OneStudying patterns of brain activitycan be used to predict epileptic seizures in patients, according to researchers. This emerges from a new analysis of data obtained from clinically approved brain implants by neuroscientists at the University of San Francisco, Bern and Geneva.
Can epileptic seizures be predicted?
For forty years, efforts to predict epilepsy seizures have focused on the development of early warning systems. In this way, patients should be able to be warned about a seizure at best within a few seconds or minutes. So this is the first time someone has been able to reliably predict multiple seizures days in advance. The new method could enable affected people and their relatives to plan their lives better. This is especially true if they are at high risk of seizures, according to the study authors. Epilepsy is a chronic disease characterized by recurrent seizures. These are brief storms of electrical activity in the brain that can cause convulsions, hallucinations, or loss of consciousness.
For decades, researchers have been working to identify patterns of electrical activity in the brain that indicate an impending seizure. However, success has so far been limited. In part, the study authors say, this is because technology has limited the field to recording brain activity for days at most and in artificial stationary environments. So the researchers wanted to test whether they could use these regular patterns to make clinically reliable predictions of risk. They developed statistical models that match patterns of recorded brain activity with subsequent seizures in 18 epilepsy patients. They observed implanted neurological devices. The team then tested these prediction algorithms using data from 157 participants who took part in the multicenter long-term treatment study between 2004 and 2018.
Study results
Looking back at the study data, the researchers were able to identify periods in which patients were almost ten times more likely to have a seizure than at the start of the study. In some patients, signs of these periods of increased risk could even be detected several days in advance. Of course, such an increased risk of seizures does not necessarily mean that epileptic seizures will occur. Epileptologists still do not fully understand what causes a seizure at any given time, although many patients report triggers such as stress, alcohol, missing medication doses or lack of sleep.This studycompares the system with the predictive models of weather forecasts, which people often use to decide what clothes to wear and whether to take an umbrella with them when they go out. Therefore, it is possible that specially designed devices can detect predictive fluctuations in brain activity in a wider range of patients. Or it could be that, as in many ways, epilepsy patients simply vary in the predictability of their risk cycles.
