Related Experiment Video
Updated: Jan 5, 2026

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Patterns of regional brain volume loss in multiple sclerosis: a cluster analysis
Juichi Fujimori1, Kazuo Fujihara2,3, Ryo Ogawa4
1Department of Neurology, Tohoku Medical and Pharmaceutical University, 1-15-1 Fukumuro, Miyagino-ku, Sendai, 983-8536, Japan. j-fujimori@tohoku-mpu.ac.jp.
Objectives:
Although whole and individual regional brain volume loss have been separately reported to correlate with disability in multiple sclerosis (MS), hierarchical cluster analyses of the whole and regional brain to find their pattern in MS are few.
Methods:
We cross-sectionally conducted high-resolution, T1-weighted volumetric MRI examinations in 75 MS patients and 21 healthy controls (HCs) to measure the volumes of whole brain and a total of 56 brain regions of interest. Using a hierarchical cluster analysis with multivariate imaging data, we classified the patients into clusters according to their brain-volume patterns. Principal component analysis was also applied. Clinical features and brain volumes were then compared among the MS clusters.
Results:
The MS patients were categorized into three major clusters (Clusters 1, 2, and 3) with increasing disability in that order. Principal component analysis also identified Clusters 1, 2 and 3. Whole brain volume and supratentorial regional brain volumes, including thalamus and corpus callosum, decreased severely in Cluster 3 and moderately in Cluster 2, while equally preserved in Cluster 1 and the HCs. Only the volumes of the ventral diencephalon and T1 white matter hypointensities significantly differed in Clusters 1, 2 and 3 and HCs. In contrast, the volumes of the cerebellar cortex and brainstem were significantly different between Clusters 3 and 1, whereas there were no significant differences between Clusters 1 and 2 and Clusters 2 and 3.
Conclusion:
We identified brain regions that exhibit different degree of atrophy in a background of global brain atrophy in MS.

