S1986
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
planned target volume (PTV), placed in marrow bone and iodine regions, were within 0.5% for D2%, Dmean, and D98%. In patient data, the maximum dose difference and overall deviation did not exceed 2%, observed in a patient whose PTV was located over bone (Fig.1). The percentage dose differences for PTVs of all the studied patients were within 0.75%, 1%, and 1.25% for V95%, Dmean, and D0.03cc, respectively. For organs at risk, D2%, Dmean, and D98% remained within 1.25% agreement (Fig.2).
International Journal of Radiation Oncology*Biology* Physics, 47(4):1121–1135, 2000 Keywords: probabilistic planning, setup uncertainty
Digital Poster 4243 Integration of Virtual Unenhanced CT Images into Radiotherapy Treatment Planning Mohammad Varasteh 1 , Marie Fargier-Voiron 1 , Nicolas Perichon 1 , Mireille Conrad 1 , Cédric De Marco 1 , Oscar Matzinger 1,2 , Maud Jaccard 1 1 Radio-oncology Department, Clinique Générale- Beaulieu et Clinique de Genolier, Swiss Medical Network, Genolier, Switzerland. 2 Health Sciences and Technology Department, ETH University, Zurich, Switzerland Purpose/Objective: Spectral maps, including iodine-suppressed (Virtual Unenhanced-VUE) images available from Dual-Energy- CT (DECT) have recently gained attention in radiotherapy (RT) workflows. The VUE images generated using the Gemstone Spectral Imaging mode on the GE Revolution DECT can potentially replace pre- contrast scans in RT planning. This study examines VUE images in dose calculation by comparing the dose distributions between pre-contrast and VUE images for ten patients. To enable a more detailed insight, an initial phantom study was carried out. Material/Methods: A multi-energy phantom (Gammex, Middleton, USA) with various inserts, including iodine and bony materials, was scanned on the Revolution DECT. To simulate a native acquisition, the first dataset was acquired without any iodine components, whereas in the second, five solid-water inserts were replaced with iodine elements of varying concentrations (0.5- 20mg/ml). Since VUE is applied to the 70 keV Virtual- Monoenergetic-Image (VMI), both VUE and 70 keV VMI images were reconstructed. TomoHelical plan generated in RayStation2024A on the native 70 keV VMI was recalculated on the VUE image and subsequently compared. For the clinical evaluation, ten patients with pelvic, brain, esophageal, and thoracic cancers - treated on Radixact and CyberKnife with SBRT and conventional fractionation - were analyzed. Results: The phantom study demonstrated that, despite effectively replacing iodine attenuation values with those of blood, the VUE algorithm reduces HU values in bony structures. While high-density bones remained largely intact, the lower-density trabecular and marrow bones showed average reductions of 30% to 60%, respectively. This yields a mass density deviation of approximately ±0.1g/cc. The dose differences in the
Conclusion: Our dosimetric comparisons suggest that VUE images might effectively replace pre-contrast scans in radiotherapy planning, with minimal impact on dose distribution. While trabecular bone may be misidentified due to similar attenuation characteristics to iodine, this implies minimal dosimetric consequence. Further investigations are needed to assess whether the observed dose differences are affected by anatomical changes between pre- and post-contrast scans, and to evaluate the use of VUE images for daily image-guided RT. Keywords: RT dose planning, Virtual Unenhanced Image, DECT
Made with FlippingBook - Share PDF online