Tutorial: a guide to diffusion MRI and structural connectomics

Ittai Shamir,Yaniv Assaf
DOI: https://doi.org/10.1038/s41596-024-01052-5
2024-09-04
Abstract:Diffusion magnetic resonance imaging (dMRI) is a versatile imaging technique that has gained popularity thanks to its sensitive ability to measure displacement of water molecules within a living tissue on a micrometer scale. Although dMRI has been around since the early 1990s, its applications are constantly evolving, primarily regarding the inference of structural connectomics from nerve fiber trajectories. However, these applications require expertise in image processing and statistics, and it can be difficult for a newcomer to choose an appropriate pipeline to fit their research needs, not least because dMRI is such a flexible methodology that dozens of acquisition and analysis pipelines have been developed over the years. This introductory guide is designed for graduate students and researchers in the neuroscience community who are interested in integrating this new methodology regardless of their background in neuroimaging and computational tools. The guide provides a brief overview of the basic dMRI methodologies but focuses on its applications in neuroplasticity and connectomics. The guide starts with dMRI experimental designs and a complete step-by-step pipeline for structural connectomics. The following section covers the basics of dMRI, including parameters and clinical applications (apparent diffusion coefficient, mean diffusivity, fractional anisotropy and microscopic fractional anisotropy), as well as different approaches and models. The final section focuses on structural connectomics, covering subjects from fiber tracking (techniques, evaluation and limitations) to structural networks (constructing, analyzing and visualizing a network).
What problem does this paper attempt to address?