An imaging AI 'epilepsy detective' could help cure treatment-resistant seizures
Experts have developed an artificial intelligence tool they say has the potential to cure treatment resistant epilepsy that causes recurring seizures.
Endearingly referred to as the AI "epilepsy detective,” the tool can identify small focal cortical dysplasias on imaging that are typically missed by radiologists, bottom-of-sulcus dysplasia in particular. These lesions are known to cause recurring seizures, but an accurate imaging assessment can assist surgeons in removing them, which often results in seizure cessation.
Emma MacDonald-Laurs, MD, a neurologist at the Royal Children’s Hospital Melbourne who helped develop the tool, believes that earlier, more precise detection of problematic focal lesions has the potential to drastically improve outcomes in children with epilepsy.
“Cortical dysplasias can be impossible for traditional MRI techniques to identify,” she said in a news announcement on her team’s findings. “Failure to locate the abnormal tissue slows the pathway to a definitive diagnosis and may stop a child being referred for potentially curative epilepsy surgery. The longer a child continues to have uncontrolled seizures, the more likely they are to develop learning difficulties, including intellectual disability.”
Researchers trained the tool using MRI and FDG-PET imaging from 54 patients who had been diagnosed with both focal epilepsy and bottom-of-sulcus dysplasia. The group extracted 12 imaging features to determine which best distinguished dysplastic from normal-appearing cortex, using those to refine the model’s detection capabilities.
Prior to the study, more than 80% of the cohort had their lesions previously missed by radiologists. When testing the tool, the team noted a 94% accuracy rate for identifying bottom-of-sulcus dysplasia. Of those patients, 12 underwent surgery; 11 of them are now seizure-free.
“Identifying the cause early lets us tailor treatment options and helps neurosurgeons plan and navigate surgery,” MacDonald-Laurs said. “With more accurate imaging, neurosurgeons can develop a safer surgical roadmap to avoid important blood vessels and brain regions that control speech, thinking and movement and removing healthy brain tissue. Children also avoid the need to have to undergo invasive testing.”
It is MacDonald-Laurs' hope that her team will be able to continue their research in pediatric hospitals across Australia.
Read more about the findings here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.