MITK v2026.06

📢 News for MITK Workbench users

We are proud to announce the release of MITK v2026.06. Thank you to everyone who contributed!

Over the past six months, we resolved around 120 tasks across approximately 1,000 commits. 🚀

The complete changelog can be found at the end of these release notes. As always, we'll start by highlighting some of the most notable improvements and new features in this release.

🚀 nnInteractive v2.5

We integrated the latest version of the groundbreaking nnInteractive segmentation tool and are already prepared for upcoming model checkpoints as they become available. Compared to nnInteractive v1, the new version delivers improved speed, higher segmentation quality, and introduces long-awaited support for remote sessions.

Multiple MITK instances can now connect to a shared GPU server, enabling lightning-fast interactive 3D segmentation even on lightweight laptops and Macs.

We've also significantly optimized MITK's integration of nnInteractive to minimize memory consumption and maximize performance. Installing or updating nnInteractive is now just a single click, and a range of new configuration options let you tailor nnInteractive to your segmentation workflow.

The scribble interaction of nnInteraction is used to quickly segment several organs and vertebra.

🌈 TotalSegmentator v2.14

We completely redesigned our TotalSegmentator integration to make it easier than ever to set up and use. Install it with a single click, choose a segmentation task, and start it with another click. Once the results are ready, simply review and confirm them.

If you have a TotalSegmentator license, you can also enter your license key to unlock additional restricted segmentation tasks.

Installing and using TotalSegmentator in MITK is just a few clicks.

🕶️ Improved live 3D visualization of segmentations

In MITK v2025.12, we introduced live 3D visualization of segmentation labels. For MITK v2026.06, we completely redesigned the rendering pipeline by replacing volume rendering with blazing-fast polygonal mesh generation.

The result is dramatically improved visual quality, significantly higher rendering performance, and a much more responsive interactive experience.

📷 Screenshot Maker Overhaul

We simplified and modernized the Screenshot Maker to make creating high-resolution screenshots easier than ever. Choose a custom background color (or use transparency) and capture publication-ready images without any enforced logo watermark.

Drag the right screenshot over another image to see that the transparency of the liver is preserved.

📈 Standardized PET SUV quantification

The PET SUV computation view is now part of the official release, together with a companion command-line app for batch and headless processing. It supports multiple SUV normalization strategies and automatic acquisition-time deduction with manual override down to the slice level.

To support standardized and reproducible quantification, MITK's SUV computation was evaluated against the IBSI SUV reference, using its SUV computation
reference as the benchmark (@ibsi thank you for this great set of DROs🙏).

🍏💄 Improved graphical user interface on macOS

The MITK Workbench has now a much better overall appearance on macOS, matching our Windows and Linux ports. It also finally supports the native fullscreen mode.

🍏⚠️ Running the MITK Workbench app on macOS

In particular since macOS 15 Sequoia, macOS imposes strict restrictions on executing the MITK Workbench application, as we are not a registered Apple developer. You may need to manually remove the MITK Workbench from quarantine to run it, e.g., allow the application to start in the security settings or run the following command on the MitkWorkbench.app file (adapt its file path as needed):

xattr -d com.apple.quarantine /Applications/MitkWorkbench.app

Note that the very first launch of the MITK Workbench on macOS can take up to 2 minutes to complete all security checks.

📢 News for developers

Recent advances in coding LLMs have significantly accelerated development, enabling us to deliver new features and fixes faster than ever. While AI has become an important productivity tool, we continue to apply the same engineering standards as before, with thorough code reviews, testing, and long-term maintainability taking precedence over pure vibe coding.

This accelerated pace of development also means that this release includes substantial changes under the hood, including several API-breaking changes. We believe that, in the age of AI-assisted development, adapting existing code to such changes is easier than ever. The resulting improvements in code quality, consistency, and modernization outweigh the temporary cost of resolving these incompatibilities.

To make migration as straightforward as possible, we have ensured that our commit messages document the rationale and details behind these changes. They provide valuable context not only for developers but also for coding LLMs, making them an excellent resource when updating existing code to the new APIs.

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

🐍 Python bindings: mitk-python package on PyPI available now!

The mitk-python package exposes the data structures and core utilities of MITK to Python. It ships as a self-contained wheel that bundles the compiled MITK runtime together with its native dependencies like ITK, VTK, CppMicroServices, and DCMTK, so you can install it with pip install mitk-python and start working with medical images without building MITK from source.

