Innovative work on artificial intelligence published in top-tier scientific journal
Christos Papadelis, Ph.D., Assistant Vice President of Neurosciences Research, and his team recently published an article in the top-tier scientific journal, Nature Portfolio Digital Medicine, titledm “Machine learning on interictal intracranial EEG predicts surgical outcome in drug resistant epilepsy.”
Hmayag Partamian, Ph.D., a post-doc research fellow, served as the first author of this work.
In this article, the Cook Children’s Neurosciences Research team proposes the development of an artificial intelligence framework that can identify – in a fully automated fashion – the brain area that generates seizures in children who have drug-resistant epilepsy. These children are unable to control their seizures with drugs and require brain surgery.
The automated nature of this innovative tool eliminates the post-acquisition preprocessing cost and time, reducing risks associated with the presurgical evaluation, such as brain infection and neurological damage from the surgery. Such a tool would be particularly useful to epilepsy centers that lack the multidisciplinary expertise to delineate accurately and precisely the brain area where seizures occur in complex cases of children with drug-resistant epilepsy. If validated prospectively, our tool would potentially replace intracranial recordings and benefit children with drug-resistant epilepsy.
The study is funded by Dr. Papadelis' R01 grant from the National Institute of Neurological Disorders and Stroke and is in collaboration with Boston Children's Hospital and Harvard Medical School.
