Abstract
We present an approach to classifying myoelectric signals (MES) that involves distinct feature extraction and classification stages. The extraction of MES features is accomplished by a self-organizing neural network composed of BCM neurons that performs Exploratory Projection Pursuit. This approach produces a meaningful representation of the input data using a limited number of features which are usefully differentiable for classification. Classification is accomplished by a simple backpropagation network. This forms the basis for a myoelectric control system that exhibits a lower initial state selection error than other neural network based approaches which have recently been reported. This system offers enhanced functionality for a myoelectric prosthesis and simplifies user training through adaptation to the individual MES characteristics of each user.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
B. Hudgins, P. Parker, and R. Scott, “The Recognition of Myoelectric Patterns for Prosthetic Limb Control,” in Ann. Intl. Conf. IEEE Engineering in Biology Society, vol. 13, pp. 2040–2041, 1991.
N. Intrator, “A Neural Network for Feature Extraction,” in Advances in Neural Information Processing Systems, vol. 2, pp. 719–726, 1990.
N. Intrator, “Feature Extraction Using an Unsupervised Neural Network,” Neural Computation, vol. 4, pp. 98–107, 1992.
C. DeLuca, “Physiology and Mathematics of Myoelectric Signals,” IEEE Trans. Biomedical Engrg., vol. 26, pp. 313–325, June 1979.
P. Parker and R. Scott, “Myoelectric control of prosthesis,” CRC Crit. Rev. Biomed. Engrg., vol. 13, no. 4, pp. 283–310, 1986.
P. Parker, J. Stuller, and R. Scott, “Processing for the multistate myoelectric channel,” Proc. IEEE, vol. 65, no. 5, pp. 662–674, 1977.
B. Hudgins, A New Approach to Multifunction Myoelectric Control. PhD thesis, University of New Brunswick, 1991.
D. Graupe and W. Cline, “Functional separation of EMG signal via ARMA identification methods for prosthesis control purposes,” IEEE Trans. Sys. Man Cyber., vol. SMC-5, no. 2, pp. 252–259, 1975.
M. Kelly, P. Parker, and R. Scott, “The Application of Neural Networks to Myoelectric Signal Analysis: A Preliminary Study,” IEEE Trans. Biomedical Engrg., vol. 37, pp. 221–230, Mar. 1990.
E. Pietka, “Feature extraction in the computerized approach to the ECG analysis,” Pattern Recognition, vol. 24, no. 2, pp. 139–146, 1991.
C. Schizas, C. Pattichis, I. Schofield, P. Fawcett, and L. Middleton, “Artificial Neural Net Algorithms in Classifying Electromyographic Signals,” in Proc. First IEE Conf. on Artificial Neural Networks, vol. 1, pp. 134–138, 1989.
P. Huber, “Projection Pursuit (with discussions),” Annals of Statistics, vol. 13, no. 2, pp. 435–475, 1985.
J. Friedman and J. Tukey, “A Projection Pursuit Algorithm for Exploratory Data Analysis,” IEEE Trans. on Computers, vol. C-23, pp. 881–890, Sept. 1974.
B. Ripley, “Statistical aspects of neural networks,” in press, 1992.
P. Diaconis and D. Freedman, “Asymptotics of graphical projection pursuit,” Annals of Statistics, vol. 12, pp. 793–815, 1984.
E. Oja, “A simplified neuron model as a principal component analyzer,” J. Math. Biology, vol. 15, pp. 267–273, 1982.
T. Leen, “Dynamics of learning in recurrent feature-discovery networks,” in Advances in Neural Information Processing Systems (R. Lippman, J. Moody, and D. Touretzky, eds.), vol. 3, pp. 70–76, 1991.
T. Sanger, “An optimality principle for unsupervised learning,” in Advances in Neural Information Processing Systems (D. Touretzky, ed.), vol. 1, pp. 11–19, 1989.
E. Bienenstock, L. Cooper, and P. Munro, “Theory for the Development of Neuron Selectivity: Orientation Specificity and Binocular Interaction in Visual Cortex,” Journal of Neuroscience, vol. 2, pp. 32–48, Jan. 1982.
M. Bear, L. Cooper, and F. Ebner, “A Physiological Basis for a Theory of Synapse Modification,” Science, vol. 237, pp. 42–48, 1987.
J. Marshall, “Self-organizing neural networks for perception of visual motion,” Neural Networks, vol. 3, pp. 45–74, 1990.
N. Intrator and G. Tajchman, “Supervised and unsupervised feature extraction from a cochlear model for speech recognition,” in Neural Networks for Signal Processing — Proceedings of the 1991 IEEE Workshop (B. Juang, S. Kung, and C. Kamm, eds.), pp. 460–469, IEEE Press, 1991.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gallant, P.J., Morin, E.L., Peppard, L.E. (1996). Improving myoelectric signal classifier generalization by preprocessing with exploratory projections. In: Chouinard, JY., Fortier, P., Gulliver, T.A. (eds) Information Theory and Applications II. CWIT 1995. Lecture Notes in Computer Science, vol 1133. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0025150
Download citation
DOI: https://doi.org/10.1007/BFb0025150
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61748-8
Online ISBN: 978-3-540-70647-2
eBook Packages: Springer Book Archive