Title :
Nonnegative Matrix Factorization for the Identification of EMG Finger Movements: Evaluation Using Matrix Analysis
Author :
Naik, Ganesh R. ; Nguyen, Hung T.
Author_Institution :
Centre for Health Technol., Univ. of Technol. Sydney, Sydney, NSW, Australia
Abstract :
Surface electromyography (sEMG) is widely used in evaluating the functional status of the hand to assist in hand gesture recognition, prosthetics and rehabilitation applications. The sEMG is a noninvasive, easy to record signal of superficial muscles from the skin surface. Considering the nonstationary characteristics of sEMG, recent feature selection of hand gesture recognition using sEMG signals necessitate designers to use nonnegative matrix factorization (NMF)-based methods. This method exploits both the additive and sparse nature of signals by extracting accurate and reliable measurements of sEMG features using a minimum number of sensors. The testing has been conducted for simple and complex finger flexions using several experiments with artificial neural network classification scheme. It is shown, both by simulation and experimental studies, that the proposed algorithm is able to classify ten finger flexions (five simple and five complex finger flexions) recorded from two sEMG sensors up to 92% (95% for simple and 87% for complex flexions) accuracy. The recognition performances of simple and complex finger flexions are also validated with NMF permutation matrix analysis.
Keywords :
bioinformatics; biomechanics; electromyography; feature extraction; feature selection; gesture recognition; image sensors; matrix decomposition; medical signal detection; medical signal processing; neural nets; patient care; patient diagnosis; patient monitoring; skin; EMG finger movement identification; NMF permutation matrix analysis; NMF-based methods; accurate sEMG feature extraction; additive signal exploitation; algorithm-based finger flexion classification; artificial neural network classification scheme; complex finger flexion recognition performance; electromyography-based finger movement identification; feature selection; functional hand status evaluation; matrix analysis-based finger movement analysis; noninvasive muscular signal recording tool; nonnegative matrix factorization; nonstationary sEMG characteristics; prosthetics; rehabilitation applications; reliable sEMG feature measurements; sEMG sensor-recorded complex finger flexion; sEMG sensor-recorded simple finger flexion; sEMG signal-based hand gesture recognition; sensor based sEMG feature measurement; sensor-based sEMG feature extraction; simple finger flexion recognition performance; skin surface muscle signal; skin surface muscular signal; sparse signal exploitation; superficial muscle signal; surface electromyography feature extraction; Artificial neural networks; Electromyography; Feature extraction; Muscles; Thumb; Training; Artificial neural network (ANN); auto regression (AR); flexions; gestures; hand gesture recognition; nonnegative matrix factorization (NMF); principal component analysis (PCA); root mean square (RMS); surface electromyography (sEMG);
Journal_Title :
Biomedical and Health Informatics, IEEE Journal of
DOI :
10.1109/JBHI.2014.2326660
