Research Project

Bayesian Joint Model for High Dimensional Health Data

Developing novel joint models to effectively and efficiently leverage complex, integrated real-world data for science

Principal Investigator
Basu, Sanjib
Co-Principal Investigator
Sun, Jiehuan
Research Area(s)
Health Data & Informatics
Funding Source
NSF DMS-2413549

Abstract

The overall objective of this project is to develop novel joint models for the integrative analysis of complex real-world data from multiple domains. The specific aims are: (1) To develop a joint model for high-dimensional longitudinal processes and a time-to-event outcome; (2) To develop a joint model for integrative analysis of high-dimensional longitudinal processes, neighborhood-level SDOH indicators, and a time-to-event outcome; (3) To develop a soft Bayesian additive regression tree-based high-dimensional joint model for integrative analysis of high-dimensional longitudinal processes, neighborhood-level SDOH indicators, and a time-to-event outcome; (4) To develop publicly available user-friendly tools with detailed documentation for our proposed methods in Aims 1, 2 and 3.