Note that the wheels on PyPI are currently available only for Python 3.12.
The documentation is hosted on Read the Docs.

import mitk
import numpy as np

img = mitk.Image(np.zeros((64, 64, 64), dtype=np.float32), spacing=(1.0, 1.0, 2.5))
img.save("output.nrrd")

loaded = mitk.Image("output.nrrd")
print(loaded.shape, loaded.spacing, loaded.origin)

🕹️ Remote control the MITK Workbench from Python

MITK v2026.06 introduces a REST API that lets external programs drive a running Workbench, e.g., with the official mitk-workbench-remote companion Python client. Install it with pip install mitk-workbench-remote, connect to a live session, and script it end to end without writing a single line of C++:

• Push data into the data storage
• Navigate scenes
• Control the Standard or MxN Displays
• Drive the crosshair and per-window cameras
• Capture screenshots

import mitk_workbench_remote as mw

wb = mw.connect("http://localhost:8080")
wb.show(image, name="case_001")   # push data into the running Workbench
png = wb.mxn.screenshot()         # capture the MxN multi-widget editor

A full example suite goes from a first connect and show up to a cohort screenshot factory that turns the MITK Workbench into a scriptable visualization service:

1. Applies an MxN layout
2. Loops over a cohort and its segmentation labels
3. Points the crosshair and cameras at each lesion
4. Captures screenshots
5. Assembles a markdown QA report

The kind of batch job that used to mean clicking through every case by hand. Combined with the layout-as-code DSL from the mitk-python package (mitk.mxn.layout), entire review layouts can be defined in Python and reused across cases.

🛠 Third-party dependency changes

Dependency	Before (v2025.12)	Now (v2026.06)
ACVD	83a0bd0b (2025-03-27)	3ca0b532 (2025-11-14)
Boost	v1.85.0 (patched)	v1.91.0
CMake (minimum)	v3.22	v3.28 (v3.31 on Windows)
cpp-httplib	v0.15.3	v0.27.0
CTK	adab52fb (2024-11-26)	db30f292 (2026-03-15)
dcmqi	v1.3.2	145b7b49 (2026-06-01)
DCMTK	v3.6.7	2dd54ca1 (2026-05-08)
GDCM	v3.0.14	v3.2.2
HDF5	v1.14.3	Use ITK's HDF5 instead
ITK	v5.4.3 (patched)	v5.4.6 (patched)
lz4	v1.9.4	v1.10.0
MatchPoint	c6188c95 (2025-04-10)	d02f5fbe (2026-06-11)
nlohmann_json	v3.11.3	v3.12.0
nnInteractive	[v1.1.2, v2.0.0)	[v2.5.0, v3.0.0)
oneTBB	v2022.1.0	v2022.3.0
OpenSSL	v3.0	v3.5...

MITK v2025.12.2

This is a hotfix release for MITK v2025.12, addressing the following issues:

1. nnInteractive failed on segmentations that did not contain a label with pixel value 1.
2. The DICOM Seg writer threw an exception when exporting a simple label without segment type modifier.

Full Changelog: v2025.12...v2025.12.2

MITK v2025.12

News for MITK Workbench users

We are proud to release MITK v2025.12. In the past 4 months we resolved about 50 tasks and issues in 140 commits. Thank you to everyone who contributed!

You can find the complete changelog at the end of these release notes.

🚀 Improved nnInteractive integration

This is the second release of MITK with the cutting-edge nnInteractive segmentation tool, with an even better integration into the MITK Workbench.

nnInteractive enables high-quality 3D segmentations from just a few 2D interactions such as points, boxes, scribbles, or contours - bringing a new level of interactivity and reliability to medical image segmentation.

✨ Key features

• Designed for medical imaging: Tailored for modalities like MRI, CT, and microscopy
• Multi-modal prompts: Supports points, scribbles, bounding boxes, and lassos
• 3D augmentation from 2D input: Generate volumetric results from minimal input
• Multi-plane interaction: Work across axial, sagittal, and coronal views
• 3D visualization: Inspect and validate results with instant full 3D rendering
• 4D frame support: Apply nnInteractive directly to segment in frames of dynamic image data

This tool has the potential to redefine your segmentation workflow — try it out directly in the MITK Workbench!

Here's a preview of how to segment a whole kidney in a second with a single click:

🕶️ Live 3D visualization of segmentation labels

Previously, labels were displayed only in the 2D views. In MITK 2025.12, we introduced live 3D volume visualization of labels, including label previews of segmentation tools. This feature can be toggled for each segmentation via the Data Manager’s right-click context menu or configured globally in the Segmentation preferences:

Of course you still can convert labels to smooth polygon models:

👆 New label selection tool

We added the Selection 2D segmentation tool, which lets you hover over the pixels of a label to either select it with a single click or to view meta information for all labels currently under the mouse cursor:

💄UX improvements of the Segmentation plugin

🧹 Streamlined plugin UI

We observed that new users can feel overwhelmed by the full feature set of the Segmentation plugin, especially on smaller screens where scroll bars may hide parts of the interface. Therefore, in the simple case where an image is present but no segmentation exists yet, the Segmentation plugin UI is streamlined to only the essential elements:

🏷️ Enhanced label suggestion system

We've significantly expanded the label suggestion capabilities to better support standardized medical terminology:

Multiple built-in vocabularies: You can now choose from multiple built-in suggestion lists, including the newly added EUCAIM v2 vocabulary. EUCAIM (European Cancer Image Alliance, https://cancerimage.eu/) provides standardized terminology specifically designed for cancer imaging research.

