MedAmor
AI-assisted medical audit system for breast cancer screening compliance, built for the MedGemma Impact Challenge on Kaggle.
An AI-assisted audit system for breast cancer screening compliance built during the MedGemma Impact Challenge on Kaggle. MedAmor uses Google's MedGemma model to extract structured compliance data from unstructured radiology reports and flags cases that need human review.
Problem
Breast cancer screening programs require strict compliance auditing report timeframes, follow-up documentation, radiologist sign-off. Auditing thousands of cases monthly by hand is slow and error-prone.
Solution
MedAmor automates compliance extraction using MedGemma, fine-tuned on 1,200 anonymized radiology reports. A Next.js review queue presents flagged cases to clinical auditors with confidence scores. Every correction is logged as training data for the next model iteration.