K1
Chair: Xiaofei Wang (Duke)
Vice Chair: Yeh-Fong Chen (FDA)
Speakers:
David Page (Duke)
Rakhi Kilaru (PPD)
Richard Payne (Lilly)
ShaAvhrée Buckman-Garner (FDA)
Keynote I – Thursday, April 22, 2021
Title: A Perspective on Progress, Challenge and Opportunity
Speaker: ShaAvhrée Buckman-Garner, MD, PhD, FAAP
ShaAvhrée Buckman-Garner, MD, PhD, FAAP is the Director of the Office of Translational Sciences (OTS), Center for Drug Evaluation and Research (CDER), U.S. Food and Drug Administration. OTS is a super office comprised of the Office of Biostatistics, Office of Clinical Pharmacology, the Office of Computational Science, and the Office of Study Integrity and Surveillance. OTS provides oversight to CDER research involving human subjects as well as CDER regulatory science research. OTS is responsible for providing coordination for cross-cutting Critical Path initiatives across CDER in partnership with individual CDER offices. Prior to serving as Director of OTS, Dr. Buckman-Garner served as Deputy Director for OTS and as medical team leader in the Division of Pediatric Drug Development, Office of Counter Terrorism and Pediatric Drug Development, CDER. Dr. Buckman-Garner received her MD and PhD degrees with an emphasis on molecular cell biology from Washington University School of Medicine. Dr. Buckman-Garner completed Pediatric specialty training at Baylor College of Medicine. She is board certified in Pediatrics and Clinical Informatics.
K2
Chair: Xiaofei Wang (Duke)
Vice Chair: Herbert Pang (Genentech)
Speaker: David Page (Duke)
Keynote II – Friday April 23, 2021
Title: Machine Learning from Electronic Health Records
Speaker: David Page, PhD
Machine learning is being used increasingly to construct predictive models from observational clinical data, especially electronic health records (EHRs). We survey how accurately a very wide variety of clinical events can be predicted. Nevertheless, many of the real applications, such as adverse drug event discovery, analysis of polypharmacy, and drug repurposing opportunities, require causal inference and not merely prediction. Therefore we also consider opportunities and challenges in this area, and we propose methods that combine approaches from the very different machine learning and causal inference traditions of statistics and artificial intelligence.
Artificial intelligence is being widely applied to health data, including a range of supervised machine learning techniques applied to a range of data types including images, text, mobile device data, and electronic health records. This talk will briefly survey leading health applications of AI and then turn its attention from prediction to causal inference. The talk will focus on ways to combine lessons from artificial intelligence approaches and statistical approaches to causal inference, especially for electronic health records.
David Page, PhD and Chair
Department of Biostatistics and Bioinformatics
Duke University School of Medicine
David Page works on algorithms for data mining and machine learning, and their applications to biomedical data, especially de-identified electronic health records and high-throughput genetic and other molecular data. Of particular interest are machine learning methods for complex multi-relational data (such as electronic health records or molecules as shown) and irregular temporal data, and methods that find causal relationships or produce human-interpretable output (such as the rules for molecular bioactivity shown in green to the side)