Guideline-grounded retrieval-augmented generation for ophthalmic clinical decision support

2026-03-23

AI Summary

Existing ophthalmic clinical decision support systems lack guideline-grounded, traceable, and precise question-answering capabilities. To address this limitation, this work proposes Oph-Guid-RAG, a multimodal visual Retrieval-Augmented Generation (RAG) system that uniquely treats guideline page images directly as evidence units during retrieval, thereby preserving their original layout information. The framework integrates query decomposition and rewriting, controllable retrieval with routing and filtering mechanisms, and multimodal reasoning to substantially enhance evidence traceability and system robustness. Evaluated on the HealthBench hard subset, Oph-Guid-RAG achieves a 30.0% improvement in overall score and a 10.4% gain in accuracy over GPT-5.2, and further outperforms GPT-5.4 by 24.4% in accuracy, significantly surpassing current baselines.

Technology Category

• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

In this work, we propose Oph-Guid-RAG, a multimodal visual RAG system for ophthalmology clinical question answering and decision support. We treat each guideline page as an independent evidence unit and directly retrieve page images, preserving tables, flowcharts, and layout information. We further design a controllable retrieval framework with routing and filtering, which selectively introduces external evidence and reduces noise. The system integrates query decomposition, query rewriting, retrieval, reranking, and multimodal reasoning, and provides traceable outputs with guideline page references. We evaluate our method on HealthBench using a doctor-based scoring protocol. On the hard subset, our approach improves the overall score from 0.2969 to 0.3861 (+0.0892, +30.0%) compared to GPT-5.2, and achieves higher accuracy, improving from 0.5956 to 0.6576 (+0.0620, +10.4%). Compared to GPT-5.4, our method achieves a larger accuracy gain of +0.1289 (+24.4%). These results show that our method is more effective on challenging cases that require precise, evidence-based reasoning. Ablation studies further show that reranking, routing, and retrieval design are critical for stable performance, especially under difficult settings. Overall, we show how combining visionbased retrieval with controllable reasoning can improve evidence grounding and robustness in clinical AI applications,while pointing out that further work is needed to be more complete.

Problem

• ophthalmology
• clinical decision support
• retrieval-augmented generation
• guideline grounding
• multimodal reasoning

Innovation

• retrieval-augmented generation
• multimodal reasoning
• visual retrieval
• clinical decision support
• evidence grounding