Room: ePoster Forums
Purpose: The purpose of this project is to propose a method to optimize the isocenter calibration process for an optical surface tracking system.
Methods: A BB phantom was first utilized to identify the relative positioning of: (a) the optimal radiation isocenter centroid, as determined from 6MV Winston-Lutz images produced by rotating the gantry and collimator only, (b) kV-CBCT imaging isocenter, and (c) the rotation axis of the linac couch. By following the routine calibration procedure for the surface imaging system, C-RAD isocenter is initially matched to the kV-CBCT isocenter. Subsequently, the C-RAD isocenter is shifted horizontally to coincide with the couch rotation axis, which had been determined to be displaced from the C-RAD isocenter by 0.6mm. A static, anthropomorphic head phantom, immobilized in an open-face frameless SRS mask, is used to quantify the accuracy of these two surface imaging isocenter calibrations at maximum couch rotation angles.
Results: Prior to shifting the C-RAD isocenter to coincide with the axis of the couch, the magnitude of initial displayed offsets between the rotated live image of the head phantom and the rotated reference images at couch 270 and 90 degrees were 0.9mm and 1.0mm, respectively. Following re-assignment of the surface imaging isocenter to the couch rotation axis position, these offsets reduced to 0.2 mm and 0.4 mm in magnitude, respectively.
Conclusion: If the surface imaging system’s isocenter is offset from the couch rotation axis, subsequent patient offsets displayed for large couch rotations will potentially include that calibration offset error. Minimizing this error could enable improved accuracy in reported submillimeter patient motion. This method could potentially benefit sites with minimal (or well-characterized) couch wobble, or sites with couch rotation axes offset from MV isocenter.