Abstract
Multileaf collimators (MLCs) need to be characterized accurately in treatment planning systems to facilitate accurate intensity‐modulated radiation therapy (IMRT) and volumetric‐modulated arc therapy (VMAT). The aim of this study was to examine the use of MapCHECK 2 and ArcCHECK diode arrays for optimizing MLC parameters in Monaco X‐ray voxel Monte Carlo (XVMC) dose calculation algorithm. A series of radiation test beams designed to evaluate MLC model parameters were delivered to MapCHECK 2, ArcCHECK, and EBT3 Gafchromic film for comparison. Initial comparison of the calculated and ArcCHECK‐measured dose distributions revealed it was unclear how to change the MLC parameters to gain agreement. This ambiguity arose due to an insufficient sampling of the test field dose distributions and unexpected discrepancies in the open parts of some test fields. Consequently, the XVMC MLC parameters were optimized based on MapCHECK 2 measurements. Gafchromic EBT3 film was used to verify the accuracy of MapCHECK 2 measured dose distributions. It was found that adjustment of the MLC parameters from their default values resulted in improved global gamma analysis pass rates for MapCHECK 2 measurements versus calculated dose. The lowest pass rate of any MLC‐modulated test beam improved from 68.5% to 93.5% with 3% and 2 mm gamma criteria. Given the close agreement of the optimized model to both MapCHECK 2 and film, the optimized model was used as a benchmark to highlight the relatively large discrepancies in some of the test field dose distributions found with ArcCHECK. Comparison between the optimized model‐calculated dose and ArcCHECK‐measured dose resulted in global gamma pass rates which ranged from 70.0%–97.9% for gamma criteria of 3% and 2 mm. The simple square fields yielded high pass rates. The lower gamma pass rates were attributed to the ArcCHECK overestimating the dose in‐field for the rectangular test fields whose long axis was parallel to the long axis of the ArcCHECK. Considering ArcCHECK measurement issues and the lower gamma pass rates for the MLC‐modulated test beams, it was concluded that MapCHECK 2 was a more suitable detector than ArcCHECK for the optimization process.PACS number(s): 87.55.Qr
Highlights
38 Kinsella, et al.: Multileaf collimators (MLCs) parameter determination fraction of the delivered monitor units (MU).(1,2) multileaf collimator (MLC) need to be characterized accurately in treatment planning systems (TPSs) to facilitate accurate intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT)
The jaws and MLCs are characterized in the X-ray voxel Monte Carlo (XVMC) model using a transmission probability filter (TPF).(4–7) While commissioning and dosimetric evaluation of Monaco XVMC has been previously described,(5,6,8–11) no specific details were provided on how the TPF MLC characteristics were determined
A representative cut-out of Fig. 1(b) is given in Fig. 1(c) and highlights that MapCHECK 2 measurements sampled the peaks and troughs of the MLC leakage at regular intervals, which facilitated straightforward evaluation of how much the TPF MLC parameters needed to be adjusted. Both film and MapCHECK 2 measurements indicated there was a greater difference between the interleaf and intraleaf transmission compared to the calculated dose with the default settings. It was found through an iterative process that the best match between calculated and measured dose distributions resulted when the ‘leaf transmission’ parameter was adjusted from the default value of 0.0120 to 0.007 and the ‘interleaf leakage’ parameter was adjusted from default value of 5.00 to 25.00
Summary
38 Kinsella, et al.: MLC parameter determination fraction of the delivered monitor units (MU).(1,2) MLCs need to be characterized accurately in treatment planning systems (TPSs) to facilitate accurate IMRT and VMAT. Nelms et al[3] showed how patient-specific quality assurance (PSQA) procedures fail to highlight errors in the MLC model, indicating the necessity to characterize them correctly at the time of commissioning. Monaco TPS (IMPAC Medical Systems, Inc., Sunnyvale, CA (an affiliate of Elekta AB, Stockholm, Sweden)) employs the X-ray voxel Monte Carlo (XVMC) dose calculation algorithm. The jaws and MLCs are characterized in the XVMC model using a transmission probability filter (TPF).(4–7) While commissioning and dosimetric evaluation of Monaco XVMC has been previously described,(5,6,8–11) no specific details were provided on how the TPF MLC characteristics were determined. The aim of this study was to explore if it was possible to optimize the TPF MLC parameters of the XVMC dose calculation algorithm and to evaluate the relative merit of MapCHECK 2 and ArcCHECK for this process
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
