Hand kinematics, high-density sEMG comprising forearm and far-field potentials for motion intent recognition

Area of Science:

• Biomedical Engineering
• Neuroscience
• Rehabilitation Engineering

Background:

• Surface electromyography (sEMG) signals offer insights into spinal motor neuron activity for human-machine interaction (HMI).
• High-density (HD) sEMG can provide robust neural drives for HMI by decomposing motor neuron potentials from forearm and wrist muscles.
• A lack of publicly available datasets combining HD sEMG of forearm-wrist (FW) muscles and hand kinematics (KIN) hinders research.

Purpose of the Study:

• To present the first comprehensive HD-FW KIN dataset, integrating HD sEMG with finger kinematics and forces.
• To validate the utility of HD sEMG for hand gesture recognition and prediction of finger movements.
• To facilitate the development of advanced prosthetic devices and wearable electronics.

Main Methods:

• Collected HD 448-channel sEMG data from forearm and wrist muscles of 21 subjects.
• Simultaneously recorded finger joint angles and finger flexion forces.
• Acquired data during 20 distinct hand gestures and 9 finger flexion tasks under varying force levels.

Main Results:

• The HD-FW KIN dataset includes detailed muscle activity and hand movement data.
• Validated the effectiveness of HD sEMG for recognizing hand gestures.
• Demonstrated the capability of HD sEMG for predicting finger joint angles and flexion forces.

Conclusions:

• The presented HD-FW KIN dataset provides a valuable resource for extracting neural drives from forearm and wrist.
• Enables the creation of more intuitive neural interfaces for advanced prosthetic hands and wrist-worn devices.
• Advances the field of HMI through robust movement intent recognition.