Meeting Banner
Abstract #3954

A modern machine-learning approach to reveal the role of multiple sclerosis lesions in inducing cortical tissue loss

Constantina Treaba1, Allegra Conti2, Ambica Mehndiratta1, Valeria Barletta1,3, Caterina Mainero1,3, and Nicola Toschi1,2
1Radiology, Massachusetts General Hospital; A. Martinos Center for Biomedical Imaging, Boston, MA, United States, 2Department of Biomedicine and Prevention, University of Rome “Tor Vergata", Rome, Italy, 3Harvard Medical School, Boston, MA, United States


In multiple sclerosis (MS), cortical atrophy could result from extensive axonal transection from distant white matter lesion and/or local cortical demyelination. The destructive potential of MS white matter lesions, however, seem to differ across lesion types, being higher for chronic active lesions detectable on susceptibility weighted images by their characteristic paramagnetic rim. Using 7T MRI that has shown increased sensitivity to cortical lesions and modern machine learning algorithms, we demonstrate that both white matter and cortical lesions are main determinants of cortical thinning. Despite their destructive capacity, chronic active lesions are not major contributors of cortical tissue loss.

This abstract and the presentation materials are available to members only; a login is required.

Join Here