AI tool has 'untapped potential' for diagnosing and managing multiple sclerosis
A new artificial intelligence tool could have “untapped potential” for gauging treatment effectiveness in patients with multiple sclerosis.
New research published in Nature Communications details the use of a tool developed by experts at University College London. MindGlide is a deep learning model that quickly analyzes brain MRIs to assess for longitudinal changes that can give providers insight into whether a patient’s MS is responding to treatment. It can extract brain region and white matter lesion volumes from any MRI of the brain, even less specialized sequences that aren’t traditionally used in the diagnosis and management of MS.
Experts are hopeful that their work could lead to better outcomes for patients living with MS, as the disease can be debilitating for many once it has progressed.
“Using MindGlide will enable us to use existing brain images in hospital archives to better understand multiple sclerosis and how treatment affects the brain,” Philipp Goebl, a PhD student at the UCL Queen Square Institute of Neurology and UCL Hawkes Institute, and colleagues suggest.
MindGlide was trained using 4,247 brain MRIs from nearly 3,000 MS patients across 592 scanners. It was externally validated on a dataset of 14,952 scans from 1,001 MS patients participating in other clinical trials studying the condition. Its use was compared alongside two other AI tools used in MS— SAMSEG (identifies and outlines different parts of the brain on imaging) and WMH-SynthSeg (detects and measures bright spots on MR exams).
MindGlide significantly outperformed both AI tools in detecting subtle changes related to plaques on imaging. It performed well across numerous sequences and images of poor quality. What’s more, the tool was able to review individual images in as little as five seconds. When done manually by human providers, this process can take hours.
“We hope that the tool will unlock valuable information from millions of untapped brain images that were previously difficult or impossible to understand, immediately leading to valuable insights into multiple sclerosis for researchers and, in the near future, to better understand a patient’s condition through AI in the clinic,” the group writes.
The team hopes to be able to use their tool in clinical practice in the next 5 to 10 years.
The study can be viewed here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.