Training wrist extensor function and detecting unwanted movement strategies in an EMG-controlled visuomotor task

Area of Science:

• Neuroscience
• Rehabilitation Engineering
• Biomedical Engineering

Background:

• Stroke frequently causes upper limb paresis, complicating early rehabilitation due to weak muscle contractions.
• Current rehabilitation often struggles with initiating motor tasks in patients with limited voluntary movement.

Purpose of the Study:

• To develop and evaluate a new training tool using electromyographic (EMG) signals for upper limb rehabilitation.
• To assess the feasibility and effectiveness of an EMG-based visuomotor task for stroke patients.

Main Methods:

• Utilized a Myo armband to develop a proportional EMG controller for modulating wrist extensor activity.
• Healthy participants performed a visuomotor task, with skill learning assessed via trial time reduction.
• An EMG-based classifier was implemented to differentiate desired movements from unwanted strategies.

Main Results:

• Significant reduction in average trial time was observed, indicating successful skill acquisition (p < 0.001).
• The EMG classifier achieved high accuracy (92.5 ± 6.9%) in distinguishing movement strategies.
• The system flagged a small percentage of trials as unwanted strategies even in healthy subjects.

Conclusions:

• The proposed EMG-based training tool shows promise for facilitating early-stage motor rehabilitation in stroke survivors.
• The system demonstrates effective skill learning and reliable classification of motor control strategies.
• Preliminary patient feedback suggests feasibility and acceptance of the rehabilitation setup.