Related Experiment Video
Updated: May 12, 2026
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Differential diagnosis tool for parkinsonian syndrome using multiple structural brain measures
Miho Ota1, Yasuhiro Nakata, Kimiteru Ito
1Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4-1-1 Ogawa-Higashi, Kodaira, Tokyo 187-8502, Japan.
Computational and Mathematical Methods in Medicine
|April 11, 2013
Summary
Distinguishing early Parkinson's disease from multiple system atrophy (MSA) is challenging. Magnetic resonance imaging (MRI) combined with discriminant analysis accurately differentiates these parkinsonian syndromes based on brain structure.
Area of Science:
- Neuroimaging
- Neurology
- Biostatistics
Background:
- Clinical differentiation between Parkinson's disease (PD), Parkinson variant of multiple system atrophy (MSA-P), and cerebellar subtype of multiple system atrophy (MSA-C) is difficult in early stages.
- Parkinsonian syndromes share overlapping early symptoms, complicating accurate diagnosis.
Purpose of the Study:
- To identify distinct patterns of brain structural changes for differentiating parkinsonian syndromes.
- To evaluate the efficacy of magnetic resonance imaging (MRI) techniques in distinguishing between PD, MSA-P, and MSA-C.
Main Methods:
- Applied discriminant analysis to T1-weighted volume and diffusion tensor imaging data from 18 MSA-C, 12 MSA-P, 21 PD patients, and 21 healthy controls.
- Utilized voxel-based morphometry and tract-based spatial statistics for brain structure evaluation.
- Employed stepwise discriminant functions for classification.
Main Results:
- Achieved a correct classification rate of 0.89 using stepwise discriminant functions.
- Combining three independent structural brain variables yielded the same classification accuracy as the stepwise method.
- Identified that each parkinsonian syndrome exhibits structural deviations across multiple brain regions.
Conclusions:
- Structural brain measures, particularly when combined, can effectively distinguish between parkinsonian syndromes.
- MRI-based discriminant analysis offers a promising tool for early and accurate diagnosis of PD and MSA subtypes.
