基于数学形态学和小波分解的QRS波群检测算法

Detection of QRS complex using mathematical morphology and wavelet transform

为了解决心电图 QRS波群检测的问题 ,研究了一种基于数学形态学及小波分解相结合来检测 QRS波群特征点的算法。数学形态学算法是基于信号局部特征的 ,可以有效地突出信号的峰谷点 ;小波分解对信号做多分辨率分解 ,可以突出信号的特征信息 ,便于 QRS波群检测。实验结果表明 :两种分析方法的结合可以有效地消除基线漂移 ,低频干扰 ,以及大 T波的影响。其 QRS波检出率达到 99.86 % 。

A electrocardiogram QRS complex detection method was developed based on mathematical morphology and wavelet transforms. The mathematical morphology operator is based on the signal shape which clarifies the signal's peaks and valleys. The wavelet transform does multi-resolution analysis on the signal which clarifies the electrocardiogram (ECG) signal characteristics to more easily detect the QRS complex. Test results show that the analytical method can remove the influence of the baseline, low frequency signals and high T complex disturbances. The result provides better analysis than conventional wavelet transforms. So that the signal-to-noise ralio can be improved, and QRS detected rate rises to 99.86%.