source_id stringlengths 15 18 | image imagewidth (px) 256 2.32k | question stringlengths 15 115 | answer stringclasses 2
values | sft_prompt stringlengths 371 471 | assistant_response stringlengths 151 314 |
|---|---|---|---|---|---|
vqa_rad_train_0 | are regions of the brain infarcted? | yes | Annotate this medical image for explainability research.
Question: "are regions of the brain infarcted?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Infarct Region [100, 250, 200, 350]
Cerebral Cortex [0, 0, 500, 500]
Lateral Ventricles [150, 150, 350, 350]
Gray Matter [0, 0, 500, 500]
White Matter [100, 100, 400, 400]
Basal Ganglia [200, 200, 300, 300]
Corpus Callosum [200, 100, 300, 200] | |
vqa_rad_train_1 | are the lungs normal appearing? | no | Annotate this medical image for explainability research.
Question: "are the lungs normal appearing?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Lungs [20, 10, 470, 470]
Heart [200, 350, 400, 450]
Spine [200, 10, 300, 470]
Ribcage [0, 0, 470, 470]
Diaphragm [100, 350, 400, 470]
Clavicles [0, 0, 470, 50]
Scapulae [0, 0, 470, 470] | |
vqa_rad_train_3 | is the lesion causing significant brainstem herniation? | no | Annotate this medical image for explainability research.
Question: "is the lesion causing significant brainstem herniation?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Brainstem [290, 420, 510, 760]
Lesion [150, 450, 620, 770]
Cerebellum [110, 450, 620, 850]
Fourth Ventricle [300, 450, 490, 550]
Spinal Cord [330, 750, 470, 850]
Temporal Lobes [10, 100, 740, 450]
Occipital Lobes [100, 450, 620, 850] | |
vqa_rad_train_9 | is this image in the transverse plane? | yes | Annotate this medical image for explainability research.
Question: "is this image in the transverse plane?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Cerebral Cortex [20, 70, 540, 660]
Lateral Ventricle [280, 100, 420, 300]
Midline [270, 70, 310, 660]
Skull [10, 30, 550, 680]
Lesion [310, 250, 420, 370]
Gray Matter [20, 70, 540, 660]
White Matter [20, 70, 540, 660] | |
vqa_rad_train_10 | are there any pulmonary findings? | no | Annotate this medical image for explainability research.
Question: "are there any pulmonary findings?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Lungs [50, 20, 520, 440]
Heart [240, 280, 420, 440]
Spine [270, 0, 340, 440]
Ribs [0, 0, 588, 440]
Diaphragm [100, 350, 520, 440]
Clavicles [0, 0, 588, 80]
Scapulae [0, 0, 588, 150] | |
vqa_rad_train_11 | are the lungs affected? | no | Annotate this medical image for explainability research.
Question: "are the lungs affected?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Lungs [50, 20, 520, 440]
Heart [240, 280, 420, 440]
Spine [280, 0, 340, 440]
Ribs [0, 0, 588, 440]
Diaphragm [100, 350, 520, 440]
Clavicles [0, 0, 588, 80]
Scapulae [0, 0, 588, 150] | |
vqa_rad_train_14 | is there herniation of the brainstem secondary to the lesion | no | Annotate this medical image for explainability research.
Question: "is there herniation of the brainstem secondary to the lesion"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Brainstem [290, 420, 500, 540]
Cerebellum [110, 450, 620, 770]
Fourth Ventricle [300, 460, 450, 530]
Spinal Cord [330, 530, 440, 760]
Posterior Fossa [100, 300, 650, 850]
Temporal Lobes [10, 100, 740, 450]
Occipital Lobes [100, 400, 650, 850] | |
vqa_rad_train_16 | is the gyral enhancement? | yes | Annotate this medical image for explainability research.
Question: "is the gyral enhancement?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Gyri [0, 0, 512, 700]
Sulci [0, 0, 512, 700]
Lateral Ventricles [180, 100, 340, 500]
Third Ventricle [250, 450, 300, 550]
Cerebral Cortex [0, 0, 512, 700]
Basal Ganglia [180, 200, 340, 400]
Brainstem [250, 550, 300, 700] | |
vqa_rad_train_25 | is it difficult to delineate the left costophrenic angle? | yes | Annotate this medical image for explainability research.
Question: "is it difficult to delineate the left costophrenic angle?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Left Costophrenic Angle [20, 750, 450, 1000]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Chest Tube [180, 40, 400, 520] | |
vqa_rad_train_27 | is this an mri? | no | Annotate this medical image for explainability research.
Question: "is this an mri?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Heart [350, 300, 650, 700]
Lungs [50, 150, 950, 950]
Spine [400, 100, 580, 900]
Ribs [20, 150, 990, 900]
Diaphragm [100, 800, 980, 1000]
Clavicles [10, 200, 220, 350] | |
vqa_rad_train_30 | are there abnormalities with in the contrast between white and grey matter? | yes | Annotate this medical image for explainability research.
Question: "are there abnormalities with in the contrast between white and grey matter?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
White Matter [0, 0, 500, 580]
Grey Matter [100, 100, 400, 480]
Ventricles [250, 200, 350, 400]
Cerebral Cortex [50, 50, 450, 550]
Corpus Callosum [250, 100, 350, 300]
Basal Ganglia [200, 300, 300, 400]
Brainstem [250, 450, 350, 500] | |
vqa_rad_train_31 | is the gray-white matter junction altered? | yes | Annotate this medical image for explainability research.
