Seizure detection using wearable electrocardiogram connected to a smartphone: a phase 3 clinical validation study

Area of Science:

• Neurology
• Biomedical Engineering
• Medical Devices

Background:

• Automated seizure detection is crucial for patient safety and objective quantification.
• Wearable seizure detection devices offer significant potential for enhancing patient care.
• Assessing the accuracy of a smartphone-connected ECG device for seizure detection and consciousness impairment is vital.

Purpose of the Study:

• To evaluate the accuracy of a wearable ECG device in detecting epileptic seizures.
• To assess the device's capability in automatically determining impairment of consciousness.
• To validate a real-time seizure detection algorithm in a clinical setting.

Main Methods:

• A prospective, blinded, multicentre clinical validation study was conducted.
• Patients with seizures exhibiting autonomic ictal manifestations wore a smartphone-connected ECG device.
• An algorithm based on heart rate variability with personalized thresholds was used for detection, triggering behavioral testing for consciousness assessment.

Main Results:

• The device achieved 90.5% overall sensitivity for 42 seizures across 17 patients.
• Bilateral tonic-clonic seizures were fully detected; focal seizures had 82.6% sensitivity.
• The mean false alarm rate was 2.5/day, with behavioral testing successfully canceling false alarms and accurately assessing consciousness.

Conclusions:

• The wearable ECG device accurately detects focal and generalized seizures.
• The device effectively assesses impairment of consciousness.
• This technology holds promise for improving epilepsy management and patient safety.