Algorithm facilitates the identification of injuries and surgical planning of patients with cortical focal dysplasia, malformation associated with a form of refractory disease with medicines
An international team of researchers has developed an AI -based tool (artificial intelligence) that identifies cortical focal dysplasia, one of the main causes of refractory epilepsy to medicines and very difficult to diagnose. The details were released in Jama Neurology.
Focal cortical dysplasia is a cortex malformation characterized by lesions in some regions of the brain. Patients with the condition tend to have anticonvulsive treatment resistance and thus surgery (to remove injured areas) remains as the only alternative to reduce the symptoms of the disease. However, these brain injuries are often very subtle and difficult to detect, which makes the treatment of cortical focal dysplasia even more challenging.
From machine learning techniques and magnetic resonance data analysis of 1,185 participants – unconnecting 703 people with cortical focal dysplasia and 482 controls – the tool, called MELD GRAPH, was able to detect abnormalities from scanning in cerebral images.
The tool, whose algorithm is available publicly, detected 64% of brain abnormalities linked to epilepsy that radiologists had not detected. As a result, Meld Graph is expected to serve more than 4 million people around the world who live with this cause of epilepsy.
International collaboration
The work was conducted within the Enigma Consortium, an international network that brings together scientists from areas such as genomic image, neurology and psychiatry. The group seeks to understand brain structure and function based on high resolution magnetic resonance imaging, genetic data and other information from patients with epilepsy, parkinson, Alzheimer, autism, schizophrenia and other neurodegenerative diseases.
The development of the AI tool was carried out by researchers at King’s College London and University College London, both from the UK. High quality magnetic resonance imaging were provided by 23 epilepsy research centers around the world that are part of the consortium. The Brazilian Institute of Neuroscience and Neurotechnology was one of the institutions with high capacity capture and image analysis to provide data for the project.
“In this type of epilepsy, patients – who can be adults or children – have 30 or even 50 epileptic seizures per day. It is severe and usually the crises are not controlled by medicines. Therefore, surgery to remove brain injuries, although very difficult to do, is the most suitable for these cases. The new tool can greatly assist these patients, giving more speed to treatment and, above all, improvement of their lives and families.”tells FAPESP Clarissa Yasuda, Professor at the Department of Neurology, Faculty of Medical Sciences, State University of Campinas.
O artigo Detection of Epileptogenic Focal Cortical Dysplasia Using Graph Neural Networks A MELD Study pode ser lido .
With information from.
