EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

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

Hongliu Yang iD, Jens Müller iD, Sotirios Kalousios iD, Georg Leonhardt iD, Matthias Dümpelmann iD, Andreas Schulze-Bonhage iD, Ronald Tetzlaff iD

Epilepsia · 2026-05 · 67 (5) · 2506-2517

DOI: 10.1002/epi.70110 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Epilepsia, online erschienen am 27.01.2026. Verfasst von Hongliu Yang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Electrical and Computer Engineering, Institute of Circuits and Systems, Technische Universität Dresden, Dresden, Germany. Hauptthemen laut MEDLINE: Seizures, Epilepsy, Brain. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Seizure risk is modulated by multiscale brain rhythms. Previous studies using cycles in electroencephalography, heart rate, and wearable data suggest the possibility of forecasting seizures days in advance. However, they commonly rely on methods requiring (days of) information from time points beyond the moment of forecast (noncausal processing). Although applicable to retrospective analyses, such an approach is not feasible for real‐time clinical applications. Here, we systematically investigated the impact of using only past data (causal processing) to estimate long‐timescale cycles of epileptic brain activity on forecasting performance.

Methods

We analyzed long‐term interictal epileptiform activity (IEA) recordings from 18 patients implanted with the RNS system. Circadian/multidien IEA cycles were extracted using two approaches: (1) noncausal filters also requiring future data; and (2) causal filters utilizing only past data, simulating real‐time scenarios. Forecasting was performed using Poisson regression, and the performance was benchmarked against established baseline models.

Results

Noncausal models achieved high performance, consistent with previous literature. However, when strict causal constraints were applied, performance deteriorated significantly (, Wilcoxon signed‐rank test). For hourly forecasting, the median area under the curve dropped from .76 to .63, time in warning increased from 35% to 60%, and patients with above‐chance performance decreased from 89% to 72%. Crucially, for daily forecasting, the causal models failed to perform better than chance for all patients. Performance loss was attributable to the input–output time lag and cycle deformation inherent to filters when analyzing data in a real‐time scenario without access to future signal trends. Significance 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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SeizuresEpilepsyBrainHumansForecastingElectroencephalographyPrediction AlgorithmsRetrospective Studies

Stichwörter der Autor:innen

clinical translationinterictal epileptiform activityperformance gapseizure prediction

Autor:innen und Institutionen

  • Faculty of Electrical and Computer Engineering, Institute of Circuits and Systems, Technische Universität Dresden, Dresden, Germany.
  • Epilepsy Center, Medical Center-University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany.
  • BrainLinks-BrainTools, Freiburg, Germany.
  • Department of Neurosurgery, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany.
  • Department of Microsystems Engineering-IMTEK, University of Freiburg, Freiburg, Germany.

ORCID-Profile: Hongliu Yang, Jens Müller, Sotirios Kalousios, Georg Leonhardt, Matthias Dümpelmann, Andreas Schulze-Bonhage, Ronald Tetzlaff

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Yang, H., Müller, J., Kalousios, S., Leonhardt, G., Dümpelmann, M., Schulze-Bonhage, A., & Tetzlaff, R. (2026). Seizure forecasting with epilepsy cycles: On the causality of forecasting pipelines. Epilepsia, 67(5), 2506-2517. https://doi.org/10.1002/epi.70110

Vancouver

Yang H, Müller J, Kalousios S, Leonhardt G, Dümpelmann M, Schulze-Bonhage A, et al. Seizure forecasting with epilepsy cycles: On the causality of forecasting pipelines. Epilepsia. 2026;67(5):2506-2517. doi:10.1002/epi.70110

BibTeX
@article{Yang2026,
  author = {Yang, Hongliu and Müller, Jens and Kalousios, Sotirios and Leonhardt, Georg and Dümpelmann, Matthias and Schulze-Bonhage, Andreas and Tetzlaff, Ronald},
  title = {Seizure forecasting with epilepsy cycles: On the causality of forecasting pipelines},
  journal = {Epilepsia},
  year = {2026},
  volume = {67},
  number = {5},
  pages = {2506--2517},
  doi = {10.1002/epi.70110}
}
Infokarte | EUVIMED

Artikeldaten derzeit nicht verfügbar.

Recherche