AI Method Reveals Hidden Multiple Sclerosis Lesions in Routine MRI Scans

A University at Buffalo-led team has developed an artificial intelligence method that detects previously invisible cortical lesions in standard MRI scans of multiple sclerosis patients. Published in Communications Medicine, the technique compares information across multiple MRI sequences to recover disease signals that conventional viewing misses. Cortical lesions in the brain's gray matter are closely linked to disability and cognitive decline in MS, but routine MRI has not reliably detected them, leaving clinicians with a significant blind spot. The AI approach could enable earlier detection of disease progression and better monitoring of treatments, many of which are designed primarily around reducing white matter lesions. The Buffalo Neuroimaging Analysis Center directed the study.