Quasi-Diffusion Imaging: Application to Ultra-High b-Value and Time-Dependent Diffusion Images of Brain Tissue

Area of Science:

• Neuroimaging
• Biophysics
• Medical Physics

Background:

• Diffusion MRI (dMRI) is crucial for non-invasively probing brain microstructure.
• Existing dMRI models face limitations in accurately representing signal behavior across broad b-value ranges.
• Understanding signal attenuation is key to extracting meaningful biological information.

Purpose of the Study:

• To introduce and validate Quasi-Diffusion Imaging (QDI) as an advanced signal representation for dMRI.
• To evaluate QDI's performance in both in vivo human and ex vivo rat brain tissue.
• To assess the accuracy and efficiency of QDI parameter estimation within clinically feasible scan times.

Main Methods:

• QDI models dMRI signal attenuation using two parameters within a Mittag-Leffler function (MLF).
• The study analyzed human and rat brain datasets across extensive b-value ranges (0-25,000 s/mm²).
• QDI's mathematical properties, including asymptotic behavior and inflection points, were investigated.

Main Results:

• QDI demonstrated excellent fits to observed dMRI signal attenuation.
• The method successfully identified signal inflection points and exhibited a negative power-law regime.
• Stable QDI parameter estimates were achieved even with clinically feasible scan times and varying diffusion times.

Conclusions:

• QDI provides a parsimonious and accurate representation of dMRI signal attenuation in brain tissue.
• The QDI model is sensitive to microstructural heterogeneity and cell membrane permeability.
• QDI enables accurate parameter estimation from reduced scan times, advancing clinical applicability.