A Symmetry-Based Method to Infer Structural Brain Networks from Probabilistic Tractography Data.

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Network Science

Background:

• Diffusion MRI and tractography yield abundant structural connectivity data.
• Existing methods often rely on arbitrary connectivity thresholds.
• Probabilistic tractography has inherent limitations regarding connection polarity.

Purpose of the Study:

• To describe and evaluate a novel method for inferring structural brain networks from tractography data.
• To address the polarity ambiguity inherent in diffusion MRI streamline data.
• To provide a robust method for mapping brain connectivity between Regions of Interest (ROIs).

Main Methods:

• The Minimum Asymmetry Network Inference Algorithm (MANIA) is proposed.
• MANIA formulates network inference as an optimization problem minimizing observed network asymmetry.
• The method leverages the bidirectional nature of tractography, regardless of true connection polarity.

Main Results:

• MANIA was evaluated using the FiberCup dataset and a synthetic network noise model.
• The algorithm successfully infers structural brain networks from diffusion MRI data.
• Application to 28 healthy subjects revealed structural networks between 18 corticolimbic ROIs.

Conclusions:

• MANIA offers a robust approach to inferring structural brain networks by addressing tractography limitations.
• The method provides a valuable tool for analyzing brain connectivity in relation to neuropsychiatric conditions.
• This work advances the understanding of structural brain networks and their clinical relevance.