ESTRO 2026 - Abstract Book PART II

S2073

Physics - Image acquisition and processing

ESTRO 2026

Digital Poster 3347 Quantitative and Clinical Evaluation of HyperSight CBCT Maria João Cardoso 1,2 , Nicolle Gomes 1 , Maria Parente 3 , Ana Soares 1 , Joao Dias 1 , Adriana Martins 1 , Ana Margarida Furtado 1 , Ana Maria Maximiano 1 , João Costa 1 , Beatriz Nunes 1 , Bruno Vieira 1 , Carlo Greco 1 1 Radiotherapy, Fundação Champalimaud, Lisboa, Portugal. 2 Radioterapia, MercuriusHealth, Lisboa, Portugal. 3 ISEL, Politecnico de Lisboa, Lisboa, Portugal Purpose/Objective: Whereas CBCT is fundamental to image-guided radiation therapy, conventional systems often suffer from scatter, noise and metal artifacts that limit accuracy. The HyperSight CBCT was designed for faster image acquisition, improved detector technology, and advanced reconstruction algorithms. This study aims to analyse image quality, consistency and clinical potential of the HyperSight system relative to prior CBCT technologies through:standardized tests to evaluate image quality; evaluation of clinical CBCTs;auto-segmentation of CBCT images;metal artifacts impact. Material/Methods: A Catphan 604 phantom was used to acquire 35 CBCT scans in different modes (Head - 100kV, Pelvis and Spotlight – 125kV) on two systems, one with HyperSight technology (Edge2_HyS) and another without (Edge1), following adequate calibration procedures. Image quality parameters were quantitatively assessed using the MyQA software (IBA).Structured surveys were administered to 8 radiotherapy professionals, to qualitatively evaluate 60 clinical CBCT images from 30 patients. Pelvic CBCT images were further processed using Limbus-AI software to evaluate automatically generated volumes of interest contours. MAR reconstruction available in HyperSight CBCT was used in patients with metal prosthesis. Results: Improvements were observed in the standardized image quality for the HyperSight CBCT compared to the conventional CBCT system. (Figure1)HyperSight CBCT met all AAPM TG-142 and specification standards. HU constancy was maintained within ±30 HU for 100 kV scans, ±40 HU for 125 kV protocols, and ±20 HU for water, except for Bone50%, with a maximum point deviation of 57.6 HU. Conventional CBCT exhibited HU constancy generally within ±50 HU, maximum deviation of 244 HU and a mean of maximum deviation of 78 ± 62 HU.Main results show an overall uniformity increase of 2% and a contrast to noise ratio improvement from 6 to 9. Pelvis CBCT with MAR reconstruction presented the best performance in HU constancy and contrast .

Conclusion: HS provides improved image quality compared to standard C-arm LINAC CBCT, with quantitative findings consistent with previous report on Halcyon HS system [2]. The enhancements achieved through updated hardware and reconstruction algorithms support its potential integration into future adaptive radiotherapy workflows. References: [1] Sijtsema ND, Penninkhof JJ, van de Schoot AJAJ, Kunnen B, Sluijter JH, van de Pol M, et al. Dose calculation accuracy of a new high-performance ring- gantry CBCT imaging system for prostate and lung cancer patients. Radiotherapy and Oncology. 2025;202: 110596. doi:10.1016/J.RADONC.2024.110596[2] Kim E, Park YK, Zhao T, Laugeman E, Zhao XN, Hao Y, et al. Image quality characterization of an ultra-high-speed kilovoltage cone-beam computed tomography imaging system on an O-ring linear accelerator. J Appl Clin Med Phys. 2024;25: e14337. doi:10.1002/ACM2.14337[3] Kerns JR. Pylinac: Image analysis for routine quality assurance in radiotherapy. J Open Source Softw. 2023;8: 6001. doi:10.21105/JOSS.06001 Keywords: Hypersight, TrueBeam, Image Quality

Made with FlippingBook - Share PDF online