Program Information
Validation of Photon Beam Models: Effective Test for Modeling Deficiency
J Xue*, H Wang , P Wu , I Das , NYU Langone Medical Center, New York, NY
Presentations
SU-I-GPD-T-433 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: This study is to explore an effective technique for diagnosing deficiencies of the photon beam model in commissioning a dose calculation algorithm.
Methods: For modern linear accelerators equipped with multileaf collimator (MLC), photon beam modeling applies appropriate leaf transmission, static leaf gap and dosimetric leaf gap (DLG) among other parameters to achieve an accurate dosimetry in 3DCRT, IMRT and VMAT delivery. We designed a series of MLC-shaped fields to understand the effect of these parameters on the goodness of a model. Measurement of both in-plane and cross-plane profiles with varied field sizes is used to compare with the calculation from TPS.
Results: Measured and calculated profiles in absolute dose can be visualized that indicate any discrepancies on varied dosimetric regions (in-field, penumbra and tails) for each photon beam. It seems 6XFFF has the worst disagreement on the tails with cross-plane profiles, which can be a measure of poor parameters in the model for DLG and static leaf gap. The10XFFF looks to have an optimal beam model since the difference for all field sizes and depths is reasonably small.Sizable to considerable difference is observed on the tails for 6X and 6XFFF, suggesting the issues on the beam model with the setting of static leaf gap and DLG. In clinic, we did observe lower Gamma pass rate on VMAT QA with both 6X and 6XFFF beams. Further investigation is warranted to understand how the modeling parameters can be tuned to improve QA pass rate.
Conclusion: We demonstrated an effective and time-saving approach to detect modeling deficiencies before performing a thorough validation test of a photon beam model, e.g., following MPPG 5.a. The suggested test is to verify static leaf gap, DLG and penumbra as they are the critical parameters in deriving an optimal beam model, particularly for IMRT and VMAT.
Contact Email: