Inherent spatial structure in myelin water fraction maps

Area of Science:

• Neuroimaging
• White Matter Tract Analysis
• Quantitative MRI

Background:

• Myelin water fraction (MWF) images are often spatially noisy, leading to averaging over large white matter (WM) volumes.
• The inherent spatial structure of MWF within WM tracts is not well understood.
• Current methods often overlook potential tract-specific information in MWF data.

Purpose of the Study:

• To investigate the existence of inherent spatial structure in brain MWF maps.
• To explore the benefits of examining MWF values along diffusion tensor imaging (DTI)-derived white matter tracts.
• To compare the spatial anisotropy of MWF with fractional anisotropy (FA).

Main Methods:

• Extracted 16 major white matter fibre bundles using DTI data from 41 healthy subjects.
• Computed MWF coefficients of variation (CoV) in sub-segments along each tract and compared them to surrounding voxels.
• Assessed MWF consistency along fibre bundles across subjects and its accuracy in age estimation.

Main Results:

• MWF and FA coefficients of variation were lower within fibre bundles compared to surrounding areas.
• Both MWF and FA showed spatial gradients of CoV, with smaller gradients aligned along fibre tracts.
• MWF profiles along WM tracts were consistent across subjects and more accurate for age estimation than whole-tract averages.

Conclusions:

• Despite visual noise, MWF distribution in white matter follows a distinct spatial pattern along fibre bundles.
• Assessing MWF in sections along white matter tracts provides a sensitive and robust method for myelin assessment.
• This tract-based analysis enhances the utility of MWF in neuroimaging research and clinical applications.