ESTRO 38 Abstract book

S965 ESTRO 38

= 0.47) and 74% for beam 2 (mean gamma = 0.77), while the dose differences at the isocenter were 9.9% for beam 1, and -6.6% for beam 2 (iViewDose compared to TPS). These differences may be at least in part attributable to the presence of very small (=2 degrees) segments with different linac states, which may be challenging for accurate dose reconstruction. Except for some deviations in leaf position and speed for the very small segments at the end of beam 1, the linac accurately delivers the control points.

Material and Methods Array and dose calibration was performed prior to measurement. ArcCheck was irradiated with 5x5, 10x10, 15x15 and 20x20 field sizes on TrueBeam machine (Varian) with different beam energies used: 6WFF, 15WFF, 6FFF and 10FFF. The ratio test was done for all beam qualities and field sizes. Calculations of TPS dose distribution was done in Eclipse (AAA 13.6.23) with different HU assigned for ArcCheck phantom. Entrance and exit dose for measurement was calculated from six diodes surrounding entrance/exit points. Standard deviation was treated as measurement uncertainty. In the second step of our study we performed analysis for clinical plans. We evaluated 10 IMRT and VMAT plans with different beam geometries: full and half arcs, IMRT with small and large field sizes. Dose calculations were performed for two different HU values – one recommended (246HU) and one achieved in former step of our work (150HU). In order to do the bias-free evaluation artificial measurements were created based on the calculations performed with 150HU and compared with dose distributions calculated for 246 HU. Artificial measurement is a file created with a Python script on the basis of TPS dicom dose distribution which mimics the real measurement file. Comparisons were done with the SNCPatient (SunNuclear, v.6.7.3) software. The 3mm/3%; 2mm/2%; 1mm/1% parameters for global and local gamma evaluation were used. Dose threshold was set to 5%. Results For 10x10 field size the HU achieved from the ratio test varies between 150HU and 246HU depending on the beam quality (see Figure 1). The ratio test gives also incoherent results for evaluated field sizes: 200-280HU for 6WFF, 160- 190HU for 15WFF, 160-230HU for 6FFF, 130-160HU for 10FFF (see Figure 2). The difference in gamma evaluation between 150HU and 246HU dose distributions depended on patient geometry and was mostly observed in exit dose region, though influencing the local gamma evaluation. Percent of passing points varied between 100% and 93% for 3mm/3% local gamma analysis and between 100% and 79% for 2mm/2% local gamma analysis.

Figure 2: IView result for beam 2.

Conclusion We have described a method to implement simple linac QA for VMAT using EPID dosimetry. The test beams have been successfully constructed and tested with benchmark measurements. The next step will be to determine reproducibility of results based on repeated measurements, in order to establish baseline values and possible action levels. EP-1784 ArcCheck HU setting influence on the uncertainty of pre-treatment verification results M. Giżyńska 1,2 , M. Bukat 1 , D. Szałkowski 1 , A. Walewska 1 1 Maria Skłodowska-Curie Memorial Cancer Center and Institute of Oncology, Medical Physics Department, Warsaw, Poland ; 2 University of Warsaw Faculty of Physics, Department of Biomedical Physics, Warsaw, Poland Purpose or Objective ArcCheck (SunNuclear) should be characterized in TPS by the HU value that would provide the best agreement of TPS vs measurement entrance to exit dose ratio (ratio test) for 10x10 field size. The purpose of this work was to examine whether the best ratio test results depend on the beam quality and field size. The influence of the HU setting on the results for gamma analysis made for clinical plans was also examined.