An artificial intelligence (AI) algorithm that can detect subtle brain abnormalities that cause epileptic seizures has been developed by a UCL-led team of international researchers, including Children’s National Hospital.
To do this, the team quantified features from MRI scans, such as how thick or folded the brain was at nearly 300,000 locations in each case.
They then trained the AI algorithm using examples labelled by expert radiologists as either a healthy brain or one with focal cortical dysplasia (FCD) based on their patterns and features.
The results, published in Brain, showed that in the main cohort of 538 patients, the algorithm was able to detect the FCD in 67% of cases.
“We put an emphasis on creating an AI algorithm that was interpretable and could help doctors make decisions. Showing doctors how the Multicentre Epilepsy Lesion Detection project (MELD) algorithm made its predictions was an essential part of that process,” said Mathilde Ripart, research assistant at UCL and the study’s co-first author.
Around 1% of the population have epilepsy and, of these, 20-30% do not respond to medications.
“We are excited to collaborate with MELD on ways to improve the treatment of pharmacoresistant epilepsy,” said Nathan Cohen, M.D., neurologist at Children’s National Hospital and co-author of the study. “This advanced imaging platform is open source and demonstrates the benefit of team science at the broadest scale.”
In children who have had surgery to control their epilepsy, FCD is the most common cause, and in adults it is the third most common cause.
Additionally, of patients who have epilepsy that have an abnormality in the brain that cannot be found on MRI scans, FCD is the most common cause.
You can read the full UCL press release here.