Premium
Using Monte Carlo methods to commission electron beams: A feasibility study
Author(s) -
Antolak John A.,
Bieda Michael R.,
Hogstrom Kenneth R.
Publication year - 2002
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 - 10.1118/1.1469626
Subject(s) - monte carlo method , dosimetry , linear particle accelerator , physics , computational physics , electron , range (aeronautics) , beam (structure) , cathode ray , nuclear medicine , optics , nuclear physics , mathematics , materials science , statistics , medicine , composite material
The purpose of this study was to investigate the feasibility of using Monte Carlo methods to assist in the commissioning of electron beams for a medical linear accelerator. The EGS4/BEAM code system was used to model an installed linear accelerator at this institution. Following an initial tuning of the input parameters, dosimetry data normally measured during the machine commissioning was calculated using the Monte Carlo code. All commissioning data was calculated for 6‐ and 12‐MeV electron beams, and a subset of the commissioning data was calculated for the 20‐MeV electron beams. On central axis, calculated percentage depth dose, cross‐beam profiles, cone‐insert ratios, and air‐gap factors were generally within 2% of D maxor 1 mm of the measured commissioning data; however, calculated open‐cone ratios were not within 2%, in most cases. Calculated off‐axis dose profiles for small fields were generally within the 2% (1‐mm) criteria; however, calculated dose profiles for larger (open cone) fields frequently failed the 2% (1‐mm) criteria. The remaining discrepancies between Monte Carlo calculations and measurement could be due to flaws in the Monte Carlo code, inaccuracies in the simulation geometry, the approximation of the initial source configuration, or a combination of the above. Although agreement between Monte Carlo calculated and measured doses was impressive and similar to previously published comparisons, our results did not prove our hypothesis that Monte Carlo calculations can generate electron commissioning data that is accurate within 2% of D maxor 0.1 cm over the entire range of clinical treatment parameters. Although we believe that this hypothesis can be proved, it remains a challenge for the medical physics community. We intend to pursue this further by developing systematic methods for isolating causes of these differences.