Biomedical Engineering Reference
In-Depth Information
but the point to be made is
independent of the color
palette
employed.)
In
panels (a), (b), and (c), the
lower end of the color
scale has been chosen to
be 25, 10 and 45 Gy
respectively. All three
panels show the same pair
of dose distributions, one
of a photon plan, the other
of a proton plan. Panel (a)
is probably a reasonable
comparison of the two
plans; panel (b) makes
photons look quite a bit
worse than protons; and
panel (c) makes protons
and photons look virtually
identical. Quite different
conclusions
could
be
drawn about the com-
parative merits of the
Figure 6.10. A comparison of two plans
using three different low-end cutoffs for
the color display (see text). Figure
courtesy of A. Niemierko, MGH, USA.
two modalities, depending
on the range of doses
included in the color-wash.
This example highlights two important points. First, dose displays
can be misleading. This is equally true when isodose contours are
used. However, the greater immediacy of color-wash makes the
perils rather starker. One must not be beguiled by pretty color
pictures. Both for the reason being brought out here, and because
there are always uncertainties in the computed dose distributions, one
must be aware that
it is not necessarily the case that “what you see is
what you get.”
The ideal is WYSIWYG; reality sometimes falls far
behind.
The second point is that, whenever one is looking at a treatment plan
whose doses are displayed over a restricted range, one must
interactively adjust the endpoints of the range to ensure that there are
no hidden surprises. I already pointed this out in Chapter 3 as it
regards the window and level with which CT and other images are
viewed; it is equally true for dose displays.
