pdaude / fsleyes

This is a mirror. Feel free to use the issue tracker. PRs welcome.

FSLeyes

FSLeyes is the FSL image viewer.

Installation

FSLeyes is a GUI application written in Python, and built on wxPython. FSLeyes requires OpenGL for visualisation.

In the majority of cases, you should be able to follow the installation instructions outlined at the FSLeyes home page:

https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLeyes.

Dependencies

All of the core dependencies of FSLeyes are listed in requirements.txt.

Some extra dependencies, which provide additional functionality, are listed in requirements-extra.txt and requirements-notebook.txt.

Dependencies for running tests and building the documentation are listed in requirements-dev.txt.

Being an OpenGL application, FSLeyes can only be used on computers with graphics hardware (or a software GL renderer) that supports one of the following versions:

• OpenGL 3.3
• OpenGL 2.1, with the following extensions:
  • EXT_framebuffer_object
  • ARB_instanced_arrays
  • ARB_draw_instanced
• OpenGL 1.4, with the following extensions:
  • ARB_vertex_program
  • ARB_fragment_program
  • EXT_framebuffer_object
  • GL_ARB_texture_non_power_of_two

Documentation

The FSLeyes user and API documentation are hosted at:

• https://open.win.ox.ac.uk/pages/fsl/fsleyes/fsleyes/userdoc/
• https://open.win.ox.ac.uk/pages/fsl/fsleyes/fsleyes/apidoc/

The FSLeyes user and API documentation is written in ReStructuredText, and can be built using sphinx:

pip install -r requirements-dev.txt
python setup.py userdoc
python setup.py apidoc

The documentation will be generated and saved in userdoc/html/ and apidoc/html/.

Credits

Some of the FSLeyes icons are derived from the Freeline icon set, by Enes Dal, available at https://www.iconfinder.com/Enesdal, and released under the Creative Commons (Attribution 3.0 Unported) license.

The volumetric spline interpolation routine uses code from:

Daniel Ruijters and Philippe Thévenaz, GPU Prefilter for Accurate Cubic B-Spline Interpolation, The Computer Journal, vol. 55, no. 1, pp. 15-20, January 2012. http://dannyruijters.nl/docs/cudaPrefilter3.pdf

The GLSL parser is based on code by Nicolas P . Rougier, available at https://github.com/rougier/glsl-parser, and released under the BSD license.

DICOM to NIFTI conversion is performed with Chris Rorden's dcm2niix (https://github.com/rordenlab/dcm2niix).

The brain_colours colour maps were produced and provided by Cyril Pernet (https://doi.org/10.1111/ejn.14430).