Improving the Performance Against Force Variation of EMG Controlled Multifunctional Upper-Limb Prostheses for Transradial Amputees

• Published in: IEEE transactions on neural systems and rehabilitation engineering Vol. 24; no. 6; pp. 650 - 661
• Main Authors: Al-Timemy, Ali H., Khushaba, Rami N., Bugmann, Guido, Escudero, Javier
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.06.2016; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adult; Amputation; Amputation Stumps - physiopathology; Amputees - rehabilitation; Artificial Limbs; Biofeedback, Psychology - instrumentation; Biofeedback, Psychology - methods; Classification; Computational efficiency; Electrodes; Electromyography; Electromyography - instrumentation; Electromyography - methods; Equipment Design; Equipment Failure Analysis; Feature extraction; Female; Force; force level variation; Hand - physiopathology; Hand - surgery; Humans; Male; Middle Aged; Movements; Muscle Contraction; myoelectric control; pattern recognition; Prostheses; Prosthetics; Radius - physiopathology; Radius - surgery; Reduction; Reproducibility of Results; Robustness; Sensitivity and Specificity; Stress, Mechanical; surface electromyogram (sEMG); Surgical implants; Task Performance and Analysis; Time-domain analysis; Training; transradial amputees; Young Adult
• ISSN: 1534-4320; 1558-0210
• DOI: 10.1109/TNSRE.2015.2445634

We investigate the problem of achieving robust control of hand prostheses by the electromyogram (EMG) of transradial amputees in the presence of variable force levels, as these variations can have a substantial impact on the robustness of the control of the prostheses. We also propose a novel set of features that aim at reducing the impact of force level variations on the prosthesis controlled by amputees. These features characterize the EMG activity by means of the orientation between a set of spectral moments descriptors extracted from the EMG signal and a nonlinearly mapped version of it. At the same time, our feature extraction method processes the EMG signals directly from the time-domain to reduce computational cost. The performance of the proposed features is tested on EMG data collected from nine transradial amputees performing six classes of movements each with three force levels. Our results indicate that the proposed features can achieve significant reductions in classification error rates in comparison to other well-known feature extraction methods, achieving improvements of ≈ 6% to 8% in the average classification performance across all subjects and force levels, when training with all forces.