Parametric representation of multiple white matter fascicles from cube and sphere diffusion MRI

Area of Science:

• Neuroimaging
• Diffusion MRI
• White Matter Tractography

Background:

• Characterizing complex brain diffusion signals is challenging.
• Existing models often focus on global diffusion profiles, limiting connectivity assessment.
• Multiple Fascicle Models (MFM) offer potential for detailed fascicle analysis but face numerical instability.

Purpose of the Study:

• To address the limitations of single-shell HARDI in estimating multiple fascicle models.
• To propose a novel acquisition scheme and estimation algorithm for accurate multi-tensor modeling.
• To enable robust investigation of white matter connectivity and fascicle diffusion properties.

Main Methods:

• Analytical demonstration of parameter co-linearity issues in single-shell acquisitions.
• Introduction of the CUSP (CUbe and SPhere) optimal acquisition scheme with multiple b-values.
• Development of a novel, efficient, and robust multi-tensor estimation algorithm.
• Evaluation using synthetic phantoms and in vivo data.

Main Results:

• Single-shell HARDI is insufficient for precise multi-tensor model estimation due to parameter co-linearity.
• The CUSP acquisition scheme enhances signal-to-noise ratio compared to multi-shell HARDI.
• The proposed algorithm combined with CUSP acquisition enables full multi-tensor model estimation.
• CUSP-MFM accurately characterizes multiple fascicles from short acquisitions.

Conclusions:

• The CUSP acquisition scheme and MFM provide a robust solution for multi-fascicle characterization.
• This approach overcomes previous numerical challenges in multi-tensor modeling.
• CUSP-MFM facilitates rapid and effective investigation of white matter in health and disease.