Correlation between lesion volume and total brain, thalami and caudate volumes obtained by automatic volumetry in multiple sclerosis patients.

Motivation: Artificial intelligence (AI) is a promising tool that must be used to its full potential. Acquiring knowledge related to its uses and limitations is essential. Goal(s): Establish the utility of new anatomic biomarkers using an AI tool for brain and lesion volumetry in patients with multiple sclerosis. Approach: Analyze the correlation between lesion burden and specific structural brain atrophy using AI segmentation and volumetry in patients with multiple sclerosis. Results: Significant statistical correlation was found between lesion burden and thalamic atrophy in patients with MS, opening the possibility of using thalamic volumetry as a biomarker for disease progression. Impact: Total lesion burden in MS patients correlates with structural brain volume loss mediated by destruction of white matter tracts and Wallerian degeneration. Establishing volumetric biomarkers using artificial intelligence may provide useful information regarding disease progression and long-term prognosis.