摘要
该文给出了一种利用高斯一阶导数小波变换对心电图(ECG)QRS综合波峰值进行检测的方法。该方法的基本思想是首先对ECG信号在小尺度上进行小波变换,然后利用系数中的极大极小值对来确定原始信号峰值(R峰或QS尖峰)的范围,进而返回原始信号定位峰值点。使用该方法对MIT-BIHArrhythmia Database[1]提供的30分钟的真实ECG记录进行仿真分析,结果显示该方法能较好的检测QRS波,并较为精确的定位峰值点。
A method based on the first derivative of Gaussian function wavelet for detecting the QRS crest of the electrocardiogram （ECG） is presented in this paper. The main idea of this method is as follows： firstyly, the wavelet transform on the signal of ECG is carried on in a small scale, and then, the range of the wave crest （R or QS crest） of original signal is determined by the minimax germinations of the coefficients, at last, it turns back to original signal to get the wave crest. Appling this method to the 30 - minute real ECG record provided by MIT - BIH Arrhythmia Database, the simulation results show that it can effectively detect QSR wave and get rather precise orientation of wave crest.
出处
《计算机仿真》
CSCD
2005年第12期147-150,共4页
Computer Simulation
关键词
心电图
小波变换
奇异点
Electrocardiogram
Wavelet transform
Singularity
