Decide whether two mammography exams belong to the same woman.
How it works
For each breast side found in both exams, the CC and MLO views of each exam pass through the same ResNet backbone with a local relational layer and non-local attention. The feature maps of the two exams are subtracted, and two heads (one for CC, one for MLO) score whether the difference comes from the same person. The two view scores are averaged per side, and the sides are averaged into the final probability.
Input
Exam A: 2–4 mammogram DICOM files (at least CC and MLO of one side)
Exam B: the same views from the other exam
Side and view are read from the DICOM tags ImageLaterality and ViewPosition
Output
Probability that both exams are of the same person (> 0.5 → same)
Per-side and per-view scores, and the images as the model saw them
Model trained on DR mammograms from one hospital (Xiangshan cohort). Typical run time: About 5–10 s.
Your data is not saved.
Uploaded files are written only to a temporary folder on the lab GPU computer and deleted the moment the computation ends (or fails, or you cancel). Results are held in memory for 15 minutes so this page can show them, then discarded. Nothing is copied to our storage or logs.
Remove patient identity information before uploading.
Strip names, IDs, birth dates, addresses, hospital and physician names from the files (for DICOM, use your PACS or a de-identification tool). This page checks DICOM headers in your browser before upload and warns you if such fields are still filled in. Research demo only — not for diagnosis or treatment decisions.
Upload limit: 400 MB.
Run it
Drop 2–4 DICOM files per exam in any order. Side and view are read from the headers and shown next to each file.