Rich label metadata: Suggestions now can include additional information beyond just the label name and color. All built-in vocabularies provide metadata such as primary anatomic structure and anatomic region. This metadata is automatically stored with your segmentation and preserved when exporting to formats like DICOM SEG or MITK MultiLabel Segmentation files, ensuring better interoperability and traceability of your annotations.

🔄 No more hidden "reinit"

In certain scenarios, such as when working with multiple or tilted images, the native pixel or voxel grid of the segmentation may not match the currently displayed pixel grid. Because proper alignment is essential for precise interaction with the segmentation tools, these tools are disabled when a mismatch occurs. Previously, a warning message instructed users to “reinit” the segmentation image - a function that is somewhat hidden in the Data Manager’s right-click context menu. We now provide a direct option within the Segmentation plugin to align the views with a single click:

🐧⚠️ Dropped support of Ubuntu 20.04

We have dropped support for Ubuntu 20.04. This version reached its end of standard support (EOL) in May 2025 and no longer meets our minimum requirements, such as providing OpenSSL v3 by default. Installers continue to be available for Ubuntu 22.04 and later.

🍏⚠️ Running the MITK Workbench app on macOS

In particular since macOS 15 Sequoia, macOS imposes strict restrictions on executing the MITK Workbench application, as we are not a registered Apple developer. You may need to manually remove the MITK Workbench from quarantine to run it:

xattr -d com.apple.quarantine /Applications/MITKWorkbench.app

Note that the very first launch of the MITK Workbench on macOS can take up to 2 minutes to complete all security checks.

News for developers

This release also introduces important modernizations under the hood, improving long-term maintainability and developer experience.

🛠 Third-party dependency changes

We didn't introduce new dependencies or updated existing dependencies this time.

However, we removed support of OpenSSL v1. This basically also ends our support of Ubuntu 20.04, which officially reached End of Standard Support (EOL) in May 2025.

On Windows, we ensure now that the minor version of OpenSSL v3.x used by the MITK superbuild matches the OpenSSL version embedded in our Python Standalone Builds dependency. For MITK v2025.12, this is OpenSSL v3.0.

Heads-up: We will update pretty much all of our third-party dependencies shortly after this release.

🔥 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.

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

Changelog since MITK v2025.08

nnInteractive

• Fixed PyTorch version to 2.8, as 2.9 is not compatible and/or massively degrades performance.
• Added advanced settings for the installation to...
  • change pip package names, versions, and index URL.
  • ignore the local pip cache resp. enforce downloads of pip packages.
• Added Settings button right next to Initialize button to quickly access preferences.
• Added preferences to...
  • change the model checkpoint.
  • change the backend (Auto, CPU, GPU including CUDA devive number).
• Added console output about found/used CUDA device including compute capability and total available VRAM.
• Improved error reporting during initialization.
• Disabled point interaction in 3D render window.
• nnInteractive can now be used on macOS when a FreeType MacPorts package is used while creating an MITK installer.
• Suppress symlink warnings of Hugging Face on Windows.

Segmentation

• Added immediate 3D visualisation of labels.
• Added label selection and identification tool.
• Improved user guide (F1 context help) of the Segmentation view.
• Improved UX to create new segmentations when no segmentations are available.
• Improved UX for errors that prevent the use of segmentation tools.
• Improved the label naming/renaming dialog.
• Allow multiple internal and external label suggestion configs.
• Extended label suggestions with label meta data and the support of maximum instance constraints.
• Label suggestions support the new MITK Stacked Segmentation Format (JSON) now.
• Introduce the EUCAIM v2 vocabulary as build-in suggestions option.
• Label presets support the new MITK Stacked Segmentation Format (JSON) now.
• Fixed label selection when renaming labels.
• Fixed label search.
• Fixed a potential exception when deleting a label.
• Fixed potential exceptions with undo/redo and label presets.
• Fixed deferred rendering update when selecting labels.
• Fixed alphabetical order of default label names with leading zeros.
• Fixed performance issues when using the "go to label" feature (double click on label instance in segmentation view)
• Improved file and directory structure of the MitkMultilabel module.
• Changed

Workbench

• Fixed reinit for segmentations.
• Fixed crash at application shutdown (introduced after MITK v2024.08).

Third party dependencies

• Made MITK compatible to Qt 6.10.
• Made MITK compatible to macOS 26 Tahoe.
• Removed support of OpenSSL v1.
• On Windows, OpenSSL v3.0 is required (matches minor OpenSSL version of Python Standalone Builds dependency).

