Download

COLlAGe is available as a prebuilt Slicer extension for Windows only (macOS/Linux in progress): https://github.com/algohary/SlicerExtension_ColiageFilter/releases

Instructions (as of 19 Nov 2018, for Slicer v4.11):

  1. Please download the latest nightly build version (*** Version 4.11.0**) of 3D slicer here: https://download.slicer.org/
  2. Download the latest release ZIP file from GitHub (containing the prebuilt CoLlAGe extension):   https://github.com/algohary/SlicerExtension_ColiageFilter/releases
  3. Extract the contents of 27526-win-amd64-COLIAGE-git8ab9ffa-2018-10-30.zip to any temporary local folder on your computer.
  4. Within this extracted folder:
    • Copy the contents in [^extracted folder]\lib\Slicer-4.11\cli-modules to C:\Program Files\Slicer 4.11.0-2018-11-16\lib\Slicer-4.11\cli-modules
    • Restart 3D Slicer
  5. Find the installed plugin bly clicking Modules -> COLlAGE Texture Radiomics

Note that just trying to load the ZIP file into Slicer does not appear to work consistently, the above steps have been tested by us as an alternative.

The source code is also available, under an Apache 2.0 license: https://github.com/algohary/SlicerExtension_ColiageFilter

Instructions for building COLlAGe from this source code is available on the wiki for this plugin.

A fully compiled Windows command line executable is also available (including sample data): https://cwru.box.com/s/drl2i1c2geffsn4cdb99lqiyv1a7gk6h

Usage for CLI executable:

../CollageFeatureExtractor2D.exe “inImage.mha” “inMask.mha” 2 “outImage.mha”

CLI executable takes the following 4 inputs:

  • Input Arg. 1: Input Greyscale, meta Image, MR Image (.mha format or ITK compatible)
  • Input Arg. 2: Input Binary, meta Image, Tumor label (.mha format or ITK compatible)
  • Input Arg. 3: Neighborhood Radius (integer)
  • Input Arg. 4: Output feature map, meta image path

If you use or publish research based on this implementation of COLlAGe, please cite:

Prasanna, P., Tiwari, P., & Madabhushi, A. (2016). Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe): A new radiomics descriptor. Scientific Reports volume 6, Article number: 37241 (2016).

Version History

Oct 30, 2018

  • V2 “release” with prebuilt Slicer plugin (tested to work with Windows only)
  • Plugin bugfixed to work with Slicer 4.11
  • Instructions for plugin install and code compiling added to GitHub and website

Sep 14, 2018

  • Public release of compiled Windows command-line executable to extract COLlAGe features, including usage information
  • V1 “release” tag of GitHub code

Jul 11, 2018

  • Public release of COLlAGe 3D Slicer plugin code
  • Compatibility: Module tested to completely compile and run within 3D Slicer across Windows and Linux
  • Functionality: Read in an MR image, compute COLlAGe features on a pixel-wise basis, visualize COLlAGe feature, save through Slicer
  • Sample data included

Please provide any feedback you may have via the Contact form or by raising an issue on GitHub.