Classification of First and Second Heart Sound based on Novel Statistical Features

In this study, we focus on the new feature extraction method to classify the first (S1) and second (S2) heart sound based on their wave shape. To perform heart sound analysis, distinguishing the representative signals S1 and S2 is one of the most important task. We designed seven formulae based on statistical features. Harr wavelet transform and Gaussian smoothing filter were applied to a phonocardiogram (PCG) recording. Among seven features, five features were extracted from the PCG signal on which Harr wavelet transform was applied and two features were extracted from the signals on which Gaussian smoothing filter was applied. Experiments are conducted on a public dataset of 135 subjects using an artificial neural network (ANN). ANN is trained with seven features extracted from the proposed methods and then tested to determine the accuracy of the classification of S1 and S2 signals. The results showed accuracy, specificity, and sensitivity, each having a value of 99%.