Embrace

The Embrace device is a wrist-worn biosensor that serves as an adjunctive monitoring system for generalized tonic-clonic seizures in adults and children (age 6+) with epilepsy, by analyzing electrodermal activity and motion data. Its core methodology relies on an onboard algorithm, validated against video-EEG in clinical settings, to detect seizure patterns during rest and automatically dispatch alerts to caregivers via a paired wireless device, while also logging data for subsequent professional review.

The Embrace is a prescription-only, wrist-worn biosensor designed as an adjunctive tool for monitoring seizures in adults and children aged 6 and older who have epilepsy or are at risk of developing it. It is intended for use in home or healthcare settings during periods of rest. The device functions by continuously capturing Electrodermal Activity (EDA) via an electrodermal sensor and motion data using an accelerometer, along with temperature readings. This multi-sensor data is processed by an on-board algorithm specifically developed to identify patterns indicative of generalized tonic-clonic seizures (GTCS). The algorithm's efficacy was validated against gold-standard video-Electroencephalogram (EEG) methodology, using data from epilepsy patients experiencing GTCS in hospital Epilepsy Monitoring Units. Upon detecting a likely GTCS event, the Embrace device transmits a signal to a paired wireless device, typically a smartphone application, which then alerts designated caregivers through calls or text messages. Beyond real-time alerting, the system records and stores comprehensive sensor data (accelerometer, EDA, temperature) for later review by healthcare professionals. Patients also have the capability to manually log seizure events. Clinical studies involving 141 patients (80 pediatric, 61 adult) demonstrated an overall Positive Percent Agreement (PPA) of 0.9815 for GTCS detection (53 out of 54 seizures) and an overall False Alarm Rate (FAR) of 0.94 false alarms per 24 hours, with some variation in FAR observed across different age cohorts.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the Embrace device differentiate between seizure-related movements and electrodermal activity changes from non-seizure activities (e.g., vigorous play in children or restless sleep) during its indicated 'periods of rest'?; What is the impact of inter-individual and intra-individual variability in baseline electrodermal activity (EDA) levels, influenced by factors such as hydration, ambient temperature, or medication, on the algorithm's sensitivity and specificity for GTCS detection?; What were the specific thresholds and feature-weighting within the on-board algorithm that determine a generalized tonic-clonic seizure detection, and how were these parameters optimized to balance Positive Percent Agreement (PPA) and False Alarm Rate (FAR) across different age groups (6-12, 13-21, >21 years)?; The clinical validation was performed in an Epilepsy Monitoring Unit (EMU); how does the device's performance, particularly the false alarm rate and detection accuracy, translate to real-world, long-term home environments where patient activities and environmental factors are less controlled?; Given the reported false alarm rates, especially the mean FAR of 1.63 for children aged 6-21, what strategies or guidance are provided to caregivers to mitigate alert fatigue and ensure appropriate response to true seizure events?