Abstract
The prominent morphometric alterations of Alzheimer's disease (AD) occur both in gray matter and in white matter. Multimodal fusion can examine joint information by combining multiple neuroimaging datasets to identify the covariant morphometric alterations in AD in greater detail. In the current study, we conducted a multimodal canonical correlation analysis and joint independent component analysis to identify the covariance patterns of the gray and white matter by fusing structural magnetic resonance imaging and diffusion tensor imaging data of 39 AD patients (23 males and 16 females, mean age: 74.91. ±. 8.13. years) and 41 normal controls (NCs) (20 males and 21 females, mean age: 73.97. ±. 6.34. years) derived from the Alzheimer's Disease Neuroimaging Initiative database. The results revealed 25 joint independent components (ICs), of which three joint ICs exhibited strong links between the gray matter volume and the white matter fractional anisotropy (FA) and significant differences between the AD and NC group. The joint IC maps revealed that the simultaneous changes in the gray matter and FA values primarily involved the following areas: (1) the temporal lobe/hippocampus-cingulum, (2) the frontal/cingulate gyrus-corpus callosum, and (3) the temporal/occipital/parietal lobe-corpus callosum/corona radiata. Our findings suggest that gray matter atrophy is associated with reduced white matter fiber integrity in AD and possibly expand the understanding of the neuropathological mechanisms in AD.
Original language | English |
---|---|
Pages (from-to) | 553-562 |
Number of pages | 10 |
Journal | Neuroscience |
Volume | 301 |
DOIs | |
Publication status | Published - 1 Aug 2015 |
Externally published | Yes |
Keywords
- Alzheimer's disease
- CCA and ICA
- Fractional anisotropy
- Gray matter
- Structural MRI and DTI