Diffusion time dependence, power-law scaling, and exchange in gray matter

Area of Science:

• Neuroimaging
• Biophysics
• Diffusion MRI

Background:

• Diffusion MRI (dMRI) is crucial for neuroimaging but signal specificity is limited, especially in gray matter (GM).
• Current biophysical models like the Standard Model (SM) and SANDI model have assumptions that need experimental validation in GM.
• The b⁻¹/² power-law is a key signature of 1D diffusion, but its absence in GM is attributed to factors like water exchange and soma signals.

Purpose of the Study:

• To experimentally validate assumptions of dMRI biophysical models in gray matter.
• To identify conditions where neurite signals dominate in GM dMRI.
• To differentiate between water exchange and restricted diffusion effects in GM.

Main Methods:

• Diffusion simulations in realistic neuron models.
• Analysis of GM dMRI signal behavior across diffusion weighting and time.
• Ultrahigh field (16.4T) dMRI experiments on ex vivo rat brain.
• Application of the Kärger exchange model and integration into SM and SANDI models.

Main Results:

• Neurite signals in simulations are masked by soma signals under typical conditions but can be isolated experimentally.
• Neurite signals in impermeable structures exhibit a power-law behavior, even with curvature and branching.
• Observed time dependence in GM dMRI data suggests significant water exchange and a sub-population of impermeable neurites.
• Exchange-driven time dependence mimics restricted diffusion, necessitating careful interpretation.

Conclusions:

• The study identifies an experimental regime to isolate neurite signals in GM dMRI.
• Findings suggest substantial water exchange and a small population of impermeable neurites in GM.
• The Kärger exchange model, integrated into SM and SANDI, aids in quantifying these GM microstructural properties.