AI algorithm can detect brain defects and help treat epilepsy disorders

The AI algorithm involved information from 300,000 regions of the patient's brain, and the results of the study, published in the journal Brain, showed that the AI algorithm identified 538 cases of FCD symptoms in the scans, including 112 cases that were not detected by radiologists.

This is very important, according to the researchers! Because currently focal cortical dysplasia (FCD) symptoms rely heavily on timely detection and then the option of surgical treatment. Dr Konrad Wagstyl, a researcher at the Institute of Neurology at Queen Square, University College London, said: "This AI algorithm could help identify more signs of hidden lesions in children and adults with epilepsy, prompting more people with epilepsy to consider brain surgery as soon as possible to cure the condition and raise awareness of the condition. Statistics show that approximately 440 paediatric patients undergo epilepsy surgery each year and improve after surgery."

The AI algorithm is suitable for patients with any underlying FCD symptoms who are aged three years or older and have undergone an MRI scan. Study co-author Dr Hannah Spitzer, a machine learning researcher at the Helmholtz Institute in Munich, Germany, said: "Our algorithm has the ability to automatically learn to detect lesions from MRI scans of thousands of patients, and it can reliably detect lesion features of different types, shapes and sizes. It even includes many lesions that were previously overlooked by radiologists."

Dr Sophie Adler of the Great Ormond Street Institute of Child Health at University College London said, "We hope that this technology will help to detect abnormalities that are easily overlooked and lead to epilepsy, ultimately allowing more people with epilepsy to be cured by brain surgery."