Learning to Be A Doctor: Searching for Effective Medical Agent Architectures

2025-04-15

AI Summary

Existing medical agent systems rely on static, manually designed workflows and struggle to adapt to dynamic diagnostic requirements. To address this, this paper proposes the first fully automated medical agent architecture search framework tailored for clinical diagnosis. Methodologically, it introduces a hierarchical medical agent search space integrating graph neural network modeling, neural architecture search (NAS)-inspired automatic architecture optimization, and reinforcement feedback-driven structural evolution—enabling editability at the node, workflow, and framework levels. Key contributions include: (i) the first realization of diagnostic-feedback-driven workflow self-evolution and cross-scenario adaptation; and (ii) a multi-granularity editable agent node design. Evaluated on skin disease diagnosis, the framework autonomously discovers optimal workflows, achieving continuously improving accuracy and significantly outperforming both handcrafted agents and baseline LLM-based approaches.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Agent/AI Theories and Architectures
• Search and Optimization: Heuristic Search

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Large Language Model (LLM)-based agents have demonstrated strong capabilities across a wide range of tasks, and their application in the medical domain holds particular promise due to the demand for high generalizability and reliance on interdisciplinary knowledge. However, existing medical agent systems often rely on static, manually crafted workflows that lack the flexibility to accommodate diverse diagnostic requirements and adapt to emerging clinical scenarios. Motivated by the success of automated machine learning (AutoML), this paper introduces a novel framework for the automated design of medical agent architectures. Specifically, we define a hierarchical and expressive agent search space that enables dynamic workflow adaptation through structured modifications at the node, structural, and framework levels. Our framework conceptualizes medical agents as graph-based architectures composed of diverse, functional node types and supports iterative self-improvement guided by diagnostic feedback. Experimental results on skin disease diagnosis tasks demonstrate that the proposed method effectively evolves workflow structures and significantly enhances diagnostic accuracy over time. This work represents the first fully automated framework for medical agent architecture design and offers a scalable, adaptable foundation for deploying intelligent agents in real-world clinical environments.

Problem

• Automating medical agent design for dynamic clinical adaptability
• Enhancing diagnostic accuracy through iterative self-improvement
• Addressing inflexibility in static medical workflow systems

Innovation

• Automated design of medical agent architectures
• Hierarchical agent search space for dynamic adaptation
• Graph-based architectures with iterative self-improvement