Part 1Objective:Two-dimensional speckle tracking echocardiography(2D-STE)is a novel and noninvasive technique.This study aimed to identify CAD individuals using 2D-STE in suspected paitentsMethods:Totally,690 patients were enrolled.Reduction in vessel diameter of>50%by stenosis in at least one major coronary artery or its main branch was considered coronary artery disease(CAD).Analysis of 2D-STE was performed using EchoPAC version 201.Results:Global longitudinal peak strain(GLPS)was analyzed in 3 layers:GLPS of the epicardium,the area under the curve(AUC)was 0.727,and the cut-off value was-16.95;sensitivity and specificity were 73.7%and 63.0%,respectively.GLPS of the middle layer,the OR was 1.260(1.192-1.333;P<0.001),the AUC was 0.732,and the cut-off value was-20.95;sensitivity and specificity were 82.4%and 56.2%,respectively.GLPS of the endocardium,the AUC was 0.708,and the cut-off value was-22.95;sensitivity and specificity were 82.9%and 52.9%,respectively.Conclusions:The findings support the clinical application of longitudinal layer strain of 2D-STE in patient populations with suspected myocardial ischemia due to CAD.Part 2Objective:Evidence suggests that screening of coronary artery disease(CAD)at an earlier stage can greatly reduce the mortality rate.This study aims to establish a machine learning model to improve the efficiency of speckle tracking echocardiography in the diagnosis of CAD.Methods:In this study,64 two-dimensional speckle tracking imaging parameters,including layer strain,strain rate and strain peak time,and 7 clinical characteristics including age,gender and history of hypertension were used.19 classifiers were integrated using R-Package Caret through model stacking.By borrowing strengths from multiple classification models though the proposed method,the new method was promoted to obtain higher diagnostic performance.Results:The accuracy of this predictive model in the diagnosis of CAD has been improved from about 70%to 87.7%.The sensitivity of the proposed method is 0.903 and the specificity is 0.843,with an AUC of 0.904,which is significantly higher than those of the individual classification models.Conclusion:The prediction model obtained by ensemble learning can effectively predict coronary artery disease,which can provide a simple,quick and effective tool for screening the disease.Part 3Objective:Myocardial work(MW)is a novel non-invasive method for assessing the left ventricular(LV)function.We aimed to investigate this method for assessing the LV function in CAD patients with different heart function.Methods:We enrolled totally 150 healthy and CAD individuals.CAD patients were divided into the normal blood pressure and hypertension(HTN)subgroups.The relationships between MW indices and conventional parameters were evaluated;MW indices were compared between the groups.Results:MW indices were strongly correlated with the left ventricular ejection fraction(LVEF).Compared with the control group,the global work index(GWI)was increased in CAD with normal LVEF subgroup with hypertension(1639.72±204.57 VS 1922.29±393.07 mmHg%,p<0.05);and decreased in all HF patients.The global waste work(GWW)was increased in all CAD sub-groups.In the CAD patients,the global constructive work(GCW)had the same tendency as GWI.Compared with the control group,GWE in CAD patients with normal LVEF was decreased(HTN:95.26±1.99 VS 91.62±3.29,p<0.05;No-HTN:95.26±1.99 VS 92.88±3.18%,p<0.05);in CAD patients with HF was also decrease(HTN:95.26±1.99 VS 78.38±8.06,p<0.05;No-HTN:95.26±1.99 VS 80.12±9.7%,p<0.05).Conclusion:MW provides an accurate quantified assessment of the LV function in CAD patients.It offers additional information on LV function regarding disease progression. |