Supervised Feature Selection for Diagnosis of Coronary Artery Disease Based on Genetic Algorithm

TL;DR

This paper introduces a genetic algorithm-based feature selection method combined with Bayesian Naive classification to improve the diagnosis accuracy of coronary artery disease, outperforming several traditional classifiers.

Contribution

A novel genetic algorithm wrapped Bayesian Naive feature selection method for CAD diagnosis, demonstrating higher accuracy than SVM, MLP, and C4.5 classifiers.

Findings

Achieved 85.50% classification accuracy for CAD diagnosis.

Outperformed SVM, MLP, and C4.5 classifiers in accuracy.

Promising results for feature selection in medical diagnosis.

Abstract

Feature Selection (FS) has become the focus of much research on decision support systems areas for which data sets with tremendous number of variables are analyzed. In this paper we present a new method for the diagnosis of Coronary Artery Diseases (CAD) founded on Genetic Algorithm (GA) wrapped Bayes Naive (BN) based FS. Basically, CAD dataset contains two classes defined with 13 features. In GA BN algorithm, GA generates in each iteration a subset of attributes that will be evaluated using the BN in the second step of the selection procedure. The final set of attribute contains the most relevant feature model that increases the accuracy. The algorithm in this case produces 85.50% classification accuracy in the diagnosis of CAD. Thus, the asset of the Algorithm is then compared with the use of Support Vector Machine (SVM), MultiLayer Perceptron (MLP) and C4.5 decision tree Algorithm. The result of classification accuracy for those algorithms are respectively 83.5%, 83.16% and 80.85%. Consequently, the GA wrapped BN Algorithm is correspondingly compared with other FS algorithms. The Obtained results have shown very promising outcomes for the diagnosis of CAD.