Filtering of surface EMG using ensemble empirical mode decomposition

doi:10.1016/j.medengphy.2012.10.009

Medical Engineering & Physics

Volume 35, Issue 4, April 2013, Pages 537-542

https://doi.org/10.1016/j.medengphy.2012.10.009 Get rights and content

Abstract

Surface electromyogram (EMG) is often corrupted by three types of noises, i.e. power line interference (PLI), white Gaussian noise (WGN), and baseline wandering (BW). A novel framework based primarily on empirical mode decomposition (EMD) was developed to reduce all the three noise contaminations from surface EMG. In addition to regular EMD, the ensemble EMD (EEMD) was also examined for surface EMG denoising. The advantages of the EMD based methods were demonstrated by comparing them with the traditional digital filters, using signals derived from our routine electrode array surface EMG recordings. The experimental results demonstrated that the EMD based methods achieved better performance than the conventional digital filters, especially when the signal to noise ratio of the processed signal was low. Among all the examined methods, the EEMD based approach achieved the best surface EMG denoising performance.

Introduction

Electromyography (EMG) signal is electrical manifestation of a contracting muscle. The acquisition of a clean EMG is a prerequisite for an appropriate interpretation and application of the signal. Surface EMG signals, like most of the electrophysiological measurements, are frequently corrupted with three categories of noise [1], i.e. power line interference (PLI), white Gaussian noise (WGN), and motion artifact or baseline wandering (BW). In particular, efforts in multiple-channel surface EMG recording have been developed in recent years using high density electrode arrays. Due to the large number of the recording electrodes and their tiny electrode-skin contact area, noise contamination emerges as an even more challenging problem. Situations are sometimes encountered where the signal to noise ratio (SNR) is poor in a few channels.

Noise contamination may compromise the efficacy of the EMG signal processing. Thus, several methods have been proposed to reduce the noise from surface EMG signals, among which the most simple and cost-efficient solution is to use conventional digital filters. Although such filters substantially reduce the noise, they also attenuate the EMG signal due to spectral overlap of the noise and the surface EMG. It, therefore, remains a challenging problem to reduce noise without distortion of the useful EMG signals, which might require more advanced methods than the conventional digital filters. For example, adaptive or nonlinear filtering has been proposed to reduce the noise contamination while minimally sacrificing sections of the surface EMG signal [2], [3]. On the other hand, taking advantages of the time-frequency resolution of the wavelet transform, wavelet thresholding has been used for noise reduction in various electrophysiological signals, including surface EMG [4]. As an alternative tool for processing nonlinear and non-stationary signals, the empirical mode decomposition (EMD) can also be used for de noising and conditioning of the electrophysiological signals [6], [7], [8], [9], [10], [11], primarily using a similar approach to the wavelet based method. In contrast to the wavelet transform, the EMD decomposes a signal into a series of intrinsic mode functions (IMFs), which are zero-mean, amplitude- and frequency-modulated (AM–FM) time series representing oscillations within the processed signal [12]. The EMD is implemented via a sifting process, which is a purely data-driven, signal-dependent iterative procedure and makes no assumption about the original signal.

In this study, a novel framework based on EMD was developed to eliminate noise contamination in surface EMG. In contrast to most of the previous de noising methods that solely target a specific category of noise, a major feature of the current study is the primary reliance on the EMD for dealing with three different types of noise often present in surface EMG, particularly in high density surface electrode array recordings. Moreover, the ensemble EMD (EEMD) was used to overcome the limitation of the mode mixing routinely induced by the regular EMD [13], thus further improving the surface EMG denoising performance. The advantages of the EMD or EEMD based methods were demonstrated by comparing them with the traditional digital filters, using signals derived from our routine electrode array EMG recordings.

