人工神经网络对左心室肥厚筛查的应用价值
|
摘要
目的本研究旨在探讨人工神经网络在左心室肥厚(LVH)筛查中的应用价值。方法共纳入健康体检者486例,将既往病史、心电图参数等11项指标作为预测因素,建立训练集和测试集,以超声心动图结果作为结局指标,建立人工神经网络模型。同时应用相应的预测因素,建立logistic回归模型,比较两个模型间的筛查诊断价值。结果人工神经网络模型预测LVH的灵敏度和特异度均高于logistic回归(LR)模型(灵敏度93%vs 89%,特异度91%vs 74%),人工神经网络ROC AUC大于log is tic回归模型[0.964,95%CI:0.921～1.000 vs 0.889,95%CI:0.831～0.948,Z=2.016,P<0.05]。结论在预测LVH上,人工神经网络模型优于logistic回归模型。
Objective To explore the application of artificial neural network(ANN) in screening of left ventricular hypertrophy(LVH). Methods A total of 486 healthy people were enrolled in this study. The training set and test set were established by using 11 indexes such as past medical history and electrocardiogram parameters as predictors. The results of echocardiography were used as outcome indicators to establish artificial neural network(ANN) model. Also, the logistic regression model was established with the same parameters. The diagnostic value of the two models was compared. Results The sensitivity and specificity of ANN model in predicting LVH were higher than those of logistic regression model(sensitivity 93% vs. 89%, specificity 91% vs. 74%). The area under ROC curve of ANN was larger than that of logistic regression model [0.964, 95% CI : 0.921 ～1.000 vs 0.889, 95% CI : 0.831 ～0.948, Z=2.016, P<0.05].Conclusion Artificial neural network model is superior to logistic regression model in predicting LVH.
Objective To explore the application of artificial neural network(ANN) in screening of left ventricular hypertrophy(LVH). Methods A total of 486 healthy people were enrolled in this study. The training set and test set were established by using 11 indexes such as past medical history and electrocardiogram parameters as predictors. The results of echocardiography were used as outcome indicators to establish artificial neural network(ANN) model. Also, the logistic regression model was established with the same parameters. The diagnostic value of the two models was compared. Results The sensitivity and specificity of ANN model in predicting LVH were higher than those of logistic regression model(sensitivity 93% vs. 89%, specificity 91% vs. 74%). The area under ROC curve of ANN was larger than that of logistic regression model [0.964, 95% CI : 0.921 ～1.000 vs 0.889, 95% CI : 0.831 ～0.948, Z=2.016, P<0.05].Conclusion Artificial neural network model is superior to logistic regression model in predicting LVH.
引文
[1]Beach JM,Mihaljevic T,Rajeswaran J,et al.Ventricular hypertrophy and left atrial dilatation persist and are associated with reduced survival after valve replacement for aortic stenosis[J].J Thorac Cardiovasc Surg,2014,147(1)∶362-369.
[2]Sani MU,Davison BA,Cotter G,et al.Echocardiographic predictors of outcome in acute heart failure patients in sub-Saharan Africa:insights from THESUS-HF[J].Cardiovasc J Afr,2017,28(1)∶60-67.
[3]Shenasa M,Shenasa H.Hypertension,left ventricular hypertrophy,and sudden cardiac death[J].Int J Cardiol,2017,237(15)∶60-63.
[4]Mancusi C,Gerdts E,de Simone G,et al.Higher pulse pressure/stroke volume index is associated with impaired outcome in hypertensive patients with left ventricular hypertrophy the LIFEstudy[J].Blood Press,2017,26(3)∶150-155.
[5]Bacharova L,Schocken D,Estes EH,et al.The role of ECG in the diagnosis of left ventricular hypertrophy[J].Curr Cardiol Rev,2014,10(3)∶257-261.
[6]Rodríguez-Padial L,Akerstrom F,Robles-Gamboa C,et al.Diagnostic accuracy of left ventricular hypertrophy in patients with myocardial infarction by omputer-assisted electrocardiography(ELEC-TROPRES)[J].Ann Noninvasive Electrocardiol,2012,18(2)∶170-180.
[7]Betjeman TJ,Rigales L,Hoffman I.Electrocardiographic Criteria for the Diagnosis of Left Ventricular Hypertrophy[J].J Am Coll Cardiol,2017,70(14)∶1829-1831.
[8]Taylor D,Harrison A,Powers D.An artificial neural network system to identify alleles in reference electropherograms[J].Forensic Sci Int Genet,2017,30∶114-126.
[9]Peixoto LA,Bhering LL,Cruz CD.Artificial neural networks reveal efficiency in genetic value prediction[J].Genet Mol Res,2015,14(2)∶6796-807.
[10]Lang RM,Badano LP,Mor-Avi V,et al.Recommendations for cardiac chamber quantification by echocardiography in adults∶an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging[J].J Am Soc E-chocardiogr,2015,28∶1-39.
[11]Germano G.Electrocardiographic signs of left ventricular hypertrophy in obese patients∶what criteria should be used?[J].High Blood Press Cardiovasc Prev,2014,22(1)∶5-9.
