The Researches Of Heart Sound Signal On The Basis Of Wavelet

Various kinds of information of heart diseases is often reflected in the heart sound.Collection and analysis of heart sound is an important method that can be used to judge the disease of cardiovascular. Heart sound is one of the most important voice signals of human body. It includes physiology and pathologic information of each part of heart.Such as atrial,ventricular, great vessels,cardiovascular and so on.Thence,accurate analysis and recognition of heart sound is significant to the diagnosis of cardiovascular diseases.Because the heart sound is a signal of the nonstationary randomness and complexity, the analysis of heart sound is indeed difficult.The typical methods include STFT, auto-regression model and the wavelet in the time-frequency analysis of the heart sound.Then this paper adopts the wavelet to analyze heart sound signal and the parts of collection and analysis are based on the development platform of LabVIEW8.6.In the course of recording heart sound,it is inevitable that many kinds of noise will be merged in the main signal.Therefore, the preprocessing is necessary to remove the noises.After the preprocessing,the quality of signal is improved and the result of denoising is expected.The main works in this paper are listed as follows:(1)Based on the development platform of LabVIEW8.6,a acquisition system of heart sound based on sound card is developed.The system is characterized by being simple carry out,friendly interface,stable performance,economical strong function and suitable.(2)In order to eliminate the noise of heart sound,we proposed one denoising methods which is Wavelet Filter. We also compared the advantages and disadvantages of the different parameter and the result of denoising is optimal.(3)In this paper, the heart sound signal was decomposed into many levels by using wavelet multi-resolution.We can accurately identify S1,S2 and murmurs from different levels.We can also classify the heart sounds signals into three categories:signal with systolic murmur, signal with diastolic murmur,or normal signal.The results show that the presented algorithm can provide an effective reference for the diagnosis of heart disease.(4)Based on Analytic Wavelet Theory, We can efficiently distinguish normal heart sounds from abnormal heart sounds and reveal the variation process of frequency along with time through established time-frequency energy spectrum.This article has provided the basic analysis method and the analysis tool to heart sound signal clinical research,lay the foundation to heart sound used in heart disease diagnosis.Finally, we made a conclusion,point out some disadvantages of the system,also propose the future research directions in this field.