AUTHOR=Shahzad Waseem , Ayaz Yasar , Khan Muhammad Jawad , Naseer Noman , Khan Mushtaq 

TITLE=Enhanced Performance for Multi-Forearm Movement Decoding Using Hybrid IMU–sEMG Interface

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 13 - 2019

YEAR=2019

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2019.00043

DOI=10.3389/fnbot.2019.00043

ISSN=1662-5218

ABSTRACT=Control of active prosthetic hands using surface electromyography (sEMG) signals is an active research area, but, despite the advances in sEMG pattern recognition and classification techniques none of the commercially available prosthetic hands provide the user with an intuitive control. One of the major reasons for this disparity between academia and industry is the variation of sEMG signals in a dynamic environment as opposed to the controlled laboratory conditions. This research investigated the effects of sEMG signal variation on the performance of a hand motion classifier due to arm position variation and also explored the effect of static and dynamic position strategies for classifier training. A wearable sensor system measured the electrical activity of the muscles and the position of the forearm while performing six classes of hand motions at static positions and under dynamic forearm movements. A set of four time domain (TD) features of sEMG signals were used to train and test a support vector machine (SVM) classifier. Results showed the dynamic training approach achieving 98.7% classification accuracy with an average improvement of 1.2% over the static training approach. Results indicated a significant dependence of classifier performance on arm position. The position aware classifiers showed an improvement of 4.1% for the static position training and an improvement of 5.8% for the dynamic position training as compared to the classifiers trained with only TD features of the sEMG. This proves the effectiveness of  dynamic training approach and sensor fusion techniques to improve the acceptability and performance of existing standalone sEMG based prosthetic control systems.