospective analysis of the dynamics of neurological symptoms following complete reduction of the L5 vertebra in groups of patients with low-grade (I–II) and high- grade (III–V, ptosis) L5 spondylolitic spondylolisthesis. Material and Methods. Thе study included 158 patients who underwent surgical treatment for the L5 spondylolitic spondylolisthesis. Objective. To perform retrospective analysis of the dynamics of neurological symptoms following complete reduction of the L5 vertebra in groups of patients with low-grade (I–II) and high- grade (III–V, ptosis) L5 spondylolitic spondylolisthesis. Material and Methods. Thе study included 158 patients who underwent surgical treatment for the L5 spondylolitic spondylolisthesis. Results. A pronounced positive dynamics was observed in the postoperative period after using the same surgical tactics in the treatment of both low- and high-degree spondylolisthesis. In isthmic spondylolisthesis, regression of pain syndrome and neurological deficit was longer in patients with high-grade spondylolisthesis in the early postoperative period. At 6 month and 1 year follow- up examination, there was no reliable difference between low- and high-grade spondylolisthesis. Ключевые слова: спондилолизный спондилолистез, неврологи- ческие осложнения, редукция L5 позвонка. Conclusion. The complete reduction of the L5 vertebra does not significantly impair the patient’s neurological status. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 © Ш.Ш. Магомедов и др., 2017 © S.S. Magomedov et al., 2017 Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 © Ш.Ш. Магомедов и др., 2017 © S.S. Magomedov et al., 2017 Результаты Сложность хирургического лече- ния спондилолистезов L5 позвонка обусловлена рядом морфологических изменений: выраженной нестабиль- ностью или, наоборот, ригидностью сегмента, диспластическими измене- ниями пояснично-крестцового отде- ла (спондилолизом и недоразвити- ем дуги позвонка, трапециевидной деформацией первого крестцового позвонка, недоразвитием ножек L5 позвонка, гипоплазией или аплазией ножек позвонка), вторичными деге- неративными изменениями смежных дисков [1–3, 11, 14, 20, 21, 27–29]. Собранная клинико-лабораторная информация всех пациентов пока- зала, что ведущие неврологические симптомы в основном представлены сегментарной моно- или полиради- кулярной компрессией, симптомами нейрогенной перемежающейся хро- моты, вертеброгенным болевым син- дромом, парезом нижних конечно- стей, нарушениями чувствительности и функции органов малого таза. Диагностический комплекс включал в себя анамнестические данные, невро- логический осмотр, данные нейровизу- ализации, МРТ, СКТ, рентгенографию в двух проекциях с функциональны- ми пробами. При большом смещении выполняли телерентгенограммы позво- ночника в двух проекциях в положе- нии пациента стоя. При дифференци- альной диагностике проводили ЭНМГ нижних конечностей. Анализировали полученные данные, оценивали сте- 46 Повреждения позвоночника Spine injuries межных 29]. роспек- евроло- ри пол- группах и высо- з) спон- звонка. 58 (муж- циентов, вное ле- спонди- –2015 гг. 5,8 года, ования: ов с I ст., 6 чело- Таблица 1 Неврологические нарушения в группах пациентов до хирургического лечения, n Неврологические нарушения Группа А (n = 128) Группа В (n = 30) Вертеброгенный синдром 15 0 Вертеброгенный синдром, корешковый синдром 54 2 Вертеброгенный синдром, корешковый синдром, нарушение чувствительности 35 3 Вертеброгенный синдром, корешковый синдром, нижний пара- или монопарез, нейрогенная перемежающаяся хромота 16 6 Вертеброгенный синдром, корешковый синдром, нарушение чувствительности, нижний пара- или монопарез, нейрогенная перемежающаяся хромота 8 13 Вертеброгенный синдром, корешковый синдром, нару- шение чувствительности, нижний пара- или монопарез, нейрогенная перемежающаяся хромота, нарушение функции органов малого таза 0 6 Таблица 1 Неврологические нарушения в группах пациентов до хирургического лечения, n Цель исследования – ретроспек- тивный анализ динамики невроло- гической симптоматики при пол- ной редукции L5 позвонка в группах с низкостепенными (I–II ст.) и высо- костепенными (III–V ст., птоз) спон- дилолизными листезами L5 позвонка. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis пень смещения позвонка по класси- фикации Meyerding [15], доработанной в 1956 г. Junge и Kuhl [12]. Для оценки качества жизни при патологии позво- ночника использовали шкалы Осве- стри (ODI) и Роланда – Морриса (RDQ), при оценке болевого синдрома – ВАШ. Эффективность оперативного лече- ния оценивали по шкале MacNаb. Для оценки динамики неврологического статуса использовали вариант модифи- цированной шкалы Nurick. Показания к операции: смещение позвонка, вер- теброгенный болевой синдром, невро- логические нарушения (радикулоише- мия, радикулопатия). Всем пациентам выполняли полную редукцию смещен- ного позвонка (рис. 1–3). Спондилолиз определяется как дефект межсуставной части дужки с одной или с двух сторон. Fujiwara et al. [8, 9] определили сегментарную нестабиль- ность как состояние, при котором происходит потеря жесткости позво- ночника, когда обычная физическая нагрузка приводит к болевому син- дрому. Спондилолизный спондило- листез встречается в 3–6 % случаев в общей популяции [24, 26], наибо- лее часто – на уровне L5–S1 (85–95 %) и L4–L5 (5–15 %) [26]. Многоуровне- вый спондилолизный спондилоли- стез регистрируют сравнительно реже. По данным литературы [18], часто- та встречаемости многоуровневого спондилолизного спондилолистеза составляет около 1,5 % от общего чис- ла патологии. век с III ст., 8 – с IV ст., 6 – с V ст. и спондилоптозом. Критерии отбора пациентов: – нестабильность в сегменте пояс- нично-крестцового отдела позвоноч- ника, подтвержденная данными СКТ, МРТ, рентгена с функциональными пробами; – симптоматическое течение забо- левания (неврологический дефицит, стойкий вертеброгенный болевой син- дром, снижение качества жизни, инва- лидизация пациента). Критерии исключения: тяжелая сопутствующая патология, бессим- птомное течение заболевания, ранее оперированные пациенты. Неврологические нарушения, встречающиеся в группах до хирур- гического лечения, представлены в табл. 1. Влияние полной редукции позвонка на развитие неврологических осложнений при хирургическом лечении спондилолизных спондилолистезов L5 позвонка с высокими степенями смещения The complete reduction of the displaced vertebra and the correction of the lumbar spine can predict the restoration of the sagittal balance with respect to the lumbar lordosis angle and the pelvic tilt, which results in restoration of the patient’s ability to work and early social adaptation. Key Words: spondylolitic spondylolisthesis, neurological complications, L5 reduction. Для цитирования: Магомедов Ш.Ш., Поляков Ю.Ю., Роминский С.П. Влияние полной редукции позвонка на развитие неврологических ослож- нений при хирургическом лечении спондилолизных спондилолистезов L5 позвонка с высокими степенями смещения // Хирургия позвоночника. 2017. Т. 14. № 4. С. 45–51. Please cite this paper as: Magomedov SS, Polyakov YuYu, Rominskiy SP. The influence of complete vertebra reduction on the development of neurolog- ical complications after surgical treatment for the L5 spondylolitic spondylo- listhesis with high degree of dislocation. Hir. Pozvonoc. 2017; 14(4):45–51. In Russian. DOI: http://dx.doi.org/10.14531/ss2017.4.45-51. DOI: http://dx.doi.org/10.14531/ss2017.4.45-51. 45 Spine injuri Повреждения позвоночника Повреждения позвоночника Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis 5 Данные качества жизни по шкале Роланда – Морриса в группе В (средние значе- ния) до операции (19,8 балла) и в послеоперационном периоде (1,2 балла через 1 год) Таблица 2 Распределение пациентов по эффективности хирургического лечения по шкале MacNab, n Результат Группа А (n = 128) Группа В ( I ст. (n = 70) II ст. (n = 58) III ст. (n = 16) IV ст. (n Отличный 59 43 11 5 Хороший 14 9 4 3 Удовлетворительный 0 0 0 0 Неудовлетворительный 0 3 1 0 ном перио консерват операции невральны тельной р корешка дл Наруше ности по мы в обеи Биомехани работе не Оценка операцион ставлена в статус ана рованной ции, через синдром з в обеих гру в наблюден логические го корешк сразу посл в четырех полной ред проводим вативного симптомат операции нальной д структур о ческого де Рис. 4 Данные качества жизни по шкале Роланда – Морриса в группе А (средние значе- ния) до операции (13,8 балла) и в послеоперационном периоде (0,9 балла через 1 год) Рис. 4 Данные качества жизни по шкале Роланда – Морриса в группе А (средние значе- ния) до операции (13,8 балла) и в послеоперационном периоде (0,9 балла через 1 год) Нарушения функциональной актив- ности по Освестри были сопостави- мы в обеих группах в срок до 1 года. Биомеханические параметры в данной работе не оценивали. До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 20 5 15 Баллы 1 год) Рис. 5 Данные качества жизни по шкале Роланда – Морриса в группе В (средние значе- ния) до операции (19,8 балла) и в послеоперационном периоде (1,2 балла через 1 год) Оценка боли до операции и в после- операционном периоде по ВАШ пред- ставлена в табл. 3. Неврологический статус анализировали по модифици- рованной шкале Nurick после опера- ции, через 3, 6 мес. и 1 год. Болевой синдром значительно уменьшился в обеих группах, достоверной разницы в наблюдении до года не было. Невро- логические нарушения в виде стойко- го корешкового болевого синдрома сразу после операции были выявлены в четырех случаях после выполнения полной редукции L5 позвонка. На фоне проводимого интенсивного консер- вативного лечения неврологическая симптоматика сохранялась. После ре- операции и дополнительной форами- нальной декомпрессии невральных структур отмечали регресс неврологи- ческого дефицита. При контрольном До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 20 5 15 Баллы Рис. Материал и методы Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 14 8 4 6 2 12 Баллы До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 20 5 15 Баллы Рис. 4 Данные качества жизни по шкале Роланда – Морриса в группе А (средние значе- ния) до операции (13,8 балла) и в послеоперационном периоде (0,9 балла через 1 год) Рис. 5 Данные качества жизни по шкале Роланда – Морриса в группе В (средние значе- ния) до операции (19,8 балла) и в послеоперационном периоде (1,2 балла через 1 год) До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 14 8 4 6 2 12 Баллы Рис. 4 Данные качества жизни по шкале Роланда – Морриса в группе А (средние значе- ния) до операции (13,8 балла) и в послеоперационном периоде (0,9 балла через 1 год) ски, удовлетворительного – регресс неврологического дефицита через неделю после выписки, неудовлетво- рительного – сохраняющийся стой- кий болевой синдром в послеопера- ционном периоде, не купирующийся консервативным лечением. До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 14 8 4 6 2 12 Баллы Пациенты с неудовлетворительным результатом хирургического лечения по шкале MacNab реоперированы из-за стойкого болевого корешкового синдрома в раннем послеоперацион- ном периоде, который не купировался консервативным лечением. При ре- операции выполняли декомпрессию невральных структур путем дополни- тельной резекции позвонка по ходу корешка для его послабления. 48 Повреждения позвоночника Spine injuries До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 До операции После операции Через 3 мес. Через 6 мес. Через 1 год 0 10 20 5 15 Баллы Рис. 4 Данные качества жизни по шкале Роланда – Морриса в группе А (средние значе- ния) до операции (13,8 балла) и в послеоперационном периоде (0,9 балла через 1 год) Рис. Материал и методы В исследование включены 158 (муж- чин – 53, женщин – 105) пациентов, которым проведено оперативное ле- чение при спондилолизном спонди- лолистезе L5 позвонка в 2010–2015 гг. Средний возраст женщин – 55,8 года, мужчин – 52,4 года. у Выделено 2 группы исследования: в группу А вошли 70 пациентов с I ст., 58 – со II ст.; в группу В – 16 чело- 46 Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Рис. 1 СКТ пояснично-крестцового отдела позвоночника: спондило- лиз межсуставной части дужки L5 позвонка, птоз L5 позвонка Функциональные рентгенограммы пациента в положении стоя при максимальном сгибании и максимальном разгибании: ригидный диспластический спондилолиз- ный антеспондилолистез L5 позвонка (V ст., спондилоптоз) Spine injuries ния позвоночника 47 Рис. 3 Рентгенограммы, демонстрирующие состояние после хирургического вмеша- тельства: устранены смещения L5 позвонка (V ст., спондилоптоз) путем его пол- ной редукции с транспедикулярной фиксацией L5, S1 позвонков и крыльев под- вздошных костей Оценку результатов хирургиче- ского лечения проводили через 3, 5, 7 дней, 3, 6 мес., 1 год после опера- тивного лечения. Оценка качества жизни по шкале RDQ в период до операции, в ран- нем послеоперационном периоде, через 3, 6 мес. и 1 год после опера- ции представлена на рис. 4, 5. Снижение показателей в раннем послеоперационном периоде, через 3, 6 мес. и 1 год после операции ото- бражает регресс неврологическо- го дефицита и повышение качества жизни. Прослеживается четкая тен- денция к восстановлению невроло- гического дефицита в послеопера- ционном периоде. Оценка эффективности хирурги- ческого лечения по шкале MacNab представлена в табл. 2. Рентгенограммы, демонстрирующие состояние после хирургического вмеша- тельства: устранены смещения L5 позвонка (V ст., спондилоптоз) путем его пол- ной редукции с транспедикулярной фиксацией L5, S1 позвонков и крыльев под- вздошных костей Критерии отличного результа- та – отсутствие неврологического дефицита в послеоперационном пе- риоде, хорошего – регресс невроло- гического дефицита ко дню выпи- 47 Spine injuries Повреждения позвоночника Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Таким образом, по их мнению, можно избежать натяжения корешков и в то же время улучшить позвоночно-тазовый баланс, добить- ся достаточного контакта позвонков для формирования костного блока [6]. По мнению А. Ротта [5], безопас- но выполнять редукцию смещенного позвонка на величину его нестабиль- ности на функциональных снимках, а при стабильных спондилолистезах – не более чем на одну степень. ей корешка подвздошно-поясничной связкой. Данные топографо-анатоми- ческих исследований Kleihues et al. [13], опубликованные в 2001 г., подтверди- ли факт компрессии L5 корешка меж- ду вырезкой боковой массы крестца с одной стороны и подвздошно-пояс- ничной связкой спереди. Высокая частота неврологических осложнений при полном устранении смещения III–V ст. вынуждает хирургов ограни- чивать объем редукции. В связи с этим в 2000-х гг. сформировалось представ- ление о том, что необходимо выпол- нять полную редукцию позвонка при I–II ст. смещения, а при большем сме- щении (III–IV ст. и выше) – не более чем на две степени [16]. Так, Boachie- Adjei et al. [6] выполняли частичную редукцию и устранение люмбосакраль- ного кифоза, проводили заднебоко- вой спондилодез и транспедикуляр- ную фиксацию. Таким образом, по их мнению, можно избежать натяжения корешков и в то же время улучшить позвоночно-тазовый баланс, добить- ся достаточного контакта позвонков для формирования костного блока [6]. По мнению А. Ротта [5], безопас- но выполнять редукцию смещенного позвонка на величину его нестабиль- ности на функциональных снимках, а при стабильных спондилолистезах – не более чем на одну степень. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Таблица 3 Оценка боли по ВАШ до и после операции в группах пациентов, баллы Срок Группа А (n = 128) Группа В (n = 30) До операции 913 (в среднем 7,1) 237 (в среднем 7,9) 3-и сут 744 (5,8) 160 (7,9) 5-е сут 340 (2,7) 85 (2,8) 7-е сут 128 (1,0) 45 (1,5) Таблица 3 Оценка боли по ВАШ до и после операции в группах пациентов, баллы Срок Группа А (n = 128) Группа В (n = 30) До операции 913 (в среднем 7,1) 237 (в среднем 7,9) 3-и сут 744 (5,8) 160 (7,9) 5-е сут 340 (2,7) 85 (2,8) 7-е сут 128 (1,0) 45 (1,5) Molinari et al. [17] рекомендуют при большом смещении усиливать прок- симальную фиксацию в крестце дополнительной установкой винтов в подвздошные кости. Преимуще- ства подобной фиксации подтверди- ли биомеханические исследования Cunningham et al. [7]. Подходы и принципы к хирурги- ческому пособию были обновлены в 2014 г. на съезде Американской ассо- циации нейрохирургов. При спонди- лолистезе и стенозе позвоночного канала было рекомендовано в каче- стве эффективного лечения хирур- гическое, хотя стандартным методом лечения определено не было. Необхо- димость полной редукции смещенно- го позвонка до сих пор дискутируется, также не определены алгоритм и стра- тегия хирургического пособия. осмотре через 3 мес. неврологических нарушений у пациентов не выявлено. Достоверной разницы в неврологиче- ском статусе у пациентов в двух груп- пах через один год после операции не было. ей корешка подвздошно-поясничной связкой. Данные топографо-анатоми- ческих исследований Kleihues et al. [13], опубликованные в 2001 г., подтверди- ли факт компрессии L5 корешка меж- ду вырезкой боковой массы крестца с одной стороны и подвздошно-пояс- ничной связкой спереди. Высокая частота неврологических осложнений при полном устранении смещения III–V ст. вынуждает хирургов ограни- чивать объем редукции. В связи с этим в 2000-х гг. сформировалось представ- ление о том, что необходимо выпол- нять полную редукцию позвонка при I–II ст. смещения, а при большем сме- щении (III–IV ст. и выше) – не более чем на две степени [16]. Так, Boachie- Adjei et al. [6] выполняли частичную редукцию и устранение люмбосакраль- ного кифоза, проводили заднебоко- вой спондилодез и транспедикуляр- ную фиксацию. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis 5 Данные качества жизни по шкале Роланда – Морриса в группе В (средние значе- ния) до операции (19,8 балла) и в послеоперационном периоде (1,2 балла через 1 год) 48 Повреждения позвоночника Spine injuries Таблица 2 Распределение пациентов по эффективности хирургического лечения по шкале MacNab, n Результат Группа А (n = 128) Группа В (n = 30) I ст. (n = 70) II ст. (n = 58) III ст. (n = 16) IV ст. (n = 8) V ст. (n = 6) Отличный 59 43 11 5 4 Хороший 14 9 4 3 2 Удовлетворительный 0 0 0 0 0 Неудовлетворительный 0 3 1 0 0 48 Повреждения позвоночника Spine injuries Таблица 2 Распределение пациентов по эффективности хирургического лечения по шкале MacNab, n Результат Группа А (n = 128) Группа В (n = 30) I ст. (n = 70) II ст. (n = 58) III ст. (n = 16) IV ст. (n = 8) V ст. (n = 6) Отличный 59 43 11 5 4 Хороший 14 9 4 3 2 Удовлетворительный 0 0 0 0 0 Неудовлетворительный 0 3 1 0 0 Заключение Ретроспективный анализ хирургиче- ского лечения пациентов со спонди- лолизным спондилолистезом позво- ляет заключить, что полная редукция L5 позвонка достоверно не ухудшает неврологического статуса пациента. По нашему опыту, при динамической нестабильности поясничного отде- ла позвоночника и при стабильном листезе с проявлениями неврологи- ческого дефицита необходимо выпол- нять полную редукцию позвонка при отсутствии отрицательных показате- лей визуального и интраоперационно- го электрофизиологического контро- ля. Особенно важна полная редукция смещенного позвонка: для увеличе- ния площади контакта и формирова- ния блока в сегменте, адекватного рас- пределения нагрузки между костными структурами и системой фиксации, для коррекции сагиттального баланса. Е й Таким образом, стремление к пол- ной редукции смещенного позвон- ка и коррекции поясничного отдела позвоночника позволяет прогнози- ровать восстановление сагиттального баланса относительно угла пояснич- ного лордоза, величины наклона таза, что приводит к восстановлению тру- доспособности и ранней социальной адаптации пациента. Исследование не имело спонсорской поддержки. Авторы заявляют об отсутствии конфликта интересов. рр Если в результате полной редук- ции смещенного позвонка отмечает- Литература/References Литература/References 1. Аганесов А.Г., Месхи К.Т. Реконструкция позвоночного сегмента при спонди- лолистезе поясничного отдела позвоночника // Хирургия позвоночника. 2004. № 4. С. 18–22. [Aganesov AG, Meskhi KT. Repair of spinal segment for spondylolisthesis in the lumbar spine. Hir. Pozvonoc. 2004;(4):18–22. In Russian]. 1. Аганесов А.Г., Месхи К.Т. Реконструкция позвоночного сегмента при спонди- лолистезе поясничного отдела позвоночника // Хирургия позвоночника. 2004. № 4. С. 18–22. [Aganesov AG, Meskhi KT. Repair of spinal segment for spondylolisthesis in the lumbar spine. Hir. Pozvonoc. 2004;(4):18–22. In Russian]. 7. Cunningham BW, Lewis SJ, Long J, Dmitriev AE, Linville DA, Bridwell KH. Biomechanical evaluation of lumbosacral reconstruction techniques for spondy- lolisthesis: an in vitro porcine model. Spine. 2002;27:2321–2327. DOI: 10.1097/01. BRS.0000030852.79881.F1. 8. Fujiwara A, Lim TN, An HS, Tanaka N, Jeon CH, Andersson GB, Haughton VM. The effect of disc degeneration and facet joint osteoarthri- tis on the segmental flexibility of the lumbar spine. Spine. 2000;25:3036–3044. DOI: 10.1097/00007632-200012010-00011. 2. Вреден Р.Р., Козловский А.А. Спондилолиз и спондилолистез: Практическое руководство по ортопедии. Л., 1936. [Vreden RR, Kozlovsky AA. Spondylolis and Spondylolisthesis. Practical Guide to Orthopedics. Leningrad, 1936. In Russian]. 2. Вреден Р.Р., Козловский А.А. Спондилолиз и спондилолистез: Практическое руководство по ортопедии. Л., 1936. [Vreden RR, Kozlovsky AA. Spondylolis and Spondylolisthesis. Practical Guide to Orthopedics. Leningrad, 1936. In Russian]. 3. Митбрейт И.М. Спондилолистез. М., 1978. [Mitbreit IM. Spondylolisthesis. Moscow, 1978. In Russian]. 3. Митбрейт И.М. Спондилолистез. М., 1978. [Mitbreit IM. Spondylolisthesis. Moscow, 1978. In Russian]. 9. Обсуждение С внедрением транспедикулярной фик- сации и метода переднего спондило- деза из трансфораминального досту- па появилась возможность выполнять максимальную редукцию смещенного позвонка, даже при больших степенях смещения [10]. Выполнение коррекции с созданием надежного костно-метал- лического блока дало возможность получить хорошие ортопедические результаты. Тем не менее, по некото- рым данным [10], после полной редук- ции позвонка у пациентов в послеопе- рационном периоде в 50 % случаев появлялись сегментарные неврологи- ческие нарушения, которые не всег- да проходили на фоне консерватив- ной терапии, приходилось прибегать к ревизионным операциям и умень- шать степень редукции. О развитии кауда-синдрома в послеоперационном периоде после коррекции спондило- листеза L5 IV ст. сообщают Schoenecker et al. [25]. За 18 лет они выявили подоб- ное осложнение у 12 пациентов. Меха- низм развития осложнений остался не ясным [7]. Noaсk и Kirgis [19], про- водя ревизионные вмешательства при неврологических нарушениях в после- операционном периоде, не наш- ли явной компрессии сегментарных корешков и предположили, что причи- на неврологического дефицита после редукции смещенного позвонка связа- на с экстрафораминальной компресси- Хирургическая тактика должна быть индивидуальной для каждого пациента, чтобы обеспечить макси- мальную эффективность при мини- мальном риске осложнений. В стра- тегию хирургического пособия необходимо включать не только сома- тическое состояние, анатомические особенности конкретного пациента, опыт хирурга в конкретной патологии позвоночника, но и социальные усло- вия пациента, систему медицинского страхования, что приводит к улучше- нию клинических результатов и каче- ства жизни [18, 23]. [ ] Нестабильность поясничного отде- ла позвоночника – сложная много- факторная проблема. Из-за отсутствия в литературе четкого алгоритма и так- тики лечения пациентов с высокосте- пенными истмическими спондилоли- стезами требуется индивидуальный подход при отработанной тактике хирургического лечения. Цели хирур- гических вмешательств: улучшение сегментарной стабильности позвоноч- ника, декомпрессия невральных струк- тур позвоночного канала и регресс неврологического дефицита, связан- ного со стенозом позвоночного кана- ла. В нашем исследовании проведен анализ хирургического лечения пато- логии позвоночника при истмическом Однако проведение лишь частич- ной редукции позвонка при тяжелых степенях его смещения часто при- водит к отсутствию формирования межтелового костного блока, пере- грузке и переломам металлоконструк- ции, потере полученной коррекции с развитием нестабильности и воз- никновением поздних неврологи- ческих осложнений. В связи с этим 49 Spine injuries Повреждения позвоночника Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis ся натяжение сегментарных корешков с изменениями интраоперционного электрофизиологического монито- ринга, следует проводить дополни- тельную декомпрессию невральных структур (резекцию поперечного от- ростка L5 позвонка, отделение под- вздошно-поясничной связки, резек- цию задних и экстрафораминальных остеофитов). Это позволяет выпол- нить полную редукцию позвонка. Современные исследования и диагно- стика патологии позвоночника могут помочь в определении будущих целей и решении дальнейших задач. 6 мес. и 1 год достоверной разницы не прослеживается. спондилолистезе всех степеней сме- щения L5 позвонка. Эффективность хирургического лечения по шкале MacNab в раннем послеоперационном периоде соста- вила в группе А 75,4 %, в группе В – 70,9 %. При оценке боли до операции и в послеоперационном периоде по ВАШ прослеживается четкий регресс болевого синдрома в послеоперацион- ном периоде в двух группах. По шка- ле RDQ снижение показателей в ран- нем послеоперационном периоде, через 3, 6 мес. и 1 год после опера- ции отображает регресс неврологи- ческого дефицита и повышение каче- ства жизни. Результаты проведенно- го исследования показывают, что при одинаковой хирургической тактике в двух группах отмечается выраженная положительная динамика в послеопе- рационном периоде. С учетом слож- ности истмического спондилолисте- за (болевого синдрома в послеопера- ционном периоде) качество жизни и неврологический дефицит в груп- пе В регрессируют дольше в раннем послеоперационном периоде. Через Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Вредена, polyakovspb@yandex.ru Address correspondence to: Polyakov Yuriy Yurievich Vreden Russian Research Institute of Traumatology and Orthopаedics, Akademika Baikova str., 8, St. Petersburg, 195427, Russia, polyakovspb@yandex.ru Статья поступила в редакцию 06.04.2017 Рецензирование пройдено 28.06.2017 Подписана в печать 05.07.2017 Received 06.04.2017 Review completed 28.06.2017 Passed for printing 05.07.2017 19. Noack W, Kirgis A. Dorsale reposition und ventrale spondylodese bei der lumbalen Spondylolisthesis. Operat Orthop Traumatol. 1992;4:31–49. DOI: 10.1007/BF02512863. 20. Panjabi MM, Oxland TR, Yamamoto I, Crisco JJ. Mechanical behav- ior of the human lumbar and lumbosacral spine as shown by three-dimen- sional load-displacement curves. J Bone Joint Surg Am. 1994;76:413–424. DOI: 10.2106/00004623-199403000-00012. 21. Pearson AM, Lurie JD, Blood EA, Frymoyer JW, Braeutigam H, An H, Girardi FP, Weinstein JN. Spine patient outcomes research trial: radiographic pre- dictors of clinical outcomes after operative or nonoperative treatment of degenerative spondylolisthesis. Spine. 2008;33:2759–2766. DOI: 10.1097/BRS.0b013e31818e2d8b. 22. Ravichandran G. Multiple lumbar spondylolyses. Spine. 1980;5:552–557. 23. Resnick DK, Watters WC 3rd, Sharan A, Mummaneni PV, Dailey AT, Wang JC, Choudhri TF, Eck J, Ghogawala Z, Groff MW, Dhall SS, Kaiser MG. Guideline update for the performance of fusion procedures for degenerative disease of the lum- bar spine. Part 9: lumbar fusion for stenosis with spondylolisthesis. J Neurosurg Spine. 2014;21:54–61. DOI: 10.3171/2014.4.SPINE14274. 24. Sakai T, Sairyo K, Suzue N, Kosaka H, Yasui N. Incident and etiology of lumbar spondylolysis: review of the literature. J Orthop Sci. 2010;15:281–288. DOI: 10.1007/ s00776-010-1454-4. Шамил Шамсудинович Магомедов, канд. мед. наук, заведующий отделением № 12, Российский НИИ травматологии и ортопедии им. Р.Р. Вредена, Санкт-Петербург, Россия, dr.shamil@mail.ru; Юрий Юрьевич Поляков, канд. мед. наук, врач отделения № 12, Российский НИИ травматологии и ортопедии им. Р.Р. Вредена, Санкт-Петербург, polyakovspb@yandex.ru; Сергей Петрович Роминский, врач отделения № 12, Российский НИИ травматологии и ортопедии им. Р.Р. Вредена, Санкт-Петербург, Россия, rominskiy@mail.ru. Shamil Shamsudinovich Magomedov, MD, PhD, head of the department No. 12, Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Petersburg, Russia, dr.shamil@mail.ru; Yuriy Yurievich Polyakov, MD, PhD, physician in the department No. 12, Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Petersburg, Russia, polyakovspb@yandex.ru; Sergey Petrovich Rominskiy, physician in the department No. 12, Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Petersburg, Russia, rominskiy@mail.ru. Yuriy Yurievich Polyakov, MD, PhD, physician in the department No. 12, Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Petersburg, Russia, polyakovspb@yandex.ru; Sergey Petrovich Rominskiy, physician in the department No. 12, Vreden Russian Research Institute of Traumatology and Orthopаedics, St. Заключение Fujiwara A, Tamai K, An HS, Kurihashi T, Lim TH, Yoshida H, Saotome K. The relationship between disc degeneration, facet joint osteoarthritis, and stability of the degenerative lumbar spine. J Spinal Disord. 2000;13:444–450. 4. Рерих В.В., Гладков А.В., Денисова Л.А. Оперативное лечение спонди- лолистеза // VII съезд травматологов-ортопедов России: Сб. науч. трудов / Под ред. Н.Г. Фомичева. Новосибирск, 2002. С. 210–211. [Rerikh VV, Gladkov AV, Denisova LA. Surgical treatment of spondylolisthesis. In: Abstracts of the 7th Congress of traumatologists and orthopedists. Novosibirsk, 18–20 September 2002:210–211. In Russian]. 4. Рерих В.В., Гладков А.В., Денисова Л.А. Оперативное лечение спонди- лолистеза // VII съезд травматологов-ортопедов России: Сб. науч. трудов / Под ред. Н.Г. Фомичева. Новосибирск, 2002. С. 210–211. [Rerikh VV, Gladkov AV, Denisova LA. Surgical treatment of spondylolisthesis. In: Abstracts of the 7th Congress of traumatologists and orthopedists. Novosibirsk, 18–20 September 2002:210–211. In Russian]. 10. Harms J, Rolinger H. [A one-stager procedure in operative treatment of spondylolis- theses: dorsal traction-reposition and anterior fusion (author`s transl)]. Z Orthop Ihre Grenzgeb. 1982;120:343–347. In German. DOI: 10.1055/s-2008-1051624. 11. Iguchi T, Kanemura A, Kasahara K, Sato K, Kurihara A, Yoshiya S, Nishida K, Miyamoto H, Doita M. Lumbar instability and clinical symptoms: which is the more critical factor for symptoms: sagittal translation or segment angulation? J Spinal Disord Tech. 2004;17:284–290. 5. Ротт А.Н. Хирургическое лечение больных со спондилолизным спондилолистезом: Дис. … канд. мед. наук. Санкт-Петербург, 2011. [Rott AN. Surgical treatment of patients with spondylolysis spondylolisthesis: MD/PhD Thesis. St.Petersburg, 2011. In Russian]. 12. Junge H, Kuhl P. [Appearance and significance of neural symptoms in lumbar spondylolisthesis and indications for operative management]. Bruns Beitr Klin Chir. 1956;193:39–58. In German. 6. Boachie-Adjei O, Do T, Rawlins BA. Partial lumbosacral kyphosis reduction, decom- pression and posterior lumbosacral transfixation in high-grade isthmic spondylo- listhesis: clinical and radiographic results in six patients. Spine. 2002;27:E161–E168. DOI: 10.1097/00007632-200203150-00019. 6. Boachie-Adjei O, Do T, Rawlins BA. Partial lumbosacral kyphosis reduction, decom- pression and posterior lumbosacral transfixation in high-grade isthmic spondylo- listhesis: clinical and radiographic results in six patients. Spine. 2002;27:E161–E168. DOI: 10.1097/00007632-200203150-00019. 50 Повреждения позвоночника Spine injuries Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis 13. Kleihues H, Albrecht S, Gill C, Reinhardt A, Noack W. Palsy of the L5 neural root following reposition of high degree spondylolisthesis and spondyloptosis – in vitro investigation. Eur Spine J. 1999;8 Suppl 1:S5. 25. Schoenecker PL, Cole HO, Herring JA, Capelli AM, Bradford DS. Cauda equine syndrome after in situ arthodesis for severe spondylolisthesis at the lumbosacral junc- tion. J Bone Joint Surg Am. 1990;72:369–377. 26. Standaert CJ, Herring SA. Spondylolysis: a critical review. Br J Sports Med. 2000;34:415–422. DOI: 10.1136/bjsm.34.6.415. 14. Leone A, Guglielmi G, Cassar-Pullicino VN, Bonomo L. Lumbar intervertebral instability: a review. Radiology. 2007;245:62–77. DOI: 0.1148/radiol.2451051359. 27. Vidal J, Marnay T. [Morphology and anteroposterior body equilibrium in spondylo- listhesis L5–S1]. Rev Chir Orthop. 1983;69:17–28. In French. 15. Meyerding HW. Spondylolisthesis: surgical fusion of lumbosacral portion of spinal column and interarticular fasets: use of autogenous bone grafs for relief of disabling backache. J Int Coll Surg. 1956;26(Part 1):556–591. 28. Weinstein JN, Lurie JD, Tosteson TD, Hanscom B, Tosteson AN, Blood EA, Birkmeyer NJ, Hilibrand AS, Herkowitz H, Cammisa FP, Albert TJ, Emery SE, Lenke LG, Abdu WA, Longley M, Errico TJ, Hu SS. Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med. 2007;356:2257– 2270. DOI: 10.1056/NEJMoa070302. 16. Molinari RW, Bridwell KH, Lenke LG, Ungacta FF, Riew KD. Complication in the surgical treatment of pediatric high-grade isthmic dysplastic spondylolisthe- sis. A comparison of three surgical approaches. Spine. 1999;24:1701–1711. DOI: 10.1097/00007632-199908150-00012. 17. Molinari RW, Bridwell RH, Lenke LG, Baldus C. Anterior column sup- port in surgery for high-grade isthmic spondylolisthesis. Clin Orthop Relat Res. 2002;(394):109–120. 29. Weinstein JN, Lurie JD, Tosteson TD, Zhao W, Blood EA, Tosteson AN, Birkmeyer N, Herkowitz H, Longley M, Lenke L, Emery S, Hu SS. Surgical compared with nonoperative treatment for lumbar degenerative spondylolisthesis. Four-year results in the Spine Patient Outcomes Research Trial (SPORT) randomized and observational cohorts. J Bone Joint Surg Am. 2009;91:1295–1304. DOI: 10.2106/ JBJS.H.00913. 18. Mummaneni PV, Dhall SS, Eck JC, Groff MW, Ghogawala Z, Watters WC, Dailey AT, Desnick DK, Choudhri TF, Sharan A, Wang JC, Kaiser MG. Guide- line update for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 11: interbody techniques for lumbar fusion. J Neurosurg Spine. 2014;21:67–74. DOI: 10.3171/2014.4.SPINE14276. Адрес для переписки: Поляков Юрий Юрьевич 195427, Россия, Санкт-Петербург, ул. Акад. Байкова, 8, Российский НИИТО им. Р.Р. Hirurgia Pozvonochnika (spine surgery) 2017;14(4):45–51 ХИРУРГИЯ ПОЗВОНОЧНИКА 2017. Т. 14. № 4. С. 45–51 Ш.Ш. Магомедов и др. Развитие неврологических осложнений при лечении спондилолизных спондилолистезов L5 позвонка S.S. Magomedov et al. The development of neurological complications after treatment for the L5 spondylolitic spondylolisthesis Petersburg, Russia, rominskiy@mail.ru. 51
https://openalex.org/W2797543562	https://zenodo.org/records/2184210/files/article.pdf	English	null	36. A Note on Certain Obsolete Utensils in North Wales.	Man	1,911	public-domain	1,862	Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at 36. A Note on Certain Obsolete Utensils in North Wales. Author(s): J. Edge-Partington Source: Man, Vol. 11 (1911), pp. 50-52 Published by: Royal Anthropological Institute of Great Britain and Ireland Stable URL: http://www.jstor.org/stable/2840882 Accessed: 27-06-2016 10:45 UTC Royal Anthropological Institute of Great Britain and Ireland, Wiley are collaborating with JSTOR to digitize, preserve and extend access to Man Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Royal Anthropological Institute of Great Britain and Ireland, Wiley are collaborating with JSTOR to digitize, preserve and extend access to Man Royal Anthropological Institute of Great Britain and Ireland, Wiley are collaborating with JSTOR to digitize, preserve and extend access to Man Royal Anthropological Institute of Great Britain and Ireland, Wiley are collaborating with JSTOR to digitize, preserve and extend access to Man FIG. 1.-SKULL FROM THES (NORMA LATERA evolved cranial form in Tliessaly, even at an early date, will be established. I may add that some of the Roussolakkos skulls from Crete (now in the museum at Candia) are quite comparable to this skull. But to judge from the Thessaliaii craiiia of modern date (to be found in the Academy at Atllens) the more usual skull form in that part of Greece is now longer and narrower than at earlier periods. In regard to its proportions, then, the specimen now under consideration would be contrasted wiLh the majority of modern Thessalian skulls, and thus there is some reason, on these grounds alone, for assigning it to an earlier epoch in hiistory. List of Measurements. Length (glabello-occipital) - 186 Frontal are - - - 130 Breadth - - - - 143 Parietal are - - - 130 Height (auricular) - 117 Lambda to inion - 65 Circumference - - - 530 Supra-auricular are - - 305 Minimum frontal breadtlh - 96- 5 Breadth index (mesaticephalic) - 76-9 W. L. H. DUCKWORTH. List of Measurements. ccipital) - 186 Frontal are - - - 130 - 143 Parietal are - - - 130 r) - 117 Lambda to inion - 65 - - 530 Supra-auricular are - - 305 lh - 96- 5 Breadth index (mesaticephalic) - 76-9 W. L. H. DUCKWORTH. ex (mesaticephalic) 76 9 W. L. H. DUCKWORTH. North Wales: Ethnography. Edge-Partington North Wales: Ethno on. ge- f U on. ge- f U g p y g g A Note on Certain Obsolete Utensils in North Wales. By J. Edge- f Partington. U Nos. .] M 11. distinctly smaller and less worn than the first. No signs of caries can be detected in either jaw. The chin is prominent, but a deep incisura submentalis reduces the height of the mandible in front. distinctly smaller and less worn than the first. No signs of caries can be detected in either jaw. The chin is prominent, but a deep incisura submentalis reduces the height of the mandible in front. The last character is almost the only distinctive feature of the specimen ; that is to say, that in the vast majority of the details observed, no clear inidication is given of the association of this skull with any well-known type. Moreover, this specimen may be of comparatively recent date, so far as the evideince of its state of preservation permits of a pronouncement on the subject. But if the evidence of its association with other objects of undoubted antiquity is good, then the presence of a highly 2.-SKULL FROM THESSALY. RMA VERTICALIS.) FIG. 1.-SKULL FROM THESSALY. FIG .-SKULL FROM THESSALY RMA VERTICALIS.) 2.-SKULL FROM THESSALY. RMA VERTICALIS.) FIG. 1.-SKULL FROM THES (NORMA LATERA This content downloaded from 198.91.37.2 on Mon, 27 Jun 2016 10:45:56 UTC All use subject to http://about.jstor.org/terms g p y g g A Note on Certain Obsolete Utensils in North Wales. By J. Edge- f Partington. U g So nmuch is and has been written about ethnographical specimens from foreign lands that those of Great Britain are apt to be overlooked; in fact, many of our most interesting industries of a bygone age have disappeared for ever, together with the implements connected with them. There are very few collectors, although there ought [ 50 1 This content downloaded from 198.91.37.2 on Mon, 27 Jun 2016 10:45:56 UTC All use subject to http://about.jstor.org/terms This content downloaded from 198.91.37.2 on Mon, 27 Jun 2016 10:45:56 UTC All use subject to http://about.jstor.org/terms 36. AN 36. 1 to be at least one in every county. Our local museums are in some way to blame for this, for if they would interest themselves more in local folklore they would soon find someone to take 'up this most important subject, thus preserving many things destined for the scrap-heap or fated to be thrown away to rot in some backyard. Most of the specimens that I have figured were obtained for me quite recently from farms in North Wales, and I think are worthy of preservation in our National Museuin. Although at present there is no room for exhibiting them, yet I hope the time is n)ot far distant when a growing interest in this subject will bring about L~~~~~~~~ 1) 4 LI t D 5/ 7aft~~~~~~~~~~~~~~~~~~~t a change in this direction, thus bringing to light many specimens of extreme interest before their final disappearance. Why should a Fiji "cannibal" bowl have more interest for an Englishman than any of the specimens here figured ? No. 1. A ram yoke, consisting of a stout bar; each end is pierced, and through the aperture passes a spring hoop, the ends of which are secured by a crossbar. One end removable to admit the animal's neck. Used during the rutting season. N. Wales. LI t D 5/ 7aft~~~~~~~~~~~~~~~~~~~t a change in this direction, thus bringing to light many specimens of extreme interest before their final disappearance. Why should a Fiji "cannibal" bowl have more interest for an Englishman than any of the specimens gured ? g No. 1. A ram yoke, consisting of a stout bar; each end is pierced, and through the aperture passes a spring hoop, the ends of which are secured by a crossbar. One end removable to admit the animal's neck. Used during the rutting season. N. Wales. N. Wales. No. 2. A spade for cutting turfs, shod with iron, with flange at right angles. N. Wales. No. 2. A spade for cutting turfs, shod with iron, with flange at right angles. N. Wales. [ 51 ] [ 51 ] * For other tales see the Journal of the Folklore Society (June, 1910), and of the Royal Society Arts (Oct. 19, 1910), and MAN (February, 1911). Africa, West. Tremearne. Hausa Folklore.* By Major A. J. IN. Tremearne, F.R.G.S., Hausa 27 Lecturer, Cambridge. UI ne. sa 27 UI arne. Africa Hausa Folklore. Lectur Africa, West. Tremearne. Hausa Folklore.* By Major A. J. IN. Tremearne, F.R.G.S., Hausa 27 Lecturer, Cambridge. UI Nos. ] M 11. No. 3. A " turfing iron;" iron blade with cutting edge on one side and at point, welded at base over the bandle. N. Wales. No. 4. An iron dish standing on three legs, one at each end of the pointed oval-shaped bowl, the third is at the end of the handle; used for holding the hot fat for dipping rushes, in the manufacture of rush-lights. N. Wales. g g No. 5. Wooden "begging bowl" used by the very poor people, enmployed in the manufacture of rush-lights, for begging food from the farms. N. Wales. No. 6. Circular wooden dish. N. Wales. No. 5. Wooden "begging bowl" used by the very poor people, enmployed in the manufacture of rush-lights, for begging food from the farms. N. Wales. No. 6. Circular wooden dish. N. Wales. No. 6. Circular wooden dish. N. Wales. No. 6. Circular wooden dish. N. Wales. No. 7. " Porringer "; this type was in general use for eatin porridge and milk. The staves are bound together by one broa wooden band with ends cut into strips and interlaced; one stav is longer than the rest and forms a handle. N. Wales. No. 8. Wooden scales used for the weighing of butter N. Wales. L 12.~~~~~~~~~~~~~~~3 !4lO.~~~~~~~~~~~~~~~~~~I . L .)$' No. 9. Shovel used in malt houses. N. Wales. No. 7. " Porringer "; this type was in general use for eating porridge and milk. The staves are bound together by one broad wooden band with ends cut into strips and interlaced; one stave is longer than the rest and forms a handle. N. Wales. No. 7. " Porringer "; this type was in general use for eating porridge and milk. The staves are bound together by one broad wooden band with ends cut into strips and interlaced; one stave is longer than the rest and forms a handle. N. Wales. No. 8. Wooden scales used for the weighing of butter. N. Wales. ~~~~~~~~~~~~3 ~~~~~~~~~~~~~~~~~I . ~~~~~~~~~~~~3 ~~~~~~~~~~~~~~~~~I . L .)$' No. 9. Shovel used in malt houses. N. Wales. No. 10. Small ditto, found in the old Kiln House. G-reywell, Hants. , y , No. 11. Rolling pin, the centre portion grooved, for crushing oat-cake. N. Wales. N 12 W d " l" (f d h & f ) Sh h , y , No. 11. Nos. Rolling pin, the centre portion grooved, for crushing oat-cake. N. Wales y No. 11. Rolling pin, the centre portion grooved, for crushing oat-cake. N. Wale g p , p g , g No. 12. Wooden "peel" (for removing dishes, &c., from oven). Shropshire. p g , , p No. 13. Iron rack for cleaning churchwarden pipes by placing them in the oven, generally after the bread was removed. Essex. g No. 13. Iron rack for cleaning churchwarden pipes by placing them in the oven, generally after the bread was removed. Essex. g No. 13. Iron rack for cleaning churchwarden pipes by placing them in the oven generally after the bread was removed. Essex. g y No. 14. Miniature barrel used by farm labourers to take their day's beer to th fields. Greywell, Hants. J. EDGE-PARTINGTON g y No. 14. Miniature barrel used by farm labourers to take their day's beer to the fields. Greywell, Hants. J. EDGE-PARTINGTON. g y No. 14. Miniature barrel used by farm labourers to take their day's beer to the fields. Greywell, Hants. J. EDGE-PARTINGTON. III.-HOW THE ILL-TREATED GIRL BECAME RICH. There was a certain man, he had two wives; they both gave birth, each brought forth a daughter. Theai one mother died, and the father said to the other, " See, 11). [ 52 ] 11). [ 52 ] This content downloaded from 198.91.37.2 on Mon, 27 Jun 2016 10:45:56 UTC All use subject to http://about.jstor.org/terms
https://openalex.org/W3195950254	https://zenodo.org/records/6573053/files/27.pdf	English	null	Learning Sparse Analytic Filters for Piano Transcription	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	6,509	1. INTRODUCTION Automatic music transcription (AMT) is an essential ca- pability for intelligent systems which analyze music [1]. The task is part of the broader music information retrieval (MIR) class of machine learning problems. AMT seeks to retrieve all of the information necessary to develop a score which accurately represents the music. However, given this complexity, the problem is typically reduced to the aim of estimating all notes, where a note is characterized by its pitch, time of onset, and duration [1]. Multi-instrument AMT is challenging, leading many to develop algorithms targeted for successful single-instrument AMT as an initial goal. In particular, piano transcription has been an active research task in AMT, given the wide availability of note annotations [2], and the consistency of piano timbre com- pared to multi-instrument ensembles [3]. We adopt the Onsets & Frames piano transcription model [2,5] and replace the feature extraction stage with a learn- able filterbank module 1 . We utilize a 1D convolutional structure for the filterbank, and experiment with several initialization strategies, model variations and regulariza- tion techniques. The filters represented in the module are complex, and can be employed with a standard hop size or stride. We compare the transcription results of each exper- iment, and provide an analysis of the learned filterbanks. We show that in general, the filters converge to sparse, unique shapes, which we speculate to be modeling various note characteristics. We also demonstrate that filterbanks initialized with random weights achieve comparable per- formance to those with fixed TFR initializations. Most machine learning methods involving AMT, and MIR in general, begin with a feature extraction stage which cal- culates a 2D time-frequency representation (TFR) for a piece of audio. Common choices for this step include the Mel-spectrogram or the constant-Q transform (CQT). In 1 All code is available at https://github.com/cwitkowitz/ sparse-analytic-filters. LEARNING SPARSE ANALYTIC FILTERS FOR PIANO TRANSCRIPTION Frank Cwitkowitz, Mojtaba Heydari, Zhiyao Duan Audio Information Research Lab, University of Rochester {fcwitkow,mheydari}@ur.rochester.edu zhiyao.duan@rochester.e ABSTRACT AMT algorithms, the features must contain at least enough information to detect and track notes. While standard TFRs are viable for this purpose, they may not be optimal. In par- ticular, most frequency analysis methods do not explicitly model note characteristics. They also require careful de- sign choices like filter shapes and other hyperparameters. Furthermore, there is fundamentally no information gained in moving to the frequency-domain. In recent years, filterbank learning has become an increas- ingly popular strategy for various audio-related machine learning tasks. This is partly due to its ability to discover task-specific audio characteristics which can be leveraged in downstream processing. It is also a natural extension of the nearly ubiquitous deep learning methods employed to tackle a diverse array of audio applications. In this work, several variations of a frontend filterbank learning mod- ule are investigated for piano transcription, a challenging low-level music information retrieval task. We build upon a standard piano transcription model, modifying only the feature extraction stage. The filterbank module is designed such that its complex filters are unconstrained 1D convolu- tional kernels with long receptive fields. Additional vari- ations employ the Hilbert transform to render the filters intrinsically analytic and apply variational dropout to pro- mote filterbank sparsity. Transcription results are com- pared across all experiments, and we offer visualization and analysis of the filterbanks. In this work, we focus on filterbank learning for piano transcription, extending the preliminary work of [4]. Fil- terbank learning is a way to circumvent the use of hand- crafted features by replacing or augmenting the TFR cal- culation with the response from a bank of learnable filters. As the learnable filters are tuned jointly with a backend model for the task at hand, ideally they can model task- and domain-specific characteristics of the input signal. Although filterbank learning is widely applicable to vari- ous audio processing problems, we target piano transcrip- tion because it is a complex task which, intuitively, may benefit from modeling notes in the time-domain. The filter- bank learning paradigm provides an opportunity to model some of the more obscure properties of musical notes, such as onset or offset behavior, harmonic structure, inharmonic- ity, or more generally timbre. Piano transcription is also a task with plenty of annotated data and consistency, two characteristics which can simplify the problem. Copyright: © 2022 Frank Cwitkowitz et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 Unported License, which permits unre- stricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) 2. RELATED WORK Music data has rich characteristics, patterns, and relationships at many layers of abstraction, including the note-level, the instrument-level, etc. As such, filter- bank learning carries significant potential for discovering, capturing, and leveraging task-specific information. level. Other approaches learn frequency-domain filters on top of spectrograms [21, 22], initializing them with har- monic relationships. We wish to discover these relation- ships naturally, rather than imposing them as a constraint. Filterbank learning has also been applied as a frontend for music data. However, this has mainly been for high- level tasks such as music auto-tagging [10±12] or various classification problems [13]. For lower-level tasks, such as AMT, filterbank learning has received only some atten- tion [14,15]. Music data has rich characteristics, patterns, and relationships at many layers of abstraction, including the note-level, the instrument-level, etc. As such, filter- bank learning carries significant potential for discovering, capturing, and leveraging task-specific information. Many approaches to piano transcription have attempted to realize a framework which can learn task-specific note characteristics. Non-Negative Matrix Factorization (NMF) has been proposed to learn properties such as the harmonic relationships, temporal evolution, attack, and decay of pi- ano notes from a TFR [23,24]. Convolutional sparse cod- ing is a similar approach which operates in the time-domain, and was proposed as a way to estimate the activation of pre-acquired note impulse responses [3,25]. More recently, deep neural networks (DNNs) have demon- strated success in learning to estimate discrete pitch activ- ity from TFRs [2, 5, 26, 27]. Some approaches have at- tempted to design DNNs such that they naturally lever- age information about note characteristics [28±31]. Since they are very powerful and efficient at learning features for many tasks, we hypothesize that DNNs can be utilized to learn better input features for acoustic models. Further- more, we believe the proposed frontend will naturally learn to model note characteristics for piano transcription. Many filterbank learning approaches learn strictly real- valued filters, and attain shift-invariance by pooling the re- sponse at small hops across time [6]. These filterbanks are analogous to standard 1D convolutional blocks in deep networks. In contrast, the filters used in fixed TFR calcula- tions, e.g., CQT, are typically complex and analytic, which means that they implicitly encode phase. As such, they are stable with relatively large hop sizes, and there is no need to apply further temporal pooling on their responses. 2. RELATED WORK Some filterbank learning approaches have extended the 1D convolutional approach to implement complex filters with grouped real-valued filters representing the real and imaginary part of the complex filter [16±18]. While the fil- ters in these examples are complex, if their analytic prop- erty is not preserved, they will have an asymmetric fre- quency response about 0 Hz and no longer exhibit shift- invariance in the magnitude response. Moreover, the 1D convolutional layer approach sometimes suffers from too little constraints, leading to noisy filters with no distinct or localized frequency response. One way to address this is to add soft constraints, e.g., initializing the filterbank with the weights approximating a standard transform [6]. 2. RELATED WORK Filterbank learning has gained traction in recent years as a means to perform audio-related machine learning tasks in an end-to-end fashion. Some of the first filterbank learn- ing attempts emerged to tackle problems within the speech Copyright: © 2022 Frank Cwitkowitz et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 Unported License, which permits unre- stricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 207 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Figure 1. Complex filterbank learning module. The real (black) and imaginary (purple) part of each filter are learned independently. The real and imaginary responses are combined channel-wise using L2 pooling to obtain a feature map. Proceedings of the 19th Sound and Music Computing Conference, June 5 12th, 2022, Saint Étienne (France) Figure 1. Complex filterbank learning module. The real (black) and imaginary (purple) part of each filter are learned independently. The real and imaginary responses are combined channel-wise using L2 pooling to obtain a feature map. community [6, 7], replacing the more widely used Mel- Frequency Cepstral Coefficients. This eventually lead to powerful models for tasks such as speech separation [8,9]. level. Other approaches learn frequency-domain filters on top of spectrograms [21, 22], initializing them with har- monic relationships. We wish to discover these relation- ships naturally, rather than imposing them as a constraint. Many approaches to piano transcription have attempted to realize a framework which can learn task-specific note characteristics. Non-Negative Matrix Factorization (NMF) has been proposed to learn properties such as the harmonic relationships, temporal evolution, attack, and decay of pi- ano notes from a TFR [23,24]. Convolutional sparse cod- ing is a similar approach which operates in the time-domain, and was proposed as a way to estimate the activation of pre-acquired note impulse responses [3,25]. community [6, 7], replacing the more widely used Mel- Frequency Cepstral Coefficients. This eventually lead to powerful models for tasks such as speech separation [8,9]. Filterbank learning has also been applied as a frontend for music data. However, this has mainly been for high- level tasks such as music auto-tagging [10±12] or various classification problems [13]. For lower-level tasks, such as AMT, filterbank learning has received only some atten- tion [14,15]. 3. METHOD Since we are primarily concerned with relationships between pitch and temporal characteristics for the AMT problem, we believe this is an appropriate initialization strategy. where nbpo is the number of bins or per octave.i straightforward way to represent complex filters in this type of framework is to allocate two adjacent filters for the real and imaginary weights, and to combine their respective re- sponses channel-wise using L2 pooling [16]: where nbpo is the number of bins or per octave. The bandwidth of each respective filter is determined by Bµ = fµ+1 −fµ + γ. In a CQT, the ratio between the center frequency fµ and the bandwidth Bµ, or the Q-factor, is constant, meaning that γ = 0. However, for large nbpo, this can lead to filters with extremely tight bands, requiring long impulse responses. Modestly utilizing the constant bandwidth offset γ relaxes the constancy constraint in the lower frequency range and provides direct control over the size of the filter receptive fields, which must all be zero- padded to the largest filter in order to be stored in a 1D convolutional layer. We also use the VQT parameters to determine the receptive field size for random initialization. X[k, µ] = q (x ∗R(θµ))2 + (x ∗I(θµ))2. (2) (2) Note that with this representation, there are actually 2 × nbins filters to learn, and they are only implicitly associ- ated via the subsequent pooling mechanism. Fig. 1 illus- trates this filterbank architecture. While this approach can model complex filters, the filters are rarely analytic, unless they are initialized as analytic filters. This means that they may have non-zero responses to negative frequencies and may not be shift-invariant with respect to time. Initializing the filterbank with the VQT weights provides a foundation to improve upon, and could even be thought of as pre-training the filterbank. Although there are mul- tiple transforms one could utilize, the VQT is chosen be- cause it is intuitively a better starting point for modeling note characteristics. Every nbpo 12 th filter is already locked onto a pitch in the Western music scale. Since we are primarily concerned with relationships between pitch and temporal characteristics for the AMT problem, we believe this is an appropriate initialization strategy. Another way to model complex filters is to learn only the real part and to infer the imaginary part such that the complex filter is analytic [32]. 