Ethical Principles and Considerations concerning the Use of Artificial Intelligence in Healthcare*

Abstract

The use of artificial intelligence (AI) in healthcare settings has become increasingly common. Many hope that AI will remove constraints on human and material resources and bring innovations in diagnosis and treatment. However, the deep learning techniques and resulting black box problem of AI raise important ethical concerns. To address these concerns, this article explores some of the relevant ethical domains, issues, and themes in this area and proposes principles to guide use of AI in healthcare. Three ethical themes are identified, including respect for person, accountability, and sustainability, which correspond to the three domains of data acquisition, clinical setting, and social environment. These themes and domains were schematized with detailed explanations of relevant ethical issues, concepts, and applications, such as explainability and accountability. Additionally, it is argued that conflicts between ethical principles should be resolved through deliberative democratic methods and a consensus building process.