S2384
Physics - Quality assurance and auditing
ESTRO 2026
(negative). This is then compared to the optimised ED values from the clinical plans, and plans based on mean structure ED or initial population- based ED values, whereby a pass or fail is assigned.
Oncology. 2022 Keywords: VMAT, PSQA, Complexity
Digital Poster 3565 Assessing the uncertainty of MR-Linac synthetic CT using a localized gamma approach Jessica Lye 1,2 , Li Wen Tseng 2 , Leah McDermott 1 , Reza Alinaghi Zadeh 1 , Sandra Fisher 1 , Rick Franich 2 , Sweet Ping Ng 1,3 1 Olivia Newton John Cancer Wellness and Research Centre, Austin Health, Melbourne, Australia. 2 School of Science, RMIT University, Melbourne, Australia. 3 School of Cancer Medicine, La Trobe University, Melbourne, Australia Purpose/Objective: One challenge with online adapted radiotherapy with the MR-Linac involves converting the daily MR image to a synthetic CT for dose calculation, when adapting to a change in shape of relevant structures. An unresolved question is the best method for comparing alternative ED assignment approaches for the synthetic CT, using dose calculation accuracy [1-3]. This work investigates an approach based on a localized gamma analysis to assess the dosimetric uncertainty when optimising treatments on a synthetic CT. Material/Methods: This work is a retrospective study of 15 patients treated for head&neck, brain, and upper abdominal cancer on the Unity MR-Linac (Elekta, Crawley, UK). The study assesses the sensitivity and specificity of a gamma approach to identify optimum electron density (ED) values that form the basis of the synthetic CT, using receiver operating characteristic (ROC) curves. The optimal synthetic CT was determined by the clinical quality assurance (QA) process. Medical physicists select ED values of relevant structures for the bulk density override during their reference QA process for each plan. Their assessment compares the reference and synthetic CT plans for differences in DVH curves, local dose difference maps and changes in dose reference points. The performance of gamma analysis as a classifier using 2%/2mm with either local or global dose differences was investigated in this study. Additionally, the diagnostic value of gamma analysis when restricted to individual PTV and OAR regions of interest was compared to diagnosis based on the total body gamma analysis. Results: The ROC curves in Figure 1 represent 2%/2mm gamma assessment for plans limited by (a) the skin contour and (b) PTV/OAR. Each point on the curve represents a specific gamma passing rate threshold for determining either a pass (positive) or fail
Figure 1 ROC for 2%/2mm gamma assessment and 10% threshold for (a) global gamma limited by skin contour and (b) local gamma limited by PTV and significant OAR with inhomogeneous lung and bowel excluded. Conclusion: The localised gamma approach is an accurate tool to assess the synthetic CT based on a set of optimised ED values. The gamma evaluation process for individual PTVs and OAR volumes shows improved sensitivity and specificity compared to global gamma across the entire patient skin volume. References: [1] Tang, B., Liu, M., Wang, B., Diao, P., Li, J., Feng, X., ... & Orlandini, L. C. (2022). Improving the clinical workflow of a MR-Linac by dosimetric evaluation of synthetic CT. Frontiers in Oncology, 12. [2] Cusumano D, Placidi L, Teodoli S, Boldrini L, Azario L. On the accuracy of bulk synthetic CT for MR-guided online adaptive radiotherapy. La Radiol Med (2019) 125(2):157–64 [3] O’Connor, L. M., Dowling, J. A., Choi, J. H., Martin, J., Warren-Forward, H., Richardson, H., ... & Greer, P. B. (2022). Validation of an MRI-only planning workflow for definitive pelvic radiotherapy. Radiation Oncology, 17(1), 1-11. Keywords: MR Linac, Synthetic CT, Gamma Analysis Poster Discussion 3621 Single-isocentre multi-target SRS: Errors that impact clinical acceptability Joerg Lehmann 1,2 , Lauren May 2 , Jessica Lye 3,4 , Mallory Glenn 5 , Fayz Kadeer 6 , José Antonio Baeza-Ortega 2 , Mohammad Hussein 7 , Peter B Greer 1,2 , Joel Poder 8,9 , Stephen Kry 5 , Catharine H Clark 10,11 1 Department of Radiation Oncology, Calvary Mater Hospital, Newcastle, Australia. 2 School of Information and Physical Sciences, University of Newcastle, Newcastle, Australia. 3 Department of Radiation Oncology, Olivia Newton John Cancer Wellness and Research Centre, Melbourne, Australia. 4 School of Health and Biomedical Sciences, RMIT University,
Made with FlippingBook - Share PDF online