Brain implant forecasts seizures days in advance
An international study is showing, for the first time, that it may be possible to predict the onset of epileptic seizures several days in advance. By analyzing data from a clinically approved brain implant designed to monitor and prevent seizures, the new research hopes to develop a model offering patients with epilepsy a seizure forecasting tool to predict the likelihood of upcoming episodes.
The research looked at data from a responsive brain stimulation implant called NeuroPace. The device was approved for clinical uses back in 2013 and it works to prevent seizures by delivering imperceptible pulses of electrical stimulation to certain parts of the brain upon detecting abnormal brain activity.
Scientists have been working on a variety of seizure prediction tools for decades. But despite some incredible advances, such as the NeuroPace device, no innovation to date has successfully shown it possible to predict seizures more than a few minutes in advance, at best.
The NeuroPace innovation offers researchers the first chance to study the relationship between seizures and brain activity using years of EEG data. The new study initially analyzed long-term data from 18 patients with the brain implant who were closely tracked for several years. From this data the researchers developed predictive algorithms to forecast seizures. These predictive algorithms were then tested on long-term data gathered from the more than 150 people who participated in the decade-long clinical trials testing the brain implant system.
Vikram Rao, co-senior author on the new study, says the data shows seizure risk could be effectively forecasted three days ahead in nearly 40 percent of subjects and one day ahead in 66 percent of subjects.
"For forty years, efforts to predict seizures have focused on developing early warning systems, which at best could give patients warnings just a few seconds or minutes in advance of a seizure,” says Rao. “This is the first time anyone has been able to forecast seizures reliably several days in advance, which could really allow people to start planning their lives around when they're at high or low risk.”
Rao does stress the current algorithm can only predict when one is at higher risk of seizure, and not specifically when a seizure will take place. A number of other unaccounted environmental triggers, from stress to erratic sleep, can play a role in the onset of a seizure. So the system currently developed is more like a weather forecast, offering probabilities designed to help guide a person’s future activities.
"I don't think I'm ever going to be able to tell a patient that she is going to have a seizure at precisely 3:17 pm tomorrow—that's like predicting when lightning will strike," explains Rao. "But our findings in this study give me hope that I may someday be able to tell her that, based on her brain activity, she has a 90 percent chance of a seizure tomorrow, so she should consider avoiding triggers like alcohol and refrain from high-risk activities like driving."
Much more work is needed before the system is ready for clinical use. This preliminary study uncovered a significant amount of variability from person to person. It is unclear why reliable forecasting could not be generated from some patient’s brain activity data. Future investigations to optimize the algorithm and perhaps incorporate multimodal physiological data may enhance the algorithm’s predictive capacity.
Plus, currently the system requires data gathered from a device requiring surgical implantation. This would limit the use of the device to only those with the most severe forms of epilepsy. More superficial subscalp EEG devices could offer a less invasive way of capturing this brain activity data over long periods of time.
"It is worth remembering that, currently, patients have absolutely no information about the future—which is like having no idea what the weather tomorrow might be—and we think our results could help significantly reduce that uncertainty for many people," adds Rao. "Truly determining the utility of these forecasts, and which patients will benefit most, will require a prospective trial, which is the next step."
The new study was published in the journal The Lancet Neurology.