Point-of-care Detection of Electrographic Seizures for Bedside Monitoring
Abstract number :
1.201
Submission category :
2. Translational Research / 2B. Devices, Technologies, Stem Cells
Year :
2024
Submission ID :
757
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Mitchell Frankel, PhD – Epitel, Inc.
Presenting Author: Mark Lehmkuhle, PhD – Epitel, Inc.
Avi Kazen, MS – Epitel, Inc.
Zoë Tosi, PhD – Epitel, Inc.
Tyler Newton, PhD – Epitel, Inc.
Vamshi Muvvala, MS – Epitel, Inc.
Tobias Loddenkemper, MD – Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
Lillian Voke, BS – UMass Chan Medical School
Edeline Jean Baptiste, BS – Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
Stephanie Dailey, BA – Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
Michele Jackson, BA – Boston Childrens Hospital
Latania Reece, BA – Boston Children’s Hospital
Claire Ufongene, MD – Boston Children’s Hospital
Mark Spitz, MD – University of Colorado - Anschutz
Laura Strom, MD – University of Colorado - Anschutz
Meagan Watson, MPH, MBAc – University of Colorado School of Medicine
Mackenzi Moore, MPH – University of Colorado - Anschutz
Trey Jouard, MS – University of Colorado - Anschutz
Lauren McCall, MS – University of Colorado - Anschutz
Kristal Biesecker, BA, R. EEG T., CLTM – University of Colorado - Anschutz
Christopher Mizenko, MS – University of Colorado - Anschutz
Michelle Sandoval, BS – University of Colorado - Anschutz
Daniel Friedman, MD – New York University Grossman School of Medicine, NYU Langone Health
Jeschke Jay, MA – New York University - Langone
Leslee Willes, MS – Willes Consulting Group, Inc.
Meredith Decker, MS – Willes Consulting Group, Inc.
Rationale: Epitel has developed a rapid, wireless, wearable EEG monitoring system (REMI) that is US FDA-cleared for use in healthcare environments, including critical care scenarios such as EDs and ICUs. To support use of this system by non-EEG-trained clinicians such as ED/ICU healthcare providers, a novel algorithm, diverse across patients and seizure types, is currently in development as a clinical decision support system to detect and notify at the bedside when the acute seizure burden reaches a set threshold, set as 10% Prevalence (% of an EEG epoch in seizure state), based on the Frequent category as defined by the ACNS 2021 Critical Care EEG Guideline.
Methods: Preliminary data was collected from patients in U.S. Epilepsy Monitoring Units. patients wore REMI wireless sensors at F7/8 and Tp9/10 locations alongside standard-of-care 19+channel video-EEG. The EEG was preprocessed to account for noise, artifacts, and cross-patient differences. The data was then windowed into 2s segments, and features were extracted in time, frequency, and complexity domains. A subset of representative patients was held out for testing, and the segmented-data features from the remaining patients were used to train a machine learning classifier. The classifier output, the probability that each segment contains seizure activity, was then fed to a post-processor trained to determine the Prevalence over the past 5-min data epoch. Ground-truth electrographic seizures were determined by consensus review of the wired-EEG from 3 independent Epileptologists and only those lasting at least 30s in duration (10% Prevalence) were included in the analysis. Algorithm performance was evaluated on the held-out data by determining the Sensitivity and False Alarm Rate (FAR = False Positives per hour [FP/hr]).
Results: There were 44 patients in the held-out data (age range: 6-73, median: 22) with 22 noted as having no abnormal EEG characteristics (5 suspected of non-epileptic seizures). The other 22 patients experienced a total of 54 electrographic seizures (range: 1-6, median: 2). Recording durations ranged from 3.4 to 72 hrs with a median of 39.9hrs. The algorithm event-level Sensitivity was 89% across all known seizures with a lower 95% confidence interval bound (CI) of 78.9% and mean per-patient Sensitivity of 93%. The event-level FAR was 0.109 FP/hr across all detections with an upper 95% CI of 0.164 FP/hr and a mean per-patient FAR of 0.117 FP/hr. For patients with no known events, the event-level FAR was 0.134 FP/hr with an upper 95% CI of 0.235 FP/hr, while for those with events, the event-level FAR was 0.083 FP/hr with an upper 95% CI of 0.125 FP/hr. A total of 22 patients had no more than 1 FP (10 non-seizure patients), and 13 patients had no FP (6 non-seizure patients).
Conclusions: Traditional EEG is not suitable or accessible to the majority of critical care environments.
The miniature, wireless REMI system, supported by an automated algorithm with low FAR and high Sensitivity that can provide crucial bedside information of seizure burden to front-line care providers, will enable broader and more equitable EEG access at the point of care.
Funding: NIH 1U44 NS121562, NIH 5SB1 NS100235
Translational Research