March 3, 2026Open Access

Seizure forecasting with epilepsy cycles: On the causality of forecasting pipelines

Key Points

Forecasting algorithms show a significant gap between prospective and retrospective abilities, indicating limitations in current methods.
This performance gap implies that previous studies may have overestimated the effectiveness of seizure forecasts using noncausal models.
Novel cycle-extraction methods are necessary to improve forecasting accuracy in real-time predictive settings, especially for epilepsy cycles.
Future research should prioritize developing robust algorithms that can work effectively with historical data in real-time.

Abstract

Clinically beneficial seizure forecasting requires algorithms that function prospectively, using only past data. Our findings reveal a substantial performance gap between theoretical (noncausal, retrospective) performance and realistic (causal, prospective) capabilities. This suggests that prior studies may have overestimated forecasting accuracy. Future research must focus on developing novel cycle-extraction methods that remain robust under real-time prospective conditions.

Seizure forecasting with epilepsy cycles: On the causality of forecasting pipelines

Key Points

Abstract

Cite This Study