Question: "is the gray-white matter junction altered?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Gray Matter Junction [120, 100, 480, 500]
White Matter [50, 50, 490, 580]
Cerebral Cortex [50, 50, 490, 580]
Lateral Ventricles [150, 200, 450, 400]
Corpus Callosum [200, 100, 350, 300]
Basal Ganglia [200, 300, 350, 450]
Brainstem [220, 450, 330, 550] | |
vqa_rad_train_32 | is there evidence of inflammation? | yes | Annotate this medical image for explainability research.
Question: "is there evidence of inflammation?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Inflamed Tissue [320, 200, 650, 480]
Kidney [320, 220, 470, 320]
Vertebral Body [470, 390, 560, 470]
Spinal Canal [480, 400, 550, 470]
Abdominal Aorta [470, 320, 540, 400]
Colon [370, 200, 680, 300] | |
vqa_rad_train_33 | is any structure inflamed? | yes | Annotate this medical image for explainability research.
Question: "is any structure inflamed?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Inflamed Structure [320, 220, 650, 480]
Liver [600, 180, 750, 350]
Kidney [320, 220, 480, 350]
Spine [480, 390, 560, 470]
Small Bowel [380, 200, 680, 300]
Large Bowel [530, 180, 710, 300]
Muscles [260, 100, 820, 550] | |
vqa_rad_train_34 | is the image in the axial plane? | yes | Annotate this medical image for explainability research.
Question: "is the image in the axial plane?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Cerebral Cortex [48, 20, 498, 580]
Lateral Ventricles [130, 150, 460, 400]
Corpus Callosum [150, 100, 440, 300]
Basal Ganglia [160, 300, 430, 450]
Thalamus [220, 350, 380, 450]
Brainstem [240, 400, 360, 500]
Cerebellum [200, 500, 390, 580] | |
vqa_rad_train_36 | is there blunting of the left costophrenic angle? | yes | Annotate this medical image for explainability research.
Question: "is there blunting of the left costophrenic angle?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Left Lung [50, 150, 450, 800]
Right Lung [500, 150, 950, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Left Costophrenic Angle [200, 750, 400, 900]
Right Costophrenic Angle [600, 750, 800, 900] | |
vqa_rad_train_40 | is this a t1 weighted image? | no | Annotate this medical image for explainability research.
Question: "is this a t1 weighted image?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Lateral Ventricles [120, 100, 400, 500]
Cerebral Cortex [0, 50, 510, 700]
Corpus Callosum [150, 100, 400, 300]
Third Ventricle [250, 450, 350, 550]
Basal Ganglia [150, 300, 400, 500]
Brainstem [250, 550, 350, 650]
Cerebellum [150, 550, 400, 700] | |
vqa_rad_train_41 | is the right renal vein visualized? | no | Annotate this medical image for explainability research.
Question: "is the right renal vein visualized?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Right Kidney [240, 250, 450, 480]
Liver [440, 240, 690, 450]
Spine [460, 370, 520, 460]
Stomach/Intestines [380, 140, 710, 330]
Pancreas [480, 300, 620, 440]
Left Kidney [520, 350, 650, 450]
Aorta/Inferior Vena Cava [460, 370, 520, 460] | |
vqa_rad_train_42 | is it possible to identify the right renal vein? | no | Annotate this medical image for explainability research.
Question: "is it possible to identify the right renal vein?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [240, 250, 450, 480]
Kidneys [450, 360, 530, 460]
Spine [500, 150, 650, 300]
Stomach/Intestines [380, 150, 700, 350]
Aorta [460, 370, 510, 450]
Inferior Vena Cava [460, 370, 510, 450]
Pancreas [500, 250, 600, 350] | |
vqa_rad_train_43 | are the kidneys visible in this image? | yes | Annotate this medical image for explainability research.
Question: "are the kidneys visible in this image?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Kidneys [120, 290, 350, 390]
Liver [10, 120, 200, 360]
Spine [200, 120, 300, 400]
Stomach [140, 120, 330, 220]
Pancreas [200, 220, 350, 300]
Spleen [350, 200, 440, 350]
Intestines [140, 120, 430, 350] | |
vqa_rad_train_44 | is air visible in the bowels? | yes | Annotate this medical image for explainability research.
Question: "is air visible in the bowels?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Air in Bowels [130, 100, 420, 270]
Liver [50, 180, 170, 350]
Kidneys [200, 280, 280, 360]
Spine [260, 100, 340, 400]
Pancreas [240, 170, 350, 250]
Stomach [130, 100, 290, 180]
Intestines [130, 100, 420, 270] | |
vqa_rad_train_45 | is there air in the bowel? | yes | Annotate this medical image for explainability research.
Question: "is there air in the bowel?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Bowel with Air [130, 90, 420, 270]
Liver [50, 180, 170, 380]
Kidney (Right) [200, 270, 280, 380]
Kidney (Left) [280, 270, 380, 380]
Spine [260, 100, 340, 400]
Pancreas [240, 170, 340, 240]
Aorta [270, 210, 290, 240] | |
vqa_rad_train_47 | are structures associated with the midbrain located in this image? | yes | Annotate this medical image for explainability research.
Question: "are structures associated with the midbrain located in this image?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Midbrain [220, 250, 320, 350]
Ventricles [150, 50, 400, 150]
Cerebellum [100, 350, 450, 500]
Thalamus [200, 200, 350, 250]
Corpus Callosum [150, 250, 400, 300]
Hippocampus [100, 300, 150, 400]
Brainstem [200, 350, 300, 450] | |
vqa_rad_train_51 | are the kidneys visualized? | yes | Annotate this medical image for explainability research.
