Combinatorial fiber-tracking of the human brain

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Graph Theory

Background:

• Diffusion imaging tractography aims to map white matter pathways.
• Conventional methods face limitations, particularly with stopping criteria and true connectivity analysis.
• Accurate reconstruction of neural pathways is crucial for understanding brain function and disease.

Purpose of the Study:

• To present a novel fiber-tracking algorithm, combinatorial tracking, for enhanced tractography.
• To model white matter connectivity using stochastic processes and global optimization.
• To enable true connectivity analysis in diffusion imaging.

Main Methods:

• Transforming white matter into a weighted grid graph with 26 connections per voxel.
• Modeling random walks using a Markov Chain model.
• Employing shortest path algorithms and calculating the mean first passage time (MFPT) matrix for fiber reconstruction.
• Introducing a simulation framework for matrix element calculation and target selection.

Main Results:

• Demonstrated reconstruction of key brain pathways: cortico-thalamic tract, pyramidal decussation, and medial cerebellar peduncle fibers.
• Showcased the ability to perform true connectivity analysis, overcoming conventional tracking limitations.
• Validated the use of diffusion tensor imaging (DTI) ellipsoids and potential for other orientation density functions (ODFs).

Conclusions:

• Combinatorial tracking offers a robust framework for brain white matter tractography.
• The algorithm enables true connectivity analysis, advancing the field beyond conventional methods.
• This approach has the potential to establish diffusion imaging tractography as a reliable measure of brain connectivity.