ESTRO 2025 - Abstract Book

S2903

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

Conclusion: This deep learning-based model offers a robust framework for predicting VMAT dose distributions in NSCLC. It has the potential as a decision-support tool to assist in optimizing radiotherapy plans and guiding treatment selection between concurrent chemoradiotherapy and standalone chemotherapy for large tumors, improving treatment efficiency and quality.

Keywords: Dose Prediction, Deep Learning, Lung Cancer References:

[1] Barragan-Montero A, Nguyen D, Lu W, Lin M, Norouzi-Kandalan R, et al. Three ‐ dimensional dose prediction for lung IMRT patients with deep neural networks: robust learning from heterogeneous beam configurations. Medical physics, (2019), 46(8), 3679-3691. [2] Shao Y, Zhang X, Wu G, Gu Q, Wang J, et al. Prediction of Three-Dimensional Radiotherapy Optimal Dose Distributions for Lung Cancer Patients with Asymmetric Network. IEEE Journal of Biomedical and Health Informatics, (2021), 1120-1127, 25(4).

3880

Digital Poster Analytical Prediction of V10Gy for SRS Plans in Brain Metastases: Volume- and Asymmetry-Based Plan Evaluation Tuba Tekin 1,2 , Lara A. Caglayan 1 , Patrick Eich 1 , Stephan Garbe 1 , Christina Leitzen 1 , Davide Scafa 1 , Gustavo Sarria 1 , Mümtaz A. Köksal 1 , Julian Layer 1 , Cas Dejonckheere 1 , Shari Wiegreffe 1 , Fabian Kugel 1 , Thomas Müdder 1 , Motaz Hamed 3 , Hartmut Vatter 3 , Ulrich Herrlinger 4 , Eleni Gkika 1 , Youness Nour 1 1 Department of Radiation Oncology, University Hospital, Bonn, Germany. 2 University Clinic for Medical Radiation Physics, Medical Campus Pius Hospital, Oldenburg, Germany. 3 Department of Neurosurgery, University Hospital, Bonn, Germany. 4 Department of Neurooncology, University Hospital, Bonn, Germany Purpose/Objective: The aim of this study was to develop an analytical solution to support plan evaluation by considering individual parameters and geometric characteristics of lesions for predicting V10Gy in linac-based SRS plans. By analyzing V10Gy