(Tuesday, 7/16/2019) 9:00 AM - 10:00 AM
Room: Pins Room | Hall 2
Purpose: VisionRT has recently introduced its Advanced Camera Optimization (ACO), aimed at improving the accuracy and stability of the 6DOF tracking. The stability of the 6DOF tracking of a phantom was tested during the delivery of a kV-CBCT as well as during the delivery of five HyperArcTM (Varian Medical Systems, Palo Alto, CA) plans with varying isocenter positions. Additionally, the accuracy during HyperArcTM patient treatments before and after the ACO was analyzed.
Methods: Five HyperArcTM plans were created on a Max-HD anthropomorphic SRS head phantom (IMT Inc., Troy, NY) with lesions located in the mid, left, right, posterior, and anterior brain. HyperArcTM allows the use of a full 360ÌŠ coplanar arc plus half arcs with couch kicks of 45ÌŠ, 315ÌŠ, and 270ÌŠ. For the mid and posterior plans all the arcs were used, while for the others only a half coplanar arc was used in addition to all the non-coplanar arcs. The phantom was aligned using a kV-CBCT, then the AlignRT baselines were re-captured. AlignRT data was collected during the delivery of all the plans. The 6DOF tracking accuracy of fifteen HyperArcTM plans, each before and after the ACO, were compared to understand the difference in clinical performance.
Results: The average translational error during the kV-CBCT delivery was 0.13-0.15 mm. During the HyperArcTM delivery the translational errors were consistent regardless of isocenter for the coplanar arcs. The largest errors were seen during the delivery of the 45ÌŠ non-coplanar, which matches the results of a couch walkout test. Additionally, the right-sided isocenter plans showed the largest errors while the left-sided isocenter showed the smallest errors. For the clinical cases, the absolute variation overall improved with the ACO except for the pitch.
Conclusion: We have seen an improvement in the tracking stability of our AlignRT system after the implementation of the ACO.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
