EMG-Based Motion Discrimination Using a Novel Recurrent Neural Network

This paper presents a pattern discrimination method for electromyogram (EMG) signals for application in the field of prosthetic control. The method uses a novel recurrent neural network based on the hidden Markov model. This network includes recurrent connections, which enable modeling time series, such as EMG signals. Weight coefficients in the network can be learned using a well-known back-propagation through time algorithm. Pattern discrimination experiments were conducted to demonstrate the feasibility and performance of the proposed method. We were able to successfully discriminate forearm motions using the EMG signals, and achieved considerably high discrimination performance compared with other discrimination methods.

[1]  Toshio Tsuji,et al.  A log-linearized Gaussian mixture network and its application to EEG pattern classification , 1999, IEEE Trans. Syst. Man Cybern. Part C.

[2]  Katsunori Shimohara,et al.  EMG pattern analysis and classification by neural network , 1989, Conference Proceedings., IEEE International Conference on Systems, Man and Cybernetics.

[3]  P J Webros BACKPROPAGATION THROUGH TIME: WHAT IT DOES AND HOW TO DO IT , 1990 .

[4]  M. Zak Terminal attractors for addressable memory in neural networks , 1988 .

[5]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[6]  Toshio Tsuji,et al.  Motion Discrimination Method from EMG Signals Using Statistically Structured Neural Networks , 1992 .

[7]  Yasuharu Koike,et al.  Estimation of Arm Posture in 3D-Space from Surface EMG Signals Using a Neural Network Model (Special Issue on Neurocomputing) , 1994 .

[8]  Mitsuo Nagamachi,et al.  A limb-function discrimination method using EMG signals for the control of multifunctional powered prostheses , 1987, Systems and Computers in Japan.

[9]  A. A. Beex,et al.  A microprocessor system for multifunctional control of upper-limb prostheses via myoelectric signal identification , 1978 .

[10]  Toshio Tsuji,et al.  Pattern classification of EEG signals using a log-linearized Gaussian mixture neural network , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[11]  R.N. Scott,et al.  The application of neural networks to myoelectric signal analysis: a preliminary study , 1990, IEEE Transactions on Biomedical Engineering.

[12]  Katsunori Shimohara,et al.  EMG pattern recognition by neural networks for prosthetic fingers control , 1992 .

[13]  Toshio Tsuji,et al.  A Human Supporting Manipulator Based on Manual Control Using EMG Signals , 2000 .

[14]  Toshio Tsuji,et al.  A recurrent log-linearized Gaussian mixture network , 2003, IEEE Trans. Neural Networks.

[15]  L. Ljung,et al.  A Microprocessor System for Multifunctional Control of Upper-Limb Prostheses via Myoelectric Signal Identification , 1978 .

[16]  Toshio Tsuji,et al.  Discrimination of Forearm Motions from EMG Signals by Error Back Propagation Typed Neural Network Using Entropy , 1993 .

[17]  Han-Pang Huang,et al.  Development of a myoelectric discrimination system for a multi-degree prosthetic hand , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).