Neural networks, bayesian networks, and logistic regression for ESI prediction

Seifu Chonde, David Nembhard, Gül E. Okudan Kremer, Omar Ashour

Research output: Contribution to conferencePaper

Abstract

Emergency Department (ED) triage is a process of determining illness severity and accordingly assigning patient priority. The Emergency Severity Index (ESI) is a 5-level acuity categorization system that aides in triage by considering patient complaints and resource needs. The objective of this paper is to compare the capabilities of predicting ESI level using ordinal logistic regression, neural networks and naïve Bayesian networks in predicting ESI levels 2 through 5. Data was obtained from Susquehanna Williamsport Hospital for 947 patients over a one month period in 2008. It contained the assigned ESI level, chief complaint, systolic blood pressure, pulse, respiration rate, temperature, oxygen saturation (SaO2) level, age, gender, and pain. A multivariate ordinal logistic regression model was fit using significant univariate logistic regression predictors. Naïve Bayesian Network (NBNs) and artificial Neural Networks (NNs), two machine learning techniques, are proposed to relax the inherent assumptions of linearity and independence in logistic regression. These three techniques are compared using incremental training dataset sizes between 50% and 100% of given data. Considerations of speed, accuracy, data utilization, model flexibility, and interpretability suggest NBNs and NNs for further research.

Original languageEnglish (US)
Pages1478-1487
Number of pages10
StatePublished - Jan 1 2013
EventIIE Annual Conference and Expo 2013 - San Juan, Puerto Rico
Duration: May 18 2013May 22 2013

Other

OtherIIE Annual Conference and Expo 2013
CountryPuerto Rico
CitySan Juan
Period5/18/135/22/13

Fingerprint

Bayesian networks
Logistics
Neural networks
Blood pressure
Learning systems
Oxygen
Temperature

All Science Journal Classification (ASJC) codes

  • Industrial and Manufacturing Engineering

Cite this

Chonde, S., Nembhard, D., Okudan Kremer, G. E., & Ashour, O. (2013). Neural networks, bayesian networks, and logistic regression for ESI prediction. 1478-1487. Paper presented at IIE Annual Conference and Expo 2013, San Juan, Puerto Rico.
Chonde, Seifu ; Nembhard, David ; Okudan Kremer, Gül E. ; Ashour, Omar. / Neural networks, bayesian networks, and logistic regression for ESI prediction. Paper presented at IIE Annual Conference and Expo 2013, San Juan, Puerto Rico.10 p.
@conference{7bf547b8eb47485787c1c07e721cfbbe,
title = "Neural networks, bayesian networks, and logistic regression for ESI prediction",
abstract = "Emergency Department (ED) triage is a process of determining illness severity and accordingly assigning patient priority. The Emergency Severity Index (ESI) is a 5-level acuity categorization system that aides in triage by considering patient complaints and resource needs. The objective of this paper is to compare the capabilities of predicting ESI level using ordinal logistic regression, neural networks and na{\"i}ve Bayesian networks in predicting ESI levels 2 through 5. Data was obtained from Susquehanna Williamsport Hospital for 947 patients over a one month period in 2008. It contained the assigned ESI level, chief complaint, systolic blood pressure, pulse, respiration rate, temperature, oxygen saturation (SaO2) level, age, gender, and pain. A multivariate ordinal logistic regression model was fit using significant univariate logistic regression predictors. Na{\"i}ve Bayesian Network (NBNs) and artificial Neural Networks (NNs), two machine learning techniques, are proposed to relax the inherent assumptions of linearity and independence in logistic regression. These three techniques are compared using incremental training dataset sizes between 50{\%} and 100{\%} of given data. Considerations of speed, accuracy, data utilization, model flexibility, and interpretability suggest NBNs and NNs for further research.",
author = "Seifu Chonde and David Nembhard and {Okudan Kremer}, {G{\"u}l E.} and Omar Ashour",
year = "2013",
month = "1",
day = "1",
language = "English (US)",
pages = "1478--1487",
note = "IIE Annual Conference and Expo 2013 ; Conference date: 18-05-2013 Through 22-05-2013",

}

Chonde, S, Nembhard, D, Okudan Kremer, GE & Ashour, O 2013, 'Neural networks, bayesian networks, and logistic regression for ESI prediction', Paper presented at IIE Annual Conference and Expo 2013, San Juan, Puerto Rico, 5/18/13 - 5/22/13 pp. 1478-1487.

Neural networks, bayesian networks, and logistic regression for ESI prediction. / Chonde, Seifu; Nembhard, David; Okudan Kremer, Gül E.; Ashour, Omar.

2013. 1478-1487 Paper presented at IIE Annual Conference and Expo 2013, San Juan, Puerto Rico.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Neural networks, bayesian networks, and logistic regression for ESI prediction

AU - Chonde, Seifu

AU - Nembhard, David

AU - Okudan Kremer, Gül E.

AU - Ashour, Omar

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Emergency Department (ED) triage is a process of determining illness severity and accordingly assigning patient priority. The Emergency Severity Index (ESI) is a 5-level acuity categorization system that aides in triage by considering patient complaints and resource needs. The objective of this paper is to compare the capabilities of predicting ESI level using ordinal logistic regression, neural networks and naïve Bayesian networks in predicting ESI levels 2 through 5. Data was obtained from Susquehanna Williamsport Hospital for 947 patients over a one month period in 2008. It contained the assigned ESI level, chief complaint, systolic blood pressure, pulse, respiration rate, temperature, oxygen saturation (SaO2) level, age, gender, and pain. A multivariate ordinal logistic regression model was fit using significant univariate logistic regression predictors. Naïve Bayesian Network (NBNs) and artificial Neural Networks (NNs), two machine learning techniques, are proposed to relax the inherent assumptions of linearity and independence in logistic regression. These three techniques are compared using incremental training dataset sizes between 50% and 100% of given data. Considerations of speed, accuracy, data utilization, model flexibility, and interpretability suggest NBNs and NNs for further research.

AB - Emergency Department (ED) triage is a process of determining illness severity and accordingly assigning patient priority. The Emergency Severity Index (ESI) is a 5-level acuity categorization system that aides in triage by considering patient complaints and resource needs. The objective of this paper is to compare the capabilities of predicting ESI level using ordinal logistic regression, neural networks and naïve Bayesian networks in predicting ESI levels 2 through 5. Data was obtained from Susquehanna Williamsport Hospital for 947 patients over a one month period in 2008. It contained the assigned ESI level, chief complaint, systolic blood pressure, pulse, respiration rate, temperature, oxygen saturation (SaO2) level, age, gender, and pain. A multivariate ordinal logistic regression model was fit using significant univariate logistic regression predictors. Naïve Bayesian Network (NBNs) and artificial Neural Networks (NNs), two machine learning techniques, are proposed to relax the inherent assumptions of linearity and independence in logistic regression. These three techniques are compared using incremental training dataset sizes between 50% and 100% of given data. Considerations of speed, accuracy, data utilization, model flexibility, and interpretability suggest NBNs and NNs for further research.

UR - http://www.scopus.com/inward/record.url?scp=84900298254&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900298254&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84900298254

SP - 1478

EP - 1487

ER -

Chonde S, Nembhard D, Okudan Kremer GE, Ashour O. Neural networks, bayesian networks, and logistic regression for ESI prediction. 2013. Paper presented at IIE Annual Conference and Expo 2013, San Juan, Puerto Rico.