Support vector machine in computer aided clinical electromyography

Motor unit action potentials (MUAPs) recorded during routine electromyography (EMG) examination provides important information for the assessment of neuromuscular disorders. In this preliminary study, support vector machines (SVMs) based on multi-class classifier is activated for the identification of normal subjects and patients suffering from motor neuron diseases (MND) and myopathies (MVO). The results in experiments prove the classification validity of SVMs which guarantee high generalization ability on the testing samples. Furthermore, its performance is compared with a back-propagation (BP) neural network. More excellent recognition accuracy indicates the potential of the SVMs techniques in clinical neuromuscular disorders evaluation.

[1]  Kuo-Chen Chou,et al.  Prediction of Protein Structural Classes by Support Vector Machines , 2002, Comput. Chem..

[2]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[3]  S. Andreassen,et al.  Quantitative Analysis of Individual Motor Unit Potentials: A Proposition for Standardized Terminology and Criteria for Measurement , 1986, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[4]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[5]  R Le Bars,et al.  Automatic classification of electromyographic signals. , 1983, Electroencephalography and clinical neurophysiology.

[6]  R. Fletcher.  Practical Methods of Optimization , 1988 .

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

[8]  E Stålberg,et al.  Quantitative motor unit potential analysis. , 1996, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[9]  Steen Andreassen,et al.  MUNIN: an expert system for EMG , 1987 .