A Unifying Framework for Spatial and Temporal Diffusion in Diffusion MRI.

Area of Science:

• Biophysics
• Neuroimaging
• Medical Physics

Background:

• Current diffusion-weighted MRI (dMRI) frameworks often fix temporal dependency, limiting analysis of tissue microstructure.
• Techniques like Axcaliber and ActiveAx highlight the importance of diffusion time in dMRI signal analysis.
• A generalized framework simultaneously modeling spatio-temporal dMRI signal dependence is lacking.

Purpose of the Study:

• To develop a novel framework for simultaneous representation of the dMRI signal across diffusion times, gradient strengths, and directions.
• To create a generalized model for the 3D+t spatio-temporal dMRI signal.
• To enable more accurate estimation of microstructural parameters.

Main Methods:

• Utilized a functional basis to fit the 3D+t spatio-temporal dMRI signal, analogous to the 3D-SHORE basis.
• Employed regularization by minimizing the analytic Laplacian norm of the basis.
• Validated on synthetic data from Callaghan et al.'s model and real ActiveAx acquisition data.

Main Results:

• The proposed method accurately describes restricted spatio-temporal signal decay in tissue models like cylindrical pores.
• Demonstrated robustness to noise in dMRI signal analysis.
• Successfully estimated axon radius distribution parameters using approaches similar to AxCaliber.

Conclusions:

• The novel framework enables comprehensive representation of the complete 3D+t dMRI signal.
• This approach offers improved insights into normal and pathological nervous tissue.
• Facilitates advanced microstructural analysis in neuroimaging.