GRAPE: Graph-Augmented Prototype Explanations for Interactive Medical Image Diagnosis
Rasul Khanbayov, Erchin Serpedin, Hasan Kurban · 2026
Read GRAPE: Graph-Augmented Prototype Explanations for Interactive Medical Image Diagnosis by Rasul Khanbayov, Erchin Serpedin, Hasan Kurban (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-30901.txt.
Description
Rasul Khanbayov, Erchin Serpedin and Hasan Kurban present GRAPE, a medical image classifier designed to explain itself through prototypes, example patches it compares new images against. Prototype systems have three clinical weaknesses: they treat findings as independent, they silently absorb bad feedback from doctors, and they need full retraining to learn a new finding. GRAPE attacks all three. A graph attention head models how anatomical concepts co-occur, and a safety check warns when the top finding inside a region drawn by the doctor clashes with the claimed label.
The safety check caught 85% of erroneous annotations, against 51% for a dropout-based rival, at no extra inference cost. On the TBX11K chest dataset, macro-F1 rose by 13.8 points over the prototype baseline, and lesion localisation was 2.6 times better. A new finding can be anchored from a single labelled image.
Catalog record
| CREATOR | Rasul Khanbayov, Erchin Serpedin, Hasan Kurban |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.30901