Reconstruction of reaching movement trajectories using electrocorticographic signals in humans

Abstract

In this study, we used electrocorticographic (ECoG) signals to extract the onset of arm movement as well as the velocity of the hand as a function of time. ECoG recordings were obtained from three individuals while they performed reaching tasks in the left, right and forward directions. The ECoG electrodes were placed over the motor cortex contralateral to the moving arm. Movement onset was detected from gamma activity with near perfect accuracy (> 98%), and a multiple linear regression model was used to predict the trajectory of the reaching task in three-dimensional space with an accuracy exceeding 85%. An adaptive selection of frequency bands was used for movement classification and prediction. This demonstrates the efficacy of developing a real-time brain-machine interface for arm movements with as few as eight ECoG electrodes.

Fig 1

A) Location of implanted ECoG contacts shown for Participant 2. Location of ECoG electrodes was identical for all participants. Contacts for the first strip were labeled 0–3 from distal to proximal relative to the electrode connector, and contacts of the second strip were similarly indexed 4–7. The functional location of Contact 1 was confirmed by electrically stimulating the cortex and observing finger or wrist movements of the contralateral upper limb. B) Movements performed by the participants of the study: reaching a target placed 30 cm to the middle, right, and left of the individual's midline.

Fig 2. Example of recordings.

A) Traces showing raw ECoG signals and EMG recorded during a reaching task for Participant 3. B) Arm velocity in three-dimensions during the task.

Fig 3. Changes in spectral density of ECoG contact located over primary motor cortex derived from activity of Contact 1.

Changes shown in dB with accompanying EMG responses. (A) Response of Participant 1 shows ERS in gamma band (center frequency 158 Hz) and slow oscillations (0–2 Hz) as well as ERD in alpha-beta band (10–30 Hz). Time required for participants to reach target and to return to initial position was approximately 2 seconds. Vertical black lines indicate frequency bands where ERS/ERD was reduced by 50% of its peak value (3 dB drop). (B) Results for Participant 2 and (C) Participant 3.

Fig 4

Duration of activity plotted versus EMG activity for Participants 1–3 together with regression lines for gamma band activity (A-C) and beta band activity (D-F).

Fig 5. Data of Participant 3.

(A) Gamma power (70–90 Hz) during 50 consecutive trials. EMG onsets are aligned at t = 0. Each trial contains both reaching and retrieval movements. (B) Time course of muscle activity and detected movement onsets. Four typical movement cycles are shown. The solid line shows power as calculated from biceps muscle activity (i.e. the square of the EMG signal) and red vertical dashed lines represent detected movement onsets.

Fig 6. Effect of bandwidth used to calculate the band power for Participant 3.

Ratio between movement (0-1sec) and pre-movement (-1 to 0sec) gamma power plotted as a function of bandwidth. Both mean and MSE are shown (solid and dotted lines respectively). The ratio declines monotonically and approaches 1 indicating poor discrimination between movement and rest when using large bandwidths.

Fig 7. Arm velocity predicted by the multiple linear regression model.

Prediction (solid line) and actual velocity (dotted line) over two different trials of reaching right (x-axis component) for Participant 3.