A new technique manages to anticipate seizures in those affected by epilepsy

By analyzing data such as heart rate, body temperature and movement, researchers found that they would have been able to predict most seizures about 30 minutes before they occurred.

These findings show that it is possible to provide reliable seizure predictions without directly measuring brain activity, the university stressed.

"I've seen these patients and I know they need something like this. When they have a lot of seizures that are resistant to medications, They have to avoid doing so many activities. We hope to help them," Nasseri said.

The study is part of the Epilepsy Foundation of the American Institute for Epilepsy Innovation and the My Seizure Gauge project, which includes international collaboration.

This is the first study that followed people during their daily activities for six to twelve months, rather than previous work that relied on recording patient data in the hospital, according to Nasseri.

They tracked down six ppeople with drug-resistant epilepsy and that they had implanted a neurostimulation device that monitors the electrical activity of the brain.

Because of the device in the brain, researchers were able to receive data indicating exactly when the seizure occurred, rather than having to rely on participants to record the time in their personal diaries, which is less reliable.

The idea is to provide a warning when a seizure is imminent.

Nasseri is contributing to the study by implementing signal processing and machine learning techniques to develop these seizure detection and prediction algorithms.

"We collect data from wrist devices and we designed a machine learning algorithm", Dijo.

The project is based in the Mayo Clinic in Rochester, Minnesota, where Nasseri worked with Dr. Benjamin Brinkmann, the study's principal investigator, before joining the UNF faculty in 2020.