File size: 2,831 Bytes
8f2f0a6 960f6b3 8f2f0a6 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | ---
tags:
- image-classification
- microbiology
- gram-stain
- medical-imaging
- research-use-only
- urine
- blood-culture
pipeline_tag: image-classification
---
# BiTTE
<p align="center">
<img src="./icon.jpg" alt="BiTTE icon" width="220">
</p>
BiTTE is an application within the CarbConnect platform. It is designed for the simple and efficient classification of microorganisms from Gram-stained microscopy images.
The app classifies findings into seven primary groups:
1. Gram-negative rods (GNR)
2. Gram-negative cocci (GNC)
3. Gram-positive rods (GPR)
4. Gram-positive cocci (GPC)
5. Yeast-like fungi
6. No bacteria
7. Multiple bacteria
BiTTE also supports more detailed subcategories beyond these primary output groups.
## Intended Uses and Limitations
BiTTE is strictly intended for **Research Use Only (RUO)**.
It is **not** intended for:
- clinical diagnostics
- medical procedures
- patient management
- therapeutic selection
- any regulated clinical use
For further details, please refer to the BiTTE Learn More page on CarbConnect.
## How to Use
A video tutorial demonstrating how to use the app is available on YouTube.
## Training Data
The model was trained on a dataset of Gram-stained images of urine and blood culture specimens generously provided by:
- the School of Medicine, Kobe University
- the National Center for Global Health and Medicine (NCGM)
Specimens were Gram-stained using either the Favor or Barmy method.
Image acquisition was performed by photographing specimens through the eyepiece of an optical microscope at **1000x magnification** using a smartphone camera.
The dataset captures frequently encountered clinical bacterial species and includes:
- 15 species in urine specimens
- 19 species in aerobic blood culture specimens
- 13 species in anaerobic blood culture specimens
## Performance and Evidence
Related publication:
Kei Yamamoto, Goh Ohji, et al. *Accuracy of classification of urinary Gram-stain findings by a computer-aided diagnosis app compared with microbiology specialists*. J Med Microbiol. 2025 Apr;74(4):002008. doi: 10.1099/jmm.0.002008.
Paper link:
https://www.microbiologyresearch.org/content/journal/jmm/10.1099/jmm.0.002008
## Citation
If you use BiTTE in research, please cite:
```bibtex
@article{yamamoto2025bitte,
author = {Yamamoto, Kei and Ohji, Goh and others},
title = {Accuracy of classification of urinary Gram-stain findings by a computer-aided diagnosis app compared with microbiology specialists},
journal = {Journal of Medical Microbiology},
year = {2025},
month = {Apr},
volume = {74},
number = {4},
pages = {002008},
doi = {10.1099/jmm.0.002008}
}
```
## Other Remarks
For guidance on achieving high-quality Gram staining, please refer to the automated gram stainer **Point of Care Gram Stainer (PoCGS)**.
|