2 Note that the filters represented with this variation may only be ap- proximately analytic, due to the limitations of discrete processing. 3. METHOD This can be done using the Hilbert Transform, which computes the imaginary counter- part to a real signal, such that the resulting complex signal is analytic 2 . This can be expressed as X[k, µ] = q (x ∗R(θµ))2 + (x ∗H(R(θµ)))2, (3) (3) An additional initialization strategy is the harmonic comb, where a set of harmonics H is defined, and separate filters with center frequency f = h × fµ are constructed for each h ∈H. All filters associated with a given µ are summed and collapsed into a single set of normalized weights for insertion into the 1D convolutional layer. In this way, each filter responds directly to the harmonics of its correspond- ing pitch, and characteristics such as inharmonicity or rel- ative harmonic strength (timbre) can be fine-tuned. where H(·) denotes the Hilbert transform. This variation learns shift-invariant filters with frequency responses con- taining only energy in the positive frequency range. 3.1 Initialization Strategy Within the framework presented above, the weights θµ are initialized randomly by default. Without inserting any prior knowledge into the filterbank, it must learn to generate fea- ture maps from scratch. This strategy has the potential to discover new filter shapes and characteristics that are not present in standard TFRs. 3. METHOD The complex filterbank learning module is implemented as a 1D convolutional layer. It accepts a 1D signal as input and produces a real and imaginary feature map, which are combined using L2 pooling, a simple mechanism for com- puting the magnitude. The output of the module is sub- sequently converted to log-amplitude and fed into a batch normalization layer. The filterbank is formulated such that an inner product is taken between a time-domain signal x and nbins filters, indexed by µ, of respective lengths lµ with weights θµ at discrete hops k spaced lh samples apart: Recently, there has been a trend to further constrain fil- terbank learning [19, 20], by learning a small number of parameters which define fixed-shape filters. While these approaches may exhibit more stability, they can only really model simple bandpass filters and thus, lower the potential for discovering meaningful patterns in data at the signal X[k, µ] = lµ−1 X n=0 x[klh + n]θµ[n]. (1) (1) Note that this operation is equivalent to convolution, or more precisely correlation, using a stride of lh. The most 208 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) where nbpo is the number of bins or per octave. The bandwidth of each respective filter is determined by Bµ = fµ+1 −fµ + γ. In a CQT, the ratio between the center frequency fµ and the bandwidth Bµ, or the Q-factor, is constant, meaning that γ = 0. However, for large nbpo, this can lead to filters with extremely tight bands, requiring long impulse responses. Modestly utilizing the constant bandwidth offset γ relaxes the constancy constraint in the lower frequency range and provides direct control over the size of the filter receptive fields, which must all be zero- padded to the largest filter in order to be stored in a 1D convolutional layer. We also use the VQT parameters to determine the receptive field size for random initialization. Initializing the filterbank with the VQT weights provides a foundation to improve upon, and could even be thought of as pre-training the filterbank. Although there are mul- tiple transforms one could utilize, the VQT is chosen be- cause it is intuitively a better starting point for modeling note characteristics. Every nbpo 12 th filter is already locked onto a pitch in the Western music scale. 3.2 Variational Dropout Variational dropout [33] is a regularization technique which allows for the learning of sparse frontend filters with long receptive fields. It is similar to Gaussian dropout, except the dropout rate, or variance, corresponding to each weight is learned. During training we treat the weights as ran- dom variables, with their true values θ being the mean, and an additional set of learned parameters σ2 being their variance. By learning the variance of each weight, vari- ational dropout induces sparsity, as it pushes less impor- tant weights to have higher variance, and more important weights to have lower variance. This leads to more sparse and interpretable filters, and can deal with very long recep- tive fields effectively. The stochasticity can also stimulate the filterbank to take interesting and unexpected shapes as it is being trained. During each forward pass, noise is sam- pled from the variance of the filterbank response and added to the mean response: Alternatively, it is possible to insert weights into the fil- terbank such that it implements a time-domain variable-Q transform (VQT) if left untrained. The complex variable-Q response for filter µ can be computed in the time-domain by making the weights θµ complex basis functions with center frequency fµ, sampling rate fs, and smoothly var- ied Q-factor Qµ: θµ,n = wµ,ne−j 2πfµn fs , n = 0, . . . , lµ. (4) (4) The filter length lµ = l Qµ fs fµ m is set such that the desired The filter length lµ = l Qµ fs fµ m is set such that the desired Q-factor, whether constant or smoothly varied, is main- tained across all filters. The windowing function w is cho- sen as a Hanning window and matches the filter length lµ for each filter. We avoid normalizing the response of each basis by filter length lµ so that the weights of each filter are all comparable and responsive to a single learning rate. The center frequency of each filter in the VQT initializa- tion is defined by selecting fmin, and applying The filter length lµ = l Qµ f fµ m is set such that the desired Q-factor, whether constant or smoothly varied, is main- tained across all filters. The windowing function w is cho- sen as a Hanning window and matches the filter length lµ for each filter. 4.4 Results We conduct several experiments to verify our implemen- tation of Onsets & Frames, and to assess the strength of various features. We use nbins = 229 and HTK frequency spacing for Mel spectrogram, as in [2]. We choose fmin ≈ 32.7 Hz, nbins = 252 and nbpo = 36, or 3 bins per semi- tone for CQT and VQT. We calculate γ = 24.7 0.108∗Q, such that all filters have a bandwidth proportional to the ERB scale by a constant factor [37] for the VQT experiment. We also experiment with an untrained filterbank initialized us- ing the harmonic comb strategy (H = {1, ..., 5}). Results for these experiments are provided at the top of Table 1. We observe that Mel spectrogram performs best in every metric besides note-onsets for MAPS, where the untrained comb filterbank slightly outperforms it. Interestingly, the comb filterbank is inferior in all other metrics.i 4. EXPERIMENTS are compared to the ground-truth at the frame-level. For note-wise evaluation, note estimates for an entire piece are compared to the ground-truth. A note is considered correct if its pitch is within half a semitone of the true value, and if its onset is within 50 ms of the true value. The note-wise with offset metric additionally compares the estimated off- set to the true value, a correct estimate being within 50 ms or 20% of the ground truth duration, whichever is larger. 4.2 Datasets MAESTRO [2] is used for training, validation, and testing. Specifically, we use version 3 (V3) of the dataset for all but one experiment, which is run on version 1 (V1). We follow the suggested partitioning for both versions. We also eval- uate on the acoustic partitions (ENSTDkAm/ENSTDkCl) of MAPS [35], to inspect the generalization potential of our method. We downsample all audio to fs = 16 kHz, and use a hop length of Lh = 512 samples. Next we conduct several filterbank learning experiments where the output of the filterbank is fed to the model in- stead of a TFR. The same VQT parameters used in the baseline experiments are used to initialize the filterbanks. We vary the filterbank variant, initialization, and regular- ization method across all experiments. For Gaussian and variational dropout, we initialize all log σ2 µ,n = −10, and for Bernoulli dropout we use a dropout rate of 10%. For variational dropout, the KL-divergence term is scaled by a factor of 0.01, such that the implicit sparsity objective does not dominate the training process. Results for these filter- bank experiments are provided at the bottom of Table 1. 4.1 Model The filterbank learning module is used as a frontend re- placement for the feature extraction stage of the Onsets & Frames model [2,5] and jointly trained for the task of piano transcription. While there have been more recent models with improved performance, e.g. [31, 34], we stick with Onsets & Frames as a baseline due to its simplicity for the purposes of filterbank analysis. The original model takes as input a Mel spectrogram and produces a frame-level salience estimate for all note pitches, independently. Note predictions are generated using the simple post-processing steps proposed in the original paper, except we do not use the note predictions to refine the salience estimate, nor do we perform the additional velocity estimation. Our exper- imental setup is the same as the original paper [5] with the subsequent improvements proposed in [2]. We utilize the same hyperparameters, running each experiment for 2000 iterations, where each music piece is accessed once per it- eration to sample a sequence of frames for a batch. 3.2 Variational Dropout Filterbank learning experiment names are formatted in terms of variant (classic (cl) / Hilbert (hb)) + initialization (random (rnd) / vqt / comb) + dropout (none / Bernoulli (brn) / Gaussian (gau) / variational (var)). Results obtained using MAESTRO V1 are indicated by . Table 1. Evaluation results for all baseline and filterbank experiments (separated by break). Filterbank learning experiment names are formatted in terms of variant (classic (cl) / Hilbert (hb)) + initialization (random (rnd) / vqt / comb) + dropout (none / Bernoulli (brn) / Gaussian (gau) / variational (var)). Results obtained using MAESTRO V1 are indicated by . 3.2 Variational Dropout We avoid normalizing the response of each basis by filter length lµ so that the weights of each filter are all comparable and responsive to a single learning rate. The center frequency of each filter in the VQT initializa- tion is defined by selecting fmin, and applying X[k, µ] ∼N(x ∗θµ, x2 ∗σ2 µ). (6) (6) The parameters σ2 µ are jointly trained with the filterbank parameters θµ, and their gradient is calculated using a very close approximation to the KL-divergence [33]. In prac- tice, we iterate upon log σ2 for improved stability. It should be noted that we only add noise to the real part of the fea- ture map when using the Hilbert filterbank variant. The parameters σ2 µ are jointly trained with the filterbank parameters θµ, and their gradient is calculated using a very close approximation to the KL-divergence [33]. In prac- tice, we iterate upon log σ2 for improved stability. It should be noted that we only add noise to the real part of the fea- ture map when using the Hilbert filterbank variant. fµ = fmin × 2 µ nbpo , (5) (5) 2 Note that the filters represented with this variation may only be ap- proximately analytic, due to the limitations of discrete processing. 209 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) MAESTRO MAPS Experiment Frame F1 Note-On F1 Note-Off F1 Frame F1 Note-On F1 Note-Off F1 mel 91.80* 95.95* 83.44* 81.40* 81.42* 59.15* mel 90.91 95.82 83.14 81.26 83.86 59.07 cqt 90.79 95.29 82.30 77.46 82.35 52.18 vqt 90.18 94.74 80.51 80.26 83.42 55.34 ﬁxed comb 87.59 92.19 75.09 80.30 84.64 57.22 cl+rnd 86.70 91.22 73.72 70.03 78.77 43.92 cl+vqt 87.53 92.24 75.64 72.90 80.90 48.06 hb+rnd 86.50 91.30 73.19 76.13 80.00 51.88 hb+vqt 87.81 92.63 76.01 75.11 80.71 50.66 hb+rnd+brn 85.23 90.19 70.75 74.81 79.48 50.57 hb+rnd+gau 85.63 90.41 71.81 75.58 80.44 51.28 hb+rnd+var 86.04 90.61 72.27 73.59 80.09 47.99 hb+vqt+var 87.45 92.39 74.92 75.62 80.80 50.55 hb+comb+var 87.52 92.27 75.69 76.74 80.98 52.57 le 1. Evaluation results for all baseline and filterbank experiments (separated by break). Filterbank learning expe f tt d i t f i t ( l i ( l) / Hilb t (hb)) + i iti li ti ( d ( d) / t / b) + d Table 1. Evaluation results for all baseline and filterbank experiments (separated by break). 4.3 Metrics We evaluate experiments using the same transcription met- rics as in [5], which include frame-wise, note-wise, and note-wise with offset scores. Each of these metrics is cal- culated in terms of precision, recall, and f1-score, using mir eval [36]. For frame-wise evaluation, detected pitches The Hilbert transform-based architecture performs simi- larly to the classic framework for random initialization, but 210 Figure 2. Examples of filters from the hb+rnd+var experiment which exhibit a high degree of sparsity. These filters have small weights and as such likely have little relevance to the downstream model. Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Figure 2. Examples of filters from the hb+rnd+var experiment which exhibit a high degree of sparsity. These filters have small weights and as such likely have little relevance to the downstream model. Figure 3. Examples of filters from the hb+rnd+var experiment which are well-localized within the receptive field and exhibit coherent shapes in both the time- and frequency-domain. Figure 3. Examples of filters from the hb+rnd+var experiment which are well-localized within the receptive field and exhibit coherent shapes in both the time- and frequency-domain. performs slightly better for VQT initialization, which tends to slightly outperform the random initialization. The vari- ational dropout regularization method performs the best among regularization methods on MAESTRO, though still under experiments with no regularization. None of the learned filterbanks outperform the fixed TFRs. Nonethe- less, they demonstrate reasonable performance, even when initialized randomly. The learned filterbanks, however, do not seem to generalize very well to the MAPS dataset. In terms of generalization to MAPS, the harmonic comb ini- tialization does the best, and variational dropout generally underperforms other regularization methods. variant with random initialization and variational dropout (hb+rnd+var), which we consider to have produced the most interesting filters. These are presented in Fig. 2- 6. In each figure, the time-domain visualization for three filters is displayed in the top row, with the corresponding frequency-domain visualization for each respective filter in the bottom row. The specific examples are chosen because they appear to belong to the same category of learned filter. In general, the filters learned for the hb+rnd+var experi- ment tend to exhibit a high degree of sparsity, with coher- ent shapes in both the time- and frequency-domain. Fig. 2 and Fig. 4.3 Metrics 3 illustrate these properties, respectively. The non-zero filter weights tend to be well-localized within the receptive field, with weights closer to zero being closer to the edges and more prominent weights being closer to the center. The tendency of the filters to be sparse is highly desirable, as it encourages more modular filters and leads 5. DISCUSSION We provide visualization of a few examples from the fil- terbank learning experiment corresponding to the Hilbert 211 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Figure 4. Examples of filters from the hb+rnd+var experiment which support multiple frequencies. Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Figure 4. Examples of filters from the hb+rnd+var experiment which support multiple frequencies. Figure 5. Examples of near-symmetric filters from the hb+rnd+var experiment comprising two impulse lobes. Figure 5. Examples of near-symmetric filters from the hb+rnd+var experiment comprising two impulse lobe Figure 6. Examples of filters from the hb+rnd+var experiment which exhibit harmonic structure. Figure 6. Examples of filters from the hb+rnd+var experiment which exhibit harmonic structure. 212 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) to more interpretable shapes in time and frequency. In- deed, it appears that the relevance of a filter can be directly estimated from its amplitude, with less important or even unused filters consisting of mostly near-zero weights rather than random noise. Furthermore, all of the Hilbert variant filters remain approximately analytic throughout the train- ing process. We do not view the artifacts in the negative frequency range to be problematic, and believe they had a negligible effect on the outcome of experiments. ture and relative strength, or inharmonicity. Given these are just hypotheses, a more objective explanation and anal- ysis of the characteristics of the filters learned with our methods is left to future work. We showed it is possible to learn sparse filters from scratch, and that resulting fil- ters differ significantly from those of the fixed transforms. Although the learned filterbanks did not surpass the perfor- mance of the fixed transforms, our method shows promise in learning to model time-domain note characteristics, and demonstrates that randomly initialized filterbanks can per- form comparably to standard TFR initializations. g g p Some additional interesting qualities of the filters are ob- served, which are presumably related to the modeling of domain-specific characteristics for piano transcription. The spectrum of the filters in Fig. 7. REFERENCES [1] E. Benetos, S. Dixon, Z. Duan, and S. Ewert, ªAuto- matic music transcription: An overview,º IEEE Signal Processing Magazine, vol. 36, no. 1, pp. 20±30, 2019. [2] C. Hawthorne, A. Stasyuk, A. Roberts, I. Simon, C.- Z. A. Huang, S. Dieleman, E. Elsen, J. Engel, and D. Eck, ªEnabling factorized piano music modeling and generation with the MAESTRO dataset,º in Pro- ceedings of ICLR, 2019. The reduced performance of the filterbank learning mod- els can potentially be attributed to too little training or inad- equate tuning of hyperparameters. We also speculate that the consistency of piano data, coupled with large model complexity, may have led to overfitting, as evidenced by the lack of generalization to the MAPS dataset. State-of- the-art methods with no filterbank learning already per- form piano transcription with reasonable proficiency. As such, it may be that a similar filterbank learning approach would show much more promise in areas with less data and less consistency, such as drum or guitar transcription.i [3] A. Cogliati, Z. Duan, and B. Wohlberg, ªPiano music transcription with fast convolutional sparse coding,º in IEEE 25th International Workshop on Machine Learn- ing for Signal Processing (MLSP), 2015. [4] F. Cwitkowitz, ªEnd-to-end music transcription us- ing fine-tuned variable-Q filterbanks,º Master’s thesis, Rochester Institute of Technology, 2019. Finally, we stress that the filterbank learning method de- scribed in this paper is not presented as a competitive al- ternative to state-of-the-art piano transcription methodolo- gies. We do not offer a comprehensive comparison to such approaches for this reason, but do acknowledge that there are many better methods for the purpose of piano transcrip- tion. One may question the utility or validity of the filter- bank learning methods due to their reduced performance. While superior performance would have been a nice out- come, we are still quite impressed with the performance of the filterbank learning models, especially the ones initial- ized randomly. From this perspective, it is clear the pro- posed methods, i.e., filterbank learning, are successful for piano transcription, albeit not at the same level of perfor- mance as standard TFRs like CQT or Mel-spectrogram. [5] C. Hawthorne, E. Elsen, J. Song, A. Roberts, I. Si- mon, C. Raffel, J. Engel, S. Oore, and D. Eck, ªOn- sets and frames: Dual-objective piano transcription,º in Proceedings of ISMIR, 2018. [6] T. N. Sainath, R. J. Weiss, A. Senior, K. W. Wilson, and O. 5. DISCUSSION 4 include multiple promi- nent fundamental frequencies in clusters, which suggests the filters may be modeling harmony or polyphonic piano sounds. The filters in Fig. 5 are nearly symmetric, with two lobes in the impulse response that have overlapping or very close frequency peaks. These filters may be modeling the attack and decay of a piano note with the same pitch. Some of these filters, in addition to the filters in Fig. 6, exhibit harmonic patterns in the frequency spectrum. In particular, odd harmonics tend to be emphasized in these filters. These observations suggest that the proposed fil- terbank learning approach is capable of capturing lower- level features which are relevant for music transcription, and that the approach is likely extensible to similar tasks where this type of information is valuable.i 7. REFERENCES Vinyals, ªLearning the speech front-end with raw waveform CLDNNs,º in Proceedings of Interspeech, 2015. [7] Y. Hoshen, R. J. Weiss, and K. W. Wilson, ªSpeech acoustic modeling from raw multichannel waveforms,º in Proceedings of ICASSP, 2015. [8] Y. Luo and N. Mesgarani, ªTasNet: Time-domain au- dio separation network for real-time, single-channel speech separation,º in Proceedings of ICASSP, 2018. Acknowledgments This work has been funded by the National Science Foun- dation grants IIS-1846184 and DGE-1922591. We would also like to thank Dr. Juan Cockburn and Dr. Andres Kwasinski for their guidance during preliminary work [4]. 6. CONCLUSION Wang, ªHigh- resolution piano transcription with pedals by regress- ing onset and offset times,º IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP), vol. 29, pp. 3707±3717, 2021. [16] N. Zeghidour, N. Usunier, I. Kokkinos, T. Schaiz, G. Synnaeve, and E. Dupoux, ªLearning filterbanks from raw speech for phone recognition,º in Proceed- ings of ICASSP, 2018. [31] Y. Yan, F. Cwitkowitz, and Z. Duan, ªSkipping the frame-level: Event-based piano transcription with neu- ral semi-CRFs,º in Advances in Neural Information Processing Systems 34 (NeurIPS), 2021. [17] N. Zeghidour, N. Usunier, G. Synnaeve, R. Collobert, and E. Dupoux, ªEnd-to-end speech recognition from the raw waveform,º in Proceedings of Interspeech, 2018. [32] M. Pariente, S. Cornell, A. Deleforge, and E. Vincent, ªFilterbank design for end-to-end speech separation,º in Proceedings of ICASSP, 2020. [18] S. Lattner, M. DÈorfler, and A. Arzt, ªLearning complex basis functions for invariant representations of audio,º in Proceedings of ISMIR, 2019. [33] D. Molchanov, A. Ashukha, and D. Vetrov, ªVaria- tional dropout sparsifies deep neural networks,º in Pro- ceedings of ICML, 2017. [19] M. Ravanelli and Y. Bengio, ªSpeaker recognition from raw waveform with SincNet,º in IEEE Spoken Language Technology Workshop (SLT), 2018. [34] C. Hawthorne, I. Simon, R. Swavely, E. Manilow, and J. Engel, ªSequence-to-sequence piano transcription with transformers,º in Proceedings of ISMIR, 2021. [20] N. Zeghidour, O. Teboul, F. d. C. Quitry, and M. Tagliasacchi, ªLEAF: A learnable frontend for au- dio classification,º in Proceedings of ICLR, 2021. [35] V. Emiya, R. Badeau, and B. David, ªMultipitch es- timation of piano sounds using a new probabilistic spectral smoothness principle,º IEEE Transactions on Audio, Speech, and Language Processing (TASLP), vol. 18, no. 6, pp. 1643±1654, 2010. [21] M. Won, S. Chun, O. Nieto, and X. Serrc, ªData-driven harmonic filters for audio representation learning,º in Proceedings of ICASSP, 2020. [36] C. Raffel, B. McFee, E. J. Humphrey, J. Salamon, O. Nieto, D. Liang, D. P. Ellis, and C. C. Raffel, ªmir eval: A transparent implementation of common MIR metrics,º in Proceedings of ISMIR, 2014. [22] Z. Zhang, Y. Wang, C. Gan, J. Wu, J. B. Tenenbaum, A. Torralba, and W. T. Freeman, ªDeep audio pri- ors emerge from harmonic convolutional networks,º in Proceedings of ICLR, 2020. [37] C. SchÈorkhuber, A. Klapuri, N. Holighaus, and M. DÈorfler, ªA matlab toolbox for efficient perfect reconstruction time-frequency transforms with log- frequency resolution,º in Audio Engineering Society (AES) 53rd Conference on Semantic Audio, 2014. 6. CONCLUSION [9] ÐÐ, ªConv-TasNet: Surpassing ideal time±frequency magnitude masking for speech separation,º IEEE/ACM Transactions on Audio, Speech, and Language Pro- cessing (TASLP), vol. 27, no. 8, pp. 1256±1266, 2019. A filterbank learning module and various techniques were analyzed for the task of piano transcription. There are sev- eral reasons to suggest that the learned filters model note characteristics such as attack and decay, harmonic struc- 213 Proceedings of the 19th Sound and Music Computing Conference, June 5-12th, 2022, Saint-Étienne (France) [10] S. Dieleman and B. Schrauwen, ªEnd-to-end learning for music audio,º in Proceedings of ICASSP, 2014. [24] T. Cheng, M. Mauch, E. Benetos, S. Dixon et al., ªAn attack/decay model for piano transcription,º in Pro- ceedings of ISMIR, 2016. [11] J. Pons, O. Nieto, M. Prockup, E. Schmidt, A. Ehmann, and X. Serra, ªEnd-to-end learning for music audio tagging at scale,º in Proceedings of ISMIR, 2018. [25] A. Cogliati, Z. Duan, and B. Wohlberg, ªPiano tran- scription with convolutional sparse lateral inhibition,º IEEE Signal Processing Letters, vol. 24, no. 4, pp. 392±396, 2017. [12] T. Kim, J. Lee, and J. Nam, ªSample-level CNN archi- tectures for music auto-tagging using raw waveforms,º in Proceedings of ICASSP, 2018. [26] S. Sigtia, E. Benetos, and S. Dixon, ªAn end-to-end neural network for polyphonic piano music transcrip- tion,º IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP), vol. 24, no. 5, pp. 927± 939, 2016. [13] J. Lee, J. Park, K. L. Kim, and J. Nam, ªSampleCNN: End-to-end deep convolutional neural networks using very small filters for music classification,º Applied Sci- ences, vol. 8, no. 1, 2018. [27] J. W. Kim and J. P. Bello, ªAdversarial learning for im- proved onsets and frames music transcription,º in Pro- ceedings of ISMIR, 2019. [14] J. Thickstun, Z. Harchaoui, and S. Kakade, ªLearn- ing features of music from scratch,º in Proceedings of ICLR, 2017. [28] R. Kelz, S. BÈock, and G. Widmer, ªDeep polyphonic ADSR piano note transcription,º in Proceedings of ICASSP, 2019. [15] R. G. C. Carvalho and P. Smaragdis, ªTowards end- to-end polyphonic music transcription: Transforming music audio directly to a score,º in IEEE Workshop on Applications of Signal Processing to Audio and Acous- tics (WASPAA), 2017. [29] T. Kwon, D. Jeong, and J. Nam, ªPolyphonic pi- ano transcription using autoregressive multi-state note model,º in Proceedings of ISMIR, 2020. [30] Q. Kong, B. Li, X. Song, Y. Wan, and Y. 6. CONCLUSION [23] E. Vincent, N. Bertin, and R. Badeau, ªAdaptive har- monic spectral decomposition for multiple pitch es- timation,º IEEE Transactions on Audio, Speech, and Language Processing (TASLP), vol. 18, no. 3, pp. 528± 537, 2010. 214
https://openalex.org/W4233624138	https://www.researchsquare.com/article/rs-9120/latest.pdf	English	null	Through service providers’ eyes: health systems factors affecting implementation of tuberculosis control in Enugu State, South-eastern Nigeria	Research Square (Research Square)	2,020	cc-by	6,321	Daniel Chukwuemeka Ogbuabor (  ogbuabordc@gmail.com ) Daniel Chukwuemeka Ogbuabor (  ogbuabordc@gmail.com ) DOI: https://doi.org/10.21203/rs.2.18500/v2 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Version of Record: A version of this preprint was published at BMC Infectious Diseases on March 6th, 2020. See the published version at https://doi.org/10.1186/s12879-020-4944-9. Page 1/16 Page 1/16 Abstract Background Well-functioning health systems are essential to achieving global and national tuberculosis (TB) control targets. This study examined health system factors affecting implementation of TB control programme from the perspectives of service providers. Methods The study was conducted in Enugu State, South-eastern Nigeria using qualitative, cross-sectional design involving 23 TB service providers (13 district TB supervisors and 10 facility TB focal persons). Data were collected through in-depth, semi- structured interviews using a health system dynamic framework and analysed thematically. Results Stewardship from National TB Control Programme (NTP) improved governance of TB control, but stewardship from local government was weak. Government spending on TB control was inadequate, whereas donors fund TB control. Poor human resources management practices hindered TB service delivery. TB service providers have poor capacity for data management because changes in recording and reporting tools were not matched with training of service providers. Drugs and other supplies to TB treatment centres were interrupted despite the use of a logistics agency. Poor integration of TB into general health services, weak laboratory capacity, withdrawal of subsidies to community volunteers and patent medicine vendors, poorly funded patient tracking systems, and ineffectual TB/HIV collaboration resulted in weak organisation of TB service delivery. Conclusion Health systems strengthening for TB control service must focus on effective oversight from NTP and local health system; predictable domestic resource mobilisation through budgets and social health insurance; training and incentives to attract and retain TB service providers; effective supply and TB drug management; and improvements in organization of service delivery. Background Well-functioning health systems are essential to achieving global and national tuberculosis (TB) control targets [1]. The global targets are 90% reduction in the number of TB deaths and 80% reduction in TB incidence rate (number of new TB cases per 100, 000 population per year) relative to 2015 levels [1]. In 2018, about 10 million people fell ill with TB, resulting in 1.5 million deaths [1]. The reduction in TB incidence rate between 2015 and 2018 was only 6.3%, significantly lower than the End TB Strategy milestone of 20% between 2015 and 2020 [1]. The End TB Strategy identified integrated, patient-centred care, bold policies and supportive systems as pillars to end TB by 2030 [2]. Comprehensive changes to policies and regulation, organisational structures, processes and relationships and effective use of resources are needed to achieve and sustain goals of TB control programme [3, 4]. Nonetheless, low- and middle-income countries (LMICs) still struggle with health system strengthening for TB control. Health systems factors that strengthened TB control include effective national strategies [5-7]; strong stewardship from district government [8]; donor funding [5, 8-13]; inclusion of TB in social health insurance schemes [14, 15]; earmarked funding for TB [16]; medical vouchers and subsidies [17, 18]; providers’ satisfaction with directly observed treatment short course [19]; incentives to programme staff [20]; patient-centred approaches and intersectoral collaboration [6, 16, 19]; motivated and dedicated healthcare workers [16, 19]; effective supply and drug management system [8, 21]; use of dedicated Page 2/16 Page 2/16 logistics agency to distribute TB drugs [9, 22]; and electronic reminders to improve treatment adherence [17]. Constraints to TB control include perception that TB control is less important than other public health programs [20, 23]; weak accountability relationship between provincial/regional and district TB programme management [7, 12, 24]; insufficient community involvement [25]; and low public spending on TB [5-7, 9-13]. TB control have been limited by lack of skilled staff [4, 10, 11, 19, 22]; lack of incentives for service providers [4, 5, 10]; lack of utilization of various levels of health staff and health facilities [25]; poor attitude and weak commitment of health workers towards deployment to TB services [22, 23], inadequate training [10, 11, 19], and lack of laboratory staff [5, 22, 23]. Poor TB data management capacity [22]; frequent revisions of reporting formats [5]; and ineffective electronic recording and reporting system [14]. Background The southeast region of Nigeria was particularly problematic with median TB case notification rate of 16% (range, 10% to 18%) [32]. In Enugu State, TB case notification rate was also 16% [32]. Yet, studies using systems thinking to analyse implementation of TB control in Nigeria are scarce. The study therefore examined health system factors affecting implementation of TB control programme in Enugu State, Nigeria from the perspectives of district-level service providers. Such evidence would inform policy changes to ensure universal access to TB services in low resource, high TB burden settings. Background Shortages of TB drug [7, 22] and unavailability of laboratory supplies and equipment hinder TB service delivery [7, 11, 22]. Poor adherence to GeneXpert algorithm, interrupted supply of cartridges, lack of replacement of damaged modules, poor maintenance, and poor recording and reporting limit usefulness of GeneXpert in TB diagnosis [14, 26, 27]. Poor adherence to national guidelines [11, 22, 23]; dilapidated infrastructure [5, 12, 23] weak TB/HIV integration [10, 12, 14, 24, 28]; lack of public-private mix [10, 12, 23] and poor integration into general health services hinder effective TB service delivery [5, 29, 30]. Nigeria’s national strategic plans for TB control emphasised health system strengthening [31]. However, after over two decades of TB control, Nigeria has about 60% funding gap and 24% TB case notification rate, and is one of ten countries accounting for 80% gap in TB notification globally [1]. Nigeria remains a high TB burden country with incidence rate of 219 per 100,000 population [1]. In 2018, about 429,000 people fell ill with TB, resulting in 157,000 deaths; domestic resource mobilisation for TB was considerably low at 8%; and about 71% of TB patients faced catastrophic costs in Nigeria [1]. The southeast region of Nigeria was particularly problematic with median TB case notification rate of 16% (range, 10% to 18%) [32]. In Enugu State, TB case notification rate was also 16% [32]. Yet, studies using systems thinking to analyse implementation of TB control in Nigeria are scarce. The study therefore examined health system factors affecting implementation of TB control programme in Enugu State, Nigeria from the perspectives of district-level service providers. Such evidence would inform policy changes to ensure universal access to TB services in low resource, high TB burden settings. Nigeria’s national strategic plans for TB control emphasised health system strengthening [31]. However, after over two decades of TB control, Nigeria has about 60% funding gap and 24% TB case notification rate, and is one of ten countries accounting for 80% gap in TB notification globally [1]. Nigeria remains a high TB burden country with incidence rate of 219 per 100,000 population [1]. In 2018, about 429,000 people fell ill with TB, resulting in 157,000 deaths; domestic resource mobilisation for TB was considerably low at 8%; and about 71% of TB patients faced catastrophic costs in Nigeria [1]. Conceptual framework The study was guided by health systems framework by Van Olmen and colleagues (33). The framework builds on World Health Organization’s six building blocks of a health system (34) and uses input, process, outcomes and goals to explain health systems dynamics (Figure 1). The goals of TB control are reductions in TB mortality and incidence. Universal coverage, treatment success and responsiveness are outcomes of optimal TB control programme. TB diagnosis and treatment is considered service delivery Page 3/16 Page 3/16 block and equals the process in the framework. Health financing, human resources, medical technology and health management information system (HMIS) serve as inputs into service delivery. Governance, which entails policy guidance, policy coordination, regulation of different functions, optimal allocation of resources and accountability, influences both the input and process. This framework was chosen because it explains how interactions of governance, inputs and process would lead to improved implementation of TB control and better health outcomes. Study population and sampling strategy Study subjects were LGTBS and facility TB focal persons in Enugu state. All 17 LGTBS were included in the study. However, data from LGTBSs was saturated after the thirteenth interview (35). To select facility TB focal persons, the 17 LGAs were categorized into low (8 LGAs) and high (9 LGAs) case finding districts based on case notification, an important indicator of successful TB control programme (1, 31). From each category, 5 LGAs were purposively selected. One TB treatment centre was purposively selected from each LGA. The TB focal persons in the selected treatment centres were interviewed. Maximum variation sampling was used to ensure that both rural and urban LGAs and all levels of facilities were represented (Table 1). Study Setting The study was conducted in Enugu State, South-east Nigeria. Enugu State is one of five geo- demographically and culturally homogenous states in southeast region of Nigeria. Enugu State was purposively chosen for this study because its 2018 TB case notification rate equals the median regional case TB notification rate of 16% (32). The state consists of 17 local government areas (LGAs); 12 of which are predominantly rural areas. The National TB Control Programme (NTP) is headed by a state control officer. Each local government area (LGA) has a TB supervisor (LGTBS). The state has 71 and 41 TB treatment centres in public and private health facilities respectively. Publicly owned TB treatment centres comprise primary, secondary and tertiary health facilities. Each treatment centre has a TB focal person. Research design The study adopted a cross-sectional, qualitative study design using in-depth, semi-structured interviews. Study population and sampling strategy The study adopted a cross-sectional, qualitative study design using in-depth, semi-structured interviews. Data collection Twenty-three participants were interviewed between January and April 2017 using pre-tested, semi- structured in-depth interview guide (Appendix 1). The interview questions were informed by the conceptual framework, adapted to TB control programme. Participants were identified using health Page 4/16 Page 4/16 officials as gatekeepers. Interviews were held in offices or health facilities, conducted in English and lasted about 60 minutes. All interviews were audiotaped, transcribed verbatim and verified by participants for accuracy of transcription. Results Findings are presented thematically using health system building blocks deduced from the study’s conceptual framework (Table 2). Data analysis Data were analysed using a framework approach. Deductive and inductive strategies were used to code the transcripts in NVivo 11 software. Main themes were deduced from health systems’ building blocks. Inductive codes were generated by familiarization with data and assigning codes to emerging themes. Coding was carried out by two persons and inconsistencies resolved by consensus. The LGTBS validated the study findings during a programme review meeting, which improved the validity of the findings. Ethical consideration The study was approved by Health Research Ethics Committee of Enugu State University Teaching Hospital, Parklane, Enugu, Nigeria. Written, informed consent was obtained from all participants. Leadership and governance Stewardship from the State and Local Governments emerged as key theme. All LGTBS received regular managerial oversight from the state TB programme. Similarly, most facility TB focal persons stated that “TB supervisors are always available when we need their advice, or we have questions or clarifications” (FP5). Quarterly review meetings offered opportunity for joint programme planning and coordination and holding LGTBS accountable. LGTBS serve under Primary Health Care Coordinators (PHCCs), who manage LG’s health system. Nonetheless, most PHCCs, except those who were LGTBS prior to their appointment as PHCC, were unsupportive of TB control. It was explained that “because money does not come from the state …, TB control program is not important” (TBS3) to local government officials. Furthermore, facility TB focal persons noted that PHCC rarely visited health facilities. Health information system Most participants indicated that recording and reporting tools were available but frequently changed without prior training of service providers. One participant stated: “they used to change the pattern and the format, which we find very difficult to cope with” (FP3). Modifications in tools were adaption to changes in TB control strategies: “we changed from eight months to six months treatment. Now we are addressing HIV, drug resistant TB and presumptive cases” (TBS5). Poor recording and reporting also resulted from work overload and lack of incentives. Electronic Tuberculosis manager (ETBM) has been rolled out to high burden facilities and LGTBS. Most participants perceived that ETBM would enable timely and complete recording and reporting. Few participants anticipated that ETBM would add to workload of service providers. Human resources for health Inadequate staffing, posting and transfer, supervision, remuneration and training emerged as key themes Inadequate staffing stemmed from unwillingness of health workers to work in NTP due to stigma and lack of incentives. Consequently, TB service providers reported high workload: “my workload is high. It is not only TB services that I provide. I am also the anti-retroviral therapy focal person” (FP7). Most participants observed that trained TB service providers are frequently transferred out TB treatment centres. Few participants noted the social restriction to the posting or transfer of female health workers. Most participants agreed that supervision was supportive: “if there are things we are doing wrong, they correct us, … and if there are things we are doing very well, they still commend us” (FP5). However, supervision is limited by “lack of functional motorbike and allowances to cover fuel and maintenance” (TBS5). Most LGAs were owing staff several months’ salaries. Yet, financial incentives are scare especially among facility TB focal persons who “do not get any incentive” (TBS3). Most participants noted that there are many untrained TB focal persons. External training was preferred to on-the-job training because “they do not give us anything, except when we go for seminar” (FP4). Financing of TB control program Page 5/16 Most participants stated that NTP has been donor-driven while public spending on TB control is low. As one supervisor stated, “TB programme is donor-driven. If I should say my mind, the government is not doing what it should” (TBS1). At the state level, allocated funds are not released to TB programme. At LG level, unlike other public health programs, TB control programme does not have a budget. Government Page 5/16 officials feel that “we are collecting dollars from these development partners; why come to look for naira” (TBS1); and “we don’t have anything (incentive) coming out of the program” (TBS10). Although LGs agreed to replace TB supervisors’ motorcycles, no LG ever supplied motorcycle to LGTBS, which limits supervision, community mobilization, and contact and defaulter tracing. One facility TB focal persons lamented “I have written the local government chairman severally through the head of department of health, for financial support to build TB laboratory. They both signed, but till now, I have not seen anything from them” (FP3). officials feel that “we are collecting dollars from these development partners; why come to look for naira” (TBS1); and “we don’t have anything (incentive) coming out of the program” (TBS10). Although LGs agreed to replace TB supervisors’ motorcycles, no LG ever supplied motorcycle to LGTBS, which limits supervision, community mobilization, and contact and defaulter tracing. One facility TB focal persons lamented “I have written the local government chairman severally through the head of department of health, for financial support to build TB laboratory. They both signed, but till now, I have not seen anything from them” (FP3). Medical / Drug Supplies Most participants stated that TB drug supplies were interrupted during the preceding year: “a quarter would pass without supply” (FP6). Although, a logistics organization was contracted by the NTP to supply drugs directly to TB treatment centres, “they cut corners by dropping all the supplies meant for facilities in an LGA at one facility and expect individual facilities to come and pick their supplies there.” Page 6/16 (TBS1). Most TB treatment centres coped by borrowing drugs from other health facilities. Few participants explained extrapulmonary TB patients and TB patients weighing above 70kg required more than one drug kit to meet their dosage requirements. A facility TB focal person remarked that “we have to de-kit in order to make it up for six months” (FP6). Human immunodeficiency virus (HIV) test kits were no longer supplied to TB treatment centres. A LGTBS asserted: “for the past 2 years, we have not received any HIV test kits” (TBLS3). Consequently, some health workers procure HIV test kits and charge fees for HIV screening. (TBS1). Most TB treatment centres coped by borrowing drugs from other health facilities. Few participants explained extrapulmonary TB patients and TB patients weighing above 70kg required more than one drug kit to meet their dosage requirements. A facility TB focal person remarked that “we have to de-kit in order to make it up for six months” (FP6). Human immunodeficiency virus (HIV) test kits were no longer supplied to TB treatment centres. A LGTBS asserted: “for the past 2 years, we have not received any HIV test kits” (TBLS3). Consequently, some health workers procure HIV test kits and charge fees for HIV screening. Service Delivery Most providers stated that TB is poorly integrated into general health services due to stigma by health workers, concern for contracting TB and lack of financial incentives. As explained, “some health workers refuse participating in TB control because of fear that they might contract TB” (TBS5); or “that nothing is coming out of it (TB programme), unlike malaria, monitoring and evaluation, and immunization” (TBS9). Most providers reported availability of functional microscopic centres. However, some providers indicated that TB diagnosis is constrained by lack of space, damaged microscope and lack of laboratory staff: “I have up to 10, but only 1 is functional” (TBLS2). Lack of dispatch riders, breakdown of machine, varying days of operations, and absence of support for sputum transport constrained use of GeneXpert for TB diagnosis. Withdrawal of incentives constrained involvement of community volunteers and patent medicine vendors. As one participant stated, “Usually, community volunteers received monthly stipends, but since GHAIN (a development partner) stopped it and government could not sustain it, all of them declined” (FP1). Also, “only very few of patent medicine vendors, who are health workers, still refer suspects to TB programme” (TBLS6). Some providers revealed that patient tracking systems is weak due to poor funding for contact and defaulter tracing, while TB/HIV collaboration was limited to comprehensive sites. Discussion Additionally, broad contextual factors underlying human resource crisis especially social restriction to deployment of female health workers and delayed payment of staff salaries warrant attention of decision makers. This study’s findings revealed that poor human resources management practices and inadequate training of TB service providers hindered TB control. These findings are consistent with lack of skilled staff (4, 10, 11, 22); lack of incentives for TB service providers (5, 10); inadequate training (10, 11, 19); and poor attitude and weak commitment of health workers towards deployment to TB services (22, 23). In contrast, evidence from European countries show that skilled and motivated healthcare workers enabled TB control (16). Strategies that aim at improving TB control workforce must address shortages of staff, stigma by health workers and lack of incentives to TB service providers. Additionally, broad contextual factors underlying human resource crisis especially social restriction to deployment of female health workers and delayed payment of staff salaries warrant attention of decision makers. The findings revealed that TB service providers have poor capacity for recording and reporting TB data due to mismatch between changes in tools and training of service providers, which is similar to evidence from Ethiopia and Nigeria (5, 22). Even though, changes in tools were adaptation to new developments in TB control program, lack of training meant that TB service providers did not completely fill tools and required constant supportive supervision. A strategy to improve data management was introduction of electronic recording and reporting system. However, experiences from South Africa suggest that operational challenges limited effectiveness of electronic recording and reporting system (14). Nigeria’s NTP would need to improve on-the-job capacity building of service providers on data management, while addressing the limitations of electronic recording and reporting system. The findings of this study that drugs and other supplies to TB treatment centres were interrupted are consistent with evidence of shortages of TB drug and laboratory supplies elsewhere (7, 11, 22), but contrast evidence of effective supply and drug management in Pakistan (8). Although use of dedicated logistics agency to distribute TB drugs improved TB drug supply system (9, 22), experiences in our study area reveal that TB drugs for an entire district were often dumped in one health facility. Equally, absence of drug kits tailored to extrapulmonary TB patients and TB patients weighing 70kg and above. Discussion Whereas stewardship from the NTP improved district TB service delivery, stewardship from local government (LG) was weak despite existence of memorandum of understanding between LGs and NTP. Effective bureaucratic relationship between state and LG TB programmes in this study contradicts evidence of poor stewardship in India, Ghana and South Africa (7, 12, 24). In our setting, accountability relationship was enhanced by programme management support, supportive supervision and quarterly performance review meetings. Nevertheless, the weak stewardship from the LG confirms previous evidence that TB control program is perceived as less important than other public health programmes (20, 23) but contrasts evidence of strong political commitment from district governments in Pakistan (8). Unmet expectation of local officials of financial incentives from NTP and perception that partners support TB control explained low responsiveness of LGs. Since governance affects input and process of TB control, LG officials need to strengthen stewardship of and become more responsive to TB control program. Page 7/16 This study also found that government spending on TB control was inadequate, which is similar to evidence from several studies where donors, instead of governments, drive TB control efforts (5-13). In this study, non-release of budgeted funds at the state-level and complete absence of TB in LG budgets resulted in persisting TB funding gap and unlikelihood of sustainability of TB control. As external funding declines, all tiers of government in Nigeria need to improve financing of TB control through budgets. Also, Nigeria’s efforts to decentralise social health insurance to states presents an opportunity to include TB in social health insurance schemes (14, 15), given that Nigerian TB patients incur catastrophic costs (36, 37). Moreover, it might be helpful to consider medical vouchers and subsidies for poor TB patients (17, 18). 18). This study’s findings revealed that poor human resources management practices and inadequate training of TB service providers hindered TB control. These findings are consistent with lack of skilled staff (4, 10, 11, 22); lack of incentives for TB service providers (5, 10); inadequate training (10, 11, 19); and poor attitude and weak commitment of health workers towards deployment to TB services (22, 23). In contrast, evidence from European countries show that skilled and motivated healthcare workers enabled TB control (16). Strategies that aim at improving TB control workforce must address shortages of staff, stigma by health workers and lack of incentives to TB service providers. Discussion Hence, service providers must augment one drug kit to meet their dosage requirements, which depletes the stock. Sustainable TB drug supply system would entail strengthening the logistics agency and improving the capacity of NTP to forecast TB drug needs. Page 8/16 Weak TB service delivery system was found as a key constraint to TB control. Consistent with findings of previous studies (5, 29, 30), integration of TB into general health services was weak. Social stigma and concern for contracting TB meant that health workers refuse deployment to and participation in TB control programmes. Also, the findings of weak laboratory capacity due to lack of space, damaged microscopes, attrition of laboratory staff and operational challenges of GeneXpert, are like evidence from previous studies (5, 11, 14, 22, 23, 26, 27). Equally, poor involvement of community volunteers and patent medicine vendors due to withdrawal of financial incentives hindered TB service delivery. Although TB service providers have used electronic reminders to improve treatment adherence similar to China’s experience (17), weak patient tracking systems resulted from poor funding and lack of transportation. Notwithstanding existence of guidelines for TB/HIV integration, weak TB/HIV collaboration is similar to evidence from other studies (10, 12, 14, 24, 28). Policies to improve TB service delivery system must address stigma among health workers, strengthen laboratory capacity, incentives to community volunteers and patent medicines vendors, improve funding of patient tracking and bolster TB/HIV integration. This study adds to the growing scholarship using health systems lens to examine disease control programmes. Particularly, application of systems thinking provided useful analytical tool to explain how health system strengthening enables or constrains TB control in high TB burden, low-resource settings. However, as the participants were limited to frontline, district-level TB service providers in one sub- national context, the findings may not be generalizable to other settings. Nonetheless, this study does not aim to generalize but to provide evidence to inform policy changes to ensure universal access to TB services in low-resource settings facing similar approaches. Given the emphasis on people-centred health system, perceptions and experiences of consumers TB service delivery systems would be an area of future research. Conclusion In conclusion, the findings of this study illustrate how health system factors drive control of TB in low resource settings. Given that governance influences both inputs and service delivery, governing TB control service delivery would entail effective oversight from NTP and local health system. Sustainable financing of TB would depend on stable and predictable domestic resource mobilisation through budgets and social health insurance. To attract and retain staff, human resource strategy must address training needs, lack of incentives to, and stigmatization of TB service providers. Furthermore, it is vital to strengthen TB drug and commodities’ supply system and organization of service delivery. Acknowledgements I am grateful to the local government supervisors and focal persons who participated in this study. Ethical approval and consent to participate Ethical clearance for this study was received from the health research ethics committee of Enugu State University Teaching Hospital, Enugu (ESUTHP/C-MAC/RA/034/VOL.11/30). Consent to publish The study did not include details, images or videos relating to individual participants, thus consent for publication were not required. Authors' contributions This article has a sole author. This article has a sole author. Abbreviations ETBM: Electronic Tuberculosis Manager; FP: Focal person; HIV: Human immunodeficiency virus; LG: Local government; LGA: Local government area; LGTBS: Local government tuberculosis supervisor; LMICs: Low- and middle-income countries; NSP-TB: National strategic plan for tuberculosis control; NTP: National TB control programme; PHCC: Primary health care coordinator; TB: Tuberculosis; TB/HIV: Page 9/16 Page 9/16 Tuberculosis/ Human immunodeficiency virus; UHC: Universal health coverage; WHO: World Health Organisation. Availability of data and material The dataset supporting the conclusions of this article is included within the article. The dataset supporting the conclusions of this article is included within the article. Competing interests The author reports no conflict of interest. References 1. WHO. Global tuberculosis report 2019. Geneva: World Health Organisation; 2019. 