Research Of Diastolic Murmurs For Coronary Artery Stenosis Based On EEMD

Coronary Artery Disease (CAD) is one of the leading causes of death in the world. Due to coronary artery disease's high incidence rate and mortality, it is very harmful to human health. With the raising of life quantity and the extending of the average longevity, morbidity for CAD is gradually increasing. Therefore, the early detection and diagnosis is very crucial. Heart sound is one of the most important physiological signals in body, before some adverse symptoms because of heart coronary artery stenosis have appeared, such as the electrocardiogram abnormalities, pain symptoms, the high frequency heart murmurs contains the reliable information of the pathological diagnosis of coronary artery stenosis. Research and analyze diastolic heart sound signals using the modern signal processing method, which has important significance for the early diagnosis of coronary artery disease.This thesis mainly analyzes the diastolic heart sounds and the feature extraction algorithm based on EEMD (Ensemble Empirical Mode Decomposition). On the basis of analyzing the relation between the coronary artery stenosis and the diastolic murmurs, firstly, the denoising and the segmentation of the heart sound are studied. Secondly, the diastolic heart sounds are analyzed by EEMD, the algorithm of the instantaneous frequency extraction by EEMD-TEO and the diastolic murmurs nonlinear dynamics analysis based on EEMD and sample entropy are proposed. At last, through the researching and analyzing the diastolic heart sounds of coronary artery disease in clinic, it effectively reveals the information hidden in the diastolic murmurs and provides an effective means for the early diagnosis of coronary artery disease.The main contents of this paper are as follows:In this paper, EEMD method is studied. The method of adaptive denoising heart sounds is proposed based on the EEMD threshold. Through analyzing the heart sound signals in clinic, the noise has been effectively filtered and the SNR (Signal to Noise Ratio) has been improved. We propose a new method to segment the heart sound based on EEMD and Hilbert transform. As studying the segmentation of a large of heart sounds in clinic, the experimental result shows that the algorithm has the strong robustness, high accuracy and no extra signal for reference.The instantaneous frequency features of diastolic murmurs are analyzed. The thesis proposes a method of analyzing diastolic murmurs by EEMD-TEO and obtains the law and the characteristic of changes of the instantaneous frequency. With the analysis of the diastolic heart sounds of coronary artery disease in clinic, the experimental result shows that the features effectively reveal the information of coronary artery stenosis and describe the characteristic of changes of diastolic heart sound signals.The paper researches the sample entropy algorithm of diastolic heart sounds on nonlinear dynamics. Properties of the sample entropy and the influences of low frequency component are analyzed. The features on nonlinear dynamics are obtained by EEMD and sample entropy. Through the research and analysis the diastolic heart sounds of coronary artery disease in clinic, the results showed that the method could effectively reveal the information hidden in the diastolic murmurs because of coronary artery stenosis.In this paper, the method of analysis the diastolic heart sounds and feature extraction algorithm was proposed based on EEMD, it effectively reveals the information of the diastolic heart sounds because of coronary artery stenosis. The research of the heart sounds lays the foundation for the further development of early diagnosis and treatment of coronary artery disease.