Image Processing Reference
In-Depth Information
the ray are combined, and illumination model level, where the illumination model
is adapted to process two volume samples directly. The second example we men-
tion is a convincing application of multi-modal volume rendering for the planning
of neurosurgical interventions, where MRI, CT, fMRI, PET and DSA data are all
combined in an interactive but high quality visualization for the planning of brain
tumor resection [
8
].
23.2.2 Therapy Planning, Predictive Simulation, and Diagnosis
Therapy planning was one of the first real applications of medical visualization and
remains important to this day. In 1993, Altobelli et al. [
1
] published their work on
using CT data to visualize the possible outcome of complicated craniofacial surgery.
By manually repositioning soft tissue fragments based on the bony surfaces under
them, in certain cases taking into account bone-skin motion ratios from literature,
the expected outcome of a craniofacial procedure could be visualized. Although
still rudimentary, this could be considered one of the earliest cases of
predictive
or outcome simulation
integrated with visualization for surgical planning. The idea
of predictive simulation, or predictive medicine, was further explored by Taylor
et al. [
66
] for cardiovascular surgery.
With the introduction of virtual colonoscopy (VC) in 1995 [
28
], medical visual-
ization also gained diagnosis as an important medical application, namely screening
for colon cancer. VC combines CT scanning and volume visualization technologies.
The patient's abdomen is imaged in a few seconds by a multi-slice CT scanner. A 3D
model of the colon is then reconstructed from the scan by automatically segmenting
the colon and employing “electronic cleansing” of the colon for computer-based
removal of the residual material. The physician then interactively navigates through
the volume rendered virtual colon employing camera control mechanisms, cus-
tomized tools for 3D measurements, “virtual biopsy” to interrogate suspicious
regions, and “painting” to support 100 % inspection of the colon surface [
29
]. VC is
rapidly gaining popularity and is poised to become the procedure of choice in lieu of
the conventional optical colonoscopy for mass screening for colon polyps—the pre-
cursor of colorectal cancer. Unlike optical colonoscopy, VC is patient friendly, fast,
non-invasive, more accurate, and a more cost-effective procedure for mass screening
for colorectal cancer.
VC technologies gave rise to the computer-aided detection (CAD) of polyps,
where polyps are detected automatically by integrating volume rendering, conformal
colon flattening, clustering, and “virtual biopsy” analysis. Along with the reviewing
physician, CAD provides a second pair of “eyes” for locating polyps [
30
]. This
work was also the basis for many other virtual endoscopy systems, such as virtual
bronchoscopy, virtual cystoscopy, and virtual angioscopy. A careful integration of
image analysis (e.g., segmentation, skeletonization), with efficient rendering (e.g.,
occlusion culling) and interaction (e.g., camera control based on predefined paths)
are major ingredients of such systems [
3
,
29
].
Search WWH ::
Custom Search