Classification of forearm EMG signals for 10 motions using optimum feature-channel combinations

Abstract

Electromyography (EMG) is the study of electrical activity in the muscles. We classify EMG signals from surface electrodes (channels) using Artificial Neural Network (ANN). We evaluate classification performance of 10 different hand motions using several feature-channel combinations with wrapper method. Highest classification accuracy of 98.7% is achieved with each feature-channel combination. Compared to previous studies, we achieve the highest accuracy for 10 classes with lower number of feature-channel combination. We reduce ANN complexity without compromising the classification accuracy for deployment in low-end hardware with limited computational power along with improving the design of a low-cost hardware for EMG signal acquisition.

Keywords:

• Electromyography
• artificial neural network
• signal classification
• feature extraction
• electrode placement

Acknowledgements

Authors would like to thank Dr. Aamir Iqbal from Capital University of Science and Technology for providing data acquisition card. We are also grateful to Mushtaq Ahmed, IST, for his assistance in circuit fabrication. Moreover, we also owe our gratitude to all the volunteers who participated in EMG signal measurements.

Disclosure statement

No potential conflict of interest was reported by the author(s).