Dr
Barbara Caccia
(Technologies and Health Dept. Istituto Superiore di Sanità and INFN, Roma (Italy))
The advances in oncological research have had an important impact on cancer patients survival, but there
has been no major or dramatic improvement in long-term survival in the lung cancer despite more
innovative treatment techniques and protocols implemented in radiation oncology. The problematic
concerning dose calculation in the presence of in-homogeneities has already been afforded by several authors
but a new challenge has been rising since radiation therapies based on respiratory gating systems have been
introduced in the clinical practice. The MonteCarlo (MC) algorithms have already been proven to be a
realistic alternative to analytical algorithms in those regions where charged particle equilibrium does not
hold. The aim of this work is to evaluate the capability of GEANT4 respect to some commercial treatment
planning systems to provide dose calculation maps with high level of accuracy also when lung densities are
changing (i.e. during different respiratory phases, from breathing out to deep inspiration). GEANT4
simulations have been compared to dosimetric data measured by radiochromic films.
In this study, a 6 MV beam from a 2100 Clinac Varian was modeled with a GEANT4 MC code.
The reported results were obtained with a modified version of the GEANT4 Advanced Example MedLinac.
Preliminary measurements were carried out for different square field sizes (20x20 cm2, 7.5x7.5 cm2 and 3x3
cm2 ) with SSD=84 cm. To study the dose distribution a series of three phantoms were constructed and
simulated, each characterized by a different lung density according to the following values: 0.030, 0.080
and 0.40 g/cm3.
The MC simulation jobs were run on a dedicated Beowulf cluster located in the Technology and Health
Department of the Italian National Institute of Health.
Two different algorithms released for clinical application from two commercial TPS were included in our
study: Varian Eclipse TPS (PB-EqTAR pencil beam with EqTAR algorithm for heterogeneities correction)
and Philips Pinnacle TPS (CCC Collapsed Cone Convolution).
Results obtained by the MC simulation and calculated data obtained from commercial TPS have been
compared with measurements performed with radiochromic films (GafchromicTM EBT). The films were
exposed at different depths perpendicularly and parallel to the beam axis at different distances. Images were
acquired with a Epson Expression 10000 XL flatbed scanner and analyzed with Picodose X PRO software.
Results show a good agreement between GEANT4 simulations and dosimetric data.
Summary
The advances in oncological research have had an important impact on cancer patients survival, but there
has been no major or dramatic improvement in long-term survival in the lung cancer despite more
innovative treatment techniques and protocols implemented in radiation oncology. The problematic
concerning dose calculation in the presence of in-homogeneities has already been afforded by several authors
but a new challenge has been rising since radiation therapies based on respiratory gating systems have been
introduced in the clinical practice. The MonteCarlo (MC) algorithms have already been proven to be a
realistic alternative to analytical algorithms in those regions where charged particle equilibrium does not
hold. The aim of this work is to evaluate the capability of GEANT4 respect to some commercial treatment
planning systems to provide dose calculation maps with high level of accuracy also when lung densities are
changing (i.e. during different respiratory phases, from breathing out to deep inspiration). GEANT4
simulations have been compared to dosimetric data measured by radiochromic films.
In this study, a 6 MV beam from a 2100 Clinac Varian was modeled with a GEANT4 MC code.
The reported results were obtained with a modified version of the GEANT4 Advanced Example MedLinac.
Preliminary measurements were carried out for different square field sizes (20x20 cm2, 7.5x7.5 cm2 and 3x3
cm2 ) with SSD=84 cm. To study the dose distribution a series of three phantoms were constructed and
simulated, each characterized by a different lung density according to the following values: 0.030, 0.080
and 0.40 g/cm3.
The MC simulation jobs were run on a dedicated Beowulf cluster located in the Technology and Health
Department of the Italian National Institute of Health.
Two different algorithms released for clinical application from two commercial TPS were included in our
study: Varian Eclipse TPS (PB-EqTAR pencil beam with EqTAR algorithm for heterogeneities correction)
and Philips Pinnacle TPS (CCC Collapsed Cone Convolution).
Results obtained by the MC simulation and calculated data obtained from commercial TPS have been
compared with measurements performed with radiochromic films (GafchromicTM EBT). The films were
exposed at different depths perpendicularly and parallel to the beam axis at different distances. Images were
acquired with a Epson Expression 10000 XL flatbed scanner and analyzed with Picodose X PRO software.
Results show a good agreement between GEANT4 simulations and dosimetric data.
Keymords
| Are you a Memeber of the Geant4 Collaboration (yes/no) |
no
|
Antonella Soriani
(Medical Physics Laboratory, Istituto Regina Elena and INFN, Roma (Italy))
Dr
Barbara Caccia
(Technologies and Health Dept. Istituto Superiore di Sanità and INFN, Roma (Italy))
Dr
Claudio Andenna
(DIPIA – ISPESL and INFN, Roma (Italy))
Dr
Gianluca Frustagli
(Technologies and Health Dept. Istituto Superiore di Sanità and INFN, Roma (Italy))
Dr
Giuseppe Iaccarino
(Medical Physics Laboratory, Istituto Regina Elena and INFN, Roma (Italy))
Dr
Lidia Strigari
(Medical Physics Laboratory, Istituto Regina Elena and INFN, Roma (Italy))
Dr
Maurizio Mattia
(Technologies and Health Dept. Istituto Superiore di Sanità and INFN, Roma (Italy))
Dr
Stefano Valentini
(Technologies and Health Dept. Istituto Superiore di Sanità and INFN, Roma (Italy))
Dr
Valeria Landoni
(Medical Physics Laboratory, Istituto Regina Elena and INFN, Roma (Italy))