2. WHO. The End TB strategy: global strategy and targets for tuberculosis prevention, care and control after 2015. Geneva: World Health Organization; 2014. 3. Shigayeva A, Coker RJ. Communicable disease control programmes and health systems: an analytical approach to sustainability. Health Policy Plann. 2015;30:368–85. 4. Kohler S, Asadov DA, Bründer A, Healy S, Khamraev AK, Sergeeva N, et al. Health system support and health system strengthening: two key facilitators to the implementation of ambulatory tuberculosis Page 10/16 Page 10/16 treatment in Uzbekistan. Health Econ Rev. 2016;6:28. treatment in Uzbekistan. Health Econ Rev. 2016;6:28. 5. Ogbuabor DC, Onwujekwe OE. Governance of tuberculosis control in Nigeria. Infect Dis Poverty. 2019;8:45. 6. Khan WM, Smith H, Qadeer E, Hassounah S. Knowledge and perceptions of national and provincial tuberculosis control programme managers in Pakistan about the WHO Stop TB strategy: a qualitative study. J Roy Soc Med Open. 2016;8(1):1-9. 7. Pai M, Daftary A, Satyanarayana S. TB control: challenges and opportunities for India. T Roy Soc Trop Med H. 2016;110(3):158-60. 8. Metzger P, Baloch NA, Kazi GN, Bile KM. Tuberculosis control in Pakistan: reviewing a decade of success and challenges. Eastern Mediterr Health J. 2010;16:S47-S53. 9. Atun R, Weil DEC, Eang MT, Mwakyusa D. Health-system strengthening and tuberculosis control. Lancet. 2010;375:2169-78. 10. Otu AA. A review of the national tuberculosis and leprosy control programme (ntblcp) of Nigeria: challenges and prospects. Ann Trop Med Pub Health. 2013;6(5):491-500. 11. Wynne A, Ritcher S, Banura L, Kipp W. Challenges in tuberculosis care in Western Uganda: Health care worker and patient perspectives. Int J Afr Nurs Sci. 2014;1:6-10. 11. Wynne A, Ritcher S, Banura L, Kipp W. Challenges in tuberculosis care in Western Uganda: Health care worker and patient perspectives. Int J Afr Nurs Sci. 2014;1:6-10. 12. Amo-Adjei J. Views of health service providers on obstacles to tuberculosis control in Ghana. Infect Dis Poverty. 2013;2:9. 12. Amo-Adjei J. Views of health service providers on obstacles to tuberculosis control in Ghana. Infect Dis Poverty. 2013;2:9. 13. Zou G, King R, Walley J, Yin J, Sun Q, Wei X. Barriers to hospital and tuberculosis programme collaboration in China: context matters. Glob Health Action. 2015;8(1):27067. 13. Zou G, King R, Walley J, Yin J, Sun Q, Wei X. Barriers to hospital and tuberculosis programme collaboration in China: context matters. Glob Health Action. 2015;8(1):27067. 14. References Churchyard GJ, Mametja LD, Hesseling AC, Reid A, Mvusi L, Babatunde S, et al. Tuberculosis control in South Africa: Successes, challenges and recommendations. S Afr Med J. 2014;104(3 (Suppl 1)):244-8. 15. Zhou C, Long Q, Chen J, Xiang L, Li Q, Tang S, et al. The effect of NCMS on catastrophic health expenditure and impoverishment from tuberculosis care in China. Int J Equity Health. 2016;15:172. 16. de Vries G, Tsolova S, Anderson LF, Gebhard AC, Heldal E, Hollo V, et al. Health system factors influencing management of multidrug-resistant tuberculosis in four European Union countries - learning from country experiences. BMC Public Health. 2017;17:334. 17. Li J, Shen X, Yeoh E, Chung P. Tuberculosis control programs and challenges in developed cities with intermediate disease burden: China experience. J Thorac Dis. 2017;9(5):E525-E8. 17. Li J, Shen X, Yeoh E, Chung P. Tuberculosis control programs and challenges in developed cities with intermediate disease burden: China experience. J Thorac Dis. 2017;9(5):E525-E8. 18. Ukwaja KN, Alobu I, Gidado M, Onazi O, Oshi DC. Economic support intervention improves tuberculosis treatment outcomes in rural Nigeria. Int J Tuberc Lung Dis. 2017;21(5):564-70. 18. Ukwaja KN, Alobu I, Gidado M, Onazi O, Oshi DC. Economic support intervention improves tuberculosis treatment outcomes in rural Nigeria. Int J Tuberc Lung Dis. 2017;21(5):564-70. 19. Diwati LV, Mavundla TR, Mbengo F. facilitators for and barriers to the implementation of national tuberculosis management guidelines. Afr J Nurs Midwifery. 2018;19:3. 19. Diwati LV, Mavundla TR, Mbengo F. facilitators for and barriers to the implementation of national tuberculosis management guidelines. Afr J Nurs Midwifery. 2018;19:3. 20. Rao KD, Ramani S, Hazarika I, George S. When do vertical programmes strengthen health systems? A comparative assessment of disease-specific interventions in India. Health Policy Plann. 2014;29:495–505. 20. Rao KD, Ramani S, Hazarika I, George S. When do vertical programmes strengthen health systems? A comparative assessment of disease-specific interventions in India. Health Policy Plann. 2014;29:495–505. Page 11/16 Page 11/16 21. Tobin-West CI, Isodje A. Quality and rural-urban comparison of tuberculosis care in Rivers State, Nigeria. Pan Afr Med J. 2016;24:60. 22. Gebreegziabher SB, Yimer SA, Bjune GA. Qualitative Assessment of Challenges in Tuberculosis Control in West Gojjam Zone, Northwest Ethiopia: Health Workers’ and Tuberculosis Control Program Coordinators’ Perspectives. Tuberc Res Treat. 2016;2016:1-8. 23. Moosazadeh M, Amiresmaili M. Challenges in case finding of tuberculosis control program in Iran: A qualitative study. Bangl J Med Sci. 2018;17(3):462-9. 24. References Hartel LA, Yazbeck AS, Osewe PL. Responding to health system failure on tuberculosis in southern Africa. Health Syst Reform. 2018;4:2. 25. Krishnan I, Kumar A, Mini GK, Aravind LR. Evidence based interventions and implementation gaps in control of tuberculosis : a systematic review in low and middle-income countries with special focus on India. Indian J Tuberc. 2019;66(2):268-78. 26. Joshi B, Lestari T, Graham SM, Baral SC, Verma SC, Ghimire G, et al. The implementation of Xpert MTB/RIF assay for diagnosis of tuberculosis in Nepal: a mixed methods analysis. PLoS ONE. 2018;13(8):e0201731. 27. Rendell NL, Bekhbat S, Ganbaatar G, Dorjravdan M, Pai M, Dobler CC. Implementation of the Xpert MTB/RIF assay for tuberculosis in Mongolia: a qualitative exploration of barriers and enablers. PeerJ. 2017;5(7):e3567. 28. Linguissi LSG, Gwom LC, Nkenfou CN, Bates M, Petersen E, Zumla A, et al. Health systems in the Republic of Congo: challenges and opportunities for implementing tuberculosis and HIV collaborative service, research, and training activities. Int J Infect Dis. 2017;56:62-7. 29. Gebreegziabher SB, Bjune GA, Yimer SA. Patients’ and health system’s delays in the diagnosis and treatment of new pulmonary tuberculosis patients in West Gojjam Zone, Northwest Ethiopia: a cross- sectional study. BMC Infect Dis. 2016;16:673. 30. Conseil A, Mounier-Jack S, Coker R. Integration of health systems and priority health interventions: a case study of the integration of HIV and TB control programmes into the general health system in Vietnam. Health Policy Plann. 2010;25(suppl_1):i32-i6. 30. Conseil A, Mounier-Jack S, Coker R. Integration of health systems and priority health interventions: a case study of the integration of HIV and TB control programmes into the general health system in Vietnam. Health Policy Plann. 2010;25(suppl_1):i32-i6. 31. FMOH. The National Strategic Plan for Tuberculosis Control: Towards Universal Access to Prevention, Diagnosis and Treatment (2015-2020). Abuja, Nigeria:National Tuberculosis and Leprosy Control Programme, FMOH; 2014. 31. FMOH. The National Strategic Plan for Tuberculosis Control: Towards Universal Access to Prevention, Diagnosis and Treatment (2015-2020). Abuja, Nigeria:National Tuberculosis and Leprosy Control Programme, FMOH; 2014. 32. FMOH. Annual TB case finding for all states, 2018. Abuja: National TB and Leprosy Control Programme, FMOH; 2019. 32. FMOH. Annual TB case finding for all states, 2018. Abuja: National TB and Leprosy Control Programme, FMOH; 2019. 33. van Olmen J, Criel B, Bhojani U, Marchal B, van Belle S, Chenge MF, et al. Analysing health system dynamics–a framework. Stud Health Serv Organ Policy. 2012;2:1. 33. References van Olmen J, Criel B, Bhojani U, Marchal B, van Belle S, Chenge MF, et al. Analysing health system dynamics–a framework. Stud Health Serv Organ Policy. 2012;2:1. 34. WHO. Everybody's business- strengthening health systems to improve health outcomes: WHO's framework for action. Geneva, Switzerland: World Health Organisation; 2007. 34. WHO. Everybody's business- strengthening health systems to improve health outcomes: WHO's framework for action. Geneva, Switzerland: World Health Organisation; 2007. 35. Fusch PI, Ness LR. Are We There Yet? Data Saturation in Qualitative Research. The Qual Rep. 2015;20(9):1408-16. Page 12/16 Page 12/16 36. Ukwaja KN, Alobu I, Igwenyi C, Hopewell PC. The high cost of free tuberculosis services: patient and household costs associated with tuberculosis care in Ebonyi State, Nigeria. PLoS ONE. 2013;8(8):e731134. 37. Ukwaja KN, Alobu I, Igwenyi C, Hopewell PC. Household catastrophic payments for tuberculosis care in Nigeria: incidence, determinants, and policy implications for universal health coverage. Infect Dis Poverty. 2013;2:21. 37. Ukwaja KN, Alobu I, Igwenyi C, Hopewell PC. Household catastrophic payments for tuberculosis care in Nigeria: incidence, determinants, and policy implications for universal health coverage. Infect Dis Poverty. 2013;2:21. Tables Table 1 Details of study participants Participants District Number of participants Number of participants by level of care/health facility Primary health centre Secondary health centre Tertiary health centre Faith- based hospital LGTBS Urban 4 4 Rural 9 9 Facility focal persons Urban 4 2 1 1 1 Rural 6 4 1 Total 23 19 2 1 1 Table 1 Details of study participants Table 1 Details of study participants Page 13/16 Page 13/16 Table 2 Summary of key health systems enablers and constraints to implementation of TB control in Enugu State, South-eastern Nigeria Key enablers Key constraints Leadership and governance Stewardship from state-level NTP Weak bureaucratic accountability from local health system. Programme management support Low government attention to TB control program Regular performance review and coordination Health financing External funding for TB control from donors Budgeted funds are not released to TB control program at the state level. Absence of TB in local governments’ budget. TB supervisors’ motorbikes are not replaced for several years. Human resources Supportive supervision of facility TB focal persons. Unwillingness of health workers to work in TB control programme Frequent re-deployment of skilled TB service providers High number of untrained TB service providers. Health workers are owed several months of salaries Health technology Use of dedicated logistics agency for drug distribution. Logistics agency dumped drugs meant for entire local government in one location. Drug kits does not meet needs of extrapulmonary TB patients and those weighing more than 70kg. Shortage of human immunodeficiency virus (HIV) test kits. Health information system Availability of recording and reporting tools Change in tools are not matched with training of service providers. Adaptation of tools to strategies in TB control. I d i f l i Health technology Health information system microscopic centre Introduction of GeneXpert Concern for contracting TB Engagement of community volunteers and patent medicine vendors Lack of incentives to attract health workers Reduction in duration of treatment from 8 to 6 months Many TB treatment centres lack of TB laboratory Poorly functioning GeneXpert. Weak patient tracking system Withdrawal of incentive for community volunteers and patent medicine vendors. Limited number of TB/HIV collaborative sites. Figures microscopic centre Introduction of GeneXpert Concern for contracting TB Engagement of community volunteers and patent medicine vendors Lack of incentives to attract health workers Reduction in duration of treatment from 8 to 6 months Many TB treatment centres lack of TB laboratory Poorly functioning GeneXpert. Weak patient tracking system Withdrawal of incentive for community volunteers and patent medicine vendors. Limited number of TB/HIV collaborative sites. Concern for contracting TB Figures Page 15/16 igure 1 ealth systems dynamic framework for tuberculosis control Figure 1 Page 15/16 Health systems dynamic framework for tuberculosis control Page 15/16 Health systems dynamic framework for tuberculosis control AppendixI.docx Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Page 16/16
https://openalex.org/W4235041687	https://www.qeios.com/read/JZ0S5A/pdf	English	null	Caliciviridae	Definitions	2,020	cc-by	77	Qeios · Definition, February 2, 2020 Open Peer Review on Qeios Open Peer Review on Qeios Caliciviridae National Cancer Institute National Cancer Institute Qeios ID: JZ0S5A · https://doi.org/10.32388/JZ0S5A Open Peer Review on Qeios Source National Cancer Institute. Caliciviridae. NCI Thesaurus. Code C14304. A family of single-stranded, non-segmented, positive-sense RNA viruses with a hexagonal/spherical capsid and no envelope. These viruses can be spread by the fecal- oral or respiratory route and infection typically leads to acute gastroenteritis. 1/1
https://openalex.org/W2795474651	https://repository.up.ac.za/bitstream/2263/66306/1/Majonga_Racial_2018.pdf	English	null	Racial Variation in Echocardiographic Reference Ranges for Left Chamber Dimensions in Children and Adolescents: A Systematic Review	Pediatric cardiology	2,018	cc-by	6,552	Abstract Echocardiography plays a critical role in the assessment of cardiac disease. Important differences in echocardiographically derived cardiac chamber dimensions have been previously highlighted in different population groups in adult studies, but this has not been systematically studied in children, whose body size changes throughout childhood. The aim of this study was to review the distribution of available reference ranges for the left cardiac chamber dimensions in older children and adoles- cents. The following electronic data bases were searched: Medline, Embase and Web of Science were searched to identify studies which have established echocardiographic reference ranges of left heart parameters in children and adolescents from 1975 to December 2017. There was no geographical limitation. All results were imported into Endnote. Retrieved articles were screened and data extracted by two independent reviewers. A total of 4398 studies were retrieved, with 36 studies finally included in this review. 29 (81%) references were from North America and European (Caucasians) populations, with only one study each from Africa and South America. Two-dimensional and M-mode techniques were the most commonly used echocardiography techniques. There were methodological variations in techniques and normalisation of references. Comparison of selected cardiac measures showed significant differences for interventricular septal thickness among Black African, Indian, German and US American children. Available echocardiographic references cannot be generalised to all settings and therefore, there is need for locally relevant reference ranges. Africa and South America are particularly under- represented. Future studies should focus on developing comprehensive echocardiographic reference ranges for children from different racial backgrounds and should use standardised techniques. Keywords Echocardiography · Reference ranges · Left ventricle · Left atrium · Children Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00246-018-1873-0) contains supplementary material, which is available to authorized users. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00246-018-1873-0) contains supplementary material, which is available to authorized users. 5 National Mycobacterial Reference Laboratory, Leibniz Research Centre, Borstel, Germany 6 Department of Pediatrics, University of Zimbabwe, Harare, Zimbabwe 7 UiT The Arctic University of Norway, Tromsø, Norway 8 Department of Public Health, Faculty of Health Sciences, University of Pretoria, Pretoria, South Africa 9 Clinical Research Department, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK 5 National Mycobacterial Reference Laboratory, Leibniz Research Centre, Borstel, Germany * Edith D. Racial Variation in Echocardiographic Reference Ranges for Left Chamber Dimensions in Children and Adolescents: A Systematic Review Edith D. Majonga1,2,9 · Gabrielle Norrish3,4 · Andrea M. Rehman1 · Katharina Kranzer1,5 · Hilda A. Mujuru6 · Kusum Nathoo6 · Jon O. Odland7,8 · Juan P. Kaski3,4 · Rashida A. Ferrand1,2 Edith D. Majonga1,2,9 · Gabrielle Norrish3,4 · Andrea M. Rehman1 · Katharina Kranzer1,5 · Hilda A. Mujuru6 · Kusum Nathoo6 · Jon O. Odland7,8 · Juan P. Kaski3,4 · Rashida A. Ferrand1,2 Received: 22 February 2018 / Accepted: 22 March 2018 / Published online: 4 April 2018 © The Author(s) 2018 Pediatric Cardiology (2018) 39:859–868 https://doi.org/10.1007/s00246-018-1873-0 Pediatric Cardiology (2018) 39:859–868 https://doi.org/10.1007/s00246-018-1873-0 REVIEW ARTICLE * Edith D. Majonga Edith.Majonga@lshtm.ac.uk Abstract Majonga Edith.Majonga@lshtm.ac.uk 6 Department of Pediatrics, University of Zimbabwe, Harare, Zimbabwe 1 London School of Hygiene and Tropical Medicine, London, UK 7 UiT The Arctic University of Norway, Tromsø, Norway 2 Biomedical Research and Training Institute, Harare, Zimbabwe 8 Department of Public Health, Faculty of Health Sciences, University of Pretoria, Pretoria, South Africa 3 Centre for Inherited Cardiovascular Diseases, Great Ormond Street Hospital, London, UK 9 Clinical Research Department, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK 4 Institute of Cardiovascular Science, University College London, London, UK (0123 1 3456789) 3 860 Pediatric Cardiology (2018) 39:859–868 Search Strategy The following electronic databases were searched: Med- line, Embase and Web of Science were searched. In addi- tion, reference lists of selected studies and other systematic reviews were manually reviewed to identify other possible studies for inclusion. The search strategy included the fol- lowing words: “echocardiography” AND “reference values” OR “normative” OR “reference standards” OR “reference intervals” AND “child” OR “children” OR “adolescent” OR “z score” (Table 1). Appropriate Boolean operators and trun- cation were used on synonyms. Both medical subjects and Inclusion Criteria and Exclusion Criteria Studies including at least 50 healthy participants aged 5–21 years that reported echocardiographic measurements at rest were included. We considered studies written in Eng- lish and published in peer-reviewed journals. Studies that only included neonates and infants, were conducted at high altitude (> 24,000 m above sea level); involved performing cardiac measures during or after exercise; or were based on autopsy specimens were excluded. Systematic reviews and meta-analyses were also excluded. The aim of this study was to systematically review racial distribution and methods used in available echo- cardiographic reference ranges for left ventricular (LV) and atrial (LA) chamber dimensions in children and ado- lescents. In addition, we compared values of selected chamber dimensions in different racial groups which have utilised the same methods for any differences. Introduction All available studies that reported echocardiographic refer- ence ranges for left cardiac chamber dimensions in healthy children and adolescents, regardless of echocardiographic technique, were considered for inclusion in this review. Echocardiography plays a critical role in the assessment of cardiac structure and function. Due to changes in body size during childhood, the evaluation of cardiac chambers is highly reliant on the availability of reference ranges, the quality of which depends largely on the availability of a representative sample of healthy subjects and the meth- ods employed to collect the data. The definition of what is “normal” varies widely according to age, body surface area (BSA), gender and race [1, 2]. Studies in adults have shown racial differences in echocardiographically derived cardiac chamber dimensions [1]. These differences may be more apparent in children whose body size changes throughout childhood, but this has not been investigated systematically. Table 1 Search strategy Materials and Methods This review was registered with the international prospec- tive register of systematic reviews (PROSPERO; registra- tion number CRD42015026030). Table 1 Search strategy Concepts Set Search terms Echocardiography 1 echocardiography.mp or exp echocardiography/ References 2 exp reference values/or normative.mp 3 Reference adj1 values 4 Reference adj1 standards) 5 Reference adj1 interval* 6 Normal values/or reference value/or reference interval.mp Children 7 Child.mp or child/ 8 adolescent.mp or exp adolescent/ 9 pediatric.mp or pediatrics/ 10 Paediatric.mp or paediatrics/ 11 Children*.mp 12 Infant.mp or exp Infant/ Z-score 13 z-score.mp 14 2 or 3 or 4 or 5 or 6 or 13 15 7 or 8 or 9 or 10 or 11 or 12 or 13 16 1 and 14 17 15 and 16 Table 1 Search strategy 1 3 Pediatric Cardiology (2018) 39:859–868 861 Table 2 Criteria for assessment of quality of studies using the New- castle-Ottawa Scale adapted for cross-sectional studies Scale of coding Representativeness of sample Truly representative = 3 Somewhat representative = 2 No description of sampling strategy = 1 Sample size Justified and acceptable = 2 Not justified = 1 Sample selection Hospital/volunteer = 3 Databases = 2 Not reported = 1 Standardisation of images American Society for Echocardiography guidelines/other = 2 Not reported = 1 Statistical methods Rigorous with clear exclusion criteria of abnormal cases = 3 Acceptable = 2 Not appropriate or incompletely described = 1 Total score 11–13 = good quality 8–10 = acceptable 5–7 = poor Table 2 Criteria for assessment of quality of studies using the New- castle-Ottawa Scale adapted for cross-sectional studies Scale of coding Representativeness of sample Truly representative = 3 Somewhat representative = 2 No description of sampling strategy = 1 Sample size Justified and acceptable = 2 Not justified = 1 Sample selection Hospital/volunteer = 3 Databases = 2 Not reported = 1 Standardisation of images American Society for Echocardiography guidelines/other = 2 Not reported = 1 Statistical methods Rigorous with clear exclusion criteria of abnormal cases = 3 Acceptable = 2 Not appropriate or incompletely described = 1 Total score 11–13 = good quality 8–10 = acceptable 5–7 = poor keywords were used. The same search strategy was adapted for all the listed databases. All results were imported into Endnote X7 (Thomas Reuter). Duplicate citations were removed. Titles and abstracts from the search results were screened independently by two reviewers (EDM) and (GN). Materials and Methods The full texts of potentially eli- gible studies were obtained and assessed in duplicate using a standardised checklist. Any disagreements about inclusion of studies were resolved by consensus. Quality of the Studies We adapted the Newcastle-Ottawa Scale for assessing qual- ity of non-randomised studies to suit cross-sectional studies in the systematic review [3]. In our tool, we assigned scores instead of stars (Table 2). The following criteria were used to determine the quality of the studies: representativeness of the sample; sample size; sample selection; standardisation of image acquisition and statistical methods used. M-mode and/or two-dimensional (2D) echocardio- graphic techniques were used in 33 (92%) studies. One study used three-dimensional (3D) echocardiography in addition to M-mode and 2D, another utilised 3D only and a third study used a rarely practiced echocardio- graphic technique called acoustic quantification [18, 21, 22]. Anthropometric and non-anthropometric measures were used for normalising the results: body surface area (BSA) in 18 studies, height (n = 4) [8, 11, 17, 28], weight (n = 3) [10, 20, 35], age (n = 2) [14, 15], and lean body mass (n = 1) [7]. The remaining studies used age and heart rate, (n = 1); [22] age and BSA, (n = 2); [26, 37] age and height, (n = 1); [27] height and BSA, (n = 1); [33] weight and BSA, (n = 1) [36] and one study used height, BSA and lean body mass [25]. Varying methods to calculate the BSA were used: 12 (33%) studies used the Dubois and Dubois method, and 4 (14%) studies used the Hay- cock. Daubeney et al. used the Boyd method; [24] Saito et al. calculated BSA using the West Nomogram [37] and five studies did not report the method used for calculat- ing BSA [25, 29, 32, 39]. 23 (64%) studies standardised Data Extraction and Analysis The following data were extracted using a standard data extraction form: author; population studied; sample size; age range; echocardiography technique; parameters measured, and type of normalisation used. A two-sample Kolmogorov–Smirnov non-parametric test was used to compare distributions of z score = 0 and 2 for selected cardiac measures of studies which used the same technique for performing echocardiography and the same method for calculation of BSA for normalisation. The selected z score represents the mean predicted value and the upper cut-off for the normal range for a cardiac meas- ure. The null hypothesis was that the compared groups were sampled from populations with identical distributions. Due to multiple testing, the chance of obtaining a significant p value when in fact there is not a true difference between distributions was high. Therefore, we used the Bonferroni adjustment, with a p value < 0.005 considered significant. of children studied [6, 8, 9, 19, 23, 25, 33, 38, 39]. Sample sizes of the studies ranged from 95 to 9858. The quality of the studies was good (n = 16) or acceptable (n = 20) in all cases. Results A total of 4398 citations were retrieved dating from 1975 to June 2017. Of these, 1193 duplicates were removed, and a further 3075 citations were excluded based on title and abstract (Fig. 1). Full texts of a total of 130 studies were reviewed and 36 studies were included. Characteristics of the included studies are in Table 3. Sixteen (44%) studies were conducted in North America, followed by Europe (n = 13, 36%) and Asia (n = 5, 14%). Only one study each was conducted in South America (Bra- zil) and Africa (Zimbabwe). Nine studies reported the race 1 1 3 862 Pediatric Cardiology (2018) 39:859–868 862 Pediatric Cardiology (2018) 39:85 Records identified through database (n =4388) Screening Included Eligibility Identification Additional records from other sources (n =10) Records after duplicates removed (n =3205) Records screened (n = 175) Records excluded (n =44) Full-text articles assessed for eligibility (n =131) Full-text articles excluded, with reasons (n =95) LV or LA function only (n=48) Only males (n=1)) Small numbers of children (n=21) No normalisation to body size (n= 6) No subgroups for younger ages (n=3) Adults or younger children only (n=3) Subjects not truly healthy (n=3) Exercise studies (n=5) Non-English (n=1) Not relevant (n=3) High altitude (n=1) Studies included in qualitative synthesis (n = 36) Fig. 1 PRISMA flow diagram for process of selecting included studies Records identified through database (n =4388) Screening Included Eligibility Identification Additional records from other sources (n =10) Records after duplicates removed (n =3205) Records screened (n = 175) Records excluded (n =44) Full-text articles assessed for eligibility (n =131) Full-text articles excluded, with reasons (n =95) LV or LA function only (n=48) Only males (n=1)) Small numbers of children (n=21) No normalisation to body size (n= 6) No subgroups for younger ages (n=3) Adults or younger children only (n=3) Subjects not truly healthy (n=3) Exercise studies (n=5) Non-English (n=1) Not relevant (n=3) Studies included in qualitative synthesis (n = 36) Additional records from other sources (n =10) Records after duplicates removed (n =3205) Records excluded (n =44) Records screened (n = 175) Studies included in qualitative synthesis (n = 36) Fig. 1 PRISMA flow diagram for process of selecting included studies Left Ventricular Dimensions images and performed measurements according to recom- mendations by the American Society of Echocardiogra- phy (ASE); one study used ASE and European Society of Cardiology recommendations and another study used ASE and Penn convention [23, 26]. The remaining stud- ies used other methods of performing measurements, including inner edge to inner edge method in five studies [24, 28, 30, 31, 34]; trailing edge to leading edge in one study [32]; leading edge to leading edge in one study [35]; standard and Penn convention in one study [37]. Three studies described how they measured parameters without stating a specific convention [21, 38, 39]. images and performed measurements according to recom- mendations by the American Society of Echocardiogra- phy (ASE); one study used ASE and European Society of Cardiology recommendations and another study used ASE and Penn convention [23, 26]. The remaining stud- ies used other methods of performing measurements, including inner edge to inner edge method in five studies [24, 28, 30, 31, 34]; trailing edge to leading edge in one study [32]; leading edge to leading edge in one study [35]; standard and Penn convention in one study [37]. Three studies described how they measured parameters without stating a specific convention [21, 38, 39]. The following LV dimensions were reported: diameter at end-diastole (LVEDD) and/or end-systole (LVESD), pos- terior wall and interventricular septal, area, length, vol- ume, LV mass and index. Saito et al. developed references for LV muscle volume, which is a rarely used measure in clinical practice [37]. 13 studies were from North Ameri- can children; 12 studies among European children; five studies from Asia and one study each from South Ameri- can and African children (Table 3). 1 3 3 Pediatric Cardiology (2018) 39:859–868 863 1 3 Table 3 Characteristics of studies Author Population studied Sample size Age range (y) Technique Parameters measured Standardisation of images Normalisation Quality of the study Lopez et al. [4] US American and Canadian 3215 0–18 2D LVEDD, IVSd, LVPWd, LV length, LV area, LV vol- ume, LVM ASE BSA (Haycock) Good Gokhroo et al. [5] Indian 746 4–15 M mode/2D LVEDD, LVESD, IVSd, IVSs, LVPWd, LVPWs, LA length ASE BSA (Haycock) Good Majonga et al. [6] Zimbabwean (Black) 282 6–16 M mode LVEDD, LVESD, IVSd, IVSs, LVPWd, LVPWs, LA ASE BSA (Dubois) Good Foster et al. Left Ventricular Dimensions [7] US Americans and Canadian 1710 5–18 M mode/2D LVM ASE Lean body mass Good Chinali et al. [8] Italian and German (Caucasians) 400 0–18 M mode/2D LVM ASE Height Good Cantinotti et al. [9] Italian (Caucasians) 1091 0–17 M mode/2D LVED area, LVES area, LVED length, LVES length, LVEDD, LVESD, LA length, LA area ASE BSA (Haycock) Good Oran et al. [10] Turkish 1200 0–17 M mode/2D LVEDD, LVESD, IVSd, LVPWd, LA ASE Weight Good Motz et al. [11] German 9858 0–19 M mode/2D LVEDD, LVESD, IVSd, LVPWd ASE Body length Acceptable Bhatla et al. [12] US American 300 0–18 2D LA volume ASE BSA (Haycock) Acceptable Kervancioglu et al. [13] Turkish 208 0–14 M mode/2D LVM ASE BSA (Dubois) Acceptable Taggart et al. [14] US American 522 0–17 2D LA volume, LA volume index ASE Age Acceptable Khoury et al. [15] US American 2273 0–18 M mode/2D LVM and LVMI ASE Age Good Pettersen et al. [16] US American 782 0–18 2D/M-mode IVSd, IVSs, LVEDD, LVESD, LVPWd, LVPWs, LA ASE BSA (Dubois) Good Foster et al. [17] US American & Canadian 440 0–17 M mode LVM ASE Height Good Poutanen et al. [18] Finnish 169 2–27 M mode/2D/3D LVM ASE BSA (Dubois) Acceptable Bonatto et al. [19] Brazilian (Blacks and whites) 595 0–12 M mode LVEDD, IVSd, LVPWd, LA, LVM, LVMI ASE BSA (Dubois) Acceptable 3 Pediatric Cardiology (2018) 39:859–868 864 1 3 ( ) Author Population studied Sample size Age range (y) Technique Parameters measured Standardisation of images Normalisation Quality of the study Overbeek et al. [20] Netherlands (Dutch) 587 0–18 M mode LVEDD, LVESD, IVSd, LVPWd ASE Weight Acceptable Poutanen et al. [21] Finnish 169 2–27 3D LA volumes, LVEDV, LVESV Other BSA (Dubois) Acceptable Brangenberg et al. [22] German 150 0–14.5 Acoustic quantifica- tion LV volume and area ASE Age/HR Good Kampmann et al. [23] German (Whites) 2036 0–18 M mode IVSd, IVSs, LVEDD, LVESD, LVPWd, LVPWs, LA ASE/ESC BSA (Dubois) Good Daubeney et al. [24] British + Australian 125 0–18 2D LV inflow, LV area Other (inner edge to inner edge) BSA (Boyd) Good Daniels et al. [25] US American (Blacks and Whites) 192 7–17 M mode LVM ASE Height/lean body mass/BSA (NR) Acceptable Huwez et al. [26] British 127 0–19 M mode IVSd, LVEDD, LVESD, LVPWd, LA, LVM, LV volume ASE & Penn conven- tion Age/BSA (Dubois) Good Malcolm et al. Left Atrium Dimensions Author Population studied Sample size Age range (y) Technique Parameters measured Standardisation of images Normalisation Quality of the study Saito et al. [37] Japanese 301 6–15 M mode LV muscle volume Other (Standard & Penn convention) Age/BSA (West Nomogram) Acceptable Henry et al. [38] US American (Cau- casians, Blacks and Oriental) 105 0–23 M mode IVSd, IVSs, LVEDD, LVESD, LVPWd, LVPWs, LA, LVM Other BSA (Dubois) Acceptable Epstein et al. [39] US American (Whites, Blacks, Oriental/East Indians) 205 0–18 M mode IVSd, IVSs, LVEDD, LVESD, LVPWd, LVPWs, LA Other BSA (NR) Acceptable M-mode motion mode, 2D two-dimensional, 3D three-dimensional, US United States, M-mode motion mode, BSA body surface area, LV left ventricle, LVEDD left ventricular diameter at end-diastole, LVESD left ventricular diameter at end-systole, LVEDV left ventricular volume at end-diastole, LVESV left ventricular volume at end-systole, IVSd interventricular septum at end- diastole, IVSs interventricular septum at end-systole, LVPWd left ventricular posterior wall at end-diastole, LVPWs left ventricular posterior wall at end-systole, LVM left ventricular mass, LVMI left ventricular mass index, LA left atrium, HR heart rate, NR not reported LA diameter references were established in twelve studies, six of which were conducted in US American children [16, 28, 30, 38, 39]. LA length references were from Indian, Italian and Swedish children and of these, two studies also reported data on LA area [5, 9, 32]. LA volume was derived in four studies and three of these were in US American chil- dren (Table 3) [12, 14, 21, 30]. Comparison of LV and LA Dimension Between Studies Reference values of selected cardiac measures from different racial groups which used M-mode technique and normalised results with BSA were identified and compared (Tables 4, 5). The p values for the compared references are shown in Supplementary Tables 1 and 2 and these are for z score = 0 and + 2 distributions. The graphical representations of the distributions for the selected cardiac measures are shown in Supplementary Figs. 1 and 2. Mean left ventricular diam- eter at end-diastole (LVEDD) among US American children and German children was similar (p = 0.906). On the other hand, Zimbabwean children had thicker mean interventricu- lar septum at end-diastole (IVSd) than German children had (p < 0.001) while US American children and German chil- dren were similar (p = 0.281). Mean LA diameter was also similar between British and German children (p = 0.699). Comparison using predicted values of z score = +2 (cut- off for upper limit of normal), significant differences were noted in the IVSd measures between Zimbabwean and German children (p = 0.001). IVSd measures between Ger- man and US American children were significantly different (p < 0.001). No significant differences were noted on the dis- tributions of LA measures between the compared studies. The predicted values of z score = +2 by Kampmann et al. progressed in a step-wise fashion so for example, a value of IVSd > 10.4 mm had a z score = +2 for BSA between 1.7 and 1.9 m2. The shape of Kampmann’s distribution for z score = +2 was therefore strikingly different from the other references [23]. Left Ventricular Dimensions [27] US American 904 6–16 M mode LVM ASE Age/height Good Nidorf et al. [28] US American 196 0–18 2D LV diameter, LV length, LA diameter Other (inner edge to inner edge) Height Acceptable Vogel et al. [29] German 95 1–17 2D LVM, LV volume ASE BSA (NR) Acceptable Pearlman et al. [30] US American 268 0–18 2D LA diameters, LA volume Other (inner edge to inner edge) BSA (Dubois) Acceptable Pearlman et al. [31] US American 268 0–18 2D LV length, LV area Other (inner edge to inner edge) BSA (Dubois) Acceptable Hanseus et al. [32] Swedish 120 0–16 2D LA length, area and width, LV diameter, area and width Other (trailing edge to leading edge) BSA (NR) Acceptable Daniels at al [33] US American (Blacks and Whites) 334 6–23 M mode LVMI ASE Height/BSA (Dubois) Acceptable Akiba et al. [34] Japanese 110 0–15 M mode LVEDD, LVESD, LVPWd, LVPWs, LV volume Other (inner to inner edge) BSA (Haycock) Acceptable Voogd et al. [35] Netherlands (Dutch) 432 4–17 M mode IVSd, IVSs, LVEDD, LVESD, LVPWd, LVPWs, LA Other (leading edge of echoes) Weight Good Henry et al. [36] US America 92 0–23 M mode LVEDD, LVESD, IVSd, IVSs, LVPWd, LVPWs, LA ASE Weight/BSA (NR) Acceptable 1 3 Pediatric Cardiology (2018) 39:859–868 865 D two-dimensional, 3D three-dimensional, US United States, M-mode motion mode, BSA body surface area, LV left ventricle, LVEDD left ventricular diameter at ventricular diameter at end-systole, LVEDV left ventricular volume at end-diastole, LVESV left ventricular volume at end-systole, IVSd interventricular septum at end- cular septum at end-systole, LVPWd left ventricular posterior wall at end-diastole, LVPWs left ventricular posterior wall at end-systole, LVM left ventricular mass, LVMI x LA left atrium HR heart rate NR not reported Discussion [23] German (Caucasian) M-mode Dubois 25 (19.2–30.8) e 5 A comparison of published echocardiographic normal references for LA dimensions in studies using M-mode a child with 1 m2 BSA) or Pop lation st died Techniq e Method for BSA ± 2SD) de motion mode, BSA body surface area, 2D two-dimensional, LA left atrium, A4C apical four chamber view ( ) US United States, M-mode motion mode, BSA body surface area, 2D two-dimensional, LA left atrium, A4C ap Differences between Zimbabwean and German children were consistently demonstrated at both mean-level and upper cut-off for normal distributions. Although no interventricular septal thickness difference was found in the mean distribu- tion of German and US American children, it was evident in the predicted values z score = +2. In practice, it is the upper cut-off for normal which is used to define abnormality rather than the mean. Our findings suggest that differences in reference ranges between different racial groups do exist but may be overlooked because of the scarcity of data e.g. Afri- can children. The use of inappropriate reference ranges may result in either under- or over-diagnosis of cardiac abnor- malities or missing of early cardiac chamber remodelling due to cardiac disease [41]. This highlights the importance of using racial-specific reference ranges in clinical practice.f for chamber dimensions in children. Most of the reference ranges in children were, however, developed in European and North American (US populations) and mainly in Cau- casians. Notably, there was only one study each from South America (Brazil) and Africa (Zimbabwe) and the latter was published very recently, implying that most African coun- tries rely on Western references in clinical practice which may not accurately represent black African children [6, 19]. Comparability of the available reference ranges was lim- ited due to substantial methodological variations, including parameters of normalisation; technique of image acquisition (e.g. 2D or M-mode); measured chamber dimension and/or method used for performing the measurements (e.g. ASE guidelines or other). Cantinotti et al. highlighted that it is imperative to standardise methods of image acquisition and consistency in normalisation of the references [40]. i In a recently published study on effect of age, sex, race and ethnicity in echocardiographic z scores of children, significant effects by all the four parameters were observed on z scores. However, the authors concluded that these were not of clinical significance [4]. 1 3 Discussion Accurate assessment of left cardiac chamber size in chil- dren relies on the availability of representative reference ranges. In this study, we have systematically reviewed dif- ferences between reference ranges for LV and LA chamber dimensions in children and adolescents. We found many studies which have established reference ranges for LV and LA chamber dimensions, reflecting the significant inter- est in, and importance of, establishing reference ranges 1 3 1 3 Pediatric Cardiology (2018) 39:859–868 866 Table 4 A comparison of published echocardiographic normal references for LV dimensions in studies using M-mode and normalised to BSA (for a child with 1 m2 BSA) Dimensions are mean (± 2SD) Author Population studied Method for BSA LVEDD(mm) LVESD (mm) IVSd (mm) IVSs (mm) LVPWd (mm) LVPWs (mm) Gokhroo et al. [5] Indian Haycock 35.02 (27.02– 42.04) 21.32 (13.81– 28.84) 7.4 (5.5–9.3) 11.0 (8.11– 13.9) 7.2 (5.4–9.1) 10.8 (10.1– 11.5) Cantinotti et al. [9] Italian Haycock 37.86 (31.56– 45.42) 22.97 (17.46– 30.20) Majonga et al. [6] Zimbabwean Dubois 37.10 (32.43– 41.76) 25.29 (20.84– 29.74) 7.0 (5.0–9.1) 9.2 (6.7–11.6) 6.8 (5.2–8.5) 9.0 (6.5–11.4) Pettersen et al. [16] US American Dubois 39.09 (32.06– 48.79) 25.1 (19.6– 32.1) 5.9 (3.9–9.0) 8.6 (6.0–12.3) 5.4 (3.7–7.9) 10.3 (7.7–13.9) Kampmann et al. [23] German Dubois 38.50 (31.70– 45.30) 24.4 (18.6– 30.2) 5.8 (4.0–7.6) 8.4 (5.1–11.7) 5.9 (3.7–8.1) 9.5 (6.8–12.2) Huwez et al. [26] British Dubois 38.27 (33.05– 43.49) 24.28 (19.79– 28.77) 7.1 (5.2–7.1) 6.4 (4.6–8.3) Dimensions are mean (± 2SD) US United States, M-mode motion mode, BSA body surface area, LV left ventricle, LVEDD left ventricular diameter at end-diastole, LVESD left ventricular diameter at end-systole, IVSd interventricular septum at end-diastole, IVSs interventricular septum at end-systole, LVPWd left ven- tricular posterior wall at end-diastole, LVPWs left ventricular posterior wall at end-systole, LVM left ventricular mass Table 5 A comparison of published echocardiographic normal references for LA dimensions in studies using M-mode and normalised to BSA (for a child with 1 m2 BSA) Dimensions are mean (± 2SD) US United States, M-mode motion mode, BSA body surface area, 2D two-dimensional, LA left atrium, A4C apical four chamber view Author Population studied Technique Method for BSA LA diameter (mm) Majonga et al. [6] Zimbabwean (Blacks) M-mode Dubois 24.05 (19.19–28.91) Huwez et al. [26] British M-mode Dubois 25.9 (20.3–31.6) Kampmann et al. References 1. Echocardiographic Normal Ranges Meta-Analysis of the Left Heart C (2015) Ethnic-specific normative reference values for echocardiographic LA and LV size, LV mass, and systolic function: the EchoNoRMAL Study. JACC Cardiovasc Imaging 8:656–665 f Most studies used the M-mode or 2D echocardiography techniques. Advanced techniques such as 3D echocardiogra- phy, which may overcome some of the technical challenges of angle dependence and other geometric assumptions associated with conventional techniques, should be used to more accu- rately quantify chamber sizes and development of reference ranges [42]. 2. Pfaffenberger S, Bartko P, Graf A, Pernicka E, Babayev J, Lolic E, Bonderman D, Baumgartner H, Maurer G, Mascherbauer J (2013) Size matters! Impact of age, sex, height, and weight on the normal heart size. Circ Cardiovasc Imaging 6:1073–1079 3. Wells G, Shea B, O’connell D, Peterson J, Welch V, Losos M, Tugwell P (2011) The Newcastle-Ottawa Scale (nos) for assess- ing the quality of nonrandomised studies in meta-analyses. [Internet]. Ottawa Hospital Research Institute, Ottawa This study is limited by the fact that many of the studies did not report the actual race of the children, and we there- fore made assumptions based on the country the study was conducted in. We compared very broad racial groups due to scarcity of data. In addition, we were unable to compare ref- erences for other cardiac measures due to varying methods used in the studies. However, in the few selected references and cardiac measures where this was possible, we were able to demonstrate significant differences in different races. We also showed similarities in same racial groups. It is also highly likely that in addition to the varying methods used in other studies, racial differences are also present. p 4. Lopez L, Colan S, Stylianou M, Granger S, Trachtenberg F, Frommelt P, Pearson G, Camarda J, Cnota J, Cohen M, Dragulescu A, Frommelt M, Garuba O, Johnson T, Lai W, Mahgerefteh J, Pignatelli R, Prakash A, Sachdeva R, Soriano B, Soslow J, Spurney C, Srivastava S, Taylor C, Thankavel P, van der Velde M, Minich L (2017) Relationship of echocar- diographic Z Scores adjusted for body surface area to age: the pediatric heart network normal echocardiogram database, Circ Cardiovasc Imaging 10: e006979 5. Conclusion This review underlines the importance of using race-specific reference ranges for children, as well as the need for standard- ising echocardiographic methods in deriving those reference ranges. Furthermore, these reference ranges need to be com- prehensive, including a wide range of cardiac measures, as some studies only reported normal values for a single cardiac measure. Future studies should focus on including 3D param- eters in addition to 2D and M-mode. 7. Foster BJ, Khoury PR, Kimball TR, Mackie AS, Mitsnefes M (2016) New reference centiles for left ventricular mass relative to lean body mass in children. J Am Soc Echocardiogr 29:441–447 8. Chinali M, Emma F, Esposito C, Rinelli G, Franceschini A, Doyon A, Raimondi F, Pongiglione G, Schaefer F, Matteucci MC (2016) Left ventricular mass indexing in infants, children, and adoles- cents: a simplified approach for the identification of left ventricu- lar hypertrophy in clinical practice. J Pediatr 170:193–198 9. Cantinotti M, Scalese M, Murzi B, Assanta N, Spadoni I, De Lucia V, Crocetti M, Cresti A, Gallotta M, Marotta M, Tyack K, Molinaro S, Iervasi G (2014) Echocardiographic nomograms for chamber diameters and areas in Caucasian children. J Am Soc Echocardiogr 27:1279 Funding This study was funded by the Wellcome Trust (Grant Number 095878/Z/11/Z). 10. Oran B, Bodur AS, Arslan D, Cimen D, Guvenc O (2014) Nor- mal M mode values in healthy Turkish children. Turk J Med Sci 44:756–763 References Gokhroo RK, Anantharaj A, Bisht D, Kishor K, Plakkal N, Agho- ram R, Mondal N, Pandey SK, Roy R (2017) A pediatric echocar- diographic Z-score nomogram for a developing country: Indian pediatric echocardiography study—the Z-score. Ann Pediatr Car- diol 10:31–38 6. Majonga ED, Rehman AM, McHugh G, Mujuru HA, Nathoo K, Patel MS, Munyati S, Odland JO, Kranzer K, Kaski JP, Ferrand RA (2017) Echocardiographic reference ranges in older children and adolescents in sub-Saharan Africa. Int J Cardiol 248:409–413 Discussion Given that this study was conducted mainly in US American and Canadian children, findings cannot be However, we were able to compare a few studies which used similar methods from different races for selected car- diac measures. There were notable differences in some of the measures, particularly interventricular septal thickness among Zimbabwean, German and US American children. 1 3 Pediatric Cardiology (2018) 39:859–868 867 generalisable to the rest of the world due to non-standardi- sation of methods in the available references. There are also other geographical confounders such as nutrition and altitude which may affect cardiovascular development [4]. References Compliance with Ethical Standards Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest. 11. Motz R, Schumacher M, Nurnberg J, Viemann M, Grafmuller S, Fiedler K, Claus M, Kronberg K (2014) Echocardiographic measurements of cardiac dimensions correlate better with body length than with body weight or body surface area. Pediatr Cardiol 35:1327–1336 Ethical Approval All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. 12. Bhatla P, Nielsen JC, Ko HH, Doucette J, Lytrivi ID, Srivastava S (2012) Normal values of left atrial volume in pediatric age group using a validated allometric model. Circ Cardiovasc Imaging 5:791–796 13. Kervancioglu P, Kervancioglu M, Tuncer MC, Hatipoglu ES (2011) Left ventricular mass in normal children and its correla- tion with weight, height and body surface area. Int J Morphol 29:982–987 Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creativeco mmons.org/licenses/by/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 14. Taggart NW, Cetta F, O’Leary PW, Seward JB, Eidem BW (2010) Left atrial volume in children without heart disease and in those 1 3 868 Pediatric Cardiology (2018) 39:859–868 cardiac chamber dimensions during development and adulthood. J Am Coll Cardiol 19:983–988 with ventricular septal defect or patent ductus arteriosus or hyper- trophic cardiomyopathy. Am J Cardiol 106:1500–1504 29. Vogel M, Staller W, Buhlmeyer K (1991) Left ventricular myocar- dial mass determined by cross-sectional echocardiography in nor- mal newborns, infants, and children. Pediatr Cardiol 12:143–149 p y p y 15. Khoury PR, Mitsnefes M, Daniels SR, Kimball TR (2009) Age- specific reference intervals for indexed left ventricular mass in children. J Am Soc Echocardiogr 22:709–714 16. Pettersen MD, Du W, Skeens ME, Humes RA (2008) Regres- sion equations for calculation of Z scores of cardiac structures in a large cohort of healthy infants, children, and adolescents: an echocardiographic study. J Am Soc Echocardiogr 21:922–934 30. Pearlman JD, Triulzi MO, King ME, Abascal VM, Newell J, Wey- man AE (1990) Left atrial dimensions in growth and develop- ment: normal limits for two-dimensional echocardiography. Compliance with Ethical Standards J Am Coll Cardiol 16:1168–1174 31. Pearlman JD, Triulzi MO, King ME, Newell J, Weyman AE (1988) Limits of normal left ventricular dimensions in growth and development: Analysis of dimensions and variance in the two- dimensional echocardiograms of 268 normal healthy subjects. J Am Coll CardioI 12:1432–1441 17. Foster BJ, Mackie AS, Mitsnefes M, Ali H, Mamber S, Colan SD (2008) A novel method of expressing left ventricular mass relative to body size in children. Circulation 117:2769–2775 18. Poutanen T, Jokinen E (2007) Left ventricular mass in 169 healthy children and young adults assessed by three-dimensional echocar- diography. Pediatr Cardiol 28:201–207 32. Hanseus K, Bjorkhem G, Lundstrom NR (1988) Dimensions of cardiac chambers and great vessels by cross-sectional echocardi- ography in infants and children. Pediatr Cardiol 9:7–15 19. Bonatto RC, Fioretto JR, Okoshi K, Matsubara BB, Padovani CR, Manfrin TCR, Gobbi M, Martino RS, Bregagnollo EA (2006) Percentile curves of normal values of echocardiographic measure- ments in normal children from the central-southern region of the State of Sao Paulo, Brazil. Arq Bras Cardiol 87:711–721 33. Daniels SR, Meyer RA, Liang YC, Bove KE (1988) Echocar- diographically determined left ventricular mass index in nor- mal children, adolescents and young adults. J Am Coll Cardiol 12:703–708 20. Overbeek LIH, Kapusta L, Peer PGM, de Korte CL, Thijssen JM, Daniels O (2006) New reference values for echocardiographic dimensions of healthy Dutch children. Eur J Echocardiogr 7:113 – 21 34. Akiba T, Yoshikawa M, Otaki S, Kobayashi Y, Nakasato M, Suzuki H, Tetsuo S (1986) Echocardiographic measurements of left ventricle in normal infants and children. Tohoku J Exp Med 149:31–37 21. Poutanen T, Jokinen E, Sairanen H, Tikanoja T (2003) Left atrial and left ventricular function in healthy children and young adults assessed by three dimensional echocardiography. Heart 89:544–549 35. Voogd PJ, Rijsterborgh H, Lubsen J, Arntzenius AC, Monsjou LK, Godijn EH (1984) Reference ranges of echocardiographic measurements in the Dutch population. Eur Heart J 5:762–70 22. Brangenberg R, Burger A, Romer U, Kozlik-Feldmann R, Netz H (2002) Echocardiographic assessment of left ventricular size and function in normal children from infancy to adolescence: acoustic quantification in comparison with traditional echocardiographic techniques. Pediatr Cardiol 23:394–402f 36. Henry WL, Gardin IM, Ware JH (1980) Echocardiographic meas- urements in normal subjects from infancy to old age. Circulation 62:1054–1061 37. Saito M, Mori C, Nishio T (1978) Normal values of echocardiog- raphy in pediatrics. I. Compliance with Ethical Standards Left ventricular muscle volume (LVMV). The Shimane heart study. Shimane J Med Sci 2:63–69 23. Kampmann C, Wiethoff CM, Wenzel A, Stolz G, Betancor M, Wippermann CF, Huth RG, Habermehl P, Knuf M, Emschermann T, Stopfkuchen H (2000) Normal values of M mode echocar- diographic measurements of more than 2000 healthy infants and children in central Europe. Heart 83:667–672 38. Henry WL, Ware J, Gardin JM, Hepner SI, Mckay J, Weiner M (1978) Echocardiographic measurements in normal subjects. Growth-related changes that occur between infancy and early adulthood. Circulation 57:278–285 24. Daubeney PEF, Blackstone EH, Weintraub RG, Slavik Z, Scanlon J, Webber SA (1999) Relationship of the dimension of cardiac structures to body size: an echocardiographic study in normal infants and children. Cardiol Young 9:402–410 39. Epstein ML, Goldberg ST, Allen HD, Konecke L, Wood J (1975) Great vessel, cardiac chamber, and wall growth patterns in normal children. Circulation 51:1124–1129 40. Cantinotti M, Kutty S, Franchi E, Paterni M, Scalese M, Iervasi G, Koestenberger M (2017) Pediatric echocardiographic nomo- grams: what has been done and what still needs to be done. Trends Cardiovasc Med 27:336–349i 25. Daniels SR, Kimball TR, Morrison JA, Khoury P, Meyer RA (1995) Indexing left ventricular mass to account for differences in body size in children and adolescents without cardiovascular disease. Am J Cardiol 76:699–701 41. Badano LP (2014) Ethnicity: a missing variable when defining normative values for reporting echocardiographic studies. Arch Cardiovasc Imaging 2(1):2 26. Huwez FU, Houston AB, Watson J, McLaughlin S, Macfarlane PW (1994) Age and body surface area related normal upper and lower limits of M mode echocardiographic measurements and left ventricular volume and mass from infancy to early adulthood. Br Heart J 72:276–280 42. Lang RM, Badano LP, Tsang W, Adams DH, Agricola E, Buck T, Faletra FF, Franke A, Hung J, de Isla LP (2012) EAE/ASE recommendations for image acquisition and display using three- dimensional echocardiography. Eur Heart J Cardiovasc Imaging 13:1–46 27. Malcolm DD, Burns TL, Mahoney LT, Lauer RM (1993) Factors affecting left ventricular mass in childhood: the Muscatine study. Pediatrics 92:703–709 28. Nidorf SM, Picard MH, Triulzi MO, Thomas JD, Newell J, King ME, Weyman AE (1992) New perspectives in the assessment of 1 3
