MultiDx: A Multi-Source Knowledge Integration Framework towards Diagnostic Reasoning

TL;DR

MultiDx is a two-stage framework that enhances diagnostic reasoning by integrating multiple knowledge sources, improving accuracy and alignment with clinical reasoning processes.

Contribution

It introduces a novel multi-source knowledge integration framework for diagnostic reasoning, leveraging web search, case data, and clinical databases.

Findings

Effective in differential diagnosis on public benchmarks.

Improves alignment with clinical reasoning trajectories.

Outperforms existing methods in diagnostic accuracy.

Abstract

Diagnostic prediction and clinical reasoning are critical tasks in healthcare applications. While Large Language Models (LLMs) have shown strong capabilities in commonsense reasoning, they still struggle with diagnostic reasoning due to limited domain knowledge. Existing approaches often rely on internal model knowledge or static knowledge bases, resulting in knowledge insufficiency and limited adaptability, which hinder their capacity to perform diagnostic reasoning. Moreover, these methods focus solely on the accuracy of final predictions, overlooking alignment with standard clinical reasoning trajectories. To this end, we propose MultiDx, a two-stage diagnostic reasoning framework that performs differential diagnosis by analyzing evidence collected from multiple knowledge sources. Specifically, it first generates suspected diagnoses and reasoning paths by leveraging knowledge from web search, SOAP-formatted case, and clinical case database. Then it integrates multi-perspective evidence through matching, voting, and differential diagnosis to generate the final prediction.~Extensive experiments on two public benchmarks demonstrate the effectiveness of our approach.