SimpleITK: a layer built on top of the Insight Toolkit (ITK), intended to simplify and facilitate ITK's use in rapid prototyping, education and interpreted languages.
Tagged on 2/2/2022 we are happy to announce SimpleITK 2.2 Release Candidate 2!
Users and developers are encouraged to test the RC before the final release and report issues, bugs, and any compatibility problems.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install --pre SimpleITK --find-links https://github.com/SimpleITK/SimpleITK/releases/tag/v2.2rc2
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simpleitk/label/dev simpleitk
Require SimpleITK_INT64_PIXELIDS for 64-bit builds. The Python ReadTransform method now returns a downcasted transform.
Update to ITK v5.3rc3 Improved Python packaging type information (PEP 561). Improved types for manual Python procedural methods: Resample, GetArrayFromImage, GetImageFromArray, GetArrayViewFromImage, ReadImage, WriteImage, SmoothingRecursiveGaussian, DiscreteGaussian.
Additional baseline test image for AArch64/ARM64 systems. Fix test failure when system has more than 127 processors. Remove failing test with 32-bit pixel and JPEG file.
Announcing the SimpleITK 2.2 Release Candidate 1!
Users and developers are encouraged to test the RC before the final release and report issues, bugs, and any compatibility problems.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install --pre SimpleITK --find-links https://github.com/SimpleITK/SimpleITK/releases/tag/v2.2rc1
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simpleitk/label/dev simpleitk
Included in this release candidate are "manylinux2014_aarch64" wheels to provide initial support for the ARMv8-A (aarch64) on Linux.
SimpleITK 2.1.1 has been released!
The patch release Updates the ITK version to 5.2.1.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
Python binary wheels are available for download from PyPI. It is important to have the latest version of pip for correct wheel compatibility and installation. To install the latest SimpleITK package:
python -m pip install --upgrade pip
python -m pip install SimpleITK
SimpleITK packages are available for conda Python package manager as a monolithic package:
conda install -c simpleitk simpleitk
or integrated into the coda-forge package dependencies:
conda install simpleitk --channel conda-forge
SimpleITK 2.1.0 has been released!
The release includes new features, API changes, documentation updates and bug fixes. This is the first SimpleITK release to provide binaries for the new Apple M1 ARM.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
Python binary wheels are available for download from PyPI. It is important to have the latest version of pip for correct wheel compatibility and installation. To install the latest SimpleITK package:
python -m pip install --upgrade pip
python -m pip install SimpleITK
Wheels with the platform tag "macxos-11_0_arm64" are available from PyPI, to provide initial support for the Apple M1 Silicon for Python 3.8 and 3.9.
SimpleITK packages are available for conda Python package manager as a monolithic package:
conda install -c simpleitk simpleitk
or integrated into the coda-forge package dependencies:
conda install simpleitk --channel conda-forge
Image::getBufferAsByteBuffer to Java interface, returns a type specific instance of java.nio.Buffer.GetBufferAsNativePointer to Image class.Image::GetSizeOfPixelComponent method.N4BiasFieldCorrectionImageFilter::GetLogBiasFieldAsImage method.Image::GetBufferAs... methods, in wrapped languages.ComposeScaleSkewVersor3DTransform class.Image class.Image's meta-data dictionary. This includes implementations for __delitem__, __contains__, and __getitem__ with a string parameter.PermuteImageFilter.LoggerBase and ITKLogger classes to interface and override ITK's output messages.ShiftScaleImageFilter, to support arbitrary type conversion with intensity scaling.sitkUInt8 mask image support to MaskNegatedImageFilter
ImageRegistrationMethod::StopRegistration methodStopRegistration method to Demons based registration filtersHistogramMatchingImageFilter to use named "referenceImage" input for second input image.GetPixelAsComplexFloat64 Image method in Python is correctly renamed to __GetPixelAsComplexFloat64__.BinaryMagnitudeImageFilter test, changed type from UInt32 to Int32 for defined C++ behavior, fixed Mac ARM testing failure.ImageViewer::SetApplication docs.* Minor grammar and spelling changes to gettingStartedAnnouncing the SimpleITK 2.1 Release Candidate 2!
Users and developers are encouraged to test the RC before the final release and report issues, bugs, and any compatibility problems.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install --pre SimpleITK --find-links https://github.com/SimpleITK/SimpleITK/releases/tag/v2.1rc2
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simpleitk/label/dev simpleitk
ImageRegistrationMethod::StopRegistration methodStopRegistration method to Demons based registration filtersHistogramMatchingImageFilter to use named "referenceImage" input for second input image.BinaryMagnitudeImageFilter test, changed type from UInt32 to Int32 for defined C++ behavior, fixed Mac ARM testing failure.Announcing the SimpleITK 2.1 Release Candidate 1!
