Guiding principles for the responsible development of artificial intelligence tools for healthcare

Badal et al. outline principles that should be adopted during the development of artificial intelligence-based healthcare tools. These principles expand upon principles proposed by several organizations by emphasizing that AI be developed to improve longstanding health care challenges. Although a number of guidelines have been put forth to enhance the use of artificial intelligence (AI) in healthcare, the importance of AI in addressing long-standing healthcare concerns has not received enough attention. We suggest that AI be developed with the following goals in mind: reducing health disparities; reporting clinically meaningful outcomes; reducing overdiagnosis and overtreatment; having high healthcare value; taking into account biographical drivers of health; being easily adaptable to the local population; encouraging a learning healthcare system; and facilitating shared decision-making. We present questions that AI developers can use when putting each of these concepts to work, and we illustrate them with cases from the study of breast cancer. Clinical medicine, biological research, public and global health, and healthcare administration are all expected to undergo radical change as a result of artificial intelligence (AI).