Evidence, Logic, and Compliance: Multi-Agent Structured Graph Reasoning with Expert Arbitration for Medical Referral
2026-08-31 • Multiagent Systems
Multiagent Systems
AI summaryⓘ
The authors explain that deciding which hospital department a patient should be referred to is complicated because it involves understanding many different types of information. They point out that current language models struggle because they either focus too much on common disease names or have unorganized teamwork that leads to mistakes. To fix this, they created MASGR, a system where different agents gather evidence from various data sources and work together by building a logical graph to make better, safer decisions. Tests show their method is better at handling complicated cases that need careful judgement between ongoing and emergency care.
Medical ReferralLarge Language Models (LLMs)Multi-Agent SystemsGraph ReasoningMultimodal DataClinical Decision-MakingKnowledge-Guided ArbitrationPatient SafetyChronic Disease ManagementEmergency Intervention
Authors
Qi Peng, Yi Cai, Jialin Cui, Tong Zhu, Yujuan Ding, Qingbao Huang, Tao Wang, Jiayuan Xie, Changmeng Zheng, Qing Li
Abstract
Medical referral (directing patients to the appropriate hospital department) is a complex decision-making process requiring the synthesis of multimodal data, including patient narratives, laboratory indicators, and radiology imaging. While Large Language Models (LLMs) have advanced medical dialogue systems, they struggle with real-world referral tasks due to two primary limitations: (1) Information Overload, where models fixate on high-frequency disease terms while overlooking subtle but critical urgency indicators; and (2) Unstructured Collaboration, where existing multi-agent frameworks rely on loose dialogue that leads to semantic drift and confirmation bias. To address these challenges, we introduce MASGR (Multi-Agent Structured Graph Reasoning), a framework that treats referral not as a classification task but as a structured graph construction problem. MASGR deploys specialized agents to extract evidence from distinct modalities and coordinates them through a clinical reasoning graph. This graph forces agents to establish explicit logical connections between conflicting evidence. Furthermore, we integrate a knowledge-guided arbitration mechanism that prioritizes patient safety rules over standard diagnostic classification. Extensive experiments on real-world medical records demonstrate that MASGR significantly outperforms state-of-the-art LLMs and existing multi-agent systems, particularly in complex cases requiring the balancing of chronic disease management and emergency intervention. The AI contribution lies in the Multi-Agent Structured Graph Reasoning framework that transforms unstructured multi-agent dialogue into a verifiable logical graph construction. The engineering application is demonstrated through its deployment in a complex healthcare decision-making system to optimize the precision of complex medical referrals.