ESTRO 2023 - Abstract Book

S244

Saturday 13 May

ESTRO 2023

radiotherapy errors, related to dose prescription, treatment plan parameters and setup equipment. Semi-structured interviews were conducted with radiotherapy experts to identify relevant variables and additional links to construct the BN topology. We simulated errors based on the failure modes reported in TG 275 of the American Association of Physicists in Medicine(2) and the reported errors in the three centres. The BN’s ability to identify errors at each centre was tested, after training it on data from another centre (cross-site validation). Results The structure of the BN network is presented in figure 1. The highest performance was observed for the BN when it was trained at the UW centre and validated at the UVMMC with an area under the receiver operating characteristic curve (AUC) value of 0.80. The different AUC values of each validation setting are summarised in table 1. High performance was observed for the gantry angle errors (AUC=0.84) while on the contrary, the BN did not perform adequately in the detection of the number of fractions errors when trained at UVMMC and validated at MAASTRO (AUC=0.61). Figure 1: Structure of thE BN

Conclusion We successfully investigated the efficacy of the BN in an international multi-centric setting. The BN has shown the efficacy to detect possible radiotherapy planning errors only from the same institution that was trained on. This result shows that the model is generalizable across different practices with data harmonization. However, the performance on specific components, such as prescription, could be affected due to the significant differences in treatment protocols across institutes. Future work is required in terms of the error simulation method and the data standardisation methodology for the performance improvement of the BN with the involvement of additional institutions. References MO-0309 Multicentre validation of the signal fading model of a new OSL film dosimeter L. Delombaerde 1 , M. Caprioli 1 , M. Both 2 , L. Michielsen 2 , E. Rault 3 , I. Silvestre Patallo 4 , A. Douralis 4 , I. Billas 4 , M. Hussein 4 , P. Leblans 5 , D. Vandenbroucke 5 , C.M. Leenders 5 , W. Crijns 6 1 UZ Leuven, Department of Radiotherapy-Oncology, Leuven, Belgium; 2 Erasmus MC, Radiotherapy, Rotterdam, The Netherlands; 3 Centre Oscar Lambret, Radiothérapie, Lille, France; 4 National Physical Laboratory, Medical Radiation Physics, London, United Kingdom; 5 Agfa NV, Innovation Office, Mortsel, Belgium; 6 UZ Leuven, Department of Radiation- Oncology, Leuven, Belgium Purpose or Objective Optically Stimulated Luminescence (OSL) film dosimeters display signal fading following irradiation due to spontaneous recombination of energetically unstable electron-hole traps. A power-law signal fading model for a new type of OSL film dosimeter was developed by Caprioli et al. (article accepted for publication). In this study we performed a multicentre validation of the model under different irradiation conditions. Materials and Methods Protocol: A measurement protocol was distributed to 3 participating centres containing a description of the measurement setup, the irradiation and read-out conditions. Participating centres had received a BaFBr:Eu2+ based OSL film, prior to onset of this study within the QUARTEL project (see Conflict-of-Interest). After irradiation the films were digitized (adjusted scanner model CR-15X, Agfa NV) by the participating centre. Digital images were collected and processed by the organizing centre. Measurements: Films were placed below 10 cm of solid water with SSD 90 cm and at least 10 cm of backscatter material. A total of 12 irradiations were performed by every centre with irradiation times (t_ir) between 2 and 20 mins and scanning times (t_scan) from 5 to 60 mins after onset of the delivery. Participants were free to select the delivered dose and dose rate, but were advised to cover a clinical range (2 – 30 Gy). A region-of-interest (ROI) of 1x1cm ² at the centre of the delivered field was used to calculate the mean signal. The linear fading model from Caprioli et al. is the following: S(Dr_m,t_ir,t_scan)=A x Dr_m x t_ir x(t_scan+ τ )^(-n) where the signal (S) is dependent on the dose rate Dr_m, t_ir and t_scan with A the sensitivity, τ and n OSL material specific parameters. For model validation the fading specific parameter ( τ and n) were kept constant. Only the parameter A was adjusted to account for a global sensitivity variation that can be expected for the different setups. A was scaled by the 1. PMID:30927253 2. PMID:31967655