Our latest TAVR publication with network meta-analysis showing among 1547 bicuspid aortic valve patients that mortality and other outcomes are comparable (except for prosthetic valve regurgitation) among diverse valve generation and expandable mechanism types. Congratulations to our multi-center team from Emory University School of Medicine, Vanderbilt University Medical Center, University of California San Diego, Mount Sinai Medical Center of Miami, Baylor College of Medicine, and University of Texas Health Science Center at Houston McGovern Medical School. Though we conducted our study according to current best-practices, is it unethically inefficient for the medical and patient community overall to have to wait years for the conclusions of randomized trials and their subsequent meta-analyses to know what treatments work best? The FDA is working with researchers and drug companies to consider real-world evidence (RWE) as part of the approval process to expedite new treatments–but what should the final guidelines looks like (i.e. to what degree if any can we use observational data instead of clinical trials and if so can we run millions of versions of a randomized trial through AI the way self-driving cars are working to teach themselves over millions of simulated miles after driving a much smaller number of real drive miles)?
GSAS Chief Data Scientist Dominique Monlezun, MD, PhD, PhD, MPH is AI+MD: our voice advocating for patients through ethical and effective AI, from the world’s first doctoral-level physician-data scientist and ethicist. Dr. Monlezun regularly breaks down the latest AI advances and their ethical aspects to ensure scientists, health professionals, policy makers, and politicians can unite to protect patients and vulnerable patient populations.