An analytical model for estimating water exchange rate in white matter using diffusion MRI

Area of Science:

• Neuroimaging
• Biophysics
• Computational Biology

Background:

• Diffusion weighted magnetic resonance imaging (DWMRI) models white matter microstructure, but water exchange rate (WXR) between intra- and extra-axonal spaces remains under-investigated.
• WXR is a critical biomarker for brain injuries including multiple sclerosis (MS), stroke, and traumatic brain injury (TBI).
• Existing diffusion models often rely on approximations like the short gradient pulse (SGP) approximation, limiting their applicability.

Purpose of the Study:

• To develop a novel analytical diffusion model incorporating WXR without the SGP approximation.
• To validate the model's accuracy in estimating WXR and microstructural parameters using simulations and experimental data.
• To assess the model's utility in detecting microstructural changes associated with TBI.

Main Methods:

• Derived a diffusion signal model for a permeable cylinder using a clinically feasible pulse gradient spin echo (PGSE) sequence.
• Validated the model with Markov Random Walk simulations, correlating estimated exchange parameters with actual WXR (R2>0.88).
• Applied the model to ex vivo rat brain diffusion data (hybrid diffusion imaging) and compared results with histological measurements.

Main Results:

• Simulations confirmed a linear correlation between the model's exchange parameter and actual WXR.
• Increasing WXR led to increased estimated axon diameter and decreased estimated volume fraction, consistent with TBI histology.
• The model accurately predicted axon diameter and volume fraction in normal rat brains (ICC=0.96).
• Significantly increased WXR and diameter, with decreased volume fraction, were observed at the TBI boundary in rats compared to controls (p<0.001).

Conclusions:

• The proposed analytical model accurately estimates WXR and microstructural parameters from DWMRI data without SGP approximation.
• The model demonstrates sensitivity to microstructural alterations in TBI, showing significant changes in WXR and axon morphology.
• This approach offers a promising tool for non-invasive assessment of brain injury and disease progression.