Build system

• Improved switch between build configurations with and without Python.

Other

• Improved default installer/package names with accurate OS names and versions.
• Completely rewrote build instructions.

MITK v2025.08

News for MITK Workbench users

We are proud to release MITK v2025.08. In the past 8 months we resolved about 130 tasks and issues in over 760 commits. Thank you to everyone who contributed!

You can find the complete changelog at the end of these release notes.

🚀 Introducing nnInteractive

This release marks a major milestone for MITK: nnInteractive, our brand-new AI segmentation tool, is now fully integrated into the MITK Workbench.

nnInteractive enables high-quality 3D segmentations from just a few 2D interactions such as points, boxes, scribbles, or contours — bringing a new level of interactivity and reliability to medical image segmentation.

✨ Key features

• Designed for medical imaging: Tailored for modalities like MRI, CT, and microscopy
• Multi-modal prompts: Supports points, scribbles, bounding boxes, and lassos
• 3D augmentation from 2D input: Generate volumetric results from minimal input
• Multi-plane interaction: Work across axial, sagittal, and coronal views
• 3D visualization: Inspect and validate results with full 3D rendering

This tool has the potential to redefine your segmentation workflow — try it out directly in the MITK Workbench!

News for developers

This release also introduces important modernizations under the hood, improving long-term maintainability and developer experience.

🛠 Third-party dependency changes

• Python Standalone Builds are now bundled with MITK. This removes the dependency on external Python installations and enables consistent Python-based workflows out of the box.
• PCRE was removed (no longer required for SWIG on Linux, where it’s available as a system package).
• TBB was added as a new dependency, serving as backend for ITK and VTK. We are also evaluating a full migration from OpenMP to TBB.

Multiple third-party dependencies were updated for performance, compatibility, and security improvements:

Dependency	Old version	New version
ACVD	18209c90 (2023-01-23)	83a0bd0b (2025-03-27)
ITK	v5.4.2	v5.4.3 (patched)
MatchPoint	56957430 (2024-06-04)	c6188c95 (2025-04-10)
nlohmann_json	v3.11.2	v3.11.3
PCRE	v8.45
POCO	v1.12.4 (patched)	v1.14.1
Python3 (python-build-standalone)		v3.12.11 (2025-07-12)
SWIG	v4.0.2	v4.3.0
TBB		v2022.1.0
VTK	v9.3.0 (patched)	40ef44cb (2025-04-09)

🔥 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.

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

Changelog since MITK v2024.12

Segmentation

• Added nnInteractive tool.
• Added live search for labels in the "Create Label" dialog.
• Added context menu entry to label overview to copy labels to other groups.
• Added support for TotalSegmentator v2.9 and licensed tasks.
• Accelerated internal preprocessing of results from TotalSegmentator.
• Improved UX for model downloads in TotalSegmentator.
• Fixed missing comma in list of TotalSegmentator tasks (reported by @chocomilk4759).
• Refactored LabelSetImage into MultiLabelSegmentation.
• Fixed undo/redo operations.
• Added support for NIfTI as file format for MITK multi-label segmentations.
• Added support for loading stacks of label masks with custom meta info as MITK multi-label segmentations.
• Added support of SAM2 to MONAI tool.
• Improved installation UX for SAM and MedSAM tools.
• Fixed unnecessary re-initialization after model change in MedSAM tool.
• Add and Subtract modes of the Live Wire and Lasso tools are now shown by default.
• Fixed static segmentations for single-slice 4-d images.

Workbench

• Improved image navigation performance.
• Fixed crash on application start when using Qt 6.6.
• Fixed crash on repeated "Close Project".
• Fixed Open/Save file dialogs on macOS Sequoia.
• Added arrow up and down hotkeys to slice through images.
• Added --MITK.fullscreen argument to start in fullscreen mode.

MxN Display

• The crosshair of the Standard Display is no longer shown in the MxN Display.
• Added group synchronization.
• Clicking on a segmentation label now jumps to its location in all render windows.
• The selection mode is now saved in layout files as well.
• Newly loaded nodes are now automatically re-selected if they were selected before.
• Navigational changes are now only shared between Standard Display and MxN Display when synchronization is toggled on.

Pharmacokinetics

• Made Dynamic DCE MR Perfusion/PET DataFit plugins compatible with multi-label segmentations.
• Constraints are now activated by default in MRI/PET perfusion fitting.
• Parameter maps are now shown with the transparent Jet LUT by default.
• Model error messages are now shown as overlay instead of using a modal dialog.
• Added black & white LUT for PET images.

Command-line apps

• Added MitkConvertToMultiLabelSegmentation app to convert images and contours into multi-label segmentations.
• Added support for image masks in MatchImage registration app.

