Robust Classification of Intramuscular EMG Signals to Aid the Diagnosis of Neuromuscular Disorders

<italic>Goal:</italic> This article presents the design and validation of an accurate automatic diagnostic system to classify intramuscular EMG (iEMG) signals into healthy, myopathy, or neuropathy categories to aid the diagnosis of neuromuscular diseases. <italic>Methods:</italic> First, an iEMG signal is decimated to produce a set of “disjoint” downsampled signals, which are decomposed by the lifting wavelet transform (LWT). The Higuchi's fractal dimensions (FDs) of LWT coefficients in the subbands are computed. The FDs of LWT subband coefficients are fused with one-dimensional local binary pattern derived from each downsampled signal. Next, a multilayer perceptron neural network (MLPNN) determines the class labels of downsampled signals. Finally, the sequence of class labels is fed to the Boyer-Moore majority vote (BMMV) algorithm, which assigns a class to every iEMG signal. <italic>Results:</italic> The MLPNN-BMMV classifier was experimented with 250 iEMG signals belonging to three categories. The performance of the classifier was validated in comparison with state-of-the-art approaches. The MLPNN-BMMV has resulted in impressive performance measures (%) using a 10-fold cross-validation—accuracy = <inline-formula><tex-math notation="LaTeX">$99.87\pm 0.25$</tex-math></inline-formula>, sensitivity (normal) = <inline-formula><tex-math notation="LaTeX">$99.97\pm 0.13$</tex-math></inline-formula>, sensitivity (myopathy) = <inline-formula><tex-math notation="LaTeX">$99.68\pm 0.95$</tex-math></inline-formula>, sensitivity (neuropathy) = <inline-formula><tex-math notation="LaTeX">$99.76\pm 0.66$</tex-math></inline-formula>, specificity (normal) = <inline-formula><tex-math notation="LaTeX">$99.72\pm 0.61$</tex-math></inline-formula>, specificity (myopathy) = <inline-formula><tex-math notation="LaTeX">$99.98\pm 0.10$</tex-math></inline-formula>, and specificity (neuropathy) = <inline-formula><tex-math notation="LaTeX">$99.96\pm 0.14$</tex-math></inline-formula>—surpassing the existing approaches. <italic>Conclusions:</italic> A future research direction is to validate the classifier performance with diverse iEMG datasets, which would lead to the design of an affordable real-time expert system for neuromuscular disorder diagnosis.

[1]  Shobha Jose,et al.  DWT-based electromyogram signal classification using maximum likelihood-estimated features for neurodiagnostic applications , 2020, Signal Image Video Process..

[2]  Roberto Merletti,et al.  Electromyography. Physiology, engineering and non invasive applications , 2005 .

[3]  Yilmaz Kaya,et al.  Detection of Parkinson's disease by Shifted One Dimensional Local Binary Patterns from gait , 2016, Expert Syst. Appl..

[4]  Yilmaz Kaya,et al.  1D-local binary pattern based feature extraction for classification of epileptic EEG signals , 2014, Appl. Math. Comput..

[5]  Guruprasad Madhavan,et al.  Electromyography: Physiology, Engineering and Non-Invasive Applications , 2005, Annals of Biomedical Engineering.

[6]  Lachit Dutta,et al.  A Multiview Discriminant Feature Fusion-Based Nonlinear Process Assessment and Diagnosis: Application to Medical Diagnosis , 2019, IEEE Transactions on Instrumentation and Measurement.

[7]  S. Vasanthi,et al.  Classification of Neuromuscular Diseases using Dominant MUAP Based on Wavelet Domain Features and Improving Its Accuracy using SVM , 2016 .

[8]  Christos D. Katsis,et al.  A novel method for automated EMG decomposition and MUAP classification , 2006, Artif. Intell. Medicine.

[9]  Abdulhamit Subasi,et al.  Effect of multiscale PCA de-noising on EMG signal classification for diagnosis of neuromuscular disorders , 2014, Journal of Medical Systems.

