“Deep Medicine” – Book review
By Simona Mellino
At the Women’s Brain Project (WBP), we are always interested in reading about advances in healthcare and technology. We recently had the pleasure to read “Deep Medicine: How AI can make Healthcare human again” (Deep Medicine) by Eric Topol, and decided to share our impressions with you.
Eric Topol is a cardiologist, digital medicine expert, and Director of the Scripps Research Translational Institute. He is a professional we follow, and earlier this year, he recently published a review for the National Health Systems (NHS) titled “Preparing the healthcare workforce to deliver the digital future”.
The book “Deep Medicine”
The book aims to provide an extensive overview of the benefits of artificial intelligence (AI) in healthcare. It is comprehensive and spans different fields, including AI applications in mental health, health systems, and virtual medical assistants. Topol defines this book as his “third” exploration about the future of medicine, after his two previous books “Creative Destruction of Medicine” and “The patient will see you now”.
What I liked most about the book:
- The book is easy to follow, even for those without an extensive background in machine learning. Topol takes the time to provide the readers with basic definitions of machine learning and AI. This makes the book easy to read for someone in the healthcare sector, even without any specific knowledge of technology. Furthermore, in his chapter “Deep Liabilities”, he explores the “potentially adverse issues” confronting deep medicine, including biases, job concerns, and ethical issues which provide some context of AI applications in real life.
- Deep Medicine covers a huge variety of examples and references. Topol did an impressive job in pulling together solid literature from the past years in each of his chapters. In fact, he provides not only examples of AI applications, but also data from US clinical practices, including misdiagnosis figures, unnecessary medical tests, and more. This comes from a review of medical literature as well as his own experience.One striking example is the Nature review of 2015, “Precision Medicine: Time for one-person trials”, showing failure rates of top medications in the market. These references clearly communicate the urge for improvement in the healthcare sector and motive the reader to keep exploring the presented content with high interest.
- Topol uses a lot of examples from his life, which I find particularly interesting. In the chapter titled “Deep Diet”, he reviews start-ups in the field of personalized nutrition. He also shows his own gut microbiome assessment taken by an Israeli company. He concludes the chapter by providing an overview of the limitations in personalized nutrition and ties them back to his own experience. This makes the book very real, and as a reader it made me feel closer to the author.
- Perhaps the part that I liked most is the final one, where Topol invites doctors to use the time freed up by technology to get to know their patients better and build a relationship with them. Topol describes the meaning of what “shallow” medicine is, a medicine where doctors do not have sufficient time to spend with their patients and rather focus on procedures: “This is where we are today: patients exist in a world of insufficient data, insufficient time, insufficient context, and insufficient presence. Or, as I say, a world of shallow medicine.”
What I liked least about the book:
- As mentioned above, the book provides an extensive overview of medical and technology-related literature. In some chapters, I failed to really understand the “so what?” of the author. At times, I felt lost in the myriad of examples provided and I asked myself what Topol’s conclusion and recommendation moving forward really was. It is obviously difficult to make bold statements given the novelty of the field.
- I would have loved to see a conclusive chapter covering current policies and regulatory hurdles for the development and deployment of novel technologies. While the book perfectly expresses the need for AI applications and the relevance in the clinical practice, I think it fails to outline the limitations to their large-scale deployment.
A focus on mental health
The book also has a dedicated chapter, called “Mental Health” covering examples of AI to diagnose and predict the response to medications, taking depression as a case study example. The author concludes with a flip side of depression, using AI to ultimately to increase happiness.
To summarize, I would highly recommend this book to people who are keen to explore and read more about digital health and its applications.
At our recent International Forum on Women’s Brain and Mental Health, we discussed the importance of considering sex and gender differences in medicine as a gateway to precision medicine, a response to today’s “shallow medicine” challenges. Understanding that there is an alternative makes it all the more important to take action and catalyze this shift away from “shallow medicine” – so at WBP we support Topol’s viewpoints and deem it part of our mission to contribute to this change.