Question: "are the kidneys visualized?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Kidneys [250, 570, 750, 750]
Spine [350, 750, 650, 860]
Intestines [180, 400, 850, 700]
Pelvic Bones [250, 750, 750, 860]
Bladder [350, 750, 650, 860]
Muscles [150, 400, 850, 900]
Fat [100, 300, 900, 950] | |
vqa_rad_train_52 | can you see the kidneys | yes | Annotate this medical image for explainability research.
Question: "can you see the kidneys"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Kidneys [250, 570, 750, 750]
Spine [450, 750, 550, 860]
Intestines [180, 400, 850, 700]
Pelvic Bones [250, 750, 750, 860]
Bladder [450, 750, 550, 860]
Muscles [150, 400, 900, 950]
Fat [100, 300, 900, 1000] | |
vqa_rad_train_58 | is the midbrain identified in this section? | yes | Annotate this medical image for explainability research.
Question: "is the midbrain identified in this section?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Midbrain [220, 290, 320, 420]
Ventricles [150, 50, 420, 250]
Cerebellum [100, 350, 480, 550]
Thalamus [180, 200, 350, 300]
Corpus Callosum [180, 150, 400, 250]
Hippocampus [120, 300, 200, 400]
Temporal Lobe [100, 350, 250, 500] | |
vqa_rad_train_62 | is there fluid in the lungs? | no | Annotate this medical image for explainability research.
Question: "is there fluid in the lungs?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [420, 150, 800, 780]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [100, 650, 800, 950]
Clavicles [0, 100, 800, 200] | |
vqa_rad_train_63 | is a pleural effusion present? | no | Annotate this medical image for explainability research.
Question: "is a pleural effusion present?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [420, 150, 800, 780]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [100, 650, 800, 950]
Clavicles [0, 100, 800, 200] | |
vqa_rad_train_64 | is this an axial cut? | yes | Annotate this medical image for explainability research.
Question: "is this an axial cut?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [120, 120, 300, 360]
Kidneys [240, 290, 350, 390]
Spine [200, 130, 300, 400]
Stomach [150, 130, 330, 220]
Pancreas [200, 220, 350, 300]
Spleen [350, 200, 430, 350]
Intestines [150, 130, 430, 350] | |
vqa_rad_train_65 | is the trachea deviated? | yes | Annotate this medical image for explainability research.
Question: "is the trachea deviated?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Trachea [470, 100, 550, 900]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [120, 680, 950, 850] | |
vqa_rad_train_66 | is tracheal deviation present? | yes | Annotate this medical image for explainability research.
Question: "is tracheal deviation present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Trachea [470, 100, 550, 900]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [120, 680, 950, 850] | |
vqa_rad_train_69 | are there more than 5 enlarged (>1 cm) lymph nodes around the stomach | yes | Annotate this medical image for explainability research.
Question: "are there more than 5 enlarged (>1 cm) lymph nodes around the stomach"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Stomach [220, 430, 510, 700]
Liver [500, 400, 850, 900]
Kidneys [280, 780, 750, 970]
Pancreas [450, 650, 700, 850]
Spine [450, 900, 550, 1000]
Intestines [200, 500, 800, 900] | |
vqa_rad_train_70 | are patchy infiltrates depicted? | yes | Annotate this medical image for explainability research.
Question: "are patchy infiltrates depicted?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Patchy Infiltrates [150, 150, 850, 650]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [200, 600, 950, 800] | |
vqa_rad_train_71 | are there patchy infiltrates present? | yes | Annotate this medical image for explainability research.
Question: "are there patchy infiltrates present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Patchy Infiltrates [150, 150, 850, 650]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [200, 600, 950, 800] | |
vqa_rad_train_72 | are the margins of the cardiac silhouette normal? | no | Annotate this medical image for explainability research.
Question: "are the margins of the cardiac silhouette normal?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Cardiac Silhouette [280, 450, 650, 700]
Right Lung [450, 150, 850, 800]
Left Lung [150, 150, 450, 800]
Mediastinum [300, 100, 650, 800]
Spine [350, 100, 580, 900]
Diaphragm [180, 550, 850, 950]
Clavicles [180, 200, 850, 350] | |
vqa_rad_train_73 | is the heart border normal? | no | Annotate this medical image for explainability research.
Question: "is the heart border normal?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Heart [320, 450, 600, 700]
Right Lung [450, 150, 850, 800]
Left Lung [150, 150, 450, 800]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [180, 550, 850, 950]
Clavicles [150, 200, 850, 350] | |
vqa_rad_train_74 | is this an axial plane | yes | Annotate this medical image for explainability research.
Question: "is this an axial plane"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Intestines [170, 390, 750, 720]
Pelvic Bones [10, 400, 990, 1000]
Vertebral Column [400, 320, 600, 1000]
Bladder [350, 650, 550, 850]
Femoral Heads [100, 650, 350, 850]
Kidneys [650, 400, 900, 700]
Muscles [0, 320, 1000, 1000] | |
vqa_rad_train_75 | is there intrahepatic ductal dilatation? | yes | Annotate this medical image for explainability research.
Question: "is there intrahepatic ductal dilatation?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Intrahepatic Ductal Dilatation [200, 150, 300, 300]
Liver [100, 100, 400, 400]
Kidneys [100, 300, 400, 400]
Spine [250, 300, 300, 400]
Pancreas [200, 200, 300, 300]
Gallbladder [250, 200, 300, 250]
Aorta [250, 250, 300, 300] | |
vqa_rad_train_76 | is intrahepatic ductal dilatation present? | yes | Annotate this medical image for explainability research.
Question: "is intrahepatic ductal dilatation present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Intrahepatic Ductal Dilatation [200, 150, 300, 300]
Liver [100, 100, 400, 400]
Kidneys [100, 300, 400, 400]
Spine [250, 300, 300, 400]
Pancreas [200, 200, 300, 300]
Gallbladder [250, 200, 300, 250]
Aorta [250, 250, 300, 300] | |
vqa_rad_train_79 | are air fluid levels depicted in this image? | yes | Annotate this medical image for explainability research.
