Time-to-event modeling for hospital length of stay prediction for COVID-19 patients

Yuxin Wen; Md Fashiar Rahman; Yan Zhuang; Michael Pokojovy; Honglun Xu; Peter McCaffrey; Alexander Vo; Eric Walser; Scott Moen; Tzu Liang (Bill) Tseng

doi:10.1016/j.mlwa.2022.100365

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients

Yuxin Wen
, Md Fashiar Rahman
, Yan Zhuang
, Michael Pokojovy
, Honglun Xu
, Peter McCaffrey
, Alexander Vo
, Eric Walser
, Scott Moen
, Tzu Liang (Bill) Tseng

Research output: Contribution to journal › Article › peer-review

17 Scopus citations

Abstract

Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

Original language	English (US)
Article number	100365
Journal	Machine Learning with Applications
Volume	9
DOIs	https://doi.org/10.1016/j.mlwa.2022.100365
State	Published - Sep 15 2022

Keywords

COVID-19
Deep learning
Length of stay
Survival analysis
Time-to-event modeling

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Computational Theory and Mathematics
Information Systems

Access to Document

10.1016/j.mlwa.2022.100365

Cite this

@article{f5c88ee083fb4d4a81d0717a890c289f,

title = "Time-to-event modeling for hospital length of stay prediction for COVID-19 patients",

abstract = "Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.",

keywords = "COVID-19, Deep learning, Length of stay, Survival analysis, Time-to-event modeling",

author = "Yuxin Wen and Rahman, \{Md Fashiar\} and Yan Zhuang and Michael Pokojovy and Honglun Xu and Peter McCaffrey and Alexander Vo and Eric Walser and Scott Moen and Tseng, \{Tzu Liang (Bill)\}",

note = "Publisher Copyright: {\textcopyright} 2022",

year = "2022",

month = sep,

day = "15",

doi = "10.1016/j.mlwa.2022.100365",

language = "English (US)",

volume = "9",

journal = "Machine Learning with Applications",

issn = "2666-8270",

}

TY - JOUR

T1 - Time-to-event modeling for hospital length of stay prediction for COVID-19 patients

AU - Wen, Yuxin

AU - Rahman, Md Fashiar

AU - Zhuang, Yan

AU - Pokojovy, Michael

AU - Xu, Honglun

AU - McCaffrey, Peter

AU - Vo, Alexander

AU - Walser, Eric

AU - Moen, Scott

AU - Tseng, Tzu Liang (Bill)

PY - 2022/9/15

Y1 - 2022/9/15

N2 - Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

AB - Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

KW - COVID-19

KW - Deep learning

KW - Length of stay

KW - Survival analysis

KW - Time-to-event modeling

UR - https://www.scopus.com/pages/publications/85164036850

UR - https://www.scopus.com/pages/publications/85164036850#tab=citedBy

U2 - 10.1016/j.mlwa.2022.100365

DO - 10.1016/j.mlwa.2022.100365

M3 - Article

AN - SCOPUS:85164036850

SN - 2666-8270

VL - 9

JO - Machine Learning with Applications

JF - Machine Learning with Applications

M1 - 100365

ER -

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this