Accurate detection of typical absence seizures in adults and children using a two‐channel electroencephalographic wearable behind the ears

Abstract

Summary Objective Patients with absence epilepsy sensitivity <10% of their absences. The clinical gold standard to assess absence epilepsy is a 24‐h electroencephalographic (EEG) recording, which is expensive, obtrusive, and time‐consuming to review. We aimed to (1) investigate the performance of an unobtrusive, two‐channel behind‐the‐ear EEG‐based wearable, the Sensor Dot (SD), to detect typical absences in adults and children; and (2) develop a sensitive patient‐specific absence seizure detection algorithm to reduce the review time of the recordings. Methods We recruited 12 patients (median age = 21 years, range = 8–50; seven female) who were admitted to the epilepsy monitoring units of University Hospitals Leuven for a 24‐h 25‐channel video‐EEG recording to assess their refractory typical absences. Four additional behind‐the‐ear electrodes were attached for concomitant recording with the SD. Typical absences were defined as 3‐Hz spike‐and‐wave discharges on EEG, lasting 3 s or longer. Seizures on SD were blindly annotated on the full recording and on the algorithm‐labeled file and consequently compared to 25‐channel EEG annotations. Patients or caregivers were asked to keep a seizure diary. Performance of the SD and seizure diary were measured using the F1 score. Results We concomitantly recorded 284 absences on video‐EEG and SD. Our absence detection algorithm had a sensitivity of .983 and false positives per hour rate of .9138. Blind reading of full SD data resulted in sensitivity of .81, precision of .89, and F1 score of .73, whereas review of the algorithm‐labeled files resulted in scores of .83, .89, and .87, respectively. Patient self‐reporting gave sensitivity of .08, precision of 1.00, and F1 score of .15. Significance Using the wearable SD, epileptologists were able to reliably detect typical absence seizures. Our automated absence detection algorithm reduced the review time of a 24‐h recording from 1‐2 h to around 5–10 min.

Noninvasive mobile EEG as a tool for seizure monitoring and management: A systematic review

“…The authors developed an automatic absence seizure‐detection algorithm based on patient‐specific modeling and achieved a sensitivity of 98.4% with 0.23 false detections per hour and a positive predictive value of 87.1%. Similarly, Swinner et al 25 applied a patient‐specific absence‐seizure algorithm in the Byteflies Sensor Dot data and achieved a sensitivity of 0.98 and 0.91 false detections per hour. Frankel et al 41 studied the accuracy of focal‐seizure detection by epileptologists, with and without the support of an automated data‐annotation algorithm applied to EEG data collected from an array of four Epilog EEG sensors attached to the scalp.…”

“…EEG data from multiple-channel EEG showed promising results for visual detection of abnormal epileptiform events 19,30 and for different clinical purposes in different clinical settings 33,34 or rural areas, 21,22 whereas low-channel solutions provided promising results regarding the possibility of visually detecting abnormal EEG patterns 23,24,25,41,65 in the EMU and, when paired with seizure-detection algorithms, 25,27,29,40,52,65,66 some of these devices detected seizures with an overall acceptable sensitivity and specificity, 27,41 especially absences. 25 Overall, evidence showed that mobile EEG is well accepted and tolerated by patients 20,24,40,[43][44][45][46][47]49,58,67 and that experts and nonexperts found these solutions easy to apply. 22,33,38 Multiple-channel solutions, which are easy to apply but with electrodes that are not fixed, such as Emotiv Epoc+ 20 and ANT neuro, 48 may not be optimal for diagnostic purposes but are useful in situations where it is important to apply the EEG easily.…”

Seizure control during the COVID-19 pandemic: Correlating Responsive Neurostimulation System data with patient reports

“…This is particularly of value during the COVID-19 pandemic, as elective admissions to characterize neurobehavioral spells are not always possible. The future of seizure-control may be devices that can record seizures and brain activity such as RNS, long-term subscalp recordings (Duun-Henriksen et al, 2020), or surface-based electroencephalography recordings ( Kappel et al, 2019 , Swinnen et al, 2021 ).…”

Design of hydrogel-based wearable EEG electrodes for medical applications

“…25 results, which gave a sensitivity of 80%, precision of 100%. 192 Japaridze et al used the wearable BrainLink EEG to measure the accuracy of fully automated absence seizure detection. (Fig.…”