About this item:

192 Views | 280 Downloads

Author Notes:

Aaron F. Struck, 7131 MFCB, 600 Highland Ave, Madison, WI 53705. Tel: +1 608‐628‐4630; Fax: 608-263-0412; E‐mail: afstruck@wisc.edu

Lawrence J. Hirsch has research support to Yale University for investigator‐initiated studies from Monteris, Upsher‐Smith, and The Daniel Raymond Wong Neurology Research Fund at Yale; consultation fees for advising from Adamas, Aquestive, Ceribell, Eisai, Medtronic and UCB; royalties for authoring chapters for UpToDate‐Neurology and from Wiley for co‐authoring the book “Atlas of EEG in Critical Care,” by Hirsch and Brenner; honoraria for speaking from Neuropace.

Monica B. Dhakar has received honorarium for consultancy from Adamas Pharmaceuticals and research support from Marinus Pharmaceuticals and UCB Biopharma for clinical trials.

The remaining authors have no conflict of interests.


Research Funding:

Development of the Critical Care EEG Monitoring Research Consortium database was supported by research infrastructure awards by the American Epilepsy Society and Epilepsy Foundation of America.


  • Science & Technology
  • Life Sciences & Biomedicine
  • Clinical Neurology
  • Neurosciences
  • Neurosciences & Neurology

Comparison of machine learning models for seizure prediction in hospitalized patients

Show all authors Show less authors


Journal Title:

Annals of Clinical and Translational Neurology


Volume 6, Number 7


, Pages 1239-1247

Type of Work:

Article | Final Publisher PDF


Objective: To compare machine learning methods for predicting inpatient seizures risk and determine the feasibility of 1-h screening EEG to identify low-risk patients (<5% seizures risk in 48 h). Methods: The Critical Care EEG Monitoring Research Consortium (CCEMRC) multicenter database contains 7716 continuous EEGs (cEEG). Neural networks (NN), elastic net logistic regression (EN), and sparse linear integer model (RiskSLIM) were trained to predict seizures. RiskSLIM was used previously to generate 2HELPS2B model of seizure predictions. Data were divided into training (60% for model fitting) and evaluation (40% for model evaluation) cohorts. Performance was measured using area under the receiver operating curve (AUC), mean risk calibration (CAL), and negative predictive value (NPV). A secondary analysis was performed using Monte Carlo simulation (MCS) to normalize all EEG recordings to 48 h and use only the first hour of EEG as a “screening EEG” to generate predictions. Results: RiskSLIM recreated the 2HELPS2B model. All models had comparable AUC: evaluation cohort (NN: 0.85, EN: 0.84, 2HELPS2B: 0.83) and MCS (NN: 0.82, EN; 0.82, 2HELPS2B: 0.81) and NPV (absence of seizures in the group that the models predicted to be low risk): evaluation cohort (NN: 97%, EN: 97%, 2HELPS2B: 97%) and MCS (NN: 97%, EN: 99%, 2HELPS2B: 97%). 2HELPS2B model was able to identify the largest proportion of low-risk patients. Interpretation: For seizure risk stratification of hospitalized patients, the RiskSLIM generated 2HELPS2B model compares favorably to the complex NN and EN generated models. 2HELPS2B is able to accurately and quickly identify low-risk patients with only a 1-h screening EEG.

Copyright information:

© 2019 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association.

This is an Open Access work distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Creative Commons License

Export to EndNote