Epileptic Seizure Prediction Using Patient-Adaptive Transformer Networks

TL;DR

This paper introduces a patient-adaptive transformer framework for epileptic seizure prediction from EEG signals, combining self-supervised learning and patient-specific fine-tuning to improve accuracy.

Contribution

It presents a novel two-stage training approach that leverages self-supervised pretraining and patient-specific fine-tuning for improved seizure forecasting.

Findings

Achieved over 90% validation accuracy in seizure prediction.

F1 scores exceeded 0.80 across evaluated patients.

Demonstrated effectiveness of combining self-supervised learning with patient adaptation.

Abstract

Epileptic seizure prediction from electroencephalographic (EEG) recordings remains challenging due to strong inter-patient variability and the complex temporal structure of neural signals. This paper presents a patient-adaptive transformer framework for short-horizon seizure forecasting. The proposed approach employs a two-stage training strategy: self-supervised pretraining is first used to learn general EEG temporal representations through autoregressive sequence modeling, followed by patient-specific fine-tuning for binary prediction of seizure onset within a 30-second horizon. To enable transformer-based sequence learning, multichannel EEG signals are processed using noise-aware preprocessing and discretized into tokenized temporal sequences. Experiments conducted on subjects from the TUH EEG dataset demonstrate that the proposed method achieves validation accuracies above 90% and F1 scores exceeding 0.80 across evaluated patients, supporting the effectiveness of combining self-supervised representation learning with patient-specific adaptation for individualized seizure prediction.