Stochastic & Multiscale Modeling and Computation Seminar by Jiehuan Sun: Learning Healthcare Delivery Network with Longitudinal Electronic Health Records Data

Title:

Learning Healthcare Delivery Network with Longitudinal Electronic Health Records Data.

Speaker: 

Jiehuan Sun, Assistant professor, University of Illinois Chicago.

Abstract:

Knowledge networks such as the healthcare delivery network (HDN), describing relationships among different medical encounters, are useful summaries of the state-of-art medical knowledge. The increasing availability of longitudinal electronic health records (EHR) data promises a rich data source for learning HDN. Most existing methods for inferring about knowledge network are based on co-occurrence patterns that do not account for temporal effects or patient level heterogeneity. Building upon multivariate Hawkes process (mvHP), we propose a flexible covariate adjusted random effects (CARE) mvHP modeling strategy for HDN construction. In this talk, I will discuss our proposed model and its application to an EHR study of type 2 diabetes patients.

 

Stochastic & Multiscale Modeling and Computation Seminar