Specialists from the United States found out that long-term monitoring with wrist sensors allows you to select patterns in patients with epilepsy, according to which the attack can be predicted 30 minutes before it occurs. The developers described their way of predicting epilepsy attacks in Scientific Reports magazine.
In the experiment, six patients with drug-resistant epilepsy, the activity of the brain of which was tracked using neurosulant implants took part in the Experiment from the Mao Clinic. They were distributed two wearable sensor and tablets to load the data from the first. Scientists instructed the tests to wear the sensors alternately to be able to collect information continuously during the day.
During the period from six months to a year, the electrical properties of the skin, body temperature, blood pressure, pulse and the movement of the participants in the experiment were monitored with the help of the sensors. This data was trained to neurallet, which found patterns, allowing to predict the epileptic attack 30 minutes before it occurs. Researchers checked prediction data using information obtained from implants.
Despite the fact that the ability to predict the epilepsy attacks using invasive brain implants has long been established, a new method will be useful to many patients who want to do without their installation, emphasize scientists.