AI Hospital: Benchmarking Large Language Models in a Multi-agent Medical Interaction Simulator

TL;DR

This paper introduces AI Hospital, a multi-agent simulation framework and benchmark for evaluating large language models in realistic medical interactions, revealing current limitations and guiding future improvements.

Contribution

It presents a novel multi-agent simulation environment and the MVME benchmark for assessing LLMs in complex clinical scenarios with multi-turn interactions.

Findings

LLMs perform significantly worse in multi-turn interactions than in one-step tasks.

The dispute resolution mechanism improves diagnostic accuracy.

Current LLMs still have substantial gaps in clinical diagnostic capabilities.

Abstract

Artificial intelligence has significantly advanced healthcare, particularly through large language models (LLMs) that excel in medical question answering benchmarks. However, their real-world clinical application remains limited due to the complexities of doctor-patient interactions. To address this, we introduce \textbf{AI Hospital}, a multi-agent framework simulating dynamic medical interactions between \emph{Doctor} as player and NPCs including \emph{Patient}, \emph{Examiner}, \emph{Chief Physician}. This setup allows for realistic assessments of LLMs in clinical scenarios. We develop the Multi-View Medical Evaluation (MVME) benchmark, utilizing high-quality Chinese medical records and NPCs to evaluate LLMs' performance in symptom collection, examination recommendations, and diagnoses. Additionally, a dispute resolution collaborative mechanism is proposed to enhance diagnostic accuracy through iterative discussions. Despite improvements, current LLMs exhibit significant performance gaps in multi-turn interactions compared to one-step approaches. Our findings highlight the need for further research to bridge these gaps and improve LLMs' clinical diagnostic capabilities. Our data, code, and experimental results are all open-sourced at \url{https://github.com/LibertFan/AI_Hospital}.