GRAPE: Graph-Augmented Explanations Boost Medical Image AI Safety and Accuracy

Prototype-based medical image classifiers have three critical limitations: they treat anatomical findings as independent, silently amplify unsafe physician feedback, and require full retraining to add a new finding. Researchers Rasul Khanbayov and Hasan Kurban present GRAPE (Graph-Augmented Prototype Explanations), a unified architecture that solves all three challenges in a single system.

GRAPE's first component, a Graph Attention Task Head, models anatomical concept co-occurrence, boosting macro-F1 by +13.8 percentage points over the prototype baseline on the TBX11K dataset. The second component introduces the first Concept-Mismatch Safety Check in prototype-based medical classifiers — it warns when the model's dominant finding inside a doctor-drawn region conflicts with the claimed label, catching 85% of erroneous annotations versus 51% for MC-Dropout, at no extra inference cost. Third, Open-Vocabulary Prototype Anchoring aligns visual prototypes to clinical text, allowing a new finding to be added from a single labeled image without modifying any other component. On NIH ChestX-ray14, one Effusion example recovers full-supervision localization accuracy; on TBX11K, prototype maps achieve 2.6x better lesion localization than end-to-end baselines. All three capabilities add only +1 ms latency at interactive batch size, making GRAPE practical for real-time clinical use.