Wearable seizure detection devices in refractory epilepsy

Area of Science:

• Neurology
• Biomedical Engineering
• Medical Devices

Background:

• Epilepsy impacts 50 million globally, affecting quality of life and treatment decisions.
• Seizure detection devices can significantly aid patient care, clinical trials, and potentially prevent SUDEP (Sudden Unexpected Death in Epilepsy).

Purpose of the Study:

• To provide an updated overview and comparison of wearable seizure detection devices (WSDDs).
• To evaluate WSDD performance based on newly proposed testing and validation standards, using the F1-score.

Main Methods:

• Reviewed and compared 16 wearable seizure detection devices (WSDDs).
• Categorized devices by sensor type (accelerometry, sEMG, EEG, multimodal) and development phase.
• Calculated and ranked devices by F1-score, a metric combining recall and precision.

Main Results:

• Significant inconsistencies were found in reported performance measures across devices.
• Devices in advanced development phases, particularly those using accelerometry (ACM) and surface electromyography (sEMG) sensors, demonstrated the highest F1-scores for detecting tonic-clonic seizures.
• The study identified 16 WSDDs, categorized by sensor technology.

Conclusions:

• Standardized testing and reporting are crucial for optimal comparison and advancement of WSDDs.
• WSDDs have the potential to enhance patient care, reduce underreporting of seizures, and improve outcomes.
• The role of neurologists and a B2B to C model are important considerations for WSDD implementation.