OpenDIVE: Diffusion Imaging Visualization for Everyone

[ElyssaMcMaster]

Title

OpenDIVE: DIffusion Visualization for Everyone

Leaders

Elyssa McMaster (@ElyssaMcMaster)
Adam Saunders (@saundersresearch)

Collaborators

No response

Project description

Despite the prolific availability of software tools to visualize diffusion MRI data, there is no standardized visualization software. Similarly, the current standard of visualizations are not accessible to people with common forms of colorblindness. We propose a software package to both standardize and improve accessibility to representations of diffusion data.

Link to project repository/sources

https://github.com/MASILab/open_dive

Concerete goals with specific tasks for Brainhack Vanderbilt 2025

1. Generate a python package that accepts slice and view inputs and outputs a consistent image display.
2. Generate a color map for tractography and DTI/ODF to be accessible to a wide range of audiences.
3. Add DTI/ODF/Tractography display functionality to the python package.

Good first issues

1. issue one: Download data and DIPY (instructions in OpenDIVE GitHub). Send a screenshot of a brain to a friend and upload as an issue in GitHub.

2. issue two: Take a look at Adam's previous research code on the OpenDIVE GitHub. Edit it to include a color bar on the visualization.

Skills

• Python: Beginner - advanced. Experience with visualization tools like Matplotlib is a plus
• GitHub: Beginner. Pull, commit, issues
• Technical writing

Onboarding documentation

https://github.com/MASILab/open_dive

What will participants learn?

• Image visualization techniques
• Basics of diffusion MRI and its models

Public data to use

We will use one of the public datasets available from DIPY -- instructions to collect data will be on the OpenDIVE GitHub.

Number of collaborators

4+

Credit to collaborators

Project collaborators will be listed in the project's README on GitHub. We will include a section that says "If you use this tool in your work, please cite it as..." with all names listed in the GitHub and in the documentation.

Image

Project Summary

Develop a software package to both standardize and improve accessibility to representations of diffusion data.

Type

visualization

Development status

0_concept_no_content

Topic

connectome, data_visualisation, tractography, other

Tools

DIPY

Programming language

Python

Modalities

DWI, MRI

Git skills

1_commit_push, 2_branches_PRs

Anything else?

Diffusion models (DTI, ODF)

Things to do after the project is submitted and ready to review.

• Add a comment below the main post of your issue saying: Hi @brainhack-vandy/project-monitors my project is ready!

[ElyssaMcMaster]

Hi @brainhack-vandy/project-monitors my project is ready!

[neurolabusc]

@ElyssaMcMaster this might be a good project for collaboration with the NiiVue project, which could benefit both Python and web applications.

• NiiVue provides many default colormaps, as shown in this live demo: in particular colormaps designed for individuals with color vision deficiency. You can specify your own colormaps.
• NiiVue is able to show the V1 from tensors, which is useful to ensure the the b-vector orientation is correct. As shown in this live demo the V1 can be modulated with FA and can also be shown as cylinders to aid insight regarding alignment. Uniquely, these cylinders require only a single GPU Texture, whereas other tools use line segments. This approach is ideal for low powered devices (e.g. phones).
• NiiVue can show streamlines as either lines or cylinders.
• Since NiiVue is based on WebGL2, it can be embedded directly into Jupyter notebooks. In contrast, the DiPy visualization uses OpenGL, so must have a separate window. ipyniivue provides an emerging Pythonic wrapper, while nilearn uses direct insertion.
• I will give a talk at the brainhack on Saturday, but happy to work with your team throughout the hack if this sounds like an approach you want to pursue.

[neurolabusc]

Here is the prototype for using pink-green-blue (PGB) instead of red-green-blue (RGB). The top row shows traditional RGB for trichromacy (left) and protanopia (right). The lower row shows PGB for trichromacy (left) and protanopia (right).