Optimizing Therapy in Epilepsy Using Seizure Forecasts Via EEG and Wearables

Summary

Mayo Clinic is conducting a research study to test a new approach to managing seizures in people with epilepsy whose seizures have not been fully controlled with regular medications. The study uses a small sensor placed just under the skin (not inside the brain) that continuously records electrical brain activity. This information is used to train a personalized computer algorithm that learns your individual seizure patterns and forecasts when a seizure is more likely to occur. If you are eligible and choose to participate, you would receive advance warning of elevated seizure risk, allowing you to take a fast-acting medication before a seizure happens. This medication would be in addition to the regular medicines prescribed by your doctor. The goal of the study is to understand whether this kind of forecasting system is safe, comfortable to live with, and capable of helping people gain more control over their seizures with a minimum amount of additional medication. Participation involves a minor surgical procedure to implant a small electrode under the skin above your ear that records your brain's electrical signals continuously over time. You would also wear a smart watch during the study. For the first six months, the system collects data to personalize your seizure forecast model and would not give you an alert. In the second six months, we would calculate your daily seizure risk and you would receive alerts on days when your risk is elevated. On these higher seizure risk days you would take a prescribed fast-acting nasal spray medication. The study is open to adults who continue to have 2 or more seizures per month in spite of having tried at least two antiseizure medications in the past, and whose seizures have shown involvement of the temporal lobe. All study-related procedures, devices, and study medication are provided at no cost to participants. Participants receive financial compensation for their time, and travel expenses to and from Mayo Clinic Rochester are reimbursed for anyone traveling more than 100 miles. If you are interested in learning more, please contact our research coordinator by email at [email protected].

Timeline

Start

2026-01-20

Primary completion

2028-12

Completion

2029-12

Publications

• Background Cloyd J, Haut S, Carrazana E, Rabinowicz AL. Overcoming the challenges of developing an intranasal diazepam rescue therapy for the treatment of seizure clusters. Epilepsia. 2021 Apr;62(4):846-856. doi: 10.1111/epi.16847. Epub 2021 Feb 22.
• Background Pal Attia T, Viana PF, Nasseri M, Duun-Henriksen J, Biondi A, Winston JS, P Martins I, Nurse ES, Dumpelmann M, Worrell GA, Schulze-Bonhage A, Freestone DR, Kjaer TW, Brinkmann BH, Richardson MP. Seizure forecasting using minimally invasive, ultra-long-term subcutaneous EEG: Generalizable cross-patient models. Epilepsia. 2023 Dec;64 Suppl 4(Suppl 4):S114-S123. doi: 10.1111/epi.17265. Epub 2022 May 4.
• Background Viana PF, Pal Attia T, Nasseri M, Duun-Henriksen J, Biondi A, Winston JS, Pavao Martins I, Nurse ES, Dumpelmann M, Schulze-Bonhage A, Freestone DR, Kjaer TW, Richardson MP, Brinkmann BH. Seizure forecasting using minimally invasive, ultra-long-term subcutaneous electroencephalography: Individualized intrapatient models. Epilepsia. 2023 Dec;64 Suppl 4(Suppl 4):S124-S133. doi: 10.1111/epi.17252. Epub 2022 Apr 16.
• Background Nasseri M, Stirling RE, Viana PF, Cui J, Nurse E, Karoly PJ, Kremen V, Dumpelmann M, Worrell GA, Freestone DR, Richardson MP, Brinkmann BH. Forecasting epileptic seizures with wearable devices: A hybrid short- and long-horizon pseudo-prospective approach. Epilepsia. 2025 Sep;66(9):3293-3308. doi: 10.1111/epi.18466. Epub 2025 May 24.

Drugs

Indications