Question: "are air fluid levels depicted in this image?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Air-Fluid Levels [100, 320, 720, 690]
Liver [750, 250, 1000, 1000]
Kidney [330, 740, 650, 850]
Spine [350, 280, 650, 1000]
Aorta [400, 800, 550, 950]
Inferior Vena Cava [550, 800, 650, 950]
Small Bowel [100, 320, 720, 690] | |
vqa_rad_train_80 | are there >5 lymph nodes located near the stomach? | yes | Annotate this medical image for explainability research.
Question: "are there >5 lymph nodes located near the stomach?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Stomach [220, 430, 510, 700]
Lymph Nodes [450, 650, 800, 850]
Liver [100, 400, 500, 1000]
Kidneys [280, 780, 750, 980]
Spine [450, 900, 550, 1000]
Pancreas [400, 600, 600, 800]
Intestines [500, 450, 850, 750] | |
vqa_rad_train_81 | is there an air fluid level present? | yes | Annotate this medical image for explainability research.
Question: "is there an air fluid level present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Air-Fluid Level [100, 560, 250, 680]
Liver [700, 300, 950, 600]
Kidney [350, 750, 550, 850]
Spine [400, 300, 600, 500]
Aorta [450, 800, 550, 900]
Inferior Vena Cava [600, 800, 700, 900]
Stomach [800, 500, 950, 650] | |
vqa_rad_train_84 | is there air in the bowel? | yes | Annotate this medical image for explainability research.
Question: "is there air in the bowel?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Bowel Gas [142, 128, 430, 310]
Liver [10, 115, 200, 370]
Kidney (Right) [245, 295, 345, 390]
Kidney (Left) [110, 295, 210, 370]
Spine [200, 100, 300, 400]
Pancreas [220, 200, 370, 300]
Stomach [140, 120, 250, 220] | |
vqa_rad_train_85 | is air present in the bowel? | yes | Annotate this medical image for explainability research.
Question: "is air present in the bowel?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Bowel Gas [142, 128, 430, 310]
Liver [10, 115, 200, 365]
Kidney (Right) [245, 295, 345, 390]
Kidney (Left) [110, 295, 210, 365]
Pancreas [180, 200, 380, 300]
Spine [200, 100, 300, 400]
Stomach [140, 120, 330, 220] | |
vqa_rad_train_91 | does the gallbladder appear distended? | yes | Annotate this medical image for explainability research.
Question: "does the gallbladder appear distended?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Gallbladder [290, 100, 380, 250]
Liver [120, 80, 300, 350]
Kidneys [170, 230, 270, 350]
Spine [320, 340, 400, 410]
Pancreas [300, 150, 450, 300]
Stomach [220, 100, 350, 200]
Intestines [200, 50, 550, 300] | |
vqa_rad_train_93 | is cardiomegaly shown? | yes | Annotate this medical image for explainability research.
Question: "is cardiomegaly shown?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Heart [320, 300, 650, 700]
Lungs [50, 150, 950, 800]
Spine [250, 0, 450, 900]
Ribs [0, 100, 810, 700]
Diaphragm [100, 650, 810, 950]
Trachea [350, 100, 450, 500]
Chest Wall [0, 0, 810, 950] | |
vqa_rad_train_94 | is the heart enlarged? | yes | Annotate this medical image for explainability research.
Question: "is the heart enlarged?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Heart [350, 250, 650, 650]
Right Lung [50, 150, 450, 800]
Left Lung [500, 150, 950, 800]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [100, 650, 800, 950]
Clavicles [10, 100, 800, 200] | |
vqa_rad_train_100 | can you see kidneys in this image | yes | Annotate this medical image for explainability research.
Question: "can you see kidneys in this image"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Kidneys [295, 375, 460, 500]
Liver [530, 170, 770, 400]
Spine [470, 400, 560, 520]
Stomach [310, 230, 460, 350]
Pancreas [450, 340, 550, 400]
Aorta [480, 480, 520, 520]
Inferior Vena Cava [520, 480, 560, 520] | |
vqa_rad_train_102 | is the gallbladder enlarged? | yes | Annotate this medical image for explainability research.
Question: "is the gallbladder enlarged?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Gallbladder [250, 100, 350, 250]
Liver [120, 80, 350, 350]
Kidneys [170, 230, 270, 350]
Spine [320, 340, 400, 410]
Pancreas [300, 150, 450, 300]
Stomach [250, 50, 350, 150]
Intestines [200, 100, 550, 300] | |
vqa_rad_train_103 | is there contrast in this image | yes | Annotate this medical image for explainability research.
Question: "is there contrast in this image"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Contrast [250, 150, 800, 550]
Liver [250, 150, 800, 550]
Kidneys [280, 370, 750, 520]
Spine [450, 400, 550, 500]
Pancreas [400, 250, 600, 400]
Stomach [300, 200, 500, 350]
Aorta [450, 350, 550, 450] | |
vqa_rad_train_104 | is pericholecystic inflammation present? | yes | Annotate this medical image for explainability research.
Question: "is pericholecystic inflammation present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Gallbladder [520, 170, 770, 410]
Pericholecystic Inflammation [450, 250, 650, 450]
Liver [250, 150, 550, 500]
Kidney (Right) [280, 370, 450, 520]
Kidney (Left) [550, 370, 750, 520]
Spine [450, 400, 550, 550]
Pancreas [450, 300, 600, 400] | |
vqa_rad_train_114 | is this image plane taken inferior to the liver? | yes | Annotate this medical image for explainability research.