Users and developers are encouraged to test the RC before the final release and report issues, bugs, and any compatibility problems.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install --pre SimpleITK --find-links https://github.com/SimpleITK/SimpleITK/releases/tag/v2.1rc1
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simpleitk/label/dev simpleitk
SimpleITK v2.1rc1 has been successfully compiled and tested on Apple OS X running native M1 ARM. Thanks to the ITK development team for providing the foundation.
For the v2.1 .0 final release, we are currently only planning to release M1 ARM binaries for the conda-forge distribution.
Image::getBufferAsByteBuffer to Java interface, returns a type specific instance of java.nio.Buffer.GetBufferAsNativePointer to Image class.Image::GetSizeOfPixelComponent method.N4BiasFieldCorrectionImageFilter::GetLogBiasFieldAsImage method.Image::GetBufferAs... methods, in wrapped languages.ComposeScaleSkewVersor3DTransform class.Image class.Image's meta-data dictionary. This includes implementations for __delitem__, __contains__, and __getitem__ with a string parameter.PermuteImageFilter.LoggerBase and ITKLogger classes to interface and override ITK's output messages.ShiftScaleImageFilter, to support arbitrary type conversion with intensity scaling.sitkUInt8 mask image support to MaskNegatedImageFilter
GetPixelAsComplexFloat64 Image method in Python is correctly renamed to __GetPixelAsComplexFloat64__.ImageViewer::SetApplication docs.SimpleITK 2.0.2 has been released! The release provides minor patches for the 2.0 release to address bugs.
Additionally Python 3.9 is supported and "manylinux2010" wheels are provided. This addresses interoperability with other "manylinux2010" wheels reported in some environments [#1252].
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install SimpleITK
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simple simpleitk
To learn more about SimpleITK go to simpleitk.org.
Image::GetBufferAsVoid, Image::GetBufferAsUInt64 and Image::GetBufferAsInt64 methods.SimpleITK 2.0.1 has been released! The release provides minor patches for the 2.0 release to address bugs.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install SimpleITK
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simple simpleitk
To learn more about SimpleITK go to simpleitk.org.
tensorflow (#1208, #1192)SimpleITK 2.0 has been released! This major version update to SimpleITK includes updating to ITK version 5, significant code improvements, performance enhancements, and API changes.
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install SimpleITK
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simple simpleitk
To learn more about SimpleITK go to simpleitk.org.
Image support for improved support of time series, and multi-channel images commonly used in microscopy.
ExtractImageFilter, JoinSeriesImageFilter, PasteImageFilter, HashImageFilter, ImageFileReader, ImageFileWriter, ImageSeriesWriter.Image and Transform classes. Please note this is designed for operations such as interprocess communications; the standard ITK IO formats should be used as portable storage formats.Show procedure calls the ImageViewer class interface which utilizes the class's defaults and configuration.The environment variables related to the Show procedure are only checked once. The Show method now is a wrapper for the ImageViewer class, and the behavior changed. The SITK_SHOW_EXTENSION and SITK_SHOW_COMMAND environment variables if defined initialize the corresponding global defaults only the first time accessed. The environment variables SITK_COLOR_COMMAND and SITK_SHOW_3D_COMMAND are removed as the ImageViewer class allows for direct configuration of the command.
Remove from all filters the Execute method with filter parameters as arguments. The Execute method with the input images remains.
Remove AddTrialPoints and ClearTrialPoints methods from FastMarchingImageFilter due to consolidation of code generation templates.
With Java wrapped Vector types whose C++ element type is an unsigned integer, the wrapped element type changed to the next larger Java signed integer type ( see 618d89 ).
Removed procedural methods taking scalar radius from the following ( see 444f9a ):
Dropped Python 3.4 support due to end of life.
Remove the sitk::ImageFilter template parameter because it is no longer utilized.
The Python installation command using setup.py changed with the location of setup.py. To install see updated Sphinx documentation.
Remove Transform::AddTransform and Transform::FlattenTransform methods. Use the CompositeTransform class.
Rewrote Show to use ImageViewer class.
For ImageViewer add user's Application folder to search path.
Use NPasteImageFilter from the SimpleITKFilters remote ITK module to replace PasteImageFilter.
Add to SimpleITK's PasteImageFilter:
DestinationSkipAxes parameterAdd method to ImageFileReader::GetImageIOFromFileName get the ITK ImageIO automatically used for file.
Update superbuild Swig version to 4.0.2. SWIG >= 4 is now required.
Update ITK version to 5.1.1.
Disable ITK 4 legacy behaviors.
Add SignedDanielssonDistanceMapImageFilter::GetVoronoiMap method.
Add wrapping for CannySegmentationLevelSetImageFilter.
