Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron

I. Elamvazuthi, N. H.X. Duy, Zulfiqar Ali, S. W. Su, M. K.A.Ahamed Khan, S. Parasuraman

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Electromyography (EMG) signals are the measure of activity in the muscles. The aim of this study is to identify the neuromuscular disease based on EMG signals by means of classification. The neuromuscular diseases that have been identified are myopathy and neuropathy. The classification was carried out using Artificial Neural Network (ANN). There are five feature extraction techniques that were used to extract the signals such as Autoregressive (AR), Root Mean Square (RMS), Zero Crossing (ZC), Waveform length (WL) and Mean Absolute Value (MAV). A comparative analysis of these different techniques were carried out based on the results. The Multilayer Perceptron (MLP) was used for carrying out the classification.

LanguageEnglish
Pages223-228
Number of pages6
JournalProcedia Computer Science
Volume76
DOIs
Publication statusE-pub ahead of print - 29 Dec 2015

Fingerprint

Electromyography
Multilayer neural networks
Muscle
Feature extraction
Neural networks

Keywords

  • Autoregressive method (AR)
  • Classification
  • Electromyography (EMG)
  • Feature Extraction
  • Multilayer Perceptron (MLP)
  • Neuromuscular Disease
  • Root mean square (RMS)
  • Waveform length (WL) and Mean Absolute Value (MAV)
  • Zero Crossing (ZC)

Cite this

Elamvazuthi, I. ; Duy, N. H.X. ; Ali, Zulfiqar ; Su, S. W. ; Khan, M. K.A.Ahamed ; Parasuraman, S. / Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron. In: Procedia Computer Science. 2015 ; Vol. 76. pp. 223-228.
@article{e17d008f942f4d7198857266f5084d1b,
title = "Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron",
abstract = "Electromyography (EMG) signals are the measure of activity in the muscles. The aim of this study is to identify the neuromuscular disease based on EMG signals by means of classification. The neuromuscular diseases that have been identified are myopathy and neuropathy. The classification was carried out using Artificial Neural Network (ANN). There are five feature extraction techniques that were used to extract the signals such as Autoregressive (AR), Root Mean Square (RMS), Zero Crossing (ZC), Waveform length (WL) and Mean Absolute Value (MAV). A comparative analysis of these different techniques were carried out based on the results. The Multilayer Perceptron (MLP) was used for carrying out the classification.",
keywords = "Autoregressive method (AR), Classification, Electromyography (EMG), Feature Extraction, Multilayer Perceptron (MLP), Neuromuscular Disease, Root mean square (RMS), Waveform length (WL) and Mean Absolute Value (MAV), Zero Crossing (ZC)",
author = "I. Elamvazuthi and Duy, {N. H.X.} and Zulfiqar Ali and Su, {S. W.} and Khan, {M. K.A.Ahamed} and S. Parasuraman",
note = "Author started at Ulster 2019",
year = "2015",
month = "12",
day = "29",
doi = "10.1016/j.procs.2015.12.346",
language = "English",
volume = "76",
pages = "223--228",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier",

}

Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron. / Elamvazuthi, I.; Duy, N. H.X.; Ali, Zulfiqar; Su, S. W.; Khan, M. K.A.Ahamed; Parasuraman, S.

In: Procedia Computer Science, Vol. 76, 29.12.2015, p. 223-228.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron

AU - Elamvazuthi, I.

AU - Duy, N. H.X.

AU - Ali, Zulfiqar

AU - Su, S. W.

AU - Khan, M. K.A.Ahamed

AU - Parasuraman, S.

N1 - Author started at Ulster 2019

PY - 2015/12/29

Y1 - 2015/12/29

N2 - Electromyography (EMG) signals are the measure of activity in the muscles. The aim of this study is to identify the neuromuscular disease based on EMG signals by means of classification. The neuromuscular diseases that have been identified are myopathy and neuropathy. The classification was carried out using Artificial Neural Network (ANN). There are five feature extraction techniques that were used to extract the signals such as Autoregressive (AR), Root Mean Square (RMS), Zero Crossing (ZC), Waveform length (WL) and Mean Absolute Value (MAV). A comparative analysis of these different techniques were carried out based on the results. The Multilayer Perceptron (MLP) was used for carrying out the classification.

AB - Electromyography (EMG) signals are the measure of activity in the muscles. The aim of this study is to identify the neuromuscular disease based on EMG signals by means of classification. The neuromuscular diseases that have been identified are myopathy and neuropathy. The classification was carried out using Artificial Neural Network (ANN). There are five feature extraction techniques that were used to extract the signals such as Autoregressive (AR), Root Mean Square (RMS), Zero Crossing (ZC), Waveform length (WL) and Mean Absolute Value (MAV). A comparative analysis of these different techniques were carried out based on the results. The Multilayer Perceptron (MLP) was used for carrying out the classification.

KW - Autoregressive method (AR)

KW - Classification

KW - Electromyography (EMG)

KW - Feature Extraction

KW - Multilayer Perceptron (MLP)

KW - Neuromuscular Disease

KW - Root mean square (RMS)

KW - Waveform length (WL) and Mean Absolute Value (MAV)

KW - Zero Crossing (ZC)

UR - http://www.scopus.com/inward/record.url?scp=84962834084&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2015.12.346

DO - 10.1016/j.procs.2015.12.346

M3 - Article

VL - 76

SP - 223

EP - 228

JO - Procedia Computer Science

T2 - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -