Newswise — ROCHESTER, Minnesota - Despite medication, surgery, and neurostimulation devices, many people with epilepsy still experience seizures. The unpredictable nature of epileptic seizures severely restricts us. Only if seizures can be predicted reliably, can people with epilepsy change their activities, take a fast-acting medication, or turn on their neurostimulator to prevent a seizure or reduce its effects.
A new study, published in the Journal of Medical Reports by Mayo Clinic researchers and their international partners, found recognizable patterns in patients who wore a special watch-shaped monitor for six to 12 months, providing approximately 30 minutes of warning before an attack occurred. This worked well most of the time for five of the six patients studied.
“Just as reliable weather forecasting helps people plan their activities, seizure forecasting can help epilepsy patients adjust their plans if they know a seizure is imminent,” says Benjamin Brinkman, Ph.D. , an epilepsy scientist at the Mayo Clinic and lead author of the study. The study using a wrist-worn device showed that providing reliable seizure predictions for patients with epilepsy is feasible without directly measuring brain activity.”
In the aforementioned study, drug-resistant epilepsy patients with an implanted neurostimulator that monitors electrical brain activity were given two wrist recorders and a tablet to upload data daily to cloud storage. The patients were then asked to wear one wristband while the other was being charged. They switch devices at a specific time every day. They were using the devices while participating in their normal activities, which provided unique long-term data for the study.
The information collected from the wearable included: the electrical properties of the skin, body temperature, blood flow, heart rate, and accelerometer data that tracks movement. The data was analyzed by artificial intelligence using a deep learning neural network method, using an algorithm for time series and frequency analysis. Because the research participants already had an implanted deep brain stimulation device for epilepsy treatment, those neurostimulators were used to check for seizures, allowing the team to measure the prediction accuracy of the wrist devices.
And while the ability to predict seizures has previously been demonstrated using brain implants, many patients do not want an invasive transplant, Dr. Brinkman said.
“We hope this research using wearable devices will pave the way towards integrating seizure prediction into clinical practice in the future,” says Dr. Brinkmann, noting that it was a preliminary study and that additional patients are recording data to expand this test.
Other authors are Mona Nasseri, Ph.D., Mayo Clinic and University of North Florida; Tal Pal Attia, from Mayo Clinic; Bonnie Joseph, MBBS, Mayo Clinic; Nicholas Gregg, M.D., Mayo Clinic; Ewan Nurse, Ph.D., Sir Medical; Pedro Viana, King's College London; Gregory Worrell, MD, Ph.D., Mayo Clinic; Matthias Dompelmann, Ph.D., University of Freiberg; Mark Richardson, Ph.D., King's College; and Dean Friston, Ph.D., from Sir Medical.
This study is part of the Epilepsy Foundation of the American Epilepsy Innovation Institute and the My Epilepsy Measurement Project, an international collaboration aimed at using wearable devices to detect seizures and predict epilepsy. Additional support was provided through the Mayo Clinic Neuroartificial Intelligence Program.
Dr. Brinkman received non-financial research support from Medtronic and licensed the intellectual property rights to Cadence Neuroscience Limited.
###
About the Mayo Clinic
Mayo Clinic is a nonprofit organization committed to innovation in clinical practice, education, and research, providing empathy, expertise, and answers to everyone who needs hospitalization. For more Mayo Clinic news, visit the Mayo Clinic News Network .
RSS