Skip to main content
VTK (Visualization Toolkit) and ITK (Insight Segmentation and Registration Toolkit) are the core libraries enabling advanced 3D rendering and medical image processing in Miele-LXIV Easy.

​
Overview

VTK - Visualization

Handles 3D rendering, volume visualization, and graphics pipeline management

ITK - Processing

Provides medical image processing algorithms including segmentation and registration

​
VTK (Visualization Toolkit)

​
What is VTK?

VTK is an open-source software system for 3D computer graphics, image processing, and visualization. In medical imaging, it provides:
  • Volume Rendering: Direct visualization of 3D medical image volumes
  • Surface Rendering: Extracting and displaying 3D surfaces from medical data
  • 2D/3D Graphics: Multi-planar reconstruction (MPR) views
  • Interactive Widgets: Tools for measuring, annotating, and manipulating images

​
Version Configuration

VTK_MAJOR=9
VTK_MINOR=0
VTK_BUILD=1
VTK_VERSION=9.0.1

​
Build Configuration

VTK is configured with these important settings:
build.sh:411-430
$CMAKE -G"$GENERATOR" \
    -D CMAKE_INSTALL_PREFIX=$BIN_VTK \
    -D CMAKE_OSX_ARCHITECTURES=$OSX_ARCHITECTURES \
    -D CMAKE_BUILD_TYPE=Release \
    -D CMAKE_OSX_DEPLOYMENT_TARGET=$DEPL_TARG \
    -D BUILD_SHARED_LIBS=OFF \
    -D BUILD_TESTING=OFF \
    -D BUILD_EXAMPLES=OFF \
    -D BUILD_DOCUMENTATION=OFF \
    -D VTK_LEGACY_REMOVE=OFF \
    $VTK_OPTIONS \
    -D VTK_USE_SYSTEM_TIFF=ON \
    -D VTK_USE_SYSTEM_JPEG=ON \
    -D VTK_MODULE_ENABLE_VTK_hdf5=NO \
    -D CMAKE_CXX_FLAGS="$COMPILER_FLAGS" \
    $SRC_VTK

​
Key VTK Options

OptionPurpose
VTK_RENDERING_BACKEND=OpenGL2Use modern OpenGL for hardware acceleration
VTK_USE_SYSTEM_TIFF=ONUse our compiled TIFF library
VTK_USE_SYSTEM_JPEG=ONUse our compiled JPEG library
VTK_LEGACY_REMOVE=OFFKeep legacy APIs for compatibility
BUILD_SHARED_LIBS=OFFBuild static libraries
VTK includes bundled versions of TIFF and JPEG, but we use our own compiled versions to ensure consistency across all libraries.

​
VTK Modules

VTK is modular. Key modules used by Miele-LXIV include:
  • VTK::CommonCore: Basic data structures
  • VTK::RenderingCore: Rendering pipeline
  • VTK::RenderingOpenGL2: OpenGL implementation
  • VTK::RenderingVolume: Volume rendering
  • VTK::InteractionWidgets: Interactive tools
  • VTK::IOImage: Image file I/O

​
Library Collapsing

VTK produces over 100 individual library files. The build system collapses them into a single archive:
build.sh:449-458
if [ $STEP_COLLAPSE_VTK ] ; then
cd $BIN_VTK
VTK_COLLAPSED=lib/libVTK.a
if [ ! -f $VTK_COLLAPSED ] ; then
    echo "=== Collapse VTK into a single library"
    ARGS=$(find lib -name '*.a' -type f)
    libtool -static -v -o $VTK_COLLAPSED $ARGS
fi
fi
This creates libVTK.a, combining all VTK modules into one file for easier linking.
The collapsed library can be quite large (500+ MB) because it includes all rendering, filtering, and I/O modules.

​
ITK (Insight Toolkit)

​
What is ITK?

ITK is an open-source toolkit for medical image analysis. It provides:
  • Image Segmentation: Identifying regions of interest in medical images
  • Image Registration: Aligning images from different time points or modalities
  • Image Filtering: Noise reduction, edge detection, morphological operations
  • Statistical Analysis: Quantitative measurements from images

​
Version Configuration

ITK_MAJOR=5
ITK_MINOR=1
ITK_BUILD=1
ITK_VERSION=5.1.1

​
Build Configuration

ITK depends on VTK and is configured after VTK is built:
build.sh:479-495
$CMAKE -G"$GENERATOR" \
    -D CMAKE_INSTALL_PREFIX=$BIN_ITK \
    -D CMAKE_OSX_ARCHITECTURES=$OSX_ARCHITECTURES \
    -D CMAKE_BUILD_TYPE=Release \
    -D CMAKE_OSX_DEPLOYMENT_TARGET=$DEPL_TARG \
    -D BUILD_SHARED_LIBS=OFF \
    -D BUILD_TESTING=OFF \
    -D BUILD_EXAMPLES=OFF \
    -D Module_ITKOpenJPEG=OFF \
    -D Module_ITKVtkGlue=ON \
    -D VTK_DIR=$BIN_VTK/lib/cmake/vtk-9.0 \
    -D CMAKE_CXX_FLAGS="$COMPILER_FLAGS" \
    $SRC_ITK

