基于神经网络的波型检测方法

QRS Detection Based on Neural-Network

介绍了一种基于神经网络白化匹配滤波器的 QRS波检测方法。我们用神经网络白化匹配滤波器来处理 ECG信号的低频成分 ,模拟其非线性及非稳态的特性。处理后的信号中含有 ECG中大部分高频成分 ,让其通过一线性匹配滤波器来检测 QRS波及其位置。对于大噪声的 ECG信号 ,在匹配滤波器后加差分滤波 ,取平方及滑动平均等处理 ,提高检测正确率。使用这种方法我们对 MIT/BIH心电信号数据库中噪声比较大的 10 5号数据进行的处理 ,检测正确率为 99.2 %。作为对比 ,用数字带通滤波器检测 ,正确率为 97.8%。

In this paper, we have developed an adaptive matched filtering algorithm based upon an artificial network (ANN) for QRS detection. We used an ANN adaptive whitening filter to model the lower frequencies of the ECG signals which are inherently nonlinear and non stationary. The residual signal which contained mostly higher frequency QRS complex energy was then passed through a linear matched filter to detect the location of the QRS complex. The results demonstrate that this ANN whitenting filter is very effective for removing the time varying, nonlinear noise characteristic of ECG signals. With this novel approach, the detection rate for a very noisy patient record in the MIT/BIH arrhythmia database is 99.2%, which compares favorably to the 97.8% achieved with a band pass filtering method.