An Impending Paradigm Shift in Motor Imagery Based Brain-Computer Interfaces

Abstract

The development of reliable assistive devices for patients that suffer from motor impairments following central nervous system lesions remains a major challenge in the field of non-invasive Brain-Computer Interfaces (BCIs). These approaches are predominated by electroencephalography and rely on advanced signal processing and machine learning methods to extract neural correlates of motor activity. However, despite tremendous and still ongoing efforts, their value as effective clinical tools remains limited. We advocate that a rather overlooked research avenue lies in efforts to question neurophysiological markers traditionally targeted in non-invasive motor BCIs. We propose an alternative approach grounded by recent fundamental advances in non-invasive neurophysiology, specifically subject-specific feature extraction of sensorimotor bursts of activity recorded via (possibly magnetoencephalography-optimized) electroencephalography. This path holds promise in overcoming a significant proportion of existing limitations, and could foster the wider adoption of online BCIs in rehabilitation protocols.

Keywords: Brain-Computer Interface (BCI); EEG; beta bursts; magnetoencephalography (MEG); motor imagery (MI); neurological rehabilitation; upper limb.

FIGURE 1

Trial-averaged discrete bursts can appear to be sustained oscillations. Simulated beta burst activity from multivariate Gaussian distributions with a time varying probability and random peak frequency, frequency span, and time duration. (A) Left: The timing of simulated bursts in each trial (N = 1,000). Right: Time-frequency decomposition analysis of each single trial level (shown for four random ones) allows for the extraction of features such as the exact timing and peak frequency (orange circle), the time duration (red vertical arrows) and the frequency span (light blue horizontal arrows) of each burst. (B) Beta band power of the same four random trials from the right panel of (A) (colored lines) depicted along with the average beta band power over all trials (black line). During each trial, beta power appears as transient peaks at varying time points. The classically described ERD and ERS phenomena emerge as a consequence of averaging over multiple trials.

FIGURE 2

Frequency specific activity can be non-sinusoidal. Mu activity from EEG electrode C4. Inset: Cycle-by-cycle analysis (Cole and Voytek, 2019) of the activity reveals that mu occurs in bursts with significant variability for measures such as peaks (magenta dots) and troughs (pink dots), or rise (yellow dots) and decay (blue dots) duration between cycles, unlike the corresponding measures of a pure 10 Hz sinusoid (orange and green dots). EEG data from BCI Competition IV (Leeb et al., 2007) https://www.bbci.de/competition/iv.