Related Experiment Video
Updated: Sep 13, 2025
09:41
Comprehensive Autopsy Program for Individuals with Multiple Sclerosis
Published on: July 19, 2019
11.6K
Cognitive profiles associated with faster thalamic atrophy in multiple sclerosis.
Moein Amin1, Keeley Scullin2, Kunio Nakamura3
1Mellen Center for Multiple Sclerosis, Cleveland Clinic, Cleveland, United States.
Multiple Sclerosis and Related Disorders
|August 3, 2025
Summary
Machine learning identified a cognitive impairment profile in multiple sclerosis linked to thalamic atrophy. This study aids understanding of neurodegeneration and cognitive decline in MS patients.
Area of Science:
- Neuroscience
- Medical Imaging
- Machine Learning
Background:
- Cognitive impairment (CI) in people with MS (pwMS) has complex pathophysiology.
- Neuropsychological testing (NPT) aids assessment but interpretation can be challenging.
- Thalamic atrophy (TA) is correlated with neurodegeneration and CI in pwMS.
Purpose of the Study:
- To leverage machine learning methods to link cognitive impairment (CI) with longitudinal neuroimaging biomarkers in pwMS.
- To explore the relationship between specific cognitive domains and neuroimaging findings.
Main Methods:
- Retrospective review of adult pwMS with NPT and at least two brain MRIs.
- K-means clustering based on NPT principal components (PC1 and PC2).
- Comparison of MRI change rates between identified clusters.
Main Results:
- 112 participants (80% relapsing-remitting MS) were analyzed.
- Two clusters emerged based on PC1, with one showing significantly more thalamic atrophy (TA) (p=0.035, p=0.002).
- Processing speed and memory were major contributors to PC1, indicating their importance in this cognitive profile.
Conclusions:
- A clustering approach successfully identified an NPT profile strongly associated with TA.
- These findings validate previous research and offer novel insights into NPT dimensionality reduction.
- The study suggests TA may drive specific patterns of cognitive impairment in MS.
