AI Tool Detects Brain Lesions, Offering Hope for Children with Epilepsy
A groundbreaking artificial intelligence (AI) tool has been developed to detect tiny brain lesions, known as cortical dysplasias, which are a common cause of epilepsy in children. This innovation promises faster diagnoses, more precise treatment, and potentially a cure for young patients suffering from this debilitating condition. Researchers at the Murdoch Children's Research Institute (MCRI) and The Royal Children's Hospital (RCH) in Melbourne, Australia, are behind this significant advancement.
The 'AI Epilepsy Detective'
The newly developed 'AI epilepsy detective' is capable of identifying brain lesions as small as a blueberry in up to 94 percent of cases, with the aid of medical imaging. These lesions, also called focal cortical dysplasias, are often missed by traditional MRI scans. The tool was trained using MRI and FDG-positron emission tomography (FDG-PET) scans.
Dr. Emma Macdonald-Laurs, a neurologist at The Royal Children's Hospital and the study lead at MCRI, emphasizes that early and accurate diagnosis of these lesions is crucial. This can lead to faster referrals for surgery, fewer seizures, and improved long-term developmental outcomes for affected children.
"Identifying the cause early lets us tailor treatment options and helps neurosurgeons plan and navigate surgery," says Dr. Macdonald-Laurs.
Improved Surgical Outcomes and Reduced Invasive Testing
The AI tool assists neurosurgeons in developing safer surgical roadmaps, helping them avoid critical blood vessels and brain regions responsible for speech, thinking, and movement. This minimizes the risk of removing healthy brain tissue and reduces the need for invasive testing in children.
The study, published in the journal Epilepsia, involved 71 children at The Royal Children's Hospital and 23 adults at the Austin Hospital, all diagnosed with cortical dysplasia and focal epilepsy. Before the AI tool was introduced, 80 percent of cases were missed by human examination of MRI results alone.
Impact on Patients and Families
Five-year-old Royal is one of the children who benefited from this technology. After scans aided by the AI detector revealed lesions on his brain, he underwent surgery and is now seizure-free. Royal's mother, Gurjinder, shared her experience, highlighting the increasing frequency and severity of Royal's seizures before diagnosis and treatment. She said Royal went from having multiple seizures a day to now being seizure free following the procedure.
"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," said Gurjinder. "We are so thankful that Royal is now back to his calm, friendly, patient self.”
The Future of Epilepsy Diagnosis and Treatment
The researchers are seeking additional funding to test the AI detector in pediatric hospitals across Australia, potentially leading to broader access to accurate diagnosis and a potential cure for children with drug-resistant epilepsy. Approximately one in 200 children suffers from epilepsy, and cortical dysplasias are a common cause of drug-resistant seizures.
The study emphasizes that failure to locate abnormal brain tissue can delay diagnosis and prevent children from being referred for potentially curative epilepsy surgery. The longer a child experiences uncontrolled seizures, the higher the risk of developing learning difficulties and intellectual disabilities.
| Key Statistic | Details |
|---|---|
| Prevalence of Epilepsy in Children | Approximately 1 in 200 |
| Missed Diagnoses (MRI Alone) | 80% prior to AI tool |
| AI Detection Success Rate | Up to 94% |
| Patients Seizure-Free Post-Surgery (Test Group) | 11 out of 12 children |