https://openalex.org/W2995775530	https://europepmc.org/articles/pmc7028824?pdf=render	English	null	Uncertainty quantification on a spatial Markov-chain model for the progression of skin cancer	Journal of mathematical biology	2,019	cc-by	14,033	Mathematical Biology Mathematical Biology Mathematics Subject Classiﬁcation 60J20 · 65C05 · 65C40 Mathematics Subject Classiﬁcation 60J20 · 65C05 · 65C40 Mathematics Subject Classiﬁcation 60J20 · 65C05 · 65C40 Uncertainty quantiﬁcation on a spatial Markov-chain model for the progression of skin cancer Fred Vermolen1 · Ilkka Pölönen2 Received: 28 September 2018 / Revised: 14 March 2019 / Published online: 19 December 2019 © The Author(s) 2019 Abstract A spatial Markov-chain model is formulated for the progression of skin cancer. The model is based on the division of the computational domain into nodal points, that can be in a binary state: either in ‘cancer state’ or in ‘non-cancer state’. The model assigns probabilities for the non-reversible transition from ‘non-cancer’ state to the ‘cancer state’ that depend on the states of the neighbouring nodes. The likelihood of transition further depends on the life burden intensity of the UV-rays that the skin is exposed to. The probabilistic nature of the process and the uncertainty in the input data is assessed by the use of Monte Carlo simulations. A good ﬁt between experiments on mice and our model has been obtained. 1 Delft Institute of Applied Mathematics, Delft University of Technology, Delft, The Netherlands 2 Faculty of Information Technology, University of Jyväskulä, 40014 Jyvaskyla, Finland Journal of Mathematical Biology (2020) 80:545–573 https://doi.org/10.1007/s00285-019-01367-y Journal of Mathematical Biology (2020) 80:545–573 https://doi.org/10.1007/s00285-019-01367-y 1 Introduction Ageing of populations in industrialised society puts a huge burden on health care. In particular, next to cardio-vascular diseases, cancer is one of the most lethal diseases developing in these societies. Reasons are a sedentary lifestyle, diets that contain large amounts of sugar, as well as bad habits such as smoking. An underestimated class of cancers concerns skin cancer. According to the WHO (2017), one in every three cancers diagnosed is classiﬁed as a skin cancer. There are three major types of skin cancers including basal-cell carcinomas (BCC), squamous- cell carcinomas (SCC) and different types of melanoma. Each of these cancers develop in different cells: BCC develops in basal cells which are squamous cells located close to B Fred Vermolen F.J.Vermolen@tudelft.nl Ilkka Pölönen ilkka.polonen@jyu.ﬁ 1 Delft Institute of Applied Mathematics, Delft University of Technology, Delft, The Netherlands 2 Faculty of Information Technology, University of Jyväskulä, 40014 Jyvaskyla, Finland 1 Delft Institute of Applied Mathematics, Delft University of Technology, Delft, The Netherlands 2 Faculty of Information Technology, University of Jyväskulä, 40014 Jyvaskyla, Finland 123 123 546 F. Vermolen, I. Pölönen the basement membrane, and metastasises rarely. SCC develops from squamous cells, and metastasises more often. Finally, melanomas, which develop from melanocytes, form the most dangerous class of skin cancers. Melanomas metastasise if they grow deeper and penetrate through the basal membrane. All these cancers are able to metas- tasise to other parts of the body. Skin cancer is often caused by a long-term exposure of skin to UV-radiation, since UV-radiation increases the risk of skin cancer. People with lighter skin and who travel to the mountains, such as during skiing when the UV radiation is large at high altitudes, and further ampliﬁed by the reﬂection of snow, are subject to dangerous portions of radiation. It should also be noted that people, who expose their skins over a long period of time to strong doses of sunlight on the beach or in water, have a high risk of developing burns and skin cancer. According to Eriksson and Tinghög (2015), the total societal cost of skin cancer in Sweden over the year 2011 were estimated at 178 million Euros, with an increase of 27% with respect to 2005. The current paper deals with a three dimensional spatial Markov chain model for the simulation of skin cancer in general. 1 Introduction Each Uncertainty quantiﬁcation on a spatial Markov-chain model… 547 Voronoi cell represents a biological cell that grows, divides or dies. By far one of the earliest cellular automaton models for cancer was developed by Qi et al. (1993), who consider cell migration, cytotoxic behaviour of the immune system and mechanical pressure of the tumor. A different class of computational models is formed by the agent-based models. In these models, the time history of each cell is taken into account, and the position of the cells is, unlike in cellular automata models, continuous over time and only becomes discrete after application of a numerical method for the time-integration. Processes like cell migration, differentiation, proliferation and mutation of each cell are simulated using this class of models. Numerous examples from the literature exist for the development and application of agent-based models. Although the list is far from complete, we mention several studies in agent-based models that were applied to diseases in the (epi)dermis. Li et al. (2013) simulate epithelial renewal in skin over a three years period. Grabe and Neuber (2007) simulate the dynamics of psoriasis as a disease in the epidermis. Sun et al. (2009), Adra et al. (2010) develop a three- dimensional model for epidermal wound healing and structuring with the interplay of the TGF-beta growth factor. Thingnes et al. (2012) simulate the temporal evolution of the melanocyte distribution on the basis of (negative) chemotaxis due to a substance that is secreted by the keratinocytes. We also note our own work on agent-based models with, among others, an application to wound contraction as a result of serious burn trauma, see Boon et al. (2016). We list three mathematical studies devoted to melanoma. The ﬁrst study is due to Gallinaro et al. (2013) and treats a model for melanoma cell migration with an elastic continuum. The second study by Morais et al. (2017) treats a 2D probabilistic Widom- Rowlinson model that shares some features with the current model and with the model by Poleszczuk and Enderling (2013). The third study by Ciarletta et al. (2001) entails a partial differential equations (PDE) based approach for the radial progression of melanoma. We will elaborate more on these models in Sect. 5. The current paper proposes a spatial Markov chain model that is entirely stochastic. 1 Introduction Mathematical modelling studies that engage themselves with skin cancer in particular are very scarce, and therefore we list some relevant mathematical studies of general solid tumors, although we admit that the list is not complete. Signiﬁcant contributions to the simulation of metastasis of general cancer using Markov chain modelling and Monte Carlo simulations were realised by Newton et al. (2012, 2013, 2015). Elements from game theory were used in the form of the Prisoner’s Dilemma in West et al. (2016). Next to these probabilistic approaches, several studies on cellular automata models applied to cancer have been performed. In this context, we mention a few important studies. Poleszczuk and Enderling (2013) considered a cellular automata model with probabilities of cell division and cell migration. They consider a two-dimensional lattice with points that are either occupied by a cancer cell or not. If the cancer cell proliferates, then it does not migrate in their model. They use constant values for the likelihoods for migration and proliferation. As far as we know, in the literature, the study by Poleszczuk and Enderling (2013) is closest to the current study. Poleszczuk and Enderling (2013) give a clear description of how their method works. Further, Butler et al. (2014) formulated a cellular-automaton model for general cancer growth in two dimensions, where processes like cell death (apoptosis) of cancer cells, immu- nity death and angiogenesis are incorporated. Their approach for cancer cell migration is similar to the study by Poleszczuk and Enderling (2013). Monteagudo and Santos (2012) present a cellular automaton model that was used by Butler et al. (2014). The latter group applied the model to predict the impact of drugs on cancer growth. Jiao and Torquato (2011) described a cellular automaton model for emergent behaviours regardinginvasivetumor growthinheterogeneous tissues. Their model is morecompli- cated and contains interactions between several cell types, cell division, degradation of extracellular matrix (ECM), mutation rates and cell mobilities. Alarcon et al. (2003) also consider tumor growth in inhomogeneous environments, in which the oxygen concentration is computed, and where blood vessels are assumed to be immobile. Though the model is based on cellular automata principles, the nature of their model is entirely deterministic. Kansal et al. (2000) propose a cellular automata model for the simulation of brain tumors, where their domain is represented by Voronoi cells. 1 Introduction Summarised, this paper contains the following key innovations: Summarised, this paper contains the following key innovations: – development of the lattice-based Markov chain model, with a mathematically sound justiﬁcation of the inter-nodal transition probability based on the states of immediate neighbours; – phenomenological incorporation of the effects of long term exposure to UV- radiation; – phenomenological incorporation of the effects of long term exposure to UV- radiation; – evaluation of the likelihood that the cancer develops to a predeﬁned state within a certain time-interval. – evaluation of the likelihood that the cancer develops to a predeﬁned state within certain time-interval. The paper is organised as follows: in Sect. 2, we formulate the physical and biolog- ical hypotheses of the model. Furthermore, some mathematical justiﬁcation is given regarding the assumptions. Section 3 deals with the numerical approach of the Markov chain process and the uncertainty quantiﬁcation on the basis of Monte Carlo simu- lations. Subsequently, we present the results in Sect. 4. Sections 5 and 6 end with a discussion and conclusions regarding the modelling framework and results. 1 Introduction The model divides the domain of computation into nodal points that have two possible states: either ‘cancer state’ or ‘not in cancer state’. The transition probability depends on the status of the neighbours in terms of distance and number of neighbours that are in ‘cancer state’. Next to this intrinsic likelihood, we propose an update of the transition probability as a function of the intensity of UV-radiation as well as the transmission of UV into the skin. As an innovation compared to the previously formulated cellular and Markov chain models for cancer development, the current model is simple in the sense that it consists of three input parameters only, and that the transmission probability of a nodal point depends on the number of neighbours that are in ‘cancer state’ and on the distance between the nodal point in consideration and neighbours. The relation is also motivatedmathematically.Furthermore,nexttopredictingtherateatwhichcancerpro- gresses, the calculations suffer from uncertainty from multiple sources. The ﬁrst source of uncertainty is the probabilistic nature of the model itself, and the second source of uncertainty comes from poor knowledge regarding the values of the input parameters. These uncertainties make that each simulation only represents one of the very many possible scenarios. Since any of these possible scenarios could describe the progres- sion of skin cancer in a patient, we are less interested in predicting the rate at which 123 548 F. Vermolen, I. Pölönen cancer grows than in predicting the likelihood that cancer reaches a certain size within a predeﬁned timeframe. Note that the development of cancer is not the same as the initiation of cancer. The current modelling framework is used to simulate the progres- sion of cancer that already initiated earlier. Hence when we speak of the development of cancer, we mean the growth of earlier initiated cancer. The issue of uncertainty regarding the growth of cancer is dealt with in the current paper using Monte Carlo simulations. This probabilistic approach is innovative in this class of modelling. 2 The mathematical model First we describe the model and motivate the choices that we make mathematically. We consider a domain D ⊂R3 that is divided into a set of N nodal points N = {1, . . . , n}, in which each node i has position xi = (xi, yi, zi). All positions are ﬁxed in time t. From a biological perspective, these nodes may correspond to single cells, or to a cluster of cells, or even a spatial part of the body. For each of these nodes, we consider its set of neighbours, indicated by Ni, which is mathematically deﬁned by Ni = { j ∈N \| j is a nearest neighbour node of node i}. Ni = { j ∈N \| j is a nearest neighbour node of node i}. Only nearest neighbours are incorporated to prevent huge computational burdens. Each node can be in two states: either in ‘cancer state’ or in ‘non-cancer state’. This is indicated by the binary switch parameter Si: Si = 0, node i is in ‘non-cancer state’; 1, node i is in ‘cancer-state’. (1) (1) This implies that if Si = 0 for all i ∈N then the tissue is absolutely free of cancer, whereas if Si = 1 for all i ∈N then the tissue is completely cancerous. All other situations are intermediate states. Next, we assume that each node, say node i, that is in ‘non-cancer state’ has a likelihood to change state to ‘cancer state’. This likelihood This implies that if Si = 0 for all i ∈N then the tissue is absolutely free of cancer, whereas if Si = 1 for all i ∈N then the tissue is completely cancerous. All other situations are intermediate states. Next, we assume that each node, say node i, that is in ‘non-cancer state’ has a likelihood to change state to ‘cancer state’. This likelihood 123 123 Uncertainty quantiﬁcation on a spatial Markov-chain model… 54 Uncertainty quantiﬁcation on a spatial Markov-chain model… 549 satisﬁes an (memory-less) exponential probability distribution with probability rate λi, that is, we have the following probability distribution satisﬁes an (memory-less) exponential probability distribution with probability rate λi, that is, we have the following probability distribution f (s, λi) = λi exp(−λi(s)(s −t)), for s ≥t. (2) (2) (2) Hence over a time interval of length τ, we have the following likelihood for transition from state {Si = 0} to state {Si = 1}: P(Si(t + τ) = 1\|Si(t) = 0)) = t+τ t λi(s) exp(−λi(s)(s −t))ds ≈1 −exp(−λi(t)τ) = λiτ + O(λiτ)2. (3) (3) Irreversibility of the transition implies P(Si(t + τ) = 0\|Si(t) = 1)) = 0. The total likelihoods are given by Irreversibility of the transition implies P(Si(t + τ) = 0\|Si(t) = 1)) = 0. Ni = { j ∈N \| j is a nearest neighbour node of node i}. The total likelihoods are given by P(Si(t + τ) = 0) = P(Si(t + τ) = 0\|Si(t) = 0)P(Si(t) = 0) +P(Si(t + τ) = 0\|Si(t) = 1)P(Si(t) = 1), P(Si(t + τ) = 1) = P(Si(t + τ) = 1\|Si(t) = 0)P(Si(t) = 0) +P(Si(t + τ) = 1\|Si(t) = 1)P(Si(t) = 1), (4) and hence we have the following probability matrix over time-interval τ: Pτ = ⎛ ⎝ P(Si(t + τ) = 0\|Si(t) = 0) P(Si(t + τ) = 0\|Si(t) = 1) P(Si(t + τ) = 1\|Si(t) = 0) P(Si(t + τ) = 1\|Si(t) = 1) ⎞ ⎠ = ⎛ ⎝ 1 − t+τ t λi(s) exp(−λi(s)(s −t))ds 0 t+τ t λi(s) exp(−λi(s)(s −t))ds 1 ⎞ ⎠≈ ⎛ ⎝ 1 −λ(t)τ 0 λ(t)τ 1 ⎞ ⎠. (5) (5) This matrix can be used to simulate the dynamics of the fraction of cancerous regions over time, by recurrently computing over time points τ, . . . , kτ y(k+1)τ = Pτykτ, (6) (6) where the ﬁrst and second component of ykτ represents the average fraction of nodes that are in ‘non cancer state’ and ‘cancer state’ respectively. Solving the average fraction in ‘cancer state’, that is the second component of ykτ, which we call p(kτ), from Eq. (6), gives p(kτ) = 1 − k j=1 1 − kτ (k−1)τ exp(−λ(s)(s −(k −1)τ)ds ≈1 − k j=1 (1 −λ(( j −1)τ)τ) . (7 − k j=1 (1 −λ(( j −1)τ)τ) . (7) (7) 123 550 F. Vermolen, I. Pölönen The fraction p(kτ) can also be interpreted as the likelihood that a node is in ‘cancer state’. From Eq. (7), we get The fraction p(kτ) can also be interpreted as the likelihood that a node is in ‘cancer state’. From Eq. (7), we get p(kτ) −p((k −1)τ) = 1 − k j=1 (1 −λ(( j −1)τ)τ) − ⎡ ⎣1 − k−1 j=1 (1 −λ(( j −1)τ)τ) ⎤ ⎦ = k−1 j=1 (1 −λ(( j −1)τ)τ) − k j=1 (1 −λ(( j −1)τ)τ) = k−1 j=1 (1 −λ(( j −1)τ)τ) − k j=1 (1 −λ(( j −1)τ)τ) = k−1 j=1 (1 −λ(( j −1)τ)τ) (1 −(1 −λ((k −1)τ)τ)) = (1 −p((k −1)τ))λ((k −1)τ)τ. Ni = { j ∈N \| j is a nearest neighbour node of node i}. (8 = k−1 j=1 (1 −λ(( j −1)τ)τ) − k j=1 (1 −λ(( j −1)τ)τ) = k−1 (1 −λ(( j −1)τ)τ) (1 −(1 −λ((k −1)τ)τ)) = k−1 j=1 (1 −λ(( j −1)τ)τ) − k j=1 (1 −λ(( j −1)τ)τ) = k−1 j=1 (1 −λ(( j −1)τ)τ) (1 −(1 −λ((k −1)τ)τ)) = (1 −p((k −1)τ))λ((k −1)τ)τ. (8) = j=1 (1 −λ(( j −1)τ)τ) (1 −(1 −λ((k −1)τ)τ)) (8) (8) Dividing the above equation by τ, gives p(kτ) −p((k −1)τ) τ = (1 −p((k −1)τ))λ((k −1)τ). (9) (9) Taking the limit τ →0 (and sending k →∞), upon (k −1)τ = t, we get Taking the limit τ →0 (and sending k →∞), upon (k −1)τ = t, we get Taking the limit τ →0 (and sending k →∞), upon (k −1)τ = t, we get p′(t) = (1 −p(t))λ(t). (10) (10) With p(0) = 0, this gives With p(0) = 0, this gives With p(0) = 0, this gives p(t) = 1 −exp − t 0 λ(s)ds . (11) (11) Proposition 2.1 Let Pτ = 1 − t+τ t λ(s) exp(−λ(s)(s −t))ds 0 t+τ t λ(s) exp(−λ(s)(s −t))ds 1 , and let r(t) = 1 −p(t), then r′(t) p′(t) = −λ(t) 0 λ(t) 0 r(t) p(t) , (12) osition 2.1 Let Pτ = 1 − t+τ t λ(s) exp(−λ(s)(s −t))ds 0 t+τ t λ(s) exp(−λ(s)(s −t))ds 1 , and let r(t) = (t) th 1 −p(t), then −p(t), then r′(t) p′(t) = −λ(t) 0 λ(t) 0 r(t) p(t) , (12) (12) r(t) = exp(− t 0 λ(s)ds), p(t) = 1 −exp(− t 0 λ(s)ds). (13) (13) Proof From the hypothesis, it follows that Proof From the hypothesis, it follows that Proof From the hypothesis, it follows that r(t + τ) = (1 − t+τ t λ(s) exp(−λ(s)(s −t))ds)r(t), r(t + τ) = (1 − t+τ t λ(s) exp(−λ(s)(s −t))ds)r(t), p(t + τ) = t+τ t λ(s) exp(−λ(s)(s −t))dsr(t) + p(t). Ni = { j ∈N \| j is a nearest neighbour node of node i}. (14) (14) 123 123 ation on a spatial Markov-chain model… 551 Uncertainty quantiﬁcation on a spatial Markov-chain model… Uncertainty quantiﬁcation on a spatial Markov-chain model… 551 The above equations can be rewritten as The above equations can be rewritten as The above equations can be rewritten as r(t + τ) −r(r) = −r(t) t+τ t λ(s) exp(−λ(s)(s −t))ds, p(t + τ) −p(t) = r(t) t+τ t λ(s) exp(−λ(s)(s −t))ds = (1 −p(t)) t+τ t λ(s) exp(−λ(s)(s −t))ds. (15) (15) From this, we get, from division by τ, taking the limit τ →0, and using de l”Hospital’s Rule From this, we get, from division by τ, taking the limit τ →0, and using de l”Hospital’s Rule r′(t) = lim τ→0 r(t + τ) −r(t) τ = −lim τ→0 r(t) t+τ t λ(s) exp(−λ(s)(s −t))ds τ = −r(t)λ(t), r′(t) = lim τ→0 r(t + τ) −r(t) τ = −lim τ→0 r(t) t+τ t λ(s) exp(−λ(s)(s −t))ds τ = −r(t)λ(t), p′(t) = lim τ→0 p(t + τ) −p(t) τ = lim τ→0(1 −p(t)) t+τ t λ(s) exp(−λ(s)(s −t))ds τ = (1 −p(t)) lim τ→0 t+τ t λ(s) exp(−λ(s)(s −t))ds τ = (1 −p(t))λ(t). (16) r(t)λ(t), p′(t) = lim τ→0 p(t + τ) −p(t) τ = lim τ→0(1 −p(t)) t+τ t λ(s) exp(−λ(s)(s −t))ds τ = (1 −p(t)) lim t+τ t λ(s) exp(−λ(s)(s −t))ds = (1 −p(t))λ(t) p′(t) = lim τ→0 p(t + τ) −p(t) τ = lim τ→0(1 −p(t)) t+τ t λ(s) exp(−λ(s)(s −t))ds τ t+τ lim →0 p(t + τ) −p(t) τ = lim τ→0(1 −p(t)) t+τ t λ(s) exp(−λ(s)(s −t))ds τ = (1 −p(t)) lim τ→0 t+τ t λ(s) exp(−λ(s)(s −t))ds τ = (1 −p(t))λ(t). (16) From this, we get r′(t) = −λ(t)r(t), p′(t) = λ(t)(1 −p(t)) = λ(t)r(t). (17) (17) From this, we get From this, we get r(t) = exp(− t 0 λ(s)ds), p(t) = 1 −exp(− t 0 λ(s)ds). This proves the assertion. ⊓⊔ This proves the assertion. ⊓⊔ Since the expected number of nodes in cancer satisﬁes a Markov chain process, see Eq. (6) and the above relation, the convergence to the steady-state of full cancer takes place asymptotically as t →∞(for λ > 0). Ni = { j ∈N \| j is a nearest neighbour node of node i}. In other words if p(t) denotes the probability that a node is in ‘cancer state’ at time t, then limt→∞p(t) = 1, whereas 0 < p(t) < 1 for t > 0. The time to reach the state where all the nodes are in ‘cancer state’ increases with the number of nodes due to the asymptotic convergence. For this reason, we only monitor the minimal time at which half of the nodes are in cancer state, that is f = 1/2, in the Monte Carlo simulations to avoid unnecessarily large computing times. It is easy to see that the dynamic system will converge to the state Si = 1 for all i ∈N as t →∞(k →∞) if λi(t) > 0 for t > 0 for all i ∈N. 123 552 F. Vermolen, I. Pölönen It is well known that cancer progresses by mutation of healthy cells, proliferation of cancer cells and by migration of cancer cells. Cell migration proceeds by several mechanisms, such as taxis processes (migration towards stiffer tissue regions (duro- taxis), chemotaxis (migration towards higher concentration of oxygen or nutrients)) or as a result of random walk, see for instance Lo et al. (2000). This is the reason why regions that are affected by cancer, that is, they are in ‘cancer state’, contaminate healthy regions. For simplicity, we consider two nodes on the x-axis, with coordinates xL = 0 and xR = 1. Let xL be in ‘cancer state’ while xR is not in ‘cancer state’. Let λ be the probability rate, from the exponential distribution, for xR to change to the ‘cancer state’ due to xL being in ‘cancer state’. Then the expected length of the time-interval to change to the ‘cancer state’ for node xR is given by E(t) = ˆt = 1 λ. Suppose now that between node xL and xR, we have the following rearrangement of nodal points: x j = xL + jh, xn = xR and nh = 1, where all these nodes, except x0 = xL are not in the ‘cancer state’. Ni = { j ∈N \| j is a nearest neighbour node of node i}. Let λ(h) be the probability rate parameter to change node x j to the ‘cancer state’, given that one of its neighbours at a distance h was in ‘cancer state’, then the expected length of the time interval to change to the ‘cancer state’ is given by Suppose now that between node xL and xR, we have the following rearrangement of nodal points: x j = xL + jh, xn = xR and nh = 1, where all these nodes, except x0 = xL are not in the ‘cancer state’. Let λ(h) be the probability rate parameter to change node x j to the ‘cancer state’, given that one of its neighbours at a distance h was in ‘cancer state’, then the expected length of the time interval to change to the ‘cancer state’ is given by ˆt j = 1 λ(h) , in node j. Note that this considers the expected length of time interval to transform a node at a distance h from a node that was in ‘cancer state’. The total expected value for the length of the time interval to change all nodes into ‘cancer state’ should be equal to the expected time to change node xR to the ‘cancer state’ and hence we have in node j. Note that this considers the expected length of time interval to transform a node at a distance h from a node that was in ‘cancer state’. The total expected value for the length of the time interval to change all nodes into ‘cancer state’ should be equal to the expected time to change node xR to the ‘cancer state’ and hence we have n j=1 ˆt j = n λ(h) = ˆt = 1 λ. (18) Herewith, we get λ(h) = nλ = λ h , (19) n j=1 ˆt j = n λ(h) = ˆt = 1 λ. (18) (18) Herewith, we get λ(h) = nλ = λ h , (19) (19) which gives the probability rate between two nodes of complementary states separated by a distance h. Ni = { j ∈N \| j is a nearest neighbour node of node i}. From the above relation, it can be observed that if h →0, then λ(h) tends to inﬁnity, which implies that the expected time for the transition is given by lim h→0 E(t) = lim h→0 ˆt j = 0, lim h→0 E(t) = lim h→0 ˆt j = 0, which means that the transition occurs immediately almost surely as h →0. We summarise the overall conclusion in Proposition 2.2. which means that the transition occurs immediately almost surely as h →0. We summarise the overall conclusion in Proposition 2.2. 123 12 123 Uncertainty quantiﬁcation on a spatial Markov-chain model… 553 Proposition 2.2 Consider xL and xR and let \|\|xL −xR\|\| = 1. Suppose that at time t one of the two points is in ‘cancer state’, whereas the other point is not in ‘cancer state’. Let λ be the probability rate from the exponential distribution to transfer the nodal point that is not in ‘cancer state’ to the ‘cancer state’, then for two points xA and xB, separated by distance \|\|xA −xB\|\| = h, where exactly one of the two points is in ‘cancer state’, the probability rate to transfer the other point to the ‘cancer state’ is given by λ(h) = λ h . This probability rate is consistent in expected value of the length of the transformation time interval. In a more-dimensional setting, a node that is not in ‘cancer state’ could have multiple neighbours that are in ‘cancer state’. Consider node j that is not in ‘cancer state’. Suppose that k of its nearest neighbours are in ‘cancer state’. For each node in ‘cancer state’, its neighbour that is not in ‘cancer state’, say node j, has a likelihood of p = t+τ t λ(h) exp(−λ(h)(s −t))ds = 1 −e−λ(h)τ, (20) (20) to change to the ‘cancer state’ during time interval τ. Hence the likelihood that this neighbour will not change its state is equal to P(Sj(t + τ) = 0\|Sj(t) = 0) = 1 −p. Consider node j having k neighbours that are in ‘cancer state’ at time t, then P(Sj(t + τ) = 0\|Sj(t) = 0) = (1 −p)k. This implies that P(Sj(t + τ) = 1\|Sj(t) = 0) = 1 −(1 −p)k. As long as p ≪1, we have P(Sj(t + τ) = 1\|Sj(t) = 0) ≈kp. Using the exponential distribution, see Eq. Ni = { j ∈N \| j is a nearest neighbour node of node i}. 20, we arrive at P(Sj(t + τ) = 1\|Sj(t) = 0) = 1 −(1 −p)k = 1 −e−kλ(h)τ = 1 −exp −k h λτ . Therefore, next to the distance of the node to the neighbours that are in ‘cancer state’, the probability rate λi = k h λ depends on the number of neighbouring nodes that are in ‘cancer state’. In order to be able to deal with various spatial rearrangements and grids, we generalise the formalism to having neighbouring nodes that have different separations in Proposition 2.3: Therefore, next to the distance of the node to the neighbours that are in ‘cancer state’, the probability rate λi = k h λ depends on the number of neighbouring nodes that are in ‘cancer state’. In order to be able to deal with various spatial rearrangements and grids, we generalise the formalism to having neighbouring nodes that have different separations in Proposition 2.3: 123 554 F. Vermolen, I. Pölönen Proposition 2.3 Consider nodal point xk, which is not in ‘cancer state’ at time t, with nearest neighbouring nodal points Nk, that are, respectively, separated from xk by distances d(xk, x j) = \|\|xk −x j\|\|, j ∈Nk. Let λ be the probability rate in the exponential distribution to change a nodal point to the ‘cancer state’ due to a nearest neighbour at unit distance. Then λk = λ j∈Nk Sj(t) d(xk, x j), (21) (21) where λk represents the probability rate in the exponential distribution for the transi- tion of nodal point xk. where λk represents the probability rate in the exponential distribution for the transi- tion of nodal point xk. Proof Consider nodal point xk, which is not in ‘cancer state’, and only consider its neighbours {xkq} that are in ‘cancer state’, hence q ∈N can k := {q ∈Nk : Sq(t) = 1}. Let pqk be the probability that nodal point q contaminates nodal point k to the ‘cancer state’ during time interval τ. Then from Proposition 2.2, we have pqk = 1 −exp − λτ d(xk, xq) . (22) (22) The probability that nodal point xk remains in ‘non cancer state’ is given by P(Sk(t + τ) = 0\|Sk(t) = 0) = q∈N can k (1 −pqk), (23) (23) where we account for the fact that each neighbour xkq, q ∈N can k may send xk to the ‘cancer state’. Ni = { j ∈N \| j is a nearest neighbour node of node i}. the life burden intensity of UV light that is experienced by node i. 3. the life burden intensity of UV light that is experienced by node i. Ni = { j ∈N \| j is a nearest neighbour node of node i}. Using the above equation, Proposition 2.2, and Eq. (20), we get where we account for the fact that each neighbour xkq, q ∈N can k may send xk to the ‘cancer state’. Using the above equation, Proposition 2.2, and Eq. (20), we get P(Sk(t + τ) = 1\|Sk(t) = 0) = 1 −P(Sk(t + τ) = 0\|Sk(t) = 0) = 1 − q∈N can k exp − λτ d(xk, xq) = 1 −exp ⎛ ⎝−λτ q∈N can k 1 d(xk, xq) ⎞ ⎠. (24) (24) The deﬁnition of N can k and Sq(t), then gives The deﬁnition of N can k and Sq(t), then gives P(Sk(t + τ) = 1\|Sk(t) = 0) = 1 −exp ⎛ ⎝−λτ j∈Nk Sj(t) d(xk, x j) ⎞ ⎠. (25) (25) Hence Hence Hence λk = λ j∈Nk Sj(t) d(xk, x j). This proves the assertion. ⊓⊔ This proves the assertion. This proves the assertion. ⊓⊔ 123 Uncertainty quantiﬁcation on a spatial Markov-chain model… Uncertainty quantiﬁcation on a spatial Markov-chain model… 555 The above result in Proposition 2.3 does not yet take into account any UV radiation. To model the effect from exposure to UV radiation, we extend the above equation with a term S The above result in Proposition 2.3 does not yet take into account any UV radiation. To model the effect from exposure to UV radiation, we extend the above equation with a term λi = λ j∈Ni Sj d(xi, x j) + αRi, (26) (26) where Ri denotes the life time burden UV intensity that is experienced by node i, and the α–parameter accounts for the sensitivity of cell i with respect to UV-radiation. Hence summarised, the model is based on the hypothesis that cancer progresses to node i, as a result of the following three factors: where Ri denotes the life time burden UV intensity that is experienced by node i, and the α–parameter accounts for the sensitivity of cell i with respect to UV-radiation. Hence summarised, the model is based on the hypothesis that cancer progresses to node i, as a result of the following three factors: 1. the number of nearest neighbours of node i in the ‘cancer state’; 2. the distance between node i and nearest neighbours that are in the ‘cancer state’; 2. the distance between node i and nearest neighbours that are in the ‘cancer state 3. 3 Numerical implementation First we explain how we apply the model in the ﬁnal calculations where we simulate with the full model. For all nodes i ∈N, we initially impose Si = 0. Then at each time-step, the light intensity Ri is computed on all the nodes. Subsequently, a sample from the exponential distribution with probability rate λi for each node is drawn over time-interval τ. Based on the resulting samples, the state of each node is changed to the ‘cancer state’ or not. Subsequently the time is incremented by the time step τ. This procedure is repeated with the exception that at the next time steps, we also check the state of the nearest neighbours. This whole procedure is repeated until half of the nodes are in ‘cancer state’. The fraction of nodes in ‘cancer state’, for a total of n nodes, is computed via f = pn(kτ) = 1 n · i∈N Si(kτ), (27) (27) at time kτ. The time T f , denoting the minimal time at which a certain predeﬁned fraction of the nodes is in ‘cancer state’, f , is stored. Inordertotestthealgorithm,thetransitionprobabilityforeachnodewastakenequal to λ, regardless the state of the neighbours. Then the solution should approximate the expression from Eq. (7), that is pn(kτ) →p(kτ) for k ∈N as n →∞. We calculate the root mean square error (RMSn) with Eq. (7), given by RMSn = nt k=1(pn(kτ) −p(kτ))2 nt , (28) (28) where nt is the time index at which all the nodes have transformed to the ‘cancer state’, that is Sj(ntτ) = 1 for all j ∈N. For parametric choices λ = 0.01, and a cubic domain of 1 × 1 × 1, and n = 8 × 8 × 8 nodes, we show the results in Fig. 1. From this ﬁgure, the exponential growth character is obvious and further pn(kτ) and p(kτ) only differ by a root mean square error of 0.01209. Note that Fig. 1 has only 123 556 F. Vermolen, I. Pölönen Fig. 1 The fraction of nodes in ‘cancer state’ as a function of time. The green curve represents the deter- ministic result from Eq. (7), whereas the blue curve represents one sample from the stochastic model result using 8 × 8 × 8 nodes with λ = 0.01 (color ﬁgure online) Fig. 1 The fraction of nodes in ‘cancer state’ as a function of time. 3 Numerical implementation The green curve represents the deter- ministic result from Eq. (7), whereas the blue curve represents one sample from the stochastic model result using 8 × 8 × 8 nodes with λ = 0.01 (color ﬁgure online) Table 1 Root mean squared errors for one sample of various numbers of nodes Number of nodes Root mean squared error 8 × 8 × 8 0.1209 × 10−1 16 × 16 × 16 0.5086 × 10−2 32 × 32 × 32 0.1774 × 10−2 64 × 64 × 64 0.1007 × 10−2 been added to illustrate graphically that the macro-Markov chain model and the lattice model are consistent. We listed the results for the root mean squared error for one sample on various numbers of nodes in Table 1. From these results, it is concluded that if n = 32 × 32 × 32 the sampling error is sufﬁciently small. In the ﬁnal computations, the Monte Carlo method is carried out by repeating the simulations 1000 times (N = 1000). The average minimal time at which a fraction f of the nodes are in ‘cancer state’ is computed by the arithmetic mean of all the sample T (i) f , hence by ˆT N f = 1 N N j=1 T ( j) f , (29) (29) where T ( j) f denotes the jth Monte Carlo trial of the minimal time at which a fraction of f of the nodes is in ‘cancer state’. These simulations are done for the full model. The above sample mean is an unbiased estimator for the expected value of the minimal time, that is E( ˆT N f ) = μ f . Here μ f = E(T f ) denotes the expected value of the where T ( j) f denotes the jth Monte Carlo trial of the minimal time at which a fraction of f of the nodes is in ‘cancer state’. These simulations are done for the full model. The above sample mean is an unbiased estimator for the expected value of the minimal time, that is E( ˆT N f ) = μ f . Here μ f = E(T f ) denotes the expected value of the 123 12 Uncertainty quantiﬁcation on a spatial Markov-chain model… 5 Uncertainty quantiﬁcation on a spatial Markov-chain model… 557 minimal time at which a fraction f of the nodes is in ‘cancer state’. 3 Numerical implementation The Monte Carlo Error (MCE) based on N trials (samples) is deﬁned by minimal time at which a fraction f of the nodes is in ‘cancer state’. The Monte Carlo Error (MCE) based on N trials (samples) is deﬁned by MCEN := Var[ ˆT N f ], (30) (30) where [ ˆT N f ] consists of m batches of N samples. The above equation requires the use of multiple batches containing a lot of simulations or the division of a Monte Carlo sequence into multiple batches. In the remainder of this section, we derive a statistically sound and practical representation of the Monte Carlo Error that can be determined using one batch of simulations only. The derivation tightly follows the footprints outlined by Koehler et al. (2009). From the Strong Law of Numbers, we have ˆT N f →μ f , as N →∞. The Central Limit Theorem implies that the above sample mean converges in distri- bution to a normal distribution via The Central Limit Theorem implies that the above sample mean converges in distri- bution to a normal distribution via √ N( ˆT N f −μ f ) −→d N(0, σ 2 T ), as N →∞, (31) (31) where σ 2 T = E((T f −μ f )2) represents the standard deviation of T f . The above relation implies Var[ √ N( ˆT N f −μ f )] →σ 2 T , as N →∞. (32) where σ 2 T = E((T f −μ f )2) represents the standard deviation of T f . The above relation implies Var[ √ N( ˆT N f −μ f )] →σ 2 T , as N →∞. (32) (32) Since Var(μ f ) = 0, the above equation implies Var[ √ N ˆT N f ] = NVar[ ˆT N f ] →σ 2 T , as N →∞. Hence we have Var[ ˆT N f ] →σ 2 T N = E((T f −μ f )2) N , (33) (33) as N →∞. The sample variance is given by as N →∞. The sample variance is given by as N →∞. The sample variance is given by as N →∞. The sample variance is given by s2 N = 1 N −1 N j=1 (T ( j) f −ˆT N f )2. 4 Computer simulations All simulations were done in an implementation in Python 2.7.5 on a MacBook Air with a 1.4 GHz Intel Core i5 processor and 8 GB 1600 MHz DDR3 memory. The spatial domain represents a generic area of skin. The top of the domain represents the surface of the epidermis and in which the bottom represents the basement mem- brane, which is a physical barrier that inhibits further progression of cancer. Further progression of cancer over the basement membrane proceeds by different biological mechanisms, such as the slower process of epithelial-mesenchymal transformation, than the ones that are considered in the current manuscript. First, we present various results of one simulation of how cancer progresses through the domain in the course of time. Subsequently, we show results from Monte Carlo simulations in two spatial dimensions, where we are interested in the length of the time interval that is needed to have 50% of the domain occupied by cancer. We will show the correlations of this length of time interval with the three input parameters. 3 Numerical implementation s2 N = 1 N −1 N j=1 (T ( j) f −ˆT N f )2. Since the sample variance is an unbiased estimator for σ 2 T (that is E(s2 N) = σ 2 N), we take σ 2 T ≈s2 N, then the right-hand side of Eq. (33) is estimated by Since the sample variance is an unbiased estimator for σ 2 T (that is E(s2 N) = σ 2 N), we take σ 2 T ≈s2 N, then the right-hand side of Eq. (33) is estimated by Var[ ˆT N f ] ≈s2 N N , (34) (34) 123 123 558 F. Vermolen, I. Pölönen Hence the MCE is estimated by Hence the MCE is estimated by Hence the MCE is estimated by ated by Hence the MCE is estimated by MCEN = sN √ N = N j=1(T ( j) f −ˆT N f )2 N(N −1) . (35) (35) The above equation is used for the estimation of the Monte Carlo Error. Note that the sample variance is an unbiased estimator for the variance σ 2 T (of T f ), then as a result of Jensen’s Inequality, the sample standard deviation, sN, is a biased, but consistent estimator of σT (that is limN→∞sN = σT ). Hence the above estimator for the MCE is biased, though consistent. Hence for large N, the above equation gives a good approximation for the MCE. Therefore, we use the above equation for the estimation of the MCE. 4.1 Time-evolution of cancer in 3D First we show model predictions of skin cancer progression over time. In the plots that we show, we only display the nodes that are in ‘cancer state’. These nodes are indicated by red dots. The other nodes, which are not in ‘cancer state’, are not displayed for the sake of visibility in the three dimensional conﬁguration. In the coming results, we present three different cases in three spatial dimensions in the coming subsections. We use 32 × 32 × 32 nodes in the closed cubic domain (x, y, z) ∈[0, 1]3 and λ = 0.01. Further, we kept αi = 0 for all i ∈N, which implies that cancer only progresses through nearest–neighbour interaction. 123 Uncertainty quantiﬁcation on a spatial Markov-chain model… 559 4.1.1 Progression of melanomas Melanomasgenerallyprogressfrommutatedmelanocytes,whicharelocatedonornear the basal membrane separating the epidermis from the dermis. Although these cancers are often enhanced by excessive exposure of the skin to UV-radiation (generally type B, which progresses deeper into the dermis), melanomas often develop as a result of genetics, or from moles. Therefore, we consider the case that one cell in the center of the domain ((x, y, z) = (0.5, 0.5, 0.5)), away from the skin surface, is in ‘cancer state’, which could represent a melanoma that spreads towards the top and the bottom of the skin surface. Several snapshots have been shown in Fig. 2, at the early and later stages. From the ﬁgures, it is observed how the cancer develops from the centre and spreads into locations further away. In Fig. 2, it is seen that at the later stages the boundary of the computational domain starts to inﬂuence the results. It can be seen how the model gives a possible representation of the geometrical development of the early stage cancer except at the ﬁnal stages, where boundary effects start to dominate, see Fig. 2i. In Fig. 3, we show the fraction of nodes that has transformed into the cancer state as a function of time for ﬁve different samples using identical input parameters and initially the central node at (0.5, 0.5, 0.5) was assumed to be in cancer state. The Fig. 2 Snapshots of early stage progression of cancer from the middle of the domain of computation. Snapshots were taken at 1, 50, 100, 200, 400, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are 0, 2.35, 3.1, 4.15, 5.3, 7.1, 11.3, 15.85, 22.3 dimensionless time units Fig. 2 Snapshots of early stage progression of cancer from the middle of the domain of computation. Snapshots were taken at 1, 50, 100, 200, 400, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are 0, 2.35, 3.1, 4.15, 5.3, 7.1, 11.3, 15.85, 22.3 dimensionless time units F. Vermolen, I. Pölönen 560 Fig. 3 Left: ﬁve sample runs of the fraction of the nodal points that transformed into the cancer state versus time using 32 × 32 × 32 = 32768 nodes; Right: a ﬁtted result to the experiments done in Demidem et al. (2001). 4.1.1 Progression of melanomas The thick solid blue curve right has been obtained by the use of the current spatial Markov chain model; the error bars denote the measured results by Demidem et al. (2001). The black curves are ﬁtted results to the Gompertz model (color ﬁgure online) Fig. 3 Left: ﬁve sample runs of the fraction of the nodal points that transformed into the cancer state versus time using 32 × 32 × 32 = 32768 nodes; Right: a ﬁtted result to the experiments done in Demidem et al. (2001). The thick solid blue curve right has been obtained by the use of the current spatial Markov chain model; the error bars denote the measured results by Demidem et al. (2001). The black curves are ﬁtted results to the Gompertz model (color ﬁgure online) curves, which do not coincide entirely as a result of the stochastic nature of the model, all exhibit the so–called S-shape in accordance with the classical Gompertz model. Although the experimental outcomes by Demidem et al. (2001) were obtained for mice, we managed to ﬁt our model to their results. Their experiments measured the mass of the tumor. Assuming a mass density of water, which is given by ρ = 1000 kg/m3, to represent the density of the tumor, implies that 5.3 gram of tumor, which is their ﬁnal tumor mass, corresponds to a tumor volume of 5.3 × 10−6 m3. This implies that we bound the tumor in a cubic box with edge length of 5.31/3 × 10−2 m. Using a bisection procedure to ﬁt our modelling results to the Gompertz curve in Demidem et al. (2001) indicated that λ ≈1.5 × 10−4 m−1 day−1. The result is shown in Fig. 3, where the fraction that we computed can be converted to actual mass by setting mT = ρ · f · VR, where f , mT and VR, respectively, denote the volume fraction of the tumor, tumor mass and total end volume. Comparison with Fig. 1 in Demidem et al. (2001) shows a good agreement between their results and our results. Note that we ﬁtted our model to the data with the highest values of the tumor weight in the study by Demidem et al. (2001). This sequence of data corresponds to non-drug treated melanoma. The thick solid blue curve in Fig. 3b has been obtained by the use of our spatial Markov chain model. 4.1.1 Progression of melanomas The error bars represent the experimental results by Demidem et al. (2001), and the thin solid black curves represent the ﬁts by Demidem et al. (2001) using the Gompertz model. This agreement demonstrates the predictive value of our model. 4.1.2 Progression from the surface: squamous basal-cell carcinomas In the modelling, we approximate the development from the upper layers by the pro- gression from the top. Squamous basal-cell carcinomas develop from the top, but not from the very top (since the squamous cells are dead at the very top and the surface is covered by keratin) of the skin surface (the top of the epidermis), and these cancers are known to develop as a result of a life time excessive exposure to UV-radiation. 12 ation on a spatial Markov-chain model… 561 Uncertainty quantiﬁcation on a spatial Markov-chain model… Uncertainty quantiﬁcation on a spatial Markov-chain model… 561 In this section, we take the enhancement of the transition from ‘non cancer state’ to ‘cancer state’ as a result of UV-radiation into account in a phenomenological way. The likelihood of transition is given by In this section, we take the enhancement of the transition from ‘non cancer state’ to ‘cancer state’ as a result of UV-radiation into account in a phenomenological way. The likelihood of transition is given by λi = ⎧ ⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎩ λ j∈Ni Sj d(xi, x j) + α, zi = 1, λ j∈Ni Sj d(xi, x j), zi < 1. (36) (36) Here α takes into account the life time burden of exposure to UV-radiation, which only acts on the nodal points that are located on the top surface of the skin. In this section, we took λ = 0.01 and α = 0.1. A small α value represents a relatively small life time burden of exposure to UV-radiation, whereas large values represent large portions of life time exposure to UV radiation. The life time burden of radiation is determined by the fraction of time the skin of a patient is exposed to sunlight and by the intensity of the UV radiation exposed to. Hence it makes a difference whether the skin is exposed to one hour of evening sun (low intensity) or one hour to the sun at the mid of the day (that is at noon, high intensity). Initially, all nodes are in the ‘non cancer’ state. The transition to the ‘cancer state’ is caused by the exposure to UV-radiation and in the current simulations, we assume that only the top cells are exposed to UV. Penetration of UV-radiation into deeper layers is neglected in the current run. The results are shown in Fig. 4.1.2 Progression from the surface: squamous basal-cell carcinomas 4, where it can be seen that cancer progresses from the top layer towards the bottom. At the initial stage, only nodes on the top can change state to ‘cancer state’. As time proceeds, the nearest neighbour interaction makes the cancer move towards the bottom. In practice, the use of sun cream could result into a lower value of α, which would give a lower initiation rate (a larger incubation time) of cancer. 4.1.3 Progression and gradually decreasing initiation with the depth of the domain Basal-cell carcinomas (BCC) generally develop from basal cells, which are predomi- nantly located in the basal membrane. Melanomas develop from melanocytes, which are also predominantly located in the basal membrane. In order to model generic forms of skin cancer, where we incorporate the attenuation and gradual decay of the life long burden of UV-radiation into the skin, we use the following phenomenological law for UV transmission R(z) = e−βˆz, (37) (37) where the top layer of skin is at the coordinate ˆz = 0, hence we have ˆz = 1−z related to the physical coordinate z. This gives the following relationship for the λ-parameter λi = λ j∈Ni Sj d(xi, x j) + αe−β ˆz j . (38) (38) 123 562 F. Vermolen, I. Pölönen 562 F. Vermolen, I. Pölönen Fig. 4 Snapshots of progression of cancer from the upper layer of the domain of computation. Snapshots were taken at 1, 100, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are in 0, 0.8, 2.7, 6.6, 15.7 and 28.4 dimensionless time units Fig. 4 Snapshots of progression of cancer from the upper layer of the domain of computation. Snapshots were taken at 1, 100, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are in 0, 0.8, 2.7, 6.6, 15.7 and 28.4 dimensionless time units Higher values of β imply that the solar irradiation is absorbed faster as the UV radiation penetrates the skin. Lower values of α represent the case that the skin is less sensitive to the UV radiation and that cancer initiation at the top of the skin is inhibited. Application of sun cream could be modelled by larger values of β and/or lower values of α. Using α = 0.1 and β = 1, we get the results from Fig. 5. Large values of β correspond to low transmissibility of UV-radiation into the skin. Well functioning sunscreen (sun cream) will increase the β–value and hence radiation decays very fast in the top layers of skin. 123 Uncertainty quantiﬁcation on a spatial Markov-chain model… 563 Uncertainty quantiﬁcation on a spatial Markov-chain model… 563 Fig. 5 Snapshots of progression of cancer from the upper layer of the domain of computation. Snapshots were taken at 1, 100, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are in 0, 0.3, 1.3, 2.85, 6.0 and 15.7 dimensionless time units Fig. 4.2.1 Scalability of the model The probabilistic nature of the model requires the assessment of its scalability. From adjusting the probability rates according to the number of immediate neighbours in ‘cancer state’ and the inter-nodal distance, it is to be expected that the expected time to reach the state in which half of the nodes are in ‘cancer state’, that is, ˆT0.5 does not depend on the spatial resolution h. Therefore, we carried out multiple simulations with various values of spatial resolution in a one-and two dimensional setting using N = 1000 samples of Monte Carlo simulations. The results have been listed in Table 2. It can be seen that for the larger stepsizes h (the smaller resolutions), there is still a dependence, but that as h becomes smaller, then the results become less dependent on h. Note that these simulations contain an error as a result of the spatial resolution, and as a result of the number of Monte Carlo samples. 4.1.3 Progression and gradually decreasing initiation with the depth of the domain 5 Snapshots of progression of cancer from the upper layer of the domain of computation. Snapshots were taken at 1, 100, 1000, 5000, 15,000 and 30,000 nodes in ‘cancer state’; Corresponding times are in 0, 0.3, 1.3, 2.85, 6.0 and 15.7 dimensionless time units Thereby the probability of cancer initiation, as well as progression can be reduced. Once again, the α-value represents the exposure of the skin surface to sunlight. Sun cream can also decrease this value due to its performance to shield the skin from UV- radiation. It can be seen that cancer develops a bit more homogeneously than in Fig. 4 in the early stages, because the probability rate λ is distributed more homogeneously and hence more nodes that do not yet have a neighbour node that is already in ‘cancer state’ are allowed to transform into ‘cancer state’ in Fig. 5. However, at the latest 123 564 F. Vermolen, I. Pölönen stages the development of cancer becomes similar to the behaviour in Fig. 4 as a result of nearest neighbour interaction. stages the development of cancer becomes similar to the behaviour in Fig. 4 as a result of nearest neighbour interaction. 4.2 Monte Carlo simulations in two dimensions The stochastic nature of the model requires a statistical assessment of the model results. First, we assess the scalability of the model with respect to consistency of the results in terms of convergence of the results as the grid size tends to zero. Secondly, we deal with the consequences from uncertainties in the model parameters. 4.2.2 Implications from uncertainties in the data The simulations so far are realisations of stochastic processes like in the case of the transition probability. Next to the uncertainty in transition, the actual values of input data such as the transition likelihood, as well as the sensitivity with respect to UV-radiation and the amount of UV radiation suffer from uncertainty. Therefore, we vary all these parameters using normal distributions and we carry out Monte Carlo simulations using the sampling values. From the simulations, we compute the expected value of the time that the cancer has grown to a predeﬁned volume fraction in the tissue, as well as the probability distribution of the time that the cancer has grown to a predeﬁned volume fraction. This allows the estimation of the likelihood that cancer Table 2 Dependence on the internodal distance h Number of nodes per dimension ˆT0.5 in R1 ˆT0.5 in R2 25 0.4495 0.2955 50 0.4155 0.2525 100 0.4076 0.2328 200 0.4075 0.2190 400 0.4058 0.2111 1 3 123 12 Uncertainty quantiﬁcation on a spatial Markov-chain model… 565 Table 3 Values of the input parameters Quantity Mean Variance λ 0.005 0.001 α 0.1 0.02 β 3 0.6 develops to a predeﬁned volume fraction within a certain amount of time. To this extent, we use the following sampling for the input parameters λ, α and β: develops to a predeﬁned volume fraction within a certain amount of time. To this extent, we use the following sampling for the input parameters λ, α and β: λ ∼N(μλ, σ 2 λ), α ∼N(μα, σ 2 α), β ∼N(μβ, σ 2 β) (39) (39) The values of the above statistical distributions have been listed in Table 3, which still only contain hypothetical values. The results of the Monte Carlo simulations for 100 × 100 nodes are displayed in Fig. 6. It can be seen that N = 1000 simulations gave good convergence of T1/2 with respect to the Monte Carlo simulations. This is conﬁrmed by the Monte Carlo Error (MCE), which can be estimated consistently using the estimator in Eq. (35) MCEN ≈0.0505. Further, the scatter plots between T1/2 and λ, α and β have been shown as well, from which it can be been seen that T1/2 tends to decrease with increasing λ and α (to a smaller extent). An increase is observed with β. Note that β represents the amount of decay of UV-radiation through the skin. 5 Discussion We propose a phenomenological model for the development of skin cancer. The model is very generic and could be applied to different types of cancers as well. The formal- ism is a two-state (binary) model and it is based on transition probabilities, which reﬂects the likelihood that a certain portion of the domain changes state. Hence the occurrence of the transition from the ‘non cancer state’ to the ‘cancer state’ depends on the states of the points surrounding the point of consideration. It is assumed that the change of state is irreversible. A strong point of the current model is that it is very generic, simple to understand, and that it contains only few steering parameters. Further, the model is easy to implement in more-dimensional settings. This facilitates a straightforward parameter sensitivity analysis. The current implementation further allows a three-dimensional evaluation of the geometry of the tumor. Demidem et al. (2001) experimentally investigate tumor growth in mice under the inﬂuence of drug treatments. Their measured tumor (melanoma) mass as a function of time is ﬁtted to the classical Gompertz model, which displays an S-shape (see Figures 1, 3 and 4 in Demidem et al. 2001). This S-shape is also reproduced by our model, see Fig. 3. Furthermore, our model also quantitatively matches with the results in Demidem et al. (2001) if the right input values are chosen. The current phenomenological model is used to simulate the progression of cancer in the epidermis as long as it does not cross the basement membrane. The process of crossing the basement membrane proceeds by different biological mechanisms. However, if the rate of progression of the cancer across the basement membrane is known then one can model this phenomenon by adjusting the transition probability rate over this barrier. In this way one can model further metastasis by using a location (and if necessary time-dependent) dependent transition probability λ. This has been omitted in the current paper. The model has a stochastic nature and this implies that each modelling outcome represents a possible scenario. Formally, using a time-step and a ﬁnite number of points, implies that the number of possibilities, that is the number of scenarios, is ﬁnite and hence the set of possible outcomes, Ω, is ﬁnite although in a continuous time-setting the number of possible outcomes is inﬁnite. 4.2.2 Implications from uncertainties in the data Using the Pearson correlation test, the correlations with their respective p-values have been listed in Table 4. These numbers conﬁrm the slight negative correlation between T1/2 and α and a stronger negative correlation between T1/2 and λ. A somewhat stronger positive correlation is observed between T1/2 and β. Hence cancer is more likely to progress for larger values of λ and larger values of α, which, respectively, represent the proliferation rate (including cancer cell division and cancer cell migration), and the intensity (or exposure) of UV-radiation. Cancer indeed progresses less quickly for larger UV-absorption (decay) by the skin. These predictions are in line with intuition and common observations and hence in this sense the model provides a reasonable description of the phenomenon. Next to the scatter plots, we present the histogram representingthefrequencyofoccurrenceof T1/2.Itcanbeseenthat T1/2 isnotnormally distributed with some kurtosis with a tail for large values of T1/2. Using the data, we compute the cumulative likelihood that T1/2 is at most equal to a certain value. That is, we compute Pr(T1/2 ≤T ) = FN(T ) = 1 N N j=1 I[0,T ](T ( j) 1/2). (40) (40) where T ( j) 1/2 denotes the jth sample, and I[0,T ](x) denotes the indicator function, which is characterised by I[0,T ](x) = 1, if x ∈[0, T ], 0, else. 123 F. Vermolen, I. Pölönen 566 566 F. Vermolen, I. Pölönen Fig. 6 a Convergence of the Monte Carlo simulations; b scatter plot between λ and T1/2; c scatter plot between α and T1/2; d scatter plot between β and T1/2; e histogram of outcomes of T1/2; f cumulative probability that T1/2 ≤τ Fig. 6 a Convergence of the Monte Carlo simulations; b scatter plot between λ and T1/2; c scatter plot between α and T1/2; d scatter plot between β and T1/2; e histogram of outcomes of T1/2; f cumulative probability that T1/2 ≤τ The results have been plotted in Fig. 6. The cumulative probability should be inter- preted as the likelihood that the tissue is seriously infected by cancer within a speciﬁc time interval. We ﬁnally remark that the uncertainty analysis could also be applied to Eq. (26), to get a statistical distribution of the λ–parameter such that one could sample from this obtained distribution. This has been omitted in the present study. 4.2.2 Implications from uncertainties in the data 123 12 Uncertainty quantiﬁcation on a spatial Markov-chain model… 567 Table 4 Correlation coefﬁcients with T1/2 Quantity Correlation p value λ −0.5502 < 0.0001 α −0.2827 < 0.0001 β 0.7389 < 0.0001 Table 4 Correlation coefﬁcients with T1/2 5 Discussion In this context one should mention that choosing a smaller time-step and a higher grid resolution enlarges the set of all possible outcomes, and the set of possible outcomes may become arbitrarily large if the numerical resolution is increased. Next to the uncertainty as a result of the stochastic nature of the model, the values of the input parameters are not known. This lack of information provides an additional source of uncertainty. If their mean and standard deviation is known as well as their statistical distribution, then samples of 123 568 F. Vermolen, I. Pölönen these values can be inserted into the stochastic model and herewith one can compute the likelihood of different scenarios. This is what we have done in the current paper and this enables us to estimate the probability that certain scenarios will happen. These scenarios are for instance the time-span needed for the cancer to have invaded half of the domain. This is done by the use of Monte Carlo simulations in the current work and this approach to estimate the likelihood of a scenario is not yet that common in the ﬁeld of mathematical biology. Since for an accurate prediction, many Monte Carlo samples are needed, these computations are relatively expensive. One could beneﬁt from a more clever application of Monte Carlo techniques by using a Multi-Level technique, which is based on taking Monte Carlo samples at various grid resolutions. It is possible to prove that the number of Monte Carlo simulations that are needed decrease with increasing resolution. More information about Multi Level Monte Carlo techniques can be found in Cliffe et al. (2011). Since the setting of the current model differs from the deterministic model that is dealt with in Cliffe et al. (2011), the investigation of Multi Level Monte Carlo techniques to this type of models in the future is of great interest. In the current simulations, we did not know any values of input parameters, hence no quantitative conclusions can be drawn from our simulations regarding the devel- opment of skin cancer in humans. Hence a calibration with experimental or clinical outcomes remains to be done in future studies, though we demonstrated by the results of Demidem et al. (2001) and Fig. 3 that such a calibration is indeed feasible. In this paper we are using a general approach to simulate any skin cancer. 