AI algorithm that detects brain abnormalities could help cure epilepsy

An artificial intelligence that can detect (AI) algorithms developed by an international team of researchers led by UCL have developed the subtle brain abnormalities that lead to seizures.

The Multicenter Epilepsy Lesion Detection Project (MELD) was developed using MRI scans of more than 1,000 patients from 22 global epilepsy centers Algorithm, which provides reports of abnormal cases of drug-resistant focal cortical dysplasia (FCD) – a major cause of epilepsy.

FCD is an abnormally developed area of ​​the brain that often results in drug-resistant epilepsy. This condition is usually treated surgically, but identifying lesions from MRI is an ongoing challenge for clinicians because MRI scans in FCD appear normal.

To develop the algorithm, the team quantified cortical features, such as thickness or folding of the cortex/brain surface, from MRI scans degree and uses approximately 300,000 locations in the brain.

The researchers then trained the algorithm on examples that radiologists labelled healthy brains or FCDs, depending on their patterns and feature.

Study results, published in Brain, found that overall, the algorithm was able to detect FCD in 67% of cases (538 participants) in the cohort.

Previously, 178 participants were considered MRI negative, meaning the radiologist couldn’t find abnormalities – but MELD The algorithm was able to identify that 63% of these cases were FCD.

This is especially important, as if a doctor could find an abnormality on a brain scan and then surgically remove it would cure it.

Co-first author Mathilde Ripart (Great Ormond Street Institute for Child Health, UCL) said, “We emphasize creating An explainable AI algorithm that can help doctors make decisions. Showing doctors how the MELD algorithm makes predictions is an important part of the process.”

, co-senior author Dr Konrad Wagstyl (UCL Queen Square Institute of Neurology) added: “This algorithm could help uncover more of these hidden lesions in children and adults with epilepsy, And get more people with epilepsy to consider surgery that can cure epilepsy and improve their brain. Cognitive development. Around 440 children in England could benefit from epilepsy surgery each year.”

About 1% of the world’s population suffers from the severe neurological disorder conditional epilepsy, characterized by frequent seizures.

In the UK, around 600,000 people are affected. Although drug therapy is available for most people with epilepsy, 20-30% do not respond to the drug.

FCD is the most common cause of epilepsy in children who have undergone surgical control, and in adults it is the third most common cause.

Additionally, FCD is most common in patients with epilepsy who have brain abnormalities that cannot be seen on MRI scans s reason.

Co-first author Dr Hannah Spitzer (Helmholtz Munich) said: “Our algorithm automatically learns from thousands of detected lesions in the MRI scans of 12 patients. It reliably detected lesions of different types, shapes, and sizes, and even detected many lesions previously missed by radiologists.”

This study on FCD detection used the largest MRI cohort of FCD to date, meaning it was able to detect all types of FCD.

The MELD FCD Classifier Tool can be run on any patient with suspected FCD who is over 3 years of age and has an MRI scan.

MRI scanners were used at 22 different hospitals worldwide participating in this study, which makes the algorithm more robust, but also May affect the sensitivity and specificity of the algorithm.