Use of genetic algorithms for neural networks to predict community-acquired pneumonia

Paul S. Heckerling, Ben S. Gerber, Thomas Gerald Tape, Robert Swift Wigton

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

Background: Genetic algorithms have been used to solve optimization problems for artificial neural networks (ANN) in several domains. We used genetic algorithms to search for optimal hidden-layer architectures, connectivity, and training parameters for ANN for predicting community-acquired pneumonia among patients with respiratory complaints. Methods: Feed-forward back-propagation ANN were trained on sociodemographic, symptom, sign, comorbidity, and radiographic outcome data among 1044 patients from the University of Illinois (the training cohort), and were applied to 116 patients from the University of Nebraska (the testing cohort). Binary chromosomes with genes representing network attributes, including the number of nodes in the hidden layers, learning rate and momentum parameters, and the presence or absence of implicit within-layer connectivity using a competition algorithm, were operated on by various combinations of crossover, mutation, and probabilistic selection based on network mean-square error (MSE), and separately on average cross entropy (ENT). Predictive accuracy was measured as the area under a receiver-operating characteristic (ROC) curve. Results: Over 50 generations, the baseline genetic algorithm evolved an optimized ANN with nine nodes in the first hidden layer, zero nodes in the second hidden layer, learning rate and momentum parameters of 0.5, and no within-layer competition connectivity. This ANN had an ROC area in the training cohort of 0.872 and in the testing cohort of 0.934 (P-value for difference, 0.181). Algorithms based on cross-generational selection, Gray coding of genes prior to mutation, and crossover recombination at different genetic levels, evolved optimized ANN identical to the baseline genetic strategy. Algorithms based on other strategies, including elite selection within generations (training ROC area 0.819), and inversions of genetic material during recombination (training ROC area 0.812), evolved less accurate ANN. Conclusion: ANN optimized by genetic algorithms accurately discriminated pneumonia within a training cohort, and within a testing cohort consisting of cases on which the networks had not been trained. Genetic algorithms can be used to implement efficient search strategies for optimal ANN to predict pneumonia.

Original languageEnglish (US)
Pages (from-to)71-84
Number of pages14
JournalArtificial Intelligence in Medicine
Volume30
Issue number1
DOIs
StatePublished - Jan 1 2004

Fingerprint

Pneumonia
Genetic algorithms
Neural networks
ROC Curve
Genetic Recombination
Momentum
Learning
Testing
Genes
Mutation
Gene Regulatory Networks
Entropy
Chromosomes
Signs and Symptoms
Backpropagation
Comorbidity
Mean square error

Keywords

  • Artificial neural network
  • Genetic algorithm
  • Pneumonia

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Artificial Intelligence

Cite this

Use of genetic algorithms for neural networks to predict community-acquired pneumonia. / Heckerling, Paul S.; Gerber, Ben S.; Tape, Thomas Gerald; Wigton, Robert Swift.

In: Artificial Intelligence in Medicine, Vol. 30, No. 1, 01.01.2004, p. 71-84.

Research output: Contribution to journalArticle

@article{8be516092f5144f7bee511ae3783dd12,
title = "Use of genetic algorithms for neural networks to predict community-acquired pneumonia",
abstract = "Background: Genetic algorithms have been used to solve optimization problems for artificial neural networks (ANN) in several domains. We used genetic algorithms to search for optimal hidden-layer architectures, connectivity, and training parameters for ANN for predicting community-acquired pneumonia among patients with respiratory complaints. Methods: Feed-forward back-propagation ANN were trained on sociodemographic, symptom, sign, comorbidity, and radiographic outcome data among 1044 patients from the University of Illinois (the training cohort), and were applied to 116 patients from the University of Nebraska (the testing cohort). Binary chromosomes with genes representing network attributes, including the number of nodes in the hidden layers, learning rate and momentum parameters, and the presence or absence of implicit within-layer connectivity using a competition algorithm, were operated on by various combinations of crossover, mutation, and probabilistic selection based on network mean-square error (MSE), and separately on average cross entropy (ENT). Predictive accuracy was measured as the area under a receiver-operating characteristic (ROC) curve. Results: Over 50 generations, the baseline genetic algorithm evolved an optimized ANN with nine nodes in the first hidden layer, zero nodes in the second hidden layer, learning rate and momentum parameters of 0.5, and no within-layer competition connectivity. This ANN had an ROC area in the training cohort of 0.872 and in the testing cohort of 0.934 (P-value for difference, 0.181). Algorithms based on cross-generational selection, Gray coding of genes prior to mutation, and crossover recombination at different genetic levels, evolved optimized ANN identical to the baseline genetic strategy. Algorithms based on other strategies, including elite selection within generations (training ROC area 0.819), and inversions of genetic material during recombination (training ROC area 0.812), evolved less accurate ANN. Conclusion: ANN optimized by genetic algorithms accurately discriminated pneumonia within a training cohort, and within a testing cohort consisting of cases on which the networks had not been trained. Genetic algorithms can be used to implement efficient search strategies for optimal ANN to predict pneumonia.",
keywords = "Artificial neural network, Genetic algorithm, Pneumonia",
author = "Heckerling, {Paul S.} and Gerber, {Ben S.} and Tape, {Thomas Gerald} and Wigton, {Robert Swift}",
year = "2004",
month = "1",
day = "1",
doi = "10.1016/S0933-3657(03)00065-4",
language = "English (US)",
volume = "30",
pages = "71--84",
journal = "Artificial Intelligence in Medicine",
issn = "0933-3657",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Use of genetic algorithms for neural networks to predict community-acquired pneumonia

AU - Heckerling, Paul S.

