A Multi-Class Cardiac Sound Diagnostic System in Deep Learning Based on PCG Signal

Heart sound classification plays a vital role in the early discovery of cardiovascular diseases, especially for small primary health care conventions. Despite that important progress has been made for heart sound bracket in recent times, utmost of them are grounded on conventional segmented features and shallow structure-grounded classifiers. These conventional aural representation and classification styles may be inadequate in characterizing heart sound, and generally suffer from a degraded performance due to the complicated and changeable cardiac aural terrain. In this paper, a new heart sound bracket system has been proposed grounded on mongrel features of heart sound signals and convolutional intermittent neural network classifier model. Then uses MFCC and HCQT features for the heart sound spectrograms. And the model categorizes five classes of heart sounds in an effective way using Convolution Neural Network models.