Effects of DTI spatial normalization on white matter tract reconstructions

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Medical Image Analysis

Background:

• Brain white matter (WM) pathways are reconstructed using tractography on diffusion tensor imaging (DTI).
• Tractography on native DTI is often preferred over normalized DTI due to potential orientation errors during non-linear transformations.
• Population analyses and tract-specific analyses (TSA) increasingly utilize tractography on normalized DTI.

Purpose of the Study:

• To evaluate and compare white matter tract reconstructions from normalized DTI versus native DTI.
• To determine if non-linear spatial normalization preserves long-range anatomical connections for accurate tractography.
• To assess the impact of normalization on the consistency and variance of tractography results.

Main Methods:

• Diffusion tensor imaging (DTI) data acquisition.
• Non-linear spatial normalization of DTI datasets.
• White matter tractography performed on both native and normalized DTI data.
• Comparison of tract reconstructions and analysis of macro/microstructural properties.

Main Results:

• Spatial normalization of DTI data successfully preserves the tract reconstructions of major white matter pathways.
• Normalization does not significantly alter the variance (individual differences) in macro and microstructural properties of WM pathways.
• Tractography on normalized DTI yields consistent results comparable to native DTI for major pathways.

Conclusions:

• Non-linear spatial normalization of DTI data is suitable for preserving major white matter pathway reconstructions.
• Quantitative and shape properties can be reliably extracted from tractography data in normalized DTI for population statistics.
• This validates the use of normalized DTI for large-scale studies investigating white matter anatomy and integrity.