The Use of Genetic Algorithm and Particle Swarm Optimization on Tiered Feature Selection Method in Machine Learning-Based Coronary Heart Disease Diagnosis System

Aug 4, 2026

Upload by Ivy

Coronary heart disease (CHD) is a leading global cause of death, making early detection critical for lowering mortality and treatment cost. Machine learning can support diagnosis using patient medical record datasets, but excessive features and irrelevant signals often degrade model performance. To address this, the study proposes a tiered feature selection framework combining genetic algorithm (GA) and particle swarm optimization (PSO), evaluated via confusion-matrix-derived metrics using CatBoost on multiple benchmark datasets.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

INTRODUCTION

Problem: CHD and limitations of feature-rich medical records

Related work on feature selection for CHD diagnosis

Computational intelligence methods (GA/PSO)

Comparison with other feature selection strategies

The use of genetic algorithm and particle swarm optimization on tiered feature selection method in machine learning-based coronary heart disease diagnosis system Wiharto1, Yasmin Mufidah1, Umi Salamah1, Esti Suryani2, Sigit Setyawan3 1Department of Informatics, Faculty of Information Technology and Data Science, Universitas Sebelas Maret, Surakarta, Indonesia 2Department of Data Science, Faculty of Information Technology and Data Science, Universitas Sebelas Maret, Surakarta, Indonesia 3Department of Medicine, Faculty of Medicine, Universitas Sebelas Maret, Surakarta, Indonesia Article Info ABSTRACT Article history: Received Oct 26, 2023 Revised Mar 7, 2024 Accepted Mar 16, 2024 Keywords: CatBoost algorithm Coronary heart disease Feature selection Genetic algorithm Particle swarm optimization Corresponding Author: Coronary heart disease (CHD) is a leading global cause of death. Early detection is the right step to reduce mortality rates and treatment costs. Early detection can be developed using machine learning by utilizing patient medical record datasets. Unfortunately, this dataset has excessive features which can reduce machine learning performance. For this reason, it is necessary to reduce the number of redundant features and irrelevant data to improve machine learning performance. Therefore, this research proposes a tiered of feature selection model with genetic algorithm (GA) and particleswarm optimization (PSO) to improve the performance of the diagnosis model. The feature selection model is evaluated using parameters derived from the confusion matrix and using the CatBoost machine learning algorithm. Model testing uses z-Alizadeh Sani, Cleveland, Statlog, and Hungarian datasets. The best results for this model were obtained on the z-Alizadeh Sani dataset with 6 selected features from 54 features and the resulting performance for accuracy parameters was 99.32%, specificity 98.57%, sensitivity 100.00%, area under the curve (AUC) 99.28%, and F1-Score 99.37% . The proposed feature selection model is able to provide machine learning performance in the very good category. The diagnostic model proposed is of excellent standard. This is an open access article under the CC BY-SA license. Wiharto Department of Informatics, Faculty of Information Technology and Data Science, Universitas Sebelas Maret Jl. Ir Sutami No. 36A, Kentingan, Jebres, Surakarta, Indonesia [Email: wiharto@staff.uns.ac.id](Email: wiharto@staff.uns.ac.id) 1. INTRODUCTION Coronary heart disease (CHD) arises from impaired function of the heart and blood vessels. It is a primary cause of mortality worldwide [1] . The World Health Organization (WHO) reported that CHD caused the deaths of up to 17.9 million individuals in 2019. Research by Alizadehsani et al. [2] indicates that 25% of individuals with CHD die unexpectedly and without any preceding symptoms. Early detection of coronary heart disease (CHD) is essential to decrease mortality rates associated with the disease [3] . Currently, CHD diagnosis is developed using machine learning models. However, the process necessitates extensive medical record data. The abundance of medical record data often results in numerous features that do not directly facilitate diagnosis [4] and can ultimately negatively influence machine learning performance. To address these issues, data mining techniques may be employed, specifically through the feature selection method. Feature selection can identify features that enhance the effectiveness of machine learning-based diagnostic models [5] . Numerous studies have been conducted on developing machine learning-based models for diagnosing coronary heart disease that uses feature selection. Kolukisa and Bakir [5] compared various feature selection models, including chi square, information gain, ReliefF, and support vector machine (SVM) . The best performance was produced when using the z-Alizadeh Sani dataset with SVM feature selection producing 25

FAQs

Why is feature selection important in machine learning-based CHD diagnosis?

Medical records often contain excessive features, including redundant and irrelevant variables. These can reduce machine learning performance by adding noise and weakening predictive ability.

What feature selection approach does this research propose?

It proposes a tiered feature selection model that combines a genetic algorithm (GA) and particle swarm optimization (PSO) to select informative subsets of features.

How is the proposed model evaluated?

Evaluation uses parameters derived from the confusion matrix and applies the CatBoost machine learning algorithm on multiple datasets, including z-Alizadeh Sani, Cleveland, Statlog, and Hungarian.

Aug 4, 2026

Upload by Ivy