Abstracts

Rationale: Long-term electroencephalography (EEG) monitoring has traditionally been a hospital- or clinic based service. Patients with epilepsy generally come into a monitoring unit for up to a few weeks of seizure assessment. During this time, a seizure may or may not occur. Moreover, once a patient is sent home, the clinician can only rely on the patient’s self-report of their seizure events, which likely leads to suboptimal treatment. Presently, there is no practical method to determine mechanism and precursor of sudden unexpected death in epilepsy (SUDEP), which is the most common non-traumatic cause of death in epileptics. Thus, we have developed a new wireless EEG headset for continuous long-term EEG monitoring which can be used in the patient’s natural environment.Methods: A novel spring-suspended EEG headset was developed using a set of 10 active dry electrodes for EEG recording, a microcontroller for signal processing, and a Bluetooth module for wireless data transfer. Gold-plated electrodes were assembled in spring-attached modules and low-noise amplifiers were mounted on the back of each electrode. Geometrical shape, dimension, and material of the electrodes were analyzed to understand the effective area of the electrode-skin interface. For remote EEG monitoring, an EEG display monitoring platform was developed and various signal processing tools (e.g., fast Fourier transform, continuous wavelet transform and support vector machines) were integrated for optimizing real-time signal processing. Recordings were obtained using both this headset and the “gold-standard” wet electrode system simultaneously from four healthy subjects. Each recording was 2-3 hours long and subjects were asked to perform various tasks (e.g., relax, mental math, listening to music, etc.). The recordings were further analyzed in separate physiological EEG frequency bands and the power in each band was compared to that obtained from by traditional wet electrodes, to determine the quality of the recording.Results: The spring-suspended technology integrated with the active dry electrodes improved comfort compared to commercial headsets, which allows easy setup for recording for a long-term usage. Impedance of the dry electrode reduced over the time due to the sweat accumulation and impedance of the wet electrode increased due to the gel's evaporation property. This wireless EEG headset was equipped with a rechargeable lithium ion battery capable of uninterrupted 24 hours recording. The remote EEG display platform provided real-time EEG analysis for feature extraction and abnormal signal pattern detection.Conclusions: This SMART (Seizure Monitoring At the Right Time) Headset will provide an invaluable tool for patients, clinicians, and researchers. This long-term EEG monitoring system will allow patients to be monitored from the comfort of their home as well as provide clinicians with sufficient data to confidently determine seizure duration, frequency and SUDEP like event in patients. This headset fills a critical gap in the current care paradigm for patients with epilepsy.