Question: "is this image plane taken inferior to the liver?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Kidneys [280, 580, 560, 750]
Small Bowel [140, 180, 900, 700]
Spine [450, 680, 580, 750]
Liver [650, 250, 850, 500]
Colon [140, 180, 700, 400]
Muscles [100, 200, 950, 850]
Fat [100, 200, 950, 850] | |
vqa_rad_train_116 | is the liver cirrhotic? | yes | Annotate this medical image for explainability research.
Question: "is the liver cirrhotic?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [100, 50, 490, 380]
Spine [250, 150, 350, 370]
Kidney (Right) [160, 250, 250, 370]
Kidney (Left) [350, 250, 450, 370]
Stomach [280, 70, 350, 200]
Pancreas [250, 250, 350, 350]
Aorta [260, 280, 290, 320] | |
vqa_rad_train_124 | is this a pa film? | yes | Annotate this medical image for explainability research.
Question: "is this a pa film?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [120, 680, 950, 850]
Ribs [0, 100, 1000, 1000] | |
vqa_rad_train_125 | is there any shift of midline structures visible in this section? | no | Annotate this medical image for explainability research.
Question: "is there any shift of midline structures visible in this section?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Midline Structures [370, 250, 440, 600]
Cervical Spine [200, 250, 650, 650]
Muscles of Neck [100, 300, 750, 650]
Trachea [370, 300, 440, 400]
Esophagus [370, 400, 440, 500]
Soft Tissues [0, 0, 800, 650]
Facial Bones [0, 0, 800, 350] | |
vqa_rad_train_127 | is this an axial plane? | yes | Annotate this medical image for explainability research.
Question: "is this an axial plane?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Vertebral Body [420, 300, 580, 500]
Kidney [330, 740, 500, 850]
Liver [100, 400, 350, 700]
Aorta [420, 800, 500, 900]
Inferior Vena Cava [600, 800, 700, 950]
Small Bowel [150, 300, 750, 600]
Muscles [0, 0, 1000, 1000] | |
vqa_rad_train_128 | is this a ct image? | yes | Annotate this medical image for explainability research.
Question: "is this a ct image?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [160, 120, 470, 370]
Kidney [160, 250, 250, 370]
Spine [250, 200, 350, 370]
Stomach [280, 70, 350, 200]
Pancreas [250, 250, 350, 300]
Aorta [250, 280, 300, 330]
Inferior Vena Cava [300, 280, 350, 330] | |
vqa_rad_train_133 | are there multiple septations within the cyst? | yes | Annotate this medical image for explainability research.
Question: "are there multiple septations within the cyst?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Cyst with Septations [320, 190, 570, 440]
Kidney (Right) [550, 320, 670, 440]
Kidney (Left) [390, 340, 520, 440]
Vertebral Body [470, 330, 570, 470]
Aorta [400, 220, 480, 300]
Inferior Vena Cava [480, 220, 540, 300]
Small Bowel Loops [450, 200, 700, 350] | |
vqa_rad_train_134 | is the cyst complex? | yes | Annotate this medical image for explainability research.
Question: "is the cyst complex?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Cyst [320, 200, 570, 440]
Kidney (Right) [550, 320, 670, 440]
Kidney (Left) [390, 340, 520, 440]
Spine [480, 320, 570, 470]
Liver [260, 200, 400, 440]
Colon [480, 200, 690, 320]
Aorta [420, 220, 480, 300] | |
vqa_rad_train_136 | is there shifting of structures across the middle? | no | Annotate this medical image for explainability research.
Question: "is there shifting of structures across the middle?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Midline [380, 250, 420, 600]
Nasal Cavity [250, 50, 450, 250]
Oral Cavity [250, 300, 550, 600]
Mandible [100, 350, 700, 650]
Maxilla [100, 100, 700, 300]
Soft Palate [300, 250, 500, 350]
Tongue [250, 400, 550, 550] | |
vqa_rad_train_139 | is there contrast in this image? | yes | Annotate this medical image for explainability research.
Question: "is there contrast in this image?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Contrast Material [270, 290, 340, 370]
Liver [100, 120, 250, 360]
Kidneys [240, 290, 350, 390]
Spine [200, 100, 300, 400]
Stomach/Intestines [140, 120, 430, 300]
Pancreas [200, 250, 300, 350]
Spleen [350, 200, 440, 350] | |
vqa_rad_train_140 | is there evidence of mesenteric lymphadenopathy? | yes | Annotate this medical image for explainability research.
Question: "is there evidence of mesenteric lymphadenopathy?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Mesenteric Lymph Nodes [250, 200, 350, 300]
Liver [100, 100, 300, 300]
Kidneys [100, 300, 400, 400]
Spine [250, 300, 300, 400]
Pancreas [200, 200, 300, 300]
Aorta [250, 250, 300, 300]
Inferior Vena Cava [200, 200, 250, 300] | |
vqa_rad_train_141 | is mesenteric lymphadenopathy present? | yes | Annotate this medical image for explainability research.
Question: "is mesenteric lymphadenopathy present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Mesenteric Lymph Nodes [250, 200, 350, 300]
Liver [100, 100, 300, 300]
Kidneys [100, 300, 400, 400]
Spine [250, 300, 300, 400]
Pancreas [200, 200, 300, 300]
Aorta [250, 250, 300, 300]
Inferior Vena Cava [200, 200, 250, 300] | |
vqa_rad_train_144 | is the right conal fascia inflammed? | yes | Annotate this medical image for explainability research.
