A registration method for improving quantitative assessment in probabilistic diffusion tractography.

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Diffusion MRI

Background:

• Diffusion MRI probabilistic tractography is vital for studying brain connectivity in health and disease.
• Quantifying white matter tract differences is challenging due to individual anatomical variability and lack of clear landmarks.
• Existing methods struggle to accurately align tracts, potentially compromising groupwise and longitudinal analyses.

Purpose of the Study:

• To quantify spatial variance in white matter tracts aligned using standard registration methods.
• To assess the impact of tract location mismatch on groupwise amplitude comparisons.
• To develop and validate an improved tract alignment method for enhanced quantitative analysis.

Main Methods:

• Seed-based probabilistic diffusion tractography was performed on motor and visual cortices projecting through the corpus callosum.
• Spatial variance of tracts was quantified using a standard whole-brain registration approach.
• A novel tract alignment method was developed and tested against the standard method.

Main Results:

• Tract amplitude decreased significantly with misalignment (74% reduction at 3.2 mm for transcallosal tracts).
• Standard whole-brain registration was insufficient for cortical tracts but adequate for subcortical tracts.
• The new alignment method substantially reduced location mismatch and improved statistical power for comparisons.

Conclusions:

• Precise tract alignment is critical for accurate amplitude comparisons in diffusion MRI tractography, especially for cortical tracts.
• Location mismatches can lead to false positives/negatives in group and longitudinal studies.
• The developed tract alignment method enhances the reliability and statistical power of quantitative connectivity analyses.