A regularized two-tensor model fit to low angular resolution diffusion images using basis directions.

Area of Science:

• Neuroimaging
• Diffusion Tensor Imaging (DTI)
• Computational Neuroscience

Background:

• Partial volume effects in DTI can cause significant artifacts.
• High angular resolution diffusion imaging (HARDI) is not yet routinely used.
• Resolving crossing fibers is crucial for accurate brain connectivity mapping.

Purpose of the Study:

• To develop a method for resolving and regularizing orientation estimates of two crossing fibers.
• To utilize conventional DTI datasets, avoiding the need for HARDI.
• To improve the accuracy of diffusion tensor imaging analysis in complex white matter regions.

Main Methods:

• A regularized two-tensor model was developed, exploiting planar diffusion profiles in crossing regions.
• A regularization scheme was applied to mitigate noise artifacts from limited DTI data.
• Relaxation labeling was used to ensure continuity of orientation estimates across neighboring voxels.

Main Results:

• The proposed two-tensor model with spatial regularization significantly improved orientation estimates.
• Results from simulations and human data showed good agreement with known anatomical structures.
• Revised fractional anisotropy (FA) and mean diffusivity (MD) values were computed.

Conclusions:

• The method successfully resolves orientational, anisotropy, and diffusivity information in regions with two crossing fibers.
• Full brain coverage scans acquired in under six minutes are sufficient for this analysis.
• This technique offers a practical approach to studying complex white matter architecture using conventional DTI.