Room: Davidson Ballroom A
Purpose: To assess the beam-level targeting accuracy of the robotic system using a scintillator/CCD phantom (XRV-124, Logos Systems International, Scotts Valley, CA).
Methods: The phantom is composed of a scintillator imaging cone coupled with a CCD camera. The CCD camera captures the two light spots where the radiation beam intersects with the scintillator imaging cone, and the recorded 2D image sequences are used to reconstruct the beam vectors. The Phantom was scanned at 0.625mm slice thickness on a GE Scanner. Three isocentric plans were created in CyberKnife MultiPlan (v5.3) using all the body path nodes with three collimators: 7.5mm fixed cone, 7.5mm Iris collimator, and 7.6mm x 7.7mm InCise2 Multi Leaf Collimator(MLC). The plans were delivered on a CyberKnife M6 system with imaging guidance on the fiducials embedded in the phantom. The captured beam vector coordinates (X, Y, Z, θ, φ) were compared with the planned coordinates.
Results: A total of 10 deliveries (4 with fixed cone, 3 with Iris, and 3 with MLC) were measured with reposition/realignment, collimator remounting, and intended extra residue alignment errors. The mean targeting errors of all 10 measurements are <0.2mm in X, Y, and Z directions and <0.3 degree in θ and φ angular directions. The mean distance errors (ΔR) for all the captured beams are 0.387±0.21 (fixed cone), 0.308±0.147 (Iris), and 0.414±0.188 (MLC); and the maximum distance errors are 1.261 mm (fixed cone), 0.803 mm (Iris), and 1.041 (MLC). With >95% confidence, the beam delivery precision is <0.2 mm in three translational directions and <0.2 degree in two angular directions.
Conclusion: The study verified overall sub-millimeter delivery accuracy of CyberKnife system at beam-level for the entire body path nodes with fiducial tracking. The XRV-124 phantom was proved to be a valuable systematic delivery QA tool for the robotic radiosurgery system.