基于小波变换的心电图QRS波群检测方法研究

Research of Wavelet Transformation Based Detection of QRS Complex in ECG

本文就心电图信号的QRS波群检测提出了一种基于小波变换的信号特征提取方法 ,此方法对心电信号中QRS波群的时变特性及几种常见的心电干扰具有较强的鲁棒性。文中我们采用两种不同性质的小波为母小波对含有噪声污染的心电信号进行多尺度的小波分解 ,在没有预先消噪处理的情况下 ,较为准确、快速地检测出QRS波群的信息 ,并且以国际上广泛承认的心电数据库MIT BIH中的记录对算法进行检验。

In this paper is presented a technique of feature extraction for QRS complex detection by using wavelet transformation based multi scales decomposition, which is robust for time varying characteristics of QRS complex and for some common kinds of EC interference. In this paper, we present a method of using two different kinds of wavelets as mother wavelet, to decompose noisy ECG signals of multi scales without pre processing of noise reduction. The results obtained were rather accurate. It was fast to detect the information in QRS waves. The algorithm was verified with records contained in MIT BIH EC database, which is widely and internationally accepted.