S1905
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Purpose/Objective: The treatment planning system (TPS) requires specific parameters to be entered by the user to optimize a plan. One of these parameters is the dose rate. However, the preset maximum dose rate is often not utilized during delivery, leading to unnecessary power consumption and wear and tear on machine parts, thereby shortening their lifespan. A potential solution is to check the control point list to determine the highest dose rate used and then adjust this parameter in the treatment field option accordingly post- optimization. This adjustment should not affect delivery times, which needs to be verified. Material/Methods: Treatment plans for various tumor sites, nominal energies, and dose rates were created in the Eclipse v16.0 treatment planning system (Varian, Palo Alto, USA). After optimization of the initial plan with the highest available dose rate of the selected energy, the dose rate was reduced to the next higher setting than the maximum used one in the control point list. If permissible, additional reduced dose rates were selected. After these reductions, the doses were once recalculated and once fully re-optimized. All variations were exported and delivered on a TrueBeam Edge v4.1 system (Varian). The time indicator at the console was recorded, and the trajectory log files were exported for later analysis in the Matlab v2023b solution (The MathWorks, Inc., USA) to assess delivery time, gantry speed, and maximum dose rate. Results: All fields, i.e. all treatment sites and variations, were delivered within the range of 1 minute 5 seconds to 1 minute 8 seconds. The most significant change in the beam time indicator was a factor of 3.6, yet the actual delivery times differed by only two seconds. All other test cases showed a 0-1 second difference in delivery times. It was discovered that the dose rate in the control point list of the TPS is, on average, underestimated by 102.2 MU/min. The gantry speed was underestimated by up to 1.2°/s. Conclusion: Based on the cases investigated, no significant increase in delivery time was observed, despite the beam time display indicating otherwise. However, the TPS used underestimates the actual values of dose rates and gantry speeds. Therefore, it may be advisable to use a dose rate two levels higher instead of the next highest level of value in the control point list to provide a safety buffer for the discovered inaccuracies.All trademarks are the property of their respective owners. QR700030407 Keywords: beam time, dose rate, post-optimization
Digital Poster 2889
Evaluation of updated automatic skin flash feature of RapidArc Dynamic in breast cancer irradiation Topi Nykänen 1,2 , Ville Raatikainen 1 , Aarno Kärnä 1 , Katariina Näkki 1 , Tuomas Koivumäki 1 1 Department of Medical physics, Hospital Nova of Central Finland, Wellbeing Services County of Central Finland, Jyväskylä, Finland. 2 Department of Physics, University of Jyväskylä, Jyväskylä, Finland Purpose/Objective: A new automatic skin flash (ASF) feature introduced in RapidArc Dynamic (RAD) has been shown to perform very similarly to the traditional virtual bolus technique (VB) [1]. However, median target dose coverages and mean skin dose were evaluated to be slightly lower with ASF and in some cases the performance was lower than with VB [1]. The aim of this study was to evaluate the updated ASF version (Eclipse v18.1 MR1, Varian Medical Systems, USA) in breast cancer. Material/Methods: The performance of the updated ASF (v18.1 MR1) was compared to VB and the earlier ASF version (v18.1) retrospectively on ten left-sided breast cancer patients with nodal involvement. For reference, a VMAT plan was optimized for each patient with a VB [2]. Four RAD plans were optimized with ASF. VMAT plans were optimized and calculated with v18.1 MR1 and RAD plans were optimized and calculated with versions v18.1 and v18.1 MR1. VMAT plans utilized four or five split arcs. Two RAD plans were created to mimic the VMAT plan and optimized with the ‘Arc Dominant’ setting. Two RAD plans were optimized with two or three split arcs on the ‘Balanced’ setting with four tangentially positioned static angle modulated ports. All plans were optimized with a 13 mm PTV extension anteriorly and laterally and a 16 mm optimization bolus (HU -350). Plans were optimized to comparable PTV V95%coverage. To evaluate the performance of the skin flash methods, three different structure sets were created with uniform 4, 8 and 12 mm expansions to simulate the breast swelling. The expansion was set to HU -100 to mimic breast tissue. Along with the simulated swelling, target structures with subtraction of 3 mm build-up region (CTVskin, PTVskin) were expanded. Mean skin dose (Dskin) was defined in the first 5 mm from the surface of the body. Results: The updated ASF (v18.1 MR1) provided the highest median CTVskin and PTVskinV95% coverages and Dskin of over 96.3, 94.1 and 90.0%, respectively, in all simulated swelling scenarios when used with the balanced setting (Table 1). The results with the arc dominant setting were slightly lower. With virtual bolus, median CTVskin and PTVskinV95% coverages and Dskin were over 93.1, 90.2 and 87.5%,
Made with FlippingBook - Share PDF online