Third party dependencies

• Upgraded ITK from v5.4.0 to v5.4.2 to make MITK compatible with the latest version of Visual Studio 2022 (contributed by @TheRisenPhoenix).
• Patched ITK to remove factory-based creation for a few small but very often instantiated classes.
• Upgraded VTK (and ACVD) from v9.3.0 (patched) to commit 40ef44cb from 2025-04-09.
• Upgraded POCO from v1.12.4 (patched) to v1.14.1.
• Upgraded nlohmann_json from v3.11.2 to v3.11.3.
• Added dependency to TBB which is now used as backend in ITK and VTK.
• Removed PCRE from SWIG dependencies (use system libs instead on Linux).
• Upgraded MatchPoint to the latest commit for compatibility with Apple Clang 17.
• Added Python Standalone Builds as our single and insolated dependency for everything Python in MITK.

Other

• Fixed many typos in documentation of code (contributed by @luzpaz).

2025 Week 26 (Very Late June)

Important

This is an unstable und unsupported pre-release.
Download the latest official release MITK v2024.12 instead.

Segmentation

• Refactored LabelSetImage into MultiLabelSegmentation
• Added nnInteractive tool (Linux and Windows only)
• Fixed undo/redo operations
• Added support for NIfTI as file format for MITK multi-label segmentations
• Added support for loading stacks of label masks with custom meta info as MITK multi-label segmentations
• Added support for TotalSegmentator v2.9 and licensed tasks
• Improved UX for model downloads in TotalSegmentator
• Added support of SAM2 to MONAI tool
• Improved installation UX for SAM and MedSAM tools
• Fixed unnecessary re-initialization after model change in MedSAM tool
• Add and Subtract modes of the Live Wire and Lasso tools are now shown by default

Workbench

• Improved image navigation performance
• Fixed Open/Save file dialogs on macOS Sequoia
• Added arrow up and down hotkeys to slice through images
• Added --MITK.fullscreen argument to start in fullscreen mode

MxN Display

• The crosshair of the Standard Display is no longer shown in the MxN Display
• Added group synchronization
• Clicking on a segmentation label now jumps to its location in all render windows
• The selection mode is now saved in layout files as well
• Newly loaded nodes are now automatically re-selected if they were selected before
• Navigational changes are now only shared between Standard Display and MxN Display when synchronization is toggled on

Pharmacokinetics

• Made Dynamic DCE MR Perfusion/PET DataFit plugins compatible with multi-label segmentations
• Constraints are now activated by default in MRI/PET perfusion fitting
• Parameter maps are now shown with the transparent Jet LUT by default
• Model error messages are now shown as overlay instead of using a modal dialog
• Added black & white LUT for PET imaages

Command-line apps

• Added MitkConvertToMultiLabelSegmentation app to convert images and contours into multi-label segmentations

Other

• Added a Undo/Redo Inspector plugin for debugging the undo/redo stack
• Fixed setting non-boolean values for new properties in Properties plugin
• MITK_SHOW_CONSOLE_WINDOW does no longer affect non-GUI apps
• Prefixed all symbols of the customized Minizip version in CppMicroServices with us_ or US_.
• Allow loading of DICOM images with invalid dates before the year 1400
• Made MITK compatible with Apple silicon (ARM64)
• Fixed a crash during mapping in a few occasions
• Fixed tutorial step 8
• Reduced number of (caught) exceptions at application startup and shutdown

Third party dependencies

• Upgraded VTK (and ACVD) from v9.3.0 (patched) to commit 40ef44cb from 2025-04-09
• Patched ITK to remove factory-based creation for a few small but very often instantiated classes
• Upgraded POCO from v1.12.4 (patched) to v1.14.1
• Upgraded nlohmann_json from v3.11.2 to v3.11.3
• Added dependency to TBB which is now used as backend in ITK and VTK
• Removed PCRE from SWIG dependencies (use system libs instead on Linux)
• Upgraded MatchPoint to the latest commit for compatibility with Apple Clang 17

2025 Week 06 (Very Early February)

Important

This is an unstable und unsupported pre-release.
Download the latest official release MITK v2024.12 instead.

Segmentation

• Added live search for labels in the "Create Label" dialog.
• Added context menu entry to label overview to copy labels to other groups.
• Accelerated internal preprocessing of results from TotalSegmentator.
• Fixed missing comma in list of TotalSegmentator tasks (reported by @chocomilk4759).

Workbench

• Fixed crash on application start when using Qt 6.6.

Third party dependencies

• Upgraded ITK from v5.4.0 to v5.4.2 to make MITK compatible with the latest version of Visual Studio 2022 (contributed by @TheRisenPhoenix).

Other

• Fixed many typos in documentation of code (contributed by @luzpaz).

