Optimizing Pharmacotherapy with Noninvasive Wearable Sensors and Subscalp EEG

Project Summary The apparently random nature of seizures is one of the most significant factors affecting quality of life for patients with epilepsy. Accurate seizure forecasting could be transformative for patients with epilepsy, allowing patients to modify activities to avoid risk, take fast-acting medications to stop seizures before they develop, and provide a sense of empowerment over their disease. Successful seizure prediction has now been established using ambulatory intracranial EEG devices. Unfortunately, no such device is currently available to patients, as no commercial vendor has been successful to date at obtaining approval for clinical use of such a device. Furthermore, invasive intracranial implants may not be appropriate for or acceptable to many patients given the associated risk of infection and hemorrhage. Subscalp EEG recording has recently emerged as a viable means for long-term monitoring of patients with epilepsy. A device is commercially available in the EU, but does not yet have FDA clearance in the US. Recently published studies show the device to be reliable and robust, and recordings longer than 6 months have been reported. In addition cycles of seizure risk have been identified in wearable device physiological signals, and these long-term cycles may be capable of contributing to the accuracy of seizure forecasts. This project will develop the ability to prospectively forecast seizures from simultaneous subscalp EEG and wrist-worn wearable signals, and will assess the safety and feasibility of administering a fast-acting supplemental medication with seizure forecasts to prevent seizures.