TY - GEN
T1 - Notice of Removal
T2 - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019
AU - Tayeb, Zied
AU - Jakovleski, Philipp
AU - Chen, Zhong
AU - Lippert, Jannick
AU - Lanillos, Pablo
AU - Lee, Dongheui
AU - Cheng, Gordon
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/5/16
Y1 - 2019/5/16
N2 - In recent time we have witnessed a major push towards providing an online closed-loop control of upper-limb hand prostheses. Notwithstanding the substantial advances that have been made in developing invasive closed-loop systems, barriers remain in achieving suitable levels of non-invasiveness and a closed-loop non-invasively controlled prosthetic hand with lifelike dexterity is still missing. In this context, this work proposes a low-cost, non-invasive system for prosthetic hands control using surface EMG (sEMG) signals. We verified the system with 10 human participants. Using the proposed system: 1) four hand movements and two force levels (low, high) were successfully classified from sEMG signals; 2) a real-time control of the prosthetic hand was achieved using the decoded sEMG activity with an average online accuracy of 86.25% using solely 10 training trials per posture; 3) measured finger-tip forces were translated into vibro-tactile stimulation of the muscles for feedback and participants were able to differentiate between three different grasped objects (soft, hard and medium objects) with an average success rate of more than 88.46% across all 10 subjects; 4) the detection of muscle fatigue from sEMG was successfully performed. In sum, our results show the potential of using low cost and non-invasive approaches for closed-loop control of upper-limb hand prostheses.
AB - In recent time we have witnessed a major push towards providing an online closed-loop control of upper-limb hand prostheses. Notwithstanding the substantial advances that have been made in developing invasive closed-loop systems, barriers remain in achieving suitable levels of non-invasiveness and a closed-loop non-invasively controlled prosthetic hand with lifelike dexterity is still missing. In this context, this work proposes a low-cost, non-invasive system for prosthetic hands control using surface EMG (sEMG) signals. We verified the system with 10 human participants. Using the proposed system: 1) four hand movements and two force levels (low, high) were successfully classified from sEMG signals; 2) a real-time control of the prosthetic hand was achieved using the decoded sEMG activity with an average online accuracy of 86.25% using solely 10 training trials per posture; 3) measured finger-tip forces were translated into vibro-tactile stimulation of the muscles for feedback and participants were able to differentiate between three different grasped objects (soft, hard and medium objects) with an average success rate of more than 88.46% across all 10 subjects; 4) the detection of muscle fatigue from sEMG was successfully performed. In sum, our results show the potential of using low cost and non-invasive approaches for closed-loop control of upper-limb hand prostheses.
UR - http://www.scopus.com/inward/record.url?scp=85066750590&partnerID=8YFLogxK
U2 - 10.1109/NER.2019.8717124
DO - 10.1109/NER.2019.8717124
M3 - Conference contribution
AN - SCOPUS:85066750590
T3 - International IEEE/EMBS Conference on Neural Engineering, NER
SP - 1021
EP - 1024
BT - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019
PB - IEEE Computer Society
Y2 - 20 March 2019 through 23 March 2019
ER -