5 Discussion Different types of skin cancer can be taken into account when the initial computational domain is created. Each of the major skin cancer types is connected to certain cell types. If the initial skin structure contains squamous, basal and melanocyte cells in the epidermal layer of skin and ﬁbroblast, macrophage, and adipocyte cells in dermal layer, it is possible to take into account skin cancer type speciﬁc modelling progression. We additionally remark that the present model can also be applied to modelling progression of other types of cancers. Regarding alternative computational models for skin cancer, we note the studies by Poleszczuk and Enderling (2013), Morais et al. (2017), Gallinaro et al. (2013) and Ciarletta et al. (2001). In Poleszczuk and Enderling (2013), a cellular-automata framework for tumor growth in general has been developed. In their model, each cancer cell occupies a single gridnode of the cellular automata mesh. Cancer cells are distinguished into cancer stem cells and non-stem cancer cells. The ﬁrst type is immortal and has an unlimited proliferation potential, whereas the second type is mortal with a ﬁnite proliferation potential. Cell division proceeds in an asymmetric way, where the likelihood that a cancer stem cell divides into a cancer stem cell is not the same as the probability of division into a non-stem cancer cell. The model explicitly incorporates the migration and division of cells. These features make their model more sophisticated than our current model, where only a Markov-chain process is used with a simple transition probability, taken into account the number of occupied (in cancer state) neighbours. In our model, each grid node does not necessarily coin- cide with a cellular position, but with a centre of a control volume which is in either cancer state or in non cancer state. Furthermore, our model incorporates the expo- sure to life burden UV-radiation by adjusting the probability of transition probability Uncertainty quantiﬁcation on a spatial Markov-chain model… 569 from one node to another. Although our own model and the model by Poleszczuk and Enderling (2013) contain some similarity by using a cellular automata model with stochastic principles, the main difference is that their model takes into account more biological processes explicitly, whereas the same biological processes are incorpo- rated implicitly in the transition probability in the current formalism in our model. Hence our own model contains fewer parameters, which makes the quantiﬁcation of uncertainty easier. 5 Discussion The formalism by Morais et al. (2017) models the relation between the proliferation of melanocytes and contact inhibition. The two-dimensional frame- work models a mono-layered cell culture, where the Widom-Rowlinson model from statistical physics is used. Two different cell types are considered: cancer cells and healthy cells. In their lattice model, each lattice vertex (or node) is occupied by one cell only. In order to simulate contact inhibition, they use an exclusion principle to bound the distance between cells of different type or of the same type from below. Their principle allows to incorporate that tumour-tumour cells are more allelophilic than tumour-healthy cells or healthy-healthy cells. Cells are more allelophilic if their minimal exclusion distance is smaller, which means that they are allowed to be located closer to each other. This model, like in the study of Poleszczuk and Enderling (2013), is based active migration and division of cells using stochastic processes. The model by Morais et al. (2017) is interesting in the sense that it incorporates more biological processes, like contact inhibition, it still is a somewhat more complicated model than our formalism with a larger number of parameters that are difﬁcult to get from lit- erature or measurements. Gallinaro et al. (2013) model melanoma cell (melanocyte) migration by the use of an elastic continuum. Their approach is based on a system of ordinary differential equation to describe the displacement of the spatial coordinate in the tissue in a one-dimensional conﬁguration. Their model contains some similarities with the morpho-elastic models, which were applied in Koppenol and Vermolen (2017) and Menon et al. (2017). The approach is entirely deterministic and one-dimensional, contrary to our current model which is entirely stochastic and three-dimensional. Cia- rletta et al. (2001) consider a partial differential equations-based model for the radial growth of melanoma. Their approach is based on mixture theory, where they incorpo- rate various phases by accounting for their volume fractions. These different phases are melanoma cells, interstitial liquid, and basal laminae. Furthermore, a balance of momentum, where inertial effects are neglected, result into a saddle-point like system (to be compared with Stokes’ ﬂow or Biot’s poroelasticity model) for the velocity vec- tors of the various components and their pressures. This framework is combined with a balance equation for the nutrients. For the movement of the interface between the tumor and non-tumor regions, a free boundary condition is imposed. 5 Discussion Their simulations suggest instability (analogous to a Mullins-Sekerka instability) of the interface with respect to long-wavelength perturbations. Their approach is interesting in the sense that the mechanisms are dealt with in a physics-based way. However, the amount of input parameters, such as for the application of the interaction forces, makes the model prone to uncertainty since the values of the input parameters are often hard, or even impossible, to get. Furthermore, the numerical approximation of the solution to the saddle point problem is often quite challenging. The application of uncertainty quantiﬁcation on their model will be much more complicated and expensive from a 123 570 F. Vermolen, I. Pölönen computational point of view than the uncertainty quantiﬁcation on our own cellular automata based model. An important weakness of the current formalism is the oversimpliﬁcation of the description of the mechanisms governing the biological process. The mechanisms involve cell division (proliferation), cell mutation, and cell migration, but also the interaction of cells of different phenotypes through mechanical and chemical signals. All these processes have been simpliﬁed through the transition probability, which represents the progression of cancer from one point to another. Though we proved how the transition probability changes with the distance between adjacent points and number of neighbouring points that are in the cancer state, it still simpliﬁes all the mechanisms behind it. Since each model represents a description of a certain process or phenomenon from the modeller’s point of view, it is very hard, and probably even impossible, to ﬁnd a model that describes such a phenomenon perfectly. However, currently we aim at formulating a simple and qualitatively useful model that describes possible scenarios of the development of skin cancer. A clear advantage of such a sim- ple model is its tractability and having just few parameters that are to be tuned or varied for the sake of uncertainty quantiﬁcation. A disadvantage is that it does not contain much of the underlying mechanisms of the phenomenon studied by us. We could have used, and we did this in different studies, more complicated models that entail many of the mechanisms. These models give more complicated mathematical problems with more input parameters which need to be determined through experiments. 5 Discussion Having such a complicated model, also yields a larger dimension of the parameter space and hence the uncertainty of such a model raises the legitimate question whether a more realistic model gives a better and more accurate description than a simple model such as the current model. If only prediction is what one is aiming at, then, it is less impor- tant that the model takes into account many of the underlying mechanisms and for these purposes a less sophisticated model could be useful as well. Furthermore, as it has been said earlier, the complicated models often involve complex mathematical problems that need to be solved by using computationally expensive procedures so that the parameter sensitivity analysis and/or prediction of likelihood of particular scenarios becomes very expensive and may even become unpractical if not enough computational power is available. More complicated models could be based on continuum-scale models based on partial differential equations, where one takes into account chemical and mechanical signals. This amounts to solving elliptic, hyperbolic and parabolic partial differential equations. An example is given in Ciarletta et al. (2001) where the radial growth of melanoma is modelled. Here one can use ﬁnite-element strategies for the approxima- tion of the solution to the partial differential equations in case of more-dimensional cases. The equations are then solved for displacements, stresses, concentration of chemicals and cell densities. One can track the cell density of cancer cells to get an impression of the development of cancer. It is known that skin cancer initiation predominantly takes place on or near the base- ment membrane, where cells are most proliferative. At the very top of the epidermis, the cells are keratinized or even dead and hence cancer is not likely to initiate there. In this view the simulations in Figs. 4 and 5 should be considered as academic cases, which have limited biological relevance. The objective of the current paper is to show Uncertainty quantiﬁcation on a spatial Markov-chain model… 571 the applicability of the model, which can be calibrated to obtain an agreement with the experiments by Demidem et al. (2001), as well as showing the straightforward quantiﬁcation of uncertainty using the current model. the applicability of the model, which can be calibrated to obtain an agreement with the experiments by Demidem et al. (2001), as well as showing the straightforward quantiﬁcation of uncertainty using the current model. 5 Discussion As it has been mentioned earlier, the current model has a very generic nature. Since the progression and development of (viral) infections is based on similar features of changing state from a ‘non-infected state’ to an ‘infected state’ depending on the states of neighbouring points, we think that our model is also applicable to modelling the infection of tissues by viruses. We ﬁnally remark that we will compare the current model to Fisher-Kolmogorov’s partial differential equation (see Murray 2004; Kaliappan 1984), in which the solution can be interpreted as the probability that a particular point in space is in cancer state at a certain time. 6 Conclusions We developed a spatial Markov chain model for the progression of cancer in skin tissue under the inﬂuence of proliferation rate, UV-radiation intensity, and absorption rate by the skin. The model has been evaluated using Monte Carlo simulations, which allow the establishment of correlation between the cancer development rate and the input parameters like cancer proliferation rate, UV radiation intensity and the UV absorption coefﬁcient by the skin tissue. Furthermore, the model allows the computation of the likelihood that a tissue becomes seriously infected by cancer within a speciﬁc time- interval. We additionally demonstrated by computations that it is possible to ﬁt our model to experiments where the tumor mass is evaluated over time. In future studies, we want to use real-world valued parameter values by calibration to experiments on humans, as well as relating λ to the migration rate of travelling waves in the context of partial differential equations like Fisher-Kolmogorov’s equation. Acknowledgements Theauthorsthanktherefereesfortheirvaluablecriticalremarks,whichhavebeneﬁtted to the manuscript. OpenAccess ThisarticleislicensedunderaCreativeCommonsAttribution4.0InternationalLicense,which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Adra S, Sun T, MacNeil S, Holcombe M, Smallwood R (2010) Development of a three-dimensional mul- tiscale computational model of the human epidermis. PLoS ONE 5(1):e8511 Alarcon T, Byrne HM, Maini PK (2003) A cellular automaton model for tumor growth in heterogeneous environment. J Theor Biol 225:257–274 Alarcon T, Byrne HM, Maini PK (2003) A cellular automaton model for tumor growth in heterogeneou environment. J Theor Biol 225:257–274 References J Theor Biol 203:367–382 pp g y Kansal AR, Torquato S, Harsh GR, Chiocca EA, Deisboeck TS (2000) Simulated brain tumor growth dynamics using a three-dimensional cellular automaton. J Theor Biol 203:367–382 Koehler E, Brown E, Haneuse SJ-PA (2009) On the assessment of Monte Carlo error in simulation-based statistical analyses. Am Stat 63(2):155–162 Koppenol DC, Vermolen FJ (2017) Biomedical implications from a morphoelastic continuum model for the simulation of contracture formation in skin grafts that cover excised burns. Biomech Model Mechanobiol 16(4):1187–1206 Li X, Upadhyay AK, Bullock AJ, Dicolandrea T, Xu J, Binder RL, Robinson MK, Finlay DR, Mills KJ, Bascom CC, Kelling CK, Isfort RJ, Haycock JW, MacNeil S, Smallwood RH (2013) Skin stem cell hypotheses and long term clone survival—explored using agent-based modelling. Sci Rep 3(1904):1–9 Lo C-M, Wang H-B, Dembo M, Wang Y-L (2000) Cell movement is guided by the rigidity of the substrate. Biophys J 79:144–152 Menon SN, Hall CL, McCue SW, McElwain DL (2017) A model for one-dimensional morphoelasticity an its application to ﬁbroblast-populated collagen lattices. arXiv:1705.00299 Monteagudo A, Santos J (2012) A cellular automaton model for tumor growth simulation. In: Rocha P, Luscombe N, Fdez-Riverola F, Corchado Rodriguez JM (eds) 6th international conference on practical applications of computational biology & bioinformatics. Advances in intelligent and soft computing. Springer, Heidelberg Morais MCC, Stuhl I, Sabino AU, Lautenschlager WW, Queiroga AS, Tortelli TC, Chammas R, Suhov Y, Ramos AF (2017) Stochastic model of contact inhibition and the proliferation of melanoma in situ. Sci Rep 7:8026 Murray JD (2004) Mathematical biology I: an introduction (Chapter 13). Springer, New York Newton PK, Mason J, Bethel K, Bazhenova L, Kuhn P (2012) A stochastic Markov chain model to describe lung cancer growth and metastasis. PLoS ONE 7(4), article number e34637 Newton PK, Mason J, Bethel K, Bazhenova L, Nieva J, Norton L, Kuhn P (2013) Spreaders and sponges deﬁne metastasis in lung cancer: a Markov chain Monte Carlo mathematical model. Cancer Res 73(9):2760–2769 Newton PK, Mason J, Venkatappa N, Jochelson MS, Hurt B, Nieva J, Comen E, Norton L, Kuhn P (2015) Spatiotemporal progression of metastatic breast cancer: a Markov chain model highlighting the role of early metastatic sites. Breast Cancer 1: article number 15018 Poleszczuk J, Enderling H (2013) A high-performance cellular automaton model of tumor growth with dynamically growing domains. arXiv:1309.6015 Qi AS, Zheng X, Du CY, An BS (1993) A cellular automaton model of cancerous growth. References 123 123 572 F. Vermolen, I. Pölönen oon WM, Koppenol DC, Vermolen FJ (2016) A multi-agent cell-based model for wound contraction. J Biomech 49(8):1388–1401 utler J, Mackay F, Denniston C, Daley M (2014) Simulating cancer growth using cellular automata to detect combination drug targets. In: Ibarra OH, Kari L, Kopecki S (eds) Unconventional computation Butler J, Mackay F, Denniston C, Daley M (2014) Simulating cancer growth using cellular automata to detect combination drug targets. In: Ibarra OH, Kari L, Kopecki S (eds) Unconventional computation and natural computation. Lecture Notes in Computer Science. Springer, Cham g g , , p ( ) p and natural computation. Lecture Notes in Computer Science. Springer, Cham and natural computation. Lecture Notes in Computer Science. Springer, Cham Ciarletta P, Foret L, Ben Amar M (2011) The radial growth phase of malignant melanoma: multi-phase modelling, numerical simulations and linear stability analysis. J R Soc Interface 8:345–368 Cliffe KA, Giles MB, Scheichl R, Teckentrupp AL (2011) Multilevel Monte Carlo methods to elliptic PDEs with random coefﬁcients. Comput Vis Sci 14:3–15 Cliffe KA, Giles MB, Scheichl R, Teckentrupp AL (2011) Multilevel Monte Carlo methods and applicatio lli i PDE i h d fﬁi C Vi S i 14 3 15 Demidem A, Morvan D, Papon J, De Latour M, Madelmont JC (2001) Cystemustine induces redifferentia- tion of primary tumors and confers protection against secondary tumor growth in a melanoma murine model. Cancer Res 61:2294–2300 Eriksson T, Tinghög (2015) Societal cost of skin cancer in Sweden in 2011. Acta Dermato Venereol 95(3):347–348 Gallinaro JV, de Godoy Marques CM, Mendes de Azevedo F, Hisayasu Suzuki DO (2013) Mathematical modelling of melanoma cell migration with an elastic continuum model for the evaluation of the inﬂuence of tumor necrosis factor-alpha on migration. J Comput Med 2013:534073 GrabeN,NeuberK(2007)Simulatingpsoriasisbyalteringtransitamplifyingcells.Bioinformatics23:1309– 1312 Jiao Y, Torquato S (2011) Emergent behaviors from a cellular automaton model for invasive tumor growth in heterogeneous microenvironments. PLoS Comput Biol 7(11):e1002314. https://doi.org/10.1371/ journal.pcbi.1002314 Kaliappan P (1984) An exact solution for travelling waves of ut = Duxx + u −uk. Physica D 11:368–37 Kaliappan P (1984) An exact solution for travelling waves of ut = Duxx + u −uk. Physica D 11:368–374 Kansal AR, Torquato S, Harsh GR, Chiocca EA, Deisboeck TS (2000) Simulated brain tumor growth dynamics using a three-dimensional cellular automaton. Sun T, Adra S, Smallwood R, Holcombe M, MacNeil S (2009) Exploring hypotheses of the actions of TGF-beta1 in epidermal wound healing using a 3D computational multiscale model of the humane epidermis. PLoS ONE 4:e8515 Thingnes J, Lavelle TJ, Hovig E, Omholt SW (2012) Understanding the melanocyte distribution in human epidermis: an agent-based computational approach. PLoS ONE 7(1):e40377 West J, Hasnain Z, Mason J, Newton PK (2016) The Prisoner’s dilemma as a cancer model. arXiv:1512.04591v4 [q–bio.PE], January 16 2016 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. World Health Organisation (2017) Skin cancer. http://www.who.int/uv/faq/skincancer/en/index1.html epidermis. PLoS ONE 4:e8515 Thingnes J, Lavelle TJ, Hovig E, Omholt SW (2012) Understanding the melanocyte distribution in human epidermis: an agent-based computational approach. PLoS ONE 7(1):e40377 References J Theor Biol 225:257–274 123 12 Uncertainty quantiﬁcation on a spatial Markov-chain model… 573 Sun T, Adra S, Smallwood R, Holcombe M, MacNeil S (2009) Exploring hypotheses of the actions of TGF-beta1 in epidermal wound healing using a 3D computational multiscale model of the humane epidermis. PLoS ONE 4:e8515 Thingnes J, Lavelle TJ, Hovig E, Omholt SW (2012) Understanding the melanocyte distribution in human epidermis: an agent-based computational approach. PLoS ONE 7(1):e40377 West J, Hasnain Z, Mason J, Newton PK (2016) The Prisoner’s dilemma as a cancer model. arXiv:1512.04591v4 [q–bio.PE], January 16 2016 World Health Organisation (2017) Skin cancer. http://www.who.int/uv/faq/skincancer/en/index1.html Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 123
https://openalex.org/W4385154629	https://ammanif.com/journal/jcp/index.php/home/article/download/74/75	English	null	Formulation and evaluation of 5-fluorouracil controlled release chronotherapeutic drug delivery system (CTDDS) for colorectal cancer	Journal of contemporary pharmacy	2,022	cc-by	4,579	FORMULATION AND EVALUATION OF 5-FLUOROURACIL CONTROLLED RELEASE CHRONOTHERAPEUTIC DRUG DELIVERY SYSTEM (CTDDS) FOR COLORECTAL CANCER Hammad Ayaz1, Muhammad Ahmad1, Mustafa Kazmi1, Husnain Ahmed1, Talib Hussain1, Ishtiaq Jeelani2 1 Institute of Pharmaceutical Sciences, University of Veterinary and Animal Sciences, Lahore Pakistan 2 Department of Medicine, Division of Endocrinology and Metabolism, University of California, San Diego, CA, USA Submitted 6th September 2022, 2nd November 2022 Submitted 6th September 2022, 2nd November 2022 ISSN 2521-0521 ISSN 2521-0521 JOURNAL OF CONTEMPORARY PHARMACY Volume 6: Issue 2: 2022 ABSTRACT Objective: As most of the drugs disintegrate and dissolve in stomach before reaching the target site and to substitute intravenous (IV) route based chrono modulated chemotherapy, oral colon target drug delivery system was formed. The aim of this study was to design, develop, and evaluate a colon targeted tablet containing 5-Fluorouracil (5-FU) to give a controlled release effect of drug for colonic cancer with a goal to increase the bioavailability and improve the patient compliance. Methods: Varied concentration of different polymers such as Xanthan gum and Eudragit were used to get an optimized formulation of 5-FU tablet. The prepared formulation was evaluated for pre compression and post compression parameters such as hardness, weight, friability and drug content uniformity. The optimized formulation was further evaluated by Fourier Transformed Infrared Spectroscopy (FTIR), Differential Scanning Calorimetry (DSC), in-vitro dissolution studies, dissolution kinetic modeling and stability analysis. Results: All pre-formulation tests were within range of USP standards. Friability of all tablets was satisfied. The interference of the polymers was ruled out by FTIR. In-vitro release studies of 5- FU tablets in phosphate buffer of pH 7.0 were performed using a modified diffusion cell that resulted sustained release (90.99% to 92.69% after 12h) and kinetic models depicted the combined diffusion and dissolution mechanism of release. The optimized tablets were found having only physical interactions based on DSC. The product was found stable when evaluated using accelerated stability studies. Conclusion: It was concluded from the studies that the colon target tablet of 5- FU prepared by different concentration of polymers were optimized and can be efficiently used to control the rate of drug release to the colon in the belief of improved therapeutic efficacy and tolerability. Therefore, it is a better alternative for intravenous route based chrono modulated chemotherapy. Keywords: 5 Fluorouracil, Eudragit S100, Colon cancer, CTDDS, pH dependent delivery system Corresponding Author. E-mail: talib.hussain@uvas.edu.pk g MATERIALS 𝑖= [(𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊1 −𝐹𝑖𝑛𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊2) ⁄ (𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊1)]𝑥 100 𝑖= [(𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊1 −𝐹𝑖𝑛𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊2) ⁄ (𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑊1)]𝑥 100 5-fluorouracil was received as gift from Pharmedic Labs Lahore, Microcrystalline cellulose (MCC), Sodium starch glycolate, Guar gum, Xanthan gum, Tri ethyl citrate, Eudragit S100 and PVP K30 were purchased from Sigma Aldrich Pakistan. Lactose, Acetone, Starch, Isopropanol, Mg Stearate, Talc, Aerosil and Pectin were purchased from Merck Pakistan. All the chemicals used were of analytical grade. Distilled water was used throughout the study. Preliminary Studies The preliminary studies were performed for the development of formulation to evaluate the purity and compatibility of the active substance of the drug. Colon targeted drug delivery system is highly desirable for local treatment of bowel diseases such as Crohn’s disease, colonic cancer, ulcerative colitis and local treatment of colonic pathologies [9]. The colonic contents have a longer retention time (up to 5 days), and the colonic mucosa is known to facilitate the absorption of several drugs, making this organ an ideal site for drug delivery [10]. However, targeting the drug to colon is a very complicated task because of its location to distal portion of alimentary canal, also because of the presence of enzymes and wide range of pH values [11]. Crystallinity A pinch of 5-fluorouracil of powder was taken onto the glass slide. The glass slide was placed onto stage of microscope. Crystal was observed by using different objective lenses. Angle of Repose Fixed the funnel on the stand and placed the paper below the funnel. A heap of powder was formed. The height of the peak was measured by the scale. The radius was calculated from the diameter where up to the powder disperse and calculated the angle of repose. Fixed the funnel on the stand and placed the paper below the funnel. A heap of powder was formed. The height of the peak was measured by the scale. The radius was calculated from the diameter where up to the powder disperse and calculated the angle of repose. Post-formulation Evaluation Tablet Hardness For each formulation, 20 tablets were randomly selected. Monsanto hardness tester was used and hardness of each tablet was checked. INTRODUCTION is clinically in use because of its mercurial bioavailability but IV use produce severe side effects due to cytotoxic effect on non-tumour cells [4]. So, a colon targeted 5-FU that overcome the side effects using various polysaccharides was formulated [5]. It used to be a common practise that every anti-cancer medication would cause effect on all sites that comes in its path before reaching the required site [6]. But now with advancement of knowledge and polymeric based products this field had successfully advanced in terms of providing site specific action. The pH specific drug release is the key feature of colon targeting drug delivery systems [7]. Colorectal cancer involves the abnormal growth of cells in colon or rectum. If diagnosed in early stage there is 91% chance of recovery but if diagnosed in late stage in which cancer has spread throughout the tissues, organs and lymph nodes there is still 72% chance of recovery [1]. Treatment usually involves surgery, radio therapy as well as chemo therapy depending upon condition [2]. 5 - Fluorouracil is among the drugs used in chemotherapy. Fluorouracil is an anticancer that is the backbone of all chemotherapeutic drugs for the patients dealing with advanced colorectal cancer [3]. It is antimetabolite that is fluorinated of the pyrimidine uracil. IV administration of this drug 65 Conventional non targeted drug delivery system may have low efficacy and undesirable side effects because drug is systemically absorbed before reaching the targeted site. Therefore, colon targeted drug delivery system was introduced to increase the bioavailability of drug and to selectively release the drug as a result of colonic environment without premature drug release in upper GI tract [8]. i.e., magnesium stearate and talc for 30 minutes. After that granules were punched by punching machine and tablets were formed. The formulation design is described in Table 1. LOD was calculated by using the following formula: LOD was calculated by using the following formula: Melting Point Placed a small amount of sample drug on the hot stage of apparatus. Observed the sample through eyepiece lens and recorded the temperature through thermometer. Drug localization at the colon helps in releasing the drug at colon as a result, major amount of the drug enters the colon. Main goal is to develop 5- fluorouracil tablet using micro emulsifying drug delivery method so to administer the drug orally to improve the patient compliance and also making available the dosage form directly to large bowel through the GIT to localize the maximum amount of drug in the colon region and also attaining the sustained release effect [12]. Thickness Twenty tablets were randomly selected from each formulation. Thickness of each tablet was measured using the micrometer. Thickness was expressed as Mean ± Standard Deviation. Friability METHODS 2% w/v starch slurry was prepared as a binder and added it into the ingredients such as Xanthan Gum and Eudragit S 100 and 5FU to form dough. Granules were formed by passing it through sieve #12. The granules were dried in hot air oven for half an hour and then granules were properly sized by passing through sieve #40. After that dried granules were kept at room temperature for 3-5 hours for moisture content adjustment. The granules were blended with glidants and lubricant g Hygroscopicity 5- Fluorouracil was weighed and placed in the China dish. Heated it with the help of oven at 80℃ for 15 minutes. Noted the difference in weight. 5- Fluorouracil was weighed and placed in the China dish. Heated it with the help of oven at 80℃ for 15 minutes. Noted the difference in weight. expressed Friability Twenty tablets were selected randomly. Initial weight of tablets (w1) was recorded. Tablets were placed in the drum of the friabilator, for 4 minutes at 25 rpm. Percent weight loss (w) was calculated as as. 𝑊= (𝑊1 −𝑊2) 𝑊1 ⁄ 𝑋 100 𝑊= (𝑊1 −𝑊2) 𝑊1 ⁄ 𝑋 100 66 Table 1: Formulation Design containing four formulations with fixed. Ingredients Concentration (mg) F1 F2 F3 F4 5- Fluorouracil 100 100 100 100 MCC 90 - 60 Sodium Starch Glycolate - 2.5 4 Guar Gum - - 150 60 Xanthan Gum 10 20 - Triethyl citrate - 0.6 - Eudragit S 100 100 60 - 75 PVP K30 - 1 - Lactose 68 80 32 Acetone - 34 - Starch 20 - 10 Isopropanol - 51.4 - Mg Stearate 3% 7 2.5 4 Talc2% 5 5 4 5 Aerosil - 5 - Pectin 50 Total Weight (mg) 300 300 300 300 Drug Content was determined as: Kinetic modeling of Drug Release Kinetic modeling of Drug Release Calibration curve was obtained by using UV absorbance. After that theoretical yield was calculated and dissolution data of four formulations was taken, the percentage release was calculated, and curve was obtained against time in hours and percentage release of formulations. By using DD Solver models were obtained to check which model was best followed by our formulations. The models applied were Zero-order, First-order, Higuchi model & Korsmeyer Peppas model to understand release kinetics of our formulation. = 𝐴𝑐𝑡𝑢𝑎𝑙 𝑌𝑖𝑒𝑙𝑑𝑇ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 𝑌𝑖𝑒𝑙𝑑 ⁄ 𝑋 100 In Vitro Drug Release Buffer of pH 7.0 was made by using distilled water. 450ml of buffer was added in round bottom flask of dissolution apparatus. One tablet was added into this round bottom flask. Samples of 1ml was taken after intervals of 0.5, 1,2,3,4,6,8,12,18 and 24 hours of placing the tablet in dissolution apparatus. Dilutions were made by adding 9ml of Phosphate buffer of pH 7.0 to each sample. Absorbance was observed of these dilutions using UV spectrophotometer at 265nm. q g 𝑆𝑤𝑒𝑙𝑙𝑖𝑛𝑔 𝐼𝑛𝑑𝑒𝑥= (𝑊∗−𝑊) 𝑊 ⁄ 𝑋 100 𝑆𝑤𝑒𝑙𝑙𝑖𝑛𝑔 𝐼𝑛𝑑𝑒𝑥= (𝑊∗−𝑊) 𝑊 ⁄ 𝑋 100 Calibration Curve Standard solution of 1mg/mL was prepared using 5FU 99.9% using distilled water. The standard solution was further diluted to prepare working standard solution of 100µg/mL. The working solution was used to prepare five dilutions of fixed concentration i.e., 2, 4, 6, 8, 10µg/mL. These dilutions were scanned (n=3) at 265nm using UV spectrophotometer. Drug Content 20 tablets from each formulation were weighed and crushed in pestle and mortar. 10% of the total powder was weighed and soaked in 10ml of phosphate buffer of pH 7.0, stirred for 30 minutes. Stored the soaked part in room temperature for 24 hours to allow the drug to penetrate back from the polymer into the solvent. The suspension was centrifuged at 4000rpm for 15 minutes and the supernatant was separated with the help of 1cc syringe. Supernatant solution was analyzed on UV spectrophotometer at wavelength of 265nm. Swelling Studies Swelling studies were performed on core tablets. Tablet from each formulation was selected, weighed (w) and after that 10 ml of phosphate buffer (pH 7.0) was used in petri dish to place the tablet [13]. Removed the tablets at regular intervals (10, 20, 40, 80, 120, 160) from petri dish and filter paper was used to remove excess of water. Reweighed (w*) the swollen tablets. Equation was used to find out swelling index: = 𝐴𝑐𝑡𝑢𝑎𝑙 𝑌𝑖𝑒𝑙𝑑𝑇ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 𝑌𝑖𝑒𝑙𝑑 ⁄ 𝑋 100 In Vitro Drug Release RESULTS Various parameters that were studied to physically characterize tablets (Table 2). Differential Scanning Calorimetry (DSC) Differential Scanning Calorimetry (DSC) 5FU, polymers and ten tablets from formulations were crushed to powder and 5mg of powder was sealed in the aluminum pan for analysis. Q1000, 67 TA software were used for enthalpy reading of every peak. At a scanning rate of 10oC/min the thermal behavior was analyzed. TA software were used for enthalpy reading of every peak. At a scanning rate of 10oC/min the thermal behavior was analyzed. Stability Studies The kinetics of drug release were evaluated by applying DDSolver based mathematical models such as zero order, first order, Higuchi and Korsemeyer Peppas. The drug release kinetics of above 4 formulations were presented in the Table 3. Stability studies for the optimized 5-FU colon- targeted tablets were conducted as per ICH guidelines by storing for 3 months at the temperature 40 ± 2 °C and 75% ± 5% RH. Physical characteristics and drug content was noted at the start and after that at the end of 1st, 2nd, 3rd months. Zero-day samples were set as control. FTIR Analysis Powder samples (5FU, plolymers, Formulation) were placed at the stage of spectrometer (Bruker FTIR Alpha spectrometer). The diamond tip of spectrometer was placed on the sample to analyze the sample. A beam of infrared was passes through the sample and curve was obtained 4000- 400cm-1 spectra wavelength. DSC Analysis y DSC analysis was performed on all the formulations which showed the absence of any incompatibility in the physical mixture of the drug with excipients. The different endothermic peaks are represented in Fig. 2. Post-formulation All the tablets formed showed the hardness within the variable range. Thickness of all tablets showed within range of 2.3 – 2.7 mm. Friability of all tablets was found out to be less than 1%. Weight variation of all tablets was within range. Calibration Curve Calibration curve was found linear with value of R2 value of 0.9995. The prepared curve is given Calibration curve was found linear with value of R2 value of 0.9995. The prepared curve is given in Fig. 1. R2 value of 0.9995. The prepared curve is given in Fig. 1. Stability Studies Accelerated stability studies were carried out on tablets of 5- Fluorouracil and drug content was calculated after 1, 2, 3 months. %CDR was found out during the drug release study conducted in simulated physiological environment of various regions of GIT after storage (Table 4). Table 4: Mean drug contents of 5-FU after storage at 40oC and 75% RH for 6 months. Duration of storage for stability Mean drug content ±S.D. (%) 1 month 99.85± 0.53 3 months 98.25 ± 0.33 6 months 96.66 ± 0.56 Table 4: Mean drug contents of 5-FU after storage at 40oC and 75% RH for 6 months. Table 4: Mean drug contents of 5-FU after storage at 40oC and 75% RH for 6 months. Table 5: %age cumulative 5-FU release after dissolution in simulated GIT and Colonic fluid when stored for 6 months. Table 5: %age cumulative 5-FU release after dissolution in simulated GIT and Colonic fluid when stored for 6 months. Drug release region of GIT Duration of storage % CDR Stomach 1 month 87.84 Small Intestine 2 months 78.44 Colon 6 months 76.89 Drug release region of GIT Duration of storage % CDR Stomach 1 month 87.84 Small Intestine 2 months 78.44 Colon 6 months 76.89 For F2 the Zero-order model was not followed as the values of R2, AIC, MSC were 0.87, 91.02, 1.97 respectively. Here AIC value was greater than 50 which was out of standard range along with MSC that is less than 6 and R2 that is less than 0.95 as well [16]. Due to deviation of values from standard range F2 formulation was not following Higuchi Model. It was following First Order Model because R2 and MSC values are in range of standard value. So, first order model is valid. It also followed Korsmeyer Peppas Model because values R2 0.97 and MSC 3.30 were in range. And according to value of n 0.671 it had non Fickian diffusion mechanism [17, 18]. For F3 formulation the First-order values were in range for R2 and MSC which were 0.97 and 3.7 but AIC value was 66.9 which was greater than 50 so it followed this model. For Zero-order and Higuchi values again varied from standard range so it didn’t follow both models. F3 obeyed the Korsmeyer Peppas model as values of R2, AIC, MSC were 0.991, 58.93, 4.41 respectively. FTIR Analysis Table 2: Physiochemical properties of 5-FU. Physical Characterization Parameters Results Crystallinity Monoclinic Melting Point 280oC Hygroscopicity Non- Hygroscopic Flow Properties Passable Flow FTIR analysis was also performed on the powder mixtures of all tablets, and it confirmed the results of DSC in showing an absence of any incompatibility. The different IR transmission peaks are represented in Fig. 3. Figure 1: Calibration curve of 5-Fluorouracil using fixed dilution method. y = 0.0654x + 0.042 R² = 0.9995 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 2 4 6 8 10 12 Absorbance Concentration (µg/ml) y = 0.0654x + 0.042 R² = 0.9995 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 2 4 6 8 10 12 Absorbance Concentration (µg/ml) Figure 1: Calibration curve of 5-Fluorouracil using fixed dilution method. 68 69 g y pp Formulations Zero Order First Order Higuchi Korsmeyer Peppas R2 AIC MSC R2 AIC MSC R2 AIC MSC R2 AIC MSC n F1 0.964 73.61 2.162 0.976 68.73 3.569 0.894 86.64 2.077 0.991 58.47 4.424 0.900 F2 0.873 91.02 2.162 0.976 71.12 3.562 0.930 83.93 2.494 0.971 74.19 3.306 0.671 F3 0.946 78.74 2.758 0.980 66.99 3.738 0.913 84.59 2.271 0.990 58.89 4.413 0.756 F4 0.922 83.58 2.387 0.976 69.31 3.577 0.926 82.96 2.439 0.987 62.90 4.111 0.717 Figure 2: DSC thermographs of (a)5-FU, (b) Physical mixture F1. Figure 3: Representation of FTIR transmittance of drug, polymer and formulation F1. Figure 2: DSC thermographs of (a)5-FU, (b) Physical mixture F1. Figure 2: DSC thermographs of (a)5-FU, (b) Physical mixture F1. Figure 3: Representation of FTIR transmittance of drug, polymer and formulation F1. 69 evaluating R2 0.976 which was in the range of being a standard value. The value of AIC was 68.74. It was deviated from standard range. The value of MSC was 3.56 (should be greater than 3). So, this formulation followed First order Model. If we talk about Higuchi model all 3 values of R2, AIC, MSC were 0.89, 86.63, 2.07 respectively which showed that all values were not following the standard range of Higuchi model. Then next model Korsemeyer Peppas was applied which showed that the values of R2, AIC, MSC were 0.9906, 58.47, 4.42 respectively were in the standard range. If the value of n is greater than 0.45 but less than 1 then it is non Fickian mechanism. FTIR Analysis If 0.45 < n indicates that it is obeying a Fickian diffusion mechanism and n=1 represents a pure relaxed controlled delivery. If it is greater than 1 than it is super case 2. The value of n was 0.900 for F1 showed it was non Fickian Case 2 diffusion mechanism [15]. In-vitro Drug Release Dissolution studies were performed to determine the percentage drug released over time by the tablets. The percentage drug release depends on the type and concentrations of polymers used. Hydrophilic polymers release drug at a faster rate leading to higher percentage release than hydrophobic polymers. Therefore, a blend is required for optimum release. In-Vitro drug release is presented in Fig. 5. Post-formulation Post-formulation The weight variation and the drug content of the prepared tablets were found to be uniform and within the pharmacopeial limits. The hardness of the tablets was within the range of 4.1 - 6.5 kg/cm2 and the friability was below 1.0%, respectively. Stability Studies The value of n was between 0.5 and 1 means it was following non Fickian diffusion mechanism. 6. Dev RK, Bali V, Pathak K. Novel microbially triggered colon specific delivery system of 5-Fluorouracil: statistical optimization, in vitro, in vivo, cytotoxic and stability assessment. Int J Pharm. 2011; 411 (1-2): 142- 51. 7. Sarangi MK, Rao MEB, Parcha V. Smart polymers for colon targeted drug delivery systems: a review. International Journal of Polymeric Materials and Polymeric Biomaterials. 2020; 70 (16): 1130-66. 8. Lee SH, Bajracharya R, Min JY, Han JW, Park BJ, Han HK. Strategic Approaches for Colon Targeted Drug Delivery: An Overview of Recent Advancements. Pharmaceutics. 2020; 12: (1). CONCLUSION The Colon Target Drug Delivery system of 5- Fluorouracil will help for the targeted and controlled release of drug in the colon. The formulations prepared showed good results for this system of 5-Fluorouracil. F1 being the best formulation provided excellent results and is optimized for the target delivery system with effective results giving controlled release effect. DSC Analysis The DSC thermal curve at a scan rate of 20oC/min of 5-FU and physical mixture of tablet excipients at a melting point of 70.8 + 0.2oC and 282.4 + 0.2oC showed a single sharp endothermic peak. Both drug peaks showed a slight peak broadening. Absence and shifting of peaks were not observed [20] ( ) 9. Mundargi RC, Tan E-L, Seo J, Cho N-J. Encapsulation and controlled release formulations of 5-fluorouracil from natural Lycopodium clavatum spores. Journal of Industrial and Engineering Chemistry. 2016; 36: 102-08. FTIR Analysis Two carbonyl functional groups at 1730 and a C- N at 1245 cm-1 were observed in FTIR analysis of pure 5-fluorouracil. Due to C-H and N-H bond the broad stretch occurs between 3150-28 cm-1. Characteristic band of methyl and methylene were Drug Uniformity g U y According to USP guidelines, the percentage drug content should lie between 85% - 115%. The drug content in the tablets were uniformly distributed. The content of tablets was found in the range of 90.9 – 92.6. From the results it was concluded that there was proper distribution of drug in all formulations. According to USP guidelines, the percentage drug content should lie between 85% - 115%. The drug content in the tablets were uniformly distributed. The content of tablets was found in the range of 90.9 – 92.6. From the results it was concluded that there was proper distribution of drug in all formulations. Stability Studies Stability studies were carried out for 3 months at the temperature of 40 ± 2 °C and RH level of 75% ± 5%, morphology of tablets didn’t change and drug content level remains unchanged. During storage time, these results stated that the drug formulation remains stable chemically and physically. [6] Kinetic Modeling of Drug Release After evaluating zero order model on formulation F1 it was found out that value of R2 was 0.9642. Which evaluated that it was in the standard range. The value of AIC was 73.64 which was not in the standard range because its reference range should be less than 50. Drawing out the conclusion by above results indicated that zero order was not followed by F2 formulation [14]. Zero order model shows rate of reaction is After evaluating zero order model on formulation F1 it was found out that value of R2 was 0.9642. Which evaluated that it was in the standard range. The last formulation F4 due to its greater MIC value than 50 and MSC value being less than 3 didn’t follow zero-order and Higuchi. On the other hand, it followed Krosmeyer Peppas model because R2 and MSC were in range and n showed it has non Fickian diffusion. So, we will take formulation was F1 as our ideal formulation because of its linearity of percentage release in graph [19]. The value of AIC was 73.64 which was not in the standard range because its reference range should be less than 50. Drawing out the conclusion by above results indicated that zero order was not followed by F2 formulation [14]. y Zero order model shows rate of reaction is independent of concentration of reactants. Moving onto next model, First-order model after 70 Figure 4: Percentage of drug release presented against time in phosphate buffer pH 7.4. 0 20 40 60 80 100 120 0 5 10 15 20 25 30 Percentage Release Time (hr) F 1 F 2 Figure 4: Percentage of drug release presented against time in phosphate buffer pH 7.4. displayed at the spectrum of ES-100 at 2997- 2952 cm-1. The other polymers show similar peak to Eudragit at 1730-1734, corresponding to carbonyl stretch of 5-FU [18] g g y ; 10. Amidon S, Brown JE, Dave VS. Colon-targeted oral REFERENCES 1. Das M, Ghosh B, Sen S, Ghosh LK. Formulation and Optimization of Controlled Release 5-Fluorouracil Tablets for Colonic Delivery. Journal of Young Pharmacists. 2017; 9 (2): 192-96. 2. Center MM, Jemal A, Smith RA, Ward E. Worldwide variations in colorectal cancer. CA Cancer J Clin. 2009; 59 (6): 366-78. 3. Longley DB, Harkin DP, Johnston PG. 5-fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer. 2003; 3 (5): 330-8. 3. Longley DB, Harkin DP, Johnston PG. 5-fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer. 2003; 3 (5): 330-8. ( ) 4. Malviya R. High Ester Pectin Based Formulation for Site Specific Delivery of 5-Fluorouracil in to Descending Colon. Pharmaceutica Analytica Acta. 2015; 07: 03. 4. Malviya R. High Ester Pectin Based Formulation for Site Specific Delivery of 5-Fluorouracil in to Descending Colon. Pharmaceutica Analytica Acta. 2015; 07: 03. 5. Singh S, Kotla NG, Tomar S, Maddiboyina B, Webster TJ, Sharma D, et al. A nanomedicine-promising approach to provide an appropriate colon-targeted drug delivery system for 5-fluorouracil. Int J Nanomedicine. 2015; 10: 7175-82. 71 drug delivery systems: design trends and approaches. AAPS PharmSciTech. 2015; 16 (4): 731-41. 16. Vilaca N, Amorim R, Machado AF, Parpot P, Pereira MF, Sardo M, et al. Potentiation of 5-fluorouracil encapsulated in zeolites as drug delivery systems for in vitro models of colorectal carcinoma. Colloids Surf B Biointerfaces. 2013; 112: 237-44. 11. Shahdadi Sardo H, Saremnejad F, Bagheri S, Akhgari A, Afrasiabi Garekani H, Sadeghi F. A review on 5- aminosalicylic acid colon-targeted oral drug delivery systems. Int J Pharm. 2019; 558: 367-79. 17. Hussain, T., Ijaz, M., Shamim, R., Hussain, K., Abbas, N., Hussain, A., Griessinger, JA. and Bukhari, NI. In vivo evaluation of a novel chitosan-polycaprolactone based mucoadhesive gastro-retentive sustained release drug delivery system for milnacipran HCl. AAPS PharmSciTech, 2020; 21(2): 1-11. 12. Entezar-Almahdi E, Mohammadi-Samani S, Tayebi L, Farjadian F. Recent Advances in Designing 5- Fluorouracil Delivery Systems: A Stepping Stone in the Safe Treatment of Colorectal Cancer. Int J Nanomedicine. 2020; 15: 5445-58. 18. Kumar A, Gulati M, Singh SK, Gowthamarajan K, Prashar R, Mankotia D, et al. Effect of co-administration of probiotics with guar gum, pectin and eudragit S100 based colon targeted mini tablets containing 5- Fluorouracil for site specific release. Journal of Drug Delivery Science and Technology. 2020; 60: 13. Gioumouxouzis CI, Chatzitaki AT, Karavasili C, Katsamenis OL, Tzetzis D, Mystiridou E, et al. REFERENCES Controlled Release of 5-Fluorouracil from Alginate Beads Encapsulated in 3D Printed pH-Responsive Solid Dosage Forms. AAPS PharmSciTech. 2018; 19 (8): 3362-75. 14. Ge M, Tang W, Du M, Liang G, Hu G, Jahangir Alam SM. Research on 5-fluorouracil as a drug carrier materials with its in vitro release properties on organic modified magadiite. Eur J Pharm Sci. 2019; 130: 44-53. 19. Nair KL, Jagadeeshan S, Nair SA, Kumar GS. Biological evaluation of 5-fluorouracil nanoparticles for cancer chemotherapy and its dependence on the carrier, PLGA. Int J Nanomedicine. 2011; 6: 1685-97. 15. Krishnaiah YSR, Satyanarayana V, Dinesh Kumar B, Karthikeyan RS, Bhaskar P. In vivo pharmacokinetics in human volunteers: oral administered guar gum-based colon-targeted 5-fluorouracil tablets. European Journal of Pharmaceutical Sciences. 2003; 19 (5): 355-62. 20. Singh P, Tyagi G, Mehrotra R, Bakhshi AK. Thermal stability studies of 5-fluorouracil using diffuse reflectance infrared spectroscopy. Drug Test Anal. 2009; 1 (5): 240-4. 72
W2312997137.txt	https://zenodo.org/records/1449336/files/article.pdf	en		Geological Results of the Mount Everest Expedition, 1921	Geographical journal	1,922	public-domain	7,648	Geological Results of the Mount Everest Expedition, 1921 Author(s): A. M. Heron Source: The Geographical Journal, Vol. 59, No. 6 (Jun., 1922), pp. 418-431 Published by: The Royal Geographical Society (with the Institute of British Geographers) Stable URL: http://www.jstor.org/stable/1780634 . Accessed: 06/02/2015 14:54 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. . The Royal Geographical Society (with the Institute of British Geographers) is collaborating with JSTOR to digitize, preserve and extend access to The Geographical Journal. http://www.jstor.org This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions GEOLOGICAL 418 RESULTS OF badly lame, and that if we stayed where we were discovery would be certain. The only course left us was to " east" part of the loads and to trust to We left behind all we could dispense with, redistributed the loads, luck. and stumbled on into the night. The rest was a nightmare of confusion, of tumble and scramble, of bruised feet, and of falling about for want of sleep, until we got down into the Wadi Araba just as the moon rose behind us over the hills which we had been clambering down for the last five After getting across the Araba depression (here 12 miles wide) hours." they experienced the same difficulty in the hills opposite; the guide came to the end of his knowledge and was sent back. But eventually they struck a well-marked track which was supposed to be the Ma'an-Kuntilla over twenty hours on end, route; then they camped after "completing with only two halts.w The next morning "a search with the glasses revealed three white objects like buildings to the north; we bore down on them, and found them to be the police post of Kuntilla." Shakespear passed on to Suez, Port Said, and Europe a contented man, having achieved his ambition, and having added his jot (no smail one) to the sum-total of human knowledge. RESULTS GEOLOGICAL OF EXPEDITION, A. M. Heron, D.Sc, Read at the Afternoon F.G.S., THE MOUNT EVEREST 1921 Geological of India 1922. Map Survey Meeting of the Society, 10 April follows p. 480. area geologically examined consists of over 8000 square miles, TPIE included within a rectangle some 120 miles from east to west and This corresponds with the Tibetan portion 70 miles from north to south. of the drainage area of the Arun river, a complicated system of valleys which unite to form the Arun before it breaks through the main Himalayan The headwaters of the range in the impressive gorge below Kharta. and the Bhutia Kosi Chu above (P6 Chu) Rongshar Nyenyam were also examined. The southern watershed is the line of great snowy peaks running from the Khombu pass south-eastwards through Everest and Makalu to the Arun, and to the east of the Arun is the continuation of the range which divides Sikkim from Tibet; a range which lies considerably to the north of the great Kangchenjunga group of peaks. The northern watershed is the extension of what has been termed the there this is hardly a Ladak or northern range of the Central Himalaya; definite range but rather a broad belt of high and much-dissected country, with a few peaks of over 20,000 feet distributed without linear arrange- This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE ment. To MOUNT the north EVEREST of this watershed EXPEDITION, short 1921 tributaries 419 drain to the Brahmaputra (Tsangpo). for the I am greatly indebted to the promoters of the Expedition HowardK. C. privilege of accompanying it, and in particular to Colonel Bury, d.s.o., the leader, for much assistance and practical interest in my to the westward, of Sir Henry work, which is virtually a continuation, Tibet Expedition of 1903-4. the Hayden's pioneer investigations during With the exception of Sir Henry Hayden no geologist had visited this part of Tibet. The geological mapping was done on a scale of \ inch to 1 mile, on maps furnished by the topographical surveyofs as their planetabling proceeded. My very cordial thanks are due to Major H. T. Morshead, r.e., d.s.o., in charge of the Survey of India detachment, for by many such facilities given and for valuable information, accompanied Over a oonsiderable specimens, from localities which I could not visit. portion of the area, however, my work had to proceed in advance of the surveys; geological boundaries in such cases were drawn on the maps It subsequently from memory, supplemented by sketch-maps and notes. was not considered advisable to arouse the suspicions of the Tibetans by too close and prolonged examination of any particular area, and the movements were unfavourable to general conditions of the Expedition's detailed work, so I endeavoured to traverse as large an area of Tibet as possible, and to lay down on the map with fair accuracy the boundaries of the different formations where they were accessible. A considerable amount of interpolation was however, necessary, and my work must be skeleton considered as a reconnaissance and nothing more. If I have the good fortune to accompany the second expedition I hope to examine more carefully the crystalline area in the neighbourhood of Mount Everest, with the assistance of Major Wheeler's map, constructed from photographic surveys on a scale of 1 inch to 1 mile, and to east some of the banded light on such problems as the origin and constitution biotite gneiss and its relationships with the metamorphosed sedimentaries. The quarter-inch map was on too small a scale to be of use in the mapping of a crystalline complex. Geologically the area is divided into two divisions: Tibetan and sediThis mentary to the north, and Himalayan and crystalline to the south. is clearly displayed in the topography resulting from the distinction underlying geological structure, for to the north we have the somewhat tame rounded and lumpy mountain ranges of Tibet, with their broad and flat-bottomed valleys, contrasting with the higher, steeper, and more rugged Himalayas on the south. The Expedition found nothing of economic interest. Stones showing the green staining of copper compounds were now and again seen on A few clear moraines, but beyond that I saw no signs of mineralization. fragments of pink tourmaline and garnet were picked up by the coolies, This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions GEOLOGICAL 420 RESULTS OF I panned but none were sufficiently free from flaws to be worth cutting. the gravels in several places for gold, but without getting a colour. The River Systems. The two main branches of the Arun river, the Bhong Chu (or Men Chu, as it is called in its upper portion) and the Yaru Chu (Ko Chu) flow from the west and the east respectively, in a general east-and-west direc? tion, uniting near the village of Lashar and then flowing south-westwards and southwards through the main Himalayan range. The Yara Chu rises in the hills to the north of Kampa Dzong and meanders through the broad plain which here lies at the northern foot of the snowy range, until at Sar This it meets a high spur of crystalline rocks projecting northwards. deflects it in a great sweep to the north-east and it finally cuts through the toe of this spur in the Rongme gorge instead of flowing round its end. The Men Chu rises on the northern slopes of Gosainthan, above the Pekhii Tang, a great plain which contains a basin of enclosed drainage, On leaving the plain it finds its way along a valley the Pekhii Tso. excavated in a syncline of Cretaceous limestones, leaves that and cuts across the intervening Jurassic shales in a fine gorge to another parallel limestone syncline, and after some 16 miles along it is deflected back again to the original syncline by a north-south barrier of pegmatite veins and hard slates. Along this it then flows as the Bhong Chu for between and 60 miles to near its junction with the Yaru. Two of its more 50 Lo northern the and the Shi Chu Chu, also have tributaries, important bands of softer their courses largely determined the of by presence, Parallel to the Bhong Chu and joining the Arun Cretaceous limestones. 20 miles below the confluence at Lashar is the Dzakar Chu, which, with its tributaries the Ding Chu and the Neo Chu, drains the mountainous In these also the synclinal origin of the valleys is district of Pharuk. The main drainage lines are therefore parallel to and dependent distinct. on the folding to which the region has been subjected ; the general strike of the folds is W.N.W.-E.S.E. At approximately drainage system right angles to the longitudinal Those from the northern slopes are a number of transverse tributaries. are turbulent glacial torrents, with straighter of the Great Himalaya Of courses and greater discharge than those from the Ladak range. the latter the more important occupy valleys intervening between tracts of high land which owe their prominence to their being composed of shales with clusters of intrusive hardened and partly metamorphosed held up by moraine dams, the tarns Except for glacial granite veins. at either of is devoid Arun region end, however, are basins of lakes; to the east, and to the west enclosed drainage, that of the Tsometretung that containing the Pekhii Tso, the Kharru Ochen Tso, and the Khumen All these are very shallow and vary greatly in extent according Tso. In the broader valleys are extensive swamps to the season of the year. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions ENTRANCE TO THE YO RI GORGE OF THE ARUN RIVER Phot. A. F, i?. Wollaston. