Johnyquest7 commited on
Commit
b973ad4
·
verified ·
1 Parent(s): e3fb751

crop update

Browse files
Files changed (1) hide show
  1. README.md +5 -1
README.md CHANGED
@@ -158,7 +158,11 @@ python evaluate_external.py \
158
  - **Single-source dataset** Performance on data from
159
  other scanners, institutions, or populations is unknown and likely lower.
160
  - **Cropped-ROI inputs.** The model expects nodule-cropped images like TN5000;
161
- whole-frame ultrasound will be out of distribution.
 
 
 
 
162
  - **Sensitivity gap at deployment threshold.** The threshold targets ≥0.95
163
  sensitivity on validation; on the test set sensitivity was 0.904 — i.e. the
164
  operating point does not perfectly transfer, and ~10% of malignant nodules
 
158
  - **Single-source dataset** Performance on data from
159
  other scanners, institutions, or populations is unknown and likely lower.
160
  - **Cropped-ROI inputs.** The model expects nodule-cropped images like TN5000;
161
+ whole-frame ultrasound will be out of distribution. The model input is a region of interest
162
+ centered on the nodule that includes the nodule together with a margin of adjacent tissue.
163
+ It is not a tight crop that excludes all surrounding parenchyma. At deployment,
164
+ an operator selects the region of interest, so region selection is a human in the loop step,
165
+ and inter operator variability in region selection may affect model output.
166
  - **Sensitivity gap at deployment threshold.** The threshold targets ≥0.95
167
  sensitivity on validation; on the test set sensitivity was 0.904 — i.e. the
168
  operating point does not perfectly transfer, and ~10% of malignant nodules