Survival prediction in the intensive care unit: A comparison of neural networks and binary logit regression

Ernest Goss, Harish Ramchandani

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

As a component of hospital cost containment policy, the National Institutes for Health recommends that hospitals limit intensive care unit (ICU) resources to patients who have a reasonable probability of recovery. In order to provide more ethical and objective measures of the likelihood of ICU recovery, hospitals have turned increasingly to decision support software such as APACHE (acute physiology and chronic health evaluation), which derives predictions from binary logit regression (BLR), a parametric estimation technique. Rapid advancements in computer software and hardware technology have encouraged researchers to use more computationally intensive non-parametric techniques such as neural networks (NNs) whose prediction capabilities are purported to be greater than those of parametric models. The present study applies this methodology to a sample of ICU patients and shows that neural networks appear to more correctly predict survival than does BLR.

Original languageEnglish
Pages (from-to)189-198
Number of pages10
JournalSocio-Economic Planning Sciences
Volume32
Issue number3
StatePublished - 1998

Fingerprint

Logit
neural network
Regression
Neural Networks
Binary
regression
Unit
Prediction
Health
Recovery
prediction
containment policy
software
Parametric Estimation
cost containment
estimation procedure
Physiology
physiology
containment
Parametric Model

All Science Journal Classification (ASJC) codes

  • Strategy and Management
  • Management Science and Operations Research
  • Statistics, Probability and Uncertainty
  • Economics and Econometrics

Cite this

Survival prediction in the intensive care unit : A comparison of neural networks and binary logit regression. / Goss, Ernest; Ramchandani, Harish.

In: Socio-Economic Planning Sciences, Vol. 32, No. 3, 1998, p. 189-198.

Research output: Contribution to journalArticle

@article{a3e58fed76ee4035afb3b248403cf9b8,
title = "Survival prediction in the intensive care unit: A comparison of neural networks and binary logit regression",
abstract = "As a component of hospital cost containment policy, the National Institutes for Health recommends that hospitals limit intensive care unit (ICU) resources to patients who have a reasonable probability of recovery. In order to provide more ethical and objective measures of the likelihood of ICU recovery, hospitals have turned increasingly to decision support software such as APACHE (acute physiology and chronic health evaluation), which derives predictions from binary logit regression (BLR), a parametric estimation technique. Rapid advancements in computer software and hardware technology have encouraged researchers to use more computationally intensive non-parametric techniques such as neural networks (NNs) whose prediction capabilities are purported to be greater than those of parametric models. The present study applies this methodology to a sample of ICU patients and shows that neural networks appear to more correctly predict survival than does BLR.",
author = "Ernest Goss and Harish Ramchandani",
year = "1998",
language = "English",
volume = "32",
pages = "189--198",
journal = "Socio-Economic Planning Sciences",
issn = "0038-0121",
publisher = "Elsevier Limited",
number = "3",

}

TY - JOUR

T1 - Survival prediction in the intensive care unit

T2 - A comparison of neural networks and binary logit regression

AU - Goss, Ernest

AU - Ramchandani, Harish

PY - 1998

Y1 - 1998

N2 - As a component of hospital cost containment policy, the National Institutes for Health recommends that hospitals limit intensive care unit (ICU) resources to patients who have a reasonable probability of recovery. In order to provide more ethical and objective measures of the likelihood of ICU recovery, hospitals have turned increasingly to decision support software such as APACHE (acute physiology and chronic health evaluation), which derives predictions from binary logit regression (BLR), a parametric estimation technique. Rapid advancements in computer software and hardware technology have encouraged researchers to use more computationally intensive non-parametric techniques such as neural networks (NNs) whose prediction capabilities are purported to be greater than those of parametric models. The present study applies this methodology to a sample of ICU patients and shows that neural networks appear to more correctly predict survival than does BLR.

AB - As a component of hospital cost containment policy, the National Institutes for Health recommends that hospitals limit intensive care unit (ICU) resources to patients who have a reasonable probability of recovery. In order to provide more ethical and objective measures of the likelihood of ICU recovery, hospitals have turned increasingly to decision support software such as APACHE (acute physiology and chronic health evaluation), which derives predictions from binary logit regression (BLR), a parametric estimation technique. Rapid advancements in computer software and hardware technology have encouraged researchers to use more computationally intensive non-parametric techniques such as neural networks (NNs) whose prediction capabilities are purported to be greater than those of parametric models. The present study applies this methodology to a sample of ICU patients and shows that neural networks appear to more correctly predict survival than does BLR.

UR - http://www.scopus.com/inward/record.url?scp=0000170514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000170514&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0000170514

VL - 32

SP - 189

EP - 198

JO - Socio-Economic Planning Sciences

JF - Socio-Economic Planning Sciences

SN - 0038-0121

IS - 3

ER -