PMID- 31353866 OWN - NLM STAT- MEDLINE DCOM- 20200513 LR - 20200513 IS - 2328-9503 (Electronic) IS - 2328-9503 (Linking) VI - 6 IP - 7 DP - 2019 Jul TI - Comparison of machine learning models for seizure prediction in hospitalized patients. PG - 1239-1247 LID - 10.1002/acn3.50817 [doi] AB - 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. CI - (c) 2019 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. FAU - Struck, Aaron F AU - Struck AF AUID- ORCID: 0000-0002-9103-1798 AD - Department of Neurology, University of Wisconsin, Madison, Wisconsin. FAU - Rodriguez-Ruiz, Andres A AU - Rodriguez-Ruiz AA AD - Department of Neurology, Emory University, Atlanta, Georgia. FAU - Osman, Gamaledin AU - Osman G AD - Department of Neurology, Henry Ford Hospital, Detroit, Michigan. FAU - Gilmore, Emily J AU - Gilmore EJ AD - Department of Neurology, Yale University, New Haven, Connecticut. FAU - Haider, Hiba A AU - Haider HA AUID- ORCID: 0000-0001-7118-3690 AD - Department of Neurology, Emory University, Atlanta, Georgia. FAU - Dhakar, Monica B AU - Dhakar MB AD - Department of Neurology, Emory University, Atlanta, Georgia. FAU - Schrettner, Matthew AU - Schrettner M AD - Department of Neurology, University of South Carolina Greenville, Greenville, South Carolina. FAU - Lee, Jong W AU - Lee JW AD - Department of Neurology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts. FAU - Gaspard, Nicolas AU - Gaspard N AD - Department of Neurology, Yale University, New Haven, Connecticut. AD - Departement de Neurologie, Universite Libre de Bruxelles, Hopital Erasme, Bruxelles, Belgium. FAU - Hirsch, Lawrence J AU - Hirsch LJ AD - Department of Neurology, Yale University, New Haven, Connecticut. FAU - Westover, M Brandon AU - Westover MB AD - Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts. CN - Critical Care EEG Monitoring Research Consortium (CCERMRC) LA - eng GR - UL1 TR001863/TR/NCATS NIH HHS/United States GR - Epilepsy Foundation of America/International GR - American Epilepsy Society/International PT - Journal Article PT - Research Support, Non-U.S. Gov't DEP - 20190627 PL - United States TA - Ann Clin Transl Neurol JT - Annals of clinical and translational neurology JID - 101623278 SB - IM MH - Aged MH - Aged, 80 and over MH - Cohort Studies MH - Critical Care MH - Electroencephalography MH - Female MH - Humans MH - *Machine Learning MH - Male MH - Monitoring, Physiologic MH - Neural Networks, Computer MH - Seizures/*diagnosis MH - Young Adult PMC - PMC6649418 COIS- 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. EDAT- 2019/07/30 06:00 MHDA- 2020/05/14 06:00 PMCR- 2019/06/27 CRDT- 2019/07/30 06:00 PHST- 2019/03/27 00:00 [received] PHST- 2019/05/21 00:00 [revised] PHST- 2019/05/23 00:00 [accepted] PHST- 2019/07/30 06:00 [entrez] PHST- 2019/07/30 06:00 [pubmed] PHST- 2020/05/14 06:00 [medline] PHST- 2019/06/27 00:00 [pmc-release] AID - ACN350817 [pii] AID - 10.1002/acn3.50817 [doi] PST - ppublish SO - Ann Clin Transl Neurol. 2019 Jul;6(7):1239-1247. doi: 10.1002/acn3.50817. Epub 2019 Jun 27.