Thinking Like a Clinician: A Cognitive AI Agent for Clinical Diagnosis via Panoramic Profiling and Adversarial Debate

2026-04-26

AI Summary

This work addresses the challenges of "tunnel vision" and diagnostic hallucination that large language models often exhibit when processing unstructured electronic health records. To mitigate these issues, the authors propose DxChain, a novel framework that emulates clinical reasoning by integrating a Profile-Then-Plan paradigm for constructing holistic patient profiles, a medical Tree-of-Thought (Med-ToT) algorithm, and an "angel-devil" adversarial debate mechanism. DxChain achieves reliable diagnosis through an iterative three-stage process involving memory anchoring, navigation, and validation. Evaluated on the MIMIC-IV-Ext cardiovascular dataset and the real-world CDM dataset, DxChain significantly outperforms existing methods, demonstrating state-of-the-art performance in both diagnostic accuracy and logical consistency.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Natural Language Processing: (Large) Language Models
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

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

Abstract

The application of large language models (LLMs) in clinical decision support faces significant challenges of "tunnel vision" and diagnostic hallucinations present in their processing unstructured electronic health records (EHRs). To address these challenges, we propose a novel chain-based clinical reasoning framework, called DxChain, which transforms the diagnostic workflow into an iterative process by mirroring a clinician's cognitive trajectory that consists of "Memory Anchoring", "Navigation" and "Verification" phases. DxChain introduces three key methodological innovations to elicit the potential of LLM: (i) a Profile-Then-Plan paradigm to mitigate cold-start hallucinations by establishing a panoramic patient baseline, (ii) a Medical Tree-of-Thoughts (Med-ToT) algorithm for strategic look ahead planning and resource aware navigation, and (iii) a Dialectical Diagnostic Verification procedure utilizing "Angel-Devil" adversarial debates to resolve complex evidence conflicts. Evaluated on two real world benchmarks, MIMIC-IV-Ext Cardiac Disease and MIMIC-IV-Ext CDM, DxChain achieves state-of-the-art performances in both diagnostic accuracy and logical consistency, offering a modular and reliable architecture for next-generation clinical AI. The code is at https://anonymous.4open.science/r/Dx-Chain.

Problem

• tunnel vision
• diagnostic hallucinations
• clinical decision support
• large language models
• electronic health records

Innovation

• DxChain
• Profile-Then-Plan
• Medical Tree-of-Thoughts
• Adversarial Debate
• Clinical Reasoning