ESTRO 2025 - Abstract Book

S2946

Physics - Image acquisition and processing

ESTRO 2025

81

Digital Poster Value of the Acorus reconstruction for conventional (non-HyperSight) Ethos systems.

Anieck E ter Braak 1 , Erik B van Dieren 2 , Sven Schwieters 1 , Esther Bosman 1 , Jeroen Veltman 3 , Judith Dasselaar 2 1 faculty of science and technology, University of Twente, Enschede, Netherlands. 2 Radiotherapy, Medisch Spectrum Twente, Enschede, Netherlands. 3 Multi-Modality Medical Imaging, University of Twente, Enschede, Netherlands Purpose/Objective: Since the introduction of CBCT in the field of radiotherapy, there has been an increasing evidence that adapting to anatomy changes during treatment may improve effectiveness of radiotherapy (1). Since 2019, CBCT based online adaptive radiotherapy has become a viable option for several anatomic sites, but low contrast due to scatter impairs easy implementation. The Acuros algorithm reconstruction for CBCTs (2) recently became available, also for non-HyperSight systems (NHES). This study examines Acuros' value for phantom and clinical pelvis CBCTs. Material/Methods: Image quality of two NHES similar systems was studied. In the first part, bi-weekly CBCT scans were made using the Kyoto phantom PBU-60 (3), in pelvis mode, for 14 weeks. CBCTs were matched to a Kyoto CT and its CT contours were propagated. Using these CBCTs, SNR, CNR, uniformity, and resolution were calculated. In the second part of the study, 53 patients treated with online adaptive radiotherapy for the pelvis were used, where the 2nd (IGRT) CBCT, just before beam on, was reconstructed using Acuros. For five patients, and 3 treatment sessions were selected for quantitative analysis, performed similarly to the phantom study. In addition, seven observers were asked to score preference for CBCT for adaptive tasks. Results: For the phantom study, the results shows that the Acuros reconstruction is superior to iCBCT in terms of contrast to noise (fig 1), signal to noise (0 (similar) to 30 better), and uniformity (5 – 6 better). In terms of spatial resolution, no clear distinction was seen. The CBCT quality of two NHES systems is fairly comparable and constant over time.

fig 1 : bi-weekly scans, low contrast detectability for iterative and Acuros for Ethos systems Variac and Soliac.

For the clinical data, Acuros contrast to noise was usually superior (fig 2). Similar results were seen for uniformity (st.dev. in Acuros HU half that of iCBCT).