基于参数建模法心律失常的计算机辅助自动诊断(英文)  

Recognition of Cardiac Arrhythmias Based on Parametric Modeling for Computer-Assisted Diagnosis

作　　者：李小梅[1] 李时辉[2] 

机构地区：[1]浙江科技学院基础部,杭州310023 [2]义乌工商职业技术学院计算机工程系,义乌322000

出　　处：《仪器仪表学报》2005年第1期1-7,共7页Chinese Journal of Scientific Instrument

摘　　要：利用多变量参数回归模型 (MAR)的数据融合技术 ,从双导联心电信号中提取特征 ,以实现计算机辅助自动诊断。在分类时 ,利用 MA R模型系数及其 K- L 变换作为信号特征 ,并采用了非线性二次判别函数 (QDF)分类器。利用文中方法对 MIT-BIH标准数据库中的正常窦性心律 (NSR)、心房早期收缩 (APC)、心室早期收缩 (PVC)、心室性心动过速 (V T) ,心室纤维性颤动 (V F) ,室上性心动过速 (SVT)信号进行了建模和测试。结果表明 ,融合双导联心电数据后取得了比只利用单导联心电数据更为满意的结果。The data fusion technique based on multivariate autoregressive (MAR) modeling was used to extract features from two-lead electrocardiogram (ECG) signals for computer-assisted diagnosis. MAR modeling was performed on two-lead ECG signals. MAR coefficients and K-L transformation of MAR coefficients have been used as ECG features. The cardiac arrhythmias in the analysis include normal sinus rhythm, atria premature contraction, premature ventricular contraction, ventricular tachycardia, ventricular fibrillation and superventricular ventricular tachycardia. A quadratic discriminant function (QDF) based algorithm was used to classify the beats into one of the six classes. The results show two-lead ECG based classification can improve the results than that of single-lead ECGs.

关 键 词：参数建模 心律失常 计算机辅助自动诊断 双导联ECG信号 二次判别函数 

分 类 号：R541.7[医药卫生—心血管疾病] R44[医药卫生—内科学]