AI Insight
Researchers developed SpaTeoGL, a machine learning framework that analyzes intracranial EEG recordings to identify seizure onset zones in epilepsy patients. The method simultaneously learns spatial relationships between brain electrodes and temporal patterns across time windows, using graph-based signal processing techniques. Testing on multi-center clinical data from patients with successful surgical outcomes showed the approach performs comparably to existing methods while providing better identification of non-seizure areas and more interpretable insights into how seizures begin and spread.
Why it matters
Accurate identification of seizure onset zones is critical for successful epilepsy surgery, as removing the wrong brain tissue can lead to continued seizures or unnecessary damage. This interpretable approach could help neurosurgeons make more informed decisions about which brain regions to remove, potentially improving surgical outcomes for drug-resistant epilepsy patients.
Understand the Science
Abstract: Accurate localization of the seizure onset zone (SOZ) from intracranial EEG (iEEG) is essential for epilepsy surgery but is challenged by complex spatiotemporal seizure dynamics. We propose SpaTeoGL, a spatiotemporal graph learning framework for interpretable seizure network analysis. SpaTeoGL jointly learns window-level spatial graphs capturing interactions among iEEG electrodes and a temporal graph linking time windows based on similarity of their spatial structure. The method is formulated within a smooth graph signal processing framework and solved via an alternating block coordinate descent algorithm with convergence guarantees. Experiments on a multicenter iEEG dataset with successful surgical outcomes show that SpaTeoGL is competitive with a baseline based on horizontal visibility graphs and logistic regression, while improving non-SOZ identification and providing interpretable insights into seizure onset and propagation dynamics.