Advanced simple and robust technique for detecting localized early muscle fatigue

Rubana H. Chowdhury, Mamun B.I. Reaz, Muhammad L. Ali, M. Sulvia

Research output: Contribution to journalArticle

Abstract

The purpose of the study was to establish a simple and robust technique that can be used to predict fatigued and non-fatigued muscle by surface electromyography (sEMG). Muscle fatigue is often caused by unhealthy work practices and recovery may take longer when the fatigue tolerable limit is exceeded. Instantly, it is difficult for the person to realize that there is a complicacy in the muscle (symptoms such as a sore throat, muscle or joint pain, and headache) before it is visible. In this study, it is shown that by using the proposed fatigue model, a near identification of the presence of fatigue is possible. Biceps curl exercises with a dumbbell (7 kg) were used to develop fatigue in the biceps brachii muscle of the upper arm. In this work, four different feature sets were considered from sEMG signals and they were considered as an input to the dimensionality reduction method PCA and then the classifier SVM. sEMG signals were recorded from a total of 34 subjects (from 21 to 32 years). A classification success rate of 94.12% was achieved by using only two time-frequency features (average instantaneous frequency after Emphirical mode decomposition and total energy by using Wavelet packet transforms) as input to the proposed model. These results also show that fewer features can be used as an alternative to more features for differentiating between fatigued and healthy muscle using the proposed method.

Original languageEnglish
Pages (from-to)3536-3550
Number of pages15
JournalJournal of Engineering Science and Technology
Volume14
Issue number6
Publication statusPublished - 1 Jan 2019

Fingerprint

Muscle
Fatigue of materials
Electromyography
Identification (control systems)
Classifiers
Decomposition
Recovery

Keywords

  • Emphirical mode decomposition
  • Muscle fatigue
  • Surface electromyography (sEMG)
  • Wavelet transform

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Advanced simple and robust technique for detecting localized early muscle fatigue. / Chowdhury, Rubana H.; Reaz, Mamun B.I.; Ali, Muhammad L.; Sulvia, M.

In: Journal of Engineering Science and Technology, Vol. 14, No. 6, 01.01.2019, p. 3536-3550.

Research output: Contribution to journalArticle

Chowdhury, Rubana H. ; Reaz, Mamun B.I. ; Ali, Muhammad L. ; Sulvia, M. / Advanced simple and robust technique for detecting localized early muscle fatigue. In: Journal of Engineering Science and Technology. 2019 ; Vol. 14, No. 6. pp. 3536-3550.
@article{eb6b73c8855b451f8decef107907131a,
title = "Advanced simple and robust technique for detecting localized early muscle fatigue",
abstract = "The purpose of the study was to establish a simple and robust technique that can be used to predict fatigued and non-fatigued muscle by surface electromyography (sEMG). Muscle fatigue is often caused by unhealthy work practices and recovery may take longer when the fatigue tolerable limit is exceeded. Instantly, it is difficult for the person to realize that there is a complicacy in the muscle (symptoms such as a sore throat, muscle or joint pain, and headache) before it is visible. In this study, it is shown that by using the proposed fatigue model, a near identification of the presence of fatigue is possible. Biceps curl exercises with a dumbbell (7 kg) were used to develop fatigue in the biceps brachii muscle of the upper arm. In this work, four different feature sets were considered from sEMG signals and they were considered as an input to the dimensionality reduction method PCA and then the classifier SVM. sEMG signals were recorded from a total of 34 subjects (from 21 to 32 years). A classification success rate of 94.12{\%} was achieved by using only two time-frequency features (average instantaneous frequency after Emphirical mode decomposition and total energy by using Wavelet packet transforms) as input to the proposed model. These results also show that fewer features can be used as an alternative to more features for differentiating between fatigued and healthy muscle using the proposed method.",
keywords = "Emphirical mode decomposition, Muscle fatigue, Surface electromyography (sEMG), Wavelet transform",
author = "Chowdhury, {Rubana H.} and Reaz, {Mamun B.I.} and Ali, {Muhammad L.} and M. Sulvia",
year = "2019",
month = "1",
day = "1",
language = "English",
volume = "14",
pages = "3536--3550",
journal = "Journal of Engineering Science and Technology",
issn = "1823-4690",
publisher = "Taylor's University College",
number = "6",

}

TY - JOUR

T1 - Advanced simple and robust technique for detecting localized early muscle fatigue

AU - Chowdhury, Rubana H.

AU - Reaz, Mamun B.I.

AU - Ali, Muhammad L.

AU - Sulvia, M.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The purpose of the study was to establish a simple and robust technique that can be used to predict fatigued and non-fatigued muscle by surface electromyography (sEMG). Muscle fatigue is often caused by unhealthy work practices and recovery may take longer when the fatigue tolerable limit is exceeded. Instantly, it is difficult for the person to realize that there is a complicacy in the muscle (symptoms such as a sore throat, muscle or joint pain, and headache) before it is visible. In this study, it is shown that by using the proposed fatigue model, a near identification of the presence of fatigue is possible. Biceps curl exercises with a dumbbell (7 kg) were used to develop fatigue in the biceps brachii muscle of the upper arm. In this work, four different feature sets were considered from sEMG signals and they were considered as an input to the dimensionality reduction method PCA and then the classifier SVM. sEMG signals were recorded from a total of 34 subjects (from 21 to 32 years). A classification success rate of 94.12% was achieved by using only two time-frequency features (average instantaneous frequency after Emphirical mode decomposition and total energy by using Wavelet packet transforms) as input to the proposed model. These results also show that fewer features can be used as an alternative to more features for differentiating between fatigued and healthy muscle using the proposed method.

AB - The purpose of the study was to establish a simple and robust technique that can be used to predict fatigued and non-fatigued muscle by surface electromyography (sEMG). Muscle fatigue is often caused by unhealthy work practices and recovery may take longer when the fatigue tolerable limit is exceeded. Instantly, it is difficult for the person to realize that there is a complicacy in the muscle (symptoms such as a sore throat, muscle or joint pain, and headache) before it is visible. In this study, it is shown that by using the proposed fatigue model, a near identification of the presence of fatigue is possible. Biceps curl exercises with a dumbbell (7 kg) were used to develop fatigue in the biceps brachii muscle of the upper arm. In this work, four different feature sets were considered from sEMG signals and they were considered as an input to the dimensionality reduction method PCA and then the classifier SVM. sEMG signals were recorded from a total of 34 subjects (from 21 to 32 years). A classification success rate of 94.12% was achieved by using only two time-frequency features (average instantaneous frequency after Emphirical mode decomposition and total energy by using Wavelet packet transforms) as input to the proposed model. These results also show that fewer features can be used as an alternative to more features for differentiating between fatigued and healthy muscle using the proposed method.

KW - Emphirical mode decomposition

KW - Muscle fatigue

KW - Surface electromyography (sEMG)

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=85077381887&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077381887&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85077381887

VL - 14

SP - 3536

EP - 3550

JO - Journal of Engineering Science and Technology

JF - Journal of Engineering Science and Technology

SN - 1823-4690

IS - 6

ER -