ESTRO 2024 - Abstract Book

S3649

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

ESTRO 2024

For prostate plans the dose coverage for PTV was also in average over the 20 patients considered slightly better for the automatically generated plans, PTV D 98% = 97.1 ± 0.3% vs 96.9 ± 0.4%. All evaluated doses to rectum (V 90% , V 77% , V 50% and mean dose) were slightly lower except for D 2% that was slightly higher 99 ± 3,3% vs 98.7 ± 2.1%. The automatically generated plans once again tended to be a little more complex with a higher total number of Monitor Units, 1633 ± 538 MU vs 1268 ± 365 MU. Unfortunately, the possibility to optimize on the number of MU is not yet available in the script function.

Conclusion:

The dose planning process can be successfully automated from approved structure delineation to final plan in a much shorter time than before and without human input. The same process was applied for palliative treatments (without RapidPlan model). Since the script is fully automated and performs all tasks without external triggers, the goal going forward is to run the script overnight for all diagnoses. The computer capacity can then be optimized and extra optimization parameters like Multi Criteria Optimization can be used without slowing down or taking extra resources for the human dose planners. We also want to extend the number of diagnoses receiving automatically generated plans, but we first need to assess the juridical questions that are of concern when using such a script in a clinical environment.

Keywords: automatic planning, eclipse scripting, rapidplan

2354

Mini-Oral

Automated, collision-free beam angle determination for oligo-metastatic VMAT treatments.

Geert De Kerf 1 , Michaël Claessens 1 , Joachim Marichal 1 , Dirk Verellen 1,2

1 Iridium Netwerk, Radiotherapy, Wilrijk, Belgium. 2 University of Antwerp, Faculty of Medicine and Health Sciences, Wilrijk, Belgium

Purpose/Objective:

Despite the focus on beam angle optimization for IMRT treatments, automatic determination of Volumetric Modulated Arc Therapy (VMAT) beam angles were not much investigated, certainly in the case of oligo-metastatic treatments. The latter is characterized by varying target locations, which may benefit from automatic beam angle determination in complement to an automated treatment planning workflow.

Material/Methods:

Automatic VMAT beam angles were determined for 595 oligo-metastatic treatments. Angles were determined by mapping the tumour location on a patient dependent grid and the location of the tumour in the grid was linked to pre-defined start and stop angles (Figure 1). In case a tumour is mapped on different grid voxels, beam angles were accumulated. After arc angle determination, a collision check was performed and the arc range was updated to avoid patient collisions. For every patient, the clinically used beam angles were compared against the