MITK v2024.12

News for MITK Workbench users

We are proud to release MITK v2024.12. In the past 6 months we resolved about 200 tasks and issues in over 300 commits. While our winter releases tend to be manily bugfix releases, we take the opportunity to highlight several new features and quality of life improvements in these release notes. You find the complete changelog at the end of this release notes.

Segmentation

Rename label groups

Labels can be organized in groups (used to be named "layers" in the past). In fact it is the only way to create overlapping labels, since labels within a group cannot share the same pixels. It is now possible to rename groups, which otherwise have generic default names like "Group 1".

Decrease label opacity

User feedback showed that opinions on a good opacity of segmentations vary greatly. If you prefer to work with less label opacity to still clearly see the labeled pixels below, you can now dial down the "Opacity factor" in the Segmentation preference page.

Create smoothed polygon models (fast)

Hidden in the context menu of the Data Manager you find the option to create smoothed polygon models for your pixel-based segmentations. It took a while to show results when using this feature for segmentations with many labels like the results of TotalSegmentator. Well, no more! The models are now generated in parallel for each label. In that regard, you are now also able to change the opacity of multiple data nodes at once, making the postprocessing of TotalSegmentator and alike for display blazingly fast.

DICOM Browser

The DICOM Browser allows users to locate and load DICOM images organized in a hierarchy of patients, studies, and series. It consists of three main components:

• Local Storage: Quickly access previously imported or retrieved DICOM images.
• Import: Scan directories to identify and manage DICOM images.
• Query/Retrieve: Connect to a PACS server to query and retrieve images.

The DICOM Browser is part of MITK since many years, but was significantly overhauled for MITK v2024.12. Give it a try to explore and import your DICOM data.

New command-line app to split 4-dimensional images

The new MitkSplit4Dto3DImages app allows you to easily split a 3d+t (4-d) image into its time steps as separate 3-d images.

News for developers

The following sections primarily address developers to summarize important or backward-incompatible changes since the last MITK release.

Migration from Phabricator to GitLab and GitHub

Since our last release, we migrated from Phabricator to GitLab for internal repository hosting and issue management. We still mirror MITK to our official GitHub page and in fact started to utilize more features of GitHub again like
issues for external reporters (including templates) and detailed release notes as well as changelogs on the releases page. In particular, our snapshot releases are now also listed on that page as pre-releases including downloads for the installers.

Reduced build times

We were able to greatly reduce the build time of MITK in particular on Windows with the combination of precompiled headers and using the Multi-ToolTask scheduler of Visual Studio.

🛠 Third-party dependency changes

The following table shows a complete list of changed third-party dependencies.

Dependency	Old version	New version
C++ REST SDK	v2.10.19 (patched)
CTK	37aff992 (patched, 2023-12-15)	adab52fb (patched, 2024-11-26)

🔥 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.

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

Changelog since MITK v2024.06.2

Segmentation

• Added timeout preference for MONAILabel segmentation tool.
• Allow custom label group names.
• Assume an empty label if a segmentation has a single label group but no label meta data.
• Fixed unhandled exception on label collisions.
• Fixed transformations of 3-d renderings of segmentations.
• Fixed visibility of 3-d renderings of segmentations.
• Fixed minimum widths of several spin boxes.
• Added option to preferences to reduce label opacity.
• Significantly reduced the execution time of "Create (smoothed) polygon model" actions when applied to multiple labels.
• Fixed a crash when changing preferences while using the Segment Anything (SAM) or MedSAM tools.
• Updated cursor icons to indicate whether tools are operating in inverted mode.
• Changed the Segment Anything (SAM) tool's input shape from points to crosshairs.
• Fixed exclusion of the highest pixel value of an image in the UL-Threshold tool.
• Fixed missing initial preview of GrowCut tool in "transfer selected labels" mode.
• Fixed error messages when using undo/redo together with manual 2-d segmentation tools.
• Fixed an unhandled exception when clicking on the active label while using the Close tool.
• Fixed excessive warning messages of SegmentAnything, MedSAM, and MONAI Label tools.
• Fixed duplicate default names for new segmentations.
• Fixed layout warnings of several tools.
• TotalSegmentator
  • Upgraded TotalSegmentator from v2.2.1 to v2.4 with support for new subtasks:
    • pleural_pericard_effusion
    • head_glands_cavities
    • head_muscles
    • headneck_bones_vessels
    • headneck_muscles
    • liver_vessels
  • Fixed incorrect label mapping of total task when other subtasks were used before.
  • Moved the TotalSegmentator installation step from the tool GUI to preferences.
  • Fixed unintended addition of invalid Python paths in the preference page.
  • Improved the visual appearance of the installation process.
• Segmentation Task Lists
  • Integrated new MITK Forms feature into Segmentation Task Lists v2 (experimental).
  • Added support for MITK scene files as task inputs in MITK Segmentation Task List files (version 3).
  • Fixed possible crash in Segmentation plugin when switching tasks in a Segmentation Task List while a segmentation tool is active.

