Thinking Like a Doctor: Conversational Diagnosis through the Exploration of Diagnostic Knowledge Graphs

TL;DR

This paper introduces a conversational diagnosis system that uses a diagnostic knowledge graph to generate and verify hypotheses through iterative questioning, improving accuracy and realism in simulated clinical interactions.

Contribution

It presents a novel knowledge graph-based approach for multi-turn diagnostic reasoning that better reflects real-world clinical conversations and patient interactions.

Findings

Improved diagnostic accuracy over baseline models

Enhanced efficiency in hypothesis verification

Physician evaluations confirm realism and clinical utility

Abstract

Conversational diagnosis requires multi-turn history-taking, where an agent asks clarifying questions to refine differential diagnoses under incomplete information. Existing approaches often rely on the parametric knowledge of a model or assume that patients provide rich and concrete information, which is unrealistic. To address these limitations, we propose a conversational diagnosis system that explores a diagnostic knowledge graph to reason in two steps: (i) generating diagnostic hypotheses from the dialogue context, and (ii) verifying hypotheses through clarifying questions, which are repeated until a final diagnosis is reached. Since evaluating the system requires a realistic patient simulator that responds to the system's questions, we adopt a well-established simulator along with patient profiles from MIMIC-IV. We further adapt it to describe symptoms vaguely to reflect real-world patients during early clinical encounters. Experiments show improved diagnostic accuracy and efficiency over strong baselines, and evaluations by physicians support the realism of our simulator and the clinical utility of the generated questions. Our code will be released upon publication.