Multiresolution MUAPs decomposition and SVM-based analysis in the classification of neuromuscular disorders

Authors:
Andrzej P. Dobrowolski;Mariusz Wierzbowski;Kazimierz Tomczykiewicz
Affiliations:
Military University of Technology, Faculty of Electronics, 2 Kaliskiego St., 00-908 Warsaw, Poland;Military University of Technology, Faculty of Electronics, 2 Kaliskiego St., 00-908 Warsaw, Poland;Military Institute of Health Service, Department of Neurology, 128 Szaserow St., 00-909 Warsaw, Poland
Venue:
Computer Methods and Programs in Biomedicine
Year:
2012

Citing 7
Cited 1

A Theory for Multiresolution Signal Decomposition: The Wavelet Representation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Ten lectures on wavelets

Ten lectures on wavelets
Support-Vector Networks

Machine Learning
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
Estimation of Dependences Based on Empirical Data: Empirical Inference Science (Information Science and Statistics)

Estimation of Dependences Based on Empirical Data: Empirical Inference Science (Information Science and Statistics)
Filter design for cancellation of baseline-fluctuation in needle EMG recordings

Computer Methods and Programs in Biomedicine

Medical decision support system for diagnosis of neuromuscular disorders using DWT and fuzzy support vector machines

Computers in Biology and Medicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a new method for the classification of neuromuscular disorders based on the analysis of scalograms determined by the Symlet 4 wavelet technique. The approach involves isolating single motor unit action potentials (MUAPs), computing their scalograms, taking the maximum values of the scalograms in five selected scales, and averaging across MUAPs to give a single 5-dimensional feature vector per subject. After SVM analysis, the vector is reduced to a single decision parameter, called the Wavelet Index, allowing the subject to be assigned to one of three groups: myogenic, neurogenic or normal. The software implementation of the method described above created a tool supporting electromyographic (EMG) examinations. The method is characterized by a high probability for the accurate diagnosis of muscle state. The method produced 5 misclassifications out of 800 examined cases (total error of 0.6%).