DiffusionImaging

MITK Diffusion

Quicklinks

• Sourcecode: Wiki-Page with instructions and web interface of the Git repository
• Download Section is here

---

DI_App|Icon|align="right"

The MITK Diffusion application offers a selection of image analysis algorithms for the processing of diffusion-weighted MR images. It encompasses the research of the Neuroimaging Group in the Division Medical and Biological Informatics at the German Cancer Research Center (dkfz).

#!wiki caution
'''Features & Highlights'''

 * Tensor and q-ball reconstruction
 * Glyph visualization
 * Quantification and partial volume clustering of tensor and q-ball images
 * Global fiber tractography, visualization, and tract post-processing
 * Brain network statistics and visualization (connectomics)
 * Interactive exploration of Tract-based spatial statistics (TBSS) results
 * Intra-voxel incoherent motion (IVIM) estimation
 * Synthetic data generation

The MITK Diffusion application is based on the MITK research platform and the most of it is open-source. The available code is embedded into the source code of MITK as a module and can be accessed through the public git repository.

Version

The current stable version is the MITK Diffusion 2013.09.04 release.

Changelog

• Redesign of the segmentation plugin
• Based on MITK 2013.09 Release
  • Update of many third party libraries
  • support for Visual Studio 2012
• Improvement of usability and workflow of several perspectives
  • New motion correction view
  • Tensor and Q-ball reconstruction will now show a preview when thresholding based on the b0 image
  • Diffusion miniApps have been bundled in one binary
• The 2013.09.04 bugfix includes important fixes improving the head motion correction stability and corrects an error in the MD calculation

User manual

Please refer to the user manual of MITK Diffusion 2013.03 or use the context help pages in the application.

Media

Publication: MITK diffusion imaging, In Methods Inf Med 2012 Oct 11;51(5):441-8<
> Authors: Fritzsche KH¹, Neher PF¹, Reicht I¹, van Bruggen T¹, Goch C¹, Reisert M², Nolden M¹, Zelzer S¹, Meinzer HP¹, Stieltjes B¹<
> Institution: ¹German Cancer Research Center, Heidelberg, Germany<
>²University Hospital Freiburg, Department of Radiology, Medical Physics, Freiburg, Germany

Publication: MITK-DI - A new diffusion imaging component for MITK. In Deserno TM, Handels H, Meinzer HP, Tolxdorff T (eds). Bildverarbeitung für die Medizin 2010. Heidelberg: Springer (2010) 246-250 <
> Authors: Fritzsche KH¹, Meinzer H-P¹<
> Institution: ¹German Cancer Research Center

Downloads

Version 2013.09 ( Release date: 10-10-2013 )

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2010: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7
• Ubuntu 12.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.8 (Mountain Lion)

The OS X 10.6 installer is compatible with OS X 10.7 (Lion) so there is no dedicated disk image build under 10.7.

Available installers

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 12.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image	10.8 OS X disk image

Installers for the 2013.09.00 version can be found here

Testing data

A set of data for testing the functionality of the mitkDiffusion application is available here.

Contact

If you have questions about the application or if you would like to give us feedback, don't hesitate to contact us.

---

MITK Diffusion - Release Archive

This section collects links and information for all released versions of the MITK Diffusion Application, but for the newest.

Available Versions	Release date
MITK Diffusion 2013.06	04 Jul 2013
MITK Diffusion 2013.03	04 Apr 2013
MITK Diffusion 2012.12	21 Dec 2012
MITK Diffusion 2012.09	02 Oct 2012
MITK Diffusion 2012	07 May 2012
MITK Diffusion 2011	02 Nov 2011

Release 2013.06

Release date: 04-07-2013

Changelog

• Based on the MITK 2013.06 Release
• Following the ITK support change in MITK, also MITK Diffusion now fully supports ITK4 ( the ITK3 support is discontinued )
• New Features
  • new mini-applications for fiber direction extraction and evaluation
• Over thirty bugs were fixed across the different modules and plugins

User manual

Please refer to the user manual of MITK Diffusion 2013.03 or use the context help pages in the application.

Downloads

Version 2013.06 ( Release date: 04-07-2013 )

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7, Windows Vista
• Ubuntu 12.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.8 (Mountain Lion)

The OS X 10.6 installer is compatible with OS X 10.7 (Lion) so there is no dedicated disk image build under 10.7.

