SU‐E‐J‐45: Verification of Beam‐Line Geometry of Truebeam Using MV‐EPID

Abstract

Purpose: Geometrical accuracy of beam‐line components of a linear‐accelerator (LINAC) is indispensible for quality of patient treatments. Positional reproducibility of the flattening filters (FF) may directly affect the beam flatness and symmetry while uncertainties in the imager position may directly impact target localization. In this study we have investigated the positional uncertainties of both of these components. Methods: Images of both flattened and flatness‐filter‐free (FFF) beams of 6 and10 MV were acquired using the megavoltage electronic portal imaging Device (MV‐EPID) over a period of time to investigate the positional uncertainties in the beam‐line geometry. Due to forward peaked nature of the MV beams FFF beams show a cone‐shaped dose distributions that was used determine the center of the X‐ray beams and used as reference points. Central portion of the flattened beams show the image of the FF. These circular objects were used to determine the center of the flatness‐filter with respect to the reference point. Imager reproducibility was also tested against the reference point which is the center of the source. Positional reproducibility of the FF and the imager were tracked over a period of two months. Results: Centers of both 6X and 10X FFs were systematically shifted by 2.2mm in X‐and ‐ 0.7 in Y‐directions. However, the shifts were magnified due to the fact that the FF being close to the MV‐source while the imager is far away (at 108cm) from the source. When corrected for this magnification, the real uncertainty in the FF position is approximately 0.2mm and ‐ 0.06 mm in X‐and Y‐directions. The imager positional reproducibility was found to be within ±1 pixel (<0.5mm). Conclusion: In this study we have shown a new method to detect positional uncertainty of the flatness‐filter and MV‐EPID. This methodology may be useful for QA of the LINACs. This project is supported by Varian Medical Systems