American Board of Artificial Intelligence in Medicine Provides Training, Certification
A new organization has been formed to provide artificial intelligence-related educational resources and ethical guidance for healthcare professionals. In a recent interview with Healthcare Innovation, the co-founders of the American Board of Artificial Intelligence in Medicine (ABAIM) discussed their progress and future goals.
Formed in 2020 by a team of clinicians and data scientists, ABAIM teaches live monthly two-day CME-accredited courses on AI concepts to physicians and nurses from around the world. It also offers an educational certification on AI in medicine to professionals that meet a standard of competency. Recipients of the certification have used it to spur change in their organizations, seek new job opportunities, or make career changes in anticipation of the upcoming wave of medical AI technology, the organization said.
Anthony Chang, M.D., is chief intelligence and innovation officer at the Sharon Disney Lund Medical Intelligence and Innovation Institute (MI3) at Children’s Hospital of Orange County, in Orange, Calif. He said one reason for launching ABAIM was that he and colleagues realized that realized that a lot of people like to think of themselves as knowledgeable in AI, but really have no basis for that belief. “We wanted to provide a baseline,” he explained. “The course makes them conversational in this area; it's no different than being conversational in a language. If you're conversational, you can get a much more in-depth appreciation for what it is. But if you're not even conversational, then you're not going to understand most of the content.”
Among the topics covered include: How will AI technologies be integrated into existing systems? How do we negate algorithmic bias on a clinician level? How do we teach rapidly changing AI topics to medical students?
Orest Boyko, M.D., Ph.D., associate professor (research) of psychology at The USC Michelson Center for Convergent Bioscience Bridge Institute, located at the University of Southern California, said that one challenges is that when new technologies enter into medicine, oftentimes they enter through a subspecialty, and that subspecialty tries to organize education. “What Anthony and I believed was that we should create an educational environment that crossed all the subspecialties.”
They reached out to a group of AI experts and subject matter experts to help in defining the critical concepts, vocabulary, terminology and lexicon that clinicians should become comfortable with over time in AI. “I think the certification says that people have learned the basic concepts of AI,” Boyko said. “Then the individual can decide how that could help them in their specific environment to advance their careers and to advance their contribution in the healthcare system.”
Boyko said ABAIM also offers weekly office hours, so any student who has attended the course can attend the office hours, which then allows them an opportunity to ask questions, or just listen to the discussion.
The initial offering is an introductory course and certification. ABAIM is now going to work on intermediate and advanced levels, Chang said. “We're working toward a clinician board certification, which is a nice final destination for all of us, including a fellowship training program. This is going to be very much in parallel with the informatics route, but perhaps democratized earlier than informatics was,” he added.
Chang also is the author of an upcoming book entitled “Intelligence-Based Medicine: Principles and Applications of Artificial Intelligence and Human Cognition in Clinical Medicine and Healthcare” and is the Editor-in-Chief of Intelligence-Based Medicine, an open access peer-review journal, published by Elsevier.
With the field of AI changing so fast, the course material needs to be updated regularly. “I swap out about 10 percent of the slides every single month,” Chang said. “That's pretty aggressive. No review course does that. I'm always giving talks, and I always have to update my talks anyway.”
The intermediate course, expected to launch before the end of the year, will include three workshops: one on how to review a paper on AI in medicine; the second one is how to design a project on your own; and the third one is coming up with an AI strategy for your group or organization. “Those three were singled out as really practical sessions. Once you have this background knowledge, and once you become conversational, then you can start having fun,” Chang said. “And of course, we're already thinking about what the advanced course would look like.”
He stressed that this is not about teaching people programming, which is probably what people expect. “It’s everything except that part, because I think the programming part you can learn on your own. And it's not the most important thing anyway. We think the most important part is appreciation for the AI as a resource, and then deployment and knowing the limitations and the ethics.”
Co-chaired by Boyko and Chang, ABAIM’s growing board comprises professionals from an array of multidisciplinary and institutional backgrounds. These currently include: Matthew Lungren, M.D., Assistant Professor of Radiology, Stanford University Medical Center; Kevin Maher, Professor of Pediatrics, Emory University School of Medicine; Spyro Mousses, Ph.D., CEO, Systems Oncology; Tanveer Syeda-Mahmood, Ph.D., IBM Fellow & Chief Scientist, IBM Research; Sharief Taraman, M.D., Health Science Associate Professor, UC Irvine School of Medicine; and Dennis Wall, Ph.D., Associate Professor of Pediatrics, Psychiatry, and Biomedical Data Science, Stanford Medical School.
