Global tractography with embedded anatomical priors for quantitative connectivity analysis

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Biomedical Engineering

Background:

• Diffusion magnetic resonance imaging (dMRI) enables non-invasive reconstruction of white matter (WM) fiber pathways.
• Tractography algorithms, categorized as local or global, analyze directional information in dMRI data.
• Current global tractography methods lack anatomical priors, leading to inaccurate reconstructions and premature tract termination in WM, limiting brain connectivity analysis.

Purpose of the Study:

• To develop a novel global tractography approach for improved brain connectivity analysis.
• To explicitly incorporate anatomical priors and tract volume into the optimization process.
• To address limitations of existing methods, such as premature tract termination and lack of quantitative tract contribution.

Main Methods:

• Proposed a novel global tractography algorithm integrating anatomical priors into the optimization process.
• Incorporated the effective contribution of each tract (i.e., volume) to the diffusion MRI signal.
• Evaluated the approach using a realistic dMRI phantom and in vivo data.

Main Results:

• The novel method demonstrated improved reconstruction of WM fiber pathways compared to existing algorithms.
• Explicitly enforcing anatomical priors ensured tracts originated in gray matter and traversed white matter.
• Considering tract volume provided a more quantitative assessment of structural connectivity.

Conclusions:

• The developed global tractography method enhances the accuracy and reliability of structural connectivity assessments.
• This approach overcomes key limitations of previous tractography techniques, offering more biologically plausible and quantitative results.
• The findings have significant implications for understanding brain networks and diagnosing neurological disorders through dMRI analysis.