Upper Limb Prosthesis Control: A Hybrid EEG-EMG Scheme for Motion Estimation in Transhumeral Subjects

Area of Science:

• Biomedical Engineering
• Neuroscience
• Rehabilitation Technology

Background:

• Upper limb prosthetics require intuitive control for improved functionality.
• Existing control systems often lack the dexterity and natural feel of biological limbs.
• Surface Electromyogram (EMG) and Electroencephalogram (EEG) offer potential biosignal sources for prosthetic control.

Purpose of the Study:

• To develop and validate a hybrid EEG-EMG strategy for estimating upper limb kinematics.
• To assess the accuracy of this hybrid approach for controlling prosthetic devices.
• To demonstrate the potential for creating more intuitive, multi-degree-of-freedom upper limb prosthesis controllers.

Main Methods:

• Utilized Kernel Least Square Tracker for 3D shoulder and elbow motion estimation from surface EMG.
• Employed a two-stage multiclass Support Vector Machine for classifying wrist, grip, and finger motions from EEG.
• Integrated EMG and EEG data in a hybrid strategy for comprehensive upper limb motion tracking.

Main Results:

• Achieved over 90% accuracy in estimating shoulder and elbow motion using EMG.
• Demonstrated 65%-70% accuracy in discerning wrist, grip, and finger motions via EEG.
• Successfully validated the hybrid EEG-EMG strategy on a transhumeral subject.

Conclusions:

• The hybrid EEG-EMG motion estimation strategy offers a robust method for upper limb control.
• This approach can significantly enhance the intuitiveness and functionality of upper limb prostheses.
• The findings pave the way for developing next-generation prosthetic controllers with multiple degrees of freedom.