Statistics and measurement

• Fixed stuck interaction of polygon-like planar figures in certain edge cases.
• Fixed a crash on the removal of a planar figure while interacting with multiple planar figures.
• Fixed a Qt 6.8 crash on application exit when the Statistics plugin was opened.

Volume Visualization

• Fixed unintentional creation of transfer function files when saving is canceled.
• Selected presets are now shown right in their drop-down list.
• Enlarged control point hitboxes of the transfer function widgets in the Volume Visualization plugin.

Image Cropper

• Fixed geometries of cropped 3d+t images.
• Fixed the unintentional removal of node names when using the override mode.

Migration from Phabricator to GitLab/GitHub

• Moved MITK-Data Git repository mirror to Helmholtz GitLab.
• Migrated Phabricator links to GitHub links.
• Created GitHub issue templates.
• Added contribution guide and code of conduct.
• Updated macOS CI status badges in README.md.

MITK Workbench

• Improved font quality in MITK Workbench, changed default font to Roboto and replaced xkcd font with Atkinson Hyperlegible.
• Fixed layout glitches of the memory indicator in the status bar.

Command-line apps

• Removed MiniApp suffix from Pharmacokinetics cmd-line apps.
• Fixed time step extraction edge case in Split4Dto3DImages cmd-line app.
• Fixed unhandled exception on unsupported data in FileConverter cmd-line app.
• Added version information to the command-line output.
• Added Split4Dto3DImages command-line app.

DICOM and image I/O

• Fixed loading of dynamic DICOM images when the timing is only encoded in the trigger time tag, but the frames also differ in the encoded acq time/date.
• Improved support of pure 2-d DICOM images.
• Fixed an issue with loading DICOM-SEG images where segments lacked labeled pixels.
• Corrected loading of DICOM images sorted by acquisition date.
• Relaxed precision-related geometry checks in the NIFTI image reader.
• Fixed several issues of the DICOM browser.
• Removed the StoreSCP feature of the DICOM browser.
• Intercept multiple loading of the same DICOM images.
• Fixed saving of DICOM parametric maps in pharmacokinetics for which sufficient metadata is available.
• The version of MultiLabelSegmentation files was increased to 2 because of the named groups feature.
• Fixed a few file readers regarding input directories with early versions of the std::filesystem library (e.g. CentOS 7).
• Expose reason why a DICOM block has been split.
• Fixed handling of multiple DICOM files of the same series without image position or orientation (e.g. screenshots).
• Fixed loading of DICOM directories when files have no extension.
• Fixed crash when a dynamic images have "empty" time steps consisting of NaN pixels only.

Pharmacokinetics

• Changed default value of the signal-to-concentration conversion factor from 0 to 1.
• Introduced a plugin for standardized uptake value (SUV) calculations...

2024 Week 50 (Early December)

Important

This is an unstable und unsupported pre-release.
Download the latest official release MITK v2024.06.2 instead.

Segmentation

• Fixed exclusion of the highest pixel value of an image in the UL-Threshold tool.
• Fixed missing initial preview of GrowCut tool in "transfer selected labels" mode.
• Fixed error messages when using undo/redo together with manual 2-d segmentation tools.
• Fixed an unhandled exception when clicking on the active label while using the Close tool.
• Fixed excessive warning messages of SegmentAnything, MedSAM, and MONAI Label tools.
• Fixed duplicate default names for new segmentations.
• Fixed layout warnings of several tools.
• TotalSegmentator
  • Fixed incorrect label mapping of total task when other subtasks were used before.
  • Fixed unintended addition of invalid Python paths in the preference page.
  • Improved the visual appearance of the installation process.

Statistics and measurement

• Fixed stuck interaction of polygon-like planar figures in certain edge cases.
• Fixed a crash on the removal of a planar figure while interacting with multiple planar figures.
• Fixed a Qt 6.8 crash on application exit when the Statistics plugin was opened.

Volume Visualization

• Fixed unintentional creation of transfer function files when saving is canceled.
• Selected presets are now shown right in their drop-down list.
• Enlarged control point hitboxes of the transfer function widgets in the Volume Visualization plugin.

Image Cropper

• Fixed geometries of cropped 3d+t images.
• Fixed the unintentional removal of node names when using the override mode.

DICOM and image I/O

• Fixed several issues of the DICOM browser.
• Removed the StoreSCP feature of the DICOM browser.
• Intercept multiple loading of the same DICOM images.
• Fixed saving of DICOM parametric maps in pharmacokinetics for which sufficient metadata is available.
• The version of MultiLabelSegmentation files was increased to 2 because of the named groups feature.
• Fixed a few file readers regarding input directories with early versions of the std::filesystem library (e.g. CentOS 7).

