HyBrain model improves seizure detection and prediction from EEG
Spatiotemporal Hyperedges for EEG Seizure Detection and Prediction
Artificial Intelligence
Summary
Seizures are hard to predict and detect because they happen rarely and involve complex brain activity across many areas. The authors present HyBrain, a new method that looks at EEG brain signals by grouping multiple channels together, rather than just pairs, to better capture these patterns over time. This method works faster and uses less computer memory than previous approaches, while improving accuracy in spotting seizures and even predicting them before they happen.
What this means in practice
- •For hospital data teams: Improve accuracy and efficiency of automated seizure monitoring by integrating HyBrain’s spatiotemporal hyperedge modeling into EEG analysis workflows.
- •For wearable device developers: Develop more reliable seizure prediction features in wearable EEG devices by using HyBrain’s compact and efficient spatiotemporal modeling.$Commercial implications: This paper enables consumer-grade wearable devices to better predict seizures, offering a valuable health monitoring product for epilepsy patients.
Authors
Hyunju Kim, Sheo Yon Jhin, Noseong Park, Nabil Imam
Abstract
Seizure detection and prediction from EEG are clinically important but challenging because seizures are rare, temporally localized, and propagate as coordinated events across multiple channels. Recent dynamic graph neural networks model this by running a temporal model over a sequence of per-time-step pairwise channel edges. However, this pairwise construction misses the spatiotemporal coupling that constitutes a seizure, at substantial training cost. We propose HyBrain, which summarizes spatiotemporal EEG evidence through a small set of soft hyperedges rather than pairwise edges. A per-channel Mamba backbone produces one token per (channel, second), and a spatiotemporal hyperedge block pools these tokens into E_h shared group embeddings through soft memberships and broadcasts them back. The same encoder serves three downstream tasks: window-based detection, one-second point-wise detection, and preictal seizure prediction. On TUSZ and CHB-MIT, HyBrain achieves the best AUROC on every reported setting against ten baselines, with the largest gap on long-clip preictal prediction. It also matches the most efficient baselines in training time and peak GPU memory. A qualitative analysis shows that even a single learned hyperedge cleanly captures the preictal -> ictal -> postictal trajectory on a real seizure clip.