医疗人工智能算法风险防范的法治化思考

Reflection on the Rule of Law for Risk Prevention of Medical Artificial Intelligence Algorithms

医疗人工智能算法应用过程中存在诊疗活动自主性受限、诊疗过程透明性匮乏、医疗算法偏见性诱导以及安全性缺失风险。在风险防治过程中出现了传统监管缺陷、主体资格界定模糊、算法权力控制失衡、责任主体追溯难等现实挑战,迫切需要运用法律手段予以纾解。因此,应构筑医疗算法风险的全面防治体系,坚持医疗算法权力驯化的向善向美属性,增强医疗算法权力对抗中主体权利的质效,厘清医疗算法责任归属中归责原则的困惑进而为医疗人工智能算法风险的防治补足法治力量。

There are risks in the application process of medical artificial intelligence algorithms,such as limited autonomy in diagnosis and treatment activities,lack of transparency in the diagnosis and treatment process,biased induction of medical algorithms,and lack of security.In the process of risk prevention and control,there have been practical challenges such as traditional regulatory deficiencies,unclear definition of subject qualifications,imbalanced control of algorithmic power,and difficulty in tracing responsible subjects,which urgently require the use of legal means to alleviate.Therefore,a comprehensive prevention and control system for medical algorithm risks should be established,adhering to the good and beautiful nature of the domestication of medical algorithm power,enhancing the quality and effectiveness of subject rights in the confrontation of medical algorithm power,clarifying the confusion of the principle of attribution in the attribution of medical algorithm responsibility,and thus supplementing the legal force for the prevention and control of medical artificial intelligence algorithm risks.