Transparent Muscle Characterization Using Quantitative Electromyography: Different Binarization Mappings

Evaluation of patients with suspected neuromuscular disorders is typically based on qualitative visual and auditory assessment of needle detected eletromyographic (EMG) signals; the resulting muscle characterization is subjective and highly dependent on the skill and experience of the examiner. Quantitative electromyography (QEMG) techniques were developed to extract motor unit potential trains (MUPTs) from needle detected EMG signals, and estimate features capturing motor unit potential (MUP) morphology and quantifying morphological consistency across MUPs belonging to the same MUPT. The aim of this study is to improve available methods for obtaining transparent muscle characterizations from features obtained using QEMG techniques. More specifically, we investigate the following. 1) Can the use of binarization mappings improve muscle categorization accuracies of transparent methods? 2) What are the appropriate binarization mappings in terms of accuracy and transparency? Results from four different sets of examined limb muscles (342 muscles in total) demonstrate that four out of the 10 investigated binarization mappings based on transparent characterization methods outperformed the multi-class characterizers based on Gaussian mixture models (GMM) and the corresponding binarization mappings based on GMM. This suggests that the use of an appropriate binarization mapping can overcome the decrease in categorization accuracy associated with quantizing MUPT features, which is necessary to obtain transparent characterizations. This performance gain can be attributed to the use of more relevant features and tuned quantization to obtain more specific binary characterizations.

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

[2]  R. Werner,et al.  Interrater reliability of the needle examination in lumbosacral radiculopathy , 2006, Muscle & nerve.

[3]  Johannes Fürnkranz,et al.  Round Robin Rule Learning , 2001, ICML.

[4]  William W. Cohen.  Fast Effective Rule Induction , 1995, ICML.

[5]  Douglas A. Reynolds,et al.  Speaker Verification Using Adapted Gaussian Mixture Models , 2000, Digit. Signal Process..

[6]  D W Stashuk,et al.  Motor unit potential characterization using "pattern discovery". , 2008, Medical engineering & physics.

[7]  Christos D. Katsis,et al.  A two-stage method for MUAP classification based on EMG decomposition , 2007, Comput. Biol. Medicine.

[8]  Nello Cristianini,et al.  Large Margin DAGs for Multiclass Classification , 1999, NIPS.

[9]  K Kunze,et al.  Discriminant classification of motor unit potentials (MUPs) successfully separates neurogenic and myopathic conditions. A comparison of multi- and univariate diagnostical algorithms for MUP analysis. , 1995, Electroencephalography and clinical neurophysiology.

[10]  Johannes Fürnkranz,et al.  Incremental Reduced Error Pruning , 1994, ICML.

[11]  Meena Abdelmaseeh,et al.  Feature selection for motor unit potential train characterization , 2014, Muscle & nerve.

[12]  Pascal Poupart,et al.  Muscle Categorization Using Quantitative Needle Electromyography: A 2-Stage Gaussian Mixture Model Based Approach , 2012, 2012 11th International Conference on Machine Learning and Applications.

[13]  D Stashuk,et al.  EMG signal decomposition: how can it be accomplished and used? , 2001, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[14]  Gérard Dreyfus,et al.  Pairwise Neural Network Classifiers with Probabilistic Outputs , 1994, NIPS.

[15]  Yang Wang,et al.  From Association to Classification: Inference Using Weight of Evidence , 2003, IEEE Trans. Knowl. Data Eng..

[16]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[17]  Robert Tibshirani,et al.  Classification by Pairwise Coupling , 1997, NIPS.

[18]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[19]  Tomoyuki Hamamura,et al.  A multiclass classification method based on multiple pairwise classifiers , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[20]  Yang Wang,et al.  Pattern discovery: a data driven approach to decision support , 2003, IEEE Trans. Syst. Man Cybern. Part C.

[21]  Wojciech W. Siedlecki.  A formula for multi-class distributed classifiers , 1994, Pattern Recognit. Lett..

[22]  Robert Sabourin,et al.  “One Against One” or “One Against All”: Which One is Better for Handwriting Recognition with SVMs? , 2006 .

[23]  Yoram Singer,et al.  Reducing Multiclass to Binary: A Unifying Approach for Margin Classifiers , 2000, J. Mach. Learn. Res..

[24]  Yang Wang,et al.  Discovery of high order patterns , 1995, 1995 IEEE International Conference on Systems, Man and Cybernetics. Intelligent Systems for the 21st Century.

[25]  D W Stashuk,et al.  Decomposition and quantitative analysis of clinical electromyographic signals. , 1999, Medical engineering & physics.

[26]  F BUCHTHAL,et al.  ACTION POTENTIAL PARAMETERS IN DIFFERENT HUMAN MUSCLES , 1955, Acta psychiatrica et neurologica Scandinavica.

[27]  Hossein Parsaei,et al.  SVM-Based Validation of Motor Unit Potential Trains Extracted by EMG Signal Decomposition , 2012, IEEE Transactions on Biomedical Engineering.

[28]  Constantinos S. Pattichis,et al.  Neural network models in EMG diagnosis , 1995 .

[29]  E. Kugelberg.  ELECTROMYOGRAPHY IN MUSCULAR DYSTROPHIES , 1949, Journal of neurology, neurosurgery, and psychiatry.