Heart Sound Signal Processing Based On DSP

Detection and analysis of heart sound is an important method that is applied to judge the state of heart and blood vessels. Cardiac auscultation is widely applied to evaluate cardiac function of patients. It has been discovered in previous research that the phonocardiogram can be used for cardiac performance assessment since the amplitude of S1 is the standard measure for myocardium contractility force. However, general phonocardiogram analysis instrument can only monitor the static patient's heart sound in a short time, ignorencing the changes of heart sound in various states and periods. Therefore, it is of much significance to develop a phonocardiogram analysis system that can be applied in real-time study of the regularity of heart sound and real-time analysis of the cardiac contractility variability in various states.For it is inevitable for noise to merge in heart sound recording, it is necessary to restrain the noise of the heart sound signals to extract the amplitude of S1 correctly. Since the heart sound signal is nonstationary, the paper applies the wavelet transform to restrain the heart sound signals'noise in order to get the better test results.The next step is the extraction of the heart sound signal after denoising. Comparing several extraction algorithms, the envelop extraction algorithm based on mathematical morphology is employed to extract heart sound signal. The results of the experiment prove the feasibility of the method employed in this paper.At last, this paper uses special assemble language of DSP to program the heart sound signals.The procedure has three modules: the wavelet module, the mathematic morphology module, the amplitude extraction module.The paper operates the algorithm in integrated development environment CCS2 and finally evaluates it on the EVM board. By comparing with MATLAB, it's known that the two results are similar.It is proved that it is effective to extract the amplitude of S1 using the algorithm and scheme proposed in the paper. It paves the way for further real-time analysis of cardiac contractility variability in practice.