Section snippets

Empirical mode decomposition (EMD)

The EMD is designed to adaptively decompose a time-series signal s(t),1 ≤ t ≤ L, into a sum of intrinsic mode functions (IMFs) c_i(t),1 ≤ i ≤ N, $s (t) = \sum_{i = 1}^{N} c_{i} (t) + T_{N} (t)$ where t is the time index, N denotes the number of IMFs, and r_N(t) is the residual signal [12]. An IMF is defined as a function satisfying two conditions, i.e. (1) the difference between the number of extrema (including both the local maxima and minima) and zero-crossings in the time-series must be no more than one; (2) the mean value of the

Results

The typical results from EMD and EEMD are presented in Fig. 1, where a relatively clean surface EMG signal contaminated primarily by PLI was decomposed into IMFs by EMD and EEMD, respectively. The IMF distribution indicates that the lower-order IMFs contain relatively high-frequency components while the higher-order IMFs contain relatively low-frequency components. Such a distribution is similar to filter bank but not exactly band-restricted, due to the fact that IMF was adaptively extracted

Discussions and conclusion

An EMD/EEMD-based IMF filtering framework was presented in this study to remove all the three common categories of noise (i.e. PLI, WGN, and BW) from surface EMG recordings. EMD is a powerful tool for processing nonlinear and nonstationary physiological signals with complex temporal-spatial structure. It can adaptively separate an original complex signal into a set of IMFs with different oscillation levels, thus offering an interpretation of the dynamic processes underlying the system

Conflict of interest

None declared.

Funding

This work was supported in part by the National Institute on Disability and Rehabilitation Research of the U.S. Department of Education under Grant H133G090093, in part by the National Institutes of Health of the U.S. Department of Health and Human Services under Grant 1R21NS075463 and Grant 2R24HD050821, in part by the Searle-Chicago Community Trust Foundation, in part by the Davee Research Foundation, in part by the National Natural Science Foundation of China under Grant 81271658, and in

Ethical approval

The study was approved by the Institutional Review Board of Northwestern University, Chicago, IL, USA (reference number: STU00014437).

References (17)

E.A. Clancy et al.
Sampling, noise-reduction and amplitude estimation issues in surface electromyography
J Electromyogr Kinesiol
(2002)
G. Aschero et al.
Denoising of surface EMG with a modified Wiener filtering approach
J Electromyogr Kinesiol
(2010)
A.O. Andrade et al.
EMG signal filtering based on empirical mode decomposition
Biomed Signal Process Control
(2006)
M. Blanco-Velasco et al.
ECG signal denoising and baseline wander correction based on the empirical mode decomposition
Comput Biol Med
(2008)
R.G.T. Mello et al.
Digital Butterworth filter for subtracting noise from low magnitude surface electromyogram
Comput Methods Prog Biomed
(2007)
R. Quian Quiroga
What is the real shape of extracellular spikes?
J Neurosci Methods
(2009)
T.W. Beck et al.
A comparison of adaptive and notch filtering for removing electromagnetic noise from monopolar surface electromyographic signals
Physiol Meas
(2009)
P. Zhou et al.
An evaluation of the utility and limitations of counting motor unit action potentials in the surface electromyogram
J Neural Eng
(2004)

There are more references available in the full text version of this article.

Cited by (86)

