Premium
Calculating output factors for photon beam radiotherapy using a convolution/superposition method based on a dual source photon beam model
Author(s) -
Liu H. Helen,
Mackie T. Rock,
McCullough Edwin C.
Publication year - 1997
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.598111
Subject(s) - superposition principle , photon , convolution (computer science) , beam (structure) , physics , optics , radiation , dosimetry , computer science , nuclear medicine , quantum mechanics , medicine , machine learning , artificial neural network
A realistic photon beam model based on Monte Carlo simulation of clinical linear accelerators was implemented in a convolution/superposition dose calculation algorithm. A primary and an extra‐focal sources were used in this beam model to represent the direct photons from the target and the scattered photons from other head structures, respectively. The effect of the finite size of the extra‐focal source was modeled by a convolution of the source fluence distribution with the collimator aperture function. Relative photon output in air( S c ) and in phantom( S cp ) were computed using the convolution method with this new photon beam model. Our results showed that in a 10 MV photon beam, the S c ,S p(phantom scatter factor), and S cpfactors increased by 11%, 10%, and 22%, respectively, as the field size changed from 3 × 3 cm 2to 40 × 40 cm 2 . The variation of the S cfactor was contributed mostly by an increase of the extra‐focal radiation with field size. The radiation backscattered into the monitor chamber inside the accelerator head affected the S cby about 2% in the same field range. The output factors in elongated fields, asymmetric fields, and blocked fields were also investigated in this study. Our results showed that if the effect of the backscattered radiation was taken into account, output factors in these treatment fields can be predicted accurately by our convolution algorithm using the dual source photon beam model.