S1876
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Simulations were performed using Monte Carlo in Monaco for both VSM1.6 and VSM2.0. Measurements were performed in a water tank and RW3 phantom. Both phantoms were created in Monaco for calculations. Beam characterization included open field dose distributions and patient plans for comparison between measurements and calculations using global gamma analysis. MLC modelling was evaluated by comparing MLC transmission, leaf tip position and penumbra, leaf offset, dosimetric leaf gap (DLG), and tongue-and-groove effects of measurements and calculations. Results: Both models showed excellent agreement with measurements for open field profiles (median gamma values VSM1.6; 0.19, 0.18 and 0.22 for 6MV, 10MV and 6MV-FFF, and for VSM2.0; 6MV: 0.18, 10MV: 0.18 and 6MV-FFF: 0.14) and patient plans (gamma pass rate >97.7% for both models). VSM1.6 and VSM2.0 showed high agreement for the leaf tip positioning and penumbra (50% isodose level within 0.3 mm for both models; Figure 1). VSM2.0 demonstrated improved accuracy for MLC transmission (VSM2.0: maximum deviation <0.1%, VSM1.6: overestimation up to <0.45%) and tongue-and-groove effects (Figure 2; illustrative of results). While VSM1.6 more accurately represented DLG (VSM1.6: difference <0.1 mm, VSM2.0: difference up to 0.45 mm with measurements) and leaf offset due to extensive clinical tuning.
significant changes relative to the original plan for both CW and WB. PTV D95% for WB IMRT plans calculated on dCTs without skin flash showed a difference of up to -6.3% while no significant changes were seen for the CW plans. Conclusion: This study provided an insight into the effect of deformation on the delivered dose for breast cancer patients. This work emphasized the importance of using skin flash for IMRT plans to achieve the desired dose coverage, especially for WB cases. References: [1] Varian Medical Systems, Inc. (2017). Image Registration and Segmentation Reference Guide (version 15.5.). Palo Alto, CA. Keywords: Breast deformation, treatment plan, IMRT, CBCT Validation of the new GPU-based Monte Carlo beam and MLC models for the Monaco treatment planning system Fasco van Ommen 1,2 , Sima Sajadi Shahrbabaki 1 , Thomas A. Foppen 1 , Jordi Saez 3 , Victor Hernandez 4 , Bram van Asselen 1 , Sara L. Hackett 1 1 Department of Radiation Oncology, UMC Utrecht, Utrecht, Netherlands. 2 Department of Medical Physics, HollandPTC, Delft, Netherlands. 3 Department of Radiation Oncology, Hospital Clínic de Barcelona, Barcelona, Spain. 4 Department of Medical Physics, Hospital Sant Joan de Reus, Tarragona, Spain Purpose/Objective: Accurate dose calculation is critical for radiotherapy treatment planning, particularly for highly modulated techniques such as IMRT and VMAT, where multileaf collimator (MLC) modeling is a major source of uncertainty. The Elekta Monaco treatment planning system (TPS) has historically used a virtual source model (VSM1.6) incorporating virtual energy fluence and analytical transmission probability filters (TPFs) to model the photon beam and beam-shaping devices. This model, however, requires user-specific Digital Poster 2153 optimization, resulting in inter-institutional variability in model accuracy. Elekta introduced a new GPU- based Monte Carlo algorithm employing a single phase-space beam model (VSM2.0) without TPFs or user-adjustable parameters. This study aimed to characterize and compare VSM2.0 with the established VSM1.6 model and experimental measurements for Elekta Versa HD linear accelerators equipped with Agility MLCs. Material/Methods: Measurements and simulations were performed for three beam energies (6 MV, 10 MV, and 6 MV-FFF).
Made with FlippingBook - Share PDF online