ESTRO 2025 - Abstract Book

S3042

Physics - Image acquisition and processing

ESTRO 2025

Theoretical Physics, Trieste, Italy. 5 Medical Physics Department, IRCCS San Raffaele Scientific Institute, Milano, Italy. 6 Radiology Unit, IRCCS Policlinico San Donato, San Donato Milanese, Italy Purpose/Objective: To evaluate the potential benefits of Photon-Counting CT (PC-CT) in radiotherapy (RT), examining its impact on image quality, artifact reduction, and dose efficiency compared to conventional Energy-integrating CT (EICT). Material/Methods: The CIRS and Catphan phantoms were used to analyse performances in contrast quantification and image quality. A metal prosthesis was applied to an anthropomorphic pelvic phantom and a water-filled cylinder to assess metal artifact reduction. Additionally, a home-made soft tissue phantom—a melon with an embedded hip prosthesis—was used to simulate realistic tissue density and assess artifact reduction in a practical RT workflow. Both conventional EICT (Philips Big Bore) and PC-CT (Siemens NAEOTOM Alpha®) imaging were performed, with protocols optimized for artifact suppression, noise control, and spatial resolution. Regions of interest (ROIs) were used to analyze Hounsfield Units (HU), noise levels, and dose-volume histogram (DVH) data, focusing on contrast media and metal artifacts influence. Furthermore, image quality was investigated in terms of task specific Spatial Resolution, Noise Power Spectrum and detectability index (D’), considering the visibility of lesions characterised by sizes and contrast meaningful for the head and abdominal region. Results: PC-CT demonstrated substantial improvements in metal artifact suppression, reducing distortions near metallic implants by up to 30%, with significantly improved soft tissue visualization in the melon phantom. Quantitative analysis revealed up to a 49% reduction in noise variance and up to 15% enhancement in spatial resolution. In the CIRS phantom, contrast quantification improved with more accurate and stable HU values across all concentrations, with a maximum difference of 90 HU out of 490 for a concentration of 20mg/mL. For D', PC-CT achieved an improvement of up to 55%, demonstrating superior lesion visibility in critical anatomical regions. DVH and monitor unit (MU) analyses confirmed comparable target coverage and dose distribution efficiency between PC-CT and EICT, even in challenging conditions with metal implants. Conclusion: This study highlights the potential of PC-CT in RT, offering significant advancements in target delineation accuracy, metallic artifact reduction, and overall image quality. The ability to achieve more precise anatomical characterization and reliable dose distribution lays the foundation for improved treatment planning and execution. Future studies will include PC-CT ability of direct electronic density reconstruction to analyse further uncertainties reduction. Furthermore, a dose measurement in the melon-phantom could test dose accuracy in treatment planning. These preliminary findings suggest that integrating PC-CT into RT workflows could significantly enhance treatment safety and precision, warranting further clinical exploration for optimizing RT practices.

Keywords: PC-CT, Treatment Planning, Metal Artifacts

3368

Digital Poster Development of a Python library for image and segmentation conversion and application for radiomic feature comparison Gabriele Palazzo 1 , Alfonso Belardo 1 , Maria Giulia Ubeira-Gabellini 1 , Emiliano Spezi 2 , Antonella del Vecchio 1 , Claudio Fiorino 1 1 Medical Physics, IRCCS San Raffaele Scientific Institute, Milan, Italy. 2 School of Engineering, Cardiff University, Cardiff, United Kingdom