Automated detection and removal of artifacts from sEMG signals based on fuzzy inference system and signal decomposition methods
2024, Biomedical Signal Processing and Control
Surface electromyography (sEMG) signal quality decreases when it is contaminated by different types of artifacts. Detection and removal of the contaminants from sEMG signals were a major and interesting task for several research works in recent years. In this paper, a novel automatic approach was designed for the detection and removal of artifacts from sEMG signals. The proposed method consists of a combined model that based on a fuzzy inference system (FIS), a signal decomposition method, and a denoising technique. Based on three spectral statistical features, which are calculated in frequency domain using power spectral density: composite multiscale entropy (CMSE), kurtosis (K), and skewness (S), FIS detects the artifactual epochs of the sEMG signal. Thereafter, those epochs are decomposed into their components by a signal decomposition. In this work, we have tested the performance of two decomposition methods namely stationary wavelet transform (SWT) and variational mode decomposition (VMD). A second FIS is used to select the contaminated components. Depending on the type of artifact, two procedures are considered, the first aims to denoise motion artifact (MOA), electrocardiography interference (ECG) and power line interference (PLI), whereas the second procedure aims to eliminate the additive white gaussian noise (AWGN). Finally, the denoised signal is reconstructed. The evaluation performance of the proposed method is compared with the performance of some existing methods using the following metrics: SNR improvement $(Δ S N R)$ , the reduction in artifact $(λ (%))$ , improvement in signal to noise and distortion ratio $(Δ S N D R)$ and improvement in the root mean square error $(Δ R M S E)$ . The results emphazed the superiority of our method over others in terms of all the used metrics and computational time.
A biosignal quality assessment framework for high-density sEMG decomposition
2024, Biomedical Signal Processing and Control
Surface electromyography (sEMG) signal is superimposed by hundreds of motor unit (MU) discharge waveforms during muscle contraction. With the development of high-density sEMG (HD-sEMG) acquisition techniques, exploring the motoneuron discharge events, which is called sEMG decomposition, has shown advantages over traditional analysis. However, experimentally recorded HD-sEMG signals are nonetheless inevitably contaminated by several kinds of noise, including baseline noise (BLN), electrocardiogram interference (ECI), powerline noise (PLN), and white Gaussian noise (WGN). These noises directly degrade the signal quality and affect the performance of decomposition, but the effect of noise type on the performance of sEMG decomposition remains to be determined. To evaluate the impact of noise, this paper presents a biosignal quality assessment framework for HD-sEMG decomposition. Firstly, the four most prominent types of noise were analyzed via simulation modeling and experimental verification. The results showed that WGN was the dominant noise factor influencing sEMG decomposition, while the other types of noise had no significant influence. Secondly, a signal-to-white Gaussian noise ratio (SWGR) estimation method based on canonical correlation analysis (CCA) was proposed, and SWGR was further used as a metric to evaluate the biosignal quality for HD-sEMG decomposition. The experimental results indicated that SWGR accurately ( $R^{2} \geq 0.72$ for decomposed MU number) reflected the performance of the sEMG decomposition. The outcomes provide guidance for the preprocessing of HD-sEMG with regard to decomposing MU discharge, potentially paving the way for a non-invasive neural interface.
Automatic selection of IMFs to denoise the sEMG signals using EMD
2023, Journal of Electromyography and Kinesiology
Surface Electromyography (sEMG) signals are muscle activation signals, which has applications in muscle diagnosis, rehabilitation, prosthetics, and speech etc. However, they are known to be affected by noises such as Power Line Interference (PLI), motion artifacts etc. Currently, Empirical Mode Decomposition (EMD) and its modifications such as Ensemble EMD (EEMD), and Complementary EEMD (CEEMD) are used to decompose EMG into a series of Intrinsic Mode Functions (IMFs). The denoised EMG can be obtained from the selected IMFs. Statistical methods are used to select the signal dominant IMFs to reconstruct the denoised signal. In this work, a novel procedure is proposed to automatically separate noisy IMFs from the original sEMG signal. For this purpose, Permutation Entropy (PE) is employed in EEMD sifting process called Partly EEMD (PEEMD), to separate the noisy IMFs from the original sEMG signal according to the preset PE threshold. PEEMD decomposes the original signal into various modes according to a preset PE threshold and the denoised signal is reconstructed from resultant IMFs. The PEEMD denoising procedure is applied on the experimental sEMG data collected from eight subjects, that include six various upper limb movement classes. The proposed denoising procedure achieved an improved denoising performance in comparison with EMD, EEMD, and CEEMD. An alternate measure called Sample Entropy (SE) is also used in place of PE, for the automated sifting process as a comparison. Signal to Noise Ratio (SNR), Root Mean Square Error (RMSE), and Reconstruction Error (RE) parameters are used to evaluate the denoising performance. The results, averaged across eight subjects, demonstrate that the proposed denoising procedure outperforms the state-of-the-art EMD techniques in terms of these performance measures on the experimentally collected sEMG data samples.
Design of robust adaptive Volterra noise mitigation architecture for sEMG signals using metaheuristic approach
2023, Expert Systems with Applications
