Rxivist logo

Stick Stippling for Joint 3D Visualization of Diffusion MRI Fiber Orientations and Density

By Ryan P Cabeen, David H Laidlaw, Arthur W. Toga

Posted 16 Jun 2020
bioRxiv DOI: 10.1101/2020.06.15.153098

This paper investigates a stick stippling approach for glyph-based visualization of complex neural fiber architecture derived from diffusion magnetic resonance imaging. The presence of sub-voxel crossing fibers in the brain has prompted the development of advanced modeling techniques; however, there remains a need for improved visualization techniques to more clearly convey their rich structure. While tractography can illustrate large scale anatomy, visualization of diffusion models can provide a more complete picture of local anatomy without the known limitations of tracking. We identify challenges and evaluate techniques for visualizing multi-fiber models and identified techniques that improve on existing methods. We conducted experiments to compare these representations and evaluated them with in vivo diffusion MR datasets that vary in voxel resolution and anisotropy. We found that stick rendering as 3D tubes increased legibility of fiber orientation and that encoding fiber density by tube radius reduced clutter and reduced dependence on viewing orientation. Furthermore, we identified techniques to reduce the negative perceptual effects of voxel gridding through a jittering and re- sampling approach to produce a stippling effect. Looking forward, this approach provides a new way to explore diffusion MRI datasets that may aid in the visual analysis of white matter fiber architecture and microstructure. Our software implementation is available in the Quantitative Imaging Toolkit (QIT). ### Competing Interest Statement The authors have declared no competing interest.

Download data

  • Downloaded 266 times
  • Download rankings, all-time:
    • Site-wide: 130,541
    • In neuroscience: 19,554
  • Year to date:
    • Site-wide: 107,740
  • Since beginning of last month:
    • Site-wide: 119,727

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide