Collaborative Medical Triage under Uncertainty: A Multi-Agent Dynamic Matching Approach
🤖 AI Summary
To address three key challenges in AI-powered triage—(1) hallucination-prone misdiagnosis due to insufficient medical expertise, (2) cross-institutional heterogeneity in departmental structures, and (3) low efficiency of detail-oriented sequential questioning—this paper proposes a multi-agent collaborative triage framework. It introduces a tripartite agent architecture comprising Reception, Interrogation, and Department agents, integrating structured symptom elicitation, rule-guided dynamic interaction, and large language model–driven medical data completion. A joint reasoning mechanism is designed, combining multi-agent reinforcement learning with department-specific clinical rules to enable rapid cross-hospital adaptation. Evaluated on 3,360 real-world cases, the framework achieves 89.2% accuracy for primary department classification and 73.9% for secondary department classification after four interaction rounds—demonstrating substantial improvements in triage accuracy, robustness, and deployment flexibility, particularly in emergency settings.
Application Category
📝 Abstract
The post-pandemic surge in healthcare demand, coupled with critical nursing shortages, has placed unprecedented pressure on emergency department triage systems, necessitating innovative AI-driven solutions. We present a multi-agent interactive intelligent system for medical triage that addresses three fundamental challenges in current AI-based triage systems: insufficient medical specialization leading to hallucination-induced misclassifications, heterogeneous department structures across healthcare institutions, and inefficient detail-oriented questioning that impedes rapid triage decisions. Our system employs three specialized agents - RecipientAgent, InquirerAgent, and DepartmentAgent - that collaborate through structured inquiry mechanisms and department-specific guidance rules to transform unstructured patient symptoms into accurate department recommendations. To ensure robust evaluation, we constructed a comprehensive Chinese medical triage dataset from a medical website, comprising 3,360 real-world cases spanning 9 primary departments and 62 secondary departments. Through systematic data imputation using large language models, we address the prevalent issue of incomplete medical records in real-world data. Experimental results demonstrate that our multi-agent system achieves 89.2% accuracy in primary department classification and 73.9% accuracy in secondary department classification after four rounds of patient interaction. The system's pattern-matching-based guidance mechanisms enable efficient adaptation to diverse hospital configurations while maintaining high triage accuracy. Our work provides a scalable framework for deploying AI-assisted triage systems that can accommodate the organizational heterogeneity of healthcare institutions while ensuring clinically sound decision-making.
Problem
Research questions and friction points this paper is trying to address.
Addresses AI-driven misclassifications in medical triage due to insufficient specialization
Solves inefficiency in triage questioning for rapid decision-making
Adapts to heterogeneous department structures across healthcare institutions
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multi-agent system for accurate medical triage
Structured inquiry mechanisms reduce misclassifications
Data imputation handles incomplete medical records
🔎 Similar Papers
No similar papers found.
