Trust in Medical AI

Abstract

Discussion on AI principles related to privacy, security and accountability are important along with the usage of artificial intelligence (AI) at various fields. There are expectations as to build "trustworthy AI system." We investigated what are the bottleneck on medical AI by creating a workflow diagram of medical treatment by doctors and questioned medical-related experts participated to seminars. As a result, issues on some dilemma situation as well as usefulness (accuracy) of technology, safety, convenience, security, cost were pointed out. In the future, we plan to deepen this qualitative data further by questionnaire survey targeting a large number of people.