Sinc-based convolutional neural networks for EEG-BCI-based motor imagery classification

TL;DR

This paper introduces Sinc-EEGNet, a lightweight CNN architecture that explicitly learns frequency domain features for EEG-based motor imagery classification, outperforming existing methods on public datasets.

Contribution

The paper proposes Sinc-EEGNet, a novel CNN architecture with learnable band-pass filters that addresses CNN limitations in EEG classification tasks.

Findings

Sinc-EEGNet outperforms reference methods in accuracy

The model is lightweight and requires fewer training data

Explicit frequency domain learning improves classification

Abstract

Brain-Computer Interfaces (BCI) based on motor imagery translate mental motor images recognized from the electroencephalogram (EEG) to control commands. EEG patterns of different imagination tasks, e.g. hand and foot movements, are effectively classified with machine learning techniques using band power features. Recently, also Convolutional Neural Networks (CNNs) that learn both effective features and classifiers simultaneously from raw EEG data have been applied. However, CNNs have two major drawbacks: (i) they have a very large number of parameters, which thus requires a very large number of training examples; and (ii) they are not designed to explicitly learn features in the frequency domain. To overcome these limitations, in this work we introduce Sinc-EEGNet, a lightweight CNN architecture that combines learnable band-pass and depthwise convolutional filters. Experimental results obtained on the publicly available BCI Competition IV Dataset 2a show that our approach outperforms reference methods in terms of classification accuracy.