Leading non-Gaussian corrections for diffusion orientation distribution function.

Area of Science:

• Neuroimaging
• Biophysics
• Medical Physics

Background:

• Diffusion Tensor Imaging (DTI) relies on a Gaussian diffusion model, limiting its accuracy in complex white matter regions.
• Intra-voxel fiber crossings and complex diffusion patterns are not fully captured by DTI's Gaussian approximation.

Purpose of the Study:

• To present an analytical representation of non-Gaussian diffusion effects for diffusion orientation distribution functions (dODFs).
• To establish a foundation for improved white matter fiber tractography using DKI.

Main Methods:

• Developed an analytical formula incorporating diffusion and diffusional kurtosis tensors estimated via DKI.
• Incorporated model-independent non-Gaussian diffusion effects beyond the Gaussian approximation of DTI.
• Validated the formula using numerical simulations and human brain data.

Main Results:

• The new analytical representation improves upon the Gaussian approximation used in DTI.
• DKI-based tractography shows potential for more accurate fiber orientation predictions.
• The method demonstrates the ability to resolve intra-voxel fiber crossings effectively.

Conclusions:

• The leading non-Gaussian corrections significantly impact white matter fiber tractography.
• This approach offers advantages over conventional DTI, particularly in regions with common fiber crossings like the centrum semiovale.