%0 Journal Article
%T 基于心电特征参数的心肌梗死疾病辅助诊断模型的建立*<br>Establishment of myocardial infarction disease-diagnosis model based on ECG characteristic parameters
%A 张清丽
%A 苏士美
%A 尹咪咪
%A 张建华
%A 刘
%A 莹
%J 郑州大学学报（医学版）
%D 2017
%R 10.13705/j.issn.1671-6825.2017.02.011
%X 目的:建立基于心电特征参数的心肌梗死疾病辅助诊断模型。方法:取PTB数据库中的158例心肌梗死患者为病例组,90例健康志愿者为对照组,提取这2组的ECG V5导联信号波形并进行预处理,用小波变换结合窗口函数的方法提取11个心电特征参数,采用独立样本t检验和精确概率法筛选特征参数,并进行归一化处理,建立logistic回归模型和支持向量机模型并比较其性能。结果:Logistic回归模型和支持向量机模型的诊断准确率分别为95.1%和96.0%。结论:Logistic回归模型和支持向量机模型对心肌梗死的分类诊断均具有重要的理论和临床价值。<br>Aim: To establish a myocardial infarction disease-diagnosis model based on ECG characteristic parameters.Methods: A total of 158 cases of myocardial infarction(case group)and 90 healthy volunteers(control group)in PTB database were chosen. ECG V5 lead signal waveforms of the 2 groups were extracted and preprocessed, 11 ECG characteristic parameters were extracted using wavelet transform combined with window function,among which, characteristic parameters were selected using independent sample t test and exact probability method, and normalized, and then the mathematical models were established based on logistic regression model and support vector machine(SVM)model, finally,the performance of the 2 models was compared.Results: The diagnostic accuracy of the logistic regression model was 95.1%,and that of the SVM model was 96.0%.Conclusion: Logistic regression model and SVM model are both of great value in classification diagnosis of myocardial infarction
%K 心肌梗死
%K 小波变换
%K logistic回归模型
%K 支持向量机模型&lt
%K br&gt
%K myocardial infarction
%K wavelet transform
%K logistic regression model
%K support vector machine model
%U http://jms.zzu.edu.cn/oa/darticle.aspx?type=view&id=201702011