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions o o o ?o > o D < H O This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE MOUNT and tracts temporarily are in fact little more. EVEREST flooded EXPEDITION, 1921 during the rains, and the so-called 421 lakes Changes in drainage lines. is little doubt that the Arun has cut back through the Great Himalaya range and has captured a river which possibly flowed east from the vicinity of Gosainthan more or less along the present courses of the Men Chu and the Bhong Chu and then through the Jikkyop gap and over the plain to the south of Kampa Dzong; this river may even, as Hayden (Mem. Geol. Surv. Ind., 36, pt. 2, p. 8) suggests, have flowed northward to join the Tsangpo, perhaps on the line of the The Dzakar Nyang Chu, the river which passes Gyantse and Shigatse. Chu, now also captured by the Arun, probably joined the above confrom on a course approximately jectural river flowing north-eastwards the Kiiyok La, above Lungme, along the present valley of the Arun between Kharkhung and Lashar. The Arun has two gorges. The lower, in which the river falls 3000 feet in the 18 miles, measured in a straight line, between Kharta and Kyimateng, is fairly straight, with walls rising 5000 feet and more in uninterrupted slopes so steep as to prevent human passage but The upper gorge is allowing bushes and trees precarious roothold. an extraordinary one, and so far I am unable to give an explanation of its origin. Where it enters the gorge the river is flowing through a fairly open valley with immense terraces of boulders and gravel, in the direction of the Kiiyok La, a low pass over comparatively soft schists. Abruptly the river turns upon itself and then plunges at a right angle into the heart of a high mountain (Y6 Ri) of hard gneiss, in a gloomy cafion with almost vertical walls. Through this gorge the river flows south for 3 miles, then swings again and flows west for 4 miles, finally emerging from the gorge on the other side of the Kiiyok La, into an open valley, which has exactly the same line and character as the Thus it cuts along two sides of a triangle in hard gneiss, original valley. in preference to along the hypotenuse in soft schists. The Rongme gorge on the Yaru Chu is somewhat similar, as the stream now cuts through the end of a northward-trending spur of gneiss and adjacent hard phyllites. It seems probable from the configuration of the country that the Yaru once flowed through the Jikkyop gap 4 miles to the north, the present course of the Chiblung Chu, and has been captured by a tributary from the east. There I was able this year to devote vicinity of glaciers, but it is advantage of a large-scale map on particular accuracy, I may be able with the investigation concurrently the only an occasional day or two to hoped that next year, with the which glaciers will be shown with to carry on glacial observations of the crystalline area, to which This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions GEOLOGICAL 422 RESULTS OF I am able to add my testimony to that of Hooker, they are confined. Blanford, Hayden, Garwood, and others, as to the former much greater extension The present glaciers are but puny repreof glaciation. sentatives of their former selves, as shown by the huge moraines Two at least of the main which encumber all the northern valleys. of glaciers of Makalu, flowing to the Kama valley, show evidences recent advance. The Himalayan Zone. of a and crystalline zone is essentially composed Himalayan biotite gneiss, usually garnetiferous, and banded intimately granite injected with dykes and sills of all sizes of a schorl-muscovite or pegmatite. The latter is often present to such an extent that it is the zone between the gneiss Forming an intermediate predominant rock. is a band of metamorphic rocks, altered and the Tibetan sedimentaries of the latter : these are also penetrated by the schorlrepresentatives in They appear to lie upon the gneiss, which granite great profusion. which is probably intrusive in them; but this point is one which I hope The foliated Other questions to settle definitely if I accompany the second expedition. which arise are, to what extent the gneiss represents very highly metaand to what extent it is an injection gneiss morphosed sedimentaries, formed by the intrusion and rolling out of granite veins along the of mica-schists. Although much the same in mineral constituents, the biotite gneiss varies so greatly in proportion, in structure, and in texture that it is Much of it is undifficult to believe that it all has the same origin. " the for from augen porphyritic example, granite, as, doubtedly derived near gneiss" type, and a less common variety, found in large amount with abundant biotite of foliae rather and thin sparse Kharta, showing In the Kharta and Dzakar felspar, forms lenticles twisted and contorted. around that resembles the Darjeeling, in which dark and gneiss valleys and biotitic felspathic respectively, alternate, and form a rock light bands, which from a little distance has the appearance of a bedded sedimentary the planes of foliation or banding have As near Darjeeling, series. Low down in this and low variety is notably garnetiferous. dips, usually some of the valleys towards the Nepal frontier, as for instance below bodies of Nyenyam and Tazang, and also probably near Kyimatang, large mica-schist are found, analogous to the schist occurring in the bottom of foliation the Teesta valley near Darjeeling and in other localities found underlying The latter have been mapped by Bose as the the gneiss of Sikkim.* the schist near the Daling Series; it is however quite uncertain whether of the modification is a or series Nepal frontier is an altered sedimentary gneiss. * Garwood, in Freshfield's ' Round Kangchenjunga,' p. 275 ; Mallet, Mem. Geol. Surv. Indiat II, pt. I ; Bose, Rec. Geol. Surv. India, 24, pt. 1, p. 46 ; pt. 2, p, 217. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE MOUNT EVEREST EXPEDITION, 1921 423 The schorl granite varies in texture from a fine homogeneous granite to a coarse porphyritic pegmatite, sometimes with graphic intergrowths of It is the latest in age of the igneous rocks, and quartz and felspar. occurs practically everywhere in the crystalline area, penetrating both The sill habit gneiss and metamorphics in veins and sills of all sizes. of the rocks is especially the with foliation concordant characteristic, intruded. Intrusion has taken place to such an extent that schorl granite is often seen to be the predominating rock, and not only so, but its lack of and cause it to resist weathering and and foliation toughness joints abrasion in moraines and streams, so that it nearly always is the main It has as accessory minerals quartz, constituent of detrital accumulations. black tourmaline plagioclase, (schorl) and muscovite, garnet, yellow and and pink tourmaline, beryl. variety of rocks, all of metamorphics comprise a considerable are certain banded or foliated in which, except quartzites, distinctly of with mineral directions determined by layers differing composition, the original stratification. on tour shortly after the As I proceeded return of the Expedition I have not had an opportunity of examining sections of these rocks. They range from quartzites fully microscopic and micaceous and to mica-schists tourmaline mica-schists, quartzites representing the arenaceous and argiilaceous sedimentaries, with crystal? line marbles and actinolite, schists from the tremolite, and epidote calcareous rocks. Graphitic schists have also been noted, but are rare. They are dislocated by profuse intrusion of vein granite, and their planes of folia? tion lie as a rule at comparatively low angles; this is striking in the field, with the intense crumpling which the same especially in comparison rocks have undergone in the Tibetan Zone. As one ascends any of the headwaters of the Dzakar Chu towards the Mount Everest group, one leaves the twisted and crumpled Jurassic shales and passes downwards in the section, as the general dip is northwards, though actually to the gently rolling limestones rising in elevation them, underlying which flatten out as they become more altered and the snowy range is In the Rongbuk valley, for instance, above the Chobu monastery, neared. are limestones much fissured and veined with crystalline calcite, underlain by a thick sill of schorl granite and pervaded by innumerable smaller sills anil streaks. Some 60 feet of the limestone immediately above the main sill has been converted into amphibole schist, and below the sill is a band of mica-schists thickly streaked and knotted with granite in lit-parlit injection, to such an extent that the result has a very strong resemblance to the banded variety of the biotite gneiss. In the gorge of the Dzakar Chu between Kal and Tsa is exposed a great thickness of flaggy lime? stones with clayey partings. At the base of the section there are great masses of schorl granite with amphibole and epidote calc-schists; upwards The This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions GEOLOGICAL 424 RESULTS OF the former become more definitely sill-like, insetbedded with calc-schists and finely crystalline and mottled limestones. remain The limestones for a horizon of the considerable above the distance crystalline topmost sill, and then pass upwards into black limestones, non-crystalline and calcite In the valleys veined, finally succeeded by Jurassic shales and quartzites. above Raphu and Chodzong alteration takes place independent of granite intrusions, calcite-veined, knotted and brecciated limestones passing downwards into tremolite, actinolite, and epidote schists. In the above described sections the change from sedimentary to metamorphic rock is very clearly seen, taking place gradually in magnificent clifF-faces with no break nor discordance in the stratification; from a short distance away it is indeed often impossible to say whether one is looking at limestone or calc-schist. Speaking generally, the valleys to the north-west and north of Everest, i.e. valleys above about 15,000 feet, are excavated in metamorphic rocks, whereas those to the north-east and east, for the most part below about in the time at my It was impossible, 15,000 feet, are in gneiss. disposal and with a small-scale skeleton map, to attempt to lay down a boundary between metamorphics and gneiss, but it would appear possible that the metamorphics form a gently northward dipping sheet underlain The gneiss is probably intrusive in the metamorphics, by the gneiss. judging from evidences of its age elsewhere in the Himalayas, and it may be possible to ascertain this definitely if I am allowed to accompany the second expedition. The group of high peaks between the Nangba La and the Rongbuk glacier and the north-western side of Everest itself up to the summit are of metamorphics, with of course much schorl granite, to the resistant is due the not the easily eroded metamorphics, and of which, power When I visited the Kharta and Kama valleys eminence of these peaks. on the east side of Everest before the end of the monsoon, the mountain was too much covered with fresh snow to show any geological structure. The base of Chomo Lonzo in the Kama valley is gneiss, but Col. HowardBury states that its upper portion is pale granite. Amphibolites. In the neighbourhood of Dak, in the Arun valley, numerous fragments both foliated and granitoid, were observed, but the of amphibolites, Their nature is therefore uncertain, but they parent mass was not found. are probably altered igneous rocks of intermediate or basic composition. The Tibetan Zone. The Tibetan Zone consists in the main of a great thickness of intensely folded Jurassic shales, the folds in general striking east and Pinched up in west, and repeated many times in complicated' fashion. these, in several very elongated and narrow synclines, are limestones belonging to the Kampa System of Hayden, of Cretaceous and Eocene This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE MOUNT EVEREST EXPEDITION, 1921 425 These synclines are closely compressed and overfolded, their axial age. force which planes dipping to the north, showing that the compressive produced them acted from the north. Along the southern border of the Tibetan Zone, below the base of the in which the Jurassic shales, is a great thickness of flaggy limestones, fossils have been destroyed and the rocks themselves converted in part into crystalline limestones and calc-schists. The age of these cannot be determined with certainty, but their character and position in the sequence indicate that they are possibly Trias or Permian. I am much indebted to my colleague, Mr. G. H. Tipper, for identifying for me the small collection of fossils which I made. From a palaeontoI was which covered the very disappointing. logist's standpoint country The Jurassic shales are almost unfossiliferous, and yielded only a few The thick and crinoid stems of little interest. ammonites, belemnites, limestones bordering the crystalline zone show, near their top, abundant signs of organisms in the form of curved layers of crystalline calcite, which in all probability are the remains of large lamellibranchs or brachiopods, but several days* search in favourable localities failed to discover a single This is par? specimen of which one could say anything more definite. to these rocks and of in view the interest unfortunate ticularly attaching their age. The Eocene and Cretaceous limestones, the zones of which have been worked out in great detail by Sir Henry Hayden in the magnificent and less disturbed sections of the Kampa ridge, here occur in much com? pressed synclines, in which fossils have been destroyed or damaged by the shearing which they have undergone, and in which zones are almost impossible to work out owing to faulting and interruption by stretches of It is only in the Tsipri ridge that a satisfactory detailed study alluvium. of the Eocene and Cretaceous rocks can be made, and this I was unable to give time for, as when I passed it I had been separated from the Ex? pedition by floods and had exhausted all my money and nearly all my It is, however, unlikely that I could have added anything of value food. to Sir Henry Hayden's description of these rocks. The Kampa System. The Kampa System is developed in two main synclines, the northern of which may be called the Tsipri syncline, from the picturesque and sacred ridge on it, and the southern the Bhong Chu, from the chief river of this area, which has excavated its valley along it; there are besides a number of small synclines. The Eocene beds above the " ferruginous sandstone" of Hayden (Mem. Geol. Surv. India, 36, Pt. z, pp. 44, 48-51) are found only in the northern syncline. In the exposures between the Yao La and Gutso this is a massive pink and white quartzite, about 100-150 feet thick, weathering into large In its degree of metamorphism it is a typical Pre-Cambrian biocks. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions 426 GEOLOGICAL RESULTS OF quartzite, although the brown shales below it and the blackish grits above and are almost unaltered; the latter contain fossil wood, dicotyledonous, The Tsipri ridge gives is the highest formation present in the section. the only fair sections of the combined Eocene and Tertiary of the Kampa System. I was unable to examine in detail, but the general section is as below : with abundant Bold scarp : Massive thick-bedded grey limestones, and Operculina, alternating with massive, white, very fineAlveolina grained and unfossiliferous limestones and thin-bedded limestones. Minor scarp : A series of limestone in regular beds of medium thick" ferruginous sandstone." ness; about the middle of this series comes the Undercliff of above : Grey flaggy limestones. Lower scarp rising from plain : Brown argillaceous limestones, in thin regular beds. Usually covered, but exposed at east end of ridge : Great thickness of grey unfossiliferous calcareous shales. Black and brown splintery shales with large East end of ridge: nodules. septarian North side of Shi Chu Valley: Grey limestone, massive quartzite, the " wall" quartzite. The upper limestones on the south side of the ridge are corrugated, and as they pass to the northern side dip steeply up to the vertical; further north, on the northern side of the Shi Chu valley, the limestones and quartzite at the base of the syncline are inverted, with the Jurassic shales overlying them and dipping to north at 300 to 8o?. The Shekar anticline formed to the north of the main hill shows a subordinate come down to At the western end the topmost limestones syncline. are seen in the and plain-level by a downward pitching of the syncline, a great and short ridge to the west of Temi with undulating dips overfold. is not so highly inIn the Tsipri ridge the ferruginous sandstone abundant it has durated in the Yeo La sections; spherical concretions of the little pebbles, iron oxide and is in certain layers finely conglomeratic, of of the size of buckshot, transparent quartz, quartzite of consisting various colours, and white chert. At the western end of the northern syncline, where it emerges from the alluvium of the Pekhu plain, the Cretaceous limestones in their upper and are thin bands of sandstone portion have numerous intercalated themselves distinctly arenaceous, indicating, with the occurrence of fossil wood in the Eocene grits above the ferruginous sandstone, the prevalence of shallower water conditions than obtain as one passes to the east. The structure is that of a recumbent isocline, of which both limbs dip north at 200 to 400, affected however by minor rollings and corrugations, and the northern margin is considerably altered by metamorphic connected with the granite intrusions of the Northern Range. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions agencies THE MOUNT EVEREST EXPEDITION, 1921 4*7 Locally the prominent sandstone quartzite band which is found else? where in the shales a little distance below the base of the limestones is This I call the " wall" quartzite. Here there is a passage into wanting. the Jurassic shales through shaly limestones. Just below these passage beds, at Menkhap Me and on the Lungchen La, fragments of ammonites of Upper Jurassic type, but not determinable with certainty, were found. East of Gutse and Menkhap Me a broad alluvium-filled river valley and a southward-trending of rocks and spur semi-metamorphic granite veins (the Burtra ridge) cut off this syncline, but there is little doubt that it is structurally continuous with that of Tsipri. This has been described above. It also is overfolded by pressure from the north. To the east of Shekar the outcrop of the syncline narrows, through the beds becoming more vertical, and as it swings to the north-east in the valley of the Lo Chu it flattens out again to a very recumbent isocline. A day's search in the Cretaceous beds round Shekar failed to yield a fossil; the beds appear to have been sheared to some extent, and are shattered and veined with calcite but not rendered crystalline. In the Lo Chu valley the shaly partings between the limestones are silvery from the production of sericite mica. The Bhong Chu syncline is also overfolded, but not to quite the same extent as the last. Like it, it extends to an unknown distance through the Pekhu plain to the westward. Where first encountered, where the Men Chu flows along a valley excavated in it, a wide plateau of undulating Jurassic shales lies to its south, on which is a shallow saucer-like syncline, " containing the wall" quartzite and a trifling thickness of limestone above it. At the edge of this plateau the shales and the " wall" quartzite roll steeply over into the Men Chu valley. On the northern bank of the stream is a fine scarp of reguiarly bedded limestone, in places crowded with small lamellibranchs (unidentifiable) and what appear to be casts of brachiopods in crystalline calcite. To the north of this a double fault is well seen, bringing this limestone against the " wall" quartzite and the Jurassic shales, these dipping vertically where meet. they Between Nelung and Tinkri, where the Bhong Chu again retums into and excavates a valley along the syncline, both limbs dip northward at about 6o?. From Tingri eastwards to where it disappears near Tsonga, its southern limb is fairly regular and the " wall" quartzite stands up conspicuously along the valley, dipping at angles of 450 to 8o?. Its boldness and continuity along this valley led me to give to this distinctive bed the field name which I have used here. It is about 120 feet thick; next above it is a thin but massive limestone, and then 300-400 feet of shales, passing into the slabby limestones which form the bulk of the visible section. The northern edge is not so regular; usually it is overfolded, but in places the dip is high but normal, and south of Shekar runs a strike fault, This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions 428 GEOLOGICAL RESULTS OF South of the Tsipri ridge the two cutting out the "wall" quartzite. synclines approach closely, with an intervening anticline of Jurassic shales All along the Bhong Chu valley exposures of the separating them. Cretaceous limestones are much disconnected by detrital deposits, and usually occur as isolated hills of bizarre form in which the beds are seen to be intensely crumpled and sheared, and fossils are represented by streaks of calcite. At Kyishong, near where it disappears, the syncline widens out owing to a subsidiary anticline rising up along its centre. The group of synclines to the south, in the Pharuk district, display great irregularities and complexity of structure, so much so that I found it impossible to map them in detail on a \|-inch scale, and have had to show them in a general and diagrammatic way. That which forms the valley of the Neo Chu and passes eastwards to near Aya is very elongated and narrow; the strata in it stand vertically or lie slightly overfolded in the usual direction, with the " wall" quartzite standing up on either side of the valley. Midway along its length a strike fault repeats it, bringing in a wedge of Jurassic shales between. At its western end it is continued by another similar syncline, not quite in line with it but parallel and a short distance to one side. In the 2 miles south of this, between Namda and Tashidzom, the " wall" quartzite and the basal beds of the Cretaceous limestones are repeated Needless again and again by sharp folds and faults of small throw. South to say they are veined with calcite and in places brecciated. of this again, from Tashidzom to Kuyul, besides the major double syncline shown on the map, small sections of the Cretaceous limestones are pinched up and faulted into the Jurassic shales. In the double syncline there is no inversion, the northern lobe being shallow, saucer-like and fairly symmetrical, while in the southern the beds are undulating and almost horizontal. rocks lies far to the The only remaining outcrop of Cretaceous north-east, a shallow syncline similar to the last, with the quartzite dipping gently inwards round the periphery and the centre occupied by horizontal and undulating sericitic limestones. The Jurassic shales. most striking features, in fact the only striking features, of the They Jurassic beds are the extent and the monotony of their outcrops. consist for the most part of dark brown and black shales and argillaceous the purer type of sandstones, with subordinate quartzites, representing more argillaceous darker and which are and limestones usually sandstone, than those of the overlying Cretaceous System. In the tract of country between the crystalline zone and the Northern the Jurassic strata are thrown into Range of the Central Himalaya, great folds and corrugated in the most fantastic fashion, and even where the general dip approaches horizontality the beds roll about irregularly. The This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE MOUNT EVEREST EXPEDITION, 1921 429 In such highly compressed thrust-faulting, country faulting, especially must be very prevalent, but where strata are so uniform in appearance it is extremely difficult to detect. The general strike of these folds is that of the "grain" of the country, i.e. in an east-west or E.S.E.-W.N.W. than in the direction, but they are subject to far more irregularities more persistent synclines of the Kampa System limestones. In the Northern Range, and also where they pass downwards into the thick limestones along the boundary of the crystallines, the shales dip less A certain amount of injection by granite variably and at lower angles. veins has taken place in the Northern Range, along with a widespread regional induration of the rocks, attaining however only a low degree of The intermediate belt, where the Cretaceous and Eocene metamorphism. and the have been compressed limestones into overfolded synclines Jurassic shales have been so intensely folded, has been a region of The alteration of the rocks weakness between two more resistant blocks. in the Northern Range extends considerably further outwards from the areas of granite intrusion than in the opposite section of the Great Pebbles of garnetiferous mica-schist and Himalaya, but is of less degree. hornblende schist (of the " feather amphibolite type ") were found in gravels below the Mon La, but the parent rock was not found in situ, nor were such highly metamorphosed types met with elsewhere. Igneous Rocks in the Tibetan Zone. The intrusive granite of the Northern Range is very similar in appearance to the schorl granite of the Himalayas, but is uniformly finegrained instead of showing the great variation in texture of the latter Like it, it is a white rock and is very tough and resistant to rock. it differs from the schorl granite in that it weathering. Mineralogically contains biotite (with muscovite as well), instead of schorl. Near Nelung, Khakyu, and between Namda and Aya, small dykes of dark rock were seen, in the last ease strung out along a line running east and west, appearing at intervals over a length of 2J miles. The dykes run 100 less or and are to feet, more, individually only up 3 feet wide. The rock is too thoroughly decomposed for determination, but is probably basic. Judging from the crushing and dislocation which the dykes have undergone, they are probably antecedent in age to the folding of the rocks. For the most part the shales have been hardened and have acquired the beginning of slaty structure, being knotted and breaking into prisms, or have developed inthem a certain amount of secondary sericite mica and aluminous silicates such as staurolite, and have in certain cases become phyllites. Often they have a baked appearance, being whitish or red, contrasting with the black or rusty brown tints of the unaltered The quartzites show no more alteration than they do amongst shales. the unaltered strata, but then in this area the usual Jurassic sandstone quartzite, fairly free from impurities, is just as hard and vitreous as any typical Pre-Cambrian quartzite. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions 43? Permo-Trias GEOLOGICAL RESULTS OF Limestones. Between the crystalline and the sedimentary zones is the outcrop of a thick series of limestones, of which 2000 or 3000 feet are exposed in a very uniform assemblage of rather thin beds of 1 to 3 feet in thickness, with shaly partings. The overlying shales, of which the major portion has been shown by Hayden to be Jurassic, pass down without any visible As has been stated, the limestones as a discordance into the limestones. all fossils having been destroyed and whole are considerably altered, More than this, they now appearing as streaks of crystalline calcite. have been extensively invaded by granite veins, converted into crystal? line limestones and calc-schists, and involved in the crystalline complex in such fashion that to lay down a true boundary on the map is impossible. and crystallines is The line which I have drawn between limestones arbitrary, and represents generally the upper and outer limit of granite intrusions: to the south of this line there is much of the limestone in the with the schorl granite. but intimately associated form of calc-schists, and their are thus obliterated The lowest portions of the limestones relation with the biotite gneiss is obscure, but it is probable that the were probably continuous The limestones latter is intrusive in them. the of the southern exposures, but from point Jurassic margin along right to point the zone of metamorphism and granite veining has encroached on them to a varying extent, in some places affecting them throughout and transgressing upwards as far as the Jurassic shales, and in others leaving a great thickness unaltered, so that their outcrop has now the irregular breadth shown upon the map. Their general dip is northward at low angles; at Yalep on the P6 Chu and at Kal are anticlinal dips undulate somewhat. flexures, and south of Raphu and Hlelung of north the The bifurcation of Tulung is, so far as I was able outcrop limestones the to to ascertain, due emerging again to the north of the The structure is, however, exposures along an anticlinal axis. of this portion of examination to be due and My faulting. doubtful, may and snowfalls hindered much was heavy mist. the area by repeated but is rocks these of The age may be put down provery doubtful, Geol. Surv. Ind., Sir Permo-Trias. as (Mem. Henry Hayden visionally the Dothak Series, of name the under has described, 36, pt. 2, p. 21) between the rocks other and of limestones sedimentary an assemblage Chumbi valley and Bhutan, which in his opinion may include part or main all of the Trias and possibly one or more of the Palaeozoic systems. He also suggests that Triassic rocks occur along the northern slopes of the Lhonak range between Tibet and Sikkim (Zoc. cil., pp. 23 and 24), and fossils typical of the Productus Shales (Upper Permian) are known from near the Kongra La, the pass which to have been collected The situation of crosses the Lhonak range south of Kampa Dzong. is zone the to very similar to that crystalline these exposures with regard area. the in of the limestones present This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions THE MOUNT EVEREST EXPEDITION, 1921 : DISCUSSION 431 Direct evidence of their age, though not very definite, is given by two sections in the ridges to east and west of Hlelung. At the base of the great series of shales which overlie the limestone, just as they pass downwards into the latter, is a thin ferruginous bed crowded with Spirifer and Productus, not however specifically These would determinable. indicate that the top of the limestones is about Upper Permian in age, if the section is a straightforward one, which there is no reason to doubt. The bulk of the limestones would then represent the Permian of the European scale, with perhaps a portion of the Carboniferous. Judging from field relationships and lithological characters, I had in my own mind considered these limestones as approximately equivalent to the Kioto Limestone of the Zangskar range in Spiti (Lower Jurassic and Upper Trias), which in that country underlies the Spiti shales (Upper Jurassic), but the fossil evidence puts them much lower in the geological scale, and indicates that the Trias is represented by the lower portion of the of shales; it is unfortunate that the absence of regreat succession cognizable fossils from the limestones themselves leaves the question so indefinite. Before the paper the President said : The paper this afternoon is one by Dr. Heron, of the Geological Survey of India. He accompanied the Mount Everest Expedition last year on the reconnaissance of the mountain, and has given us the result of that reconnaissance from the geological point of view. He is very energetic, and managed to cover a great deal of country in the region of Mount Everest. Though I do not believe he actually got to the top of the mountain, he did, however, what is from a geological point of view much more interesting?he explored all the region north of Mount Everest towards the Brahmaputra, and also to the east of it. Dr. Heron, unfortunately for us, is not able to be present as he is still out in India. He was hoping to rejoin the Mount Everest Expedition this year also, but I am not sure that he has been able to do so. We are, however, fortunate in having with us Mr. Fox, who also belongs to the Geological Survey of India, and has worked for the Survey in the Himalayas. Mr. C. S. Fox, of the Geological Survey of India, gave an account of the above paper by Dr. Heron, and added : The remarkable parallelism between the trough of the Tsangpo and the axis of the main Himalayan range has always been taken to indicate a common origin for both, that is, crust squeeze or pressure at right angles to these features. The same forces would account for the closely corrugated and crumpled sedimentary beds, and for the great double synclinal sag of the Cretaceous and Tertiary strata. It is, however, difficult to explain thetectonics concerning the metamorphism of the sedimentary rocks at the junction of the Tibetan and Himalayan zones. There are certain peculiarities in the valleys of the Men Chu, Bhong Chu, and Dzakar Chu which require comment. These streams occupy synclinals or structural troughs of Cretaceous strata for part of their courses. Such structural courses are as a rule uncommon. Most streams in mountainous country occupy valleys carved along anticlinal folds. It is therefore thought that these above streams began their courses along anticlinal folds, but owing This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions map m. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions Hfe. This content downloaded from 128.235.251.160 on Fri, 6 Feb 2015 14:54:40 PM All use subject to JSTOR Terms and Conditions
https://openalex.org/W2516226176	https://gastro.zaslavsky.com.ua/index.php/journal/article/download/247/676	Ukrainian	null	Особливості мікробіоценозу товстої кишки та рівня фекального кальпротектину у хворих на хронічні запальні захворювання кишечника залежно від нутритивного статусу	Gastroenterologìa/Gastroenterologìâ	2,021	cc-by	4,017	ОСОБЛИВОСТІ МІКРОБІОЦЕНОЗУ ТОВСТОЇ КИШКИ ТА РІВНЯ ФЕКАЛЬНОГО КАЛЬПРОТЕКТИНУ У ХВОРИХ НА ХРОНІЧНІ ЗАПАЛЬНІ ЗАХВОРЮВАННЯ КИШЕЧНИКА ЗАЛЕЖНО ВІД НУТРИТИВНОГО СТАТУСУ ОСОБЛИВОСТІ МІКРОБІОЦЕНОЗУ ТОВСТОЇ КИШКИ ТА РІВНЯ ФЕКАЛЬНОГО КАЛЬПРОТЕКТИНУ У ХВОРИХ НА ХРОНІЧНІ ЗАПАЛЬНІ ЗАХВОРЮВАННЯ КИШЕЧНИКА ЗАЛЕЖНО ВІД НУТРИТИВНОГО СТАТУСУ Резюме. Стаття присвячена актуальній проблемі — стану мікрофлори товстої кишки та рівня фекального кальпротектину у хворих на хронічні запальні захворювання кишечника з різним нутритивним статусом. Проведено обстеження 76 пацієнтів із хронічними запальними захворюваннями кишечника, серед яких 28 хворих на хворобу Крона та 48 — на неспецифічний виразковий коліт. Проаналізовано отримані дані залежно від нозології та нутритивного статусу. Встановлено, що серед хворих досліджених груп пере- важали пацієнти з субкомпенсованою формою дисбіозу. Зниження концентрації біфідобактерій у вмісті товстої кишки у хворих ІІІ групи виявляли найчастіше — у 78,9 % випадків. Дефіцит лактобактерій най- частіше спостерігався у хворих ІІ групи (92,3 %). Вміст кальпротектину був суттєво підвищеним у 84,2 % пацієнтів, позитивно корелював із ступенем тяжкості перебігу захворювання і не залежав від їх нутритив- ного статусу. Ключові слова: хронічні запальні захворювання кишечника, нутритивний статус, дисбіоз, мікрофлора кишечника, кальпротектин. Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology УДК 616.345-008.1+612.86/577.486 ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧАН О.В. Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна ГАСТРОЕНТЕРОЛОГIЯ GASTROENTEROLOGY Îðèã³íàëüí³ äîñë³äæåííÿ Original Researches Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology УДК 616.345-008.1+612.86/577.486 ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧАН О.В. Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна ГАСТРОЕНТЕРОЛОГIЯ GASTROENTEROLOGY Îðèã³íàëüí³ äîñë³äæåííÿ Original Researches ГАСТРОЕНТЕРОЛОГIЯ GASTROENTEROLOGY Îðèã³íàëüí³ äîñë³äæåííÿ Original Researches Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology УДК 616.345-008.1+612.86/577.486 ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧАН О.В. Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна УДК 616.345-008.1+612.86/577.486 ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧАН О.В. Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна УДК 616.345-008.1+612.86/577.486 УДК 616.345-008.1+612.86/577.486 ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧА Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна СТОЙКЕВИЧ М.В., ВІННИК Н.В., ТАТАРЧУК О.М., СТОЙКЕВИЧ М.В., КУДРЯВЦЕВА В.Є., СОРОЧ Лабораторія мікробіології та імунології, ДУ «Інститут гастроентерології НАМН України», м. Дніпропетровськ, Україна КУДРЯВЦЕВА В.Є., СОРОЧАН О.В. Вступ ся невідомою, а механізми патогенезу — недостатньо з’ясованими. ХЗЗК вважаються багатофакторними за- хворюваннями з генетичною схильністю, що дозволяє Хронічні запальні захворювання кишок (ХЗЗК) за- лишаються однією з найбільш вагомих проблем гастро- ентерології і становлять близько 10,0 % всієї хронічної патології органів травлення. Безперервний рецидиву- ючий перебіг цих захворювань із прогресуючим пору- шенням структури та функцій слизової оболонки ки- шечника супроводжується значним зниженням якості життя хворих, що ставить ХЗЗК у ряд важливих меди- ко-соціальних проблем і передбачає пошук нових на- прямків у вивченні етіології й патогенезу, удосконален- ня профілактики та лікування цієї патології [1]. Адреса для листування з авторами: Вінник Н.В. Лабораторія мікробіології та імунології, ДУ «Інститут гастро- ентерології НАМН України»; просп. Слобожанський, 96, м. Дніпропетровськ, 49074, Україна Е-mail: gastrodnepr@i.ua © Вінник Н.В., Татарчук О.М., Стойкевич М.В., Кудрявцева В.Є., Сорочан О.В., 2016 © «Гастроентерологія», 2016 © Заславський О.Ю., 2016 Лабораторія мікробіології та імунології, ДУ «Інститут гастро- ентерології НАМН України»; просп. Слобожанський, 96, м. Дніпропетровськ, 49074, Україна Е-mail: gastrodnepr@i.ua Е-mail: gastrodnepr@i.ua © Вінник Н.В., Татарчук О.М., Стойкевич М.В., Кудрявцева В.Є., Сорочан О.В., 2016 © «Гастроентерологія», 2016 © Заславський О.Ю., 2016 Незважаючи на велику кількість досліджень, при- свячених ХЗЗК, етіологія неспецифічного виразко- вого коліту (НВК) та хвороби Крона (ХК) залишаєть- 21 № 2 (60), 2016 www.gastro.org.ua Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology Ï реалізуватися досі ще невідомим уражуючим агентам. Вторинні ефекторні механізми, що виникають на фоні симбіозу генетичних і зовнішніх факторів, призво- дять до розвитку неспецифічного імунного запалення в слизовій оболонці кишок — поверхневого при НВК і трансмурального при ХК [2, 3]. реалізуватися досі ще невідомим уражуючим агентам. Вторинні ефекторні механізми, що виникають на фоні симбіозу генетичних і зовнішніх факторів, призво- дять до розвитку неспецифічного імунного запалення в слизовій оболонці кишок — поверхневого при НВК і трансмурального при ХК [2, 3]. Для пацієнтів із запальними захворюваннями ки- шечника в гострій фазі характерний розвиток недо- статності харчування, що проявляється зниженням маси тіла за рахунок як жирової, так і м’язової маси, негативним балансом азоту, гіповітамінозами, дефіци- том заліза, кальцію та інших мікроелементів [12, 13]. Згідно з сучасними уявленнями, головну роль у ме- ханізмах розвитку НВК і ХК відіграють порушення імунітету з дисбалансом цитокінової регуляції. Окрім того, широко обговорюється й роль мікробних фак- торів у розвитку захворювань, зокрема ешерихій, кло- стридій, ієрсиній, кишкового кампілобактера, псевдо- туберкульозних мікобактерій, а також низки вірусів, у тому числі вірусу кору. Однак численні спроби довести інфекційну природу ХЗЗК не мали успіху. Вступ Проте роль кишкової мікрофлори в їх розвитку не заперечується, але не як етіологічного фактору, а як одного зі сти- мулів, що призводять до ініціації запального процесу, активуючи синтез макрофагами первинних медіаторів запалення. Результати численних робіт із мікроекології людського організму доводять, що дисбіотичні зміни травного каналу, які можуть бути частіше вторинними, ніж первинними, поглиблюють основне захворюван- ня, а деколи набувають превалюючого значення в па- тології, впливають на тяжкість і тривалість основного захворювання [2, 4]. Порушення нутритивного статусу у хворих на НВК та ХК обумовлено рядом причин: Порушення нутритивного статусу у хворих на НВК та ХК обумовлено рядом причин: — втратою нутрієнтів, води та електролітів із ча- стим випорожненням; — зменшенням всмоктувальної поверхні слизової оболонки внаслідок запального процесу або резекції частини кишечника; — ферментативною недостатністю (синдром мальабсорбції, синдром «короткої кишки»); — обмеженням харчування за рахунок больового синдрому та інтоксикації; — порушенням моторики кишечника; — підвищеною втратою харчових речовин (хроніч- на крововтрата, ексудація кишечником білків плазми крові); — підвищенням енергозатрат у зв’язку з розвитком системного запального процесу (підвищення темпера- тури тіла, збільшення частоти пульсу та дихання, син- тез білків «гострої фази»); — синдромом надлишкового бактеріального росту [14, 15]. В основі патогенезу ХЗЗК є порушення імунної від- повіді, що призводить до розвитку неспецифічного запа- лення в стінках та слизової оболонки кишки. Дія імунних комплексів і медіаторів запалення (цитокінів, гістаміну, активних форм кисню, NO) на клітини стінки кишечни- ка сприяє її ушкодженню та деструкції тканини. Основними наслідками недостатності харчуван- ня є ускладнення перебігу основного захворювання, пролонгування фази загострення та, як наслідок, тривале перебування хворого в стаціонарі, збільшен- ня частоти інфекційних ускладнень через вторинний імунодефіцит, а також зниження якості життя па- цієнтів [16]. Кальпротектин сімейства S100 протеїнів уперше був виявлений I. Dale і співавт. (1983) у цитоплазмі грануло- цитів як протеїн, що чинить протимікробну дію. Він та- кож має бактеріостатичну, противірусну та фунгіцидну дію. Ряд авторів вважають, що фекальний кальпротек- тин (ФК) інгібує ріст мікроорганізмів за рахунок кон- курентного зв’язування цинку. Кальпротектин стано- вить близько 60 % від загальної маси солютабного білка цитоплазми нейтрофілів людини й також локалізується в моноцитах, макрофагах і епітеліальних клітинах [5– 8]. Після зв’язування з кальцієм стає стійким до високої температури та до розщеплення під впливом лейкоци- тарних і мікробних ферментів [9, 10]. Виділення з ка- лом ФК — білка нейтрофілів, який становить частину запального інфільтрату при ХЗЗК, відображає перехід клітин запалення в просвіт кишечника. Вступ Встановле- но, що його концентрація корелює з інтенсивністю нейтрофільної інфільтрації слизової оболонки кишеч- ника, у зв’язку з чим ФК і був запропонований як неін- вазивний маркер запалення кишечника [5–8]. Було доведено, що при зниженні маси тіла хворого за час перебування в стаціонарі на 5–10 % тривалість госпіталізації збільшується у 2 рази, а частота усклад- нень зростає в 3 рази та більше. Метою дослідження є визначення рівня ФК та дисбіотичних порушень мікробіоценозу товстої кишки у хворих на ХЗЗК залежно від нутритивного статусу. Матеріали та методи Обстежено 76 хворих на ХЗЗК, які перебували на лікуванні у відділенні захворювань кишечника Ін- ституту гастроентерології, з них 28 осіб мали ХК, 48 — НВК. Серед пацієнтів було 39 жінок та 37 чоловіків, середній вік яких становив (38,4 ± 1,33) року. Оцінка стану компонентного складу тіла пацієнтів здійснюва- лася за допомогою біоімпедансного аналізатора фірми «МЕДАСС» (Москва, Росія). На підставі визначен- ня відсотка жирової маси тіла за даними імпедансме- трії досліджені хворі були розподілені на три групи: І — зі зниженим нутритивним статусом (n = 44), ІІ — із нормальним нутритивним статусом (n = 13) та ІІІ — із підвищеним нутритивним статусом (n = 19). У дослідженнях Степанова, Татьяниной та ін. пока- зано, що у міру наростання ендоскопічної активності в пацієнтів як із НВК, так і з ХК підвищувався вміст ФК [6–8, 11]. Також показано, що максимальні зна- чення рівня ФК були у хворих із тяжким ступенем пе- ребігу захворювання, мінімальні — у хворих із легким ступенем перебігу [6]. Дослідження видового та кількісного складу мі- крофлори вмісту товстої кишки (ТК) проводили ме- 22 Гастроентерологія, ISSN 2308-2097 № 2 (60), 2016 Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology тодом посіву десятикратних розведень (10–1–10–9) фекалій на стандартний набір елективних і дифе- ренційно-діагностичних поживних середовищ для виділення аеробних і анаеробних мікроорганізмів. Оцінку глибини дисбіотичних порушень проводили з використанням градації за ступенем, що свідчить про амплітуду відхилень мікробіоценозу ТК: відсутність відхилень — еубіоз; наявність відхилень — дисбактеріоз (I ступінь — компенсований, легкий; II ступінь — суб- компенсований, середній; III ступінь — декомпенсова- ний, тяжкий). У процесі дослідження були виділені умовно-пато- генні ентеробактерії родів Enterobacter (7,9 %), Serratia (2,6 %), Klebsiella (21,1 %), Citrobacter (2,6 %), Proteus (10,5 %). Майже у третини хворих їх концентрація сяга- ла lg 7,5 — lg 8,9 КУО/г. У 35,5 % обстежених був підви- щений рівень дріжджоподібних грибів роду Candida ( lg 4,0 КУО/г). Крім того, у 10 % хворих висівали ге- молітичні біовари кишкової палички, яких за нормою не повинно бути у вмісті ТК, при цьому у половині ви- падків спостерігалось їх домінування над кишковою паличкою з нормальною ферментативною активністю. Аналіз досліджених показників за нозологіями ви- явив різницю у вираженості дисбіотичних порушень між хворими на НВК і ХК. Порівнювали середні загальні показники еубіозу, трьох ступенів дисбактеріозу та відхилення у складі мі- крофлори ТК загальної групи хворих на ХЗЗК, окремо по групах (хворі на ХК і НВК), а також по групах хво- рих на ХЗЗК залежно від нутритивного статусу. Матеріали та методи Розподіл хворих за глибиною дисбіотичних розладів показав, що у пацієнтів при обох нозологіях домінувала субкомпенсована форма дисбіозу: ХК (50,0 %) та НВК (45,8 %), тоді як компенсована та декомпенсована форми зустрічались з однаковою частотою: ХК по 25 % та НВК — по 27,1 %. Рівень кальпротектину в калі визначали імунофер- ментним аналізом (ІФА) з використанням тест-набору R-Biopharm (Німеччина). ІФА виконували за допомо- гою імуноферментного аналізатора Stat Fax 303 Plus (США). Як при НВК, так і при ХК виявили вірогідне зни- ження рівня біфідо- і лактобактерій. Однак у хворих на ХК частота виявлення цих порушень була вищою і дорівнювала 78,6 % проти 70,8 % (р < 0,001) та 89,3 % проти 83,3 % (р < 0,001) відповідно. Усі вихідні показники, отримані при виконан- ні роботи, з метою оптимізації математичної оброб- ки вводилися у базу даних, побудовану за допомогою електронних таблиць Microsoft Excel. Статистична об- робка результатів досліджень здійснювалася методами варіаційної статистики, реалізованими за стандартним пакетом прикладних програм Statistica for Windows 6.0. Підвищення концентрації дріжджоподібних грибів роду Candida у вмісті ТК переважало у хворих на НВК і становило 37,5 %. Але у хворих на ХК була декіль- ка вищою частота виділення гемолітичних біоварів кишкової палички (висівалася у 10,7 % хворих на ХК і у 8,3 % — на НВК). Також слід відзначити, що у 10,7 % випадків у хворих на ХК було виділено патогенний ста- філокок. Результати дослідження Проведені мікробіологічні дослідження вмісту ТК у хворих на ХЗЗК показали наявність глибоких змін якісного та кількісного складу мікрофлори у 100,0 % хворих. Серед досліджених хворих переважали пацієн- ти з субкомпенсованою формою дисбіозу (47,4 %). Дисбіотичні порушення головним чином були обу- мовлені різким зниженням чисельності основних сим- біонтів товстокишкового мікробіоценозу та зростан- ням концентрації умовно-патогенної мікрофлори. Так, вірогідне зниження рівня біфідобактерій спостеріга- лось у 56 (72,7 %) пацієнтів (р < 0,05), знижена концен- трація лактобактерій виявлена у 66 (86,4 %) пацієнтів (р < 0,001). Аналіз показників мікрофлори ТК у групах хворих з різним нутритивним статусом показав суттєві зміни стану мікрофлори ТК у досліджених групах і виявив різницю в розподілі хворих за глибиною дисбіотичних порушень (рис. 1). Відповідно до рис. 1, серед хворих досліджених груп переважали пацієнти з субкомпенсованою фор- мою дисбіозу: І група — 36,4 %, ІІ група — 76,9 %, ІІІ група — 52,6 %. Середні показники кількості мікроор- ганізмів у вмісті ТК і частота виявлених змін їх кон- 36,4 % 76,9 % 52,6 % ȱɝɪɭɩɚ ȱȱɝɪɭɩɚ ȱȱȱɝɪɭɩɚ ȱɫɬɭɩɿɧɶɞɢɫɛɿɨɡɭ ȱȱɫɬɭɩɿɧɶɞɢɫɛɿɨɡɭ ȱȱȱɫɬɭɩɿɧɶɞɢɫɛɿɨɡɭ 29,5 % 15,4 % 15,8 % 34,1 % 7,7 % 31,6 % Рисунок 1 — Стан мікробіоценозу ТК у хворих на ХЗЗК залежно від нутритивного статусу ȱɝɪɭɩɚ Рисунок 1 — Стан мікробіоценозу ТК у хворих на ХЗЗК залежно від нутритивного статусу 23 № 2 (60), 2016 Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology Ï Таблиця 1 — Кількість хворих зі зміненою концентрацією мікроорганізмів та її середні показники залежно від нутритивного статусу Мікроорганізми (показники концентрації) Кількість хворих зі змінами у складі мікрофлори (%) Середні показники концентрації (М ± m) (КУО/г) Групи І ІІ ІІІ І ІІ ІІІ Контроль Bifidobacterium (< lg 8,0 КУО/г) 72,7 69,2 78,9 6,24 ± 0,34* 6,74 ± 0,53* 6,34 ± 0,45* 9,58 ± 0,57 Lactobacillus (< lg 6,0 КУО/г) 86,4 92,3 84,2 2,88 ± 0,28* 2,48 ± 0,50* 3,16 ± 0,37* 6,93 ± 0,29 Candida ( lg 4,0 КУО/г) 31,8 23,1 47,4 1,60 ± 0,35 1,18 ± 0,63 1,87 ± 0,48 3,2 ± 0,3 Умовно-патогенні ентеробактерії ( lg 7,0 КУО/г) 22,7 15,4 26,3 1,68 ± 0,48 1,15 ± 0,78 1,85 ± 0,73 4,3 ± 0,3 Примітка. * — Р < 0,001 порівняно з контролем. я 1 — Кількість хворих зі зміненою концентрацією мікроорганізмів та її середні показники залежно від нутритивного статусу підвищений в 5,7 раза, у хворих на ХК — в 6,1 раза порівняно з контролем (рис. 2). центрації залежно від трофологічного статусу наведені в табл. 1. Висновки Серед хворих досліджених груп переважали пацієн- ти з субкомпенсованою формою дисбіозу. Зниження концентрації біфідобактерій у вмісті ТК у хворих ІІІ групи виявляли найчастіше — у 78,9 % випадків. Дефі- цит лактобактерій найчастіше спостерігався у хворих ІІ групи (92,3 %). При аналізі показника залежно від нозології вста- новлено, що у хворих на НВК його рівень вірогідно ɏɄ ɇȼɄ ɏɁɁɄ Ʉɨɧɬɪɨɥɶ ɦɤɤɝ 0 50 100 150 200 250 300 350 * * * Примітка. * — Р < 0,001 порівняно з контролем. Рисунок 2 — Концентрація фекального кальпро- тектину у хворих на ХЗЗК У 15,8 % хворих на ХЗЗК спостерігали помірне підвищення рівня ФК (від 50 до 120 мг/кг), у 84,2 % хворих вміст кальпротектину був суттєво підвище- ним, позитивно корелював зі ступенем тяжкості пе- ребігу захворювання і не залежав від їх нутритивного статусу. У 15,8 % хворих на ХЗЗК спостерігали помірне підвищення рівня ФК (від 50 до 120 мг/кг), у 84,2 % хворих вміст кальпротектину був суттєво підвище- ним, позитивно корелював зі ступенем тяжкості пе- ребігу захворювання і не залежав від їх нутритивного статусу. Результати дослідження За даними аналізу, суттєві відмінності у складі нормофлори вмісту ТК спостерігались у хворих усіх досліджених груп. Зниження концентрації біфідобак- терій у вмісті ТК у хворих ІІІ групи виявляли найчасті- ше — у 78,9 % випадків, тоді як у хворих І групи цей показник становив 72,7 % та 69,2 % — у хворих ІІ гру- пи. Дефіцит лактобактерій найчастіше спостерігався у хворих ІІ групи (92,3 %). Вірогідної відмінності рівня кальпротектину залеж- но від нутритивного статусу не виявлено. Рівень ФК був вірогідно підвищеним у хворих І групи до (295,30 ± 16,36) мг/кг, у ІІ — (306,10 ± 30,83) мг/кг та у ІІІ — (301,2 ± 27,9) мг/кг щодо контролю. Рівень кальпротектину позитивно корелював зі сту- пенем тяжкості перебігу захворювання (r = +0,24; р < 0,05), швидкістю зсідання еритроцитів (r = +0,33; р < 0,01) та негативно корелював зі ступенем недостат- ності харчування (r = –0,32; р < 0,01). Крім того, у ІІІ групі хворих вищою була часто- та виділення дріжджоподібних грибів роду Candida: на 15,6 % вище, ніж у І групі, та на 24,3 % — ніж у ІІ групі, а також частота виявлення умовно-патогенних ентеробактерій: на 3,6 та 10,9 % вище відповідно. Отримані дані співпадають із даними інших дослі- джень [1, 2, 4, 7] і вказують на можливість застосування ФК як маркера запалення кишечника. Рівень ФК був підвищеним у всіх досліджених хво- рих на ХЗЗК і коливався від 51,6 до 403,3 мг/кг. У 12 (15,8 %) хворих спостерігали помірне підвищення рів- ня (від 50 до 120 мг/кг), у 64 (84,2 %) хворих вміст каль- протектину був суттєво підвищеним. Список літератури 1. Inflammatory Bowel Disease diagnostic and therapeutic strategies / G. Adler, C. Fiocchi, L.B. Lazebnik, G.I. Vorobiev. — Springer, 2007. — 237 p. 2. Румянцев В.Г. Язвенный колит: руководство для вра- чей / B.Г. Румянцев. — М.: Медицинское информационное агентство, 2009. — 424 с. 3. Каншина О.А. Неспецифический язвенный колит у де- тей (и взрослых) / О.А. Каншина, Н.Н. Каншин. — М.: Биоин- формсервис, 2002. — 212 с. Примітка. * — Р < 0,001 порівняно з контролем. 4. Гастроэнтерология: национальное руководство / Под ред. B.Т. Ивашкина, Т.Л. Лапиной. — М.: ГЭОТАР-Медиа, 2008. — 704 с. Рисунок 2 — Концентрація фекального кальпро- тектину у хворих на ХЗЗК 24 № 2 (60), 2016 Гастроентерологія, ISSN 2308-2097 Ïàòîëîã³ÿ êèøå÷íèêà / Bowel Pathology 5. Абатуров О.Є. Значення фекального кальпротектину при ротавірусній інфекції в дітей раннього віку / О.Є. Аба- туров, Ю.Ю. Степанова // Гастроентерологія. — 2014. — № 4(54). — С. 106-109. 11. Roseth A.G. Assessment of disease activity in ulcerative colitis by faecal calprotectin, a novel granulocyte marker protein / A.G. Roset, E. Aadland, J. Jahnsen et al. // Digestion. — 1997. — № 58. — P. 176-180. 6. Степанов Ю.М. Содержание фекального кальпротек- тина у больных хроническими воспалительными заболевани- ями кишечника / Ю.М. Степанов, Н.С. Федорова // Сучасна гастроентерологія. — 2010. — № 2. — С. 44-48. 12. Bistrian B. Role of the systemic inflammatory response in the development of protein»energy malnutrition in Inflammatory bowel disease / B. Bistrian // Inflammatory Bowel Diseases: Nestle Nutrition Workshop Series Clinical & Performance Programme. — 1999. — Vol. 2. — P. 1-6. 6. Степанов Ю.М. Содержание фекального кальпротек- тина у больных хроническими воспалительными заболевани- ями кишечника / Ю.М. Степанов, Н.С. Федорова // Сучасна гастроентерологія. — 2010. — № 2. — С. 44-48. ями кишечника / Ю.М. Степанов, Н.С. Федорова // Сучасна гастроентерологія. — 2010. — № 2. — С. 44-48. 7. Татьянина О.Ф. Маркеры кишечного воспаления при за- болеваниях кишечника: обзор литературы / О.Ф. Татьянина, А.С. Потапов, Л.С. Намазова, Е.Г. Цимбалова, А.Г. Кучерен- ко, А.Н. Сурков // Педиатрическая фармакология. — 2008. — Т. 5, № 3. — С. 39-45. 7. Татьянина О.Ф. Маркеры кишечного воспаления при за- болеваниях кишечника: обзор литературы / О.Ф. Татьянина, А.С. Потапов, Л.С. Намазова, Е.Г. Цимбалова, А.Г. Кучерен- ко, А.Н. Сурков // Педиатрическая фармакология. — 2008. — Т. 5, № 3. — С. 39-45. 13. Синдеева Л.В. Антропометрия и биоимпедансме- трия: параллели и расхождения / Л.В. Синдеева, Г.Н. Список літератури Каза- кова // Фундаментальные исследования — 2013. — № 9. — С. 476-480. 14. Луфт В.М. Трофологическая недостаточность и кри- терии ее диагностики / В.М. Луфт, Е.И. Ткаченко // Воен.- мед. журн. — 1993. — № 12. — С. 21-24. 8. Татьянина О.Ф. Фекальный кальпротектин в неинва- зивной диагностике воспалительных заболеваний кишечника у детей / О.Ф. Татьянина, А.С. Потапов, Л.С. Намазова, Е.Г. Цимбалова, А.Г. Кучеренко, М.М. Лохматов // Педиа- трическая фармакология. — 2008. — Т. 5, № 3. — С. 46-51. 15. Analysis of fat and muscle mass in patients with inflammatory bowel disease during remission and active phase / R. Rocha, G.O. Santana, N. Almeida, A.C. Lyra // Br. J. Nutr. — 2009. — Vol. 101. — P. 676-679. 9. Долгих Т.И. Фекальный кальпротектин — неинвазивный биомаркер воспалительного процесса кишечника / Т.И. Дол- гих // Спецвыпуск «Лаборатория». — 2013. — № 3. — С. 44-46. 16. Anthropometry in body composition: An overview / J. Wang, J.C. Thornton, S. Kolesnik, R.N. Pierson // Ann. N.Y. Acad. Sci. — 2000. — Vol. 904. — P. 317-326. 10. Roset A.G. Normalization of faecal calprotectin: a predictor of healing in patients with inflammatory bowel disease / A.G. Roset, E. Aadland, K. Grzyb et al. // Scand. University Press. — 2004. — № 9. — P. 1012-1017. Отримано 22.03.16 Винник Н.В., Татарчук О.М., Стойкевич М.В., Кудрявцева В.Е., Сорочан Е.В. Лаборатория микробиологии и иммунологии, ГУ «Институт гастроэнтерологии НАМН Украины», г. Днепропетровск, Украина Винник Н.В., Татарчук О.М., Стойкевич М.В., Кудрявцева В.Е., Сорочан Е.В. Лаборатория микробиологии и иммунологии, ГУ «Институт гастроэнтерологии НАМН Украины», г. Днепропетровск, Украина 5. Абатуров О.Є. Значення фекального кальпротектину при ротавірусній інфекції в дітей раннього віку / О.Є. Аба- туров, Ю.Ю. Степанова // Гастроентерологія. — 2014. — № 4(54). — С. 106-109. 6. Степанов Ю.М. Содержание фекального кальпротек- тина у больных хроническими воспалительными заболевани- ями кишечника / Ю.М. Степанов, Н.С. Федорова // Сучасна гастроентерологія. — 2010. — № 2. — С. 44-48. 5. Абатуров О.Є. Значення фекального кальпротектину при ротавірусній інфекції в дітей раннього віку / О.Є. Аба- туров, Ю.Ю. Степанова // Гастроентерологія. — 2014. — № 4(54). — С. 106-109. 6. Степанов Ю.М. Содержание фекального кальпротек- тина у больных хроническими воспалительными заболевани- ями кишечника / Ю.М. Степанов, Н.С. Федорова // Сучасна гастроентерологія. — 2010. — № 2. — С. 44-48. 7. Татьянина О.Ф. Маркеры кишечного воспаления при за- болеваниях кишечника: обзор литературы / О.Ф. Татьянина, А.С. Потапов, Л.С. Намазова, Е.Г. Цимбалова, А.Г. Кучерен- ко, А.Н. Сурков // Педиатрическая фармакология. — 2008. — Т. 5, № 3. — С. 39-45. 8. Татьянина О.Ф. Фекальный кальпротектин в неинва- зивной диагностике воспалительных заболеваний кишечника у детей / О.Ф. Татьянина, А.С. Потапов, Л.С. Намазова, Е.Г. Цимбалова, А.Г. Кучеренко, М.М. Лохматов // Педиа- трическая фармакология. — 2008. — Т. 5, № 3. — С. 46-51. 9. Долгих Т.И. Фекальный кальпротектин — неинвазивный биомаркер воспалительного процесса кишечника / Т.И. Дол- гих // Спецвыпуск «Лаборатория». — 2013. — № 3. — С. 44-46. 10. Roset A.G. Normalization of faecal calprotectin: a predictor of healing in patients with inflammatory bowel disease / A.G. Roset, E. Aadland, K. Grzyb et al. // Scand. University Press. — 2004. — ОСОБЕННОСТИ МИКРОБИОЦЕНОЗА ТОЛСТОЙ КИШКИ И УРОВНЯ ФЕКАЛЬНОГО КАЛЬПРОТЕКТИНА У БОЛЬНЫХ ХРОНИЧЕСКИМИ ВОСПАЛИТЕЛЬНЫМИ ЗАБОЛЕВАНИЯМИ КИШЕЧНИКА В ЗАВИСИМОСТИ ОТ НУТРИТИВНОГО СТАТУСА Резюме. Статья посвящена актуальной проблеме — состоя- нию микрофлоры толстой кишки и уровню фекального каль- протектина у больных хроническими воспалительными заболе- ваниями кишечника с разным нутритивным статусом. Прове- дено обследование 76 пациентов с хроническими воспалитель- ными заболеваниями кишечника, среди которых 28 больных болезнью Крона и 48 — с неспецифическим язвенным колитом. Проанализированы полученные данные в зависимости от нозо- логии и нутритивного статуса. Установлено, что среди больных исследованных групп преобладали пациенты с субкомпенсиро- ванно бактер выявл рий ча ние ка пацие жести Клю кишеч шечни ванной формой дисбиоза. Снижение концентрации бифидо- бактерий в содержимом толстой кишки у больных III группы выявляли чаще всего — в 78,9 % случаев. Дефицит лактобакте- рий чаще наблюдался у больных II группы (92,3 %). Содержа- ние кальпротектина было существенно повышенным у 84,2 % пациентов, что положительно коррелировало со степенью тя- жести заболевания и не зависело от их нутритивного статуса. р ф р ур ф протектина у больных хроническими воспалительными заболе- ваниями кишечника с разным нутритивным статусом. Прове- дено обследование 76 пациентов с хроническими воспалитель- ными заболеваниями кишечника, среди которых 28 больных болезнью Крона и 48 — с неспецифическим язвенным колитом. Ключевые слова: хронические воспалительные заболевания кишечника, нутритивный статус, дисбиоз, микрофлора ки- шечника, кальпротектин. Проанализированы полученные данные в зависимости от нозо- логии и нутритивного статуса. Установлено, что среди больных исследованных групп преобладали пациенты с субкомпенсиро- FEATURES OF COLON MICROBIOCENOSIS AND FAECAL CALPROTECTIN LEVEL IN PATIENTS WITH CHRONIC INFLAMMATORY BOWEL DISEASES DEPENDING ON THE NUTRITIONAL STATUS Reducing the concentration of bifidobacteria in the colon content of patients from group III was detected more often — in 78.9 % of cases. Deficiency of lactobacilli was most often observed in patients of group II (92.3 %). The content of calprotectin was significantly increased in 84.2 % of patients, it positively correlated with the degree of severity of the disease and did not depend on their nutritional status. Summary. The article deals with an actual problem — the state of the microflora of the colon and faecal calprotectin level in patients with chronic inflammatory bowel diseases with different nutritional status. The study involved 76 patients with chronic inflammatory bow- el diseases, including 28 patients with Crohn’s disease and 48 — with ulcerative colitis. We analyzed the findings depending on the nosology and nutritional status. It was found that among the patients of studied groups, patients with subcompensated form of dysbiosis dominated. Key words: chronic inflammatory bowel diseases, nutritional sta- tus, dysbiosis, microflora of the bowel, calprotectin. 25 № 2 (60), 2016 www.gastro.org.ua
