发表机构
Hunan University; ByteDance; Duke University; Westlake University; The University of Hong Kong; Nanyang Technological University; Institute of Automation, Chinese Academy of Sciences; University of Macau; The Ohio State University（湖南大学; 字节跳动; 杜克大学; 西湖大学; 香港大学; 南洋理工大学; 中国科学院自动化研究所; 澳门大学; 俄亥俄州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究探讨大语言模型等对医疗智能体的重塑，从临床部署出发，将其形式化为决策系统并给出自主性分类。沿统一框架扩展，强调临床环境扩展为关键方向，定位临床自我进化为前沿，还研究了多领域应用及挑战，提供医学成像系统路线图。
AI 中文摘要
大语言模型和视觉语言模型联合解释和推理图像与文本的能力不断增强，正在重塑医疗智能体，使其从特定任务预测器向能在临床环境中感知、推理、规划、记忆和行动的自主系统转变。本研究从临床部署出发，探讨医疗智能体在实际应用中所需的任务、抗污染基准和交互式训练环境。将医疗智能体形式化为部分可观测下的序列决策系统，并给出了辅助、合作和完全自主操作的三级自主性分类法。沿着由框架扩展、能力扩展和环境扩展组成的统一扩展框架，临床环境扩展被视为在PACS、EHR和FHIR生态系统中运行的智能体最具可行性但未充分探索的方向。临床自我进化被定位为关键研究前沿，借鉴自我改进智能体、智能体训练环境和测试时计算扩展的见解。研究了放射学、病理学、眼科和医院工作流程中的应用以及包括幻觉、级联故障和公平性在内的部署挑战。通过整合300多篇参考文献，特别是2025年至2026年的进展，为实际临床实践中可信的、自我改进的医学成像系统提供了路线图。
英文摘要
The growing ability of large language models and vision-language models to jointly interpret and reason over images and text is reshaping medical imaging AI, moving it from task-specific predictors toward autonomous agents that perceive, reason, plan, remember, and act in clinical environments. This survey departs from the capability-first perspective of existing literature and instead begins from clinical deployment, asking what tasks, contamination-resistant benchmarks, and interactive training environments are required before medical agents can be trusted in practice. Medical agents are formalized as sequential decision-making systems under partial observability, together with a three-level autonomy taxonomy spanning assisted, cooperative, and fully autonomous operation. The field is organized along a unified scaling spine consisting of framework scaling, capability scaling, and environment scaling. Within this framework, clinical environment scaling, the integration of tools, data, and clinical gyms, is identified as the most actionable yet underexplored direction for agents operating in PACS, EHR, and FHIR ecosystems. Clinical self-evolution, where agents improve through interaction with their environments rather than parameter scaling alone, is further positioned as a key research frontier, drawing insights from self-improving agents, agent gyms, and test-time compute scaling. Applications across radiology, pathology, ophthalmology, and hospital workflows are examined together with deployment challenges including hallucination, cascading failures, and fairness. By consolidating more than 300 references, with particular emphasis on advances from 2025 to 2026, this survey provides a roadmap toward trustworthy, self-improving medical imaging systems for real clinical practice.
CommentsProject page: https://github.com/zhcz328/Awesome-Medical-Agents
