A Survey of LLM-based Agents in Medicine: How far are we from Baymax?

2025/02/16 by Wen‐Xuan Wang, Z. Ma, Wang, Wenxuan +12 · 39 citations
Computer Science · Decision Sciences · Medicine · Psychology · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Data science #FOS: Computer and information sciences #Medicine #Psychology #Scientific Computing and Data Management #Statistical and Computational Modeling #Traditional medicine

paper · pdf · doi:10.48550/arxiv.2502.11211

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2025/02/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

Large Language Models (LLMs) are transforming healthcare through the development of LLM-based agents that can understand, reason about, and assist with medical tasks. This survey provides a comprehensive review of LLM-based agents in medicine, examining their architectures, applications, and challenges. We analyze the key components of medical agent systems, including system profiles, clinical planning mechanisms, medical reasoning frameworks, and external capacity enhancement. The survey covers major application scenarios such as clinical decision support, medical documentation, training simulations, and healthcare service optimization. We discuss evaluation frameworks and metrics used to assess these agents' performance in healthcare settings. While LLM-based agents show promise in enhancing healthcare delivery, several challenges remain, including hallucination management, multimodal integration, implementation barriers, and ethical considerations. The survey concludes by highlighting future research directions, including advances in medical reasoning inspired by recent developments in LLM architectures, integration with physical systems, and improvements in training simulations. This work provides researchers and practitioners with a structured overview of the current state and future prospects of LLM-based agents in medicine.

Cited by

• Lumos: Let there be Language Model System Certification
• Mirror, Mirror on the Wall -- Which is the Best Model of Them All?
• MTBBench: A Multimodal Sequential Clinical Decision-Making Benchmark in Oncology
• Human or LLM as Standardized Patients? A Comparative Study for Medical Education
• BioVerge: A Comprehensive Benchmark and Study of Self-Evaluating Agents for Biomedical Hypothesis Generation
• Prompt-R1: Collaborative Automatic Prompting Framework via End-to-end Reinforcement Learning
• Diagnosing Hallucination Risk in AI Surgical Decision-Support: A Sequential Framework for Sequential Validation
• EdgeRunner 20B: Military Task Parity with GPT-5 while Running on the Edge
• Generative AI for Healthcare: Fundamentals, Challenges, and Perspectives
• Embracing Trustworthy Brain-Agent Collaboration as Paradigm Extension for Intelligent Assistive Technologies
• Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agents
• Do LLMs Recognize Your Latent Preferences? A Benchmark for Latent Information Discovery in Personalized Interaction
• A Call for Collaborative Intelligence: Why Human-Agent Systems Should Precede AI Autonomy
• A Vision for Access Control in LLM-based Agent Systems
• Traj-CoA: Patient Trajectory Modeling via Chain-of-Agents for Lung Cancer Risk Prediction
• Doctor-R1: Mastering Clinical Inquiry with Experiential Agentic Reinforcement Learning
• Policy-as-Prompt: Turning AI Governance Rules into Guardrails for AI Agents
• CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding
• Safety Compliance: Rethinking LLM Safety Reasoning through the Lens of Compliance
• Backdoor Attribution: Elucidating and Controlling Backdoor in Language Models
• Scaling LLM-Driven Multi-Agent Systems: Design Principles and Architectural Scalability Analysis
• AEAS: Actionable Exploit Assessment System
• OpenLens AI: Fully Autonomous Research Agent for Health Infomatics
• H²R: Hierarchical Hindsight Reflection for Multi-Task LLM Agents
• Baichuan-M2: Scaling Medical Capability with Large Verifier System
• Are Large Language Models Dynamic Treatment Planners? An In Silico Study from a Prior Knowledge Injection Angle
• Critique of impure reason: Unveiling the reasoning behaviour of medical large language models
• Infherno: End-to-end Agent-based FHIR Resource Synthesis from Free-form Clinical Notes
• DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents
• TRiSM for Agentic AI: A Review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems
• MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
• AUTOCT: Automating Interpretable Clinical Trial Prediction with LLM Agents
• EndoBench: A Comprehensive Evaluation of Multi-Modal Large Language Models for Endoscopy Analysis
• MedSentry: Understanding and Mitigating Safety Risks in Medical LLM Multi-Agent Systems
• CT-Agent: A Multimodal-LLM Agent for 3D CT Radiology Question Answering
• LLM-Powered AI Agent Systems and Their Applications in Industry
• EviSearch: A Human in the Loop System for Extracting and Auditing Clinical Evidence for Systematic Reviews
• LLM-Based Human-Agent Collaboration and Interaction Systems: A Survey
• From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review
• SAGA: A Security Architecture for Governing AI Agentic Systems
• PrivacyPeek: Auditing What LLM-Based Agents Acquire, Not Just What They Say

Related

• Large Language Models Illuminate a Progressive Pathway to Artificial Healthcare Assistant: A Review
• MedAgentBench: A Realistic Virtual EHR Environment to Benchmark Medical LLM Agents
• Large Language Models as Agents in the Clinic
• Large Language Models for Medicine: A Survey
• A Perspective for Adapting Generalist AI to Specialized Medical AI Applications and Their Challenges