Pre-trained Vision Transformers for Seizure Prediction: A Reproducible Baseline with Event-Based Evaluation and Statistical Validation

Background: Scalp electroencephalography (EEG) based seizure prediction plays a critical role in improving the quality of life for patients with drug-resistant epilepsy, offering the potential for real-time warnings and timely interventions. Despite its clinical significance and decades of research, the field still lacks an open benchmark with reproducible baselines and deployment-oriented event-level evaluation. Most prior work relies on the small, outdated Children’s Hospital Boston (CHB-MIT) dataset and reports only window-level metrics, leaving the false-alarm burden of a real warning system underspecified. In seizure prediction, the cost of a false alarm is significantly high since patients may receive painful electrical stimulation to suppress seizures. Hence, false alarms per hour (FA/h) and partial AUC (pAUC) are the most deployment-relevant metrics, reflecting alarm burden and discriminability in the low-false-alarm operating region that a usable warning system can realistically tolerate. However, few studies have systematically reported such metrics. In addition, vision transformers’ event-level performance under deployable FA/h constraints remains underexplored, and newer backbones such as MambaVision have yet to be evaluated under this setting. Methods: In this work, we introduce a reproducible 5-fold benchmark derived from the Temple University Hospital EEG Seizure Corpus (TUSZ) dataset, and evaluate models using a pseudo-real-time event pipeline, reporting event-level sensitivity, false alarms per hour (FA/h) and partial AUC (pAUC). All models are compared to random predictors for statistical validation. We benchmark pre-trained vision transformers (SegFormer and MambaVision) under three EEG-to-image encoding methods, including a self-proposed Temporal-Patchify encoding for SegFormer. Results: Our proposed Temporal-Patchify encoding method achieves state-of-the-art performance. We achieved 0.61 pAUC, which is 16.2% higher than the baseline Temporal-Tile SegFormer of Parani et al. The false-alarm burden (0.40 FA/h) is 44.4% lower than the Temporal-Tile SegFormer baseline while maintaining clinically usable sensitivity (60.7%). We further perform statistical validation against a matched Poisson random predictor, confirming performance exceeds chance. Finally, we report end-to-end inference throughput up to 920 windows/s, confirming MambaVision’s fastest inference speed, exceeding SegFormer by over 20%. Conclusions: This work bridges the gap between seizure prediction algorithms and clinically usable seizure prediction systems in real-world settings. Our findings indicate that pre-trained vision transformers, when coupled with appropriate EEG encoding methods, can achieve robust performance in low–false-alarm operating regimes, which is critical for real-world deployment. This benchmark and evaluation framework may facilitate more clinically meaningful and reproducible seizure prediction research.