ESTRO 2024 - Abstract Book

S4926

Physics - Quality assurance and auditing

ESTRO 2024

2. Huq MS, Fraass BA, Dunscombe PB, et al. The report of Task Group 100 of the AAPM: Application of risk analysis methods to radiation therapy quality management. Medical Physics. 2016;43(7):4209-4262.

2928

Proffered Paper

Dosimetric impact of deviations in machine delivery parameters.

Sergei Diuzhenko 1,2 , Andreas Smolders 3,4 , Antony John Lomax 3,4 , Damien Charles Weber 3,5,6 , Francesca Albertini 3 , Kristjan Anderle 1 1 Cosylab, Medical Services, Ljubljana, Slovenia. 2 University of Ljubljana, Faculty of mathematics and Physics, Ljubljana, Slovenia. 3 Paul Scherrer Institute, Center for proton therapy, Villigen, Switzerland. 4 ETH Zurich, Department of Physics, Zurich, Switzerland. 5 University Hospital Zurich, Department of Radiation Oncology, Zurich, Switzerland. 6 Inselspital, Bern University Hospital, Department of Radiation Oncology, Bern, Switzerland

Purpose/Objective:

One of the sources of error in particle therapy (PT) is the deviation of machine delivery parameters from planned values, due to dynamic beam delivery effects as well as random noise in accelerator beamline components and inaccuracies in patient positioning equipment. It's therefore necessary to examine whether the impact of these errors should and can be minimized, particularly in the context of online adaptive PT where other major source of uncertainties (e.g., anatomical variation) are expected to become less dominant in the overall uncertainty budget. In this work we aim to characterize the impact of different machine errors and whether the errors are systematic.

Material/Methods:

1. Dosimetric impact of deviations.

For all the fractions the delivery log-files of 20 patient cases, treated at Paul Scherrer Institute (PSI) Gantry 2 with proton therapy in 2022 were analyzed. Patients were selected to cover typical indications treated with PT, and different treatment plan modality, inclduing Single Field Optimized (SFO) and Intensity Modulated Proton therapy (IMPT) Plans. The nominal dose distribution was compared for each fraction with the dose distributions recalculated from the delivery log-files on the patient anatomy. To isolate the dosimetric impact of the deviations for the specific parameters, the log-file analysis recalculation was repeated by analyzing one parameter at a time, keeping all the others to their nominal values. The following parameters were considered: spot monitor unit count (MU), spot position, gantry rotation angle, couch spatial position. The comparison of back-calculated dose distributions was done using gamma analysis (1%/1mm and 3%/3mm) with local normalization, 10% dose threshold and excluding dose in the air outside of a patient. Comparisons were made at different levels of complexity: for individual fields, for fields combined into treatment plans, and for plans combined into treatment series. Doses recalculation were performed with PSI in-house analytical ray-casting GPU based algorithm.