Motor Imagery Classification based on CNN-GRU Network with Spatio-Temporal Feature Representation

TL;DR

This paper introduces a CNN-GRU based model that effectively classifies EEG signals by extracting spatio-temporal features, achieving superior accuracy over existing methods in motor imagery classification tasks.

Contribution

The study proposes a novel combined CNN-GRU model utilizing covariance matrices for improved spatio-temporal EEG feature extraction and classification.

Findings

Achieved 77.70% accuracy on BCI competition IV_2a dataset.

Outperformed baseline methods in motor imagery EEG classification.

Demonstrated the effectiveness of separating spatial and temporal processing.

Abstract

Recently, various deep neural networks have been applied to classify electroencephalogram (EEG) signal. EEG is a brain signal that can be acquired in a non-invasive way and has a high temporal resolution. It can be used to decode the intention of users. As the EEG signal has a high dimension of feature space, appropriate feature extraction methods are needed to improve classification performance. In this study, we obtained spatio-temporal feature representation and classified them with the combined convolutional neural networks (CNN)-gated recurrent unit (GRU) model. To this end, we obtained covariance matrices in each different temporal band and then concatenated them on the temporal axis to obtain a final spatio-temporal feature representation. In the classification model, CNN is responsible for spatial feature extraction and GRU is responsible for temporal feature extraction. Classification performance was improved by distinguishing spatial data processing and temporal data processing. The average accuracy of the proposed model was 77.70% for the BCI competition IV_2a data set. The proposed method outperformed all other methods compared as a baseline method.