| Part One Heart Model A.I.artificial intelligence three-dimensional echocardiography quantitative evaluation of left ventricular function and parameter settingObjective: This study aims to explore the feasibility of Heart Model A.I.(HM)artificial intelligence(A.I.)three-dimensional echocardiography(3DE)to make quantitative evaluation of left ventricular function and discover suitable border parameter settings.Methods: A total of 113 participants that underwent echocardiography in our hospital were eligible for inclusion.The HM 3DE(HM method)and conventional 3DE(3DE method)were used to analyze echocardiography images.The HM border parameter settings end-diastolic(ED)and end-systolic(ES)were set to 40 and 20,60 and 30,60 and 50,70 and 30,70 and 40,74 and 68,80 and 40,90 and 50,100 and 50 to assess different left ventricular systolic function parameters including left ventricular end-diastolic volume(LVEDV),left ventricular end-systolic volume(LVESV),left ventricular ejection fraction(LVEF),and left ventricular diastolic function parameters including maximal left atrium volume(LAVMAX).Performed the correlation,agreement,reproducibility assessment of HM and 3DE measurements.Results:1.LVEDV,LVESV and LVEF measured with HM border parameter settings that ED and ES were set to 40 and 20,60 and 30,60 and 50,70 and30,70 and 40,74 and 68,80 and 40,90 and 50,100 and 50 were strongly correlated with the 3DE method.For LVEF,the reading using the HM method with ED and ES = 70 and 30 and 70 and 40 showed the best correlation with the 3DE method(r = 0.92,0.92,P < 0.05).For LVEDV,the reading using the HM method with ED and ES = 40 and 20 showed the best correlation with the3 DE method(r = 0.75,P < 0.05).For LVESV,the reading using the HM method with ED and ES = 40 and 20 showed the best correlation with the 3DE method(r = 0.85,P < 0.05).2.For LVEF,the reading using the HM method with ED and ES = 70 and30 showed the best agreement with the 3DE method,and the difference in the readings was not statistically significant(Bias = 0.12%,P > 0.05).For LVEDV,the reading using the HM method with ED and ES = 40 and 20 showed the best agreement with the 3DE method,the difference in the readings were statistically significant(Bias = 19.62 m L,P < 0.05).For LVESV,the reading using the HM method with ED and ES = 40 and 20 showed the best agreement with the 3DE method,the difference in the readings were statistically significant(Bias = 8.67 m L,P < 0.05).3.For LAVMAX,the correlation between the HM method and 3DE method was strong(r = 0.70,P < 0.05),but the agreement was not satisfied(Bias = 17.61 m L,P < 0.05).4.The measurements using the HM method were more reproducible than those using the 3DE method(P < 0.05).5.The measurement time when using the HM method was significantly less than the 3DE method(P < 0.05).Conclusions: Evaluation of the left ventricular function using HM method is feasible,saves time,and is reproducible.The accuracy of LVEF was better when the border parameter setting of ED and ES = 70 and 30.Part Two Heart Model A.I.artificial intelligence three-dimensional echocardiography quantitative evaluation of left ventricular systolic function in patients with myocardial infarctionObjective: This study aims to investigate the feasibility of using 3D cardiac anatomical model automatic quantitative technique Heart Model A.I.(HM)artificial intelligence(A.I.)three-dimensional echocardiography(3DE)to assess left ventricular systolic function in patients with myocardial infarction.Methods: A total of 62 participants that underwent echocardiography in our hospital were eligible for inclusion,which were divided into two groups:30 in the myocardial infarction group(case group)and 32 in the normal control group(control group).M-mode echocardiography(MME),two-dimensional echocardiography(2DE),HM 3DE(HM method)and conventional 3DE(3DE method)were used to collect echocardiographic images.Left ventricular systolic function parameters including left ventricular end-diastolic volume(LVEDV),left ventricular end-systolic volume(LVESV),and left ventricular ejection fraction(LVEF)were obtained and performed correlation,agreement,reproducibility assessment.Results:1.Compared with the control group,the LVEDV and LVESV measured by MME,2DE,HM,and 3DE in the case group were higher than those in the control group,and the LVEF in case group were lower than those in the control group(P < 0.05).2.Compared with MME,2DE,HM and 3DE,HM method had the best correlation with 3DE method in evaluating LVEF,and the case group was the same as the control group(r = 0.99,0.99,P < 0.05).3.Compared the agreement of MME,2DE,HM and 3DE,HM method and 3DE method had the best agreement in evaluating LVEF,and the control group was better than the case group(Bias =-1.00%,-2.53%,P < 0.05).4.Compared the reproducibility of MME,2DE,HM and 3DE,the reproducibility of HM was the best,and the control group was better than the case group(P < 0.05).5.Compared the measurement time of MME,2DE,HM and 3DE,HM took the shortest time,and the control group was better than the case group(P< 0.05).Conclusions: HM automatic quantitative technique can evaluate left ventricular systolic function in patients with myocardial infarction,and it is simple,rapid and reproducible.The measurement of LVEF showed good correlation and reproducibility between HM and 3DE,... |
PDF Full Text Request