Angular smoothing and radial regularization of ODF fields: application on deterministic crossing fiber tractography

Area of Science:

• Neuroimaging
• Diffusion MRI
• Computational Neuroscience

Background:

• High angular resolution diffusion imaging (HARDI) offers insights into complex brain pathways.
• Clinical HARDI acquisition faces challenges like low resolution and high noise levels.
• Delineating crossing and branching fibers remains difficult under these constraints.

Purpose of the Study:

• To present a novel spatial filter for Orientation Distribution Function (ODF) fields.
• To address noise and resolution limitations in clinical HARDI.
• To improve tractography accuracy, especially in regions with complex fiber architectures.

Main Methods:

• Development of a spatial filter for ODF fields utilizing local structural information.
• Implementation of a directionally selective method for angular smoothing and radial regularization (ASRR).
• Proposal of a dynamic regularization scheme applied selectively to multimodal ODFs.

Main Results:

• The ASRR method effectively reduces noise in ODF fields.
• Improved accuracy of diffusion peaks and enhanced signals for non-dominant fibers were observed.
• Successful demonstration of improved tractography for challenging pathways in a human in vivo dataset.

Conclusions:

• The proposed spatial filter (ASRR) significantly enhances HARDI data quality under clinical conditions.
• This method improves the tractography of complex and non-dominant white matter pathways.
• The technique holds promise for more accurate neuroimaging analysis in clinical settings.