Jie Xu, Ph.D.

My research interests are focused on the intersection of machine learning and health informatics, with a particular emphasis on disease progression subtyping, predictive modeling, and federated learning techniques. Prior to joining UF, I conducted postdoctoral research at Weill Cornell Medicine, working closely with Dr. Fei Wang and Dr. Jyotishman Pathak. I earned my Ph.D. in Electrical Engineering from Xidian University in 2018, during which I participated in a joint doctoral program at the University of Pittsburgh and the University of Texas at Arlington from 2016 to 2018. I am enthusiastic about continuing my research at UF and contributing to healthcare advancement through innovative machine learning approaches.

I have been working on developing novel computational algorithms, including advanced AI methods for analyzing various kinds of healthcare data, including Electronic Health Records (EHRs), medical and pharmacy claims data, and medical imaging data. One research direction that I am pursuing in fundamental AI methods development is patient similarity evaluations, i.e., assessing the clinical similarity between patients based on their medical histories. Such AI-driven approaches can be applied to advance disease surveillance, disease sub-phenotyping, and comparative effectiveness research. I also have strong expertise in predictive modeling, including developing models for various risk predictions, such as mortality, disease onset, and disease state change. Furthermore, I am actively investigating federated learning (FL) techniques, which allow for training an algorithm across multiple decentralized institutions without sharing their data samples. My research has yielded over 30 peer-reviewed publications in both machine learning and health informatics venues, with papers appearing at top-tier artificial intelligence and machine learning conferences such as NeurIPS, AAAI, KDD, IJCAI, at health informatics conferences such as AMIA, as well as at journals such as JHIR, JMIR.