Description:

Speaker:  Jiehuan Sun, PhD, University of Illinois Chicago (IUC)

Title:  Some Statistical Methods for Analyzing Longitudinal Electronic Health Records Data

The increasing adoption of electronic health records (EHR) by healthcare institutions has provided a rich data source for developing tools to aid disease management and clinical decision support. Many of the data captured in the EHR may be viewed as point process data, which traditional statistical methods cannot handle. Novel statistical methods are in great demand for analyzing EHR data.

In this talk, I will discuss two of our recently developed methods: one for studying the association between a point process predictor and a scalar response through a generalized functional linear model framework and the other for inferring healthcare delivery networks based on the multivariate Hawkes process. Their applications to EHR datasets will also be discussed.

Bio:  Jiehuan Sun received his B.E. in Bioinformatics from Huazhong University of Science and Technology in 2010 and his Sc.M. in Biostatistics from Johns Hopkins University in 2012. He then completed his Ph.D. in Biostatistics at Yale University in 2017. After graduation, he did two-year postdoctoral training at the Department of Biostatistics at Harvard University. Since 2019, he has been an Assistant Professor of Biostatistics at the University of Illinois Chicago (UIC). He is also a member of the Biostatistics Shared Resource Core at the University of Illinois Cancer Center. His primary research interests are to develop statistical methods to deal with high-dimensional (big) biomedical data, including electronic health records data and genomics data. Specifically, his research areas include biomarker discovery, disease subtype identification, dynamic risk prediction, and network modeling.