作者：董金陇, 杨录, 刘康驰, 高斌
作者单位：中北大学信息与通信工程学院 电子测试国家重点实验室，山西 太原 030051
关键词：生物医学工程学;改进EMD算法;单周期对比;脉搏信号伪基线
摘要：
针对脉搏信号中伪基线的去除问题，提出基于单周期对比的EMD改进算法。结合相关系数分析方法，判定并去除IMF分量中的伪分量；通过原始脉搏信号的最大单周期与IMF分量的最小单周期进行对比，分离出属于脉搏信号伪基线的低频IMF分量；最后重构其余的IMF分量得到去除伪基线的脉搏信号。该方法分离出的脉搏信号和伪基线信号与理想信号的相关系数分别为0.9997和0.9602；与EMD经验法相比信噪比（SNR）提升约两倍，均方根误差（RMSE）降低两个数量级。实验结果表明：该方法有效去除脉搏信号伪基线，且能较好保留脉搏信号的特征点，对后续脉搏信号的特征提取与准确处理有重要意义。
Research on pulse signal pseudo-baseline removal method
DONG Jinlong, YANG Lu, LIU Kangchi, GAO Bin
State Key Laboratory of Electrical Testing, Institute of Information and Communications Engineering, North University of China, Taiyuan 030051, China
Abstract: Aiming at the problem of removing pseudo-baseline in pulse signal, an improved EMD algorithm based on single cycle comparison is proposed. Combining with the correlation coefficient analysis method, the pseudo components in IMF component are determined and removed. The low-frequency IMF component belonging to the pseudo-baseline of pulse signal is separated by comparing the maximum single period of original pulse signal with the minimum single period of IMF component. Finally, the remaining IMF components are reconstructed to obtain the pulse signal with pseudo-baseline removed. The correlation coefficients of pulse signal and pseudo-baseline signal separated by the method are 0.9997 and 0.9602, respectively. Compared with EMD empirical method, SNR is improved by about two times and RMSE is reduced by two orders of magnitude. The experimental results show that this method can effectively remove the pseudo-baseline of pulse signal and keep the feature points of pulse signal, which is of great significance for the feature extraction and accurate processing of subsequent pulse signals.
Keywords: biomedical engineering;improved EMD algorithm;single period comparison;pulse signal pseudo-baseline
2021, 47(4):19-25 收稿日期: 2020-06-18;收到修改稿日期: 2020-08-07
基金项目: 山西省自然科学基金资助项目（201901D111161）
作者简介: 董金陇（1996-），男，山西垣曲县人，硕士研究生，专业方向为生物信号处理
参考文献
[1] 张兢, 曾建梅, 杨超, 等. 改进的经验模态分解算法对脉搏信号的处理[J]. 自然科学, 2014, 28(8): 71-75
[2] 李霞, 陆琤, 刘星星. 脉搏波分析研究进展[J]. 北京生物医学工程, 2015, 34(5): 544-547
[3] 行鸿彦, 许瑞庆. 基于经验模态分解的脉搏信号去噪[J]. 计算机应用与软件, 2009, 26(8): 156-158
[4] 王杰华, 夏海燕, 孙万捷, 等. 基于EMD和SVD的光电容积脉搏波信号去噪方法[J]. 现代电子技术, 2018, 41(4): 65-69,74
[5] 朱菊华, 杨新, 李俊. 一种改进的自适应保细节中值滤波算法[J]. 计算机工程与应用, 2001(3): 93-95
[6] 龙雨, 朱希安. 基于小波变换的容积脉搏波信号去噪研究[J]. 科技通报, 2016, 32(1): 141-144
[7] 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
[8] SORNMO L. Time-varying filtering for removal of baseline wander in exercise ECGs[C]//Computers in Cardiology. IEEE, 1991.
[9] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903-995
[10] WENG B, BLANCO-VELASCO M, BARNER K E. Baseline wander correction in ECG by the empirical mode decomposition[C]//Bioengineering Conference. IEEE, 2006.
[11] 韩庆阳, 李丙玉, 王晓东. 一种同时消除脉搏波信号中呼吸基线漂移和高频噪声的方法[J]. 中国医学物理学杂志, 2014, 31(2): 4801-4805
[12] 苏志刚, 吕江波, 郝敬堂, 等. 改进的经验模态分解法去除脉搏波基线漂移[J]. 计算机测量与控制, 2018, 26(7): 229-232,237
[13] RILLING G, FLANDRIN P, GONCALVES P. On empirical mode decomposition and its algorithms[C]//IEEE-EURASIP Workshop on Nonlinear Signal and Image Processing, 2003.
[14] 郭树言, 王黎明, 牛晓东, 等. 基于EMD的自适应ECG信号基线漂移消除方法研究[J]. 中国医学物理学杂志, 2020, 37(3): 342-349
[15] 林丽, 余轮. 基于相关系数的EMD改进算法[J]. 计算机与数字工程, 2008, 36(12): 28-29,38
