Stabilizing High-Dimensional Prediction Models Using Feature Graphs

Author

Gopakumar, Shivapratap ; Truyen Tran ; Tu Dinh Nguyen ; Dinh Phung ; Venkatesh, Svetha

Author_Institution

Centre for Pattern Recognition & Data Analytics, Deakin Univ., Geelong, VIC, Australia

Volume

19

Issue

3

fYear

2015

fDate

5/1/2015 12:00:00 AM

Firstpage

1044

Lastpage

1052

Abstract

We investigate feature stability in the context of clinical prognosis derived from high-dimensional electronic medical records. To reduce variance in the selected features that are predictive, we introduce Laplacian-based regularization into a regression model. The Laplacian is derived on a feature graph that captures both the temporal and hierarchic relations between hospital events, diseases, and interventions. Using a cohort of patients with heart failure, we demonstrate better feature stability and goodness-of-fit through feature graph stabilization.

Keywords

Laplace equations; cardiology; diseases; electronic health records; feature selection; graphs; medical diagnostic computing; regression analysis; Laplacian-based regularization; clinical prognosis; diseases; feature graph stabilization; goodness-of-fit; heart failure; hierarchic relations; high-dimensional electronic medical records; hospital events; interventions; regression model; selected features; stabilizing high-dimensional prediction models; temporal relations; Data models; Feature extraction; Heart; Indexes; Predictive models; Stability criteria; Biomedical computing; electronic medical records; predictive models; stability;

fLanguage

English

Journal_Title

Biomedical and Health Informatics, IEEE Journal of

Publisher

ieee

ISSN

2168-2194

Type

jour

DOI

10.1109/JBHI.2014.2353031

Filename

6887285