抄録
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C), Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3300000, indirect: 990000)
Surface electromyography (sEMG) has been widely investigated as a biological signal from which motion intentions can be recognized to control prosthetic hands. The availability and quality of sEMG can limit the usability and intuitiveness of advanced prosthetic hands that can restore most necessary hand movements. This study introduces force myography (FMG) as a supplementary signal and develops a hybrid sensor to measure sEMG and FMG signals simultaneously. Furthermore, a layer-fusion convolutional neural network (CNN) was proposed to analyze the sEMG and FMG signals. The recognition results of hand motion showed a significantly improved classification accuracy (CA) of the hybrid sEMG and FMG with respect to individual modality due to the rubustness of FMG. The FMG-assisted sEMG sensing approach can effectively offer great potential in the clinical application of sophisticated prosthetic hands without increasing burden to the user.