ESTRO 38 Abstract book

S1142 ESTRO 38

test was generated to investigate whether the deviations reported by the Pointer-WL test are actually reflecting the geometrical shift of the measured prescription isodose line respect to the target center (d 80%-center ). In addition, the film dose distributions were compared with the theoretical distributions computed by the Eclipse, to evaluate the entire delivery accuracy (E2E test) in terms of isodose shifts (d E2E ). The Pointer-WL, Dose-WL and E2E tests were performed in 10 different sessions using CBCT imaging. For each session, the d ball-field,max value was compared to the three patient’s anatomical components (Left-Right or L/R; Anterior-Posterior or A/P; and Superior-Inferior or S/I) of the d 80%-center and d E2E vectors. Results The deviations (mean ± SD) found for the Dose-WL (L/R: 0.5 ± 0.4 mm; A/P: 0.5 ± 0.4 mm; S/I: 0.6 ± 0.2 mm) and the E2E tests (L/R: 0.7 ± 0.5 mm; A/P: 0.5 ± 0.4 mm; S/I: 0.6 ± 0.2 mm) were in most cases less than 1 mm, and they were quite less than the maximum deviations registered with the traditional Pointer-WL tests (d ball-field,max : 1.8 ± 0.3 mm). On average, d ball-field,max exceeded 1.2 mm (in R/L and SI) and 1.3 mm (in A/P) the deviations reported by the 10 Dose-WL tests. Analogously, an average excess of 1.0 mm (in R/L), 1.3 mm (in A/P) and 1.1 mm (in S/I) were noted for the 10 E2E tests. Conclusion In this study we have proved that deviations greater than 1 mm as reported by the analysis of a traditional WL test do not necessarily imply shifts greater than 1 mm in the prescription isodose line respect to the target to be treated. According to the results of this works, it is advisable to move to a dose-based WL test by replacing the classical WL test involving a metallic ball that only provides information about geometric misalignments of the isocenter. EP-2072 Automatic analyzation of patient specific QA measurements made with the Octavius verification device Purpose or Objective Patient specific 3D QA measurements with the PTW Octavius phantom are well accepted. A small drawback of the method is, that only the 3d dose distribution can be evaluated what makes it difficult to understand possible deviations found. We present a tool that compares the measurement file to the DICOM RT plan file and detects deviations automatically. Material and Methods The measurement of the PTW Octavius system results in a time sample structured file, based on the settings given for the system. This file includes the dose matrix together with gantry and time information. The tool presented detects the dose information and puts them together as dose groups by eliminating the beam off times. Calibration and correction factors are also applied. The DICOM RT plan file is searched for MU, MLC, Gantry information which will be sorted to fit the order of the measurement data. Knowing the geometry, RT-plan data are transferred to dose matrixes using a simple scatter model (Gaussian curve based). Measured values are given in the Octavius/array resolution needing an interpolation and correction of partial volume effects, which is also taken care of by a scatter model. The tool now analysis both data sets for dose and position errors which are indicated only if given limitations are exceeded and finally calculates a gamma pass rate per segment/gantry step. 12 patient cases from a Siemens Artiste have been analyzed so far with the IDL (RSI) written tool. Excel sheets then help to understand and display the generated data. Results P. Haering 1 , M. Splinter 1 , C. Lang 1 1 DKFZ, e040, Heidelberg, Germany

Table 1: DSC and MAE computed between the sCTs and planning CT

Figure 1: Typical example of a) planning CT b) intensity- based sCT c) contour-based sCT d) hybrid sCT Conclusion In this project, we have tested multi-atlas based deformable registration workflows using commercially available RTP software. Our results indicate that hybrid of intensity- and contour-based deformable registration methods MIM provides the best solution for generating sCTs in the prostate region. Reference: 1. Sjölund, J., Forsberg, D., Andersson, M., & Knutsson, H. (2015). Generating patient specific pseudo-CT of the head from MR using atlas- based regression. Physics in Medicine & Biology, 60(2), 825. EP-2071 Redefining the classical Winston-Lutz test for cranial radiosurgery in terms of dose J.F. Calvo Ortega 1 , S. Moragues Femenia 1 , J. Casals 1 1 Hospital Quirónsalud Barcelona, Radiotherapy, Barcelona, Spain Purpose or Objective To analyze the dosimetric impact of the deviations (ball center-field center) revealed by the traditional Winston- Lutz (WL) test. Material and Methods A pointer-based WL plan containing 15 non-coplanar fields is routinely delivered in our department for SRS QA purpose (Pointer-WL test). The maximum deviation between the ball center and the radiation field center (d ball-center,max ) over the 15 taken portal images gives the result of this test. Two acrylic phantoms able to support a radiochromic film in coronal and sagittal orientations, respectively, were home-made. Both phantoms were CT- scanned and transferred to the Eclipse treatment planning system (TPS). The WL plan was recreated in the Eclipse TPS over each phantom to be delivered (Dose-WL test). This film-based Electronic Poster: Physics track: Implementation of new technology, techniques, clinical protocols or trials (including QA & audit)