Automatic determination of primary electron beam parameters in Monte Carlo simulation

Abstract

In order to obtain realistic and reliable Monte Carlo simulations of medical linac photon beams, an accurate determination of the parameters that define the primary electron beam that hits the target is a fundamental step. In this work we propose a new methodology to commission photon beams in Monte Carlo simulations that ensures the reproducibility of a wide range of clinically useful fields. For such purpose accelerated Monte Carlo simulations of 2 x 2, 10 x 10, and 20 x 20 cm2 fields at SSD = 100 cm are carried out for several combinations of the primary electron beam mean energy and radial FWHM. Then, by performing a simultaneous comparison with the correspondent measurements for these same fields, the best combination is selected. This methodology has been employed to determine the characteristics of the primary electron beams that best reproduce a Siemens PRIMUS and a Varian 2100 CD machine in the Monte Carlo simulations. Excellent agreements were obtained between simulations and measurements for a wide range of field sizes. Because precalculated profiles are stored in databases, the whole commissioning process can be fully automated, avoiding manual fine-tunings. These databases can also be used to characterize any accelerators of the same model from different sites.

An investigation into the use of MMCTP to tune accelerator source parameters and testing its clinical application

“…Initially starting at large field ($30\times 30\hspace{0.17em}{\text{cm}}^2$) dose profiles and PDDs and progressing to examination of output factors for a range of field sizes, focusing particularly on smaller field sizes for finer tuning. Small fields are more sensitive to changes in the FWHM of the radius of the incident electron beam, ( 3 , 15 ) which allows tuning to be performed more quickly and accurately. It was decided to focus on the larger field sizes initially as these are more sensitive to changes in the mean energy of the initial electron source when looking at profiles at 1.5 cm depth ( 3 , 16 ) — again improving efficiency and precision of the tuning process being developed.…”

“…In order to get a lower uncertainty and hence decide upon the optimum values for the initial electron beam, the simulations would become very long and the tuning process unacceptably time‐consuming for clinical implementation (Table 1). Output factors for the $30\times 30\hspace{0.17em}{\text{cm}}^2$ field were used to further finetune the energy as output factors are more sensitive to changes in the primary beam parameters ( 3 ) and, as a result, require less computation time than extended simulations and profile comparisons. The output factor was calculated from the dose scored in a $0.5\times 0.5\times 0.5\hspace{0.17em}{\text{cm}}^3$ voxel on the central axis at 5 cm depth for the Varian Clinac, as this was the depth the output factors were measured at in the clinic.…”

“…The output factors for square fields of $40\times 40\hspace{0.17em}{\text{cm}}^2$, $30\times 30\hspace{0.17em}{\text{cm}}^2$, $5\times 5\hspace{0.17em}{\text{cm}}^2$, $4\times 4\hspace{0.17em}{\text{cm}}^2$, and $3\times 3\hspace{0.17em}{\text{cm}}^2$ were considered. The inclusion of smaller field sizes allowed for a more accurate determination of the FWHM of the beam radius ( 3 ) as these are the most sensitive to changes in the spatial distribution, while larger fields are more sensitive to energy variations. From this it was determined that the optimum value for the Gaussian distribution of the radius was 0.05 cm, because this radius value provided output factor values that matched measurements within 1% for all the field sizes investigated (Fig.…”

“…The angular spread of electrons resulting from scatter in air from a narrow collimated beam, for most practical applications, can be approximated by a Gaussian radial distribution. ( 3 , 7 , 8 ) In order for the BEAMnrc ( 9 , 10 ) code to be used to calculate accurate dose distributions, the accelerator model must first be benchmarked. The initial electron energy and full width half maximum (FWHM) of the radius of the initial electron beam incident on the target was varied to find the percentage depth dose, dose profile curves, and output factors that match the hospital‐measured data, providing output factors that match within 1% of measured output factor values.…”

“…In the work by Pena et al ( 3 ) for the Siemens PRIMUS (Siemens AG, Erlangen, Germany) and Varian 2100 CD (Varian Medical Systems, Palo Alto, CA), the initial electron energy is varied in steps of 0.25 MeV over an energy range of 5.5 to 6.5 MeV. For the Siemens KD, Varian Clinac, and Elekta SL25 (Elekta, Stockholm, Sweden), Sheikh‐Bagheri and Rogers ( 4 ) vary the energy in steps of 0.1 MeV over a range from 5.5 to 6.6 MeV, and Tzedakis et al ( 5 ) use increments of 0.2 MeV for the energy tuning from 5.0 to 7.0 MeV for the Philips/Elekta SL75/5.…”