Page 67 - AIH-1-4

Artificial Intelligence in Health

ORIGINAL RESEARCH ARTICLE
Heartbeat classification using various machine

learning models: A comparative study

2
Marc Nshimiyimana 1 , Jovial Niyogisubizo * , and Jean de Dieu Ninteretse 3
1 Department of Bridge, Tunnel and Underground Engineering, School of Civil Engineering, Southeast
University, Nanjing, China
2 Shenzhen Key Laboratory of Intelligent Bioinformatics and Center for High-Performance Computing,
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
3 Department of Construction and Real Estate, School of Civil Engineering, Southeast University,
Nanjing, China

Abstract
Cardiac arrhythmias, known as irregular heartbeats, pose a notable health threat
that necessitates prompt diagnosis, as untreated arrhythmias can lead to severe
heart complications. Among the various methods for arrhythmia detection,
electrocardiography is the most prevalent due to its non-invasive monitoring of
heart activity. However, manual electrocardiogram (ECG) analysis is inefficient and
prone to errors, prompting the exploration of machine learning (ML) models for
ECG feature recognition. Integrating ML models with ECG analysis can revolutionize
cardiac diagnostics by improving healthcare efficiency and outcomes by enhancing
the accuracy and consistency of existing approaches as well as their processing speed
for large datasets. Unfortunately, current ML methods encounter two key limitations:
*Corresponding author: prolonged training times and the need for manual feature selection. To address these
Jovial Niyogisubizo
(jovial@siat.ac.cn) issues, we propose using ML models enhanced with innovative techniques such as
the Fourier transform (FT) and Gaussian noise injection for improved cardiac health
Citation: Nshimiyimana M, assessment. To validate this approach, we utilized statistical tools, including Pearson
Niyogisubizo J, Ninteretse JDD.
Heartbeat classification using correlation and p-values, to uncover relationships within the data. In addition, we
various machine learning models: A employed the FT technique to extract and analyze frequency-domain features. Our
comparative study. Artif Intell comparative study of different ML models relied on metrics such as accuracy, precision,
Health. 2024;1(4):61-72.
doi: 10.36922/aih.3543 recall, F1 score, and receiver operating characteristic area under the receiver operating
characteristic curve, demonstrating XGBoost’s impressive average recall of 0.956 with
Received: April 30, 2024
99.96% overall accuracy. An average precision of 0.956 further underscored the accuracy
Accepted: September 3, 2024 of XGBoost’s predictions, indicating its high level of reliability in distinguishing various
Published Online: October 14, cardiac conditions. These results highlight the considerable potential of ML techniques
2024 for precise ECG-based clinical diagnoses, helping healthcare professionals make more
Copyright: © 2024 Author(s). accurate and timely decisions in patient care.
This is an Open-Access article
distributed under the terms of the
Creative Commons Attribution Keywords: Electrocardiogram; Fourier transform; Gaussian noise; Heartbeat classification;
License, permitting distribution, Machine learning; Pearson correlation
and reproduction in any medium,
provided the original work is
properly cited.
Publisher’s Note: AccScience
Publishing remains neutral with 1. Introduction
regard to jurisdictional claims in
published maps and institutional Electrocardiography generates an electrocardiogram (ECG or EKG), a record of the
affiliations. heart’s electrical activity, which can be monitored using both traditional medical devices

Volume 1 Issue 4 (2024) 61 doi: 10.36922/aih.3543