SeizureFormer: A Transformer Model for IEA-Based Seizure Risk Forecasting
This work addresses personalized epilepsy management by developing an interpretable forecasting tool, though it is incremental as it applies a Transformer model to a specific clinical domain.
The paper tackled seizure risk forecasting using interictal epileptiform activity and long episode biomarkers from neurostimulation systems, achieving state-of-the-art performance with mean ROC AUC of 79.44% and mean PR AUC of 76.29% across five patients and multiple prediction windows.
We present SeizureFormer, a Transformer-based model for long-term seizure risk forecasting using interictal epileptiform activity (IEA) surrogate biomarkers and long episode (LE) biomarkers from responsive neurostimulation (RNS) systems. 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 to 14 days), SeizureFormer achieved state-of-the-art performance with mean ROC AUC of 79.44 percent and mean PR AUC of 76.29 percent. Compared to statistical, machine learning, and deep learning baselines, it demonstrates enhanced generalizability 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.