https://openalex.org/W2905250743	https://itms-journals.rtu.lv/article/download/itms-2018-0007/pdf	English	null	Security Aspects of Information Structures in the Information Warfare Context	Information Technology and Management Science	2,018	cc-by	3,896	II. THE EVOLUTION OF THE CONCEPT “INFORMATION WARFARE” Abstract – In the modern sense, the concept of information warfare includes the use and management of information and communication technologies to achieve a competitive advantage compared to the opponent. Information warfare is manipulation with information that trusts a goal without an objective understanding, so that the goal is to take decisions against its own interests in the interests of the opponents. Information structures are considered to be systems that produce and process various types of information, provide the storage of information and access to users. Such information structures may include neural networks, adaptive learning systems, etc. They must be prepared to train, respond to threats and ensure the safety of their existence, which is very topical during modern information warfare. This analytical article will cover more theoretical aspects related to the security of information systems from the system theory point of view. Knowledge base of the information structure can be a neural network, in which training should be provided from external threats. The author considers artificial neural networks to be one of the potential threats in the context of information warfare. The term “information warfare”, as the 4th generation war, appeared in the late 1980s and became popular very quickly. Thus, in the early 1990s, the first theoretical, and later, practical works appeared, where various definitions of the “information warfare” were given. Currently, the term “cyberwar” that is often endowed with the content and meanings attributed to “information warfare” is also actively used. The first profound definition of the term “information warfare” was given in the report of the American RAND Corporation “Strategic Information Warfare: a New Face of War” in 1996 [2]. According to it, “Information warfare is a war in the information space” i.e., to the existing at that time 3 military spaces (land, naval and air), a new information space was added. Keywords – artificial neural networks, cyberwar, information structures, information warfare, neural networks. Subsequently, in the cooperative document developed by headquarters of “Joint Doctrine for Information Operations” in 1998 [3] the definition of the information warfare was given – “information operations – a conflict in which crucially important and strategically important resource is information that is to be developed or destructed”. This is a multi- dimensional concept, which is only one of the aspects, the dimension of which is purely military. ©2018 Pēteris Grabusts. This is an open access article licensed under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0). Security Aspects of Information Structures in the Information Warfare Context Pēteris Grabusts Rezekne Academy of Technologies, Rezekne, Latvia ISSN 2255-9094 (online) ISSN 2255-9086 (print) December 2018, vol. 21, pp. 49–53 doi: 10.7250/itms-2018-0007 https://itms-journals.rtu.lv ISSN 2255-9094 (online) ISSN 2255-9086 (print) December 2018, vol. 21, pp. 49–53 doi: 10.7250/itms-2018-0007 https://itms-journals.rtu.lv Information Technology and Management Science II. THE EVOLUTION OF THE CONCEPT “INFORMATION WARFARE” The term “information operations” makes it possible to more accurately, than the traditional term “information warfare”, investigate the place and role of information confrontation as components of global confrontations. • cybernetic wars in virtual space. The goal of information warfare is to impact society through information. The signs of information warfare include: In a more modern interpretation, the information operation (Info Ops) is understood as an integrated use of the possibilities of electronic weapons, computer network operations (CNO), psychological operations (PSYOP), operations with military disinformation and disorganisation and security operations (OPSEC) to use the possibilities of influencing the human consciousness with the aim of destroying, corrupting, or in general intercepting the influence on the decision-making processes of the enemy, while protecting one’s own (decision) [6] (see Fig. 1). • restriction of access to certain information: blocking web resources, television programmes, printed publications; • creating a negative background on specific issues (fake news etc.); • the infiltration of information into various spheres of society. IV. MEANS OF IMPLEMENTATION OF INFORMATION ATTACKS Fig. 1. Components of information operations. Today there are many ways and methods of information warfare. The author singles out software and media. Software can be classified according to the tasks performed with their help on the means of information collection, means of distortion and destruction of information and means of influence on the functioning of information systems. Some means can be universal and used for distorting or destroying information, and for influencing the functioning of information systems. The main methods and techniques of using information weapons can be as follows: • damage to separate elements of the information infrastructure; • destruction or damage of information, programme resources of the enemy, overcoming protection systems, introduction of viruses, programme bookmarks and logical bombs; • the impact on the software and databases of information systems and management systems with the aim of distorting or modifying them; Fig. 1. Components of information operations. • seizure of media channels with the aim of spreading disinformation, rumours, demonstrating strength and bringing their demands; I. INTRODUCTION American researcher Harold Dwight Lasswell [1] can be called the leading theorist of the information warfare of the first half of the 20th century. He actively used the methods of social psychology, psychoanalysis and psychiatry in the study of political behaviour and propaganda, revealing the role of mass media during the information warfare of various states of the world for power. He singled out four main functions of mass media: • observing the world (collecting and spreading the information); There are many other definitions, both official and not official. According to the work “Information Warfare and Security” by D.E. Denning [4], “Information war is a set of operations aimed at or exploiting information resources”. • “editing” (selecting and commenting the information); • editing (selecting and commenting the information); • public opinion formation; • public opinion formation; • public opinion formation; The most profound definition of information warfare was suggested by the American theorist M. Libicki in his work “What Is Information Warfare?” dated 1995, where he singled out 7 types of information warfare [5]: • the spread of culture. Obviously, all these functions are active components of information warfare. The strategy of waging the information warfare through impact on public opinion presupposes awareness of the moods of all social, confessional and ethnic groups, awareness of the real state of things. Hence, on the one hand, there is informational-psychological impact through all possible channels, and on the other – a thorough study of public opinion, i.e., the identification of a reaction – the relationship of the elite and the population to informational-psychological influences, so that it is possible to make adjustments to the parameters of the impact. • military confrontation for monopolizing command- control functions; • confrontation of intelligence service and counterintelligence; • confrontation in the electronic sphere; • psychological operations; • organised spontaneous hacker attacks on information systems; • informational-economic wars for controlling the trade of information products and monopolizing the • informational-economic wars for controlling the trade of information products and monopolizing the 49 49 Information Technology and Management Science 2018/21 warfare depends on well-organized agitation based on the feelings and desires of members of society. information that is necessary to overcome the competitors; p • cybernetic wars in virtual space. • cybernetic wars in virtual space. III. TYPES AND CHARACTERISTICS OF INFORMATION WARFARE Information warfare can be used among military and among civilian population. One separate type of information warfare or a set of events can be used for this purpose. The types of information confrontation include: • destruction and suppression of links, artificial overloading of switching nodes; • impact on computer equipment to disable it. Typically, the media uses various methods of negative information impact: Typically, the media uses various methods of negative nformation impact: 1. information warfare on the internet – different and often contradictory information is offered, used to disorient the enemy; • use of compromising information in order to create a negative image of the politician; 2. psychological operations – selection and offering of the information, which seems like a counterargument to the moods existing in society; • special reduction of selected negative facts about this or that phenomenon, creation of numerous special TV programmes and headings in newspapers. 3. disinformation is the promotion of false information with the purpose to direct the enemy on the wrong track; First, a “plurality of opinions” is constructed. Then a more definite opinion about the event is expressed, providing it with “expected results”. 4. destruction is the physical destruction or blockage of electronic systems important to the enemy; At the third stage – in debates and disputes – unnecessary arguments are discarded in favour of one definite and unchanging decision. electronic systems important to the enemy; 5. security steps – strengthening the protection of the resources in order to save plans and intentions; The fourth stage is the introduction of the fact or evaluation of the event, which is presented as “the prevailing conviction” in the form of “people’s tendency to unanimity” in the interests of ordinary citizens. 6. direct information attacks – a mixture of false and truthful information. Information warfare can be waged both within the state and between different countries. The effectiveness of information 50 Information Technology and Management Science 2018/21 Therefore, the information is structured, i.e., it is written in a definite scheme. The policy of targeted influence on public opinion presupposes awareness of the moods of the broad masses of people realising the real situation. Hence, on the one hand, there is information and psychological impact on all possible channels, and on the other – a thorough study of public opinion. V. TRENDS AND DEVELOPMENT OF INFORMATION WARFARE Information warfare accompanies the history of all mankind. Propaganda can be admitted as the first element of information warfare. An information system is a system that performs obtaining input data, processing the data, the output of a result or a change in its external state. The French sociologist J. Ellul suggested distinguishing between vertical and horizontal propaganda. Vertical is a classic variant of propaganda – an information flow from the top down with a passive audience response [7]. Information warfare between two information systems is the open and hidden targeted informational impacts of two systems on each other with the aim of obtaining a certain prize. Horizontal propaganda is realized in the group, but does not come from above. In this situation all participants are equal. Information impact is implemented by using information weapons, i.e., such means, which allow carrying out the conceived actions with the transmitted, processed, created, destroyed and perceived information. Today’s business actively uses the methods of propaganda under other names – public relations (PR) and advertising. J. Stein in 1995 published the study “Information Warfare” [8], where he emphasises that information warfare deals with ideas. Concerning more specific goals, he states the following: “The target of information warfare, then, is the human mind, especially those minds that make the key decisions of war or peace and, from the military perspective, those minds that make the key decisions on if, when, and how to employ the assets and capabilities embedded in their strategic structures”. Information weapons are directly related to algorithms. Therefore, it is possible to refer any system to an information system – an information warfare object, if it is capable of processing an algorithm based on input data. One of the key issues that leads to the insolvability of the problem of winning information warfare is the following: “Is the information system capable of determining that information warfare has been launched against it?”. In his book “War and Anti-war”, A. Toffler gives examples of what is most often used to influence others [9]: In his book “War and Anti-war”, A. Toffler gives examples of what is most often used to influence others [9]: The author [12] considers that the problem of constructing an algorithm for determining the beginning of information warfare, in general, is algorithmically unsolvable (see Fig. 2). V. TRENDS AND DEVELOPMENT OF INFORMATION WARFARE • accusations of atrocities; • bid hyperbolization; The scheme above does not represent all possible approaches and techniques to the organisation and conduct of operations on information impact. • demonization and dehumanization of the opponent; • demonization and dehumanization of the opponent; • polarization; • polarization; • divine sanctions; It can be concluded that information weapons are primarily an algorithm. To use the information weapon means to select the input data for the system with the aim to activate certain algorithms in it, and in case of their absence, to activate algorithms for generating the necessary algorithms. • meta-propaganda, which discredits the propaganda of the other side. J. Arquilla [10] has formulated the rule: only the network structure can effectively work against the network structure; therefore, the hierarchical structures that belong to the state will always lose to the network structures. He has formulated the following three rules of this struggle: The following information will concern the information structures – training systems – in the simplified assumption it could be artificial neural network (ANN) and social networks. It is assumed that the information structure is a knowledge carrier and knowledge of the information system is expressed through its structure. Then, to evaluate the amount of information perceived by the system, it is logical to use such a notion as the degree of structure modification by input data. For example, the weight coefficients of neural links have changed – this is one information, but when elements have been cancelled or appeared – other information. • hierarchies find it difficult to fight networks; • you need networks to fight with networks; • those who master the first network forms will have significant advantages. The current situation of information operations can be seen in a series of publications [2], [11]. The current situation of information operations can be seen in a series of publications [2], [11]. III. TYPES AND CHARACTERISTICS OF INFORMATION WARFARE Information structure is now the most common term for those aspects of a sentence meaning that have to do with the way in which the hearer integrates the information into already existing information. To put it more simply, information structure is the domain of language structure and language study that is concerned with notions such as topic, comment, presupposition, and focus [12]. VI. INFORMATIONAL WARFARE – INFLUENCE ON INFORMATION STRUCTURES The information is tried to be saved in such a way that it can be easily navigated with a possibility of quickly finding the needed information element. 51 Information Technology and Management Science 2018/21 2018/21 Orienting the input impacts to those structures and elements that submit programming and reprogramming Information warfare strategy Study of opponent’s system Selection of input data corresponding to the standard Organisation of the presentation of the material “with one’s own eyes” Orienting the input actions to figures of art and culture Creation of chaos in all spheres of the functioning system Definition of a reprogramming chain, i.e the standard Organisation of the presentation of emotional materials (from the arts and culture) Identification of the basic elements and organisation of their behaviour modelling Fig. 2. Typical information warfare strategy. The structure of the system determines not only the resistance to external and internal effects, not only the time of reaction to this or that threat, but even such parameters as information security, the tendency to corruption, the level of complexity of the tasks solved by the organisation. Information warfare strategy Study of opponent’s system Conversely, the most stable system can be considered a system where the structure has the maximum connections, each element is connected to each, i.e., each element is basic. ANN, in general, cannot be considered stable information structures. It is connected with various training algorithms that work mostly on “black box” principle, which can make them vulnerable to various external threats. Organisation of the presentation of emotional materials (from the arts and culture) Artificial neural networks are a popular approach in the field of machine learning and perception. Traditionally, they attribute the properties of self-learning, self-organisation, and the ability to process figurative information in opposition to conventional algorithms, which are also traditionally considered to be hard-coded, untraceable, and intended for processing symbolic information. Identification of the basic elements and organisation of their behaviour modelling The more complex the network, the more parameters it has, the more data is required for its training. Usually we do not understand the connection between the trained neural network and the simulated phenomenon. It is not clear in detail why it works and we cannot predict in what cases it can fail. Fig. 2. Typical information warfare strategy. The issue of limiting AI has been raised in recent years [13], [14]. VI. INFORMATIONAL WARFARE – INFLUENCE ON INFORMATION STRUCTURES An AI box is a hypothetical isolated computer system where a possibly dangerous AI is kept constrained in a “virtual prison” and not allowed to manipulate events in the external world. Such a box would be restricted to minimalist communication channels. Unfortunately, even if the box is well-designed, a sufficiently intelligent AI may nevertheless be able to persuade or trick its human keepers into releasing it, or otherwise be able to “hack” its way out of the box [13]. It can be stated that the information structure is resistant against external effects, if the number of its elements does not undergo sharp fluctuations from these effects. What structure should a system have in order to prevent the number of its elements from experiencing sharp fluctuations? This is structure A, in which there are several groups of elements that are closely interconnected, but the relationships between groups are very unstable, for example, (see. Fig. 3) [12]: The author presents his viewpoint of AI as the protection information structure in the context of information warfare. Structure А: 1 – (2, 3, 4), 2 – (1, 3, 4), 3 – (1, 2, 4), 4 – (1, 2, 3, 5), 5– (4, 6, 7), 6 – (5, 7), 7 – (5, 6). In the context of information warfare, against the certain AI system (ANN or social network based on it), a certain threshold is set up which, apparently, should be calculated by some methodology, taking into account various activities within the framework of the system (fake news, social surveys etc.). The importance of the problem should be taken into account by the system’s developer (corporation) and, in case of a critical situation, by the government. Fig. 3. An example of stable information structure A. In any case, the system should have an inbuilt mechanism that could be called a trigger, which should respond to an extraordinary intrusion into its structure in the context of the information warfare. At the same time, the system is learning, re-learning and self-learning. Fig. 3. An example of stable information structure A. In structure A, it is enough to destroy an element with number 4 and the number of elements of the system is reduced by two times. Clearly, this structure is not stable (any structure is unstable, in which there are single elements carrying out a bunch of group elements). REFERENCES [1] H. Lasswell, “The Structure and Function of Communication in Society,” in The Communication of Ideas, L. Bryson, Ed. Institute for Religious and Social Studies, 1948, p. 117. [2] R. C. Molander, A. Riddile and P. A. Wilson, “Strategic Information Warfare: a New Face of War,” RAND Corporation, 1996. [Online]. Available: https://www.rand.org/pubs/monograph_reports/MR661.html [Accessed: September 29, 2018]. [ p ] [3] “Joint Publication 3-13/Information Operations,” Oct. 9, 1998. [Online]. Available: http://www.c4i.org/jp3_13.pdf. [Accessed: September 29, 2018]. Fig. 4. Potential reaction structure in case of information warfare attack. [4] D. E. Denning, Information Warfare and Security. Addison-Wesley, 1999. 1. trigger “on” – the self-destroyed mechanism is launched – the network activity is paralyzed, links are destroyed. The AI box protocol is interrupted; [5] M. C. Libicki, What Is Information Warfare?, National Defense University, Institute for National Strategic Studies, 1995. University, Institute for National Strategic Studies, 1995. [6] “Information Operation Roadmap,” Oct. 30, 2003. [Online]. Available: http://news.bbc.co.uk/1/shared/bsp/hi/pdfs/27_01_06_psyops.pdf. [Accessed: September 29, 2018]. y p p 2. trigger “off” – the attack is treated as false alarms and the system continues to work in the previous mode under the AI box protocol; [ p , ] [7] J. Ellul, Propaganda: The Formation of Men's Attitudes, Vintage Books, New York, 1965. 3. trigger “neutral” – the attack is treated as an unknown alert and the system continues to work in the previous mode under the AI box protocol, but by intensifying the analysis of the causes of the attack and trying to identify and prevent future threats; [8] G. J. Stein, “Information Warfare,” 1995. [Online]. Available: http://www.iwar.org.uk/iwar/resources/airchronicles/stein.htm. [Accessed: September 29, 2018]. [ p ] [9] A. Toffler, War and anti-war. Survival at the dawn of the 21st century, Little Brown & Co., 1993. identify and prevent future threats; [10] J. Arquilla and D. Ronfeldt, “The Advent of Netwar, “RAND Corporation, 2001, [Online]. Available: https://www.rand.org/pubs/monograph_reports/MR1382.html [Accessed: September 29, 2018]. 4. trigger “counterattack” – self-learning allows the system to exit the AI box protocol framework and the effects are not predictable. [11] D. Ventre, Information Warfare, 2nd Edition, Wiley, 2016. https://doi.org/10.1002/9781119004721 VI. INFORMATIONAL WARFARE – INFLUENCE ON INFORMATION STRUCTURES In case the information warfare attacks against the information structures, the trigger responds and there could be four possible situations (see Fig. 4): 52 Information Technology and Management Science 2018/21 Fig. 4. Potential reaction structure in case of information warfare attack. • export of the most prospective and high technologies from the country. • export of the most prospective and high technologies from the country. • export of the most prospective and high technologies from the country. The research presents a description of a potential counteraction against the threats of information warfare against information systems (AI based on neural networks). The research presents a description of a potential counteraction against the threats of information warfare against information systems (AI based on neural networks). The research presents a description of a potential counteraction against the threats of information warfare against information systems (AI based on neural networks). VII. CONCLUSION [12] S. P. Rastorguev, Information Warfare, M: Radio and Communication, 1998 (in Russian). Information warfare is a war of technologies; it is a war in which the structures of systems, as bearers of knowledge, interfere. It is necessary to talk about methods of information warfare because the understanding of the techniques of information warfare allows one to transfer it from the category of hidden threats into evident ones, with which one can deal. [13] D. Chalmers, “The Singularity: A Philosophical Analysis,” Journal of Consciousness Studies, vol.17, no. 7–65, Jan. 2010. [14] R. V. Yampolskiy, “What to Do with Singularity Paradox?,” in Philosophy and Theory of Artificial Intelligence, vol. 5, V. C. Muller Ed. Berlin, Germany: Springer-Verlag, 2013, pp. 397–413. https://doi.org/10.1007/978-3-642-31674-6_30 Consequences of information warfare: Pēteris Grabusts received his Dr. sc. ing. degree in Information Technology from Riga Technical University in 2006. Since 1996, he has been working at Rezekne Academy of Technologies. Since 2014, he has been a Professor at the Engineering Faculty. His research interests include data mining technologies, neural networks and clustering methods. His current research interests include ontologies. • death and emigration of part of the population; • destruction of industry; • loss of territory; • political dependence on the winner; he destruction (sharp reduction) of the army or the ban • the destruction (sharp reduction) of the army or the ban on one’s own army; E-mail: peteris.grabusts@rta.lv on one’s own army; 53
https://openalex.org/W4245581208	https://www.degruyter.com/document/doi/10.1515/dema-2010-frontmatter1/pdf	English	null	Titelei	Demonstratio Mathematica	2,010	cc-by	2,876	EDITORS Krzysztof Chelminski, Andrzej Fryszkowski, Stanislaw Janeczko, Janina Kotus, Zbigniew Lone, Zbigniew Pasternak-Winiarski, Anna Romanowska, Jacek Wesolowski, Ewa Zadrzynska-Pi^tka ASSOCIATE EDITORS ASSOCIATE EDITORS Hans-Dieter Alber (TU Darmstadt), Marek Bozejko (Wroclaw Univ.), Wlodzimierz Bryc (Univ. of Cincinnati), Vijay Gupta (New Delhi IT), Shyuichi Izumiya (Hokkaido Univ.), Jonathan Smith (Iowa State Univ.), Grzegorz Swiqtek (Warsaw Techn. Univ.), Miroslaw Truszczynski (Univ. of Kentucky), Wojciech Zaj^czkowski (IM PAN), Anatol Odzijewicz (Univ. of Bialystok) Naklad 430 + 20 egz. Ark. wyd. 19. Ark druk. 15. Druk ukoriczono w lutym 2010 r. Zam. nr 50/2010. Oficyna Wydawnicza Politechniki Warszawskiej, ul. Polna 50, 00-644 Warszawa All correspondence should be sent via e-mail to: demmath@mini.pw.edu.pl Directions for Authors. DEMONSTRATIO MATHEMATICA publishes original papers in mathematics. Manuscripts in the English language should be e-mailed to the Editorial Office at demmath@mini.pw.edu.pl , in the form of a pdf file. The covering letter should include all authors' names, the title of the paper and 2000 MSC codes. Papers should not exceed 20 typed pages. A short abstract should be included. The manuscript should contain the affiliation of all authors, including complete postal and e-mail addresses, and 2000 Mathematics Sub ject Classification codes. After the paper is accepted for publication, the author will be requested to e-mail the final TEX version to the Editorial Office. Authors receive 25 reprints of their articles free of charge. © Copyright by Faculty of Mathematics and Information Science, Warsaw University of Technology Typeset in T^X at the Faculty Printed and bound by Publishing House of the Warsaw University of Technology ul. Polna 50, 00-644 Warszawa Naklad 430 + 20 egz. Ark. wyd. 19. Ark druk. 15. Druk ukoriczono w lutym 2010 r. Zam. nr 50/2010. Oficyna Wydawnicza Politechniki Warszawskiej, ul. Polna 50, 00-644 Warszawa Warsaw University of Technology Faculty of Mathematics and Information Science DEMONSTRATIO MATHEMATICA Pl. Politechniki 1 00-661 Warsaw, POLAND DEMONSTRATIO MATHEMATICA Warszawa 2010 Warszawa 2010 ASSOCIATE EDITORS Hans-Dieter Alber (TU Darmstadt), Marek Bozejko (Wroclaw Univ.), Wlodzimierz Bryc (Univ. of Cincinnati), Vijay Gupta (New Delhi IT), Shyuichi Izumiya (Hokkaido Univ.), Jonathan Smith (Iowa State Univ.), Grzegorz âwiq.tek (Warsaw Techn. Univ.), Miroslaw Truszczynski (Univ. of Kentucky), Wojciech Zaj^czkowski (IM PAN), Anatol Odzijewicz (Univ. of Bialystok) Barbara Traczyk All correspondence should be sent via e-mail to: demmath@mini.pw.edu.pl Directions for Authors. DEMONSTRATIO MATHEMATICA publishes original papers in mathematics. Manuscripts in the English language should be e-mailed to the Editorial Office at demmath@mini.pw.edu.pl , in the form of a pdf file. The covering letter should include all authors' names, the title of the paper and 2000 MSC codes. Papers should not exceed 20 typed pages. A short abstract should be included. The manuscript should contain the affiliation of all authors, including complete postal and e-mail addresses, and 2000 Mathematics Subject Classification codes. After the paper is accepted for publication, the author will be requested to e-mail the final TEX version to the Editorial Office. Authors receive 25 reprints of their articles free of charge. C o n t e n t s S. ABBAS, M. BENCHOHEA: Impulsive partial hyperbolic differential inclu- sions of fractional order 775-797 J. ADAMUS: Effective regularity criteria for analytic mappings 235-262 J. AGGARWAL, J. K. KOHLI, D. SINGH: see J. K. KOHLI, D. SINGH, J. AGGARWAL M. AKKOUCHI, M. H. LALAOUI RHALI: Solutions of some functional equa- tions related to the determinant of certain matrices 97-106 M. AKKOUCHI, V. POPA: Well-posedness of fixed point problem for map- pings satisfying an implicit relation 923-929 G. A. ANASTASSIOU, A. ARAL: On Gauss-Weierstrass type integral opera- tors 841-849 M. K. AOUF, A. SHAMANDY, A. O. MOSTAFA, S. M. MADIAN: Certain class of p-valent functions associated with the Wright generalized hyper- geometric function 39-54 A. ARAL, G. A. ANASTASSIOU: see G. A. ANASTASSIOU, A. ARAL K. N. ARORA, D. KUMAR: see D. KUMAR, K. N. ARORA T. ASHIKAGA: Local signature of fibered complex surfaces via moduli and monodromy 263-276 G. AYKANAT, S. TEMIR: see S. TEMIR, G. AYKANAT N. BANO, N. A. MIR: Strong convergence of averaged iteration for asymp- totically non-expansive non-self mappings 681-690 A. BARTOSZEWICZ, M. FILIPCZAK, T. POREDA: On density with respect to equivalent measures 21-28 Z. BARTOSZEWSKI: Remarks on the convergence of an iterative method of solution of generalized least squares problem 931-938 F. BA§AR, O. TALO: see O. TALO, F. BA§AR M. BENCHOHRA, S. ABBAS: see S. ABBAS, M. BENCHOHRA J. BRZDIJK: A note on stability of the Popoviciu functional equation on re- stricted domain 635-641 H. CHANDRA, P. KUMAR: Ascent and descent of composition operators on lp spaces 161-165 S. S. CHENG, K. W. WEN, G. Q. WANG: see K. W. WEN, G. Q. Barbara Traczyk WANG, S. S. CHENG S. CHINO, S. IZUMIYA: Lightlike developables in Minkowski 3-space . . . . 387-399 G. AYKANAT, S. TEMIR: see S. TEMIR, G. AYKANAT LEMMERT: On coupled first and second order boundary value problems in ordered Banach spaces 561-574 N. T. T. HUYEN, N. M. TUAN: On the solvability of systems of linear equa- tions on commutative semigroup 865-876 S. IZUMIYA, S. CHINO: see S. CHINO, S. IZUMIYA D. JARUSZEWSKA-WALCZAK: Difference methods for infinite systems of quasilinear parabolic functional differential equations 213-230 S. JAYARAMAN, K. C. SIVAKUMAR: Matrix interval monotonicity 1-10 Z. JELONEK, M. FARNIK: see M. FARNIK, Z. JELONEK T. KAMRAN, Q. KIRAN: see Q. KIRAN, T. KAMRAN M. KASEDOU: Spacelike submanifolds in de Sitter space 401-418 S. M. KHAIRNAR, M. MORE: Properties of certain classes of analytic func- tions defined by Srivastava-Attiya operator 55-79 M. KHARE, A. K. SINGH: The range of non-atomic measures on effect alge- bras 497-510 Q. KIRAN, T. KAMRAN: Fixed point and homotopy results for generalized contractions on spaces with two metrics 151-160 S. IZUMIYA, S. CHINO: see S. CHINO, S. IZUMIYA G. AYKANAT, S. TEMIR: see S. TEMIR, G. AYKANAT L. I. COTIRLA: Properties of analytic functions defined by an integral operator 799 803 K. DENECKE, S. LEKKOKSUNG: Nd-hypersubstitutions of many-sorted al- gebras 519-537 M. B. DHAKNE, H. L. TIDKE: see H. L. TIDKE, M. B. DHAKNE S. S. DRAGOMIR: Inequalities for the numerical radius in unital normed al- gebras 129-137 S. S. DRAGOMIR: Some inequalities for the maximum of the spectrum for the real part of two operators product in Hilbert spaces 665-680 S. S. DRAGOMIR, H. B. THOMPSON: A two points Taylor's formula for the generalised Riemann integral 827-840 M. FARNIK, Z. JELONEK: A complete variety with infinitely many maximal quasi-projective open subsets 277-284 I. FIDYTEK: Some properties of the curves xn + yn = 1 with even exponents 89-96 M. FILIPCZAK, A. BARTOSZEWICZ, T. POREDA: see A. BARTOSZEWICZ, M. FILIPCZAK, T. POREDA T. FUKUNAGA: On homotopy classification of phrases and its application . 419-432 R. S. GUPTA, P. K. PANDEY, A. SHARFUDDIN: see P. K. PANDEY, R. S. GUPTA, A. SHARFUDDIN J. GWOZDZIEWICZ, A. LENARCIK, A. PLOSKI: Polar invariants of plane curve singularities: intersection theoretical approach 303-323 G. HERZOG, R. LEMMERT: On coupled first and second order boundary value problems in ordered Banach spaces 561-574 N. T. T. HUYEN, N. M. TUAN: On the solvability of systems of linear equa- tions on commutative semigroup 865-876 S. IZUMIYA, S. CHINO: see S. CHINO, S. IZUMIYA D. JARUSZEWSKA-WALCZAK: Difference methods for infinite systems of quasilinear parabolic functional differential equations 213-230 S. JAYARAMAN, K. C. SIVAKUMAR: Matrix interval monotonicity 1-10 Z. JELONEK, M. FARNIK: see M. FARNIK, Z. JELONEK T. KAMRAN, Q. KIRAN: see Q. KIRAN, T. KAMRAN M. KASEDOU: Spacelike submanifolds in de Sitter space 401-418 S. M. KHAIRNAR, M. MORE: Properties of certain classes of analytic func- tions defined by Srivastava-Attiya operator 55-79 M. KHARE, A. K. SINGH: The range of non-atomic measures on effect alge- bras 497-510 Q. KIRAN, T. KAMRAN: Fixed point and homotopy results for generalized contractions on spaces with two metrics 151-160 M. KOBAYASHI: A survey on polygonal portraits of manifolds 433-445 M. FILIPCZAK, A. BARTOSZEWICZ, T. POREDA: see A. BARTOSZEWICZ, M. FILIPCZAK, T. POREDA T. FUKUNAGA: On homotopy classification of phrases and its application . 419-432 R. S. GUPTA, P. K. PANDEY, A. SHARFUDDIN: see P. K. PANDEY, R. S. GUPTA, A. SHARFUDDIN J. GWOZDZIEWICZ, A. LENARCIK, A. PLOSKI: Polar invariants of plane curve singularities: intersection theoretical approach 303-323 G. HERZOG, R. S. IZUMIYA, S. CHINO: see S. CHINO, S. IZUMIYA D. JARUSZEWSKA-WALCZAK: Difference methods for infinite systems of quasilinear parabolic functional differential equations 213-230 S. JAYARAMAN, K. C. SIVAKUMAR: Matrix interval monotonicity 1-10 Z. JELONEK, M. FARNIK: see M. FARNIK, Z. JELONEK T. KAMRAN, Q. KIRAN: see Q. KIRAN, T. KAMRAN M. KASEDOU: Spacelike submanifolds in de Sitter space 401-418 S. M. KHAIRNAR, M. MORE: Properties of certain classes of analytic func- tions defined by Srivastava-Attiya operator 55-79 M. KHARE, A. K. SINGH: The range of non-atomic measures on effect alge- bras 497-510 Q. KIRAN, T. KAMRAN: Fixed point and homotopy results for generalized contractions on spaces with two metrics 151-160 M. KOBAYASHI: A survey on polygonal portraits of manifolds 433-445 L. KOCZAN, M. SOBCZAK-KNEC, P. ZAPRAWA: On functions convex in the direction of the imaginary axis with real coefficients 545-559 J. K. KOHLI, D. SINGH, J. AGGARWAL: R-supercontinuous functions . . . 703-723 S. KOIKE, A. PARUSINSKI: Some questions on the Fukui numerical set for complex function germs 285-302 A. KOWALIK, I. MICHALIK: Quasi-simplicity of some homeomorphism and diffeomorphism groups 11-19 D. KUMAR, K. N. ARORA: Growth and approximation properties of gener- alized axisymmetric potentials 107-116 P. KUMAR, H. CHANDRA: see H. CHANDRA, P. KUMAR R. KUMAR, R. RANI, R. K. NAGAICH: Some properties of lightlike subman- ifolds of semi-Riemannian manifolds 691-701 L. KUSHNER, M. J. SAIA: Geometry of pre quasi homogeneous polynomials of type (1/2,1/4) 461-471 M. H. LALAOUI RHALI, M. AKKOUCHI: see M. AKKOUCHI, M. H. LALAOUI RHALI S. LEKKOKSUNG, K. DENECKE: see K. DENECKE, S. LEKKOKSUNG R. LEMMERT, G. HERZOG: see G. HERZOG, R. LEMMERT A. LENARCIK, J. GWOZDZIEWICZ, A. PLOSKI: see J. GWOZDZIEWICZ, A. LENARCIK, A. PLOSKI W. LENSKI, B. SZAL: On the degree of strong approximation of functions and conjugate functions 117-128 N. T. LONG, L. T. P. NGOC, L. X. TRUONG: see L. T. P. NGOC, L. X. TRUONG, N. T. LONG S. M. MADIAN, M. K. AOUF, A. SHAMANDY, A. O. MOSTAFA: see M. K. AOUF, A. SHAMANDY, A. O. MOSTAFA, S. M. MADIAN G. MAJCHROWSKA, CHR. MOURATIDES: Blaschke products, a-curves and Qp spaces 29-38 H. MAKOWIECKA: On some operations on finite affine planes applied to the theory of regular configurations 167-185 M. VAN MANEN, D. SIERSMA: see D. SIERSMA, M. VAN MANEN J. MATKOWSKI: Generalizations of Lagrange and Cauchy mean-value theo- rems 765-774 L. MEZRAG: Characterization of /r-dominated m-linear operators 851-863 I. MICHALIK, A. KOWALIK: see A. KOWALIK, I. MICHALIK N. S. IZUMIYA, S. CHINO: see S. CHINO, S. IZUMIYA A. MIR, N. BANO: see N. BANO, N. A. MIR M. MORE, S. M. KHAIRNAR: see S. M. KHAIRNAR, M. MORE A. O. MOSTAFA, M. K. AOUF, A. SHAMANDY, S. M. MADIAN: see L. KOCZAN, M. SOBCZAK-KNEC, P. ZAPRAWA: On functions convex in the direction of the imaginary axis with real coefficients 545-559 J. K. KOHLI, D. SINGH, J. AGGARWAL: R-supercontinuous functions . . . 703-723 S. KOIKE, A. PARUSINSKI: Some questions on the Fukui numerical set for complex function germs 285-302 A. KOWALIK, I. MICHALIK: Quasi-simplicity of some homeomorphism and diffeomorphism groups 11-19 D. KUMAR, K. N. ARORA: Growth and approximation properties of gener- alized axisymmetric potentials 107-116 P. KUMAR, H. CHANDRA: see H. CHANDRA, P. KUMAR CHR. MOURATIDES, G. MAJCHROWSKA: see G. MAJCHROWSKA, CHR. MOURATIDES A. MUCKA: Subdirectly irreducible fibered automata 739-753 R. K. NAGAICH, R. KUMAR, R. RANI: see R. KUMAR, R. RANI, R. K. NAGAICH L. T. P. NGOC, L. X. TRUONG, N. T. LONG: High-order iterative methods for a nonlinear Kirchhoff wave equation 605-634 T. NISHIMURA: Singularities of pedal curves produced by singular dual curve germs in 5 " 447-459 T. NOIRI, V. POPA: The unified theory of certain types of generalizations of Lindelof spaces 203-212 T. OKUMA: The geometric genus, the Casson invariant conjecture and splice type singularities 325-338 P. K. PANDEY, R. S. GUPTA, A. SHARFUDDIN: B. Y. Chen's inequalities for bi-slant submanifolds in Kenmotsu space forms 887-898 A. PARUSINSKI, S. KOIKE: see S. KOIKE, A. PARUSINSKI E. PAWELEC, M. SZULARZ, A. SMOKTUNOWICZ: see M. SZULARZ, A. SMOKTUNOWICZ, E. PAWELEC P. PENNER, S. L. WISMATH: Equational bases for P-compatible identities 511-518 J.-P. PENOT: A short proof of the separable reduction theorem 653-663 A. PLOSKI, J. GWOZDZIEWICZ, A. LENARCIK: see J. GWOZDZIEWICZ, A. LENARCIK, A. PLOSKI V. POPA, M. AKKOUCHI: see M. AKKOUCHI, V. POPA V. POPA, T. NOIRI: see T. NOIRI, V. POPA T. POREDA, A. BARTOSZEWICZ, M. FILIPCZAK: see A. BARTOSZEWICZ, M. FILIPCZAK, T. POREDA J. K. PRAJAPAT, R. K. RAINA: New sufficient conditions for starlikeness of analytic functions involving a fractional differintegral operator 805-813 R. K. RAINA, J. K. PRAJAPAT: see J. K. PRAJAPAT, R. K. RAINA R. RANI, R. KUMAR, R. K. NAGAICH: see R. KUMAR, R. RANI, R. K. NAGAICH M. J. SAIA, L. KUSHNER: see L. KUSHNER, M. J. SAIA G. S. SALUJA: Convergence theorems of a family of uniformly (L, a)-Lipschitz asymptotically quasi-nonexpansive type mappings 877-886 M. S. SARSAK: On Bi-metacompact spaces 911-921 A. SHAMANDY, M. K. AOUF, A. O. MOSTAFA, S. M. MADIAN: see M. K. AOUF, A. SHAMANDY, A. O. MOSTAFA, S. M. MADIAN A. SHARFUDDIN, P. K. PANDEY, R. S. GUPTA: see P. K. PANDEY, R. S. GUPTA, A. SHARFUDDIN D. SIERSMA, M. VAN MANEN: Medial axis and a moment map 375-385 CHR. MOURATIDES, G. MAJCHROWSKA: see G. MAJCHROWSKA, CHR. MOURATIDES A. K. SINGH, M. KHARE: see M. KHARE, A. K. SINGH D. SINGH, J. K. KOHLI, J. AGGARWAL: see J. K. KOHLI, D. SINGH, J. AGGARWAL K. SHIBUYA: A set of integral elements of higher order jet spaces 473-486 K. C. SIVAKUMAR, S. JAYARAMAN: see S. JAYARAMAN, K. C. SIVAKU- MAR A. SMOKTUNOWICZ, M. SZULARZ, E. PAWELEC: see M. SZULARZ, A. SMOKTUNOWICZ, E. PAWELEC M. SOBCZAK-KNEC, L. KOCZAN, P. ZAPRAWA: see L. KOCZAN, M. SOBCZAK-KNEC, P. ZAPRAWA B. SOBEK: Pexider equation on a restricted domain 81-88 G. SOJKA: Determination of convex bodies by —oo-chord functions 187-202 B. SZAL, W. LENSKI: see W. LENSKI, B. SZAL P. SZCZUKA: The connectedness of arithmetic progressions in Furstenberg's, Golomb's, and Kirch's topologies 899-909 M. SZULARZ, A. SMOKTUNOWICZ, E. PAWELEC: An inverse structured perturbation problem for the linear system AT Ax = 6 755-763 0. TALO, F. BA§AR: Certain spaces of sequences of fuzzy numbers defined by a modulus function 139-149 S. TEMIR, G. AYKANAT: Errata to Convergence of implicit iterative process for finite family of I-nonexpansive mappings 231 H. B. THOMPSON, S. S. DRAGOMIR: see S. S. DRAGOMIR, H. B. THOMPSON H. L. TIDKE, M. B. DHAKNE: Existence and uniqueness of mild and strong solutions of nonlinear Volterra integrodifferential equations in Banach spaces 643-652 1.-C. TI§E: A fixed point approach to functional-integral set equations . . . . 815-826 T. TOMARU: Complex surface singularities and degenerations of compact complex curves 339-359 L. X. TRUONG, L. T. P. NGOC, N. T. LONG: see L. T. P. NGOC, L. X. TRUONG, N. T. LONG N. M. TUAN, N. T. T. HUYEN: see N. T. T. HUYEN, N. M. TUAN A. WALENDZIAK, M. WOJCIECHOWSKA-RYSIAWA: Bipartite pseudo-BL algebras 487-496 J. WALEWSKA: The second jump of Milnor numbers 361-374 G. Q. WANG, K. W. WEN, S. S. CHENG: see K. W. WEN, G. Q. WANG, S. S. CHENG K. W. WEN, G. Q. WANG, S. S. CHENG: Oscillations of a higher order M. SOBCZAK KNEC, P. ZAPRAWA B. SOBEK: Pexider equation on a restricted domain 81-88 G. SOJKA: Determination of convex bodies by —oo-chord functions 187-202 B. SZAL, W. LENSKI: see W. LENSKI, B. SZAL P. SZCZUKA: The connectedness of arithmetic progressions in Furstenberg's, Golomb's, and Kirch's topologies 899-909 M. SZULARZ, A. SMOKTUNOWICZ, E. PAWELEC: An inverse structured perturbation problem for the linear system AT Ax = 6 755-763 0. TALO, F. CHR. MOURATIDES, G. MAJCHROWSKA: see G. MAJCHROWSKA, CHR. MOURATIDES BA§AR: Certain spaces of sequences of fuzzy numbers defined by a modulus function 139-149 S. TEMIR, G. AYKANAT: Errata to Convergence of implicit iterative process for finite family of I-nonexpansive mappings 231 H. B. THOMPSON, S. S. DRAGOMIR: see S. S. DRAGOMIR, H. B. THOMPSON H. L. TIDKE, M. B. DHAKNE: Existence and uniqueness of mild and strong solutions of nonlinear Volterra integrodifferential equations in Banach spaces 643-652 1.-C. TI§E: A fixed point approach to functional-integral set equations . . . . 815-826 T. TOMARU: Complex surface singularities and degenerations of compact complex curves 339-359 L. X. TRUONG, L. T. P. NGOC, N. T. LONG: see L. T. P. NGOC, L. X. TRUONG, N. T. LONG N. M. TUAN, N. T. T. HUYEN: see N. T. T. HUYEN, N. M. TUAN A. WALENDZIAK, M. WOJCIECHOWSKA-RYSIAWA: Bipartite pseudo-BL algebras 487-496 J. WALEWSKA: The second jump of Milnor numbers 361-374 G. Q. WANG, K. W. WEN, S. S. CHENG: see K. W. WEN, G. Q. WANG, S. S. CHENG K. W. WEN, G. Q. WANG, S. S. CHENG: Oscillations of a higher order nonlinear differential equations with impulses 575-603 W. WILCZYNSKI: The set of points of discontinuity of X-approximately con- tinuous functions 539-544 S. L. WISMATH, P. PENNER: see P. PENNER, S. L. WISMATH M. WOJCIECHOWSKA-RYSIAWA, A. WALENDZIAK: see A. WALENDZIAK, M. WOJCIECHOWSKA-RYSIAWA P. ZAPRAWA, L. KOCZAN, M. SOBCZAK-KNEC: see L. KOCZAN, M. SOBCZAK-KNEC, P. ZAPRAWA Preface 233-234
https://openalex.org/W4390675158	https://www.researchsquare.com/article/rs-3833422/latest.pdf	English	null	Association of the predicted free blood concentration of teicoplanin with the development of renal dysfunction	Research Square (Research Square)	2,024	cc-by	3,893	Association of the predicted free blood concentration of teicoplanin with the development of renal dysfunction Kyohei Sugiyama (  kyohei-sugiyama@i.shizuoka-ph Shizuoka General Hospital Keita Hirai Shinshu University Graduate School of Medicine Yukako Suyama Shizuoka General Hospital Kento Furuya Shizuoka General Hospital Kenta Ito Shizuoka General Hospital Research Article Posted Date: January 8th, 2024 DOI: https://doi.org/10.21203/rs.3.rs-3833422/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Additional Declarations: No competing interests reported. Version of Record: A version of this preprint was published at European Journal of Clinical Pharmacology on January 29th, 2024. See the published version at https://doi.org/10.1007/s00228-024-03638-0. Page 1/14 Page 1/14 Abstract Purpose: In clinical practice, teicoplanin (TEIC) is typically administered at a trough concentration of 15– 40 μg/mL. TEIC has a protein binding rate of approximately 90%, and its concentration rarely exceeds 40 μg/ml. Nevertheless, an increase in the free blood trough concentration of TEIC may result in renal dysfunction. However, the relationship between the free blood trough concentration and occurrence of renal dysfunction remains unclear. This study aimed to examine the effect of predicted free blood concentrations on the development of renal dysfunction. Methods: This retrospective study included patients who underwent TEIC treatment and at least one trough concentration measurement. The association between the frequency of renal dysfunction and the predicted free blood concentration was evaluated using the following equation: free TEIC concentration = total TEIC concentration/(1+1.78×serum albumin level). Results: Of the 170 patients included in this study, 18% (31/170) developed renal dysfunction. The predicted free trough concentration was significantly higher in the renal dysfunction onset group than that in the non-onset group. However, the total trough concentration was not significantly associated with the development of renal dysfunction. The odds ratio for developing renal dysfunction was 4.2 (95% confidence interval: 1.71–10.4, P=0.002) when the predicted free trough concentration was >4.0 μg/mL. Conclusion: Elevated free trough concentrations of TEIC are associated with an increased risk of renal dysfunction. Controlling the increase in the predicted free blood concentration of TEIC may effectively prevent the development of renal dysfunction. Introduction Teicoplanin (TEIC) is a glycopeptide antibacterial agent recommended for methicillin-resistant Staphylococcus aureus (MRSA) infections [1]. TEIC has comparable therapeutic efficacy to vancomycin (VCM) and is associated with a lower incidence of renal dysfunction [2]. Therefore, TEIC is a more suitable treatment option for patients with MRSA infection and reduced renal function. TEIC has a high protein-binding rate of approximately 90% [3], which can lead to increased free blood concentrations in patients with low albumin levels. This may cause whole-blood trough concentrations to be lower in patients with hypoalbuminemia owing to increased tissue distribution [4], whereas the free-blood trough concentration remains high. Consequently, it is advisable to refrain from establishing target trough concentrations solely based on whole-blood trough concentrations [5]. A meta-analysis exploring the risk factors for renal dysfunction in patients undergoing TEIC treatment reported that a decrease in the serum albumin level was a risk factor for the development of renal dysfunction [6]. Although a previous study examined the total trough concentrations predicted from the serum albumin levels in patients with hematologic malignancies [7], no report has shown a direct association between elevated free blood trough concentrations and the development of renal dysfunction. According to the current guidelines for therapeutic drug monitoring (TDM) via TEIC for the Page 2/14 treatment of MRSA infection, the recommended total trough concentration is 15–40 µg/mL [5]. A infection, the recommended total trough concentration is 15–40 µg/mL [5] treatment of MRSA infection, the recommended total trough concentration is 15–40 µg/mL [5]. Theoretically, the target free blood concentration is 1.5–4.0 µg/mL when the protein binding rate is 90% However, there are no established recommendations for target-free blood concentrations in terms of th efficacy or adverse drug reactions. Theoretically, the target free blood concentration is 1.5–4.0 µg/mL when the protein binding rate is 90%. However, there are no established recommendations for target-free blood concentrations in terms of the efficacy or adverse drug reactions. Considering the aforementioned facts and our clinical experience, which indicates that TEIC rarely deviates from the recommended total trough concentration, we speculated that an increased free blood concentration might be implicated in the emergence of renal dysfunction. To investigate this possibility, we employed a prediction formula [8] that calculates the free TEIC concentration from the measured total trough concentration using the following equation: free TEIC concentration = total TEIC concentration/(1 + 1.78×serum albumin level). Study design and patients This retrospective observational study was conducted at the Shizuoka General Hospital, Shizuoka, Japan. The Ethics Committee of Shizuoka General Hospital approved the study protocol (approval number: SGHIRB#2023027). The study included patients who were admitted to Shizuoka General Hospital between January 2016 and April 2023 and underwent TEIC treatment with at least one trough concentration measurement. The exclusion criteria were an estimated glomerular filtration rate (eGFR) < 10 mL/min/1.73 m2, renal replacement therapy, age younger than 18 years, patients who received TEIC for less than three days, and a trough concentration below the detection limit (< 3.0 µg/mL); additionally, the serum creatinine (SCr), eGFR, and serum albumin levels were not measured within three days before TEIC administration. Introduction This formula was derived from a dataset comprising 198 samples collected from 68 Japanese subjects, in whom bound and free blood trough concentrations were determined using a fluorescence polarization immunoassay. The aim of this study was to elucidate the association between the predicted free blood concentration of TEIC and the development of renal dysfunction. Outcomes and variables The primary endpoint was the relationship between the frequency of renal dysfunction and minimum, maximum, and mean predicted free blood trough concentrations. Renal dysfunction was defined as an increase of 0.5- or 1.5-fold in the baseline SCr level [6, 9]. The patients were followed up for three days after the completion of TEIC treatment. The upper limit of the recommended total trough concentration is 40 µg/mL, which corresponds to a free trough concentration of 4.0 µg/mL when TEIC protein binding is 90%. Therefore, we also evaluated the association between a predicted free blood concentration greater than 4.0 µg/mL and the frequency of renal dysfunction. We analyzed the influence of age, sex, weight, body mass index (BMI), number of days of TEIC administration, loading dose, maintenance dose, and total trough concentration. The total trough concentration of TEIC was determined by latex agglutination Page 3/14 turbidimetry. We also recorded the use of loop diuretics, thiazide diuretics, nonsteroidal anti-inflammatory drugs (NSAIDs), cyclosporine, tacrolimus, angiotensin II receptor blocker (ARB), angiotensin-converting enzyme inhibitor (ACEI), cisplatin, liposomal amphotericin B, foscarnet, tenofovir, acyclovir, valacyclovir, aminoglycosides, trimethoprim-sulfamethoxazole, and tazobactam piperacillin [10] prescribed for more than three days during the TEIC treatment period, which could affect the renal function. Notably, the Scr, eGFR, and serum albumin levels were obtained from laboratory tests. Statistical analysis The Wilcoxon rank-sum test was used to analyze continuous variables, whereas Fisher's exact test was used for categorical variables to evaluate the dissimilarities between the groups characterized by the presence or absence of renal dysfunction. A logistic regression analysis was conducted to calculate the odds ratio (OR) for the development of renal impairment and the corresponding 95% confidence interval (CI). The statistical significance was set at P < 0.05. All statistical analyses were performed using the R software (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria). Characteristics of the study patients A total of 170 eligible patients were included in this study, of which 31 (18%) patients developed renal dysfunction (Fig. 1). The characteristics of patients with and without renal dysfunction, including sex, weight, body mass index (BMI), TEIC dosage, number of days on TEIC, and concomitant medications that may affect renal function, were not significantly different (Table 1). Page 4/14 Table 1 Table 1 Comparison of groups with and without renal dysfunction Development of renal dysfunction P value Onset (n = 31) Non-onset (n = 139) Sex, Male 20 (65%) 83 (60%) 0.688 Age (years) 80 (69, 84) 66 (52, 76) < 0.001 Weight (kg) 53.7 (45.5, 63.3) 52.3 (42.9, 60.3) 0.463 BMI (kg/m2) 19.8 (18.6, 23.1) 20.5 (18.9, 23.2) 0.632 Baseline serum creatinine level 1.52 (0.74, 2.16) 0.74 (0.53, 1.1) < 0.001 Baseline eGFR 32 (22, 69) 77 (44, 108) < 0.001 Days of TEIC administration 11 (8, 16.5) 11 (8, 15) 0.824 Loading dose (mg/kg) 11.2 (9.97, 12.2) 11.1 (10.1, 12) 0.815 Maintenance dose (mg/kg) 6.3 (5.1, 7.4) 6.64 (5.61, 7.94) 0.378 Total trough concentration Minimum 18.7 (14.2, 22.9) 19.0 (14.3, 23.8) 0.942 Maximum 24.1(19.3, 29.7) 22.7 (17.7, 27.5) 0.297 Mean 21.9 (17.4, 25.4) 20.8 (16.2, 25.5) 0.544 Predicted free blood concentration