Question: "is the right conal fascia inflammed?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [250, 150, 500, 500]
Kidney (Right) [530, 170, 770, 410]
Kidney (Left) [290, 370, 460, 500]
Spine [450, 400, 580, 550]
Stomach [320, 230, 460, 350]
Pancreas [450, 300, 580, 400]
Inflamed Right Conal Fascia [530, 170, 770, 410] | |
vqa_rad_train_145 | is inflammation present at the right conal fascia? | yes | Annotate this medical image for explainability research.
Question: "is inflammation present at the right conal fascia?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [250, 150, 800, 550]
Kidney [530, 170, 770, 410]
Spine [450, 400, 580, 550]
Stomach [300, 230, 460, 360]
Pancreas [450, 300, 600, 400]
Inflamed Right Conal Fascia [550, 350, 650, 450] | |
vqa_rad_train_147 | is there evidence of a small bowel obstruction with perforated viscus on this ct? | yes | Annotate this medical image for explainability research.
Question: "is there evidence of a small bowel obstruction with perforated viscus on this ct?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Small Bowel [120, 180, 900, 750]
Perforation Site [450, 500, 550, 600]
Liver [650, 200, 900, 500]
Kidney [250, 500, 400, 650]
Spine [450, 650, 550, 750]
Colon [150, 180, 350, 350]
Muscles [100, 700, 900, 850] | |
vqa_rad_train_154 | is there hilar adenopathy? | yes | Annotate this medical image for explainability research.
Question: "is there hilar adenopathy?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Hilar Adenopathy [650, 250, 850, 450]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900] | |
vqa_rad_train_155 | is the et tube properly placed? | yes | Annotate this medical image for explainability research.
Question: "is the et tube properly placed?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Endotracheal Tube [350, 0, 520, 820]
Right Lung [480, 100, 800, 620]
Left Lung [150, 150, 450, 600]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [250, 600, 980, 820] | |
vqa_rad_train_156 | is there perinephric fat stranding? | yes | Annotate this medical image for explainability research.
Question: "is there perinephric fat stranding?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Perinephric Fat Stranding [240, 120, 800, 540]
Liver [240, 120, 500, 540]
Kidney (Right) [580, 320, 730, 460]
Kidney (Left) [350, 430, 670, 540]
Spine [420, 380, 580, 500]
Stomach/Gastrointestinal Tract [410, 120, 650, 250]
Pancreas [350, 250, 550, 400] | |
vqa_rad_train_158 | is there evidence of a pneumoperitoneum secondary to a perforation? | yes | Annotate this medical image for explainability research.
Question: "is there evidence of a pneumoperitoneum secondary to a perforation?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Free Air (Pneumoperitoneum) [260, 270, 360, 370]
Small Bowel [140, 180, 900, 750]
Colon [140, 180, 900, 750]
Liver [650, 250, 900, 500]
Kidney [280, 580, 400, 700]
Spine [450, 680, 580, 750]
Muscles [100, 150, 950, 850] | |
vqa_rad_train_160 | is there cardiomegaly? | yes | Annotate this medical image for explainability research.
Question: "is there cardiomegaly?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Heart [280, 300, 650, 750]
Right Lung [450, 100, 850, 800]
Left Lung [50, 100, 450, 800]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Ribcage [0, 0, 1000, 800]
Clavicles [240, 0, 900, 200] | |
vqa_rad_train_161 | are the sulci visible in this image? | no | Annotate this medical image for explainability research.
Question: "are the sulci visible in this image?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Lateral Ventricles [120, 140, 360, 380]
Cerebral Cortex [60, 20, 400, 400]
Skull [50, 10, 410, 420]
Thalamus [200, 180, 260, 260]
Basal Ganglia [180, 220, 280, 300]
Midline Structures [200, 180, 260, 320]
Cerebrospinal Fluid (CSF) Spaces [120, 140, 360, 380] | |
vqa_rad_train_163 | is the trachea midline? | yes | Annotate this medical image for explainability research.
Question: "is the trachea midline?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Trachea [320, 0, 400, 811]
Right Lung [0, 100, 350, 811]
Left Lung [350, 100, 699, 811]
Heart [300, 300, 550, 700]
Spine [300, 0, 450, 811]
Diaphragm [100, 600, 699, 811]
Ribs [0, 0, 699, 811] | |
vqa_rad_train_164 | are there any fractures in this image? | no | Annotate this medical image for explainability research.
Question: "are there any fractures in this image?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [362, 100, 610, 650]
Left Lung [40, 100, 310, 600]
Heart [250, 300, 450, 600]
Mediastinum [250, 100, 450, 600]
Spine [280, 100, 380, 700]
Ribcage [0, 50, 640, 650]
Diaphragm [150, 550, 580, 700] | |
vqa_rad_train_165 | was the patient given iv contrast? | yes | Annotate this medical image for explainability research.
Question: "was the patient given iv contrast?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Liver [220, 430, 500, 950]
Kidneys [280, 790, 720, 970]
Spine [450, 940, 550, 1000]
Intestines [250, 430, 800, 800]
Aorta [450, 750, 550, 850]
Inferior Vena Cava [550, 750, 650, 850]
Pancreas [400, 600, 600, 750] | |
vqa_rad_train_166 | is there contrast in the patient's vasculature? | yes | Annotate this medical image for explainability research.
Question: "is there contrast in the patient's vasculature?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Vasculature [280, 650, 820, 980]
Liver [200, 400, 550, 700]
Kidneys [280, 750, 820, 980]
Spine [450, 400, 550, 1000]
Stomach/Intestines [250, 450, 800, 750]
Pancreas [400, 600, 600, 800]
Spleen [700, 450, 850, 700] | |
vqa_rad_train_167 | is there pleural thickening in the r lung? | yes | Annotate this medical image for explainability research.
