医疗人工智能临床应用的伦理思考OA北大核心

Medical artificial intelligence(AI)is a new type of application formed by the combination of machine learning,computer vision,natural language processing,and other technologies with clinical medical treatment.With the continuous iteration and development of relevant technologies,medical AI has shown great potential in improving the efficiency of diagnosis and treatment,and service quality,but it also increases the possibility of triggering ethical issues.Ethical issues resulting from the clinical application of medical AI were analyzed,including the lack of algorithmic interpretability and transparency of medical AI,leading to information asymmetry and cognitive discrepancies;the concerning status of security and privacy protection of medical data;and the complex and unclear division of responsibilities due to the collaborative participation of multiple subjects in the clinical application of medical AI,resulting in increased difficulty in the identification of medical accidents and clarification of responsibilities.The paper proposed the principles of not harming patients'interests,physician's subjectivity,fairness and inclusiveness,and rapid response.It also explored the strategies and implementation paths for responding to the ethical issues of medical AI from multiple perspectives,including standardizing the environment and processes,clarifying responsibility attribution,continuously assessing the impact of data protection,guaranteeing data security,ensuring model transparency and interpretability,carrying out multi-subject collaboration,as well as the principles of being driven by ethical values and adhering to the"human health-centeredness."It aimed to provide guidance for the healthy development of medical AI,ensuring technological progress while effectively managing and mitigating accompanying ethical risks,thereby promoting the benign development of medical AI technology and better serving the healthcare industry and patients.