Medium density EMG armband for gesture recognition DOI Creative Commons

Eisa Aghchehli,

Milad Jabbari, Chenfei Ma

et al.

Frontiers in Neurorobotics, Journal Year: 2025, Volume and Issue: 19

Published: April 30, 2025

Electromyography (EMG) systems are essential for the advancement of neuroprosthetics and human-machine interfaces. However, gap between low-density high-density poses challenges to researchers in experiment design knowledge transfer. Medium-density surface EMG offer a balanced alternative, providing greater spatial resolution than while avoiding complexity cost arrays. In this study, we developed research-friendly medium-density system evaluated its performance with eleven volunteers performing grasping tasks. To enhance decoding accuracy, introduced novel spatio-temporal convolutional neural network that integrates information from additional sensors temporal dynamics. The results show significantly improve classification accuracy compared maintaining same footprint. Furthermore, proposed outperforms traditional gesture approaches. This work highlights potential as practical effective solution, bridging low- systems. These findings pave way broader adoption research clinical applications.

Language: Английский

Medium density EMG armband for gesture recognition DOI Creative Commons

Eisa Aghchehli,

Milad Jabbari, Chenfei Ma

et al.

Frontiers in Neurorobotics, Journal Year: 2025, Volume and Issue: 19

Published: April 30, 2025

Electromyography (EMG) systems are essential for the advancement of neuroprosthetics and human-machine interfaces. However, gap between low-density high-density poses challenges to researchers in experiment design knowledge transfer. Medium-density surface EMG offer a balanced alternative, providing greater spatial resolution than while avoiding complexity cost arrays. In this study, we developed research-friendly medium-density system evaluated its performance with eleven volunteers performing grasping tasks. To enhance decoding accuracy, introduced novel spatio-temporal convolutional neural network that integrates information from additional sensors temporal dynamics. The results show significantly improve classification accuracy compared maintaining same footprint. Furthermore, proposed outperforms traditional gesture approaches. This work highlights potential as practical effective solution, bridging low- systems. These findings pave way broader adoption research clinical applications.

Language: Английский

Citations

0