EEG-Titans: Long-Horizon Seizure Forecasting via Dual-Branch Attention and Neural Memory

TL;DR

EEG-Titans introduces a dual-branch neural network with memory mechanisms for long-horizon seizure prediction from EEG, achieving high sensitivity and reduced false alarms in challenging clinical scenarios.

Contribution

The paper presents a novel dual-branch architecture with neural memory for improved long-term EEG seizure forecasting, addressing the trade-off between local pattern detection and long-range context.

Findings

Achieves 99.46% sensitivity on CHB-MIT dataset.

Reduces false alarms to 0.00 FPR/h in high-noise cases.

Demonstrates robustness in clinically constrained evaluation.

Abstract

Accurate epileptic seizure prediction from electroencephalography (EEG) remains challenging because pre-ictal dynamics may span long time horizons while clinically relevant signatures can be subtle and transient. Many deep learning models face a persistent trade-off between capturing local spatiotemporal patterns and maintaining informative long-range context when operating on ultralong sequences. We propose EEG-Titans, a dualbranch architecture that incorporates a modern neural memory mechanism for long-context modeling. The model combines sliding-window attention to capture short-term anomalies with a recurrent memory pathway that summarizes slower, progressive trends over time. On the CHB-MIT scalp EEG dataset, evaluated under a chronological holdout protocol, EEG-Titans achieves 99.46% average segment-level sensitivity across 18 subjects. We further analyze safety-first operating points on artifact-prone recordings and show that a hierarchical context strategy extending the receptive field for high-noise subjects can markedly reduce false alarms (down to 0.00 FPR/h in an extreme outlier) without sacrificing sensitivity. These results indicate that memory-augmented long-context modeling can provide robust seizure forecasting under clinically constrained evaluation