Time-to-event predictive modeling for chronic conditions using electronic health records

Yu Kai Lin, Hsinchun Chen, Randall A. Brown, Shu Hsing Li, Hung Jen Yang

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Although electronic health records (EHRs) hold promise for supporting clinical decision making, few studies have used them to model the progression of chronic conditions. To examine the feasibility of EHR-based predictive models for chronic conditions and to mitigate the associated data challenges, the authors develop a time-to-event predictive modeling framework consisting of five analytical steps: guideline-based feature selection, temporal regularization, data abstraction, multiple imputation, and extended Cox models. Using concept- and temporal-abstracted features, the proposed model attained significantly improved performance over the model using only base features.

Original languageEnglish (US)
Article number6813395
Pages (from-to)14-20
Number of pages7
JournalIEEE Intelligent Systems
Volume29
Issue number3
DOIs
StatePublished - 2014

Fingerprint

Health
Feature extraction
Decision making

Keywords

  • chronic conditions
  • EHR
  • electronic health records
  • intelligent systems
  • prognostic modeling
  • time-to-event predictive modeling

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications

Cite this

Time-to-event predictive modeling for chronic conditions using electronic health records. / Lin, Yu Kai; Chen, Hsinchun; Brown, Randall A.; Li, Shu Hsing; Yang, Hung Jen.

In: IEEE Intelligent Systems, Vol. 29, No. 3, 6813395, 2014, p. 14-20.

Research output: Contribution to journalArticle

Lin, Yu Kai ; Chen, Hsinchun ; Brown, Randall A. ; Li, Shu Hsing ; Yang, Hung Jen. / Time-to-event predictive modeling for chronic conditions using electronic health records. In: IEEE Intelligent Systems. 2014 ; Vol. 29, No. 3. pp. 14-20.
@article{69fb0252b36b492eabb5b85c219b6dff,
title = "Time-to-event predictive modeling for chronic conditions using electronic health records",
abstract = "Although electronic health records (EHRs) hold promise for supporting clinical decision making, few studies have used them to model the progression of chronic conditions. To examine the feasibility of EHR-based predictive models for chronic conditions and to mitigate the associated data challenges, the authors develop a time-to-event predictive modeling framework consisting of five analytical steps: guideline-based feature selection, temporal regularization, data abstraction, multiple imputation, and extended Cox models. Using concept- and temporal-abstracted features, the proposed model attained significantly improved performance over the model using only base features.",
keywords = "chronic conditions, EHR, electronic health records, intelligent systems, prognostic modeling, time-to-event predictive modeling",
author = "Lin, {Yu Kai} and Hsinchun Chen and Brown, {Randall A.} and Li, {Shu Hsing} and Yang, {Hung Jen}",
year = "2014",
doi = "10.1109/MIS.2014.18",
language = "English (US)",
volume = "29",
pages = "14--20",
journal = "IEEE Intelligent Systems",
issn = "1541-1672",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Time-to-event predictive modeling for chronic conditions using electronic health records

AU - Lin, Yu Kai

AU - Chen, Hsinchun

AU - Brown, Randall A.

AU - Li, Shu Hsing

AU - Yang, Hung Jen

PY - 2014

Y1 - 2014

N2 - Although electronic health records (EHRs) hold promise for supporting clinical decision making, few studies have used them to model the progression of chronic conditions. To examine the feasibility of EHR-based predictive models for chronic conditions and to mitigate the associated data challenges, the authors develop a time-to-event predictive modeling framework consisting of five analytical steps: guideline-based feature selection, temporal regularization, data abstraction, multiple imputation, and extended Cox models. Using concept- and temporal-abstracted features, the proposed model attained significantly improved performance over the model using only base features.

AB - Although electronic health records (EHRs) hold promise for supporting clinical decision making, few studies have used them to model the progression of chronic conditions. To examine the feasibility of EHR-based predictive models for chronic conditions and to mitigate the associated data challenges, the authors develop a time-to-event predictive modeling framework consisting of five analytical steps: guideline-based feature selection, temporal regularization, data abstraction, multiple imputation, and extended Cox models. Using concept- and temporal-abstracted features, the proposed model attained significantly improved performance over the model using only base features.

KW - chronic conditions

KW - EHR

KW - electronic health records

KW - intelligent systems

KW - prognostic modeling

KW - time-to-event predictive modeling

UR - http://www.scopus.com/inward/record.url?scp=84905282118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905282118&partnerID=8YFLogxK

U2 - 10.1109/MIS.2014.18

DO - 10.1109/MIS.2014.18

M3 - Article

AN - SCOPUS:84905282118

VL - 29

SP - 14

EP - 20

JO - IEEE Intelligent Systems

JF - IEEE Intelligent Systems

SN - 1541-1672

IS - 3

M1 - 6813395

ER -