Decoding neural activity to predict rat locomotion using intracortical and epidural arrays

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Science

Background:

• Restoring voluntary gait after spinal cord injury (SCI) requires effective motor cortical commands.
• Functional electrical stimulation (FES) driven by cortical activity can restore movement.
• Stimulation synchronized with voluntary effort may enhance recovery.

Purpose of the Study:

• To evaluate the neural information needed to drive a cortically-controlled FES brain-machine interface (BMI) in rats.
• To compare the accuracy of predicting hindlimb electromyograms (EMG) and kinematics using intracortical versus epidural arrays.

Main Methods:

• Rats were trained to walk on a treadmill.
• Intracortical or epidural arrays were implanted in the hindlimb sensorimotor cortex.
• EMG, kinematics, and neural data were recorded and decoded using linear and nonlinear methods.

Main Results:

• Intracortical arrays successfully predicted hindlimb EMGs and kinematics from multiunit activity and local field potentials (LFPs).
• Epidural arrays provided essentially uninformative signals for predicting movement.
• Somatosensory evoked potentials (SSEPs) confirmed neural activity but not useful predictive information from epidural arrays.

Conclusions:

• Accurate EMG prediction from intracortical recordings is sufficient to drive an FES-BMI.
• This rat model will be used to evaluate cortically-controlled FES for restoring locomotion after SCI.
• FES-BMI holds potential as a rehabilitative technology for improving motor function.