S1975
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
2. Agostinelli, S., et al. (2003) Geant4-A simulation toolkit. Nucl. Instrum. Meth. A. 506(3):250–303. doi:10.1016/S0168-9002(03)01368-8 3. Li, Z., et al. (2022). Implementation of the EPICS2017 database for photons in Geant4. Phys. Med. 95:94-115. doi:10.1016/j.ejmp.2022.01.008 4. Zankl, M., et al. (2021) The ICRP recommended methods of red bone marrow dosimetry. Rad. Meas. 146(21):106611. doi: 10.1016/j.radmeas.2021.106611 HU-RED CBCT calibration curves validation and feasibility for dose calculation verification for linear accelerators with aSi1200 vs. HyperSight imager Patrycja Borowska 1 , Marta Paluszy ń ska 1 , Krzysztof Matuszewski 1 , Urszula Sobocka-Kurdyk 1,2 1 Zak ł ad Fizyki Medycznej, Wielkopolskie Centrum Onkologii, Pozna ń , Poland. 2 Wydzia ł Medyczny, Uniwersytet Kaliski, Kalisz, Poland Purpose/Objective: The aim of this study was to determine whether kV- CBCT images aquired at the treatment unit are feasible for dose calculation. It was subject to Digital Poster 4044 verification if it is possible to assign universal HU-RED calibration curve for different CBCT modalities. It was evaluated whether the use of advanced image acquisition system (Hypersight) provides superior dose calculation reliability to the standard – aSi-1200 detector. Material/Methods: In this study, HU-RED/physical density calibration curves were measured for two types of image detectors (standard – aSi- 1200 and advanced – Hypersight) integrated with TrueBeam linear accelerators. Measurements were performed with Cheese Phantom for all available CBCT modalities (head, pelvis and thorax). The results were compared to the HU-RED calibration curve assigned in the TPS to the CT scanner used for routine dose calculations. For dose distribution calculation purpose, all investigated CBCT modalities and detectors as well as CT images of Catphan 604 phantom were acquired. Both open and modulated arc field plans were calculated and dose distributions analyzed. Doses were compared against reference (calculated on CT-scans of Phantom) in terms of point dose value at selected reference point, mean doses in volumes of interest (VOI) as well as 3D local gamma dose distribution comparison (with 2%, 2mm pass-criteria). Results: HU-physical density calibration curves for all CBCT modalities are plotted against CT calibration curve in fig. 1.
Figure 1: Two-field plan calculated using the updated hybrid-MC engine. The GPU accelerated calculations took 5 mins per field. A sub-voxel dose calculation is shown for a subsection of the dose field. The D t calculation in figure 2 shows a smooth dose distribution similar to D w and in contrast with D m which shows strong dose enhancement in the bone. Unlike the D w calculation, D t accurately predicts the attenuation of photons through the patient. D t better approximates the dose deposited to soft tissues in bone and produces dose distributions comparable to MV photon radiotherapy.
Figure 2: Patient femur case comparing MRT valley D t , D w , and D m calculation methods. Conclusion We have developed a fast MC engine for the hybrid algorithm capable of computing MRT dose distributions in a few minutes on capable graphics hardware. The improvements in calculation times open the door to developing inverse planning methods for MRT. References 1. Donzelli, M., et al. (2018). Hybrid dose calculation: a dose calculation algorithm for microbeam radiation therapy. Phys. Med. Biol. 63:045013. doi:10.1088/1361- 6560/aaa705
Made with FlippingBook - Share PDF online