[12]Hopkins CB,Suleman J,Cook C.An artificial neural network for the electrocardiographic diagnosis of left ventricular hypertrophy[J].Crit Rev Biomed Eng,2000,28(3-4)∶435-438.
[13]Satoh S,Omura S,Inoue H,et al.Gender differences in factors influencing electrocardiographic findings of left ventricular hypertrophy in severe aortic stenosis[J].Heart Vessels,2014,29(5)∶659-666.
[14]Cuspidi C,Rescaldani M,Sala C,et al.Left-ventricular hypertrophy and obesity∶a systematic review and meta-analysis of echocardiographic studies[J].J Hypertens,2014,32(1)∶16-25.
[15]Tang ZH,Liu J,Zeng F,et al.Comparison of prediction model for cardiovascular autonomic dysfunction using artificial neural network and logistic regression analysis[J].PLo S One,2013,8∶e70571.
[16]Gu W,Vieira AR,Hoekstra RM,et al.Use of random forest to estimate population attributable fractions from a case-control study of salmonella enterica serotype enteritidis infections[J].Epidemiol Infect,2015,143∶2786-2794.
[17]余琳琳,曾少丹.窦性心律不齐在年龄层分布规律及性别差异分析[J].心电与循环,2018,37(2)∶121-123.
[18]黄进宇.慢性心力衰竭的药物治疗进展[J].心电与循环,2018,37(2)∶83-85.
[2]Sani MU,Davison BA,Cotter G,et al.Echocardiographic predictors of outcome in acute heart failure patients in sub-Saharan Africa:insights from THESUS-HF[J].Cardiovasc J Afr,2017,28(1)∶60-67.
[3]Shenasa M,Shenasa H.Hypertension,left ventricular hypertrophy,and sudden cardiac death[J].Int J Cardiol,2017,237(15)∶60-63.
[4]Mancusi C,Gerdts E,de Simone G,et al.Higher pulse pressure/stroke volume index is associated with impaired outcome in hypertensive patients with left ventricular hypertrophy the LIFEstudy[J].Blood Press,2017,26(3)∶150-155.
[5]Bacharova L,Schocken D,Estes EH,et al.The role of ECG in the diagnosis of left ventricular hypertrophy[J].Curr Cardiol Rev,2014,10(3)∶257-261.
[6]Rodríguez-Padial L,Akerstrom F,Robles-Gamboa C,et al.Diagnostic accuracy of left ventricular hypertrophy in patients with myocardial infarction by omputer-assisted electrocardiography(ELEC-TROPRES)[J].Ann Noninvasive Electrocardiol,2012,18(2)∶170-180.
[7]Betjeman TJ,Rigales L,Hoffman I.Electrocardiographic Criteria for the Diagnosis of Left Ventricular Hypertrophy[J].J Am Coll Cardiol,2017,70(14)∶1829-1831.
[8]Taylor D,Harrison A,Powers D.An artificial neural network system to identify alleles in reference electropherograms[J].Forensic Sci Int Genet,2017,30∶114-126.
[9]Peixoto LA,Bhering LL,Cruz CD.Artificial neural networks reveal efficiency in genetic value prediction[J].Genet Mol Res,2015,14(2)∶6796-807.
[10]Lang RM,Badano LP,Mor-Avi V,et al.Recommendations for cardiac chamber quantification by echocardiography in adults∶an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging[J].J Am Soc E-chocardiogr,2015,28∶1-39.
[11]Germano G.Electrocardiographic signs of left ventricular hypertrophy in obese patients∶what criteria should be used?[J].High Blood Press Cardiovasc Prev,2014,22(1)∶5-9.
[12]Hopkins CB,Suleman J,Cook C.An artificial neural network for the electrocardiographic diagnosis of left ventricular hypertrophy[J].Crit Rev Biomed Eng,2000,28(3-4)∶435-438.
[13]Satoh S,Omura S,Inoue H,et al.Gender differences in factors influencing electrocardiographic findings of left ventricular hypertrophy in severe aortic stenosis[J].Heart Vessels,2014,29(5)∶659-666.
[14]Cuspidi C,Rescaldani M,Sala C,et al.Left-ventricular hypertrophy and obesity∶a systematic review and meta-analysis of echocardiographic studies[J].J Hypertens,2014,32(1)∶16-25.
[15]Tang ZH,Liu J,Zeng F,et al.Comparison of prediction model for cardiovascular autonomic dysfunction using artificial neural network and logistic regression analysis[J].PLo S One,2013,8∶e70571.
[16]Gu W,Vieira AR,Hoekstra RM,et al.Use of random forest to estimate population attributable fractions from a case-control study of salmonella enterica serotype enteritidis infections[J].Epidemiol Infect,2015,143∶2786-2794.
[17]余琳琳,曾少丹.窦性心律不齐在年龄层分布规律及性别差异分析[J].心电与循环,2018,37(2)∶121-123.
[18]黄进宇.慢性心力衰竭的药物治疗进展[J].心电与循环,2018,37(2)∶83-85.