Data mining techniques utilizing latent class models to evaluate emergency department revisits

Ofir Ben-Assuli, Joshua R. Vest

Research output: Contribution to journalArticle

Abstract

Background: The use of machine learning techniques is especially pertinent to the composite and challenging conditions of emergency departments (EDs). Repeat ED visits (i.e. revisits) are an example of potentially inappropriate utilization of resources that can be forecasted by these techniques. Objective: To track the ED revisit risk over time using the hidden Markov model (HMM) as a major latent class model. Given the HMM states, we carried out forecasting of future ED revisits with various data mining models. Methods: Information integrated from four distributed sources (e.g. electronic health records and health information exchange) was integrated into four HMMs which capture the relationships between an observed and a hidden progression that shift over time through a series of hidden states in an adult patient population. Results: Assimilating a pre-analysis of the various patients by applying latent class models and directing them to well-known classifiers functioned well. The performance was significantly better than without utilizing pre-analysis of HMM for all prediction models (classifiers(. Conclusions: These findings suggest that one prospective approach to advanced risk prediction is to leverage the longitudinal nature of health care data by exploiting patients’ between state variation.

Original languageEnglish (US)
Article number103341
JournalJournal of Biomedical Informatics
Volume101
DOIs
StatePublished - Jan 2020
Externally publishedYes

Fingerprint

Data Mining
Data mining
Hospital Emergency Service
Hidden Markov models
Classifiers
Health
Electronic Health Records
Health care
Learning systems
Delivery of Health Care
Composite materials
Population

Keywords

  • Electronic health records
  • Emergency department revisit
  • Health information exchange
  • Hidden Markov Models
  • Predictive analytics

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

Data mining techniques utilizing latent class models to evaluate emergency department revisits. / Ben-Assuli, Ofir; Vest, Joshua R.

In: Journal of Biomedical Informatics, Vol. 101, 103341, 01.2020.

Research output: Contribution to journalArticle

@article{36e022ccca2e459d9ce70c0aa2da33fe,
title = "Data mining techniques utilizing latent class models to evaluate emergency department revisits",
abstract = "Background: The use of machine learning techniques is especially pertinent to the composite and challenging conditions of emergency departments (EDs). Repeat ED visits (i.e. revisits) are an example of potentially inappropriate utilization of resources that can be forecasted by these techniques. Objective: To track the ED revisit risk over time using the hidden Markov model (HMM) as a major latent class model. Given the HMM states, we carried out forecasting of future ED revisits with various data mining models. Methods: Information integrated from four distributed sources (e.g. electronic health records and health information exchange) was integrated into four HMMs which capture the relationships between an observed and a hidden progression that shift over time through a series of hidden states in an adult patient population. Results: Assimilating a pre-analysis of the various patients by applying latent class models and directing them to well-known classifiers functioned well. The performance was significantly better than without utilizing pre-analysis of HMM for all prediction models (classifiers(. Conclusions: These findings suggest that one prospective approach to advanced risk prediction is to leverage the longitudinal nature of health care data by exploiting patients’ between state variation.",
keywords = "Electronic health records, Emergency department revisit, Health information exchange, Hidden Markov Models, Predictive analytics",
author = "Ofir Ben-Assuli and Vest, {Joshua R.}",
year = "2020",
month = "1",
doi = "10.1016/j.jbi.2019.103341",
language = "English (US)",
volume = "101",
journal = "Journal of Biomedical Informatics",
issn = "1532-0464",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Data mining techniques utilizing latent class models to evaluate emergency department revisits

AU - Ben-Assuli, Ofir

AU - Vest, Joshua R.

PY - 2020/1

Y1 - 2020/1

N2 - Background: The use of machine learning techniques is especially pertinent to the composite and challenging conditions of emergency departments (EDs). Repeat ED visits (i.e. revisits) are an example of potentially inappropriate utilization of resources that can be forecasted by these techniques. Objective: To track the ED revisit risk over time using the hidden Markov model (HMM) as a major latent class model. Given the HMM states, we carried out forecasting of future ED revisits with various data mining models. Methods: Information integrated from four distributed sources (e.g. electronic health records and health information exchange) was integrated into four HMMs which capture the relationships between an observed and a hidden progression that shift over time through a series of hidden states in an adult patient population. Results: Assimilating a pre-analysis of the various patients by applying latent class models and directing them to well-known classifiers functioned well. The performance was significantly better than without utilizing pre-analysis of HMM for all prediction models (classifiers(. Conclusions: These findings suggest that one prospective approach to advanced risk prediction is to leverage the longitudinal nature of health care data by exploiting patients’ between state variation.

AB - Background: The use of machine learning techniques is especially pertinent to the composite and challenging conditions of emergency departments (EDs). Repeat ED visits (i.e. revisits) are an example of potentially inappropriate utilization of resources that can be forecasted by these techniques. Objective: To track the ED revisit risk over time using the hidden Markov model (HMM) as a major latent class model. Given the HMM states, we carried out forecasting of future ED revisits with various data mining models. Methods: Information integrated from four distributed sources (e.g. electronic health records and health information exchange) was integrated into four HMMs which capture the relationships between an observed and a hidden progression that shift over time through a series of hidden states in an adult patient population. Results: Assimilating a pre-analysis of the various patients by applying latent class models and directing them to well-known classifiers functioned well. The performance was significantly better than without utilizing pre-analysis of HMM for all prediction models (classifiers(. Conclusions: These findings suggest that one prospective approach to advanced risk prediction is to leverage the longitudinal nature of health care data by exploiting patients’ between state variation.

KW - Electronic health records

KW - Emergency department revisit

KW - Health information exchange

KW - Hidden Markov Models

KW - Predictive analytics

UR - http://www.scopus.com/inward/record.url?scp=85075288532&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075288532&partnerID=8YFLogxK

U2 - 10.1016/j.jbi.2019.103341

DO - 10.1016/j.jbi.2019.103341

M3 - Article

C2 - 31747623

AN - SCOPUS:85075288532

VL - 101

JO - Journal of Biomedical Informatics

JF - Journal of Biomedical Informatics

SN - 1532-0464

M1 - 103341

ER -