S2044
Physics - Image acquisition and processing
ESTRO 2026
kVp analysis exceeded 99 % (2%/2mm) and 98 % (1%/1mm). Two-way ANOVA revealed no significant interaction between kVp and treatment technique (IMRT/VMAT).Conversely, CatPhan-based calibrations consistently underestimated DVH metrics versus CIRS (p < 0.001), especially in high-density tissues. This dosimetric difference was confirmed by γ -passing rates, which dropped to 88–95 % (1%/1mm) when comparing phantoms.
Fig. 1. Top: HU-RED and HU-MD relationships across kVp for CIRS and CatPhan. Bottom: Effect of kVp variations on the HU (left), and proposed polynomial fits for the highest-density tissues (right).
Fig. 2. DVH metrics across CT calibration curves using AXB (top) and AAA (bottom), comparing CIRS (green) versus CatPhan (blue); nplan=160. Stars ( ★ ) indicate p<0.05 for kVp comparisons, and phantom comparisons are p<0.001 (Wilcoxon test). Conclusion: Phantom selection (plastic vs tissue-equivalent) has a stronger dosimetric influence than dCT parameter variations. These findings support the suitability of using dCT, regardless of kVp, for IMRT/VMAT planning, enabling resource-efficient workflows like simulation-
Made with FlippingBook - Share PDF online