Detection of arrhythmias and myocardial infarction using SVM and ANN algorithms

Authors

Mohammad Issa, Faculty of Biomedical Engineering, Al Andalus University for Medical Science, Tartous, Syria.Follow
Ghada Saad, Faculty of Biomedical Engineering, Al Andalus University for Medical Science, Tartous, Syria / Department of Biomedical Engineering, Faculty of Mechanical and Electrical Engineering, Tishreen University, Latakia, Syria.Follow
Mohammad Abdo, Faculty of Biomedical Engineering, Al Andalus University for Medical Science, Tartous, Syria.Follow
Aous Mohammad, Department of Computer and Automatic Control Engineering, Faculty of Mechanical and Electrical Engineering, Tishreen University, Latakia, Syria.Follow

Abstract

The heart is one of the most vital organs in humans, and any defect in its function is reflected in the general health of the patient. Therefore, the heart and its diseases have been largely studied in order to help the doctors diagnose heart diseases and reduce errors as much as possible. This study aims to investigate and suggest a method for diagnosing heart diseases from electrocardiogram (ECG) signals. Initially, noise was removed from the signal and the morphological and dynamic features of the ECG signal were extracted by an appropriate feature extraction algorithm and wavelet transformation. A support vector machine (SVM) classifier was then proposed to identify the sound signals from the pathological signals, followed by the use of an appropriate neural classifier in order to classify the pathological signals and extract the results. The system has reached an accuracy of 97 % of the diseased varieties from the healthy ones, out of 150 samples. The proposed system was also able to identify 96% of the number of pathological samples, and attributing them to 3 categories (myocardial infarction, arrhythmias, and other categories).

Keywords

ECG features extraction, myocardial infraction, neural network classifier, SVM, Wavelet transformation

Article Type

Article

How to Cite this Article

Issa, Mohammad; Saad, Ghada; Abdo, Mohammad; and Mohammad, Aous (2024) "Detection of arrhythmias and myocardial infarction using SVM and ANN algorithms," Baghdad Science Journal: Vol. 21: Iss. 8, Article 2.
DOI: https://doi.org/10.21123/bsj.2024.8168