Segmentation of heart sound signal into cycles based on time properties of the heart sound

Abstract

A simple alternative method to segment the heart sound signal into cycles is proposed in this paper. The segmentation is done based on the time properties of the heart sound signal only, without the assistant of other biomedical signal such as ECG and pulse signal. The procedures are including selecting the heart sound envelogram to find significant peaks, recovering missing peaks, removing unwanted peaks and examining cycle’s sequence. 52 over 62 (83.87%) set of heart sounds signal is successfully segmented with segmentation accuracy over 80%.