Can EEG resting state data benefit data-driven approaches for motor-imagery decoding?

TL;DR

This study explores integrating resting-state EEG data into motor imagery decoding models to enhance BCI performance, but finds limited benefits for model generalization across users, highlighting the need for further research.

Contribution

Proposes a feature concatenation method combining resting-state EEG with EEGNet to improve motor imagery decoding and develop user-generalized models.

Findings

Improved mean accuracy in within-user scenarios on two datasets.

Limited benefits observed for across-user generalization.

Highlights need for further investigation into model interpretability and data concatenation effects.

Abstract

Resting-state EEG data in neuroscience research serve as reliable markers for user identification and reveal individual-specific traits. Despite this, the use of resting-state data in EEG classification models is limited. In this work, we propose a feature concatenation approach to enhance decoding models' generalization by integrating resting-state EEG, aiming to improve motor imagery BCI performance and develop a user-generalized model. Using feature concatenation, we combine the EEGNet model, a standard convolutional neural network for EEG signal classification, with functional connectivity measures derived from resting-state EEG data. The findings suggest that although grounded in neuroscience with data-driven learning, the concatenation approach has limited benefits for generalizing models in within-user and across-user scenarios. While an improvement in mean accuracy for within-user scenarios is observed on two datasets, concatenation doesn't benefit across-user scenarios when compared with random data concatenation. The findings indicate the necessity of further investigation on the model interpretability and the effect of random data concatenation on model robustness.