ESTRO 2025 - Abstract Book

S2694

Physics - Dose calculation algorithms

ESTRO 2025

Conclusion:

GDoTA effectively addresses angular dependencies without requiring per-beamlet CT interpolation. Tests conducted on a water phantom confirmed robustness across diverse geometries and beamlet configurations. These findings highlight GDoTA’s potential to enhance treatment planning accuracy and efficiency. However, lower performance near boundaries of differing densities remains a limitation. Further research is needed for addressing the current limitations and adaptivity to full treatment plan dose calculation.

Keywords: Deep Learning, PBS PT, Adaptive RT, AI

References: 1. Li, P., Wang, S., Li, T., Lu, J., HuangFu, Y., & Wang, D. (2020). A Large-Scale CT and PET/CT Dataset for Lung Cancer Diagnosis (Lung-PET-CT-Dx) 2. Souris K, Lee JA, Sterpin E. Fast multipurpose Monte Carlo simulation for proton therapy using multi- and many core CPU architectures. Med Phys. 2016 Apr;43(4):1700. PMID: 27036568.