Designed and developed to support internal research efforts, provide a viable and robust environment for clinical applications.
The MILXView platform was developed by the Biomedical Imaging team at the Australian e-Health Research Centre (AEHRC) and we acknowledge all staff, past and present, that have contributed to its development.
The MILXView platform was designed and developed to support internal research efforts, provide a viable and robust environment for clinical applications, and to meet the needs of the AEHRC. The AEHRC required a dedicated 3D medical imaging visualization platform which could be easily customized to deliver project-specific clinical applications, and met a diverse set of requirements that ranged from:
- musculoskeletal imaging;
- prostate radiotherapy; to
- small animal imaging.
Each of these areas requires specific requirements and analysis methods.
MILXView has a core framework, which includes viewer components and basic user interface such as slicing and zooming of 2D and 3D images. In addition to the core framework, MILXView has a large number of plug-in components, which can be loaded or unloaded as necessary, that add visualization, image analysis functions and complex image processing pipelines.
MILXView provides a flexible and intuitive interface for manipulation of 3D voxel images. The default view appears in coronal, sagittal, axial and 3D views.
MILXView is a standalone application, written in C++ on Linux and windows operating systems. It comprises:
- A core framework that includes viewer components and basic user interface.
- A large number of core visualization and analysis plugin components that adds visualization tools, user interface panels, basic image analysis functions or other functionalities
- Advanced image processing, segmentation and registration pipelines i.e. co-registration, spatial normalization, partial volume correction, atlas creation, groupwise statistical analysis
MILXView interface displaying a CT scan for a prostate cancer patient with their radiation treatment dose overlaid (darker colours indicate higher dose). A rendered view of the patient’s bony anatomy from the CT scan is shown in the lower right image quadrant.
- Customised plugins for particular applications and anatomical regions i.e. cortical Thickness Estimation, deep grey-matter segmentation, Knee cartilage segmentation, lower-back muscle segmentation, prostate segmentation, breast density estimation and small animal imaging
The framework has been designed this way to enable multiple projects to run simultaneously, developing their own plugins, without interrupting one another. Essentially, plugins are developed as autonomous components, with their own IP. This allows for a large collaborative team of scientists to implement their own algorithms within the same software application, whilst still being able to manage IP. It also enables students to actively participate in the addition of new features.
MILXView features can be separated into two distinct parts: a core framework containing the basic GUI and visualization framework for the application, and add-ons comprising more than 30 plug-in components that add visualization, image analysis functions and complex image processing pipelines.
- Support for Linux(Ubuntu) and Windows distributions from a single source-base
- Fully user-customizable layout and user interface
- System Functions
- Identify the current operating system and user
- Modify the location of data and tmp directories
- View the plugins currently loaded into MILXView
- Modify number of concurrent tasks and threads per task
- Identify task status
- Display log messages from MILXView
- View the MILXView handbook
- Standard Imaging Functions
- Import/ export 2D and 3D data from all standard image formats (DICOM, Nifti, Analyze, bmp, jpg, tiff)
- Histogram inspection
- Metadata inspection using a configurable multi-panel and multi-tab viewer
- 3D multi-slicing screenshot
- Meshing Functions
- Load, save, generate, visualise and manipulate 3D polygon meshes
- Mesh Image Plugin – generate a mesh
- Register Image Plugin – inflate and register a population of cortical hemisphere meshes to an atlas
- Viewing Options Plugin – visualise, import, export and manipulate meshes
- Image Masking – mask the cerebellum from the brain prior to meshing
- Cortical Thickness Estimation (CTE 2.0) Functions
- Segmentation using multiple atlases for the vote approach
- Partial volume estimation
- Topology correction
- Thickness Estimation
- Image Functions
- Segment a T1 weighted MR image, and to measure the thickness of gray matter
- Calculate image and volume properties using a label map and a grey scale image
- Modify the intensity of the image
- Apply a colour map to an image
- Compare two or more images using different methodologies; blended, added, min/max, checkerboard, background, colour
- Flip, rotate, translate and scale an image, using three interpolation modes
- View information about an image
- View the slice of the image at a given position, that maybe modified via the x,y, z axis
- Denoise an image with Non-Local Means
- Create an average representation of a set of N images
- Align and overlay two or more images of the same region of the body
- View images loaded and their properties
- Align and position an image
- Registration Functions
- Image co-registration
- Image registration and automatic labeling
- Image Atlas creation
Example of co-registered CT and MR image of pelvis in MILXView and surface renderings of automatically obtained segmentations
||uses MILXView to develop algorithms and software to process and analyse MRI and PET scans both qualitatively and quantitatively.
||uses MILXView to load, visualise, analyse and compare multi-modality images.
||uses MILXView to automatically calculate of electron density from MR scans (with 98.5% accuracy).
|Small Animal Imaging
||uses MILXView to automatically segment organs from small rodents and estimate the main pharmaco-kinetics parameters of the new imaging compounds by using small animal SPECT, PET and CT systems.
||uses MILXView to manually or automatically estimate of the amount of dense tissue in the breast.