Longitudinal analysis of cerebellum volume changes in a multiple sclerosis cohort

Multiple sclerosis (MS) is a degenerative disease of the central nervous system that affects millions worldwide. The cerebellum, crucial for motion coordination, is affected by this disease. In this study, we employed a modified deep-learning-based cerebellum parcellation framework, ACAPULCO, to investigate the longitudinal patterns of cerebellar volume in cohorts with relapsing-remitting MS (RRMS) and secondary progressive MS (SPMS). We identified and corrected ACAPULCO errors in our cohort with a post-processing step. Statistical analysis was performed with a linear mixed-effect model on the volumetrics of the parcellated regions. Our findings indicate that Lobules IV, V, and VI are the most affected by disease progression, agreeing with previous clinical research.