|
|
Producing images of the mucosal surface of the colon involves four steps. First, the patient's colon must be cleansed, which is similar to what is required for a barium enema or colonoscopy. Second, air is inflated into the colon to produce better contrast between the colon wall and the lumen. The third step is to take a helical CT scan of the abdomen. In the fourth step, visualization techniques are employed to generate the images, interactive flythroughs, and animations of the colon.
In our study, starting two days prior to the 3D virtual colonoscopy, the patient is asked to drink 100cc of liquid barium with each meal, so that retained stool can be differentiated from polyps or tumors. On the evening prior to the day of the procedure, the patient takes a standard colonic cleansing routine consisting of drinking one gallon of Golytely. After the patient arrives at the CT scanning suite, a small tube is placed into the rectum and approximately 1,000cc air is pumped into the colon to distend the colon. A spiral CT scan with an X-Ray beam of 5mm width, 1:1 or 1:2 pitch, and 40cm field-of-view is then performed from the level of the top of the splenic flexure of the colon to the rectum based upon the landmarks obtained from the digital scout image. The scanned data is next reformatted into 5mm thick slices at increments of 1mm or 2.5mm, with each slice represented as a matrix of 512 by 512 pixels. In this way, a large number of 2D slices are generated depending on the length of the scan. This varies from 107 to 280 in our experiments (see Section 6). The actual data acquisition time is between 30 and 45 seconds. Later this set of 2D slices is reconstructed into a 3D volume to be visualized using the VolVis system.
VolVis is a comprehensive, diversified, and high performance volume visualization system developed at SUNY Stony Brook [1]. It supports numerous visualization algorithms and methods within a consistent and well-organized framework, ranging from fast rough approximations, to compression-domain rendering, to accurate volumetric ray tracing and radiosity, to irregular grid rendering. Input primitives accepted by VolVis include 3D scalar volumetric data as well as 3D volume-sampled and classical geometric models. The VolVis system consists of several primary components which are intended to meet the various needs of volume visualization users. These components are File I/O, Filters, Object Control, Image Control, Rendering, Navigation, Animation, Quantitative Analysis, and Input Devices. In this study, we employ extensively the Navigation and Animation components of the VolVis system to provide interactive object manipulation and straightforward specifications of complex animations [2] for the visualization of the colon.