S2368
Physics - Quality assurance and auditing
ESTRO 2026
Digital Poster 2834 A novel methodology for optimizing single- isocentre multi-target stereotactic radiosurgery quality assurance procedures Michael Reiner 1 , Daniela Eulenstein 2 , Philipp Freislederer 3 , Claus Belka 1,4 1 Department of Radiation Oncology, University Hospital, LMU Munich, Munich, Germany. 2 PTW Freiburg, PTW Freiburg, Freiburg, Germany. 3 Brainlab SE, Brainlab SE, Munich, Germany. 4 Bavarian Cancer Research Center (BZKF), Bavarian Cancer Research Center (BZKF), Munich, Germany Purpose/Objective: Regardless of patient immobilization, relevant intrafractional motion occurs during single-isocentre multi-target stereotactic radiosurgery (SIMT-SRS), which may have a significant impact on the desired dose distribution. To improve QA routines for SIMT- SRS a novel methodology combining film and point dose measurements was established detecting geometric misalignments less than 1 mm which is the institutional PTV margin. Material/Methods: All measurements were conducted using the RUBY head phantom (PTW Freiburg, Germany) with two different inserts, either for a film in coronal plane or point dose detectors at three different positions. Gafchromic EBT4 films (Ashland, USA) and a microDiamond TN60019 (PTW Freiburg, Germany) were used as detectors. Two non-coplaner SIMT-SRS plans with different target diameters were generated with Elements Multiple Brain Mets SRS (Brainlab SE, Germany) and irradiated to the phantom on a VersaHD linac (Elekta AB, Sweden). X-ray imaging with ExacTrac Dynamic (Brainlab SE, Germany) at all couch angles was used to position the phantom.Simultaneous shifts in x-, y- and z-direction of 1 and 2 mm were introduced (distance to original isocenter 1.7 mm and 3.5 mm, respectively) and measured point doses were compared to the calculated Dmean in the active detector volume. For the film measurements, shifts of 1, 2 or 3 mm in only one direction (either lateral or longitudinal) were applied. Films were scanned, calibrated und compared to the corresponding coronal dose plane of the DICOM dose cube using FilmAnalyze and Verisoft (PTW Freiburg, Germany). Results: Results for the point dose measurements are found in the table. Taking into account our commissioning data dose deviations up to 3 % are considered inconspicuous. Larger deviations (red) are only found for the smallest target for the 1.7 mm shift; for a 3.5 mm shift also larger targets show significant dose deviations. In contrast films were able to detect the
correctly identify all suitable sCTs, and 95.7% specificity for detecting unacceptable sCTs(fig2a). Performance remained >95% balanced accuracy when changing the verification method to CBCTcorr from sCT-DL, confirming the method-agnostic behavior of the approach(fig2b). The few false positives (sCT- DL:n=1; CBCTcorr:n=2) had GPR slightly below the 95% target (lowest ≈ 93.3%). Total evaluation time was <1- minute for both metrics.
Conclusion: We presented a simple, fast, method-independent QA- pipeline that converts CBCT-based sCT use in PT into an objective accept/reject step for plan re-
optimization. References:
[1] Håkansson K, Giannoulis E, Lindegaard A, Friborg J, Vogelius I. CBCT-based online adaptive radiotherapy for head and neck cancer–dosimetric evaluation of first clinical experience. Acta Oncologica 2023;62:1369–74. https://doi.org/10.1080/0284186X.2023.2256966.[2] Galapon A, Wagenaar D, Langendijk JA, Both S. Transfer learning for DL-based Synthetic CT after reconstruction algorithm upgrade in a proton therapy clinic. Med Phys 2025;52. https://doi.org/10.1002/MP.17937. Keywords: adaptive PT therapy,quality assurance,synthetic ct
Made with FlippingBook - Share PDF online