KERAP: A Knowledge-Enhanced Reasoning Approach for Accurate Zero-shot Diagnosis Prediction Using Multi-agent LLMs

TL;DR

KERAP is a multi-agent framework that enhances large language models with knowledge graphs to improve zero-shot medical diagnosis prediction, addressing hallucinations and reasoning limitations for more reliable and interpretable results.

Contribution

This paper introduces KERAP, a novel multi-agent approach that integrates knowledge graphs with LLMs for accurate zero-shot medical diagnosis prediction.

Findings

KERAP improves diagnostic accuracy over baseline LLMs.

The framework reduces hallucinations and increases interpretability.

Experimental results show scalability and efficiency in medical diagnosis tasks.

Abstract

Medical diagnosis prediction plays a critical role in disease detection and personalized healthcare. While machine learning (ML) models have been widely adopted for this task, their reliance on supervised training limits their ability to generalize to unseen cases, particularly given the high cost of acquiring large, labeled datasets. Large language models (LLMs) have shown promise in leveraging language abilities and biomedical knowledge for diagnosis prediction. However, they often suffer from hallucinations, lack structured medical reasoning, and produce useless outputs. To address these challenges, we propose KERAP, a knowledge graph (KG)-enhanced reasoning approach that improves LLM-based diagnosis prediction through a multi-agent architecture. Our framework consists of a linkage agent for attribute mapping, a retrieval agent for structured knowledge extraction, and a prediction agent that iteratively refines diagnosis predictions. Experimental results demonstrate that KERAP enhances diagnostic reliability efficiently, offering a scalable and interpretable solution for zero-shot medical diagnosis prediction.