Phenotyping

MITK Phenotyping is a selection of different methods and algorithms that can be used for Radiomics-related research and application building. It is mainly developed within the Division Medical Image Computing at the German Cancer Research Center (DKFZ).

MITK Phenotyping is part of the main MITK Software. Currently, it contains some software part that are not yet integrated in the main release. It is planned to merge both, MITK Phenotyping and MITK Workbench into a single software package.

MITK Phenotyping is also described in this paper:

Michael Götz, Marco Nolden, Klaus Maier-Hein, MITK Phenotyping: An open-source toolchain for image-based personalized medicine with radiomics , Radiotherapy and Oncology 2018

https://doi.org/10.1016/j.radonc.2018.11.021

Features & Highlights

Support for most established medical imaging file formats

• Images: DICOM, NIFTI, NRRD and other
• VTK file formats (for meshed, point sets etc..)

Image preprocessing

• Registration using MatchPoint
• Image Filtering and processing
  • Laplacian of Gaussian (LoG)
  • Wavelet-Transformation
  • Mathematical Operations

Semiautomatic Segmentation

• Multilabel Support
• Several manual and semiautomatic segmentation methods
• Extra Tools for Segmentation Cleaning / Postprocessing

Radiomics Feature Calculation

• Wide range of possible Features
• Tested in combination with the IBSI initiative
• Possibility of encoding feature parameters in feature names

Documentation and Help

The modules and Tools are documented within the MITK documentation. The following list directly lists the most importants parts of this documentation:

• Gemeral introduction into MITK Phenotyping
• Feature overview: All classes for the calculation of radiomics features are derived from AbstractGlobalImageFeature Class, including
  • Image-descriping features
  • Volumetric features
  • First order features
  • Histogram-based first order features
  • Cooccurence matrix based features
  • Run-length matrix based features

Downloads

If you encounter any bugs, please report them in our bugtracking system or use the MITK-users mailing list. We are grateful for any feedback!

Please be aware that MITK Phenotyping is still work in progress and so far not based on a tested stable version of MITK.

Latest installers (2018-10-18)

This installers include the graphical and command line version of MITK Phenotyping. For obtaining scripting support, you currently need to build MITK Phenotyping by yourself with the respective option activated.

Windows 7, Windows 10	MS Windows (64 bit) installer
Windows 7, Windows 10	MS Windows (64 bit) zip archive
Linux x64	Build on Ubuntu 14.04 (64 bit), tar.gz archive

Building MITK Phenotyping from source

1. Install Qt on your system (>= 5.11.1).
2. Clone MITK from out git repository using Git version control.
3. Configure the MITK Superbuild using CMake (>= 3.10).
   1. Choose the source code directory and an empty binary directory.
   2. Click "Configure".
   3. Check the option MITK_USE_SWIG.
   4. Check the option MITK_USE_MatchPoint.
   5. Click "Generate".
4. Build the project
   1. Linux: Open a console window, navigate to the build folder and type "make -j8" (optionally supply the number threads to be used for a parallel build qith -j).
   2. Windows (requires visual studio): Open the MITK Superbuild solution file and build all projects.
5. The build may take some time and should yield the binaries in "your_build_folder/MITK-build/bin"
6. After the build, open the "your_build_folder/MITK-build" folder in Cmake
   2. Check the option BUILD_BasicImageProcessingMiniApps
   3. Check the option BUILD_ClassificationMiniApps
   4. Check the option BUILD_org.mitk.gui.qt.radiomics
   5. Check the option BUILD_org.mitk.gui.qt.multilabelsegmentation
   6. Check the option BUILD_org.mitk.gui.qt.segmentation
   7. Check the options BUILD_org.mitk.gui.qt.matchpoint.* depending on which elements you want in your view.

More detailed build instructions can be found in the documentation.