S1812
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
10.1016/j.radonc.2023.109675.[2] Nakao M, et al. Tolerance levels of mass density for CT number calibration in photon radiation therapy. J Appl Clin Med Phys 2019;20:45–52. DOI: 10.1002/acm2.12601.We thank all participating centres for facilitating the site visits, for their time and effort, and for sharing the data. We thank Kenneth Ruchala (Sun Nuclear) for kindly providing the phantom. Keywords: Direct Density, CT scanner Digital Poster 881 Validation of a Python-Integrated Module for X-ray Beam Characterization in PenRed Verónica Ribes 1 , Sandra Oliver 2 , Belén Juste 2 , Gumersindo Verdú 2 , Vicent Giménez 3 1 Departamento de Sistemas Informáticos y Computación, Universitat Politècnica de València, València, Spain. 2 Instituto de Seguridad Industrial Radiofísica y Medioambiental (ISIRYM), Universitat Politècnica de València, València, Spain. 3 Dpto. de Estadística e Investigación Operativa Aplicadas y Calidad, Universitat Politècnica de València, València, Spain Purpose/Objective: Accurately characterizing X-ray beams is a complex challenge, as existing tools are often limited. While they can determine the energy spectrum for simple configurations, they typically fail to capture the full spatial distribution of the beam, omitting critical non- uniform features like the heel effect. The core of the problem lies in the difficulty of modeling intricate physical processes, such as electron transport within the anode and the significant influence of the tube's specific geometry.Although Monte Carlo (MC) simulation is a powerful method capable of providing a complete and detailed model, its practical use is hindered by significant barriers. These simulations demand long computation times and a highly complex setup, which becomes even more pronounced when modeling beam interaction with objects or patients. Furthermore, they require deep expertise not only in the underlying physics but also in the specific simulation code and its parameters.To facilitate the use of MC simulations for this purpose, two utilities have been implemented in PenRed. Material/Methods: The implemented utilities are:- anodeSim: Simulates an electron beam impinging on an anode without a filter and records the resulting photon spectrum and spatial distribution.- deviceSim: Simulates an electron beam impinging on an anode and records the resulting photon spectrum and spatial distribution after the beam passes through the tube filters.Leveraging the novel pyPenred module, which
The maximum difference between clinical and guide- generated LUT for non-bone tissues was 4.4% while for bone it reached 24% (Figure 2). The latter, larger differences can be expected since institutions either extrapolate based on bone tissue equivalent inserts or interpolate based on metal inserts in this region. A systematic positive difference for bone tissues (median 1.1% and (1Q -1.0%; 3Q +10.3%)) is explained by institutions using only the body phantom setup measurements, contrary to the guide’s recommendation to average out body and head data. Tolerance levels [2] for RED and MD differences were met for the lung, adipose, and soft tissue regions (4.4%, 2.2%, and 2.2%), but differences exceeding the constraint were observed in bone (4.4%). Despite the expected limited therapeutic photon dose impact due to the small thickness of cortical bone overall in the human body, these high differences should be reduced. Further investigation using patient data is needed.
Conclusion: Standardization of DD LUT generation reduced variation for comparable CT-software configurations. The validity of clinically implemented LUT must be checked with DD kernel improvements. References: [1] Peters N, et al. Consensus guide on CT-based prediction of stopping-power ratio using a Hounsfield look-up table for proton therapy. Radiother Oncol 2023;184:109675. DOI:
Made with FlippingBook - Share PDF online