Artificial intelligence to support people with epilepsy



The neuroscientists are collaborating with industry to devise seizure detection algorithms based on changes in specific physiological signals. 

Throughout the world, 50 million people are living with epilepsy—a neurological condition characterized by recurrent seizures brought about by excessive electrical discharges in the brain. Patients experience different types of seizures, from strange sensations to loss of consciousness and disabling convulsions. While 65% of sufferers are able to control their symptoms with medication, the remaining 35% continue to have seizures that can cause injuries, falls and even death. But according to neurologist and researcher Dang Khoa Nguyen, professor in the Department of Neurosciences at Université de Montréal and researcher at the CHUM Research Centre, artificial intelligence may change the lives of individuals with epilepsy by predicting the onset of their seizures and thus facilitating effective interventions.

Nguyen and his team worked to develop prediction algorithms by studying electrical signals before, during and after an epileptic seizure. They collected data by placing electrodes in the brains of patients with epilepsy and then combined the results with figures from American and Australian databases. Now that they have established the proof of concept, the researchers dream of building a system that will send a seizure alert to the sufferer and his or her caregiver on their smartphones and perhaps even automatically administer an additional dose of medication directly into the patient's brain. 

As they focus on developing their application, the neuroscientists are also collaborating with industry to devise seizure detection algorithms based on changes in specific physiological signals (heart and breathing rates, muscle activity and perspiration) that could be sensed by a wearable device, such as a piece of clothing or an accessory. An alert sent out by the wearable could help parents, caregivers, nurses and doctors act quickly to stop a seizure and minimize the risk of injury or death for epilepsy patients at home or in a hospital or long-term care facility.