EpiGraph: A Knowledge Graph and Benchmark for Evidence-Intensive Reasoning in Epilepsy

2026-05-10

AI Summary

This study addresses the critical need for integrating multi-source, heterogeneous clinical knowledge to support evidence-based reasoning in epilepsy diagnosis and treatment, a domain currently hindered by the lack of structured knowledge frameworks and systematic evaluation benchmarks. To bridge this gap, the authors present EpiGraph, the first large-scale epilepsy knowledge graph, which synthesizes 48,166 scientific articles and seven types of clinical resources across five layers of clinical knowledge, along with EpiBench, a comprehensive multi-task benchmark. Building upon this foundation, they propose a Graph-RAG framework to enhance the precision of large language models in neuro-medical reasoning. Experimental results demonstrate that incorporating EpiGraph substantially improves model performance, yielding a 30–41% increase in accuracy for pharmacogenomic inference, thereby validating the efficacy of structured knowledge in augmenting clinical decision-making.

Abstract

Epilepsy diagnosis and treatment require evidence-intensive reasoning across heterogeneous clinical knowledge, including biosignal patterns, genetic mechanisms, pharmacogenomics, treatment strategies, and patient outcomes. In this work, we present \textsc{EpiGraph}, a large-scale epilepsy knowledge graph and benchmark for evaluating knowledge-augmented clinical reasoning. \textsc{EpiGraph} integrates 48,166 peer-reviewed papers and seven clinical resources into a heterogeneous graph containing 24,324 entities and 32,009 evidence-grounded triplets across five clinical layers. Built upon this graph, \textsc{EpiBench} defines five clinically motivated tasks spanning clinical decision-making, EEG report generation, pharmacogenomic precision medicine, treatment recommendation, and deep research planning. We evaluate six LLMs under both standard and Graph-RAG settings. Results show that integrating \textsc{EpiGraph} consistently improves performance across all tasks, with the largest gains observed in pharmacogenomic reasoning (+30--41\%). Our findings demonstrate that structured epilepsy knowledge substantially enhances evidence-grounded clinical reasoning and provides a practical benchmark framework for evaluating knowledge-augmented LLMs in real-world neurological settings. Our code is available at: https://github.com/LabRAI/EEG-KG.

Problem

• epilepsy
• evidence-intensive reasoning
• knowledge graph
• clinical decision-making
• pharmacogenomics

Innovation

• Knowledge Graph
• Evidence-Intensive Reasoning
• Graph-RAG
• Pharmacogenomics
• Clinical Benchmark