​
Key ITK Options

OptionPurpose
Module_ITKVtkGlue=ONEnable ITK-VTK integration
Module_ITKOpenJPEG=OFFUse our OpenJPEG build instead
VTK_DIRPoint to VTK installation for integration
BUILD_SHARED_LIBS=OFFBuild static libraries
The ITKVtkGlue module is critical - it allows ITK to convert its image data structures to VTK formats for visualization.

​
ITK Modules

ITK has a modular architecture with 200+ modules. Key modules include:
  • ITKCommon: Core data structures and algorithms
  • ITKIOImageBase: Image file I/O
  • ITKImageFilterBase: Basic filtering operations
  • ITKSegmentation: Segmentation algorithms
  • ITKRegistration: Image registration methods
  • ITKVtkGlue: VTK integration

​
Library Collapsing

Like VTK, ITK produces many individual libraries:
build.sh:510-519
if [ $STEP_COLLAPSE_ITK ] ; then
cd $BIN_ITK
ITK_COLLAPSED=lib/libITK.a
if [ ! -f $ITK_COLLAPSED ] ; then
    echo "=== Collapse ITK into a single library"
    ARGS=$(find lib -name '*.a' -type f)
    libtool -static -v -o $ITK_COLLAPSED $ARGS
fi
fi
This creates libITK.a, combining all ITK modules.

​
VTK-ITK Integration

VTK and ITK work together in Miele-LXIV:

​
Workflow Example

  1. ITK loads a DICOM series into an itk::Image
  2. ITK processes the image (e.g., applies filters, performs segmentation)
  3. ITKVtkGlue converts the ITK image to a vtkImageData
  4. VTK renders the image using volume rendering or surface extraction
  5. User interacts with the visualization through VTK widgets

​
Build Time Considerations

Build Time: VTK and ITK are large projects that take significant time to compile:
  • VTK: 30-60 minutes on modern hardware
  • ITK: 45-90 minutes on modern hardware
Plan accordingly and ensure you have sufficient disk space (10+ GB for build artifacts).

​
Parallel Building

The build script uses all available CPU cores:
build.sh:118
MAKE_FLAGS="-j $(sysctl -n hw.ncpu)"
This significantly reduces build time on multi-core systems.

​
Dependencies

Both VTK and ITK depend on image format libraries:

​
Configuration in Kconfig

Kconfig-miele:41-48
config DOWNLOAD_SOURCES_VTK
    bool "VTK"
    default y

config DOWNLOAD_SOURCES_ITK
    bool "ITK"
    default y
Kconfig-miele:243-259
config COMPILE_VTK
    bool "Build VTK"
    default y

config COLLAPSE_VTK
    bool "Single VTK library"
    default y
    depends on COMPILE_VTK

config BUILD_ITK
    bool "Build ITK"
    default y

config COLLAPSE_ITK
    bool "Single ITK library"
    default y
    depends on BUILD_ITK

​
Common Use Cases in Medical Imaging

​
Volume Rendering (VTK)

vtkSmartPointer<vtkVolume> volume = vtkSmartPointer<vtkVolume>::New();
vtkSmartPointer<vtkVolumeRayCastMapper> mapper = 
    vtkSmartPointer<vtkVolumeRayCastMapper>::New();
mapper->SetInputData(imageData);
volume->SetMapper(mapper);

​
Image Filtering (ITK)

using FilterType = itk::MedianImageFilter<ImageType, ImageType>;
FilterType::Pointer filter = FilterType::New();
filter->SetInput(inputImage);
filter->SetRadius(2);
filter->Update();

​
ITK to VTK Conversion

using ConverterType = itk::ImageToVTKImageFilter<ITKImageType>;
ConverterType::Pointer converter = ConverterType::New();
converter->SetInput(itkImage);
converter->Update();
vtkImageData* vtkImage = converter->GetOutput();

​
Further Reading

VTK Documentation

Official VTK guides and API reference

ITK Software Guide

Comprehensive ITK textbook (PDF)

VTK Examples

Hundreds of VTK code examples

ITK Examples

ITK example code and tutorials

Build docs developers (and LLMs) love

Get started for freeTalk to us