MITK v2024.06

Updated 72 Days AgoPublic

Actions

Previous changelog • Next changelog

News for MITK Workbench users

We are proud to release MITK v2024.06, the most feature-heavy release in years! In the past 6 months we resolved about 230 tasks and issues in over 660 commits. We are already excited about your feedback on many improvements and features.

Customizable MITK Workbench

You are now optionally greated by a dialog at application start to tailor the MITK Workbench to your specific needs.
Select from a few pre-defined plugin presets or easily define your own to only display the tools and plugins you actually need for your work.

Segmentation

MR image support with TotalSegmentator v2.2

The TotalSegmentator segmentation tool was updated to version 2.2 of TotalSegmentator. Besides the already supported 117 classes for CT images, 56 classes for MR images are now supported with the total_mr task. See the following official image from https://github.com/wasserth/TotalSegmentator for an overview of all new MR image classes:

MedSAM

The new MedSAM AI-based 2-d segmentation tool is similar to the Segment Anything tool, but is explicitly tailored for medical images. Segmenting anatomical structures and pathologies with a few clicks is now easier than ever.

MONAI Label

We are excited to announce the introduction of the MONAI Label tool, developed focusing on inferencing capabilities. MITK now supports both automatic and click-based interaction modes, allowing users to leverage the strengths of AI-driven segmentation in particular using the Radiology app.

Utilities

The Segmentation Utilities plugin supports multiple labels per image now in all of its tools like Boolean Operations, Morphological Operations, Image Masking, Image/Contour/Mesh-to-Segmentation conversion and the completely new Extract from Segmentation tool, which can be used to extract labels into separate binary images.

General improvements

• Labels can be easily identified in an instant now when hovering over them in the Segmentation plugin, since they are displayed highlighted in the image accordingly.
• While we support multiple labels in segmentation interpolation since MITK v2023.04, it is much more robust and reliable now.
• Many segmentation tools can be used more intuitively now and contribute to an overall improved user experience.

Statistics

The Statistics plugin supports multiple labels per image now and is performing faster than ever.

ModelFit and Perfusion

We are excited to announce the integration of ModelFit into the latest MITK release!
This new feature as part of the Workbench provides robust tools for model-fitting including an extensive framework for the specialized application of model-fitting in the area of dynamic contrast-enhanced MRI.

Please have a closer look at the key features here.

We also added the following command-line apps to our official installers:

• PixelDumpMiniApp: Dump (potentially masked) pixel values of passed images into a CSV file.
• Fuse3Dto4DImageMiniApp: Fuse several 3-d images with the same geometry into a dynamic 4-d (3d+t)
• GenericFittingMiniApp: Make a pixel-based fitting on the intensity signal over time for a given model function.

MxN image viewer

To simplify the comparison of images in the MxN view, we have added the option to set a "Data-Based Layout".
This allows users to select a set of images and automatically get a window layout containing all the view directions for each image.
The possible amount of windows is only limited by your imagination (and the size of your screen)!

News for developers

See the past five changelogs starting from 2023 Week 49 (Early December) for a comprehensive, developer-centric overview of changes.

Qt 6

We finally switched from Qt 5 to Qt 6 (minimum v6.6.2).

The migration from Qt 5 to Qt 6 potentially has the biggest impact on your code since the last release. A lot of mainly minor adaptions were necessary on our side. We recommend to frequent the official Qt documentation and porting guide.

MITK-IGT

Just like MITK Diffusion a few years ago, MITK-IGT followed to move into a separate MITK extension. MITK-IGT provides highly valuable and proven image-guided therapy functionality.

Improved integration of MITK extensions into the MITK superbuild

It is now possible to have (more) influence on MITK's third-party dependencies. Individual dependencies of MITK extensions can be optionally processed before MITK's own dependencies, allowing them to be used in MITK's dependencies. For example, if you add OpenCV to your MITK extension dependencies, you are now able to also activate the OpenCV bridge functionality of ITK, since it is configured and built after OpenCV.

MITK extensions also have improved documentation capabilities now, as you can customize Doxygen's EXAMPLE_PATH to provide your own examples.

Command-line apps moved into apps subfolder

For the sake of clarity we moved all command-line apps in our installers and packages from the top level into the apps subfolder. Please adapt your scripts and workflows accordingly.

Command-line argument style on Windows

On Windows, the MITK Workbench and alike exclusively used Windows-style command-line arguments like /BlueBerry.clean in the past. Please use Unix-style command-line arguments like --BlueBerry.clean from now on on all supported platforms.

🛠 Third-party dependency changes

The following table shows a complete list of changed third-party dependencies. The most notable change is the upgrade from Qt 5 to Qt 6 (minimum v6.6.2).

We recommend to use the Qt online installer to install the following required Qt 6 components:

• Qt 5 Compatibility Module
• Qt State Machines
• Qt WebEngine

On Windows, the Qt installer offers a welcome and straight forward way to install OpenSSL v3.x as well.

🔥 Disclaimer for API-breaking changes

We discontinued the extensive listing of API-breaking changes as the vast majority of them are straight forward to resolve or do not affect the majority of developers at all. The ratio between the time and effort spent for writing these reports and actual developer feedback turned out to be greatly imbalanced.

Complex or critical API changes are still accompanied by migration guides in the five changelogs since the last release.

In case you experience any trouble while migrating to the latest version of MITK, please do not hesitate to contact us or to create an issue on our GitHub mirror.

Previous changelog • Next changelog