Extracting kinematic parameters for monkey bipedal walking from cortical neuronal ensemble activity.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Robotics

Background:

• Neurological injuries and diseases significantly impair walking ability.
• Brain-machine interfaces (BMIs) have shown success in restoring upper-limb function but not walking.
• Restoring locomotion is a critical unmet need for individuals with paralysis.

Purpose of the Study:

• To investigate the potential of BMIs to restore walking by decoding cortical activity.
• To evaluate the real-time prediction of bipedal walking kinematics using neural ensembles.
• To assess the feasibility of using neural decoding for gait restoration.

Main Methods:

• Chronic recordings of cortical neuron ensembles in rhesus macaques.
• Development of linear decoders to extract 3D leg joint coordinates and muscle activity from neural data.
• Implementation of a switching decoder to adapt to different walking paradigms (speed, direction).

Main Results:

• Accurate offline and real-time prediction of bipedal walking kinematics was achieved.
• Larger neuronal populations were required for decoding more complex walking patterns.
• The switching decoder significantly improved decoding accuracy across various gaits.

Conclusions:

• Cortical neuron ensembles can be decoded to predict walking kinematics.
• BMIs offer a potential future solution for restoring walking in paralyzed patients.
• This research paves the way for neuroprosthetic devices to enable ambulation.