ESTRO 2025 - Abstract Book

S4382

RTT - Treatment planning, OAR and target definitions

ESTRO 2025

Conclusion: The quality of AI-contoured OARs for the RTT-only workflow for bladder and cervical cancer radiotherapy using the Ethos 2.0 Emulator demonstrates clinical applicability and is a promising starting point. However, as the HD reveals larger deviations at certain points, particularly for certain structures, visual inspection –and, where necessary, adaptation– of the contours remains essential. Future studies with larger datasets are needed to validate these findings on CBCTs and further optimize AI contouring algorithms for broader implementation in oART workflows.

Keywords: AI, CBCT-guided oART, RTT-only workflow

References: [1] Azzarouali S, et al . Radiat Oncol . 2024 Oct 7;19(1):138. doi: 10.1186/s13014-024-02526-2 [2] Åström LM, et al. Phys. Imaging Radiat. Oncol. , vol. 31, Jul. 2024, doi: 10.1016/j.phro.2024.100640 [3] Goudschaal K, et al. CtRO ., vol. 50, Jan. 2025, doi: 10.1016/j.ctro.2024.100884 [4] Sibolt P, et al. Phys Imaging Radiat Oncol . Dec. 2020;17:1-7, doi: 10.1016/j.phro.2020.12.004 [5] De Jong R, et al. Radiat Oncol. Jul. 2021;16:136, doi: 10.1186/s13014-021-01866-7 [6] Crum WR, et al. IEEE Trans Med Imaging . Nov. 2006;25(11):1451–61 doi: 10.1109/TMI.2006.880587 [7] Taha AA, et al. BMC Med Imaging . Aug. 2015;15(1):29 doi: 10.1186/s12880-015-0068-x