The Doctor and the Algorithm

Scott highlights the potential synergies of often-siloed literatures from bioethics, critical algorithm studies, and government regulation. Importantly, he encourages experts to engage in conversation by introducing key concepts from their respective fields, articulating common goals, and proposing a rubric for assessing how well a particular technology meets the standards for a just AI...This book is most suitable for scholars and researchers who are already familiar with at least one of the literatures on which the author draws. * Choice * A seminal and groundbreaking study that is timely, well written, impressively informative, exceptionally thought provoking, and enhanced for the reader with the inclusion of an eight page Technical Appendix, twenty four pages of Notes, a sixteen page listing of References, and a six page Index, The Doctor and the Algorithm: Promise, Peril, and the Future of Health AI is unreservedly recommended for personal, professional, community, medical school, college, and university library Health/Medicine and AI collections and supplemental curriculum studies lists. * Library Bookwatch * In this book, S. Scott Graham meticulously lays out examples of AI models being used in medical care, critically examines their use, and contrasts what is known about AI from the scientific literature with claims being made by AI model developers, medical experts, and futurists. In doing so, Dr. Graham brings health AI to life by painting a rich picture of how AI is developed, studied, and marketed, and why these processes fundamentally incentivize hype as a core component of adoption. This book is refreshing because it directly takes experts to task for their claims on medical AI that result in extreme hype or cynicism. For AI newcomers and experts alike, this book will leave you with a deep appreciation for the complex social phenomena that underlie the lifecycle of AI models from development to deployment. * Karandeep Singh, MD, MMSc, Chair, Michigan Medicine Clinical Intelligence Committee, University of Michigan *