Question: "is there pleural thickening in the r lung?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900] | |
vqa_rad_train_168 | is iv contrast used? | yes | Annotate this medical image for explainability research.
Question: "is iv contrast used?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Contrast Enhancement [210, 270, 300, 330]
Kidneys [160, 150, 440, 400]
Spine [210, 300, 300, 410]
Liver [350, 150, 470, 300]
Intestines [160, 150, 440, 400]
Aorta [260, 200, 330, 250]
Inferior Vena Cava [260, 250, 330, 300] | |
vqa_rad_train_169 | is there edema in the patient's right temporal lobe? | yes | Annotate this medical image for explainability research.
Question: "is there edema in the patient's right temporal lobe?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Temporal Lobe (Right) [200, 100, 300, 300]
Temporal Lobe (Left) [50, 100, 150, 300]
Ventricles [100, 200, 250, 350]
Cerebellum [100, 300, 250, 400]
Brainstem [150, 250, 200, 350]
Orbit (Right) [250, 50, 300, 150]
Orbit (Left) [50, 50, 100, 150] | |
vqa_rad_train_170 | is this process bilateral? | yes | Annotate this medical image for explainability research.
Question: "is this process bilateral?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Cerebral Cortex [100, 90, 400, 430]
Lateral Ventricles [120, 100, 380, 420]
Midline [230, 90, 250, 430]
Gray Matter [100, 90, 400, 430]
White Matter [120, 120, 380, 400]
Subcortical Structures [120, 120, 380, 400]
Skull [70, 60, 430, 450] | |
vqa_rad_train_171 | is the sigmoid colon visualized at this axial section? | yes | Annotate this medical image for explainability research.
Question: "is the sigmoid colon visualized at this axial section?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Sigmoid Colon [200, 60, 820, 420]
Small Intestines [200, 60, 820, 420]
Vertebral Body [400, 300, 600, 420]
Psoas Muscles [130, 100, 250, 400]
Abdominal Aorta [450, 300, 550, 400]
Inferior Vena Cava [450, 300, 550, 400]
Pelvic Floor Muscles [300, 350, 700, 420] | |
vqa_rad_train_172 | is there a pneumothorax? | yes | Annotate this medical image for explainability research.
Question: "is there a pneumothorax?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [250, 40, 450, 350]
Left Lung [50, 40, 250, 350]
Pneumothorax [250, 40, 450, 350]
Heart [150, 300, 300, 450]
Mediastinum [150, 100, 300, 500]
Spine [150, 10, 300, 500]
Diaphragm [100, 350, 400, 500] | |
vqa_rad_train_173 | is the abnormality hyper intense? | yes | Annotate this medical image for explainability research.
Question: "is the abnormality hyper intense?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Abnormality (Hyperintense Lesion) [120, 160, 180, 240]
Cerebral Cortex [0, 0, 230, 300]
Ventricular System [80, 100, 160, 280]
Basal Ganglia [100, 120, 150, 200]
Midline Structures [100, 100, 140, 280]
Temporal Lobe [10, 100, 120, 300]
Frontal Lobe [180, 100, 230, 300] | |
vqa_rad_train_174 | is there ring-enhancement? | yes | Annotate this medical image for explainability research.
Question: "is there ring-enhancement?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Enhancing Lesion [200, 50, 270, 150]
Lateral Ventricles [150, 100, 200, 300]
Third Ventricle [160, 250, 190, 300]
Cerebral Cortex [0, 0, 330, 380]
Basal Ganglia [100, 150, 250, 300]
Brainstem [150, 300, 200, 350]
Cerebellum [100, 350, 300, 400] | |
vqa_rad_train_175 | is there air in the lesion? | no | Annotate this medical image for explainability research.
Question: "is there air in the lesion?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Lesion [420, 590, 650, 760]
Small Bowel [100, 400, 970, 800]
Liver [100, 100, 970, 500]
Spine [450, 750, 600, 900]
Kidney [400, 600, 550, 700]
Fat Plane [100, 700, 970, 900]
Muscles [100, 100, 970, 900] | |
vqa_rad_train_179 | is there a shift of midline structures? | no | Annotate this medical image for explainability research.
Question: "is there a shift of midline structures?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Midline Structures [150, 100, 250, 400]
Third Ventricle [160, 250, 240, 350]
Lateral Ventricles [50, 100, 350, 400]
Cerebral Hemispheres [0, 0, 350, 500]
Corpus Callosum [150, 100, 250, 250]
Brainstem [150, 350, 250, 450]
Cerebellum [150, 400, 250, 500] | |
vqa_rad_train_180 | is there evidence of cytotoxic edema in the right temporal lobe? | yes | Annotate this medical image for explainability research.
Question: "is there evidence of cytotoxic edema in the right temporal lobe?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Temporal Lobe [200, 100, 300, 300]
Ventricles [150, 150, 250, 300]
Cerebral Cortex [50, 50, 300, 350]
Basal Ganglia [150, 150, 250, 250]
Brainstem [150, 250, 250, 350]
Skull [0, 0, 320, 320]
Sinuses [50, 50, 270, 100] | |
vqa_rad_train_182 | are the cerebellar infarcts hyper attenuated? | yes | Annotate this medical image for explainability research.
Question: "are the cerebellar infarcts hyper attenuated?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Cerebellum [200, 350, 650, 600]
Vermis [350, 380, 450, 550]
Fourth Ventricle [350, 380, 450, 450]
Brainstem [350, 450, 450, 550]
Occipital Lobe [150, 300, 650, 400]
Temporal Bone [50, 100, 150, 300]
Orbital Cavities [100, 100, 700, 300] | |
vqa_rad_train_183 | is this patient intubated? | no | Annotate this medical image for explainability research.
