Classification of PCG Signals using Fourier-based Synchrosqueezing Transform and Support Vector Machine

In this paper, a new approach based on the Fourier-based synchrosqueezing transform (FSST) is proposed to classify the normal and pathological Phonocardiogram (PCG) signals. The time-frequency (TF) matrices of the PCG signals are evaluated using FSST, and features such as energy and entropy are computed from the TF matrices obtained from the PCG signals. The most relevant features are selected and then classified using support vector machine (SVM) with different kernel functions. The effectiveness of the proposed approach is evaluated on the publicly available datasets, and the classification yields a result of 96.33% accuracy with 90.66% specificity and 98.33% sensitivity. The experimental results demonstrate that the proposed approach effectively classifies PCG signals into normal and pathological cases. Hence, this approach can be used as a diagnostic tool by the medical practitioner.