Key result
Attention-based graph neural networks achieve ~93% EEG accuracy, outperforming methods ignoring spatial correlations.
Why the study?
Existing deep learning methods for EEG signal classification rarely consider the correlation and connections between different brain regions and electrodes.
A multi-scale attention-based graph convolutional neural network improves EEG-based motion imagery classification accuracy to 93.22% by incorporating spatial correlations between electrodes.
Answers are grounded in this paper and cite their sources.
May enhance BCI development for motor imagery; leaves open clinical validation before neurorehabilitation use.
Recently, deep learning has been widely used in the classification of EEG signals and achieved satisfactory results. However, the correlation between EEG electrodes is rarely considered, which has been proved that there are indeed connections between different brain regions. After considering the connections between EEG electrodes, the graph convolutional neural network is applied to detect human motor intents from EEG signals, where EEG data are transformed into graph data through phase lag index, time-domain and frequency-domain features with different signal bands. Meanwhile, a multi-scale attention mechanism is proposed to the network to improve the accuracy of classification. By using the multi-scale attention-based graph convolutional neural network, the accuracy of 93.22% is achieved with 10-fold cross-validation, which is higher than the compared methods which ignore the spatial correlations of EEG signals.
No takes yet. Share an insight, caveat, or question.
Zhu et al. (2023) studied EEG-Based Motion Imagery Classification. Multi-scale attention-based graph convolutional neural network vs. Methods which ignore the spatial correlations of EEG signals was evaluated on Classification accuracy. A multi-scale attention-based graph convolutional neural network achieved a classification accuracy of 93.22% with 10-fold cross-validation, outperforming methods that ignore spatial correlations.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: