Abstract
A method for quickly re-rendering volume data consisting of several distinct materials and intermixed with moving geometry is presented. The technique works by storing depth, color and opacity information, to a given approximation, which facilitates accelerated rendering of fixed views at moderate storage overhead without re-scanning the entire volume. Storage information in the ray direction (what we have called super-z depth buffering), allows rapid transparency and color changes of materials, position changes of sub-objects, dealing explicitly with regions of overlap, and the intermixing or separately rendered geometry. The rendering quality can be traded-off against the relative storage cost and we present an empirical analysis of output error together with typical figures for its storage complexity. The method has been applied to the visualization of medical image data for surgical planning and guidance, and presented results include typical clinical data. We discuss the implications of our method for haptic (or tactile) rendering systems, such as for surgical simulation, and present preliminary results of rendering polygonal objects in the volume rendered scene.
Highlights
Visualization plays a central role in the presentation of 3D medical image data, such as Magnetic Resonance (MR), Computerized Tomography (CT), MR Angiography (MRA), or a combination of these (Wells et al, 1996b), to the radiologist or surgeon for diagnosis and surgery planning
A technique for quickly re-rendering volume data consisting of several distinct materials and intermixed with moving geometry was presented
We presented rendering results using both synthetic volumes and models, and typical clinical data which was used for surgery planning
Summary
We describe a volume compositing scheme that uses a specialized depth buffer which facilitates the rapid re-rendering of fixed viewpoints without the traversal of the entire volume. Surface rendering has the advantage that accelerated hardware is commonly available, but the disadvantage that isosurfaces have to be defined in the volume data in order to construct the triangle meshes (Lorensen and Cline, 1987) This may be satisfactory when the data to be visualized has already been segmented, e.g. using accurate statistical methods (Wells et al, 1996a) (Warfield et al, 1995), since the model surfaces have been defined. For intraoperative segmentation and virtual surgery simulation, physical tissue modeling using voxels is being advocated (Gibson, 1997) - the need for volume graphics Such systems raise the question of speed of update, i.e. being able to quickly reflect changes in the voxel volume, whether these are intermediate results from an elastic matching scheme or more the result of “volume sculpting”. An important characteristic of both these situations, which we have sought to exploit in this work, is the observer’s viewpoint being fixed, or not changed during interaction
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
