EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Learning to generalize seizure forecasts

Marc G Leguia iD, Vikram R Rao iD, Thomas K Tcheng iD, Jonas Duun-Henriksen iD, Troels W Kjaer, Timothée Proix, Maxime O Baud iD

Epilepsia · 2023-12 · 64 (S4)

DOI: 10.1111/epi.17406 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Epilepsia, online erschienen am 22.09.2022. Verfasst von Marc G Leguia und 6 weiteren Autor:innen. Beteiligte Einrichtung: Wyss Center Fellow, Sleep-Wake-Epilepsy Center, Center for Experimental Neurology, NeuroTec, Department of Neurology, Inselspital Bern University Hos…. Hauptthemen laut MEDLINE: Seizures, Epilepsy. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Epilepsy is characterized by spontaneous seizures that recur at unexpected times. Nonetheless, using years‐long electroencephalographic (EEG) recordings, we previously found that patient‐reported seizures consistently occur when interictal epileptiform activity (IEA) cyclically builds up over days. This multidien (multiday) interictal–ictal relationship, which is shared across patients, may bear phasic information for forecasting seizures, even if individual patterns of seizure timing are unknown. To test this rigorously in a large retrospective dataset, we pretrained algorithms on data recorded from a group of patients, and forecasted seizures in other, previously unseen patients.

Methods

We used retrospective long‐term data from participants ( N = 159) in the RNS System clinical trials, including intracranial EEG recordings (icEEG), and from two participants in the UNEEG Medical clinical trial of a subscalp EEG system (sqEEG). Based on IEA detections, we extracted instantaneous multidien phases and trained generalized linear models (GLMs) and recurrent neural networks (RNNs) to forecast the probability of seizure occurrence at a 24‐h horizon.

Results

With GLMs and RNNs, seizures could be forecasted above chance in 79% and 81% of previously unseen subjects with a median discrimination of area under the curve (AUC) = .70 and .69 and median Brier skill score (BSS) = .07 and .08. In direct comparison, individualized models had similar median performance (AUC = .67, BSS = .08), but for fewer subjects (60%). Moreover, calibration of pretrained models could be maintained to accommodate different seizure rates across subjects. Significance Our findings suggest that seizure forecasting based on multidien cycles of IEA can generalize across patients, and may drastically reduce the amount of data needed to issue forecasts for individuals who recently started collecting chronic EEG data. In addition, we show that this generalization is independent of the method used to record seizures (patient‐reported vs. electrographic) or IEA (icEEG vs. sqEEG).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SeizuresEpilepsyHumansRetrospective StudiesElectroencephalographyElectrocorticography

Stichwörter der Autor:innen

intracranial EEGmultidienseizure forecastingsubscalp EEGtransfer learning

Autor:innen und Institutionen

  • Wyss Center Fellow, Sleep-Wake-Epilepsy Center, Center for Experimental Neurology, NeuroTec, Department of Neurology, Inselspital Bern University Hospital, University of Bern, Bern, Switzerland.
  • Department of Neurology and Weill Institute for Neurosciences, University of California, University of California, San Francisco, California, USA.
  • NeuroPace, Mountain View, California, USA.
  • UNEEG Medical, Allerød, Denmark.
  • Department of Neurology, Zealand University Hospital, Roskilde, Denmark.
  • Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark.

ORCID-Profile: Marc G Leguia, Vikram R Rao, Thomas K Tcheng, Jonas Duun-Henriksen, Maxime O Baud

Zitieren

APA

Leguia, M. G., Rao, V. R., Tcheng, T. K., Duun-Henriksen, J., Kjaer, T. W., Proix, T., & Baud, M. O. (2023). Learning to generalize seizure forecasts. Epilepsia, 64(S4). https://doi.org/10.1111/epi.17406

Vancouver

Leguia MG, Rao VR, Tcheng TK, Duun-Henriksen J, Kjaer TW, Proix T, et al. Learning to generalize seizure forecasts. Epilepsia. 2023;64(S4). doi:10.1111/epi.17406

BibTeX
@article{Leguia2023,
  author = {Leguia, Marc G and Rao, Vikram R and Tcheng, Thomas K and Duun-Henriksen, Jonas and Kjaer, Troels W and Proix, Timothée and Baud, Maxime O},
  title = {Learning to generalize seizure forecasts},
  journal = {Epilepsia},
  year = {2023},
  volume = {64},
  number = {S4},
  doi = {10.1111/epi.17406}
}
Kostenlos starten →