MEDDxAgent: A Unified Modular Agent Framework for Explainable Automatic Differential Diagnosis

2025-02-26

AI Summary

Existing automated differential diagnosis (DDx) methods suffer from limited clinical utility and interpretability due to evaluation on single datasets, isolated module optimization, reliance on complete electronic health records, and assumptions of single-step diagnosis. This work proposes a modular agent framework for interactive DDx, introducing a novel tripartite collaborative architecture—comprising DDxDriver, a history-elicitation simulator, and two domain-specialized agents—that enables multi-turn, question-driven knowledge retrieval and dynamic diagnostic ranking, thereby transcending the conventional single-step paradigm. The method integrates LLM-powered modular design, simulated interactive consultation, domain-knowledge augmentation, and a multi-disease, interactive evaluation benchmark. Empirical evaluation across respiratory, dermatological, and rare disease tasks demonstrates >10% absolute improvement in diagnostic accuracy over single-turn baselines, maintains compatibility with LLMs of varying scales, and provides fully traceable, stepwise reasoning provenance.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Differential Diagnosis (DDx) is a fundamental yet complex aspect of clinical decision-making, in which physicians iteratively refine a ranked list of possible diseases based on symptoms, antecedents, and medical knowledge. While recent advances in large language models have shown promise in supporting DDx, existing approaches face key limitations, including single-dataset evaluations, isolated optimization of components, unrealistic assumptions about complete patient profiles, and single-attempt diagnosis. We introduce a Modular Explainable DDx Agent (MEDDxAgent) framework designed for interactive DDx, where diagnostic reasoning evolves through iterative learning, rather than assuming a complete patient profile is accessible. MEDDxAgent integrates three modular components: (1) an orchestrator (DDxDriver), (2) a history taking simulator, and (3) two specialized agents for knowledge retrieval and diagnosis strategy. To ensure robust evaluation, we introduce a comprehensive DDx benchmark covering respiratory, skin, and rare diseases. We analyze single-turn diagnostic approaches and demonstrate the importance of iterative refinement when patient profiles are not available at the outset. Our broad evaluation demonstrates that MEDDxAgent achieves over 10% accuracy improvements in interactive DDx across both large and small LLMs, while offering critical explainability into its diagnostic reasoning process.

Problem

• Unified Modular Agent Framework
• Explainable Automatic Differential Diagnosis
• Iterative Learning in Clinical Decision-Making

Innovation

• Modular framework for diagnosis
• Iterative learning process
• Specialized agents integration