Available installer

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 12.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image	10.8 OS X disk image

Release 2013.03

Release date: 04-04-2013

• Based on the MITK 2013.03 Release

Changelog

• New features in FiberFox
  • Fiber Definition
    • Added functions to copy, translate, rotate and scale fibers including the corresponding fiducials
    • Fiducial alignment with voxel grid
    • Fiber definition with only two ROIs
• Signal Generation
  • Extended diffusion models: Stick, Zeppelin, Tensor, Ball, Astrosticks, Dot
  • Extended two-compartment model to four-compartment model
  • Realistic simulation of relaxation effects. Parameters: T2 (per compartment), TE, T2' and line readout time
  • Explicit definition of fiber radius used for the calculation of compartment volume fractions
  • Changed the relative parameter SNR to the absolute parameter noise variance
  • Removed internal signal normalization --> double voxel size in one dimension results in twice the signal strength
• Complete refactoring of the Fiberfox help page
• Fiber Processing View reodered and splitted into separate views: Fiber Processing and Fiber Extraction
• Modularization of the Diffusion Module into
  • new submodules DiffusionCore, FiberTracking, Connectomics, Quantification
• New, user defined visualisation properties for connectomics networks added
• Several bugs fixed in both registration views

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7, Windows Vista
• Ubuntu 12.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.8 (Mountain Lion)

The OS X 10.6 installer is compatible with OS X 10.7 (Lion) so there is no dedicated disk image build under 10.7.

Available installer

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 11.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image	10.8 OS X disk image

Documentation

The user manual for the MITK Diffusion 2013.03 application can be found here.

Release 2012.12

Release date: 21-12-2012

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7, Windows Vista
• Ubuntu 11.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.7 (Lion)

Available installer

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 11.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image	10.7 OS X disk image (.dmg)	10.8 OS X disk image (.dmg)

Release 2012.09

Release date : 02 Oct 2012

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7, Windows Vista
• Ubuntu 11.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.7 (Lion)

Available installer

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 11.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image	10.7 OS X disk image (.dmg)

Release 2012

Release date : 07 May 2012

Changelog

A list of the major changes from the 2011 to the 2012 release of MITK Diffusion.

2>|Connectomics	Added global clustering computation
Added generator for synthetic networks
4>|Fiber Processing	much more information about the selected fiber bundle is displayed
extraction of fibers passing through a 3D ROI (mitkSurface, STL mesh) is possible
fiber smoothing using spline interpolation added
fiber coloring by scalar map added
3>|Gibbs Tracking	tracking results can be saved automatically
major bug that caused only few fibers to be reconstructed was fixed
input of gfa image is not needed anymore
Stochastic Tracking	gui completely refactored
3>|mitkFiberBundleX	mitkFiberBundle is replaced by mitkFiberBundleX which is essentially a vtkPolyData which results in a much better performance and easier handling
fiber bundle are saved as vtkPolyData
old xml based fiber files can still be loaded and are automatically converted into the new mitkFiberBundleX
3>|ODF Details	refactored gui
tensor data is displayed
normalization of the ODFs is applied correctly
Partial volume analysis	Export of clustering result as image is now possible
2>|Preprocessing	the number of gradients of the selected image can be reduced while preserving an equal distribution of the gradients (as far as possible)
the gradient directions can be displayed as pointset
Q-Ball Reconstruction	removed erroneous numerical reconstruction method
Segmentation	fast interactive 3D interpolation added

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the minimal OS version changed to Ubuntu 11.04 because the compiler version available in 10.04 does not provide features needed by MITK
  • the Qt framework, version 4.6.2 or later

Tested systems

• Windows 7, Windows Vista
• Ubuntu 11.04 and newer
• OS X 10.6 (Snow Leopard), OS X 10.7 (Lion)

Documentation

You can access the MITK Diffusion online documentation here

Available installer

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 11.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	OS X 10.6 and newer	10.6 OS X disk image (.dmg)	10.7 OS X disk image (.dmg)

Release 2011

Release date : 02 Nov 2011

Requirements

Additional system specific requirements:

• Windows
  • If you have problems running the Windows application, please install the Microsoft Redistributable Packages for VS 2008: 32 bit or 64 bit
• Linux
  • the Qt framework, version 4.6.2 or later
  • the tar.gz archive needs to be extracted into a folder whithout whitespaces in the path

Documentation

You can access the MITK Diffusion 2011 online documentation here

You can download the MITK Diffusion information brochure here (PDF)

Available Installer

The installer is available for platforms enlisted in the table. Support for other systems is only experimental.

Windows	Vista, Windows 7	MS Windows (32 bit) installer	MS Windows (64 bit) installer
Linux	Ubuntu 10.04 and newer	Ubuntu (32 bit), tar.gz archive	Ubuntu (64 bit), tar.gz archive
Mac OS X	10.6 Snow Leopard and newer	OS X disk image (.dmg)