ESTRO 2025 - Abstract Book

S2616

Physics - Detectors, dose measurement and phantoms

ESTRO 2025

Conclusion: Both AAA and AXB model superficial doses well when dose delivered directly to the optics and calculate scattered radiation acceptably. AXB is more accurate with a 1mm grid size when calculating dose from scattered radiation.

Keywords: superficial dose, Acuros XB algorithm, TLD

References: Han T, Mourtada F, Kisling K, et al. Experimental validation of deterministic Acuros XB algorithm for IMRT and VMAT dose calculations with the Radiological Physics Center's head and neck phantom. Med Phys. 2012;39(4):2193-2202. Kan MW, Leung LH, Yu PK. Verification and dosimetric impact of Acuros XB algorithm on intensity modulated stereotactic radiotherapy for locally persistent nasopharyngeal carcinoma. Med Phys. 2012;39(8):4705-4714. Harnett AN, Hungerford JL, Lambert GD, et al. Improved external beam radiotherapy for the treatment of retinoblastoma. Br J Radiol. 1987;60(716):753-760.

2826

Poster Discussion Development and validation of a robust dataset using commercial TPS and radiochromic films for deep learning in transit dosimetry Carlotta Mozzi 1 , Lorenzo Marini 2,3 , Michele Avanzo 4 , Aafke Kraan 3 , Francesca Lizzi 3 , Livia Marrazzo 1,5 , Icro Meattini 1,6 , Alessandra Retico 3 , Cinzia Talamonti 1,5,7 1 Department of Experimental and Clinical Biomedical Sciences “Mario Serio”, University of Florence, Florence, Italy. 2 Computer Science, University of Pisa, Pisa, Italy. 3 Istituto Nazionale di Fisica Nucleare, INFN, Pisa, Italy. 4 Centro di Riferimento Oncologico, CRO, Aviano, Italy. 5 Medical Physics Unit, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy. 6 Radiation Oncology Unit, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy. 7 Istituto Nazionale di Fisica Nucleare, INFN, Florence, Italy Purpose/Objective: The aim of this study is to validate Monte Carlo dose calculation algorithm within Monaco treatment planning system (TPS) by comparing it with radiochromic film dosimetry. This validation underpins radiotherapy patient transit dosimetry using Electronic Portal Imaging Device (EPID). Upon successful validation, a robust dataset was developed to train a deep learning (DL)-based supervised model, enabling the conversion of EPID responses into portal dose (PD) images. This approach facilitates the implementation of transit dosimetry in radiotherapy [1][2]. Material/Methods: Transit dosimetry by EPID was simulated by calculating dose distribution at EPID level, positioned 60 cm from isocenter, using Monaco TPS, widely utilized in clinical environments. For EPID geometry simulation in the TPS, the calculation area was extended beyond the computed tomography scan to reach EPID level. The EPID panel was simulated as a geometric structure with a thickness of 4.4 cm and a relative electron density equal to 1. Exported dose map was centrally positioned. To validate these simulations, Gafchromic (GAF) EBT3 films were placed at EPID level to compare experimental dose measurements with the simulated TPS data [3][4]. Experimental setup was constructed using water-equivalent thickness in place of the EPID detector (Fig.1). To date, 5 GAF films have been used to verify the agreement between measured and calculated dose distributions. Homogeneous and multi-plug phantoms containing insertions of titanium, bone, solid water, and lung were irradiated using beams with various geometries and dose levels. Once the TPS response was confirmed to be accurate, a robust dataset was constructed. Over 200 EPID images, paired with their corresponding simulated dose images, were generated. These datasets incorporated a wide range of phantom types and beam geometries to support DL training models.