Classification Of Heart Sound Signals Using Scalogram And Convolutional Neural Network

Cardiovascular diseases (CVDs) are the most leading causes of death every year in world. The threat CVDs can be decreased and controlled with early diagnoses. Therefore, interpreting heart sounds is considered as one common ways to diagnose cardiovascular system. Heart sound signal known phonocardiogram (PCG) provides useful information about condition, which used diagnostic, helps physicians detection several abnormalities. technology development helped appearance new diagnosis techniques, combines advanced processing techniques deep learning algorithms. Thus, classification becoming a crucial task modern healthcare field. In this work learning-based method was proposed. Using PCG database contains five different classes taken from cases valve defects. Scalogram signals time-frequency representation create scalogram image extracted database. A convolutional neural network Direct Acyclic Graph structure (DAG CNN) evaluation performance indicated that accuracy 99,6\%. comparative results manifest proposed had better compared other previous works same used.