生物医学工程学杂志

生物医学工程学杂志

基于样本熵和个体化阈值的表面膈肌肌电信号起点检测

查看全文

膈肌是人体最主要的呼吸肌,表面膈肌肌电(sEMGdi)信号的动作区间起点检测可用于呼吸康复训练,但心电(ECG)信号的存在增加了其检测难度,故本文对此提出了基于样本熵(SampEn)和个体化阈值的起点检测方法,简称样本熵法。该方法涉及样本熵特征的提取,样本熵特征参数 wr0 的优化,个体化阈值的选取以及判断条件的设立。同时还选用其他三种常用方法与本文所提的样本熵法进行起点检测方面的比较,即利用小波变换(WT)去噪后再分别使用均方根(RMS)和能量算子(TKE)的起点检测方法,以及不做小波变换而直接使用 TKE 的起点检测方法。本文共采集 12 名健康受试者在 2 种呼吸状态下的 sEMGdi 信号,用于信号合成和算法检测。最后以误差的绝对值累加和作为评价起点检测精度的指标。最终结果表明,样本熵法在稳定性和精度两方面皆优于其他三种方法,是一种能适应个体间差异,无需提前对 sEMGdi 信号进行 ECG 信号去噪便可获得较高精度的起点检测方法,为基于 sEMGdi 信号的呼吸康复训练和实时交互提供了依据。

The diaphragm is the main respiratory muscle in the body. The onset detection of the surface diaphragmatic electromyography (sEMGdi) can be used in the respiratory rehabilitation training of the hemiparetic stroke patients, but the existence of electrocardiograph (ECG) increases the difficulty of onset detection. Therefore, a method based on sample entropy (SampEn) and individualized threshold, referred to as SampEn method, was proposed to detect onset of muscle activity in this paper, which involved the extraction of SampEn features, the optimization of the SampEn parameters w and r0, the selection of individualized threshold and the establishment of the judgment conditions. In this paper, three methods were used to compare onset detection accuracy with the SampEn method, which contained root mean square (RMS) with wavelet transform (WT), Teager-Kaiser energy operator (TKE) with wavelet transform and TKE without wavelet transform, respectively. sEMGdi signals of 12 healthy subjects in 2 different breathing ways were collected for signal synthesis and methods detection. The cumulative sum of the absolute value of error τ was used as an judgement value to evaluate the accuracy of the four methods. The results show that SampEn method can achieve higher and more stable detection precision than the other three methods, which is an onset detection method that can adapt to individual differences and achieve high detection accuracy without ECG denoising, providing a basis for sEMGdi based respiratory rehabilitation training and real time interaction.

关键词: 表面膈肌肌电; 起点检测; 样本熵; 心电干扰

Key words: surface diaphragmatic electromyography; onset detection; sample entropy; electrocardiograph

登录后 ,请手动点击刷新查看全文内容。 没有账号,
登录后 ,请手动点击刷新查看图表内容。 没有账号,
1. 刘火根, 吴爱萍, 杨毅, 等. 膈肌电活动的监测与应用. 中国呼吸与危重监护杂志, 2010, 9(4): 447-450.
2. Emayavaramban G, Amudha A. sEMG based classification of hand gestures using artificial neural network. Indian J Sci Technol, 2016, 9(35): 1-10.
3. Al Harrach M, Carriou V, Boudaoud S A, et al. Analysis of the sEMG/force relationship using HD-sEMG technique and data fusion: a simulation study. Comput Biol Med, 2017, 83: 34-47.
4. 谢平, 宋妍, 苏崇钦, 等. 脑卒中患者表面肌电信号与痉挛性肌张力关系分析. 生物医学工程学杂志, 2015, (4): 795-801.
5. 周乐, 莫鸿强, 田联房, 等. 一种基于提升小波的膈肌肌电信号在线去噪方法. 中国生物医学工程学报, 2010, 29(3): 331-335.
6. Avila A, Chang J Y. EMG onset detection and upper limb movements identification algorithm. Microsystem Technologies, 2014, 20(8/9): 1635-1640.
7. 冼莹, 王思华, 陆尧胜, 等. 膈肌肌电信号的处理和分析. 医疗卫生装备, 2005, 26(4): 22-23, 28.
8. Liu Jie, Ying Dongwen, Rymer W Z, et al. Robust muscle activity onset detection using an unsupervised electromyogram learning framework. PLoS One, 2015, 10(6): e0127990.
9. 李瑞辉, 范志坚, 赵翠莲, 等. 利用sEMG能量高斯分布特性提取动作信号的方法. 中国医疗器械杂志, 2014, (3): 177-180.
10. Li Xiaoyan, Zhou Ping, Aruin A S. Teager-Kaiser energy operation of surface EMG improves muscle activity onset detection. Ann Biomed Eng, 2007, 35(9): 1532-1538.
11. 罗国, 邓嘉仪, 陈润杰. 应用小波能量时谱去除膈肌肌电信号中的心电干扰. 中山大学研究生学刊: 自然科学.医学版, 2014, 35(4): 25-34.
12. 田絮资, 杨建, 黄力宇. 心电信号去噪的数学形态学滤波器. 计算机工程与应用, 2012, 48(2): 124-126.
13. Wu Feiyun, Tong Feng, Yang Zhi. EMGdi signal enhancement based on ICA decomposition and wavelet transform. Appl Soft Comput, 2016, 43: 561-571.
14. Zhang Xu, Zhou Ping. Sample entropy analysis of surface EMG for improved muscle activity onset detection against spurious background spikes. Journal of Electromyography and Kinesiology, 2012, 22(6): 901-907.
15. 成娟, 陈勋, 彭虎. 基于样本熵的肌电信号起始点检测研究. 电子学报, 2016, 44(2): 479-484.
16. Estrada L, Torres A, Sarlabous L, et al. Improvement in neural respiratory drive estimation from diaphragm electromyographic signals using fixed sample entropy. IEEE J Biomed Health Inform, 2016, 20(2): 476-485.
17. Estrada L, Torres A, Sarlabous L, et al. Onset and offset estimation of the neural inspiratory time in surface diaphragm electromyography: a pilot study in healthy subjects. IEEE J Biomed Health Inform, 2018, 22(1): 67-76.
18. 李营, 陈帅, 王丽. 基于近似熵和样本熵的思维脑电信号分析对比. 重庆工商大学学报:自然科学版, 2013, 30(6): 44-47, 78.
19. 刘澄玉, 赵莉娜. 熵理论发展史及其在生物医学信号分析中的作用. 北京生物医学工程, 2012, 31(5): 539-543.
20. Wolf A, Swift J B, Swinney H L, et al. Determining lyapunov exponents from a time series. Physica D, 1985, 16(3): 285-317.
21. Estrada L, Torres A, Sarlabous L, et al. Influence of parameter selection in fixed sample entropy of surface diaphragm electromyography for estimating respiratory activity. Entropy, 2017, 19(9): 460-474.