ESTRO 2022 - Abstract Book

S395

Abstract book

ESTRO 2022

Conclusion PT reduces dose to OAR compared to RT and may enable further dose escalation to the target. All NSCLC patients with heterogeneous dose distributions were model-based selected for PT. However, the patients benefiting the most from PT presented with left-sided tumors at the heart level.

OC-0451 Comparative of different dose prediction and robust mimicking strategies for automatic IMPT planning

E. Borderias-Villarroel 1 , M. Huet-Dastarac 1 , A.M. Barragán-Montero 1 , M. Holmstrom 2 , X. Geets 1,3 , E. Sterpin 4,5

1 UCLouvain, MIRO, Brussels, Belgium; 2 RaySearch Laboratories, Machine Learning, Stockholm, Sweden; 3 Cliniques Universitaires Saint Luc, Radiation therapy , Brussels, Belgium; 4 UCLovain, MIRO, Brussels, Belgium; 5 KULeuven, Laboratory of Experimental Radiotherapy, Department of Oncology, Leuven, Belgium Purpose or Objective Automatic planning based on dose prediction with deep learning followed by dose mimicking is becoming popular. The term dose mimicking refers to the process of finding, through conventional optimization algorithms, the machine parameters that deliver a reference three-dimensional dose distribution. In proton therapy, dose mimicking is even more challenging, because the optimization algorithm needs to both accurately reproduce the dose distribution and the associated robustness against uncertainties. This study compares the performance of three different automatic planning workflows: (1) our in- house prediction and mimicking solutions (IH = IHpred-IHmim), (2) the ones from the commercial treatment planning system