Axial and radial axonal diffusivities and radii from single encoding strongly diffusion-weighted MRI.

Area of Science:

• Neuroimaging
• Biophysics
• Computational Neuroscience

Background:

• Diffusion MRI (dMRI) is crucial for studying white matter microstructure.
• Strong diffusion weightings approximate white matter signal as axonal contributions.
• Current methods struggle to estimate axial diffusivity (AD) due to signal averaging.

Purpose of the Study:

• To develop a novel method for estimating per-axon axial and radial diffusivities from diffusion-weighted MRI data.
• To improve the accuracy of radial diffusivity (RD) estimation compared to spherical averaging.
• To enable more precise modeling of white matter axons, particularly in multi-compartmental models.

Main Methods:

• Utilized single-encoding, strongly diffusion-weighted pulsed gradient spin echo MRI data.
• Introduced kernel zonal modeling for estimating both AD and RD.
• Applied the method to data from the MGH Adult Diffusion Human Connectome Project.

Main Results:

• Successfully enabled estimation of per-axon axial diffusivity.
• Improved estimation of per-axon radial diffusivity compared to spherical averaging.
• Reported reference values for axonal diffusivities across 34 subjects and derived axonal radii estimates.

Conclusions:

• The kernel zonal modeling approach provides accurate per-axon AD and RD estimates.
• This method offers potential for bias-free estimates, reducing partial volume effects.
• The findings contribute to more refined white matter microstructure analysis and modeling.