S1834 Gy. A repeated - measures analysis showed a significant effect of skin-protection on dose fall - off (p<0,01). Pairwise comparison with Bonferroni correction revealed a significant difference between the two plan variations with- and without - skin - protection, with higher dose fall - off values in PTV_Boost without skin- protection, M ₍ diff ₎ = –0,64, SE = 0,18, p = 0,006, 95%-CI [–1,04, –0,23]. No significant effect was observed across dose levels (p>0,05). Both the heterogeneity index (HI_PTV_Boost) and conformity index (CI_PTV_Boost) differed significantly across dose levels for both plan variants (Friedman-test, p<0,05), and between the two plan variants with- and without - skin - protection (Wilcoxon-test, p<0,05). The dose distribution shows higher HI and CI in PTV_Boost in plans without - skin - protection and lower boost dose levels. Conclusion: These results indicate that dose escalation and skin- protection affect dose homogeneity and conformity within the boost volume. Here, we demonstrate adequate dose fall-off beyond tumor margins in the planned tumor bed while maintaining high dose heterogeneity and conformity in plans without-skin- protection for PTV_Boost. Further treatment planning could focus on optimization of high dose fall-off in the intended surgical tumor bed to evaluate side-effects. However, prospective studies confirming the clinical relevance of dose fall-off are currently missing. Moreover, it remains debated whether optimizing dose distribution or minimizing acute and late effects to reduce wound complications and fibrosis should be prioritised. Keywords: dose fall-off, sarcoma, ultrahypofractionation Digital Poster 1452 First clinical evaluation of automatic exposure control (AEC) for HyperSight CBCT Helen Grimes, Miranda Frizzelle Radiotherapy physics, University College London NHS Trust, London, United Kingdom Purpose/Objective: In March 2025, a Varian Halcyon linac with HyperSight imaging was commissioned for clinical use. HyperSight delivers superior CBCT image quality and HU accuracy compared with standard systems and incorporates Automatic Exposure Control (AEC) via Auto mAs to adapt dose to patient size.While previous studies have highlighted its image quality and dose-calculation potential, AEC performance has not been evaluated. This study commissioned and assessed a pelvis CBCT protocol using Auto mAs
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Material/Methods: The standard pelvis protocol for HyperSight uses 125 kV and 470 mAs. Auto mAs determines patient-specific exposure using:mAs(D)=mAs(D_0)e^( μ (D-D_0))where mAs(D ₀ ) is the reference exposure for a reference water-equivalent diameter D ₀ , μ is the attenuation coefficient, and D is the patient’s maximum water- equivalent diameter across the CBCT field, derived from the planning CT. Both D ₀ and μ are fixed system parameters; mAs(D ₀ ) is user-defined.RTTs confirmed that HyperSight image quality was clinically acceptable, so the aim was to standardise this quality level. As no manufacturer method existed for estimating D, the reference mAs was determined empirically. Twenty pelvis patients were reviewed to establish the average patient width, and a representative dataset was used to iteratively adjust mAs(D ₀ ) until Auto mAs reproduced 470 mAs.The protocol was evaluated on 13 patients to determine the predicted mAs required for standardised image quality. Imaging was performed using the standard pelvis protocol, RTTs rating image quality as ‘better than required’, ‘optimised’, ‘acceptable but not optimal’, and ‘poor’. Image quality from the standard mAs was compared with the mAs predicted by AEC, to verify predicted values were consistent and clinically appropriate prior to implementation. Results: The reference mAs was determined as 360 mAs, corresponding to an average patient diameter of 38 cm. Table 1 summarises predicted mAs values and RTT image-quality assessments.
For cases rated better than required, Auto mAs predicted lower exposures, with one case showing a 48% reduction. For acceptable but not optimal cases, Auto mAs predicted higher mAs, indicating potential image-quality improvement.Figure 1 demonstrates the variability in image quality observed with fixed-mAs protocols
Conclusion: This is the first study showing that Auto mAs for HyperSight CBCT can optimise imaging dose and standardise image quality. An AEC CBCT protocol was commissioned for clinical use to enable patient- specific dose optimisation, comparable to CT, resulting in more consistent image quality for patient cohorts.
Made with FlippingBook - Share PDF online