Minimum 4.0 (3.6, 4.6) 3.2 (2.7, 4.1) 0.001 Maximum 5.1 (4.1, 6.8) 3.9 (3.1, 4.9) < 0.001 Mean 4.7 (4.0, 5.5) 3.6 (2.9, 4.4) < 0.001 Minimum > 4.0 µg/mL 16 (52%) 38 (27%) 0.011 Maximum > 4.0 µg/mL 24 (77%) 64 (46%) 0.002 Mean > 4.0 µg/mL 22 (71%) 49 (35%) < 0.001 Serum albumin level Minimum 1.9 (1.5, 2.2) 2.5 (2.1, 3.1) < 0.001 The data are presented as frequencies (percentages) or medians (interquartile ranges). BMI, body mass index; eGFR, estimated glomerular filtration rate; ARB, angiotensin II receptor blocker; ACEI, angiotensin-converting enzyme inhibitor. Characteristics of the study patients Page 5/14 Development of renal dysfunction P value Onset (n = 31) Non-onset (n = 139) Maximum 2.1 (1.9, 2.5) 2.7 (2.3, 3.2) < 0.001 Mean 2.1 (1.7, 2.3) 2.6 (2.2, 3.1) < 0.001 Drugs at risk for renal impairment Loop diuretics 11 (35%) 28 (20%) 0.096 Thiazide diuretics 1 (3.2%) 1 (0.7%) 0.332 Cyclosporine 0 (0%) 1 (0.7%) 1.000 Tacrolimus 0 (0%) 9 (6.5%) 0.368 ARB 4 (13%) 7 (5%) 0.117 ACEI 0 (0%) 4 (2.9%) 1.000 Liposomal amphotericin B 4 (14%) 13 (9.4%) 0.518 Aminoglycoside 3 (10%) 22 (16%) 0.575 Piperacillin/Tazobactam 6 (21%) 40 (29%) 0.657 Sulfamethoxazole-trimethoprim 2 (6.5%) 30 (21.6%) 0.073 Acyclovir 0 (0%) 7 (5%) 0.352 Valacyclovir 1 (3.2%) 5 (3.6%) 1.000 Foscarnet 0 (0%) 1 (0.7%) 1.000 Impact of trough concentration on serum albumin levels and predicted free blood concentration The differences in the total trough concentrations between patients with and without renal dysfunction were not statistically significant for the minimum, maximum, or mean concentrations (Fig. 2a). In contrast, there was a notable difference in the minimum, maximum, and mean predicted free-trough concentrations between the two groups, with significantly greater values in patients who developed renal dysfunction than in those who did not (Fig. 2b). Page 6/14 Page 6/14 As indicated in Table 1, a minimum, maximum, and mean predicted free trough concentration higher than 4.0 µg/mL was found to be a statistically significant factor in patients with renal dysfunction compared to those without renal dysfunction. Furthermore, the serum albumin concentration and baseline renal function were significantly lower in the group with renal dysfunction. Factors associated with the development of renal dysfunction Factors associated with the development of renal dysfunction A logistic regression analysis indicated that age, baseline renal function, and predicted free blood concentration were significantly associated with the development of renal dysfunction (Table 2). The predicted free blood concentration, adjusted for age and baseline Scr also showed significant associations with the development of renal dysfunction for all factors: minimum, maximum, and mean. A predicted free blood concentration > 4.0 µg/mL was also found to be a statistically significant factor for the risk of developing renal dysfunction (Table 2). A receiver operating characteristic (ROC) curve analysis revealed that the predicted mean free blood trough concentration could predict the incidence of renal dysfunction, with an area under the curve (AUC) of 0.74 (95% CI; 0.65, 0.83), a sensitivity of 71%, and a specificity of 71%. The best cutoff value for predicting the mean free blood concentration was 4.2 µg/mL (Fig. 3). Page 7/14 Table 2 Logistic regression analysis of factors associated with the development of renal dysfunction Factors Univariate Adjustment with age and serum creatinine level OR (95% CI) P value OR (95% CI) P value Age (years) 1.01 (1.01, 1.11) 0.001 Baseline serum creatinine level 2.21 (1.41, 3.41) < 0.001 Predicted free blood concentration Minimum 1.41 (1.11, 1.81) 0.009 1.41 (1.06, 1.88) 0.017 Maximum 1.51 (1.21, 1.91) < 0.001 1.47 (1.18, 1.84) < 0.001 Mean 1.61 (1.21, 2.11) < 0.001 1.58 (1.19, 2.09) 0.001 Minimum > 4.0µg/mL 2.81 (1.31, 6.31) 0.01 2.61 (1.12, 6.11) 0.026 Maximum > 4.0µg/mL 4.01 (1.61, 9.91) 0.002 3.83 (1.46, 10.0) 0.006 Mean > 4.0µg/mL 4.51 (1.91, 10.51) < 0.001 4.22 (1.71, 10.4) 0.002 OR: odds ratio; CI: confidence interval. Table 2 Discussion The results of this study indicate that the total trough concentration of TEIC did not correlate with the emergence of renal dysfunction. In all patients, dosage optimization was performed using TDM, and there were no significant differences in the dosage. The median total trough concentration in the group that developed renal dysfunction was approximately 20 µg/mL, which is markedly different from a previous report suggesting that renal dysfunction is a risk factor for trough concentrations ≥ 60 µg/mL [11]. Page 8/14 By contrast, the predicted free blood concentration was significantly higher in the group that developed renal dysfunction. The factors associated with the development of renal impairment were older age, lower baseline renal function, and lower serum albumin levels. Older age is correlated with lower baseline renal function [12], which is considered a risk factor for drug-induced kidney injury [13]. However, even after accounting for the influence of age and baseline renal function, an increase in the predicted free blood concentration was associated with an increased incidence of renal dysfunction. Previous studies have Page 8/14 Page 8/14 reported an association between renal dysfunction and hypoalbuminemia [6, 14], which was also observed in the present study. Specifically, low serum albumin levels increase the incidence of renal dysfunction. Low serum albumin levels may lead to higher predicted free blood levels, thereby increasing the risk of renal dysfunction. reported an association between renal dysfunction and hypoalbuminemia [6, 14], which was also observed in the present study. Specifically, low serum albumin levels increase the incidence of renal dysfunction. Low serum albumin levels may lead to higher predicted free blood levels, thereby increasing the risk of renal dysfunction. Although this study assumed that the protein-binding rate of TEIC was approximately 90%, inter- individual differences in the protein-binding rate may exist [8, 15]. Therefore, in clinical practice, the measurement of free blood levels [16] may be beneficial for preventing the development of renal dysfunction. However, this approach poses a challenge because it requires specialized equipment and techniques, and may not be easily implemented in general hospitals. In addition, free blood concentration, which indicates efficacy, was not confirmed in the present study. It is not clear whether the total or unbound concentrations are more important for the efficacy of TEIC [17]. Therefore, we believe that a total blood concentration ≥ 15 µg/mL (≥ 20 µg/mL in severe cases) is an appropriate target value for efficacy [5]. Discussion In the present study, an elevated predicted free blood concentration (> 4.0 µg/mL as a specific indicator) was found to be a risk factor for renal dysfunction, which was justified by the fact that the optimal cutoff value for the mean free blood trough concentration to predict renal dysfunction according to an ROC analysis was 4.2 µg/mL. This finding suggests that the upper trough concentration limit may contribute to preventing the development of renal dysfunction by using free blood concentration as an indicator rather than by evaluating the total blood concentration. This study had several limitations. First, this was a single-center study with a relatively small sample size. Second, this approach did not eliminate all potential background factors that may contribute to the development of renal dysfunction. Third, the accuracy of the formula used to calculate the predicted free blood concentrations was not evaluated. Conclusions A more effective strategy for preventing the development of renal dysfunction in TEIC may be to control the upper limit of the predicted free blood concentration of TEIC to less than 4.0 µg/mL rather than to control the total trough concentration. Authors' contributions K.S. and K.H. designed the study. K. S., Y. S., K. F., and K. I. performed the experiments. K.S. and K.H. analyzed the data. K. S. and K. H. wrote the manuscript. All the authors have read and approved the final version of the manuscript. Availability of data and material Availability of data and material Page 9/14 Ethics approval This study was approved by the Ethics Committee of Shizuoka General Hospital (approval number: SGHIRB#2023027). This study was approved by the Ethics Committee of Shizuoka General Hospital (approval number: SGHIRB#2023027). Funding This study received no specific grants from funding agencies in the public, commercial, or non-profit sectors. Competing Interests The authors declare that they have no conflicts of interest. References 1. Brown NM, Goodman AL, Horner C et al (2021) Treatment of methicillin-resistant Staphylococcus aureus (MRSA): Updated guidelines from the UK. J Antimicrob Chemother 76:1377–1378. https://doi.org/10.1093/jac/dkab036 2. Rolston KVI, Nguyen H, Amos G et al (1994) A randomized double-blind trial of vancomycin versus teicoplanin for the treatment of gram-positive bacteremia in patients with cancer. J Infect Dis 169:350–355. https://doi.org/10.1093/infdis/169.2.350 3. Assandri A, Bernareggi A (1987) European journal of clinical pharmacology binding of teicoplanin to human serum albumin. Eur J Clin Pharmacol 33:191–195. https://doi.org/10.1007/BF00544566 4. Ueda T, Takesue Y, Nakajima K et al (2020) Clinical efficacy and safety in patients treated with teicoplanin with a target trough concentration of 20 µg/mL using a regimen of 12 mg/kg for five doses within the initial 3 days. BMC Pharmacol Toxicol 21:50. https://doi.org/10.1186/s40360-020- 00424-3 4. Ueda T, Takesue Y, Nakajima K et al (2020) Clinical efficacy and safety in patients treated with teicoplanin with a target trough concentration of 20 µg/mL using a regimen of 12 mg/kg for five doses within the initial 3 days. BMC Pharmacol Toxicol 21:50. https://doi.org/10.1186/s40360-020- 00424-3 5. Hanai Y, Takahashi Y, Niwa T et al (2022) Clinical practice guidelines for therapeutic drug monitoring of teicoplanin: A consensus review by the Japanese Society of Chemotherapy and the Japanese Society of Therapeutic Drug Monitoring. J Antimicrob Chemother 77:869–879. https://doi.org/10.1093/jac/dkab499 5. Hanai Y, Takahashi Y, Niwa T et al (2022) Clinical practice guidelines for therapeutic drug monitoring of teicoplanin: A consensus review by the Japanese Society of Chemotherapy and the Japanese Society of Therapeutic Drug Monitoring. J Antimicrob Chemother 77:869–879. https://doi.org/10.1093/jac/dkab499 Page 10/14 6. Hirai T, Hosohata K, Ogawa Y, Iwamoto T (2022) Clinical predictors of nephrotoxicity associated with teicoplanin: Meta-analysis and meta-regression. Basic Clin Pharmacol Toxicol 130:110–121. https://doi.org/10.1111/bcpt.13679 7. Byrne CJ, Parton T, McWhinney B et al (2018) Population pharmacokinetics of total and unbound teicoplanin concentrations and dosing simulations in patients with haematological malignancy. J Antimicrob Chemother 73:995–1003. https://doi.org/10.1093/jac/dkx473 8. Yano R, Nakamura T, Tsukamoto H et al (2007) Variability in teicoplanin protein binding and its prediction using serum albumin concentrations. Ther Drug Monit 29:399–403. https://doi.org/10.1097/FTD.0b013e3180690755 9. Rybak MJ, Le J, Lodise TP et al (2020) Therapeutic monitoring of vancomycin for serious methicillin- resistant Staphylococcus aureus infections: A revised consensus guideline and review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. References Deltombe O, Mertens T, Eloot S, Verstraete AG (2019) Development and validation of an ultra-high performance liquid chromatography – high resolution mass spectrometry method for the quantification of total and free teicoplanin in human plasma. Clin Biochem 65:29–37. https://doi.org/10.1016/j.clinbiochem.2018.12.010 17. Dykhuizen RS, Harvey G, Stephenson N et al (1995) Protein Binding and Serum Bactericidal Activities of Vancomycin and Teicoplanin. Antimicrob Agents Chemother 39:1842–1847. https://doi.org/10.1128/AAC.39.8.1842 17. Dykhuizen RS, Harvey G, Stephenson N et al (1995) Protein Binding and Serum Bactericidal Activities of Vancomycin and Teicoplanin. Antimicrob Agents Chemother 39:1842–1847. https://doi.org/10.1128/AAC.39.8.1842 References Am J Health Syst Pharm 77:835–864. https://doi.org/10.1093/ajhp/zxaa036 10. Workum JD, Kramers C, Kolwijck E et al (2021) Nephrotoxicity of concomitant piperacillin/tazobactam and teicoplanin compared with monotherapy. J Antimicrob Chemother 76:212–219. https://doi.org/10.1093/JAC/DKAA385 10. Workum JD, Kramers C, Kolwijck E et al (2021) Nephrotoxicity of concomitant piperacillin/tazobactam and teicoplanin compared with monotherapy. J Antimicrob Chemother 76:212–219. https://doi.org/10.1093/JAC/DKAA385 11. Tobin CM, Lovering AM, Sweeney E, MacGowan AP (2010) Analyses of teicoplanin concentrations from 1994 to 2006 from a UK assay service. J Antimicrob Chemother 65:2155–2157. https://doi.org/10.1093/jac/dkq266 11. Tobin CM, Lovering AM, Sweeney E, MacGowan AP (2010) Analyses of teicoplanin concentrations from 1994 to 2006 from a UK assay service. J Antimicrob Chemother 65:2155–2157. https://doi.org/10.1093/jac/dkq266 12. Weinstein JR, Anderson S (2010) The Aging Kidney: Physiological Changes. Adv Chronic Kidney Dis 17:302–307. https://doi.org/10.1053/j.ackd.2010.05.002 12. Weinstein JR, Anderson S (2010) The Aging Kidney: Physiological Changes. Adv Chronic Kidney Dis 17:302–307. https://doi.org/10.1053/j.ackd.2010.05.002 13. Naughton CA (2008) Drug-Induced Nephrotoxicity. Am Fam Physician 78:743–750 13. Naughton CA (2008) Drug-Induced Nephrotoxicity. Am Fam Physician 78 n CA (2008) Drug-Induced Nephrotoxicity. Am Fam Physician 78:743–750 13. Naughton CA (2008) Drug-Induced Nephrotoxicity. Am Fam Physician 78:743–750 14. Cheng T, Wang X, Han Y et al (2023) The level of serum albumin is associated with renal prognosis and renal function decline in patients with chronic kidney disease. BMC Nephrol 24:57. https://doi.org/10.1186/s12882-023-03110-8 14. Cheng T, Wang X, Han Y et al (2023) The level of serum albumin is associated with renal prognosis and renal function decline in patients with chronic kidney disease. BMC Nephrol 24:57. https://doi.org/10.1186/s12882-023-03110-8 15. Roberts JA, Stove V, De Waele JJ et al (2014) Variability in protein binding of teicoplanin and achievement of therapeutic drug monitoring targets in critically ill patients: Lessons from the DALI Study. Int J Antimicrob Agents 43:423–430. https://doi.org/10.1016/j.ijantimicag.2014.01.023 15. Roberts JA, Stove V, De Waele JJ et al (2014) Variability in protein binding of teicoplanin and achievement of therapeutic drug monitoring targets in critically ill patients: Lessons from the DALI Study. Int J Antimicrob Agents 43:423–430. https://doi.org/10.1016/j.ijantimicag.2014.01.023 16. Deltombe O, Mertens T, Eloot S, Verstraete AG (2019) Development and validation of an ultra-high performance liquid chromatography – high resolution mass spectrometry method for the quantification of total and free teicoplanin in human plasma. Clin Biochem 65:29–37. https://doi.org/10.1016/j.clinbiochem.2018.12.010 16. Figures Page 11/14 Figure 1 Flow chart of the selection of patients who underwent TEIC and selection by onset or non-onset of rena dysfunction. Figure 1 Flow chart of the selection of patients who underwent TEIC and selection by onset or non-onset of renal dysfunction. Page 12/14 Page 12/14 Figure 2 Figure 2 Comparison of minimum, maximum, and mean total trough concentrations in patients with onset and non-onset renal dysfunction (a). Comparison of minimum, maximum, and mean trough concentrations of the predicted free blood concentration in patients with onset and non-onset renal dysfunction (b). Comparison of minimum, maximum, and mean total trough concentrations in patients with onset and non-onset renal dysfunction (a). Comparison of minimum, maximum, and mean trough concentrations of the predicted free blood concentration in patients with onset and non-onset renal dysfunction (b). Comparison of minimum, maximum, and mean total trough concentrations in patients with onset and non-onset renal dysfunction (a). Comparison of minimum, maximum, and mean trough concentrations of the predicted free blood concentration in patients with onset and non-onset renal dysfunction (b). Page 13/14 Figure 3 Receiver operating characteristic (ROC) curve for the onset of renal dysfunction using the predicted mean ree blood concentration Figure 3 Receiver operating characteristic (ROC) curve for the onset of renal dysfunction using the predicted mean free blood concentration. Receiver operating characteristic (ROC) curve for the onset of renal dysfunction using the predicted mean free blood concentration. Receiver operating characteristic (ROC) curve for the onset of renal dysfunction using the predicted mean free blood concentration. Page 14/14 Page 14/14
https://openalex.org/W4246862860	https://www.ocl-journal.org/articles/ocl/pdf/2014/03/ocl140016.pdf	French	null	Introduction	Oilseeds and fats, crops and lipids/OCL. Oilseeds & fats crops and lipids	2,014	cc-by	538	OCL 2014, 21 (3) D301 c⃝B. Guy-Grand, Published by EDP Sciences 2014 DOI: 10.1051/ocl/2014016 OCL 2014, 21 (3) D301 c⃝B. Guy-Grand, Published by EDP Sciences 2014 DOI: 10.1051/ocl/2014016 Introduction – Dossier Open Access VITAMIN D, VITAMIN OR HORMONE? La vitamine D, vitamine ou hormone ? VITAMIN D, VITAMIN OR HORMONE? La vitamine D, vitamine ou hormone ? Introduction Dossier Bernard Guy-Grand Président du Groupe Lipides et Nutrition Le Groupe Lipides et Nutrition a organisé sa journée scientiﬁque en novembre 2013 autour de la vitamine D. La vitamine D est connue depuis longtemps et largement utilisée depuis des décennies : peu d’enfants n’ont-ils pas, il y a plus d’un demi siècle, absorbé de l’huile de foie de morue pour assurer la prévention du rachitisme et des préparations pharma- ceutiques au goût plus neutre font maintenant partie des prescriptions systématiques. À la lumière des progrès réalisés depuis un peu plus de dix ans dans la connaissance des mécanismes d’action de la vitamine D qui exerce des eﬀets endocrines, paracrines et intracrines sur toute une série de tissus, il était semble-t-il utile de faire un point, non seulement sur ses eﬀets osseux qui demeurent majeurs, mais aussi sur de nombreux processus importants pour l’homéostasie générale. g Même s’il existe encore des débats sur la façon de déﬁnir les déﬁcits en vitamine D, le problème est d’autant plus impor- tant qu’il semble bien qu’une grande partie de la population soit plutôt déﬁcitaire et que le rôle de cet état puisse jouer dans l’émergence de nombreux désordres physiopathologiques, d’où le problème des supplémentations. g p y p g q , p pp Les lecteurs d’OCL trouveront dans ce numéro les articles rédigés par l’intégralité des orateurs in – Vitamine D : sources, métabolisme et mécanismes d’action Jean-François Landrier – Vitamine D et santé osseuse Véronique Coxam, Marie-Jeanne Davicco, Yohann Wittrant – Vitamine D et santé osseuse Véronique Coxam, Marie-Jea – Vitamine D et santé osseuse ique Coxam, Marie-Jeanne Davicco, Yohann Wittrant – Actualités sur la vitamine D Jean-Claude Souberbielle – Actualités sur la vitamine D Jean-Claude Souberbielle – Vitamine D, recommendations, fortiﬁcation in France, and communication – Vitamine D, recommendations, fortiﬁcation in France, and communication Amélie Dhaussy – Vitamine D and cancer Jean-François Doré, Marie-Christine Chignol – Vitamine D and cancer Jean-François Doré, Marie-Christine Chignol – Les eﬀets neurocognitifs de la vitamine D chez la personne âgée Cédric Annweiler – Vitamine D et santé cardiovasculaire Marie Courbebaisse, Catherine Cormier – Rôle de la vitamine D et risque de maladies auto-immunes/cancers – Rôle de la vitamine D et risque de maladies auto-imm Laure Esterle, Anya Rothenbuhler, Agnès Linglart Laure Esterle, Anya Rothenbuhler, Agnès Linglart La prochaine journée du Groupe Lipides et Nutrition, le 18 novembre 2014, sera consacrée au cholestérol. Introduction La prochaine journée du Groupe Lipides et Nutrition, le 18 novembre 2014, sera consacrée au cholestérol. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecom which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is p
W4390062472.txt	https://journals.wlb-stuttgart.de/ojs/index.php/zh/article/download/8733/8634	de		Dr. Friedrich August Lehner	Zollerheimat	2,023	cc-by	5,276	Dr. Friedrich August Lehner Fürstlic Hohenzollerischer Hofrat, Bibliothekar und Museumsdirektor 1864—1894 Von Prof. Dr. Hans Lehner in Bonn gs wird wohl nur noc wenige ßeute in Sigmaringen ober sonst in Hohenzollern geben, die sic des rüstigen alten Herrn entsinnen werden, den man täglic zu einer bestimmten Nachmittagsstunde feinen Erholungsspaziergang am Mühlberg und Brenzkoferberg in Sigmaringen machen sehen sonnte, während er sonst den übrigen Tag in feinem schönen Amtszimmer in der Fürstlichen Hofbibliothet mit amtlichen und wissenschaftlichen Arbeiten ausfüllte, jenes bescheidenen Mannes, der jeder Streberei und Selbstüberhebung abhold, schlicht und freundlich, besonders gern mit einsamen ßeuten in ihrer Sprache, die er als Schwabe beherrschte, sic unterhielt, dessen liebste Erholung in jüngeren Jahren gelegentlich ein einsamer Pürschgang ober ein Jagdausflug mit weidgerechten Jägern war, unb ber nichts jein wollte, als ein treuer Beamter feines fürstlichen Herrn, an bem er mit innigster Verehrung hing. Allgemein besannt unb beliebt war er, aber nur bie Wenigsten ahnten, das er weit über bie Grenzen Hohenzollerns unb Schwabens hinaus in ganz Deutschland unb auc im Ausland in wissenschaftlichen Streifen als Autorität ersten Ranges allseitiger Achtung unb hoher Verehrung genoßz, ber sich auc wissenschaftlic Andersdenkende rückhaltlos anschlossen, weil bie Lauterkeit seiner Gesinnung, bie vornehme Sachlichkeit unb Serabheit seines Charakters, bie Liebenswürdigkeit feines arglosen Wesens jeden in ihren Bann schlugen, ber mit ihm näher besannt wurde. Seboren am 10. Oktober 1824 in bem Dörfchen Geislingen bei Balingen in Württemberg, also dicht an ber hohenzollernscheu Grenze, als Sohn eines kleinen württembergischen Forstbeamten, wuchs er als Ältester einer zahlreichen Kinderschar in schlichten, ja bürftigen Verhältnissen auf. Da ber be= gabte Knabe aber für ben geistlichen Staub bestimmt wurde, so fanb sic bie Möglichkeit zum Gesuch ber Lateinschule in Balingen unb bann bes Gymnasiums in Rottweil, nach desjen Absolvierung er bie Universität Hübingen bezog, um zunächst Theologie zu ftubieren. Wenn er bann glücklicherweise rechtzeitig erkannte, ba^ ber geistliche Beruf ihm nicht lag, so mar das theologische Stubium hoch nicht vergeblich gewesen, sondern sollte, wie mir sehen merben, noch in feinen späteren Jahren reiche Früchte zeitigen. Aber zunächst manbte er sich jetzt ber Philologie unb Archäologie zu, unb eignete sich in ber ausgezeichneten Hübinger Schule jene gründliche philologische Methode an, bie ihm später bei feinen wissenschaftliehen Arbeiten in so vortrefflicher Weise zustatten kam. Im Jahre 1847 ging er von ber Unioerfität ab unb muhte sich, ber Not gehorchend, bem, mie er selbst sagte, nicht immer dornenlosen Beruf bes Erziehers in abligen Läufern widmen, ber ihn über Stuttgart unb München nach Wien führte. Mit rastlosem Fleiß setzte er überall neben ber beruflichen Tätigfeit seine archäologisch-kunstgeschichtlichen Stubien fort, sobah es ihm nac Überwindung ber materiellen Schwierigkeiten endlich 1863 möglich mar, an ber Unioerfität Leipzig bie Doktorwürde zu erlangen. In Wien hatte er bas Glück, in einen Kreis bebeutenber Männer ber Kunst unb Wissenschaft eintreten zu bürfen, bereu Umgang zum Heil unmittelbar bestimmend für feine Zukunft mürbe. Als besonders fruchtbar hat er stets eine Studienreise zur Aufnahme ber Bau unb Kunstdenkmale nach Ungarn betrachtet, bie er unter ber Oberleitung bes berühmten Gotiters unb Dombaumeisters am Stefansbome Schmid unternehmen bürste, unb bereu Ergebnisse er in einer Reihe von Schilderungen herausgab, bie unter bem Titel „Oberungarische Stäbtebilber" in ber Dsterreichischen Revue im Anfang ber sechziger Jahre erschienen. In ber ersten Hälfte bes Jahres 1864 bürste er sich unter ber Oberleitung von Eitelberger unb o. Falke als Hilfsarbeiter an ber Begründung unb Einrichtung bes Österreichischen 24 Museums für Kunst unb Industrie beteiligen, welche Tätigseit bann ber unmittelbare Anlaß zu feiner Berufung an bas Museum bes Fürsten von Hohenzollern nac Sigmaringen mürbe. In Wien hatte er inzwischen auch feine LebensGefährtin gefunben, bie ihm bie Überfiebelung aus ber Weltstabt in bie enge Sphäre ber Kleinstadt Sigmaringen erträglic zu machen in erster ßinie berufen mar. Die Verhandlungen mit bem fürsten Karl Anton von Hohenzollern begannen schon im Januar 1864. Vom 19. Januar basiert bas erste ausführliche Schreiben bes Fürsten an ßehner betreffenb bie Übernahme ber Bibliothekar- unb Konfervatorstelle in Sigmaringen. Die meitere Korrespondenz führte bann zur Anstellung am 1. Juli 1864, mo ßehner bie Aufgabe zufiel, in bas bereits fertige Kunstsammlungsgebäube, nachdem es von Prof. Andreas Müller aus Düsselborf ausgemalt mar, mit ben Sammlungen einzuziehen. So entftanb jene Kunstsammlung, melche mit unwesentlichen Veränderungen in ber Form verblieb, bie ßehner ihr gegeben hatte, bis zu ihrer Auflösung im Jahr 1928. Ein in ber Leipziger Illustrierten Zeitung von 1872 Nr. 1517 erschienenes Bild bes stattlich wirkenden, an eine gotische Kirche erinnernben Raumes gibt schon ben harmonischen Gesamteindruc mieber, ber jebem Besucher bis in bie letzten Jahre bes Bestehens ber Kunstsammlung oertraut mar, unb beffen vornehmer Festlichkeit sich niemanb entziehen sonnte. Spätere Hände haben kleine Veränderungen her Aufstellung vorgenommen, mie sie sic aus bem Zuwachs unb Abgang jeder Sammlung von selbst ergeben, aber bie Gejamtaufstellung mar unb blieb oon Lehner bestimmt. So schön unb feierlich dieser Gesamteindruc ber Kunsthalle mar, so sonnte es bem an moberne Museumsbauten geroöhnten Beschauer, namentlieh bem Fachmann nicht entgehen, bah Manches nicht so zur Geltung fam, mie man es seinem Wert unb feiner Bedeutung nach gewünscht hätte. Besonders bie kostbaren Altartafeln ber schwäbischen Malerschule mußten stets unter der nicht sehr glücklichen Gestaltung bes Gebäudes leiben, inso- fern sie, an den einzig für ihre Aufstellung in Betracht kommenden hohen Langwänden des kirchenschiffartigen Raumes eine ungünstige Beleuchtung hatten. Aber das war nun einmal ein Mangel, der daraus entsprang, daß der neuernannte Museumsdirektor in einen ohne fein Zutun fertig gestellten Raum einziehen und sic nun mit der Schöpfung des Architekten abfinden muszte, — ein Los, welches das Sigmaringer Museum bekanntlich mit manchem anderen zu teilen gehabt hat. Auc die tastbare Holzplastif hätte wohl durc eine andere Raumgestaltung besser zur (Bettung fommen sönnen, als es unter den gegebenen Umständen möglic war. Um so glänzender und erfreulicher war die Aufstellung aller der Abteilungen, welche von der architektonischen Gesamtgestaltung weniger abhängig maren, der Kleinkunst, des Kunstgewerbes. Da baten in den vornehm-schlichten Vitrinen bes Hauptraumes die schönen, rheinischen, schwäbischen, fränfischen keramischen Sammlungen, die rheinischen und Limager Schmelzarbeiten, barunter einzigartige Stücke von unschätzbarem Wert, bie Gläser, Elfenbeinschnitzereien unb runtwaffen sic bem Beschauer in voller Pracht bar, unb in zwei intim ausgestatteten Seitentabinetten fanb man italienische Majoliken unb erlesene Stücke ber Email-, Elfenbein- unb Metallkunst aus ber Gesamtheit herausgehoben. Das Ganze, durchaus auf aesthetische Wirkung gestimmt, spiegelte in glücklichster Weise ben vornehmen, streng sachlichen, aber einer feiner sozialen Stellung entsprechenden unaufdringlichen Prachtentfaltung zugeneigten Sinn bes Schöpfers ber Sammlung, bes Fürsten Karl Anton wider, bem Lehner nun nac Beendigung ber Aufstellung ben in ber Kunsthalle verewigten Wahlspruc in ben Mund legte: Mit Gott hab ic dies Haus erbaut, Manc Kleinod hab ic ihm vertraut, Ic freu mic des, doch nicht allein, Auc du sols mir willkommen sein. All was du siehst, is dein wie mein, Auc mic kann ja das Schau’n nur freu’n, Zwar, nähmst du was, hätt ich's nicht gern, Doc nimm das beste mit dir: lern! Diesem liberalen, menschenfreundlichen Grundsaß entfpre= chend, wurde in einer weit entgegenfommenben Besuchsordnung bie Kunstsammlung, bie ja doc eigentlich engster PBrivatbesitz bes fürstlichen Hauses war. Jedem zugänglich gemacht, ber Kunst genieren unb ftubieren wollte, wovon freilic bie Bewohner bes Residenzstädtchens nur sehr bescheideneu Gebrauch machten, während bie auswärtigen Besucher bes schönen Donautales in immer fteigenbem Masze bie herrliche Kunstsamlmung bes Firsten von Hohenzollern zu schätZen wuszten. Aber nicht bie Kunsthalle allein mar ben Besuchern geöffnet. In nächster Nachbarschaft mar eine in wuchtigen, schweren Architekturformen gehaltene Halle bes Schlosses selbst ber Waffensammlung eingeräumt. Da stauben Ritterrüstungen, ein gepanzerter Ritter zu Pferde, Schießz-, Hhieb- unb Stichwaffen aus allen Jahrhunderten bes Mittelalters unb ber Neuzeit, Trommeln unb anbere kriegerische Musikinstrumente, Fahnen, Stanbarten unb Schilder in malerischer Anorbnung, für bie breite Masse ber Schloszbesucher fast ein noc gröberer Anziehungspunkt, als bie bem allgemeinen Verständnis mehr entrückte Kunstsammlung, welche übrigens eine Auslese ber kunstvollsten Waffenstücke in gefolgerter Aufstellung zeigte. In bem Obergeschoß ber Kunsthalle, leiber aus räumlichen Gründen dem allgemeinen Bejuc nicht zugänglich, fanb bie bebeutenbe Sammlung vaterländischer Altertümer eine übersichtliche, für bas Stubium ber Fachleute vollkommen ausreichende Aufstellung, unb in ben Räumen ber ebenfalls von Lehner georbneten unb verwalteten fürstlichen Bibliothe? sonnte man in großen Pultvitrinen einen nicht sehr umfangreichen, aber erlesenen Schat von alten Handschriften mit herrlichen Miniaturen bewundern. Eben ba mar auc eine Kupferstich- unb eine Münzsammlung untergebracht. Diese Gesamtheit ber kunst- unb kulturgeschichtlichen Schötze ber fürstlichen Sammlung, zu ber noc eine ganze Reihe in ben Räumen bes Schlosses verstreuter Gegenstände gehörte, unb bie, menn in einem einzigen Gebäude vereinigt, ein ganz groszes Museum dargestellt hätte, mar ber Obhut Lehners unterstellt unb bedurfte ber ständigen Bflege, bie bei ber räumlichen Zersplitterung ber einzelnen Abteilungen doppelt schwer mar, beburfte aber zunächst einmal ber Inoentarifierung unb Katalogisierung. Vor allem traf bies auc zu für bie schon ermähnte fürstliche Bibliothek. Sie umfaszte etma 15 000 Bände, als Lehner sie übernahm, unb mar bei feinem Dienstaustritt auf etma 50 000 Bände angewachfen. Schon aus diesem Umfang ist ersichtlich, baß bie Biblivthet nicht, mie manche Sigmaringer annahmen, lediglich einem oberflächlichen Unterhaltungsbedürfnis biente, sondern sie umfaßte, ber Geistesrichtung ihres Begründers entsprechend, vor allem politische, geschichtliche, biographische, kultur- unb kunstgeschichtliche Fachliteratur. Vor allem bürste Lehner sie als bas notmenbige wissenschaftliche Rüstzeug für bie Bearbeitung ber Kunstsammlungen ausbauen. Wer je auf fünfte unb kulturgeschichtlichem Gebiete in ber fürstlichen Bibliothek gearbeitet hat, ber mar erstaunt über bie Reichhaltigkeit, welche sie auf diesem Gebiet an Einzelwerken unb Zeitschriften aufmies. Auc für Altertumskunde mar gut gesorgt, kostbare Serienpublikationen, mie bas ganze Corpus inscriptionum Latinarum, bie Archäologische Zeitung und ähnliche maren vorhanden, vor allem aber viele jener grosfen kostspieligen kunstgeschichtlichen Vrachtwerke, beren jedes Einzelne enorme Summen kostete. In grosszügigster Weise kam hier ber Fürst ben Wünschen unb Bedürfnissen feines Bibliothekars entgegen. In mustergültiger Art hat Lehner bie Bibliothek geordnet unb einen doppelten Zettelkatalog davon angefertigt, unb zwar bewältigte er diese, mie bie nachher erwähnten Arbeiten, ganz allein ohne wissenschaftliche ober technische Hilfskraft unb Schreibhilfe. Nur ein jogenannter Galeriediener, aus bem Bestand ber fürstlichen ^ofbebienfteten entnommen, besorgte bie mechanischen HilfeLeistungen unb bie Führung unb Überwachung ber Sammlungsbesucher. Und nun ging Lehner mit Feuereifer an bie Katalogisierung ber Kunstsammlungen. Bis auf bie Sammlung Vaterländischer Altertümer, für welche ber besonnte vortreffliche illustrierte Katalog von ß. Lindenschmit bereits seit 1860 oorlag, mar für bie Sammlungen noc feine wissenschaftliche Behandlung vorhanden. Das muszte alles von Grund auf neu geschaffen merben. Zunächst wurde ein Gesamtinventar angelegt, unb bann begann bie wissenschaftliche Veröffentlidjung. Bereits 1868 hatte Lehner bei Eduard Ebner in Stuttgart „Fünfzig ber bebeutenberen Gemälde ber fürstlichen Sammlung" in photographischen Reproduktionen herausgegeben. Im Jahre 1871 folgte nun bas genaue beschreibenbe „Verzeichnis ber Gemälde", welches damals 210 Nummern umfaßte unb 1883 in zweiter vermehrter Auflage erschien. Dasselbe Jahr 1871 zeitigte noc bas „Verzeichnis ber Schnitwerke" (377 Stück) unb bas „Verzeichnis ber Tonarbeiten" (550). Im Jahr 1872 erschienen nicht weniger als fünf Oktavbände, bie Emailwerke (72), bie sogenannten Kleinobien (214), bie Gläser (351), bie Metallarbeiten (662) unb bie Handschriften (357). Im Jahr 1874 endlich famen noc bie Verzeichnisse bes Mobiliars, fomeit es zu ben Schaustücken bes Museums gehörte (156) unb ber Tertilarbeiten, nicht sehr zahlreich (59 Stück), aber zum Teil sehr foftbare Stücke aus bem Mittelalter umfassend. Diese sämtlichen Kataloge erschienen in ber Hofbuchhandlung von C. Tappen (später K. Liehner) in Sigmaringen. Sie maren so eingerichtet, bah sie unmittelbar vor ben Gegenständen selbst, bereu Numerierung auf sie Bezug nahm, als „Führer" benutzt merben sonnten. Über bie Weiterentwickelung ber einzelnen Sammlungsabteilungen hat Lehner regelmäßig in ber Museographie ber Westdeutschen Zeitschrift für Geschichte unb Kunst Bericht erstattet. Bald griff aber Lehners wissenschaftliche unb literarische Tätigkeit über bie durc sein Amt unmittelbar gezogene Grenze hinaus. Im Jahre 1868 übernahm er ben Vorsis bes 25 Vereins für Geschichte und Altertumskunde in Hohenzollern, bessert von ihm mitbegründete jährlich erscheinende Zeitschrift er bis 1886 redigierte. Ausgrabungen römischer Gutshöfe in der Umgebung von Sigmaringen haben ihn mehrfach be= schäftigt, und ihre Ergebnisse hat er in der Vereinszeitschrift niedergelegt. Durc eine Menge von Vorträgen und Mitteilungen antiquarisch historischen Inhaltes belebte er die regelmäßigen Sitzungen des Altertumspereins. Die berühmten Zeitblomschen Altartafeln der Pfarrkirche von Bingen bei Sigmaringen veröffentlichte er 1867 in einer Prachtausgabe mit elf photographischen Abbildungen in Folioformat. Kaum besannt bürste es fein, baß Lehner dem Fürsten auc bie Begründung eines naturgeschichtlichen Hohenzollernschen Heimatmuseums vorschlug, wie aus einem Priese des Fürften vom 11. 3. 65, in welchem er diesem plane zustimmte, hervorgeht. Warum bie Gründung nicht zustande fam, ist unbekannt. Bei einer so umfassenden wissenschaftlichen Tätigkeit, welche auc zu einer ausgedehnten Korrespondenz mit Fachgenoffen des In- unb Auslandes führte, sonnte es nicht ausbleiben, daß bie wissenschaftliche Welt mehr unb mehr auf bie Sigmaringer Sammlungen unb ihren ßeiter aufmerksam wurde. Hatte ihm bie verständnisvolle Munifizenz des Fürften häufige Reisen nach Perlin, München, Dresden unb in andere Zentren deutschen Geisteslebens ermöglicht, unb ein halbjähriger Urlaub im Winter 1868—69 ihn nach Südfrankreic unb Italien als Reisebegleiter eines jungen Aristofraten, ben er in Wien kennen gelernt hatte, ,311 einer fruchtbaren Studienreise geführt, wobei er überall auc persönliche Beziehungen zu ben Fachgenossen anzufnüpfen Gelegenheit hatte, so führte anbererfeits ber wachsende Ruhm ber fürstlichen Sammlungen in stets fteigenbem Masze Besucher aus Wachsreifen nac Sigmaringen, bie natürlich niemals versäumten, mit bem ßeiter ber Sammlungen in persönliche Verbindung zu treten. Aber nicht nur auf ben Kreis ber engeren Fachgenossen blieben diese anregenden Besuche beschränkt. Nachdem ber Fürst im Jahr 1871 seinen ftänbigen Wohnsitz oon Düsseldorf wieder nach Sigmaringen verlegt hatte, suchte er sich für bas geistig angeregte ßeben, bas er in ber rheinischen Kunststadt zu führen gewohnt gewesen mar, wenigstens einen kleinen Ersah zu schaffen, indem er geistig hochstehende Menschen aller möglichen Berufsarten zu längeren unb kürzeren Besuchen als feine Gäste empfing. Sigmaringen ist in jenen glücklichen Wahren öfter nicht mit Unrecht mit gewissen italienischen Höfen ber Renaissancezeit verglichen worden. Namhafte ßiteraten, bilbenbe Künstler, Politiker, Gelehrte ber verschiedensten Disziplinen unb Richtungen lösten sich da in bunter Folge ab unb scharten sic um ben geistvollen Fürsten im Genuß einer vornehmen GastFreundschaft, die durc bie reichen Kunstschätze ber Sammlungen unb bie natürlichen Vorzüge ber Gegend noch ihre besonderen Reize erhielt. Mit all diesen Besuchern brachte schon fein Perus auc ben Museumsdirektor unb Bibliothekar in engen Konnex, fobaß bie öbe ber Kleinstadt wenigstens in jenen glücklichen Wahren säum empfunden wurde. Aber bie enge unb bauernbe Perührung mit ben auswärtigen Fachgenossen sollte ßehner noch höheren Gewinn einbringen. Schon 1872 bürste er bie schöne Genugtuung für feine vorbildliche Museumstätigkeit erfahren, daß er in ben Permaltungsrat des Germanischen Nationalmuseums in Nürnberg gemahlt mürbe. War bie Ernennung zum Mitglied einer so hoc angesehenen Körperschaft eine hohe Ehre für ben ßeiter einer fürstlichen Privatsammlung, so mar sie ihm gleichzeitig eine Quelle reichster Anregung, bie er bei ben alljährlich in Nürnberg ftattfinbenben Sitzungen aus bem Meinungsaustausch mit hervorragenden Wachmännern schöpfen bürste. Unb er mar befonbers gerne gesehen in diefern illustren Gremium. War er hoch im Staube, nicht nur mit feiner Erfahrung unb feinem Rat die ideellen Wntereffen des Nürnberger Museums zu förbern, sondern oermöge seiner guten Beziehungen auch materiell manchen großen Nutzen zu stiften, was natürlich befonbers bantbar anersannt mürbe. 26 Schon früh, lange oor feiner Sigmaringer Zeit, hatte sich ßehner mit Fragen ber christlichen Archäologie beschäftigt. Besonders hatte ihn bas Problem bes allmählichen Wachsens unb Werdens bes Marientultus oon feinen ersten Anfängen bis an bie Schwelle bes frühen Mittelalters mehr unb mehr gefesselt. Schon im Wahre 1862 hatte er in ben Mitteilungen ber K. K. Centralkommission VII S. 119 ff. einen Portrag, ben er im Wiener Altertumsverein über bie älteste Entwickelung bes Marienkultus gehalten hatte, veröffentlicht. Aber nun reifte, start geförbert durc feine Studienreise nach Italien, allmählich ber plan 311 einer groß angelegten Monographie über bie Entwickelung bes frühesten Marienkultus, welche in Sigmaringen Gestalt annahm. Wm Wahr 1881 erichien bas Wert „Die Marienverehrung in ben ersten Jahrhunderten" bei W. G. Eotta in Stuttgart mit acht Doppeltafeln in Steinbrucf. Pon ben ersten Erwähnungen Marias in ben frühen kanonischen Evangelien an mirb bas Werden unb Wachsen ihres Bildes in ber gläubigen Phantasie der ersten Eßriften, welches bann namentlich bie Stählungen ber apokryphen Evangelien ftarf befruchteten, bis zu ben gelehrten Spekulationen ber Kirchenväter 00m 2. bis 4. Jahrhundert unb bis zum Konzil von Ephesus (431) verfolgt. Unb es mirb bann gezeigt, mie oon diesen Borstellungen, bie sic auf bie Eigenschaften Marias als Wungfrau, Mutter, Josephs Weib etc. bezogen, bie frühchristliche Poesie unb endlich bie frühchristliche bilbenbe Kunst beeinflußt wurben. Es ergab sich, baß zwar bie Poesie schon früh von fast allen Ausgestaltungen, bie bas Marienideal in ben erwähnten Schriften erhalten hatte, Nahrung empfing, baß dagegen bie frühe bilbenbe Kunst sich auf einen ziemlich engen Bezirk dieser Vorstellungen beschränkte. Die Mutter mit bem Kinde ist bie früheste unb häufigste Darstellung, meist aber nicht isoliert für sich, sondern als Teil einer Komposition, entweder mit Joseph ober einem Propheten, ober am häufigsten mit ben Magiern, seltener bei ber Darstellung im Tempel, ber Auffindung" bes zwölfjährigen Jesus im Tempel, ober mit bem erwachsenen Sohn beim Weinwunder oon Kana. Es fommen bann, meit seltener, Darstellungen ber Verfünbigung, ber Vermählung, ber Heimsuchung, ober ganz allein als „Orans" oor, mobei allerdings bie Deutung nicht immer gesichert ist. Ebenso steht bie Deutung vereinzelter Darstellungen als Wungfrautönigin im Himmel nicht ganz fest. Es kann natürlich in diesem Zusammenhang ber Inhalt bes Werkes nur ganz flüchtig skizziert merben. Es mar eine streng wissenschaftliche kunstarchäologische Untersuchung, feine Apologie bes Marienkultus, aber auc feine pelemif gegen ihn. Mit mohltuenber Objektivität ist bas heikle Thema behandelt, unb diese Art ber Behandlung ist auc seinerzeit in fast allen Besprechungen oon Kritikern aller Konfessionen unb Richtungen anerfannt morben. Das Puch erregte beim Erscheinen in Wachsreifen bebeutenbes Aufsehen, von bem bie mehr als 50 Besprechungen, bie bem Verfasser besannt wurben, ein höchst ehrenvolles Zeugnis ablegen. Rastlos hat ber Verfasser auch später an bem Wert meiter gearbeitet, eine zweite Titularauflage 00m Wahr 1886 brachte schon allerlei Änderungen unb Zusätze, unb fein hinterlassenes Handeremplar bemeift durc zahlreiche handschriftliche Eintragungen, baß ßehner sich mit einer brüten Auflage be= schäftigte, zu ber es aber nicht mehr gefommen ist. Gewiß ist in ben runb 50 Wahren seit bem Erscheinen bes Werkes oieles burch neuere Forschungen überholt, in manchem haben sic bie wissenschaftlichen Anschauungen über ben Gegenstand grundsätzlich geändert; aber für feine Zeit bebeutete biefe erstmalige Zusammenfassung einen Markstein in ber frühchristlichen archäologischen Forschung unb ist als solcher auc gemertet morben. Die frühchristliche Archäologie, welche namentlich auf bem Gebiet ber Katakombenforschung gerabe in ben achtziger Wahren einen ftarfen unb glücklichen Auftrieb erhielt, ließ ßehner bis zu feinem Enbe nicht mehr los. Eine ganze Reihe oon Aufsätzen, meist in Form ausführlicher Besprechungen, neuerschienener Quellenroerfe, erschien aus feiner Feder größtenteils in ber wissenschaftlichen Peilage zur Münchener Allgemeinen Zeitung bis in die letzten Monate feines Lebens mar es bem Siebziger vergönnt, fein otium cum dignitatè hinein und gab ihm auc Gelegenheit, feinen grundsätzlichen in Stuttgart, wohin er sich zurückgezogen hatte, zu geniessen. Standpunkt auf dem Gebiete der Katakombenforschung Mitten in wissenschaftlicher Arbeit raffte ihn eine Lungengegenüber gewissen Richtungen, die ihm abwegig schienen, entzündung, bie er sich im (Befolge einer Erkältung 3130g unb ber bas franse Herz nicht mehr Widerstand leistete, am zum Teil mit grober Schärfe zu formulieren. Der Fürst Karl Anton und ebenso sein Nachfolger Fürst 3. Juni 1895 hinweg. — Noc an feinem Todestage biftierte er, ber sein Ende Leopold förderten in jeder Weise diese über die unmittelbare Amtspflicht hinausgehende wissenschaftliche Tätigkeit, die mit stoischer Ruhe in vollem Bewusztsein herankommen nur dazu angetan war, bas Ansehen ber Stelle selbst nac fühlte, einige kurze fachliche Eingaben über sein Leben m t auszen hin zu heben, was auc Karl Anton ausdrücklich in bem Wunsch, bah sie „ohne überflüssige weitere Betrachtuneinem an Lehner gerichteten Brief ausgesprochen hat. Ein gen" nach seinem Tode in ber Beilage ber Münchener Algereger über 20 Jahre sic hinziehender Briefwechsel dieses meinen Zeitung erscheinen sollten. Sein Wunsc ist wörtlic Fürsten mit Lehner gibt überhaupt ein schönes Bild des erfüllt morben, sein Diktat erschien ohne Zusätze in Dir 120 hohen Vertrauens, beffen ber Fürs feinen Beamten wür- bes genannten Blattes vom 6. Juni 1895. Lehner hat wohl in ber Überzeugung, bah sein Lebensbigte. Um nur ein Beispiel für viele anzuführen, so fei hier werk, bas Sigmaringer Museum, stets von ihm unb feiner zum ersten Mal ber Öffentlichkeit mitgeteilt, daß Lehner vom Fürsten Karl Anton 1870 beauftragt mar, bem König Wil- Tätigfeit zeugen merbe, auf einen ausführlichen Nachruf heim bie Verzichtleistung des Erbprinzen Leopold auf ben verzichten zu sönnen geglaubt. Aber er sollte nicht Recht bespanischen Thron persönlich nach Ems zu überbringen. Unter halten. Daß mehr unb mehr fein Andenken gleichzeitig mit ben hinterlassenen Rapieren Lehners fanb sich die eigenhän- bem Verschwinden feiner Zeitgenossen unb Freunde dahinbige Instruktion des Fürsten für dieses Mission, basiert vom fchmanb, bas ist Menschenlos, bas nicht zu ändern ist. Das 13. Juli 1870. Mit hingebender Verehrung und Treue und Bitterste ist ihm wenigstens zu feinen Lebzeiten erspart gemit rückhaltloser Offenherzigkeit mürbe bas Vertrauen des blieben: bah das Werk selbst, das er geschaffen unb an bem Fürsten von feinem Beamten erwidert. Das Verhältnis Leh- er mit ganzem Herzen gehangen hatte, untergehen sollte. ners zu feinem Vorgesetzten läszt sich am besten burch ben Erst im Jahr 1928 ist bies Letzte erfolgt: bas schöne, koitbare, einzigartige Sigmaringer Kunstmuseum ist in alle Winde Vers aus Goethes Tasso kennzeichnen: Zerstreut! — Wir mollen nicht mit bem Schicksal rechten; „Gar leicht gehorcht man einem edlen Herrn, in ber traurigen Zeit, bie mir alle durchleben muszten, ist so Der überzeugt, inbem er uns gebietet". Nac bem Tode Karl Antons 1885 hat Lehner beffen Nach- vieles zerbrochen unb eingestürzt, was ein monumentum folger Leopold noc bis zum Jahre 1894 treu gebient unb aere perennius zu fein geschienen hatte, unb bies Los hat muhte im Frühjahr feines vorletzten Dienstjahres 1893 noch eben auc bie Schöpfung Friedric August Lehners teilen ben Schmerz erleben, bas alte schöne Felsenschloß in Sigma- müssen. Aber menn fein Wert auch verschwunden, sein ringen burch einen furchtbaren Brand in Trümmer unb Schrifttum vergessen, sein Andenken fast ausgelöscht ist, so Ajche sinken zu sehen. Wie burch ein Wunder blieb bie mit missen mir ja doch, bah, wer ben Besten feiner Zeit genug viel leicht entzündlichem Material gebaute Kunsthalle unb bie getan, für alle Zeiten gelebt hat; ein versöhnender Gedanke, Bibliothek verschont; mit schwerer eigener Lebensgefahr lei- ber auch von fern anklingt aus ben schönen unb tiefen Wortete ber 69jährige Direktor bie ganze Nacht hindurch bie ten, welche einst Carmen Sylva in bas Fremdenbuc ber Rettungsarbeiten unb bie vorläufige Bergung ber am mei- Sigmaringer Kunstsammlung geschrieben hat: „Jeder gebe sein Bestes her ftcn gefährdeten Kostbarkeiten. Neuaufitellungs- unb NeuUnb meihle so kühn, unb feile so zart, versicherungsarbeiten füllten ben Rest feiner Dienstzeit reichUnb merfe getrost fein Tun ins Meer — lich aus, dann abcv zwang ihn, der sonst körperlich unb vor Einst kehrt es mieber unb leuchtet hehr. allem geistig noch sehr rüstig mar, unb ben fein Herr nur sehr Von mogenber Ewigkeit umschart". ungern gehen ließ, ein sic mehr unb mehr fühlbar machenI (Q uellen: Familienpapiere, Nachlasz.) bes Herzleiden, sein Amt niederzulegen. Nicht ganz ein Sahr Kleine Mitteilungen „zur botanischen Erforschung Hohenzouerns" freuen mir uns — als Vachtrag zu S. 13 — darauf hinweisen zu können, bah soeben als Ersaß für ben veralteten Kirchner unb Eichler ein ganz neues Bestimmungsbuc von K. unb 3 r. Bertic „Flora von Württemberg unb Hohen:zollern" (1933, 311 S.) erschienen ist, bas ganz Hohenzollern umfaßt unb bestens zu empfehlen ist (vgl. bie Besprechung). — D. Schmeils „Flora von Deutschland" ist 1932 in 44. EdiHon mit 449 Seiten unb 1000 Abbildungen herausgekommen. Hohenzolerijche Münzen, Ziedailien usw. Den Freunden hohenzollerischer Münzen, Medaillen, Drben usw. zur Nachricht, bah die Hohenzollerische Heimatbücherei Literatur zu diesem Sammelgebiete besitzt, bie neben ben menigen Druckschriften Hinweise auf Zeitichriftenartikel ud sonstige Fachliteratur unb bisher unveröffentlichte Arbeiten umfaszt. Bejonders in ben zuletztgenannten Schriften ist versucht, bas ganze Sammelgebiet zu ersassen unb ben wirklichen Liebhabern zu gänglic zu machen. Vor allem soll bamit bie Freude an einem schönen unb wertvollen Sondergebiete ber Heimatkunde geweckt merben, das bisher bei uns start vernachlässigt mürbe. Wir sind natürlich bestrebt, unsere Materialiensammlung zu erweitern unb bitten alle Freunde ber Gache um Unterstützung, wie auch mir gerne Auskunft geben. Für Mitarbeiter möge als Anhalt bienen, daß in ber Buchliteratur bisher berücksichtigt finb: Die Münzen, aber nicht erschöpfend; bas Notgeld nur zum Teil; Medaillen bes Fürstlichen Hauses, doc äuszerst lückenlaft; Medaillen ber Vereine unb Ausstellungen (also Schützentaler und Preismedaillen aller Art, sowie auc Erinnerungsmedaillen) über baupt nicht; Orden unb Ehrenzeichen, bie „bene merenti , Fürstliche Erinnerungszeichen unb Geschenkstüicke nur höchst bürftig. Sobalb über einzelne Abteilungen genügend Material beiJammen ist, soll an dieser Stelle barüber eingehend berichtet merben. S). Fazbender. Zur „Vorgeschichte der Abtretung Hohenzolerns an Preus sen“ macht Herr Dr. G. Hebeije n (Mitteil., 62., 1931, 53 62) ben Versuch, bie Abtretung auf bie Verschuldung bes Hauses hohenzollern - Hechingen zurüczuführen, bie burch bas Eintreten dieses Hauses „für Kaiser unb Reic unb vor allem bes katholischen Wesens halber" (S. 53 unb 62) verur facht märe. Die Feststellung ber Verschiebung unb ihr Zurückgehen auf bie Tätigfeit bes Fürsten Johann Georg im Dienste ber Kaiser ist in banfensmerter Weise völlig geglückt. In gar seiner Weise jedoc ist es bem Verfasser gelungen, obige These irgendwie zu belegen. Was seststeht ist lediglich biefes, bah bie Geldnöte ber Hechinger Linie zu einem Briefwechsel mit Brandenburg führten, in beffen Verlauf biefes, nachdrücklich auf bie bestehende Verwandtschaft unb bie daraus von Hechingen abgeleitete Unterstützungspflicht aufmertfam gemacht, so praktisch mar, an Successionsperträge zu denken (S. 55 56) unb diese auc 1695 unb 1707 zu erreichen. 27 Aber nicht nur enthalten diese Verträge selbst (58 59) feine Abmachungen über finanzielle Unterstützungen, sondern, was viel mehr bedeuten will, zwischen 1670 und 1731 vermag H. nirgends eine solche Hülfeleistung anzuführen, während die Verschuldung dauernd steigt. Ist es auch möglich, daß Hechingen bei dem Vertrage an finanzielle Vorteile gedacht hat, das völlig schuldenfreie Sigmaringen hat dies nicht und es ist doc naheliegender, baß dieselben Gründe, bie Sigmaringen zu dem Vertrage bestimmten — welc einen politischen Rückhalt bot das mächtige Brandenburg ja doc auf alle Fälle! —, auc für Hechingen wesentlich mehr maßgebend waren als feine Schulden, um beren Begleichung es mit ober ohne Vertrag bitten sonnte und gebeten hat. Jedenfalls: für ben wichtigsten Teil Hohenzollerns, für Sigmaringen, bas ja g e r a b e unb o o r allem unser ßanb 1850 abtrat, spielt bie Hechinger Verschuldung als Motiv für ben Vertragsabschluß gar feine Rolle unb bamit bricht bie ganze Beweisführung H ‘ 5 völlig 3 u f a m m e n ! Unb es ist historisch niemals zulässig, in ber bloszen Tatsache, baß die Verschuldung Hechingens dieses 1663 zu einer wenn auc noc fc wichtigen Unterstüzzungsbitte an Brandenburg bewog, nun b i e „Ursache für bie Abtretung" (S. 53 u. 62) zu erblicken. Solche inbirefte „Fernwirkungen" finb spielende Konstruktionen! Der am Ende gegebene kurze überblick über bie Gründe ber Revolution von 1848 49 (ber so oiel geschmähte Würth ist übrigens sein Konstanzer) zeigt nur, wie bringenb not uns eine exakte, archivalisc belegte, b. h. q u e 11 e n m ä f f i g e Darlegung dieser Periode, ihrer Verwaltungs-, Berfassungs- unb Rechtsgeschichte, ihrer wirtschaftlichen unb sozialen Zustände, ihrer ganzen Geistesgeschichte, tut. Von bem Verbleib ber Hechinger Schulden (nur bie an Muri werden 1885 nochmal genannt) hören mir leiber 1731 zum leiten mal. Hat Hechingen sie nicht durc bie reiche Heirat mit ber Leuchtenbergerin zum Teil löschen sönnen? Hat Sigmaringen sie (wie es bei ber an Muri geschah) übernommen unb warum? Hat Preuszen 1850 dabei geholfen? Jedenfalls, bie ganze Abtretungsfrage bebarf erst noch ihrer allseitigen Unterjuchung unb Klärung! Da bie Arbeit S’s bisher ohne Einwände hingenommen wurde, möchten biefe Zeilen dazu bienen, eine historische Legenbenbilbung über ihr Thema zu verhindern. Dr. Senn. zum Bestimmen nichtblühender Wasserpflanzen. Für bie Braris bürsten bie beigegebenen Bestimmungsschlüssel für bas $013 ber einheimischen ßaubbäume unb für bie Getreidearten unb Wiesengräser nac ben Blättern von besonderem Wert fein. Die neue Flora mirb für alle, bie sich von Berufs megen ober aus Freude unb ßiebhaberei mit unserer heimischeu Pflanzenwelt beschäftigen, unentbehrlich fein. Dr. Willi Maier, Tübingen, Votan. Institut. Hoffmann, G.: Kirchenheilige in Württemberg (Darstellungen aus ber Württ. Geschichte, 23. Bd.). (Stgt., Kohlhammer, 1932, 8 ", 325 S., 1 Karte, 6 M.) Der württembergische Pfarrer unb Kirchenhifiorifer Bossert ist es gewesen, ber bereits 1893 auf bie grunblegenbe Bedeutung ber Patrozinienforschung hingewiesen unb selbst Bahnbrechendes barin geleistet hat. Was bie Ortsnamen für bie älteste Giedelungsgeschichte bebeuten, bas finb bie Batrozinien für bie älteste und ältere Zeit ber Christianisierung unb kirchlichen Organisation, für die anbere Quellen oft fehlen. Hierin liegt ihre unschätzbare Wichtigkeit! Bosserts Anregungen unb feiner Einleitung ist nun Hoffmann gefolgt und hat in mühevollster Kleinarbeit bas vorliegende Wert, bas erste über ein größeres Gebiet Deutschlands, geschaffen. Den Hauptteil bes Buches (45/258) nimmt, nac Bistümern georbnet, bie Zusammenstellung ber Orte ein, bei benen für jede Kirche, jebe Kapelle, jeben Elitär (!) ber dazugehörige Heilige, bem sie geweiht maren, mit genauem Beleg für bas erste urfunbliche Vorkommen desselben ermittelt ist. Ein Heiligenregister mit Hinweis auf bie Orte unb ein Ortsregifter erschliessen bas Geleistete vollends, bas auc burd) eine Karte ber Diözesans- unb Landkapitelgrenzen mit Hervorhebung ber Martins- unb Michaelskirchen illustriert mirb. Die Unsumme an Arbeit, bie hier geleistet wurde, ist staunenswert! interessant finb bie Hinweise, bie $). für bie Methodik dieser oft sehr schmierigen Ermittelungen (Benütjung von Archivalien, Inschriften, Skulpturen unb Malereien, Gewölbeschluszsteinen, Orts- unb Flurnamen, Siegeln etc.) gibt. 3n berfelben Einleitung (1 44) gibt er auch feine Hauptergebnisse, bespricht bie Gründe für bie Wahl ber Batrozinien, erörtert heidnische Einflüsse (Michaels- u. Stephanuskirchen) unb solche ber Fränkischen Besiedelung (Martinskireßen — Vebeutung für alte Straszenzüge!), er zeigt bie Bebeutung ber alten Klöster vom 8. Jahrhundert ab unb finbet in den Patrozinien ben Niederschlag all ber verschiedenen Besprechungen Geistesströmungen ber mittelalterlichen Kirche. Auc bie Bertic, Rar unb Franz: Flora von Württemberg unb Ho- schwierige Frage bes Batroziniumswechsels mirb eingeßenb henzollern. Zum Gebrauche auf Wanderungen, in Schu- besprochen. — Hohenzollern ist nur auf ber Karte mitbehanlen unb beim Selbstunterricht. (München, ßehmann, 1933, belt unb naturgemäß unvollständig. Dennoch ist bas Buc auch für unsere Kirchengeschichte grundlegend unb ein zwin311 S., 55 Abb., 6.80 M.) Nachdem bie alte Kirchner unb Eichlersche „Erkursionsflora genbes Vorbilb: es verlangt bringenb nach einer Ergänzung für Württemberg unb Hohenzolern" vergriffen mar, gab es für bas hohenzollerische unb badische Gebiet! Im Zusammenoiele Jahre lang fein umfassendes Bestimmungsbuc flang bes ganzen Südwestens werden sic weitere ErgebDr. 5. Stehle. für bie Pflanzenwelt unserer engeren Heimat. Durch bie in niffe herausschälen lassen. diesem Frühjahr im Verlag I. F. Lehmann, München, erWissenschaftliche Anfragen schienene „Flora von Württemberg unb Hohenzottern" von K a r l unb F r a n 3 B e r t f c ist nun ertreulicherweije dieBoltsliedfammlung in Hohenzolern fer seit langem empfunbene Mangel behoben morben. Zur Förderung ber hohenzollerischen Volfsliedforschung Das überaus reichhaltige, aber trotzdem sehr handliche bitte all unsere Heimatfreunde um Nachforschungen nac Buc umfaszt bie Gefäszkryptogamen unb bie Blütenpflan- alten ic Liederheften vereinzelten Aufzeichnungen von Liegen. Zahlreiche ausführliche Bestimmungsschlüssel ermög- bern, beren genaueunb Sichtung durc Fachleute manch brauchließen, zum Teil mit Hilfe von Abbildungen, ein leichtes unb baren Fund ergeben sann. Vefonbers in ben schriftlichen sicheres Bestimmen ber einzelnen Arten. Für ben mit ben Nachlässen unserer Lehrer, in Schul-, Musikbotanischen Srunbbegriffen weniger Vertrauten mirb ber vereinsbüchereien bürste sich vielleicht noch diesunbundGesangjenes Gebrauch des Buches durc eine eingehende Erklärung ber finben. Bei leihweiser Überlassung mirb balbige Rücksendung verwendeten Fachausdrücke wesentlich erleichtert. Zahlreiche erfolgen, bei geschenkweiser mirb für zentrale unb einwandStanbortsangaben geographischer unb geologischer Natur Aufbewahrung Sorge getragen. Daß auch sonstige geben einen Anhalt über bie Verbreitung ber weniger häufi- freie Aufzeichnungen Volfsgutes sehr willgen Bflanzen. Besonders begrüßenswert finb bie Tabellen kommen mären, brauche alten ich wohl nicht erst zu versichern. zur Veftimmung von Bäumen unb Sträuchern nac ben Dr. Senn, Konstanz, Seeftr. 15. Blättern unb nac ben Winterknospen, sowie ber Schlüsse! Herausgegeben mit Unterstützung von Heimatfreunden vom Verlag der Hohenzollerischen Blätter G. m.b. H. Hechingen. —Verantwortlich Walter Sauter in Hechingen. — Nachdruck sämtlicher Originalartikel verboten Preis im Jahr R22. 2.50 zuzüglich 30 Pfg. Versendungstosten, zahlbar an Hohenzollerische Blätter, Postichec 28 27083 Amt Stuttgart.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.