Use of fuzzy neural network to predict coronary heart disease in a Malaysian sample

Authors:
Basir Abidin;Rosma Mohd Dom;A. Rashid A. Rahman;Roselina A. Bakar
Affiliations:
Cyberjaya University College of Medical Sciences, Malaysia;Faculty of Information Technology & Quantitative Science, MARA University of Technology, Malaysia;Cyberjaya University College of Medical Sciences, Malaysia;Institute of Medical Research, Malaysia
Venue:
TELE-INFO'09 Proceedings of the 8th Wseas international conference on Telecommunications and informatics
Year:
2009

Citing 3
Cited 0

Readings in medical artificial intelligence: the first decade

Readings in medical artificial intelligence: the first decade
Neuro-fuzzy approach versus rough-set inspired methodology for intelligent decision support

Information Sciences—Informatics and Computer Science: An International Journal
Artificial Intelligence Illuminated

Artificial Intelligence Illuminated

Quantified Score

Hi-index	0.00

Visualization

Abstract

The purpose of this study was to evaluate the ability of fuzzy neural network model to predict the likelihood of coronary heart disease for individuals based on knowledge of their biomarkers, risk habits and demographic profiles. The prediction performance of fuzzy neural network models were measured in terms of percentage accuracies and compared with the prediction performance of logistic regression models. Provisionary results showed that four markers namely body mass index, systolic blood pressure, total cholesterol level, and age are the appropriate markers for the prediction of coronary heart disease in the sample studied. Fuzzy neural network models prediction performance were found to be superior to the logistic regression performance as well as to other results reported in related literature.