SU-F-BRE-09: Linac Isocenter Quality Assurance: A Stereotactic Approach

Purpose: A quantitative method was designed to independently determine the couch/collimator/gantry isocenters, as well as the overall LINAC mechanical isocenter in 3D. Methods: Performed on both a Varian TrueBeam™ STx and Trilogy with the gold standard front pointer positioned at 100 cm SAD, and a Radionics XKnife™ (RX) attached to the couch. At gantry and couch 0°, the RX laser alignment attachment (RXLAA) was centered to the front pointer using the micrometers (0.1-mm precision) on the RX. The 3D coordinates of the micrometers were recorded. The collimator was rotated to 90° and 270°. At each collimator rotation, the RXLAA was re-centered to the front pointer and the micrometer coordinates recorded. At collimator and gantry 0°, the process was repeated for couch angles 0°/90°/270°. Finally, at collimator and couch 0°, the steps were repeated for gantry rotations 0°/90°/180°/270°. The centers/radii of the smallest bounding spheres for the collimator, couch and gantry walkout were calculated (using MatLab™). The smallest bounding sphere containing the collimator, couch and gantry walkout spheres was then calculated. The center of this all-encompassing sphere is the overall mechanical isocenter of the LINAC. This position was dialed in on the RX. LINAC mechanical and radiation isocenter coincidence was determined by performing Winston- Lutz test at four cardinal gantry angles. Results: TrueBeam and Trilogy mechanical isocenters had overall walkout radii of 0.8 mm and 1.5 mm, respectively. For the TrueBeam and Trilogy 6-MV beams, the radii of radiation isocenter were 0.3 mm and 0.4 mm, respectively, with distances between LINAC mechanical and radiation isocenters of 0.5 mm and 0.9 mm, respectively. Conclusion: This efficient and simple method allows for an independent and reliable quantitative assessment of LINAC isocenter in 3D with equipment typically available in a radiation oncology clinic. It can easily be performed for LINAC commissioning and annual QA