Classification of Heart Sound Signals Using Multi-modal Features

Abstract

Abstract Cardiac auscultation is a technique of listening to heart sounds. Any abnormality in the heart sound may indicate some problem in the heart. In this paper, the phonocardiogram (PCG) signal i.e. the digital recording of the heart sounds has been studied and classified into three classes namely normal signal, systolic murmur signal and diastolic murmur signal. Total number of samples used for this study are 144 out of which 60 are normal signals, 45 are diastolic murmur signals and 39 are systolic murmur signals. Various features have been extracted for the classification. A total of 28 features have been extracted and then reduced to 7 most significant features using feature reduction technique. The selected features have been used to classify the signal into various classes using classifiers. The classifiers which have been used in this study are k-NN (k Nearest Neighbour) , fuzzy k-NN and Artificial Neural Network (ANN) . Both k-NN and fuzzy k-NN as classifiers have the highest accuracy of 99.6%.