Biomedical Engineering Reference
In-Depth Information
if: (1) the lines at which each section intersects the others are shown;
shown; and (2) the user can move around the 3D space by, say,
dragging the lines of intersection about and having the display update
immediately. Such an display is shown in Figure 6.6.
Interactivity
The importance of being able to maneuver rapidly through these large
data sets cannot be sufficiently emphasized. One needs not only a
high resolution color
6
display device, but a system powerful enough
to rapidly show changes in the images as one changes some
controlling parameter. A refresh rate of at least 10 per second is
desirable.
Time variation of dose
So far, I have described the use of sequential displays as a tool to
explore the third spatial dimension. However, of course, sequentially
displayed 2D images can be used to make evident the dimension of
time. One would then see, in a movie loop display as described
above, how the dose distribution in a given 2D section varies with
time. By switching between sections, one can explore the full 4D
space.
There is a subtlety regarding the representation of dose as a function
of time. What one is primarily interested in is the dose at some
anatomic point
as a
function of time. In general that cell will move about in time - that is,
the 2D section in which the cell lies will move and distort with time.
This introduces an added complication. What one wants is for the
anatomy to appear to stay static and the dose display (e.g., the color-
wash display or isodose lines) to vary with time. This requires that:
(1) the images taken at different times be spatially registered with
one another using, for example, a deformable registration technique
as discussed in Chapter 3; and (2) the dose display for any particular
time be mapped onto the anatomic information at some reference
6
Tufte has written a series of most interesting topics on the representation of
−
let us say at the position of a particular cell
−
quantitative data (Tufte 1990, 1997, 2001). He makes the point that
everything in an image should serve a clear purpose. Colors should not be
used just because they look pretty. However, in radiation oncology, the
use of color is virtually essential. It is needed to code such things as
anatomy (using different colors for different anatomic structures), or dose
(using different colors for different dose ranges), or both.
