arXiv 2503.164632025-02-24cs.AIcs.CLcs.HC

通过临床经验学习提升大语言模型智能体的交互式诊断能力

Improving Interactive Diagnostic Ability of a Large Language Model Agent Through Clinical Experience Learning

Research Center of Computational Life Science, Zhejiang Lab（浙江实验室计算生命科学研究中心）；TransTek Medical（TransTek医疗）；Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University（浙江大学医学院附属邵逸夫医院）；College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）

Zhoujian Sun, Ziyi Liu, Cheng Luo, Jiebin Chu, Zhengxing Huang

备注：: 30 pages

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对大语言模型在交互式诊断中性能骤降的问题，本研究定位缺陷在初始诊断阶段，利用350多万份中美电子病历开发即插即用增强型LLM智能体，较基线提升超30%，准确率接近完整临床数据水平。

摘要：

大语言模型（LLM）的最新进展在医疗诊断领域展现出可观前景，部分研究显示其在特定场景下的表现优于人类医师。但LLM的诊断能力往往被高估，在需要主动收集信息的交互式诊断场景中，其性能会大幅下降。本研究探究了这一性能下降现象的内在机制并提出解决方案。我们发现，LLM的主要缺陷存在于初始诊断阶段，尤其是信息收集效率和初始诊断形成环节，而非后续的鉴别诊断阶段。为解决这一局限，我们利用中美医疗机构的350多万份电子病历，开发了一种即插即用增强型（PPME）LLM智能体。该方法将经监督学习和强化学习技术训练的初始疾病诊断与现病史问诊专用模型集成到系统中。实验结果显示，PPME LLM较基线模型提升了30%以上，其在交互式诊断场景下的最终诊断准确率接近使用完整临床数据时的水平。这些发现为开发自主诊断系统提供了可观潜力，但仍需进一步验证研究。

英文摘要：

Recent advances in large language models (LLMs) have shown promising results in medical diagnosis, with some studies indicating superior performance compared to human physicians in specific scenarios. However, the diagnostic capabilities of LLMs are often overestimated, as their performance significantly deteriorates in interactive diagnostic settings that require active information gathering. This study investigates the underlying mechanisms behind the performance degradation phenomenon and proposes a solution. We identified that the primary deficiency of LLMs lies in the initial diagnosis phase, particularly in information-gathering efficiency and initial diagnosis formation, rather than in the subsequent differential diagnosis phase. To address this limitation, we developed a plug-and-play method enhanced (PPME) LLM agent, leveraging over 3.5 million electronic medical records from Chinese and American healthcare facilities. Our approach integrates specialized models for initial disease diagnosis and inquiry into the history of the present illness, trained through supervised and reinforcement learning techniques. The experimental results indicate that the PPME LLM achieved over 30% improvement compared to baselines. The final diagnostic accuracy of the PPME LLM in interactive diagnostic scenarios approached levels comparable to those achieved using complete clinical data. These findings suggest a promising potential for developing autonomous diagnostic systems, although further validation studies are needed.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。