Surface Electromyogram (sEMG) signals, like other electrophysiological measurements, get corrupted by several artefacts; much critical helpful information regarding a person’s clinical conditions may alter or be lost entirely. Therefore, the sEMG signal must be filtered to minimise such artefacts. The drive of this paper is to propose an efficient adaptive Volterra noise mitigation architecture (AVNMA) for the sEMG signal. Further, the optimal coefficients for designing the Volterra filter architecture are achieved by applying the recently proposed metaheuristic algorithm, gannet optimisation algorithm and compared to the performance of other benchmark optimisation algorithms, namely harmony search optimisation algorithm and teaching learning-based optimisation algorithm to the proposed architecture. The quantitative analysis based on the proposed architecture is measured in terms of several metrics such as mean squared error, normalised root mean square error, peak reconstruction error, mean difference, maximum error and signal-to-noise ratio at various noisy environments in the presence of additive white Gaussian noise, including artefacts such as muscle noise, baseline wandering, electrode misplacements, and electrical interference. The outcomes of experimental research ( $SNR$ = 102.236 dB and $MSE$ = 6.71E-09 for healthy-sEMG) ensure the superiority of the gannet optimisation algorithm based-AVNMA over the other metaheuristic algorithm based-AVNMAs. The proposed approach is verified parametrically and non-parametrically in a statistical sense with the help of two sample t-tests and the Mann-Whitney U test, respectively.
Feature Layer Fusion of Linear Features and Empirical Mode Decomposition of Human EMG Signal
2022, Journal of Electronic Science and Technology
Citation Excerpt :
Empirical mode decomposition is a nonlinear method for signal processing proposed by Huang et al. in 1998. The main idea is to decompose the signal to be analyzed into a signal composed of multiple intrinsic mode function (IMF) components [9,10]. IMF contains different specific time scales of signals, and its decomposition function is derived from the data itself, so the result itself ensures the non-stationarity of the information considered.
To explore the influence of the fusion of different features on recognition, this paper took the electromyogram (EMG) signals of rectus femoris under different motions (walk, step, ramp, squat, and sitting) as signals, linear features (time-domain features (variance (VAR) and root mean square (RMS)), frequency-domain features (mean frequency (MF) and mean power frequency (MPF)), and nonlinear features (EMD) of the signals were extracted. Two feature fusion algorithms, the series splicing method and complex vector method, were designed, which were verified by a double hidden layer error back propagation (BP) neural network. Results show that with the increase of the types and complexity of feature fusions, the recognition rate of the EMG signal to actions is gradually improved. When the EMG signal is used in the series splicing method, the recognition rate of time-domain + frequency-domain + empirical mode decomposition (TD + FD + EMD) splicing is the highest, and the average recognition rate is 92.32%. And this value is raised to 96.1% by using the complex vector method, and the variance of the BP system is also reduced.
A spasticity assessment method for voluntary movement using data fusion and machine learning
2021, Biomedical Signal Processing and Control
Citation Excerpt :
This paper verifies the validity and feasibility of this paper by comparing the filtering effect and classification accuracy. As shown in Fig. 5, although IIR filter is a common practice for all sEMG preprocessing, it has limited effect on sEMG with low SNR, which is consistent with the results of [52,53]. Therefore, in the process of active movement of patients, IIR filtering is not effective in improving the classification accuracy.
The assessment of spasticity under voluntary movement is helpful for the therapist to comprehensively assess the patient's dyskinesia. However, current researches focus on spasticity evaluation based on passive motion. We propose a new method for evaluating spasticity under active motion. Our method is based on the following three steps: (i) Empirical Mode Decomposition (EMD) is used to reduce involuntary movement noise in patients' active movement; (ii) Extract voluntary movement segments of each muscle for feature extract and fusion; (iii) Use machine learning methods to evaluate the degree of spasm in patients. To investigates the feasibility of the method proposed in this paper, An experiment of elbow flexion and extension against gravity is designed, and the electromyographic signal of brachioradialis (BR), biceps brachialis (BB), triceps brachialis (TB) and elbow motion data of 13 subjects were collected. We compared the classification effect of filter method, window length and classifier type. Moreover, we analyze the improvement of classification effect by data fusion. The results showed that the random forest with a window length of 256 ms had the best effect (F1-score = 0.952). Compared with the electromyographic signal (F1-score = 0.756) or motion signal only (F1-score = 0.7053), the method presented in this paper had better classification accuracy. Result demonstrated the feasibility of our method. This study can assist doctors to evaluate patients' spasmodic state under active movement, and has the application potential of wearable devices.

View all citing articles on Scopus

View full text

Technical noteFiltering of surface EMG using ensemble empirical mode decomposition

Abstract

Introduction

Section snippets

Empirical mode decomposition (EMD)

Results

Discussions and conclusion

Conflict of interest

Funding

Ethical approval

J Electromyogr Kinesiol

J Electromyogr Kinesiol

Biomed Signal Process Control

Comput Biol Med

Comput Methods Prog Biomed

J Neurosci Methods

A comparison of adaptive and notch filtering for removing electromagnetic noise from monopolar surface electromyographic signals

Physiol Meas

An evaluation of the utility and limitations of counting motor unit action potentials in the surface electromyogram

J Neural Eng

Technical note
Filtering of surface EMG using ensemble empirical mode decomposition