Medical Image Analysis
Alzheimer's Disease
Prostate RadiotherapyIntroduction
MILX View is a 3D medical imaging analysis and visualization platform developed by the Biomedical Imaging team at the Australian e-Health Research Centre (AEHRC). It was designed and developed to support internal research efforts, and provide a viable and robust environment for clinical applications.
MILX View comprises of a core framework (viewer components and basic user interface) and a large number of plug-in components that add visualization, image analysis functions and complex image processing pipelines.
MILX View provides a flexible and intuitive interface for manipulation of 3D voxel images
MILX View 3.0
Each release of MILX View improves on the previous version. MILX View 3.0 is the latest release of the MILX View platform and provides researchers and clinicians with a rich image analysis toolkit.
Features
- Support for a variety of Linux distributions from a single source-base (currently developed under Ubuntu)
- Standard medical imaging functions: windowing, histogram inspection, panning, slicing, zooming, metadata inspection using a configurable multi-panel and multi-tab viewer, and Import and export 2D and 3D data from all standard image formats (DICOM, Nifti, Analyze, bmp, jpg, tiff).
- Fully user-customizable layout and user interface.
- 3D visualization tools: maximum intensity projection (MIP) views, volume and surface rendering, 3D multi-slicing screenshot, color map, overlay, blending, checkerboard. Loading, saving, visualization, registration and manipulation of 3D polygon meshes.
MILX View has fully featured support for meshing and visualisation of scalar mesh data
- Data manipulation in voxel or scanner space: manual rotation scaling, zooming and translation. Navigation through multiple datasets simultaneously.
- Manual linking and registration of two or more volumetric images.
- Plugin architecture for new algorithms developed by the BioMedIA Lab including Cortical Thickness Estimation(CTE) and Non-Rigid Registration (NRR).
- Cortical Thickness Estimation (CTE 2.0) plug-in has had its user interface modified to allow for segmentation using multiple atlases for the vote approach, partial volume estimation, Topology correction, and Thickness Estimation.
Cortical Thickness Estimation (CTE) plugin results within MILX View 3.0
- A number of new "Analysis" plug-ins have been added to MILX View 3.0 including Partial Volume Correction, Region of Interest, and a statistics plug-in to allow for design and calculation of statistics stored in a MySQL database.
- A new "Registration" menu has been created to cater for the new plug-ins: Image co-registration, Image registration and automatic labeling and image Atlas creation.
- Improved and increased the multi-threaded architecture of the platform including a "preferences" plug-in where the total number of threads and the maximum number of threads per task can be altered, and A task manager plug-in showing the status of any running process, their queued, start and running times and their current progress.
- MILX View 3.0 is now packaged with an online handbook that can be viewed from the MILX View 3.0 application or from a standard browser.
Software Design
MILX View is built upon the MILX libraries, which are a repository of CSIRO image processing algorithms in C++. The libraries provide an interface to three main third party libraries;
- The Insight Segmentation and Registration Toolkit (ITK), which provides extensive medical image analysis functionalities.
- The Visualization Toolkit (VTK), which supplies comprehensive rendering and visualization libraries for volumetric and mesh-structured images.
- The WxWidgets cross-platform user interface library.
MILX View is configurable to build standalone applications that can be shared with collaborators.
Source Control
All source code and related resources are maintained in a subversion (SVN) repository. The SVN repository can be configured to allow external access by collaborators. Under this scheme, collaborators can checkout files from the MilxView repository, and link against existing software libraries thus allowing them to create new plug-ins that can be incorporated in later versions of MILX View.
Continuous Integration
The MILX View code is built everynight if new code has been checked in that day. Unit tests and valgrind are automatically run on the code to verify the code is operating correctly and there are no outstanding memory management and threading bugs. The unit tests also provide detailed analysis of the code coverage of the system. DOXYGEN documentation is also automatically generated as a result of the build together with a source code style checker (KWStyle). Dashboard (CDash) is used to display code metrics and the results of continuous integration tests.
The MILX View Dashboard (CDash) displays code statistics and the results of continuous integration testing
