Evo-RAD: Navigating Rare Retinal Disease Diagnosis via Self-Evolving Agentic Retrieval
2026-06-22 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors noticed that big medical AI models often struggle to diagnose rare diseases because there isn't much data about them. To fix this, the authors created Evo-RAD, a smart system that treats finding helpful medical examples like a step-by-step game where it can add or remove cases to get better information. This system learns how to pick the most relevant and consistent cases to improve diagnosis, especially for rare retinal diseases. Their tests showed Evo-RAD works much better than other methods at identifying rare conditions.
foundation modelsrare disease diagnosisretrieval-augmented diagnosishubness problemMarkov Decision Process (MDP)graph-based agentGroup Relative Policy Optimization (GRPO)diagnostic homogeneityretinal diseasesparameter-efficient fine-tuning
Authors
Wangding Xia, Ye Du, Jiashi Lin, Meng Wang, Danli Shi, Shujun Wang
Abstract
Large-scale pretrained foundation models have revolutionized general medical screening, but often falter on rare diseases because such conditions are underrepresented in real-world clinical datasets. While retrieval-augmented diagnosis attempts to mitigate this, conventional static methods frequently succumb to the hubness problem, retrieving visually similar but semantically incorrect common diseases. To address this, we propose Evo-RAD, a self-evolving agentic framework that transforms evidence acquisition into a dynamic decision-making task. We formulate retrieval as a Markov Decision Process (MDP) where a graphbased agent observes the reference set state and executes actions to purge discordant evidence (DELETE), acquire pathologically consistent samples (INSERT), or conclude the evolution (TERMINATE). Optimized via Group Relative Policy Optimization (GRPO) with a homogeneityaware reward, the agent learns to maximize the diagnostic homogeneity of the support reference set. Experiments on retinal disease benchmarks show that Evo-RAD substantially improves rare-disease diagnosis, outperforming retinal foundation models by +21.04%, while also surpassing retrieval-based and parameter-efficient fine-tuning methods by +3.56%. Code is available at https://github.com/SDH-Lab/Evo-RAD.