基于小波双阈值滤波算法的膈肌肌电信号去噪方法-《中国医学物理学杂志》

文章信息/Info

Title:: Denoising of diaphragmatic electromyogram signals using dual-threshold filtering algorithm

作者:: 李双妙1; 李志为1; 张涵1; 张建恒2; 1.华南师范大学电子与信息工程学院，广东广州 510631； 2.广州医科大学附属第一医院呼吸内科，广东广州 520120

Author(s):: LI Shuangmiao1; LI Zhiwei1; ZHANG Han1; ZHANG Jianheng2; 1. School of Electrical and Information Engineering, South China Normal University, Guangzhou 510631, China 2. Respiratory Department, the First Affiliated Hospital of Guangzhou Medical University, Guangzhou 520120, China

Keywords:: Keywords: diaphragmatic electromyogram electrocardiogram contamination wavelet coefficient dual-threshold filtering

摘要:: 针对传统去心电信号（ECG）干扰算法处理异常ECG信号效果欠佳的问题，提出一种基于小波双阈值滤波算法的膈肌肌电（EMGdi）信号降噪方法。该方法以小波系数为基础，构造QRS群波中R峰的检测阈值，进而准确定位R峰位置。在此基础上，采用干扰区间两侧小波平均能量作为干扰区间阈值计算依据，对干扰区间进行平滑处理。通过临床EMGdi信号去ECG干扰实验，结果表明本文方法在去干扰性能上优于传统算法，尤其针对弱EMGdi信号去干扰优势更加明显。

Abstract:: Abstract: Given that the traditional algorithms for elimination of electrocardiogram (ECG) contamination have a poor performance on abnormal ECG signals, a denoising algorithm for diaphragmatic electromyogram (EMGdi) signals based on wavelet dual-threshold filtering is presented. The method constructs the detection threshold of R peak in QRS group wave based on wavelet coefficient for accurately locating the position of the R peak, and takes the average energy on both sides of one interference range as the threshold of this interference range for eliminating ECG contamination. Experimental results of eliminating ECG contamination from clinical EMGdi signals show that the proposed algorithm surpasses the traditional algorithms in interference removal, especially for weak EMGdi signals.