ESTRO 2021 Abstract Book

S1329

ESTRO 2021

Purpose or Objective In dosimetry of LINAC megavoltage fields and dose calculation, the knowledge of the equivalent square field of rectangular fields is of great importance. Still today the equivalent square field of rectangular fields is calculated according to the geometric mean or the sterling area‐to‐perimeter equation, which are both purely geometrical considerations and do not take any physics of the radiation process into account. In this study we use the pencil-beam model established by Ahnesjö (Med. Phys. 1992;19:263-73) together with the extension of Nyholm (Radiother. Oncol. 2006;78:347-351) to develop a method for calculating the equivalent square field size to a given rectangular field at specified depth and TPR20,10. We compared our method with measurements and calculations according to geometric mean and sterling equation. Materials and Methods Based on Monte-Carlo data, Ahnesjö established a radial pencil-beam kernel model for mega-voltage beams: The first term corresponds to the primary dose and the second one to the scatter dose. A z , a z , B z and b z are depth z and TPR20,10 dependent fit parameters. To calculate the equivalent square field we implemented these two models in a Python program. The program integrates numerically the kernel over the rectangular respectively square area of the considered fields to calculate the dose in the middle of the field. To get the equivalent square field of a considered rectangular one the dose for the rectangular field is calculated and compared to doses of square fields. The square field with the smallest deviation in dose is picked as equivalent square. For comparison we measured the dimensions and doses of several square and rectangular fields with an Exradin W2 1x3 scintillator in a PTW water-tank using an ELEKTA Versa HD LINAC at 6MV and 18MV. The dimensions of the square fields ranging from 0.5cm to 2cm with intervals of 1mm. The rectangular fields were measured for two series, with one fixed side at 0.5cm and 1cm respectively and the other side ranging from 0.5cm up to 5cm. Results Figure 1 shows the difference between measured and calculated equivalent length for our method, the geometric mean and the sterling equation. The measurement uncertainty is approximately 0.5mm. Our model fitted the measured data best, followed by the sterling equation. Both geometric methods fail especially at highly unsquared fields. Between the two energies, we found differences only for very elongated fields.

Conclusion For rectangular fields with nearly squared dimensions the common sterling equation may still be used. The geometrical mean is not an accurate approach. For highest accuracy, especially for elongated rectangular fields we recommend our new model, considering even the small depth and TPR20,10 dependence. We will publish our program with a simple to use graphical interface with versions for Linux, OSX and Windows. PO-1609 Clinical validation of a GPU-based MC dose engine of a commercial TPS for PBS proton therapy F. Fracchiolla 1 , E. Engwall 2 , M. Janson 2 , F. Tamm 3 , S. Lorentini 1 , F. Fellin 1 , M. Bertolini 1 , C. Algranati 1 , R. Righetto 1 , P. Farace 1 , M. Amichetti 1 , M. Schwarz 1,4 1 Centro di Protonterapia, Trento Hospital, Trento, Italy; 2 RaySearch Laboratories AB, RaySearch , Stockholm, Sweden; 3 RaySearch Laboratories AB, RaySearch, Stockholm, Sweden; 4 TIFPA , Trento Institute for Fundamental Physics and Applications, Trento, Italy Purpose or Objective To perform the validation of the GPU-based (Graphical Processing Unit based) proton Monte Carlo (MC) dose engine implemented in a commercial TPS and to report dose calculation times for clinical cases. Materials and Methods 440 patients treated at our Proton Therapy Center between 2018 and 2019 were selected for this study. 636