Biomedical Engineering Reference
In-Depth Information
each type of tissue. The models had several variables and we varied
them all. Figure 8.8 shows graphs of the difference in NTCP between
the two plans, for the two tissue architectures, and for two of the more
significant variables in the analysis in each case.
What one sees is that, in the case of a serial tissue architecture, the
rotation plan appears to be better for all combinations of the two
variables shown. However, in the case of a parallel architecture
tissue, while there are combinations of the two variables for which the
rotation plan would be preferred, there are also regions where the 3-
field plan is better. These regions are those in which the D
50
of the
tissue is small compared with the tumor dose - i.e., where the parallel
architecture tissues are quite radiosensitive
−
and/or the critical volume
is large.
NTCP(3-fields) - NTCP(360° rotation)
Figure 8.8. Graphs of the difference in NTCP between a 3-field and
rotation plan: (a) normal tissues have a parallel architecture, and (b)
normal tissues have a serial architecture. Shades of blue are negative
differences, implying that the 4-field plan results in a lower NTCP.
Shades of red are positive differences, implying that the rotation plan
results in a lower NTCP.
One can perform the same kind of analysis using the EUD model
described in Chapter 5 which has only one parameter,
a
. In this case,
one finds that the rotation plan is preferred for tissues that have
a
> 1
and the 3-field plan is preferred for the lesser number of tissues that
have
a
< 1. Interestingly, the two approaches are predicted to be
equally good in the case of
a
= 1. For that value, the average normal
tissue dose determines the NTCP - and the average normal tissue
dose and the integral dose are the same in the two plans.
