Purpose: With the rise of highly conformal, stereotactic procedures as well as multi-isocenter treatments and non-coplanar arcs in VMAT the accuracy of the couch motions for patient positioning is increasingly important for accurate delivery and patient safety. Common QA procedures rely on manually moving the couch and physically measuring the displacement. This process can be time consuming, and the manual nature allows for errors and inter-operator variabilities. Many treatment vaults are equipped with 3D optical imaging systems which can provide an accurate and automated alternative.
Methods: A script was generated to automatically drive the couch through a trajectory describing a 20cm cube at couch rotation angles of 0Â° and Â±5Â°, with dwell times at each of the vertices for statistical averaging, using the developer mode feature on a TrueBeam LINAC (Varian Medical Systems). A standard IR reflective marker block was placed on the couch and tracked with a Polaris (NDI Medical) imaging system. By analyzing the vertex locations and cube edge lengths in the optical measurement, the trajectory design provides 12 measurements of translation accuracy for each direction of motion, 24 measurements of the angle between the respective motion axes, and 3 measurements of the couch rotation accuracy as well as the alignment of the couch rotation axis with the vertical translation axis.
Results: The translation errors were measured to be 0.26(std=0.07), 0.53(std=0.10) and 0.02(std=0.10) mm in the vertical, lateral and longitudinal directions respectively. The angle between the each of the three translation axes was found to be 90.00Â°(std=0.17Â°, 0.06Â° and 0.04Â° respectively. Couch rotation errors were <0.02Â° in each direction and the angle between the rotation axes and the vertical translation direction was 0.19Â°.
Conclusion: The results show optical tracking can provide accurate measurements for automated QA, including components difficult to measure physically such as axis orthogonality and alignment.
Funding Support, Disclosures, and Conflict of Interest: Funding was provided in part by NIH grant R01CA227124.