Biomedical Engineering Reference
In-Depth Information
weeks, as when the patient loses weight or the tumor shrinks. Only
recently have tools such as 4DCT studies become available to follow
the short-term motions of the patient. Long-term changes can be, and
traditionally have been, tracked by repeating imaging studies after,
say, the first weeks of therapy.
As I mentioned above, even semitransparent 3D data are simply too
overwhelming to take in, and the best way we have at present of
viewing time-varying dose distributions is to look at a 2D section of
the patient taken at a certain time, and then sequence through the
sections at progressively later times, showing a time sequence of
snap-shots in a kind of movie-loop.
2D dose distributions
I am jumping over the presentation of 3D dose distributions, as 2D
dose displays are the bedrock of dose viewing, and 3D presentation
is largely based on 2D presentation. The solution has already been
shown, for example in Figure 6.1 above. Dose and anatomic
information are superimposed in a 2D image, on the screen or on
paper, as here. The anatomic information (e.g., CT Hounsfield units)
is represented by the image intensity at any point. Experienced users
are familiar enough with normal anatomy to be able to interpret the
images and identify the important normal structures as well as,
sometimes, the tumor. These may be enhanced by overlaying the
outlines of any delineated structures, such as the target volume as
Figure 6.3. Dose superposed on a CT section with outlines of
OARs and the PTV (shown as closed color-coded contours), using
isodose contours (
left
) and color-wash (
right
) to represent the dose
in the chosen section. Figure courtesy of A. Lomax, PSI, CH.
