Scalable Brain Network Construction on White Matter Fibers

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Graph Theory

Background:

• Diffusion Tensor Imaging (DTI) enables non-invasive characterization of human brain structural connectivity.
• Whole brain tractography generates extensive datasets of white matter fiber tracts, forming large 3D graphs.
• Standardized methods for constructing brain structural network graphs from tractography data are lacking.

Purpose of the Study:

• To present a novel, scalable framework for building brain structural network graphs.
• To apply this framework to investigate abnormal brain connectivity in autism spectrum disorder.

Main Methods:

• Development of a scalable iterative framework termed the ε-neighbor method.
• Application of the ε-neighbor method to construct network graphs from large-scale DTI tractography data.
• Utilizing the generated graphs to test for abnormal connectivity patterns.

Main Results:

• The ε-neighbor method provides a scalable approach for network graph construction from DTI data.
• The framework successfully generated brain structural networks for analysis.
• The method was applied to identify potential connectivity abnormalities relevant to autism.

Conclusions:

• The ε-neighbor method offers a robust and scalable solution for building brain structural networks.
• This approach facilitates the investigation of brain connectivity in neurological and psychiatric disorders.
• The framework has potential applications in understanding the neural basis of autism.