Disease has several causes and, in about 30% of cases, is related to structural anomalies of the brain
Australian researchers announced the development of a capable of identifying small brain bads difficult to detect in children with epilepsy. Technology can expand access to surgeries that change the lives of patients whose drug treatment is not effective.
Epilepsy has several causes and, in about 30% of cases, is related to structural anomalies of the brain. Often, these changes do not appear in conventional magnetic resonance imaging (sister), especially when they are small or hidden in brain folds.
The team led by pediatric neurologist Emma MacDonald Laurs, from the Melbourne Real Children’s Hospital, trained AI with brain images of children to locate the size of a blueberry or even smaller. “Often, they go unnoticed and many children are not considered candidates for surgery,” said Laurs, during a presentation of the study of the study in Epilepsia magazine.
According to the researchers, the tool does not replace radiologists or experts, but it works as a detective, helping to identify more quickly the cases where surgery can be indicated. In the tests, technology analyzed resonance and pet scan exams, reaching 94% precision in one group and 91% in another. Among 17 children evaluated, 12 underwent surgery and 11 of them were free from crises.
The next step will be to apply the system in hospital environments, with new patients not yet diagnosed. Epilepsy affects approximately one in 200 children, and one third of them does not respond to medicines. Innovation can accelerate diagnoses and offer alternatives to families awaiting more effective treatments.
*With information from AFP
