Heart arrhythmia classification using extracted features in poincare plot of RR intervals

Abstract

Early diagnosis of cardiac arrhythmia is important for a better management of arrhythmia. The main goal of this article was to compare conventional and new features extracted from Poincare plot between normal sinus rhythm, atrial fibrillation, acute myocardial infarction, and congestive heart failure. Furthermore, extracted features from Poincare plot were used w k-nearest neighbor (KNN) classifier for classification of NSR and different arrhythmia. The classification sensitivity, specificity, and accuracy of 94.63%, 98.21%, and 97.31% were achieved using extracted features with KNN.