ESTRO 2025 - Abstract Book

S2649

Physics - Detectors, dose measurement and phantoms

ESTRO 2025

available software. Further PSQA workflow tests considering a larger sample of clinical carbon ion treatment plans will be conducted.

Keywords: particle therapy, PSQA, detector

3779

Digital Poster CT number stability for HyperSight cone-beam CT on seven TrueBeam accelerators Lone Hoffmann 1,2 , Mai-Britt Linaa 1 , Camilla Byskov 1 , Rune Hansen 1 , Anne IS Holm 1 , Morten B Jensen 1 , Ditte S Møller 1,2 , Vicki T Taasti 2,3 1 Department of Oncology, Aarhus University Hospital, Aarhus, Denmark. 2 Department of Clinical Medicine, Aarhus University, Aarhus, Denmark. 3 Danish Center for Particle Therapy, Aarhus University Hospital, Aarhus, Denmark Purpose/Objective: The daily delivered dose may be calculated based on the cone beam-CT (CBCT) images and may be used to determine if treatment adaptation is needed. This study aimed to evaluate the stability of CT numbers for HyperSight CBCT imaging at seven TrueBeam accelerators. Material/Methods: The CBCT systems at seven identical TrueBeam accelerators were upgraded to HyperSight. For each accelerator, CBCT images of the Gammex AED phantom were acquired using the Head-and-Neck (HN, 100kVp, 150mAs, half trajectory, iterative reconstruction (iCBCT)), Thorax (125kVp, 265mAs, full trajectory, Feldkamp-Davis-Kress (FDK) reconstruction), and Pelvis (125kVp, 1061mAs, full trajectory, iCBCT) modes. For the HN mode, the inner cylindrical phantom (diameter=20cm) was used, while for the thorax and pelvis modes, the full phantom (height=30cm, width=40cm) was used. The phantom length was 16cm, which was extended by 15cm of solid water at each end, and the CBCT scan length was collimated to 20cm. Consecutive CBCT scans were acquired while changing the central insert between fourteen density inserts (0.3-1.9 g/cm 3 ). For two CBCT systems, the CBCT scans were repeated three times within a period of 6 weeks. For one CBCT system, the effect of collimation was evaluated based on the full phantom using collimation widths of 8, 10, 16, and 25cm. Additionally, consecutive scans of the Gammex phantom with the solid water insert were performed. Results: The CT number variation of the ten measurements on the seven CBCT systems was compared to the variation seen for nine conventional TrueBeam CBCT systems [1], see Figure 1. The variation was much smaller for the new HyperSight systems, especially when applying iCBCT reconstruction. For consecutive measurements of the solid water insert, a CT number variation of approximately 5 HU was seen. However, comparing the CT numbers to the fan-beam CT used for treatment planning, it is still apparent that dedicated conversion curves are needed for CBCT to ensure accurate dose calculation. The low variance over the individual HyperSight CBCT systems allows for a single conversion curve shared between the CBCT systems. The effect of the collimation width was most pronounced for FDK reconstruction, while only a slight CT number variation was seen for iterative reconstruction. Conclusion: CBCT phantom scans at seven TrueBeam accelerators with HyperSight CBCT systems showed small inter- and intra CBCT-imager variations. No time-dependent difference was observed. This allows for a single CT-number-to relative-electron-density conversion curve for each CBCT mode (HN, thorax, pelvis) to be used for all HyperSight CBCT systems, easing daily dose re-calculations.