[10]  S Micera,et al.  Control of Hand Prostheses Using Peripheral Information , 2010, IEEE Reviews in Biomedical Engineering.

[11]  C. Pattichis,et al.  Autoregressive and cepstral analyses of motor unit action potentials. , 1999, Medical engineering & physics.

[12]  Richard G. Baraniuk,et al.  Nonlinear wavelet transforms for image coding via lifting , 2003, IEEE Trans. Image Process..

[13]  B.S. Raghavendra,et al.  Signal characterization using Fractal Dimension , 2008, TENCON 2008 - 2008 IEEE Region 10 Conference.

[14]  Anju Krishna,et al.  Classification of EMG Signals Using Spectral Features Extracted from Dominant Motor Unit Action Potential , 2015 .

[15]  Ram Bilas Pachori,et al.  Computer aided detection of abnormal EMG signals based on tunable-Q wavelet transform , 2017, 2017 4th International Conference on Signal Processing and Integrated Networks (SPIN).

[16]  Abdulhamit Subasi,et al.  Medical decision support system for diagnosis of neuromuscular disorders using DWT and fuzzy support vector machines , 2012, Comput. Biol. Medicine.

[17]  H. Adeli,et al.  Fractality analysis of frontal brain in major depressive disorder. , 2012, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[18]  Manuel G. Penedo,et al.  Heartbeat classification fusing temporal and morphological information of ECGs via ensemble of classifiers , 2019, Biomed. Signal Process. Control..

[19]  Steven W. Su,et al.  Electromyography (EMG) based Classification of Neuromuscular Disorders using Multi-Layer Perceptron , 2015 .

[20]  Nikolaos Arikidis,et al.  Interscale wavelet maximum - a fine to coarse algorithm for wavelet analysis of the EMG interference pattern , 2002, IEEE Transactions on Biomedical Engineering.

[21]  T. Higuchi.  Approach to an irregular time series on the basis of the fractal theory , 1988 .

[22]  Abdulhamit Subasi,et al.  Comparison of decision tree algorithms for EMG signal classification using DWT , 2015, Biomed. Signal Process. Control..

[23]  B. S. Raghavendra,et al.  Complexity analysis of EEG in patients with schizophrenia using fractal dimension , 2009, Physiological measurement.

[24]  Editedby Eleanor Criswell,et al.  Cram's Introduction to Surface Electromyography , 2010 .

[25]  Abdulhamit Subasi,et al.  Classification of EMG signals using PSO optimized SVM for diagnosis of neuromuscular disorders , 2013, Comput. Biol. Medicine.

[26]  Ali Motie Nasrabadi,et al.  EEG-Based Mental Task Classification in Hypnotized and Normal Subjects , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[27]  Ganesh R. Naik,et al.  Single-Channel EMG Classification With Ensemble-Empirical-Mode-Decomposition-Based ICA for Diagnosing Neuromuscular Disorders , 2016, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[28]  Muhammad Yazid,et al.  Variable step dynamic threshold local binary pattern for classification of atrial fibrillation , 2020, Artif. Intell. Medicine.

[29]  Reza Boostani,et al.  A Multi-Classifier Approach to MUAP Classification for Diagnosis of Neuromuscular Disorders , 2014, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[30]  Jamileh Yousefi,et al.  Characterizing EMG data using machine-learning tools , 2014, Comput. Biol. Medicine.

[31]  Leonardo Duque-Muñoz,et al.  Neuromuscular disease detection by neural networks and fuzzy entropy on time‐frequency analysis of electromyography signals , 2018, Expert Syst. J. Knowl. Eng..

[32]  John J. Soraghan,et al.  Lower Arm Electromyography (EMG) Activity Detection Using Local Binary Patterns , 2014, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[33]  Varun Bajaj,et al.  Classification of amyotrophic lateral sclerosis disease based on convolutional neural network and reinforcement sample learning algorithm , 2017, Health Inf. Sci. Syst..