Pharmacokinetics

• Fixed inactive "Start modelling" button in Curve Descriptive Parameters and Dynamic PET plugins.
• DCE MR Perfusion Data Fit plugin
  • Added FAQs to the context help page.
  • Fixed bogus "Could not open AIF file!" message.
  • Display parameter units for constraints.
  • Fixed node filter for images as start parameters.

Build system

• Reduced build times of MITK using precompiled headers (MITK_PCH advanced CMake cache variable).
• Updated build instructions and prerequisites for MITK with Qt 6.8.

Other improvements

• Allow larger settings for thick-slice image rendering (MIP).
• Added rename action to Data Manager context menu for data nodes.
• Fixed unnecessarily truncated image names in the Data menus of the MxN Display.
• Fixed constricted spin boxes on Windows.
• Reset of render window editor preferences are now applied immediately for all related settings.
• Fixed Qt includes that were not compliant with the Qt API.

2024 Week 47 (Mid November)

Important

This is an unstable und unsupported pre-release.
Download the latest official release MITK v2024.06.2 instead.

Segmentation

• Added option to preferences to reduce label opacity.
• Significantly reduced the execution time of "Create (smoothed) polygon model" actions when applied to multiple labels.
• Fixed a crash when changing preferences while using the Segment Anything (SAM) or MedSAM tools.
• Moved the TotalSegmentator installation step from the tool GUI to preferences.
• Updated cursor icons to indicate whether tools are operating in inverted mode.
• Changed the Segment Anything (SAM) tool's input shape from points to crosshairs.
• Added support for MITK scene files as task inputs in MITK Segmentation Task List files (version 3).

Pharmacokinetics

• Introduced a plugin for standardized uptake value (SUV) calculations in PET-CT images.
• Display parameter units in the DCE MR Perfusion Data Fit and ModelFit Inspector plugins.
• Made the info box in the Curve Descriptive Parameters plugin read-only.

DICOM and image I/O

• Fixed an issue with loading DICOM-SEG images where segments lacked labeled pixels.
• Corrected loading of DICOM images sorted by acquisition date.
• Relaxed precision-related geometry checks in the NIFTI image reader.

Command-line apps

• Added version information to the command-line output.

Build system

• Reduced Windows build times by 33–50% using the Multi-ToolTask scheduler in Visual Studio.

Other improvements

• Added coordinate and time step details to the Pixel Value plugin.
• Labeled the time slider in the Image Navigator plugin as "Time step".
• Relaxed precision-related geometry checks in the Statistics plugin.
• Made the "Constrained zooming and panning" preference effective immediately without requiring a restart.
• Improved the user guide for the Point Set Interaction plugin.
• Added support for the Qt command-line option -style.

2024 Week 43 (Very Late October)

Important

This is an unstable und unsupported pre-release.
Download the latest official release MITK v2024.06.2 instead.

Segmentation

• Integrated new MITK Forms feature into Segmentation Task Lists v2 (experimental)
• Added timeout preference for MONAILabel segmentation tool
• Allow custom label group names
• Assume an empty label if a segmentation has a single label group but no label meta data
• Fixed unhandled exception on label collisions
• Fixed transformations of 3-d renderings of segmentations
• Fixed visibility of 3-d renderings of segmentations
• Fixed minimum widths of several spin boxes
• Upgraded TotalSegmentator from v2.2.1 to v2.4 with support for new subtasks:
  • pleural_pericard_effusion
  • head_glands_cavities
  • head_muscles
  • headneck_bones_vessels
  • headneck_muscles
  • liver_vessels

Migration from Phabricator to GitLab/GitHub

• Moved MITK-Data Git repository mirror to Helmholtz GitLab
• Migrated Phabricator links to GitHub links
• Created GitHub issue templates
• Added contribution guide and code of conduct
• Updated macOS CI status badges in README.md

MITK Workbench

• Improved font quality in MITK Workbench, changed default font to Roboto and replaced xkcd font with Atkinson Hyperlegible
• Fixed layout glitches of the memory indicator in the status bar

Command-line apps

• Removed MiniApp suffix from Pharmacokinetics cmd-line apps
• Fixed time step extraction edge case in Split4Dto3DImages cmd-line app
• Fixed unhandled exception on unsupported data in FileConverter cmd-line app

DICOM

• Fixed loading of dynamic DICOM images when the timing is only encoded in the trigger time tag, but the frames also differ in the encoded acq time/date.
• Improved support of pure 2-d DICOM images

Pharmacokinetics

• Changed default value of the signal-to-concentration conversion factor from 0 to 1

Build system

• Fixed auto-detection of MITK_OPENSSL_SSL_DLL and MITK_OPENSSL_CRYPTO_DLL on Windows with older versions of CMake
• Fixed FlowBenchSegmentationRelease configuration preset