Next Article in Journal
Coronavirus Disease (COVID 19): Handling Challenges in Kuwait
Previous Article in Journal
Automatic Detection of Dynamic and Static Activities of the Elderly using a Wearable Sensor and Support Vector Machines
Open AccessArticlePost Publication Peer ReviewVersion 1, Original

A Comparison of Denoising Methods in Onset Determination in Medial Gastrocnemius Muscle Activations during Stance (Version 1, Original)

1
Machine Learning Algorithms Team, Glassdoor, Mill Valley, CA 94941, USA
2
Department of Physical Therapy, Crean College of Health and Behavioral Sciences, Chapman University, Orange, CA 92866, USA
3
Department of Electrical Engineering and Computer Sciences, Fowler School of Engineering, Chapman University, Orange, CA 92866, USA
4
School of Biological and Health Systems Engineering, Arizona State University, Tempe, AZ 85281, USA
*
Author to whom correspondence should be addressed.
Received: 11 May 2020 / Accepted: 14 May 2020 / Published: 3 June 2020
(This article belongs to the Section Wearable Biomedical Systems)
Peer review status: 2nd round review Read review reports

Reviewer 1 Michele Vecchio Università degli Studi di Catania, Catania, Italy Reviewer 2 Chi Hwan Lee Department of Mechanical Engineering, Purdue University, West Lafayette, IN, USA
Version 1
Original
Approved with revisions
Authors' response
Approved with revisions
Authors' response
Version 2
Revised
Approved Approved
Version 2, Revised
Published: 7 July 2020
DOI: 10.3390/sci2030053
Download Full-text PDF

Version 1, Original
Published: 3 June 2020
DOI: 10.3390/sci2020039
Download Full-text PDF
One of the most basic pieces of information gained from dynamic electromyography is accurately defining muscle action and phase timing within the gait cycle. The human gait relies on selective timing and the intensity of appropriate muscle activations for stability, loading, and progression over the supporting foot during stance, and further to advance the limb in the swing phase. A common clinical practice is utilizing a low-pass filter to denoise integrated electromyogram (EMG) signals and to determine onset and cessation events using a predefined threshold. However, the accuracy of the defining period of significant muscle activations via EMG varies with the temporal shift involved in filtering the signals; thus, the low-pass filtering method with a fixed order and cut-off frequency will introduce a time delay depending on the frequency of the signal. In order to precisely identify muscle activation and to determine the onset and cessation times of the muscles, we have explored here onset and cessation epochs with denoised EMG signals using different filter banks: the wavelet method, empirical mode decomposition (EMD) method, and ensemble empirical mode decomposition (EEMD) method. In this study, gastrocnemius muscle onset and cessation were determined in sixteen participants within two different age groups and under two different walking conditions. Low-pass filtering of integrated EMG (iEMG) signals resulted in premature onset (28% stance duration) in younger and delayed onset (38% stance duration) in older participants, showing the time-delay problem involved in this filtering method. Comparatively, the wavelet denoising approach detected onset for normal walking events most precisely, whereas the EEMD method showed the smallest onset deviation. In addition, EEMD denoised signals could further detect pre-activation onsets during a fast walking condition. A comprehensive comparison is discussed on denoising EMG signals using EMD, EEMD, and wavelet denoising in order to accurately define an onset of muscle under different walking conditions. View Full-Text
Keywords: ensemble empirical mode decomposition (EEMD); denoising; mode mixing; electromyographic (EMG) signals; filtering; wavelet method ensemble empirical mode decomposition (EEMD); denoising; mode mixing; electromyographic (EMG) signals; filtering; wavelet method
Show Figures

Figure 1

MDPI and ACS Style

Zhang, J.; Soangra, R.; E. Lockhart, T. A Comparison of Denoising Methods in Onset Determination in Medial Gastrocnemius Muscle Activations during Stance. Sci 2020, 2, 39.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1

Reviewer 1

Sent on 16 Jun 2020 by Michele Vecchio | Approved with revisions
Università degli Studi di Catania, Catania, Italy

1.A greater number of subjects would fit better this kind of study

2.Is it unclear if and how subjects’ skin was prepared before the application of the surface EMG electrodes; furthermore, did Authors follow any guideline for the correct position of them?

3. Regarding the choice of reporting only gastrocnemius medialis: Authors wrote that collected more muscles data. Probably it would have been useful to analyze at least one other muscle (for example an antagonist like the anterior tibial) in comparison.

