Correction for diffusion MRI fibre tracking biases: The consequences for structural connectomic metrics

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Network Science

Background:

• Diffusion MRI streamlines tractography is key for brain connectome reconstruction.
• Streamline counts for structural connectivity are limited by termination ambiguity and non-quantitative nature.

Purpose of the Study:

• To investigate how advanced connectome construction methods impact structural brain network analysis.
• To evaluate the effects of Anatomically Constrained Tractography (ACT) and Spherical-deconvolution Informed Filtering of Tractograms (SIFT) on network metrics.

Main Methods:

• Utilized state-of-the-art connectome construction techniques, including ACT and SIFT.
• Applied graph theoretical approaches to analyze structural brain networks derived from diffusion MRI.
• Compared network characteristics obtained using different tractogram reconstruction methods.

Main Results:

• Connectome variability, global network metrics, small-world attributes, and network hubs were significantly altered by improved tractogram accuracy.
• Anatomically Constrained Tractography (ACT) and Spherical-deconvolution Informed Filtering of Tractograms (SIFT) enhanced biological accuracy.
• Standard inverse length correction for connection density proved insufficient.

Conclusions:

• Advanced tractogram reconstruction techniques like ACT and SIFT are crucial for accurate structural connectomics.
• Improvements in streamline tractogram accuracy lead to more reliable structural brain network characterization.
• Relying solely on streamline counts or basic corrections is inadequate for robust connectomics research.