Revolutionary AI Technology Identifies Small Brain Lesions in Children with Epilepsy

A groundbreaking artificial intelligence (AI) tool has been developed to detect minute brain lesions responsible for severe epilepsy in children. This innovative technology enhances the speed and accuracy of diagnosis, enabling timely and targeted treatment options that could potentially lead to a cure.

Created by researchers at Murdoch Children's Research Institute (MCRI) and The Royal Children's Hospital (RCH), the "AI epilepsy detective" is capable of identifying focal cortical dysplasias—the abnormal brain tissue associated with epilepsy—that are as small as a blueberry. Remarkably, this system has demonstrated success in detecting lesions in up to 94% of cases when supported by medical imaging.

Dr. Emma Macdonald-Laurs, a neurologist at RCH leading the project, explained that accurate early diagnosis of cortical dysplasia is crucial for improving treatment outcomes. Faster identification allows for quicker referrals for epilepsy surgery, which can significantly reduce seizure frequency and improve long-term developmental prospects for affected children.

The approach leverages advanced imaging techniques, combining MRI scans with FDG-PET scans, and was trained on data from both children and adults. During testing, the AI tool outperformed traditional methods, with a success rate of 94% in identifying lesions in a cohort of Pediatric patients. Among these children, most who underwent surgical removal of the lesion experienced seizure freedom.

Epilepsy affects approximately 1 in 200 children, often presenting as seizures that begin unexpectedly during early childhood. When caused by cortical dysplasia, these seizures can become resistant to medication, making surgical intervention a vital option. However, conventional MRI techniques frequently miss these lesions, especially the smaller ones, delaying diagnosis and treatment.

This AI system addresses this challenge, enabling clinicians to visualize otherwise undetectable lesions, thus streamlining the pathway to surgery and reducing reliance on invasive testing. The enhanced imaging capability can also help neurosurgeons plan safer procedures by accurately mapping brain structures and avoiding vital regions involved in speech, movement, and cognition.

Published in the journal Epilepsia, the study included 71 children and 23 adults with cortical dysplasia. The findings highlight the importance of integrating AI into diagnostic protocols and open doors for widespread application across pediatric hospitals in Australia. With additional funding, researchers hope to expand testing and validation, further improving epilepsy management and outcomes for children.

This advancement marks a significant step forward in the fight against drug-resistant epilepsy in young patients, offering hope for more precise diagnosis, effective treatment, and potential cures.