Disorders in the functions of the heart cause heart diseases or arrhythmias in the cardiovascular system. Diagnosis in cardiac arrhythmias is realized utilizing the Electrocardiogram which is an electrophysiological signal. In this study, a three-class, K-means clustering-based arrhythmia detection method, distinguishing the cardiac arrhythmia type Right Bundle Branch Block and Left Bundle Branch Block from normal heart-beats, is proposed. Data from the MIT-BIH Arrhythmia Database were analyzed for clustering-based arrhythmia analysis. Feature Set 1 was created by extracting the features from the Electrocardiogram signal with the help of QRS morphology, Heart Rate Variability and statistical metrics. The RELIEF feature selection algorithm was used for dimensionality reduction of the obtained features and Feature Set 2 was obtained by determining the most appropriate features in Feature Set 1. Overall performance results for Feature Set 1 were obtained as 99,18% accuracy, the sensitivity of 98,78% and 99,39% specificity while overall performance results for Feature Set 2 were provided as 95,37% accuracy, the sensitivity of 92,99% and 96,54% specificity. In this study, the computational cost was decreased by reducing the processing complexity and load utilizing the reduced feature data set FS2 and an arrhythmia detection method having a satisfactory level of high performance was proposed.
Arrhythmia detection, Electrocardiogram (ECG), K-means clustering, Machine learning, RELIEF feature selection algorithm