Question: "is this patient intubated?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Trachea [480, 200, 540, 900]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Diaphragm [200, 800, 900, 1000] | |
vqa_rad_train_185 | is the spleen present? | yes | Annotate this medical image for explainability research.
Question: "is the spleen present?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Spleen [260, 100, 400, 260]
Liver [100, 50, 300, 250]
Kidney [350, 100, 420, 200]
Spine [200, 50, 300, 300]
Stomach [250, 100, 350, 200]
Pancreas [200, 150, 300, 250]
Diaphragm [100, 200, 400, 300] | |
vqa_rad_train_190 | is there a rib fracture? | no | Annotate this medical image for explainability research.
Question: "is there a rib fracture?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [500, 150, 950, 650]
Left Lung [50, 150, 500, 650]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Clavicles [420, 0, 780, 150]
Scapulae [20, 100, 1000, 300] | |
vqa_rad_train_191 | does the mass have a smooth appearing border? | yes | Annotate this medical image for explainability research.
Question: "does the mass have a smooth appearing border?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Mass [120, 460, 620, 760]
Vertebral Body [230, 420, 580, 750]
Intervertebral Disc [290, 620, 520, 700]
Spinal Cord [300, 460, 500, 700]
Sacrum [200, 700, 550, 850]
Pelvic Bones [0, 400, 750, 850]
Rectum [300, 700, 450, 850] | |
vqa_rad_train_192 | are there decreased lung markings in the right upper lung field? | yes | Annotate this medical image for explainability research.
Question: "are there decreased lung markings in the right upper lung field?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Ribcage [0, 0, 1000, 832]
Diaphragm [100, 650, 990, 832] | |
vqa_rad_train_193 | is there an brain bleed? | no | Annotate this medical image for explainability research.
Question: "is there an brain bleed?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Cerebral Cortex [60, 70, 500, 620]
Lateral Ventricles [100, 100, 450, 550]
Corpus Callosum [250, 100, 350, 550]
Gray Matter [60, 70, 500, 620]
White Matter [100, 100, 450, 550]
Brainstem [250, 550, 350, 620]
Subcortical Structures [100, 100, 450, 550] | |
vqa_rad_train_195 | are there increased interstitial markings? | yes | Annotate this medical image for explainability research.
Question: "are there increased interstitial markings?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Increased Interstitial Markings [0, 0, 1000, 1000]
Right Lung [500, 150, 950, 800]
Left Lung [50, 150, 450, 800]
Heart [300, 300, 650, 700]
Mediastinum [350, 100, 600, 800]
Spine [400, 100, 580, 900]
Ribs [0, 0, 1000, 1000] | |
vqa_rad_train_197 | is the corpus callosum involved? | yes | Annotate this medical image for explainability research.
Question: "is the corpus callosum involved?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Corpus Callosum [440, 160, 540, 270]
Lateral Ventricles [350, 320, 650, 480]
Cerebral Cortex [350, 90, 650, 500]
Skull [250, 50, 780, 520]
Midline [450, 100, 550, 500]
Third Ventricle [480, 350, 520, 400]
Basal Ganglia [400, 300, 580, 450] | |
vqa_rad_train_199 | is there pneumoperitoneum? | no | Annotate this medical image for explainability research.
Question: "is there pneumoperitoneum?"
Answer: "no"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: X-ray
Diaphragm [20, 650, 900, 1000]
Lungs [0, 0, 920, 1000]
Heart [300, 200, 600, 700]
Spine [350, 0, 580, 1000]
Ribs [0, 0, 920, 1000]
Clavicles [0, 0, 920, 150]
Trachea [400, 0, 500, 300] | |
vqa_rad_train_201 | is there grey-white matter differentiation? | yes | Annotate this medical image for explainability research.
Question: "is there grey-white matter differentiation?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Grey Matter [120, 70, 410, 430]
White Matter [120, 70, 410, 430]
Ventricles [180, 150, 340, 300]
Cerebral Cortex [120, 70, 410, 430]
Skull [80, 60, 440, 450]
Sinuses [100, 100, 420, 350]
Subarachnoid Space [120, 70, 410, 430] | |
vqa_rad_train_202 | is the mass well-defined? | yes | Annotate this medical image for explainability research.
Question: "is the mass well-defined?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: MRI
Mass [150, 420, 620, 770]
Sacrum [240, 420, 580, 770]
Vertebral Body [240, 420, 580, 770]
Pelvic Bones [0, 100, 750, 860]
Bladder [150, 650, 300, 770]
Rectum [240, 420, 580, 770]
Femoral Heads [100, 100, 700, 450] | |
vqa_rad_train_203 | is there any lesion in the ventricles? | yes | Annotate this medical image for explainability research.
Question: "is there any lesion in the ventricles?"
Answer: "yes"
Tasks:
1. Identify the imaging MODALITY (CT, MRI, or X-ray)
2. List 5-7 anatomical structures with bounding boxes
FORMAT:
Modality: <modality>
StructureName [x_left, y_top, x_right, y_bottom]
Coordinates: 0-1000 scale. (0,0)=top-left, (1000,1000)=bottom-right.
OUTPUT: | Modality: CT
Lateral Ventricles [300, 200, 710, 660]
Third Ventricle [480, 200, 550, 350]
Fourth Ventricle [500, 500, 550, 550]
Cerebral Hemispheres [290, 190, 710, 660]
Skull [260, 80, 750, 730]
Brainstem [480, 450, 550, 600]
Cerebellum [480, 550, 550, 650] |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 20