Resolving power for the diffusion orientation distribution function.

Area of Science:

• Neuroimaging
• Diffusion MRI
• Computational Neuroscience

Background:

• The diffusion orientation distribution function (dODF) is essential for reconstructing white matter (WM) fiber architecture using diffusion MRI.
• Accurate tractography relies on the dODF's ability to resolve complex fiber configurations, such as crossings.

Purpose of the Study:

• To investigate the resolving power of the diffusion orientation distribution function (dODF) for white matter fiber tractography.
• To evaluate the performance of different dODF approximations against an exact dODF model using a simple two-fiber bundle configuration.

Main Methods:

• The Sparrow criterion was employed to quantitatively assess the resolving power of various dODF models.
• Evaluated models included the exact dODF, q-space imaging (QSI), q-ball, and kurtosis approximations.
• Theoretical and numerical calculations were performed to determine resolving power dependencies.

Main Results:

• Resolving power was found to be dependent on the eigenvalues of the diffusion model and the radial weighting of the dODF.
• The resolving power of QSI and q-ball dODFs improved with increasing b-value.
• The kurtosis dODF demonstrated a resolving power comparable to the exact dODF.

Conclusions:

• All dODFs, exact and approximate, possess finite resolving powers that inherently limit sensitivity to complex fiber crossings.
• The resolving powers established in this study serve as valuable benchmarks for comparing the performance of different dODF methods.
• Findings provide insights into the limitations and comparative strengths of various dODF approximations in diffusion MRI tractography.