Incoming Faculty, UC San Diego
Chief Health AI Officer, UC San Diego Health
Associate CMIO for Inpatient Care, UC San Diego Health
TITLE OF TALK
Why Health AI Implementations Fail
ABSTRACT
In this talk, Dr. Singh will use real-world examples to illustrate common implementation roadblocks that can lead AI models to fail. AI model implementations can fail for many reasons, including issues related to models (reproducibility, transportability, net benefit, modifiable vs. non-modifiable risk) and issues related to interventions (lack of efficacy, wrong end-users, increasing workload, readiness for change, and resource constraints). I will demonstrate how we can anticipate these issues, estimate their impact, and better design interventions to make a positive impact on patient care.
LEARNING OBJECTIVES
• Identify common reasons why artificial intelligence models fail to perform well in real-world health care settings.
• Describe implementation challenges that prevent AI models from making a clinical impact when implemented.
• Review ethical frameworks, regulatory guidance, and educational efforts that are underway to support the implementation of AI models in health care.
ABOUT THE SPEAKER
Karandeep Singh, MD, MMSc, was an Associate Professor of Learning Health Sciences, Internal Medicine, Urology, and Information at the University of Michigan and has recently been hired as an incoming faculty member, Chief Health AI Officer, and Associate CMIO for Inpatient Care at UC San Diego Health. He directed the Machine Learning for Learning Health Systems (ML4LHS) Lab, which focused on translational issues related to the implementation of machine learning models within health systems. He has also served as an Associate Chief Medical Information Officer of Artificial Intelligence for Michigan Medicine and has been the Associate Director for Implementation for Precision Health at the University of Michigan.
He completed his internal medicine residency at UCLA Medical Center, where he served as chief resident, and a nephrology fellowship in the combined Brigham and Women’s Hospital/Massachusetts General Hospital program in Boston, MA. He completed his medical education at the University of Michigan Medical School and holds a master’s degree in medical sciences in Biomedical Informatics from Harvard Medical School. He is board-certified in internal medicine, nephrology, and clinical informatics.
His work in machine learning and digital health has been published in leading journals in their respective fields, including the New England Journal of Medicine, Lancet, British Medical Journal, JAMA Internal Medicine, Nature Machine Intelligence, Health Affairs, Clinical Journal of the American Society of Nephrology, Ophthalmology, Radiology, and European Urology.
