An AI-powered tool can detect 64% of brain abnormalities linked to epilepsy that radiologists miss, according to a new study.
Scientists from King’s College London and University College London developed the MELD Graph, which helps detect brain abnormalities known as focal cortical dysplasia (FCD), which can cause epilepsy.
Researchers say the tool will speed up diagnosis times, get people surgical treatment quicker and reduce costs to the NHS by up to £55,000 per patient.
According to the researchers, FCDs are a common cause of epilepsy in which seizures are usually hard to control with medication, but surgery can be effective.
However, they explain that FCDs can be difficult to see with the human eye and up to half of these are missed by radiologists.
‘Making the NHS more efficient’
The study used MRI data from 1,185 patients from 23 epilepsy centres around the world who took part in the Multicentre Epilepsy Lesion Detection project (MELD). Of the participants, half were children, and 703 had FCD and 482 did not.
The researchers used the data to train the AI tool to detect subtle brain abnormalities that might not be seen by a doctor.
Study author and project lead, Dr Konrad Wagstyl (pictured with project leads Dr Sophie Adler and Dr Mathilde Ripart), from King’s College London, said radiologists are inundated with images for review. “Using an AI-powered tool like MELD Graph can support them with their decisions, making the NHS more efficient, speeding time to treatment for patients and relieving them of unnecessary and costly tests and procedures.”
Prof Helen Cross, co-author, president of the International League Against Epilepsy, and paediatric neurologist, said: “Many of the children I see have experienced years of seizures and investigations before we find a lesion.
“Initiatives such as MELD have the potential to rapidly identify abnormalities that can be removed and potentially cure epilepsy.”
Healthcare crisis
The tool is not yet clinically available, but the research team have released it as an open-source software and are running workshops to train clinicians in how to use it.
Health improvement and research manager at Epilepsy Action, Tom Shillito, welcomed the news but warned that there is still a need for more specialists. He said: “It’s really exciting to hear about the potential this new AI tool has in driving forward better detection and diagnosis for people with epilepsy.
“We’re hopeful that this will be a useful tool that can empower professionals to give people faster diagnosis and help manage their epilepsy.
“However, this does not solve the problem of the healthcare crisis people with epilepsy are facing in the UK. Using NHS workforce data across England, Epilepsy Action found that the amount of epilepsy specialist nurses is 78% lower than the NICE-recommended rate of nine per 500,000 people.”
There is more information on epilepsy causes and treatments on the Epilepsy Action website.
More articles