[34]  M. J. Katz,et al.  Fractals and the analysis of waveforms. , 1988, Computers in biology and medicine.

[35]  Lachit Dutta,et al.  An automatic feature extraction and fusion model: application to electromyogram (EMG) signal classification , 2018, International Journal of Multimedia Information Retrieval.

[36]  Abdulhamit Subasi,et al.  Classification of EMG signals using combined features and soft computing techniques , 2012, Appl. Soft Comput..

[37]  Srdjan Kesic,et al.  Application of Higuchi's fractal dimension from basic to clinical neurophysiology: A review , 2016, Comput. Methods Programs Biomed..

[38]  Kiran Pu,et al.  TQWT Based Features for Classification of ALS and Healthy EMG Signals , 2018 .

[39]  Clodoaldo Ap. M. Lima,et al.  Assessing fractal dimension methods as feature extractors for EMG signal classification , 2014, Eng. Appl. Artif. Intell..

[40]  W. J. Hennessey,et al.  Needle electromyography. , 1995, Archives of physical medicine and rehabilitation.

[41]  Brian Litt,et al.  A comparison of waveform fractal dimension algorithms , 2001 .

[42]  Constantinos S. Pattichis,et al.  Genetics-based machine learning for the assessment of certain neuromuscular disorders , 1996, IEEE Trans. Neural Networks.

[43]  Mustafa Yilmaz,et al.  Classification of EMG signals using wavelet neural network , 2006, Journal of Neuroscience Methods.

[44]  E W Abel,et al.  Neural network analysis of the EMG interference pattern. , 1996, Medical engineering & physics.

[45]  Pornchai Phukpattaranont,et al.  Feature reduction and selection for EMG signal classification , 2012, Expert Syst. Appl..

[46]  Abdulhamit Subasi,et al.  A decision support system for diagnosis of neuromuscular disorders using DWT and evolutionary support vector machines , 2013, Signal, Image and Video Processing.

[47]  Dheeraj Sharma,et al.  An efficient method for analysis of EMG signals using improved empirical mode decomposition , 2017 .

[48]  Wim Sweldens,et al.  Lifting scheme: a new philosophy in biorthogonal wavelet constructions , 1995, Optics + Photonics.

[49]  Ellen Poliakoff,et al.  Machine learning algorithm validation with a limited sample size , 2019, PloS one.

[50]  Edilson Delgado Trejos,et al.  Discrete wavelet transform and k-nn classification in EMG signals for diagnosis of neuromuscular disorders , 2014, 2014 XIX Symposium on Image, Signal Processing and Artificial Vision.

[51]  S. A. Fattah,et al.  Neuromuscular disease classification based on mel frequency cepstrum of motor unit action potential , 2014, 2014 International Conference on Electrical Engineering and Information & Communication Technology.

[52]  Wim Sweldens,et al.  The lifting scheme: a construction of second generation wavelets , 1998 .

[53]  Norman Poh,et al.  One dimensional local binary patterns of electroencephalogram signals for detecting Alzheimer's disease , 2017, 2017 22nd International Conference on Digital Signal Processing (DSP).

[54]  Sabri Koçer,et al.  Classification of Emg Signals Using Neuro-Fuzzy System and Diagnosis of Neuromuscular Diseases , 2010, Journal of Medical Systems.

[55]  Robert S. Boyer,et al.  MJRTY: A Fast Majority Vote Algorithm , 1991, Automated Reasoning: Essays in Honor of Woody Bledsoe.

[56]  Richard Simon,et al.  Bias in error estimation when using cross-validation for model selection , 2006, BMC Bioinformatics.

[57]  Anil Kumar,et al.  Classification of normal, ALS, and myopathy EMG signals using ELM classifier , 2016, 2016 2nd International Conference on Advances in Electrical, Electronics, Information, Communication and Bio-Informatics (AEEICB).

[58]  C. Tricot.  Curves and Fractal Dimension , 1994 .