Microscopic anisotropy misestimation in spherical-mean single diffusion encoding MRI.

Area of Science:

• Neuroimaging
• Diffusion MRI
• Biophysics

Background:

• Microscopic fractional anisotropy (µFA) is crucial for differentiating microstructural properties from orientation dispersion in biological tissues.
• Double diffusion encoding (DDE) MRI is the established method for accurate µFA estimation.
• Recent proposals suggest using single diffusion encoding (SDE) signals with standard models (SM) for µFA estimation.

Purpose of the Study:

• To evaluate µFA estimation using powder-averaged SDE signals with the spherical mean technique (SMT) constraints.
• To assess the general applicability of powder-averaged SM signals for µFA estimation.

Main Methods:

• Ex vivo mouse brain SDE MRI experiments were conducted at 16.4 T (Δ/δ = 12/1.5 ms).
• µFA maps from SDE were compared to model-free µFA maps from DDE-MRI (Δ/τ/δ = 12/12/1.5 ms).
• Theoretical analysis and simulations explored heterogeneity effects on estimations.

Main Results:

• Powder-averaged SDE signals showed significant deviations from ground truth µFA in both gray and white matter.
• Simulations indicated that unmodeled factors like inter- and intracompartmental kurtosis contribute to misestimations.
• The SMT and 2-component SM models failed to accurately report µFA and other microstructural parameters.

Conclusions:

• Powder-averaged SMT and SM are currently inadequate for accurate µFA quantification in ex vivo tissues.
• Model assumptions and constraints can severely impact the specificity of microstructural parameter estimation.
• Further research and validation are essential before clinical or preclinical application of these SDE-based methods.