Title: Predictive Modeling with Complex Healthcare Data

Abstract: The recent years have witnessed a surge of interests on healthcare data analytics. Predictive modeling, which aims at building computational models to predict the likelihood of specific outcomes (e.g., mortality or onset of specific condition), is one fundamental and challenging problem. In this talk I will introduce the research that has been done by my group in recent years on predictive modeling with complex healthcare data. Here the complexities can come from either input features or predicted outcomes. Specifically, I will talk about tensor based modeling for input complexity and heterogeneous target regression for outcome complexity. I will also talk about other potential issues or challenges for healthcare data modeling and discuss about future directions.

Bio: Fei Wang is an Associate Professor in Department of Healthcare Policy and Research at Weill Cornell Medicine. His research interest lies in the intersection between data mining and computational medicine. He has extensive publications on top CS venues such as NIPS, ICML, KDD, AAAI and biomedical venues such as JAMA Internal Medicine. His h-index is 50. Dr. Wang is the winner of 2016 Parkinson's Progression Markers Initiative data challenge organized by Michael J. Fox Foundation and 2017 NIPS/Kaggle challenge on Personalized Medicine organized by Memorial Sloan Kettering Cancer Center. Dr. Wang is the recipient of NSF CAREER award in 2018, and his research has been supported by NSF, NIH, ONR, PCORI, AHA, MJFF and industries such as Amazon.