Speaker:
Abstract: This lecture gives an historical overview of the fundamental principles of AI/ML, followed by a discussion of how these fundamental principles apply to modern biomedical implementations and how data sources, structure and topology lead to potential biases. The lecture will demonstrate these issues with specific examples from our work at Utah showing the importance of using real word data (RWD), the use of conformal prediction to quantify uncertainty, and the importance of using temporal ML approaches.
Bio: Dr. Facelli attended the University of Buenos Aires where he got his Ph.D. in physics in 1982. In 1983 he did post-doctoral research at the University of Arizona and the following year he joined the University of Utah. At the University of Utah, he was the Director of the Center for High Performance Computing from 1995 to 2013, and he is currently, Distinguished Professor of Biomedical Informatics, and Associate Director for Biomedical Informatics at the Utah Clinical and Translational Science Institute. He served as Chair of the Coalition for Scientific Computing (CASC, https://casc.org/) during 2003 and 2004 and as President of the University of Utah Academic Senate from 2019-20. He was elected as a Fellow of the American College of Medical Informatics (ACMI) in 2014 and elected a Fellow of the Academy of Science Health Educators in 2017. He is co-author of more than 280 international peer review publications in diverse fields of computational sciences. His work has been funded by NSF, NIH, DoE and DoD.
Join Zoom Meeting: https://duke.zoom.us/j/98328208324?pwd=8f2WRs8ExQHSLhF3BbtlOJXPX0L5Z2.1
Meeting ID: 983 2820 8324
Passcode: 103794