Combining Euclidean Alignment and Data Augmentation for BCI decoding

TL;DR

This paper evaluates how combining Euclidean Alignment and Data Augmentation techniques enhances deep neural network performance in EEG signal classification, leading to significant accuracy improvements across models and datasets.

Contribution

It demonstrates that integrating EA and DA synergistically improves EEG decoding accuracy, especially when using shared models with fine-tuning.

Findings

Combining EA and DA improves model accuracy across datasets.

Shared models with fine-tuning benefit most, with an 8.41% accuracy increase.

Synergistic effects of EA and DA enhance EEG classification performance.

Abstract

Automated classification of electroencephalogram (EEG) signals is complex due to their high dimensionality, non-stationarity, low signal-to-noise ratio, and variability between subjects. Deep neural networks (DNNs) have shown promising results for EEG classification, but the above challenges hinder their performance. Euclidean Alignment (EA) and Data Augmentation (DA) are two promising techniques for improving DNN training by permitting the use of data from multiple subjects, increasing the data, and regularizing the available data. In this paper, we perform a detailed evaluation of the combined use of EA and DA with DNNs for EEG decoding. We trained individual models and shared models with data from multiple subjects and showed that combining EA and DA generates synergies that improve the accuracy of most models and datasets. Also, the shared models combined with fine-tuning benefited the most, with an overall increase of 8.41\% in classification accuracy.