ABSTRACT
Background: Patient’s conebeam computer tomography (CBCT) images have suggested a possibility for adaptive radiotherapy although the dose delivery is of structural complexity. It is of practical importance to verify and test the intensity-modulated radiation (IMRT) planning system for radiation therapy. Objective: Verify accuracy of dose calculations based on CBCT imaging. Materials and methods: Electron density calibration curve was generated for planning CT and CBCT data set using two CT phantoms (Gammex RMI® and Catphan® 600). Anthropomorphic head and neck phantom images were acquired from planning CT and CBCT. The routine IMRT technique was generated on the planning CT, which was applied to the CBCT. Dose distributions were computed. All LiF TLD-100 dosimeters were calibrated with gamma-ray. Forty-eight TLD measuring points were chosen in five different slices of the phantom. Measurements were repeated four times, and the average dose was compared to the reading doses on both CT and CBCT plans. Dose volume histograms (DVH) of various structures were generated, and dose statistics were analyzed. Results: Hounsfield unit obtained from Catphan phantom was similar between planning CT and CBCT. IMRT dose calculations based on the planning CT and CBCT agreed well with reading doses at 48 points. Statistical point doses by DVH calculation on CBCT were about 3% lower than those by the conventional CT. Dose ratios calculated over measured ones ranged from 0.82 to 1.09. Conclusion: Point doses and DVH calculations based on the planning CT and on-board CBCT were in acceptable agreement. CT phantom specifically designed for CBCT is recommended to improve accuracy of IMRT dose calculation on CBCT images.
ABSTRACT
Objective: Setup verification is the critical part to make sure of the accuracy for Intensity-modulated radiotherapy in nasopharyngeal cancer patients. This pilot study was designed to answer whether and how much the kV-CBCT in addition to 2DkV is more accurate than 2DkV alone in terms of verification parameters. Methods: Images acquisition: Offline images were displayed in the matched position between reference images. 2D and 3D matches: The bony matches were done by using the location of the tumor in the nasopharyngeal and upper neck regions. The distances displaced from the isocenter were recorded in x-y-z directions. Analysis: The distance of the isocenter shift in each direction (X, Y, Z) were presented as point estimations. The alignment between the two methods was assessed with Pearson’s and Spearman’s correlation. The 3 mm difference within 90% is considered as an acceptable range of non-inferiority of 2DkV, compared with CBCT. Results: 11 nasopharyngeal cancer patients were included into this study. The correlation between 2DkV and kV-CBCT were 0.46, 0.11 and 0.16 for Superior-inferior (SI), Anterior-posterior (AP) and Left-right (LR) directions, respectively. The central value for the kV-CBCT; SI, AP and LR directional shift were 0.07, 0.06 and 0.03 cm, respectively, whereas the central value for 2DkV; SI, AP and LR directional shift were 0.05, 0.07 and 0.04 cm. For the difference shift < 3 mm, the results > 90% were within acceptable value: 100% and 96.96% for SI and LR directions whereas the AP direction was 87.87%. Conclusion: Compared with kV-CBCT by using our criteria, 2DkV images are accurate enough for treatment verification in nasopharyngeal cancer patients.