AI-powered Detector Improves Epilepsy Diagnosis and Treatment in Children
A new artificial intelligence (AI) tool is showing promising results in detecting subtle brain abnormalities that cause drug-resistant epilepsy in children, perhaps leading to faster diagnoses and more effective surgical interventions.Developed by researchers from The Royal Children’s Hospital, the University of Melbourne, Florey Institute of Neuroscience and Mental Health, Harvard Medical School, and the Austin Hospital, the detector identifies cortical dysplasias – often developing in utero – which are a common cause of epilepsy unresponsive to medication.
Currently,cortical dysplasias are challenging to identify using standard MRI scans,with less than half being detected on a child’s initial scan. the AI tool focuses on identifying these lesions, which can be “impossible for traditional MRI techniques to identify,” according to Dr. emily Macdonald-Laurs. This delay in diagnosis can lead to prolonged seizures and increased risk of learning difficulties,including intellectual disability.
In a recent test group of 17 children, 12 underwent surgery guided by the AI-aided scans, and a remarkable 11 are now seizure-free. Approximately one in 200 children experiences epilepsy, and this new technology offers hope for those with drug-resistant forms linked to cortical dysplasias.
The impact of this technology is powerfully illustrated by the case of Royal, a 5-year-old boy who suffered increasingly frequent seizures, escalating to as many as 19 in a two-hour period. Despite antiseizure medication, his seizures persisted, and a standard MRI failed to pinpoint the cause. Following scans aided by the detector, a cortical dysplasia was located and successfully removed at the royal Children’s Hospital (RCH). His mother, Gurjinder, reports that Royal has gone from multiple seizures daily to being completely seizure-free, regaining his appetite and returning to his normal personality.
“Without the assistance of the detector, it would have taken so much longer to achieve a diagnosis and Royal’s health would have continued to deteriorate,” Gurjinder stated.
Dr. Macdonald-Laurs and her team are now seeking additional funding to expand testing of the detector to paediatric hospitals across Australia. The research, published in Epilepsia (Macdonald‐Laurs, E., et al. (2025). Automated detection of bottom‐of‐sulcus dysplasia on magnetic resonance imaging-positron emission tomography in patients with drug‐resistant focal epilepsy. epilepsia. doi.org/10.1111/epi.18628), represents a meaningful step forward in the diagnosis and treatment of epilepsy in children.
Source: https://www.mcri.edu.au/