SeizureFormer: A Transformer Model for IEA-Based Seizure Risk Forecasting

Unlike raw scalp EEG-based models, SeizureFormer leverages structured, clinically relevant features and integrates CNN-based patch embedding, multi-head self-attention, and squeeze-and-excitation blocks to model both short-term dynamics and long-term seizure cycles. Tested across five patients and multiple prediction windows (1-14 days), SeizureFormer achieved state-of-the-art performance with mean ROC AUC of 79.44% and mean PR AUC of 76.29% . Compared to statistical, machine learning, and deep learning baselines, it demonstrates enhanced generaliz-ability and seizure risk forecasting performance under class imbalance. This work supports future clinical integration of interpretable and robust seizure forecasting tools for personalized epilepsy management. Introduction Epilepsy is a chronic neurological disorder affecting around 50 million people worldwide, marked by recurrent and unpredictable seizures that significantly disrupt daily life.