Classification of Arrhythmia Potential using the K-Nearest Neighbor Algorithm

Abstract

According to the World Health Organization in 2021, cardiovascular diseases caused around 17.9 million deaths worldwide, making them the leading cause of global mortality. This data indicates that deaths due to heart disease remain very high, largely due to a lack of tools and technology for early detection of heart conditions. In this study, the K-Nearest Neighbor (KNN) machine learning algorithm is used with Electrocardiogram (ECG) signal data. The study was conducted on 30 subjects, where the heart activity data of each subject was recorded using an ECG device. The collected data is classified into four categories: high potential for arrhythmia, potential for arrhythmia, normal, and abnormal. The implementation of the KNN algorithm resulted in an accuracy rate of 93%. The high accuracy of the KNN algorithm is expected to make a significant contribution to the early detection of cardiovascular diseases.