MMedAgent: Learning to Use Medical Tools with Multi-modal Agent

Authors:
  • BLBinxu Li
    Binxu Li
  • TYTiankai Yan
    Tiankai Yan
  • YPYuanting Pan
    Yuanting Pan
  • ZXZhe Xu
    Zhe Xu
  • JLJie Luo
    Jie Luo
  • RJRuiyang Ji
    Ruiyang Ji
  • SLShilong Liu
    Shilong Liu
  • HDHaoyu Dong
    Haoyu Dong
  • ZLZihao Lin
    Zihao Lin
  • YWYixin Wang
    Yixin Wang

Abstract
Multi-Modal Large Language Models (MLLMs), despite being successful, exhibit limited generality and often fall short when compared to specialized models. Recently, LLM-based agents have been developed to address these challenges by selecting appropriate specialized models as tools based on user inputs. However, such advancements have not been extensively explored within the medical domain. To bridge this gap, this paper introduces the first agent explicitly designed for the medical field, named \textbf{M}ulti-modal \textbf{Med}ical \textbf{Agent} (MMedAgent). We curate an instruction-tuning dataset comprising six medical tools solving seven tasks, enabling the agent to choose the most suitable tools for a given task. Comprehensive experiments demonstrate that MMedAgent achieves superior performance across a variety of medical tasks compared to state-of-the-art open-source methods and even the closed-source model, GPT-4o. Furthermore, MMedAgent exhibits efficiency in updating and integrating new medical tools.