Continuous sensorimotor rhythm based brain computer interface learning in a large population

James R Stieger,Stephen A Engel,Bin He

doi:10.1038/s41597-021-00883-1

James R Stieger, Stephen A Engel + Show 1 more

Open Access

https://doi.org/10.1038/s41597-021-00883-1

Copy DOI

Abstract

Brain computer interfaces (BCIs) are valuable tools that expand the nature of communication through bypassing traditional neuromuscular pathways. The non-invasive, intuitive, and continuous nature of sensorimotor rhythm (SMR) based BCIs enables individuals to control computers, robotic arms, wheel-chairs, and even drones by decoding motor imagination from electroencephalography (EEG). Large and uniform datasets are needed to design, evaluate, and improve the BCI algorithms. In this work, we release a large and longitudinal dataset collected during a study that examined how individuals learn to control SMR-BCIs. The dataset contains over 600 hours of EEG recordings collected during online and continuous BCI control from 62 healthy adults, (mostly) right hand dominant participants, across (up to) 11 training sessions per participant. The data record consists of 598 recording sessions, and over 250,000 trials of 4 different motor-imagery-based BCI tasks. The current dataset presents one of the largest and most complex SMR-BCI datasets publicly available to date and should be useful for the development of improved algorithms for BCI control.

Highlights

Background & SummaryMillions of individuals live with paralysis[1]
Each individual participant completed 7–11 online Brain computer interfaces (BCIs) training sessions, and the dataset includes on average 4,340 trials and 9.9 hours’ worth of EEG data per participant. This dataset contains roughly 4.5 times as many trials and 10 times as much data as the largest dataset currently available for public use[44]. We believe this dataset should be of particular value to the field for four reasons: (1) the amount of EEG data is sufficient to train large decoding models, (2) the sample size permits tests of how well decoding models and signal processing techniques will generalize, (3) the BCI decoding tasks are challenging, and (4) the longitudinal study design enables tests of how well decoding models and signal processing techniques adapt to session by session changes
The main goals of our original study were to characterize how individuals learn to control sensorimotor rhythm (SMR)-BCIs and to test whether this learning can be improved through behavioral interventions such as mindfulness training[30]

Summary

Introduction

Background & SummaryMillions of individuals live with paralysis[1]. The emerging field of neural prosthetics seeks to provide relief to these individuals by forging new pathways of communication and control[2,3,4,5,6]. This dataset comprises 62 participants, 598 individual BCI sessions, and over 600 hours of high-density EEG recordings (64 channels) from 269,099 trials.

Results

Conclusion