Key result
A graph neural network with self-supervised pre-training achieved an AUROC of 0.875 for seizure detection and a weighted F1-score of 0.749 for seizure classification, outperforming previous methods.
Why the study?
Prior automated EEG seizure detection and classification models struggled to represent non-Euclidean data structures, accurately classify rare seizure types, and quantitatively localize seizures.
Absolute Event Rate: 0.875% vs 0.796%
A self-supervised graph neural network approach significantly improves automated seizure detection, classification, and localization on EEG compared to existing CNN methods.
May enhance rare seizure classification and EEG localization on non-Euclidean data; leaves open prospective clinical validation.
Automated seizure detection and classification from electroencephalography (EEG) can greatly improve seizure diagnosis and treatment. However, several modeling challenges remain unaddressed in prior automated seizure detection and classification studies: (1) representing non-Euclidean data structure in EEGs, (2) accurately classifying rare seizure types, and (3) lacking a quantitative interpretability approach to measure model ability to localize seizures. In this study, we address these challenges by (1) representing the spatiotemporal dependencies in EEGs using a graph neural network (GNN) and proposing two EEG graph structures that capture the electrode geometry or dynamic brain connectivity, (2) proposing a self-supervised pre-training method that predicts preprocessed signals for the next time period to further improve model performance, particularly on rare seizure types, and (3) proposing a quantitative model interpretability approach to assess a model's ability to localize seizures within EEGs. When evaluating our approach on seizure detection and classification on a large public dataset, we find that our GNN with self-supervised pre-training achieves 0.875 Area Under the Receiver Operating Characteristic Curve on seizure detection and 0.749 weighted F1-score on seizure classification, outperforming previous methods for both seizure detection and classification. Moreover, our self-supervised pre-training strategy significantly improves classification of rare seizure types. Furthermore, quantitative interpretability analysis shows that our GNN with self-supervised pre-training precisely localizes 25.4% focal seizures, a 21.9 point improvement over existing CNNs. Finally, by superimposing the identified seizure locations on both raw EEG signals and EEG graphs, our approach could provide clinicians with an intuitive visualization of localized seizure regions.
No takes yet. Share an insight, caveat, or question.
Tang et al. (2021) studied Seizures (n=5,499). Graph Neural Network (GNN) with self-supervised pre-training vs. Existing CNNs and RNNs (e.g., Dense-CNN, LSTM) was evaluated on Seizure detection (AUROC) and seizure classification (weighted F1-score). A graph neural network with self-supervised pre-training achieved an AUROC of 0.875 for seizure detection and a weighted F1-score of 0.749 for seizure classification, outperforming previous methods.