Agentic AI in Healthcare & Medicine: A Seven-Dimensional Taxonomy for Empirical Evaluation of LLM-based Agents

This paper proposes a seven-dimensional taxonomy for the empirical evaluation of Large Language Model (LLM)-based agents in healthcare and medicine. It reviews 49 studies, categorizing their capabilities across cognitive, knowledge management, interaction, adaptation, safety, ethics, framework, and core task dimensions, revealing key asymmetries in their development.