Datasets:
Upload lesion_annotations/dwi_channel_mapping_explanation.txt with huggingface_hub
Browse files
lesion_annotations/dwi_channel_mapping_explanation.txt
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
This JSON represents a mapping between unique MRI study identifiers and corresponding lists of MRI image identifiers.
|
| 2 |
+
|
| 3 |
+
Structure:
|
| 4 |
+
- The JSON object contains key-value pairs.
|
| 5 |
+
- Each key is a identifier from the original dataset.
|
| 6 |
+
- Each value is a list of strings. Each string represents a specific MRI scan, labeled with a "BreastDivider_<number>" format.
|
| 7 |
+
|
| 8 |
+
Image Ordering:
|
| 9 |
+
- Index 0 → Pre-contrast MRI image.
|
| 10 |
+
- Index 1 → Post-contrast 1 MRI image.
|
| 11 |
+
- Index 2 → Post-contrast 2 MRI image.
|
| 12 |
+
- Index 3 → T2-weighted MRI image.
|
| 13 |
+
- Index 4 and onwards → Dynamic post-contrast images (Post 3, Post 4, and so on).
|
| 14 |
+
|
| 15 |
+
Example:
|
| 16 |
+
"Duke_Breast_MRI_001": [
|
| 17 |
+
"BreastDivider_06181", # Pre-contrast
|
| 18 |
+
"BreastDivider_06886", # Post 1
|
| 19 |
+
"BreastDivider_05421", # Post 2
|
| 20 |
+
"BreastDivider_06513", # T2
|
| 21 |
+
"BreastDivider_09657", # Post 3
|
| 22 |
+
"BreastDivider_09604" # Post 4
|
| 23 |
+
]
|