Mixed Results for Wearable Device’s Seizure Detection

TOPLINE:

A wrist-worn seizure detection device used in an epilepsy monitoring unit reliably detected all 16 focal to bilateral tonic-clonic seizures (FBTCSs) and generalized tonic-clonic seizures (GTCSs) that occurred in a new study, enabling timely intervention during off hours. However, the device missed all focal seizures that occcurred and triggered some false alarms.

METHODOLOGY:

A prospective observational study included 72 patients (mean age, 39 years; 57% women) admitted to a tertiary-care EMU in Canada between 2024 and 2025 to evaluate the real-world performance and clinical integration of Embrace2, a US FDA-cleared wrist-worn seizure detection device.
Participants wore the device in active mode on their nondominant hand during continuous EEG monitoring.
The device transmitted alerts upon seizure detection in real time via Bluetooth to a paired smartphone connected to the hospital Wi-Fi, which then notified designated caregivers and nursing staff through the Alert mobile application. Monitoring spanned 373 days.
Outcomes included seizure detection sensitivity, false alarm rate per 24 hours, and delays in nurse or technician intervention measured from alert generation to bedside response.

TAKEAWAY:

The device demonstrated high sensitivity and successfully detected all 16 FBTCSs and GTCSs that occurred while it was worn and functioning; but it did not detect any of the 510 focal seizures that occurred.
Real-time notifications or alerts were generated with a median detection latency of 25 seconds from the onset of bilateral tonic-clonic progression and reached caregivers in 15 of the 16 detected seizures; one alert failed due to a Wi-Fi disconnection.
Seven of the successful alerts occurred during overnight shifts when EEG-trained technicians were not on duty, with nurses responding in three of those within a median of 23 seconds of alert generation.
Twenty-nine false alarms occurred (0.08 per 24 hours), most commonly triggered by rhythmic or repetitive motor activities, such as brushing teeth.

IN PRACTICE:

“The Embrace2 wrist-worn device demonstrated high sensitivity in detecting FBTCSs, with acceptable latency and a low false alarm rate. Importantly, the device facilitated timely clinical intervention only during overnight and weekend shifts when specialized EEG personnel are typically unavailable,” the investigators wrote.

SOURCE:

The study was led by Amirhossein Jahani, PhD, Université de Montréal, Montreal, Quebec, Canada. It was published online on January 7 in Epilepsia.

LIMITATIONS:

The implementation of the device required extensive coordination between research and clinical staff. Technical issues included unreliable Wi-Fi connectivity and potential instances of staff not carrying designated phones for alerts. Despite their relative infrequency, false alarms may have influenced staff confidence in the system, potentially affecting long-term adoption.

DISCLOSURES:

The study was funded by the Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, Fonds de Recherche du Québec-Santé, and Canada Research Chairs. Some investigators reported having financial ties with various sources, which are fully listed in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.