AU - Gerber, Ben S.

AU - Tape, Thomas Gerald

AU - Wigton, Robert Swift

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Background: Genetic algorithms have been used to solve optimization problems for artificial neural networks (ANN) in several domains. We used genetic algorithms to search for optimal hidden-layer architectures, connectivity, and training parameters for ANN for predicting community-acquired pneumonia among patients with respiratory complaints. Methods: Feed-forward back-propagation ANN were trained on sociodemographic, symptom, sign, comorbidity, and radiographic outcome data among 1044 patients from the University of Illinois (the training cohort), and were applied to 116 patients from the University of Nebraska (the testing cohort). Binary chromosomes with genes representing network attributes, including the number of nodes in the hidden layers, learning rate and momentum parameters, and the presence or absence of implicit within-layer connectivity using a competition algorithm, were operated on by various combinations of crossover, mutation, and probabilistic selection based on network mean-square error (MSE), and separately on average cross entropy (ENT). Predictive accuracy was measured as the area under a receiver-operating characteristic (ROC) curve. Results: Over 50 generations, the baseline genetic algorithm evolved an optimized ANN with nine nodes in the first hidden layer, zero nodes in the second hidden layer, learning rate and momentum parameters of 0.5, and no within-layer competition connectivity. This ANN had an ROC area in the training cohort of 0.872 and in the testing cohort of 0.934 (P-value for difference, 0.181). Algorithms based on cross-generational selection, Gray coding of genes prior to mutation, and crossover recombination at different genetic levels, evolved optimized ANN identical to the baseline genetic strategy. Algorithms based on other strategies, including elite selection within generations (training ROC area 0.819), and inversions of genetic material during recombination (training ROC area 0.812), evolved less accurate ANN. Conclusion: ANN optimized by genetic algorithms accurately discriminated pneumonia within a training cohort, and within a testing cohort consisting of cases on which the networks had not been trained. Genetic algorithms can be used to implement efficient search strategies for optimal ANN to predict pneumonia.

AB - Background: Genetic algorithms have been used to solve optimization problems for artificial neural networks (ANN) in several domains. We used genetic algorithms to search for optimal hidden-layer architectures, connectivity, and training parameters for ANN for predicting community-acquired pneumonia among patients with respiratory complaints. Methods: Feed-forward back-propagation ANN were trained on sociodemographic, symptom, sign, comorbidity, and radiographic outcome data among 1044 patients from the University of Illinois (the training cohort), and were applied to 116 patients from the University of Nebraska (the testing cohort). Binary chromosomes with genes representing network attributes, including the number of nodes in the hidden layers, learning rate and momentum parameters, and the presence or absence of implicit within-layer connectivity using a competition algorithm, were operated on by various combinations of crossover, mutation, and probabilistic selection based on network mean-square error (MSE), and separately on average cross entropy (ENT). Predictive accuracy was measured as the area under a receiver-operating characteristic (ROC) curve. Results: Over 50 generations, the baseline genetic algorithm evolved an optimized ANN with nine nodes in the first hidden layer, zero nodes in the second hidden layer, learning rate and momentum parameters of 0.5, and no within-layer competition connectivity. This ANN had an ROC area in the training cohort of 0.872 and in the testing cohort of 0.934 (P-value for difference, 0.181). Algorithms based on cross-generational selection, Gray coding of genes prior to mutation, and crossover recombination at different genetic levels, evolved optimized ANN identical to the baseline genetic strategy. Algorithms based on other strategies, including elite selection within generations (training ROC area 0.819), and inversions of genetic material during recombination (training ROC area 0.812), evolved less accurate ANN. Conclusion: ANN optimized by genetic algorithms accurately discriminated pneumonia within a training cohort, and within a testing cohort consisting of cases on which the networks had not been trained. Genetic algorithms can be used to implement efficient search strategies for optimal ANN to predict pneumonia.

KW - Artificial neural network

KW - Genetic algorithm

KW - Pneumonia

UR - http://www.scopus.com/inward/record.url?scp=0346848886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0346848886&partnerID=8YFLogxK

U2 - 10.1016/S0933-3657(03)00065-4

DO - 10.1016/S0933-3657(03)00065-4

M3 - Article

VL - 30

SP - 71

EP - 84

JO - Artificial Intelligence in Medicine

JF - Artificial Intelligence in Medicine

SN - 0933-3657

IS - 1

ER -