Research On Analysis Of Electro-mechanical Characteristics Of Cardiovascular System Based On Surface Signals

The incidence and morbidity of cardiovascular diseases(CVD)are increasing year by year,and its prevalence in young people is also elevated.Non-invasive and non-destructive early detection of CVD and reducing cardiovascular events are of great social significance.Surface cardiovascular signals contain abundant physiological information and surface signal analysis is one of the effective methods for non-invasive and non-destructive detection of CVD.Studies on cardiovascular cardiac electrical and mechanical characteristics analysis(such as heart rate variability analysis and pulse wave transit velocity analysis)have attracted extensive attention from researchers.The combined analysis of electrical and mechanical characteristics can describe the cardiovascular function as a whole and reveal the interaction between the components of the system.However,relevant issues remains to be further studied.In this dissertation,we aimed to systematically study the electromechanical characteristics of the cardiovascular system based on surface electrocardiogram,heart sound and pulse wave signals,and explore effective information to diagnose cardiovascular diseases.Main works and innovations were as follows:(1)Dynamic pattern analysis was introduced to analyze the cardiac electrical characteristics of patients with congestive heart failure(CHF)and coronary artery disease(CAD);the role of cardiac electrical characteristics analysis in cardiovascular disease detection was discussed.Results indicated that compared with healthy people,QT interval of CHF patients was significantly prolonged,the complexity significantly reduced.Ventricular depolarization and repolarization activity became unstable in CHF patients,but the electrical characteristics of CAD patients did not show significant change.In QTV analysis,it was necessary to pay attention to the coupling relationship between QT interval and RR interval.Dynamic pattern analysis could capture the hidden details of QT time series,which provides a new method for detecting cardiac function changes.(2)The cardiac electrical and mechanical characteristics variability and cardiac electro-mechanical coupling characteristics in CAD patients were systematically studied.The results showed that,compared with the healthy control group,the standard deviation of S1S2 series in the CAD group increased significantly,and the systolic intervals changed more irregularly.The coupling of QT-S1S2,TQ-S2S1 and QQ-S1S1 were significantly reduced in the CAD group.The synchronization of electrical and mechanical activities in CAD patients was impaired,and the immediate response of cardiac mechanical activities to electrical excitement was weakened.The results suggested that the combined analysis of electro-mechanical characteristics was more effective in non-invasive and non-destructive detection of CAD than single electrocardiogram or electrocardiogram signal analysis.(3)The cardiac electro-mechanical delay variability was proposed,and machine learning algorithm was performed to realize the automatic classification of CAD group and healthy group.Results showed that,compared with the healthy control group,the electro-mechanical delay of CAD patients significantly prolonged,sample entropy significantly decreased,and the proportion of oscillating mode in the dynamic pattern significantly increased.It further indicated that atherosclerotic lesions would lead to the dyssynchrony of electro-mechanical activities of myocardial cells.The SVM results showedthat the addition of electro-mechanical delay variability indices significantly improved the classification accuracy from 72.9%based on signal electrical and mechanical characteristics to 95.8%.These results suggested that the cardiac electromechanical delay characteristics were highly sensitive and specific to CAD,and it could potentially be helpful for noninvasive detection CAD.(4)Based on the pulse wave signal of the cuff,the influence of age and CAD on the pulse wave transit time(PTT)between different sites of the human body and bilateral similarity were analyzed for thr first time.The results showed that compared with the healthy control group,BaPTT was significantly shortened in the CAD group,while the mean of HbPTT and HaPTT did not change significantly.The brachial-ankle PTT(BaPTT),heart-brachial PTT(HbPTT)and heart-ankle PTT(HaPTT)were significantly shortened in elderly group,which suggesting that aging could lead to degeneration of vascular elastic function.The young students showed a good matching between left and right PTT series,and both aging and CAD led to a significant increase in bilateral PTT difference and a significant decrease in bilateral similarity.These results suggested that BaPTT analysis and similarity analysis of bilateral cardiovascular time series are helpful for the assessment of vascular elasticity function and non-invasive and nondestructive detection of CAD.