4. In the discussion Authors mentioned the phenomenon of fiber switching into type I especially in gastrocnemius muscle in elders: a reference would be appropriate.

Best regards

Response to Reviewer 1

Sent on 12 Jul 2020 by Jian Zhang, Rahul Soangra, Thurmon E. Lockhart

First of all, we thank the reviewers for their insightful comments and have improved the manuscript, as per the suggestions. The suggestions from reviewers were great and will help us design future study. Response: We agree with the reviewer, a larger population with different age groups would have been very useful. We plan on a future study with larger sample size.

Thank you for your comment. We followed the standard skin preparation procedure. At first, we shaved the part of the body to remove hairs, then used sand paper and alcohol wipes to clean the skin surface. Sometimes it is reported to be itchy but, participants were informed about this during their IRB Consent.

Thank you for this comment. The purpose of our study was to compare effects of denoising methods on muscle onset detections during stance. We plan to consider this suggestion for our future study.

Thank you for your comment. We have added the references now. 1. Wilkinson, D.J., M. Piasecki, and P.J. Atherton, The age-related loss of skeletal muscle mass and function: Measurement and physiology of muscle fibre atrophy and muscle fibre loss in humans. Ageing Research Reviews, 2018. 47: p. 123-132. 2. Kramer, I.F., et al., Extensive Type II Muscle Fiber Atrophy in Elderly Female Hip Fracture Patients. The Journals of Gerontology: Series A, 2017. 72(10): p. 1369-1375. 3. Tieland, M., I. Trouwborst, and B.C. Clark, Skeletal muscle performance and ageing. Journal of Cachexia, Sarcopenia and Muscle, 2018. 9(1): p. 3-19.

Reviewer 2

Sent on 22 Jun 2020 by Chi Hwan Lee | Approved with revisions
Department of Mechanical Engineering, Purdue University, West Lafayette, IN, USA

The authors explored four different denoising techniques to determine the onset and cessation events of muscle activity under different conditions. It is interesting that the detection results of onset showed different behavior by denoising methods as well as walking speed. The authors explained the theory and experimental results to classify the pros and cons of various denoising methods and analyzed well. In the below, I listed several questions and comments to further improve the manuscript.

 

  1. In figure 3, authors marked onset and cessation by red stars. However, they are hard to see. Please use clearer marks.
  2. In figure 5, the authors insisted that they did not find any significant differences among older and younger participants’ muscle activation time for any denoising method. Please try to match with the y-axis scale between the two graphs in figure 5. It seems the activation time of young participants is faster than older.
  3. In the section of 2.3.6. Wavelet Denoising, the authors found that a decomposition level of 6 was adequate and would not remove EMG signal artifacts. What is the level and why was level 5 adequate? Please explain them.
  4. I agree with one of conclusion in the article that relying on only low pass filtering is not the solution to determine true muscle onset based on results. Then, the authors mentioned that “delayed onset of lower extremity muscle is an indication that the activation is stimulated by a stretch stimulus, rather than central nervous system control. Delay of gastrocnemius action until late terminal stance implies the influence of passive stretch during dorsiflexion.” I could not find that information or arguments in this article. How do you derive the conclusion? Please explain more details or use references to better understand the conclusion.
  5. The values, letters, and characters of x-axes and y-axes in all figures are too small which makes it difficult to see. Please enlarge and modify them to be clear.

 

Response to Reviewer 2

Sent on 12 Jul 2020 by Jian Zhang, Rahul Soangra, Thurmon E. Lockhart

First of all, we thank the reviewers for their insightful comments and have improved the manuscript, as per the suggestions. The suggestions from reviewers were great and will help us design future study. Response: Thank you for your comments. We have enlarged the figure now for better visibility.

Response: Thank you for this comment. We agree with the review that activation time of young participants looks faster in the figure, however, we did not find statistical differences.

Response: Thank you for this comment. The RMS of the signal is considered the ground truth of the EMG signal. Since this is common practice to evaluate the onset and cessation points using RMS of EMG signals. We conducted pilot studies for denoising at different level of EMG and further decided for the decomposition level which is most closest to the RMS signals were considered for EMG signal analysis.

Response: Thank you for the comment. We have cited the book below (page 152) Barnes, W.P., Feedback and Motor Control in Invertebrates and Vertebrates. 2012: Springer Netherlands. (Page 152)

Response: We have Stretched the figures to improve clarity.

Back to TopTop