Abstract
Gaussian processes (GPs) are a versatile nonparametric model for nonlinear regression and have been widely used to study spatiotemporal phenomena. However, standard GPs offer limited interpretability and generalizability for datasets with naturally occurring hierarchies. With large-scale, rapidly-updating electronic health record (EHR) data, we want to study patient trajectories across diverse patient cohorts while preserving patient subgroup structure. In this work, we partition our cohort of over 2000 COVID-19 patients by sex and ethnicity. We develop and apply a hierarchical Gaussian process and a mixture of experts (MOE) hierarchical GP model to fit patient trajectories on clinical markers of disease progression. A case study for albumin, an effective predictor of COVID-19 patient outcomes, highlights the predictive performance of these models.
Original language | English (US) |
---|---|
Title of host publication | Pacific Symposium on Biocomputing 2022, PSB 2022 |
Editors | Russ B. Altman, A. Keith Dunker, Lawrence Hunter, Marylyn D. Ritchie, Teri E. Klein |
Publisher | World Scientific |
Pages | 266-277 |
Number of pages | 12 |
ISBN (Electronic) | 9789811250460 |
DOIs | |
State | Published - 2022 |
Event | 27th Pacific Symposium on Biocomputing, PSB 2022 - Kohala Coast, United States Duration: Jan 3 2022 → Jan 7 2022 |
Conference
Conference | 27th Pacific Symposium on Biocomputing, PSB 2022 |
---|---|
Country/Territory | United States |
City | Kohala Coast |
Period | 1/3/22 → 1/7/22 |
All Science Journal Classification (ASJC) codes
- Biomedical Engineering
- Computational Theory and Mathematics
Keywords
- COVID-19
- electronic health record
- Gaussian processes
- patient trajectories