The Simian Project

My focus within computer graphics is in the subfield of scientific visualization, specifically the extraction and display of features from three dimensional data sets.

I am presently working on the Simian volume rendering tool. The Simian allows for the visualization of three dimensional data sets generated using imaging techniques such as computerized axial tomography and magnetic resonance imaging as well as the results of physical simulations. The image at right is a treatment of CT data. Using Simian the user can highlight specific volume elements (voxels) within the data set based on each voxel's density and density gradient.

In the screen shot to the right, the box below the image of the subject's head is the present interface allowing the user to define an opacity assignement function (also called a transfer function) appropriate for the data set and the features to be emphasized. The large rectangular sub-tool on the right is responsible for assigning opacity to the densest voxels in the dataset, those corresponding to the bone. The triangular tool on the left assigns a lesser opacity to the those voxels where there is a meeting of air and flesh, i.e., those corresponding to the subject's skin.

Joe Kniss is another student of Chuck's who put together the last fully-functional release fo the Simian. A new version is in the works. You can find the image at right and others like it on the site he maintains (linked below.)

My research this semester will explore techniques to poll the data provided to the tool to determine what type of transfer function tools should be used and where they should be placed in order to extract features of interest automatically. At present Simian requires significant time to master. If an appropriate transfer function could be assigned without user input, or with only minimal adjustments, Simian could see real use in medical research and other research disciplines.

Links

Joe's Simian Website

UCSD Neuron and Sub-Cellular Structure images produced with the Simian