基于小波变换和数据融合的多导联ECG信号QRS波群精确检测算法

Wavelet Transform and Data Fusion based Algorithm for Precision Detection of QRS Complex in Multilead ECG Signals

本文首先对单导联心电图（ Ｅ Ｃ Ｇ）信号的小波变换进行假设检验，得到 Ｑ Ｒ Ｓ波群的位置和检验的虚警漏报概率，然后对 １２ 导联检测结果作峰点比对融合和决策加权融合处理，最终得到精确的 Ｑ Ｒ Ｓ位置。实验表明这种方法对各种主要病症的 Ｑ Ｒ Ｓ波群均能准确检出，正确率在 ９９９５％ 以上。

First,the localization of QRS complex in each lead is detected by using hypothesis testing on electrocardiograph(ECG) signal's wavelet transform,and the probability of false alarm and the probability of miss are also computed.Then,the peak comparison fusion and decision weightiness fusion are applied to 12 leads' decision to obtain the precision lccation of each QRS complex.The experiments show that we are able to detect the QRS complex correctly for various pathological ECGs by this method,and the correct detection rate of QRS complex can be above 99.95%.