大语言模型驱动的医疗智能体研究与应用综述  

A Review on Research and Application of Large Language Model-based Medical Agents

作　　者：李鹏飞[1] 吴静依 郑悦闻 张路霞[1] LI Pengfei;WU Jingyi;ZHENG Yuewen;ZHANG Luxia(National Institute of Health Data Science,Peking University,Beijing 100191,China)

机构地区：[1]北京大学健康医疗大数据国家研究院,北京市100191 [2]浙江省北大信息技术高等研究院,浙江省杭州市311215

出　　处：《中国卫生信息管理杂志》2025年第2期179-186,194,共9页Chinese Journal of Health Informatics and Management

基　　金：国家重点研发计划“基于AI大数据驱动和可信安全计算的创新药物筛选系统研发与应用”(2022YFF1203002);国家自然科学基金“气候变化对人群健康影响的精准评估、动态预警和管理决策研究”(72474010);国家自然科学基金“大数据驱动的重大疾病防控管理”(72125009)。

摘　　要：基于大语言模型(LLM)的医疗智能体(agent)在辅助临床决策、提升医疗服务效率与质量等方面展现出巨大的应用潜力。本文系统梳理了LLM驱动下医疗智能体的研究发展历程,深入分析了主流模型的基本结构及核心前沿技术;概述了其在医疗数据智能治理与挖掘、健康教育与居家健康管理、智能诊断与随访、临床辅助决策等多场景中的研究进展与应用前景;全面梳理了其在临床实践转化过程中面临的主要挑战,并对未来的研究方向与应用前景进行了展望,为推动医疗智能体的深入研究和实际应用提供了参考。Large language model(LLM)-based medical agents exhibit significant potential for supporting clinical decision-making and enhancing the efficiency and quality of healthcare services.This study systematically reviews the development trajectory of LLM-driven medical agents,providing an in-depth analysis of their foundational architectures and core cutting-edge technologies.We summarize the research progress and applications of LLM-based agents across various healthcare scenarios,including intelligent governance and analysis of medical data,health education and self-management,intelligent diagnosis and follow-ups,and clinical decision support.Finally,we identify key challenges in translating LLM-based medical agents into clinical practice and explore future research directions and application prospects.This review aims to serve as a valuable reference for advancing the research and implementation of LLM-based medical agents in healthcare.

关 键 词：医疗智能体 大语言模型 医疗健康 

分 类 号：R197.323[医药卫生—卫生事业管理] R319[医药卫生—公共卫生与预防医学]