摘要
研究适于远程心电诊断,基于ARMA模型的高精度的心电信号(ECG)直接分类方法.利用ARMA模型系数作为特征对ECG信号进行分类和压缩.在对信号特征分类时,采用了BP神经网络分类的方法.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(VT)、室上性心动过速(SVT)和心室纤维性颤动(VF)各300个样本信号进行了测试,获得了96.51%~98.38%的分类精度.
A direct classification method with high accuracy that is suitable for telediagnosis of electrocardiogram (ECG) signals is studied. A simpler autoregressive-moving average (ARMA) based data fusion technique is proposed to classify and compress ECGs in this paper. ARMA coefficients were used to represent the ECGs. The data in the analysis included normal sinus rhythm (NSR), premature ventricular contraction (PVC), atrial premature contraction (APC), ventricular tachycardia (VT), ventricular fibrillation (VF), and superventricular tachycardia(SVT). Three hundred data samples from each of them have been utilized to classify and test. The accuracy of detecting these arrhythmias proposed were 96.51%-98.38% using BPNN.
出处
《科技通报》
北大核心
2004年第6期569-572,共4页
Bulletin of Science and Technology