Update levelset segmentation filters to support named inputs.
Add 4D support toSliceImageFilter and to the Python slice indexing.
Add complex pixel support to Image::GetBufferAs... methods.
Add wrapping for BinaryPruningImageFilter.
Add per label measurements to LabelOverlapMeasuresImageFilter.
ResampleImageFilter add option to use nearest neighbor extrapolation.
Add ReturnBinMidPoint parameter to OtsuThresholdImageFilter. Defaults to false which may change results.
Add support to specify compression level and compression algorithm to ImageFileWriter and ImageSeriesWriter classes.
BSplineTransform Python support construction and SetCoefficientImage with list-like series of Images.
MeanImageFilter directly filters vector images, improving performance.
Add to LabelShapeStatisticsImageFilter per label method GetIndexes and GetRLEIndexes.
Add Image::IsUnique method.
Add wrapping for DICOMOrientImageFilter.
Add complex pixel type support to GetBufferAs methods.
Improve checking and error messages when Image inputs require the dimensions, sizes or pixel types to match.
Add LevelSetMotionRegistrationImageFilter::InitialDisplacementField method.
Refactor of the template code generation system to enable numerous new features and C++ style enhancements.
Add boolean in_place field to all JSON filter descriptions.
Add C++ automatic "in place" ITK filter execution with r-value image input.
Add C++ Image operators with automatic in-place filter execution with r-value Image classes.
Use native SWIG/R enumeration support.
Add named filter inputs to morphological reconstruction image filters.
Replace the CMake configuration option of "SimpleITK_4D_IMAGE" with "SimpleITK_MAX_DIMENSION"
Default configuration now include 5D Image support.
The following filters support up to "SimpleITK_MAX_DIMENSION":
Refactored MemberFunctionFactory's hashing function.
Add ExtractImageFilter support to extract an abitrary dimension e.g. 3D image from 5D image.
Add sitkScaleVersor enum.
Add Transform::GetTransfromEnum method.
Add CompositeTransform class.
Update to ITKv5 threading model behaviors for the following ProcessObject methods:
SetGlobalDefaultThreader and GetGlobalDefaultThreader
SetGlobalDefaultNumberOfThreads and GetGlobalDefaultNumberOfThreads
SetNumberOfWorkUnits and GetNumberOfWorkUnits
Python
DiscreteGaussian, SmoothingRecursiveGaussian, ReadImage and WriteImage to support keyword arguments.WriteImage Python procedure the keyword only arguments imageIO and compressor.__setitem__ for [] operator support:
GetArrayFromImage, GetImageFromArray and GetArrayViewFromImage methods.Image class, ensuring the copy is unique with lazy copying implementation.Image and Transform classes.Resample procedure.Image class's in-place operators to use C++ in-place operators.__setitem__ support for assigning sub-regions via the PasteImageFilter.Transform.Downcast method.CSharp
GetConstBufferAs....Fix the ImageRegistrationMethod exception to use GetPixelIDTypeAsString
Update ndarray.tostring to tobyte ( tostring deprecated in numpy 1.19.0 )
Add missing deepcopy implementation for Image with RLE "Label" pixel types.
Fixed Python pickling for missed transform classes.
Refactored GetImageFromVector method to retain reference to original container with a command object. Fixing image buffer aliasing with BSplineTransform and DisplacementFieldTransform.
Make TranslationTransform conversion constructor explicit.
Remove implicit conversion from Image to Transform class.
Update StatisticsImageFilter input convention for ITKv5.
Remove WarpImageFilter matching image size requirement.
The results of the LiThresholdImageFilter changes with ITK, baseline test results were updated. ( See ITK commit e3ce37 for details. )
Add additional baseline images for registration results, due to change in ImageRegistration's smoothing algorithm. ( See ITK commit 569a47 for details. )
Correct LabelShapeStatistics baseline for OrientedBoundingBoxVertices results. ( See ITK commit 50c695 for details. )
Fix potential double memory free of pixel container in results from internal image to vector image conversions.
Fix RelabelComponentImageFilter incorrect object size computation with sorting enabled. ( See ITK commit 162101 for details. )
Fix potential ITK pipeline execution error when a filter execute on a vector image by per component.
Add C++11 move semantics support to the Image class.
Add direct support for C++11 lambda command to the ProcessObject class.
Add GetBufferAsVoid method to the Image class.
Add Decay parameter to the MirrorPadImageFilter.
Update doc strings for Python and JAVA from JSON filter descriptions.
Update examples to use slice based assignment instead of PasteImageFilter.
Update the tutorials page.
Update FAQ.
Improve WriteImage Doxygen documentation.
Update RTD Show procedure documentation.
Add custom HTML Doxygen header with dynamic timestamp insertion, and privacy policy.
