Exploration of cardiac valvular hemodynamics by heart sound analysis of hypertensive cardiac disease background patients

Heart sounds reflect heart valve activities, cardiac muscle contractions, and hemodynamics of heart. Auscultation technique is one of valuable methods for diagnosis. There are limitations in traditional auscultation stimulate utilization of signal processing methods to extract important characteristics of the heart sounds. We proposed Continuous Wavelet Transform (CWT) as time-frequency analysis for exploration of cardiac valvular hemodynamics of two normal subjects with hypertensive heart disease history. Decimation and a wavelet denoising were used for filtering. A normalized average Shannon energy was used for heart sound signal segmentation. Time-scale maps resulted by calculation of CWT were processed using thresholding method to localize temporal and frequency-related information of valvular activities of each cardiac sub-cycle. The temporal and frequency-related parameters were spaced time between systolic (S1) and diastolic (S2), activities of mitral (M1) and tricuspid (T1) valves during systolic period, aortic (A2) and pulmonary (P2) valves during diastolic period, and split time between valve activities in each cycle. Results of this study were considered to be valuable to explain the cardiac valvular hemodynamics of heart sounds precisely. The results indicate the benefit of the developed method to be applied in analyzing heart sound characteristics. The results are very helpful information for clinical use. Next topic of our research was addressed to expand the analytical method to other pathological background. The research work would be finalized by development of software analysis of cardiac pathology diagnostic system.