Essential Insights
-
An AI tool named MELD Graph, developed by researchers from King’s College London and UCL, detects 64% of brain abnormalities associated with epilepsy that human radiologists often miss, significantly enhancing diagnostic accuracy.
-
The tool aims to expedite diagnosis and surgical treatment for approximately 30,000 patients in the UK and 4 million globally, ultimately reducing NHS costs by up to £55,000 per patient.
-
MELD Graph was trained on MRI data from 1,185 participants and has shown effectiveness in identifying subtle lesions, including cases where patients previously endured daily seizures without improvement from medications.
- Although not yet clinically available, MELD Graph is open-source, and the research team is conducting global workshops to train healthcare professionals, fostering international collaboration to improve epilepsy diagnosis and treatment.
AI Tool Aims to Revolutionize Diagnosis of Brain Abnormalities in Children with Epilepsy
Scientists at King’s College London and University College London (UCL) have made a significant advancement in epilepsy diagnosis. They developed MELD Graph, an AI-powered tool that detects brain abnormalities in children. Remarkably, this tool identifies 64% of abnormalities linked to epilepsy that human radiologists often miss.
Epilepsy affects 1 in 100 individuals in the UK. Furthermore, 1 in 5 of those with epilepsy have seizures caused by structural abnormalities in the brain. Focal cortical dysplasia (FCD) is one of the leading causes of these seizures, and the challenges in accurately diagnosing FCD can lead to delayed treatment.
The study, published today in JAMA Neurology, underscores the potential of MELD Graph to expedite diagnosis and treatment. Researchers emphasize that the AI tool could significantly speed up the identification process, allowing patients to receive surgical treatment more quickly. Additionally, the use of this technology could save the NHS up to £55,000 per patient.
Project lead Dr. Konrad Wagstyl explained the burden faced by radiologists. Currently, they review numerous scans, which can lead to oversight. MELD Graph can assist by highlighting subtle lesions, improving efficiency within the NHS.
In a striking example, co-author Dr. Luca Palma recounted the case of a 12-year-old boy who had suffered daily seizures. Traditional evaluations overlooked a subtle lesion, but MELD Graph identified it, potentially paving the way for effective surgical intervention.
While MELD Graph is not yet available for clinical use, the research team has released it as open-source software. They also conduct workshops to train clinicians around the globe, including locations like Great Ormond Street Hospital and the Cleveland Clinic.
Dr. Mathilde Ripart, the study’s first author, noted the positive feedback from international doctors using the tool to benefit their patients. Professor Helen Cross, an expert in childhood epilepsy, highlighted the ongoing struggle within the epilepsy community. Speeding up diagnosis and treatment through innovations like MELD Graph can make a real difference for many children who have endured prolonged seizures.
Co-lead Dr. Sophie Adler remarked on the importance of global collaboration. With efforts from 75 researchers and clinicians, they aspire to reach a goal: no missed epilepsy lesions worldwide.
Overall, this groundbreaking development promises to reshape the future of epilepsy care. The potential for quicker diagnosis and improved treatment options excites the medical community and offers hope to countless children and their families.
Expand Your Tech Knowledge
Explore the future of technology with our detailed insights on Artificial Intelligence.
Discover archived knowledge and digital history on the Internet Archive.
SciV1
