< < |
Stradview is a user-friendly medical 3D data visualisation package which can run under Windows, and implements many of the algorithms described in the research page. It was originally developed from Stradwin, but without any of the live ultrasound acquisition features, which makes it much more appropriate for those who only need to visualise and analyse existing medical data.
For more information, and to obtain the software, see the Stradview page.
Stradwin was originally developed for freehand 3D ultrasound, but also incorporates many other techniques for visualising and analysing medical data. It has been developed with the aim of encouraging clinical research in this area. Those who are only interested in visualisation and analysis of data should use Stradview instead.
This is available free - see the stradwin documentation page for much more information about availability and technical content.
wxDicom was originally written as a simple tool to demonstrate various image interpolation and filtering techniques in a teaching setting. It has since been expanded to allow it to load medical data from various sources (including DICOM) in addition to several standard image formats. Since it can also save sequences of images, it can be used to load, process, then save DICOM data to a standard image sequence. It also implements the Bitonic filter and Refined Metal Artefact Reduction for CT data. For more information, and to obtain the software, see the wxDicom page.
This is a filter combining robust structurally varying (adaptive) morphological operations and an adaptive anisotropic Gaussian, with high noise reduction performance, particularly for very high noise levels. It is founded on preserving signal bitonicity, and since this concept only involves the order of values, not the image levels, it is naturally edge-preserving.
This is a filter for removing noise in signals and images whilst preserving edges and other bitonic data, ie. anything with either one local maximum or minimum within the range of the filter. It has better edge-preserving properties than a median filter, whilst also removing noise similarly to a Gaussian filter. It has wide applicability and very few parameters, with no need for training or prior knowledge of the noise variance, and is hence particularly suited to situations where the noise is varying.
This can be used for extracting triangulated iso-surfaces from a block of data - for instance the skin and skeleton which have been extracted from the female cadaver of the Visible Human Project data set, shown below. This is an image of a full 3-D model.
The software implements Regularised Marching Tetrahedra to triangulate the iso-surface, and Maximal Disc Guided Shape Based Interpolation to interpolate the thresholded data if required. The input is from a regular block of data, which can be scaled differently in each direction, and the surface can be extracted at any resolution. The output is a surface triangulation and a set of cross-sectional contours. For more information, and to obtain the software, see the IsoSurf page.
This can be used for generating triangulated surfaces from implicit functions, for instance those shown below.
The function is given as a command line string, for instance the commands to generate these surfaces were (left to right):
eqnsurf "(3-3*x)^2*exp(-x^2-(y+1)^2) - 10*(x/5-x^3-y^5)*exp(-x^2-y^2) - 1/3*exp(-(x+1)^2-y^2) - 3*z" -c 0,0,2 -l 3,3,4 -r 0.1
eqnsurf "(x^2+y^2+z^2-0.5)*(x-0.2)*(y-0.1)" -t 0.01 -r 0.04
eqnsurf "cos(x^2) + ln(y^2) + z^2 - 0.9" -l 2,2,2 -r 0.07
The software implements Regularised Marching Tetrahedra to triangulate the iso-surface, which can be extracted at any resolution. The visualisation requires both the OpenGL (or Mesa) and GLUT libraries. For more information, and to obtain the software, see the EqnSurf page.
This uses the same algorithm as used in IsoSurf for interpolating cross-sections to surfaces, but this time applied to the interpolation of surfaces to form time sequences. This means it can be used to 'morph' one surface into another, for example the queen and pawn shown below, which was generated in less than two minutes from the original objects (click the middle image to see a low quality movie of this morph sequence):
The surfaces are supplied as closed polygonal meshes in the OOGL format, however there is also a converter supplied from the more widely used 3DS format. Polygonal meshes are sampled (with antialias correction) to a volume, then interpolated, then re-triangulated using Regularised Marching Tetrahedra, before being displayed as a real time morph sequence that can be manipulated in 3D. The visualisation requires both the OpenGL (or Mesa) and GLUT libraries. For more information, and to obtain the software, see the VolMorph page. |