基于多分辨率分析的心电图信号去噪算法

Denoising Algorithm for ECG Signals Based on Multi-Resolution Analysis

在心电信号的采集过程中 ,不可避免地会混入肌电噪声、基线漂移和 5 0Hz工频干扰等噪声信号 ,不同程度地造成了常规心电图 (ECG)特征识别方法对心电信号特征点提取的误判和漏判。根据二进小波在各个尺度上所具有的不同带宽的带通滤波特性 ,提出了二进小波去噪和特征提取的算法。所提算法能很好地抑制各种干扰 ,可消除基线漂移 ,为运用阈值法等识别算法对心电信号进行实时准确的特征识别提供了理想的波形。采用MIT/BIH数据库中的心电数据对算法进行检验 。

In the course of collecting ECG signal data, electromyographic interference, baseline drift and 50Hz power line interference will be introduced inevitably, which often makes it difficult to identify the characteristics of ECG signals to some degree by using conventional identification methods. The research indicates that the dyadic wavelet of WTS is a set of band pass filters having different frequency bands in every scale. An algorithm of characteristic identification is proposed in this paper. Various interferences can be efficiently restricted by utilizing this algorithm, which provides ideal waves for the real time and exact characteristic identification of ECG signals by utilizing various algorithms, for example, valve value technique. The algorithm is tested using MIT/BIH arrhythmia database and proved to have satisfactory results.