405 / 1084
Validity of the margin recipe
•
Assumes
homogeneous patient population
(identical SD)
•
Assumes
many fractions
•
Assumes
spherical symmetry
More or less OK if CTV >> σ
•
Assumes
“ideal” conformation
ie preparation errors have the same impact in all directions
•
Rotations and shape variations
have been ignored
•
Uncertainties were assumed to be isotropic
The concept can be generalized to 3D by separating x, y, and z
directions.
•
The different sources of uncertainties are assumed to be
statistically independent
As most of the uncertainties are introduces at different stages of the
treatment, this assumption seems OK
•
And again:
normal probability distributions
are assumed.
The PTV 2017 - D. Verellen