作者：农汉彪, 曾巧妮

作者单位：百色学院，广西 百色 533000

关键词：心电信号;基线漂移;聚类分析;中值滤波;DBSCAN

摘要：

针对心电信号基线漂移信号拟合点难于提取问题，根据心电信号统计特性提出基于聚类分析的方法来消除心电信号基线漂移。信号数据通过最优分段后，统计分段数据的方差和峭度，并进行归一化，再进行基于DBSCAN的聚类分析，选取分布在零点附近的分类所对应的数据段的中值为拟合点。经过仿真数据和实测数据的验证分析并与常用方法比较，结果表明所提出的方法有效且性能优于其他常用方法。在均方根误差上仅为中值滤波的31.5%，消除后的信噪比比其他方法高出10 dB。

Baseline wander removing of ECG signal based on clustering analysis
NONG Hanbiao, ZENG Qiaoni
Baise University, Baise 533000, China
Abstract: Aiming at the problem that it is difficult to extract the fitting points of ECG baseline drift, a method based on cluster analysis is proposed to eliminate the baseline wander of ECG, according to the statistical characteristics of ECG signal. After dividing the signal data into segments with an optimal block size, the variance and kurtosis of the data segments are counted and normalized, and then cluster analysis based on DBSCAN algorithm, then select the median value of data segments which is distributed near the zero point as the fitting point. Through verification analysis of the simulation data and measured data, and comparing with the common methods, the results show that the proposed method is effective and better than other common methods. The root mean square error is only 31.5% of the median filter, and the SNR is 10 dB higher than other methods.
Keywords: electrocardiogram (ECG) signal;baseline wander;clustering analysis;median filter;DBSCAN
2022, 48(9):22-28  收稿日期: 2021-03-10;收到修改稿日期: 2021-06-01
基金项目:
作者简介: 农汉彪（1982-），男，广西百色市人，工程师，博士，研究方向为信号分析与处理
参考文献
[1] 林金朝, 刘乐乐, 李国权, 等. 基于改进EEMD的心电信号基线漂移消除方法[J]. 数据采集与处理, 2018, 33(5): 880-890
[2] YAO L, PAN Z. A new method based CEEMDAN for removal of baseline wander and powerline interference in ECG signals[J]. Optik (Stuttgart), 2020, 223: 165566
[3] 刘春, 谢皓, 肖奕霖, 等. EWT算法在ECG信号滤波中的研究[J]. 电子测量与仪器学报, 2017, 31(11): 1835-1842
[4] BODA S, MAHA M, DUTTA P K. A hybrid method for removal of power line interference and baseline wander in ECG signals using EMD and EWT[J]. Biomedical Signal Processing and Control, 2021, 67: 102466
[5] 崔善政, 郭艳珍, 梁钊, 等. 变分模态分解在去除心电图信号基线漂移中的应用[J]. 电子测量与仪器学报, 2018, 32(2): 167-171
[6] SINGHAL A, SINGH P, FATIMAH B, et al. An efficient removal of power-line interference and baseline wander from ECG signals by employing Fourier decomposition technique[J]. Biomedical Signal Processing and Control, 2020, 57: 101741
[7] ROMERO F P, PIÑOL D C, VÁZQUEZ C R. Deep Filter: An ECG baseline wander removal filter using deep learning techniques[J]. Biomedical Signal Processing and Control, 2021, 70: 102992
[8] 万相奎, 唐文普, 张赖, 等. 改进的三次样条插值心电基线漂移滤波法[J]. 生物医学工程学杂志, 2016, 33(2): 227-231
[9] 徐万松, 陈天武. 基于形态学消除心电信号基线漂移方法的研究[J]. 中国医学工程, 2019, 27(9): 8-12