Edited by Shai Ben-David, Giuseppe Curigliano, David Koff, Barbara Alicja Jereczek-Fossa, Davide La Torre, and Gabriella Pravettoni, Elsevier Press, 2024, ISBN: 9780443136719 (Paperback), ISBN: 9780443136726 (eBook), 281 +xii pages, $180 (bundle)
This text from Elsevier is composed of 19 chapters, written by a total of 69 listed authors from an international community of medical and other practitioners. Each chapter covers some aspects of artificial intelligence (AI) and/or machine learning (ML) as applied in one or more areas of medicine. While a few chapters include introductory materials (to AI and ML), most cover specific methods and applications relevant to the author’s fields. Most chapters are overviews, rather than instructive of specific methodologies to be used or developed.
This reviewer’s overview of chapters: 1) an impressive overview of AI applied to cancer research and treatment, with introductory material on the basics of AI and ML; 2) useful outcomes using whole slide imaging in pathology; 3) a nice review of current constraints and applications of AI in new drug design; 4) a discussion of AI and ML use to consider drug repurposing in lung cancer research; 5) a brief overview of potential AI interactions involving virtual reality, health care, and wearables; 6) a brief evaluation of the acceptance of AI in enhancing processes surrounding childbirth; 7) a brief overview of the use of AI in diagnostic and predictive pathology; 8) current and probable impacts of AI on the process of oncology workflows; 9) a helpful review of current applications of ML to decision-making in aspects of child and youth mental health diagnosis and care; 10) AI applied to cancer detection using hyperspectral imaging of whole slides; 11) a sampling and analysis of a recent 10-year epoch of publications and regulations involving the uses of AI and ML; 12) an overview of the status of AI in radiology, specifically the interactions between innovation, ethics, and regulations; 13) a brief (eight-page) introduction to AI in radiology; 14) an overview of the need to integrate multiple measures (DNA to patient record data) to improve deep learning in solving biomedical problems; 15) a discussion of AI and behavior health economics, with the suggestion that it be used to support rather than force medical decision making; 16) a brief (eight-page) essay on the use of AI to allow physicians more patient interaction; 17) another introduction to ML, with application to radiation oncology, with complaints regarding lack of training data to promote improved outcomes; 18) a good overview of applications of AI in Alzheimer’s disease, Parkinson’s, epilepsy, and related neurological diseases; and 19) a brief (eight-page) essay on the believability and explainability of AI in medicine, terming it “black box” in need of more data for training.
Considering the topics considered in this book, and the elevated level at which it is written, this reviewer suggests that the text will be useful to upper-level entrants to the field, such as graduate students in fields that use or enhance the methodology discussed. Implied, in every chapter, is the promise that new and/or better applications of AI and ML in medicine await development and deployment. As the chapters are very well referenced, it is suggested that the purchaser at least obtain the electronic version of this text to easily access the plentiful electronic references.
—Reviewed by Paul H. King
Vanderbilt University