Viewing time: 5 minutes | The caption is auto-generated and may contain errors.
“Our results support the application of machine learning (ML) in multiple sclerosis (MS), which is the first major step. [The findings] We demonstrated that ML can identify subtypes and stages of MS starting from MRI data. Also, [the study] Supporting the new hypothesis of MS as a continuum, it is more than a clear disease between recurrence and progressiveness. ”
Multiple sclerosis (MS) is a heterogeneous condition traditionally classified by clinical features, and is increasingly incorporating MRI metrics into classification schemes. Recent research published in
In this study, researchers trained sustain, an unsurveillanced ML algorithm, using MRI scans containing T1-weighted and T2-flair images, to identify disease subtypes and model temporal progression. Of the 250 patients with complete MRI data from the study cohort, the investigators identified two different MRI-driven subtypes. Findings showed that subtype 1 (n = 235) was lesion-driven, showing early increases in lesion volume and T1/flare abnormalities followed by atrophy of the cerebellum, corpus callosum, and spinal cord. For comparison, subtype 2 (n = 15) exhibited a burden of low lesions with early diffused cortical gray matter degeneration.
At Ectrims 2025, Bianchi, a postdoctoral researcher at the University of Siena in Italy, provided background for the interview study NeurologyLive®. In the conversation, she noted that early stage clustering of patients with clinically isolated syndrome and late mapping of patients with progressive MS may support a new view of MS as a disease continuum. Furthermore, she highlights that disease stages are correlated with both disease duration and expansion of disability status scale scores, highlighting the possibility of improving ML classification and staging.
