Regulation of medical artificial intelligence in public health: governance lessons and implications for China

With the deep integration of artificial intelligence (AI) technologies into public health and clinical care, challenges including dynamic algorithm iteration, data security risks, and ambiguous liability allocation pose multifaceted threats to traditional regulatory frameworks. Among these challenges, this study focuses specifically on the tension between static pre-market approval and the continuous iteration of AI-enabled medical systems. Globally, the European Union (EU), the United States (US), and Japan have developed distinct regulatory models that offer valuable lessons for China. This paper conducts a comparative analysis of medical AI regulatory logics and institutional practices across jurisdictions, with particular attention to algorithm modification, lifecycle oversight, and continuous post-market evaluation. Building on international experience, we propose an optimized institutional pathway for China that centers on differentiated algorithm-change control, continuous lifecycle evaluation, and risk-based regulatory adjustment, while treating liability allocation, risk classification, and data governance as supporting dimensions. This framework provides theoretical support and practical guidance for improving China's medical AI regulatory system, promoting standardized technology development, and supporting the orderly development of public health services in the digital health era.