Add procedural example to IO page.
Updating the main Doxygen page.
Update Sphinx filter list.
Add example for reading raw image files.
Add DICOM series writing example with floating point pixels.
Add FAQ entry for the differences between ITK and SimpleITK.
Rename Sphinx "installation" page to "getting started".
Update examples from using scalar radius to vector.
Use Doxygen interactive SVG output, and enable client side MathJax for Latex rendering.
Add C++ CMake example for using SimpleITK.
Update examples to use CompositeTransform.
Update copyright to NumFOCUS.
Add casting to N4BiasFieldCorrection example.
Typo fixes.
Create Docker images for generating Doxygen.
Add C# ConnectedThresholdSegmentation example.
Remove references to next branch in documentation.
Update R installation instructions.
Add C# ImageReistrationMethod2 example.
Add multi-lingual examples for FastMarchingSegmentationImageFilter.
Uset stopping time parameter in FastMarchingSegmentation examples.
Add C# CannySegmentationLevelSetImageFilter example.
Improve documentation to GetImageFromArray about isVector parameter.
Update referenced tutorials.
Add acknowledgment section to readme.
Update Java and CSharp installation instructions.
Move Doxygen pages to Sphinx documents, remove other out dated pages.
Move Sphinx documentation to docs directory.
Address signed to unsigned comparison compiler warnings.
Fix PYTHON_EXECUTABLE being undefined in CMake for Doxygen.
Add Doxygen MathJAX URL as CMake option.
Continue to address compilation warnings.
Pin Sphinx version for Read-The-Docs.
Remove EXPORT specification for static member function declaration.
Support pass TBB_DIR to ITK in the Superbuild.
Require C++11 standard for compilation.
Replace C++ tr1 usage with standard C++11 classes.
Replace compiler depended defines with C++11 keywords.
Prefer using C++11 lambda over std::bind when std::placeholders are not needed.
Improve support on OSX for isysroot flag and CMAKE_OSX_SYSROOT variable.
Support additional CMAKE_GENERATOR_*, CMAKE_VS_PLATFORM_TOOLSET_* variables is superbuild.
Added AWS S3 buckets for data mirroring.
Use SHA512 hash files as index for downloaded source code.
Use CMake for creating zip archives.
Improve finding of Lua interpreter.
Update Lua superbuild version to 5.3.5, require Lua version 5.2 or 5.3 for code generation.
Update PCRE superbuild version to 8.43.
Add USE_CCACHE cmake option to automatically use ccache as launcher if available.
Enable GNU gold linker by default if available.
Update GTest vesrion to 1.10.0 in superbuild.
Fix incorrect version of virtualenv detected.
ImageViewer class and Show procedure.ImageRead and ImageWrite procedures.nullptr over NULL.Announcing the SimpleITK 2.0 Release Candidate 3!
This is anticipated to be the last release candidate before 2.0 final. Users and developers are encouraged to test the RC before the final release and report issues, bugs, and any significant compatibility problems.
Additional release notes can be found in prior candidates:
Complete instructions on getting started with SimpleITK including downloading binaries or building SimpleITK can be found on the SimpleITK Read the Docs web page.
python -m pip install --upgrade pip
python -m pip install --pre SimpleITK --find-links https://github.com/SimpleITK/SimpleITK/releases/tag/v2.0rc3
Conda packages are available from Anaconda Cloud on the SimpleITK channel. These can be installed with:
conda install -c simpleitk/label/dev simpleitk
Show method are only checked once. The Show method now is a wrapper for the ImageViewer class, and the behavior changed. The SITK_SHOW_EXTENSION and SITK_SHOW_COMMAND environment variables if defined initialize the corresponding global defaults only the first time accessed. The environment variables SITK_COLOR_COMMAND and SITK_SHOW_3D_COMMAND are removed as the ImageViewer class allows for direct configuration of the command.Show to use ImageViewer class.ImageView application search path for Mac.DiscreteGaussian, SmoothingRecursiveGaussian, ReadImage and WriteImage to support keyword arguments.WriteImage Python procedure the keyword only arguments imageIO and compressor.NPasteImageFilter from the SimpleITKFilters remote ITK module to replace PasteImageFilter.PasteImageFilter:
DestinationSkipAxes parameter__setitem__ for [] operator support:
ImageFileReader::GetImageIOFromFileName get the ITK ImageIO automatically used for file.ImageRegistrationMethod exception to use GetPixelIDTypeAsString
ndarray.tostring to tobyte ( tostring deprecated in numpy 1.19.0 )PasteImageFilter.WriteImage Doxygen documentation.Show procedure documentation.PYTHON_EXECUTABLE being undefined in CMake for Doxygen.ImageViewer class and Show procedure.ImageRead and ImageWrite procedures.