ESTRO 2025 - Abstract Book

S2658

Physics - Detectors, dose measurement and phantoms

ESTRO 2025

Material/Methods: A set of 32 3D-printing filaments, varying in color and composition, were printed using different settings (e.g., density, flow rate). They were scanned with dual-energy CT (DECT; SOMATOM Confidence; Siemens Healthineers) at a tube voltage of 80 and 140 kV to quantify material properties: Hounsfield Unit (HU), effective atomic number (Z eff ), Relative Electron Density (RED), and proton Stopping Power Ratio (SPR). Materials were printed as solid cylindrical rods fitting the Gammex phantom (Model 1472; Sun Nuclear). DECT calibration (1) and material property assessment were performed in AMIGOpy, an open-source and in-house developed software. Filament material properties were compared with ICRU44-based metrics and selected on their resemblance with human-tissue for phantom printing. One patient-size pelvic- and one thoracic (including deformable lungs) phantom were 3D-printed from a generated realistic anatomical model (XCAT). 3DP-settings were optimized based on the 3D-printed rod results to showcase higher tissue-equivalence using this application. Results: The ICRU44 Z eff ranged between (6.2 – 6.6) for adipose and was (7.5) and (13.41) for muscle and cortical bone-tissue, respectively. The printed rods showed Z eff -values between 5.9 and 11.8 which were not affected by 3DP-settings. Different 3DP-settings did influence HU, RED and SPR. Phantom printing was limited to the use of two simultaneous filaments, a soft tissue mimicking filament was chosen with a Z eff of 6.9 (equidistant between adipose and muscle tissue) and a filament approximating cortical bone (Z eff =11.78). Different 3DP-settings were used to increase HU, RED and SPR contrast, highlighting muscle and adipose tissue differences inside the phantoms. In figure 2 the pelvic-phantom shows visually and quantitively this improved contrast in HU, RED and SPR as a result of these different 3DP-settings.