Evaluation of the Residual Rotation Error of a 6 DOF Robotic Couch Using Optical Surface and CBCT.

Abstract

Rotation setup errors for SRS treatments are typically less than two degrees which usually causes less dosimetry discrepancy for a typical SRS treatment as the target size is likely less than 4 cm. However this might not be true for the single isocenter and multiple metastatic targets’ technique where, the distances between the isocenter of the treatment and those tumor centers can be very large. A 6DOF couch can be used to correct this setup error by using optical surface and/or Conebeam CT (CBCT) images. The accuracy and consistence between optical surface and CBCT images were studied. The CBCT system and optical surface images of an SRS phantom were taken at different pitch, roll, and rotation angles varied from -3 to 3 degrees by intentionally changing values on the service console. A total 39 combinations of pitch, roll, and rotations using the 6DOF couch were generated. The optical surface images were taken and Vision RT software was used to determine the rotation angles and isocenter shifts. Eight CBCT images corresponding to +- 3 degree changes in pitch, roll, and rotation were taken. The center positions of the 5 tungsten carbide BBs contained in the SRS QA phantom were determined using in-house software. A least square fit was used to determine translation/rotation parameters. Among the 39 calculations from optical images, the maximum rotation angle differences were 0.2, 0.1, and -0.2 degrees for pitch, roll, and couch rotation, respectively. The maximum isocenter shifts were -0.80 mm, 0.70 mm, and -0.10 mm in lateral, longitudinal, and vertical axes, respectively. The maximum residual rotation angles from analysis of CBCT images were 0.20 degree for pitch, roll, and rotation. By using 6DOF couch, the rotation errors can be reduced dramatically compared to initial rotation errors. To overcome this rotation errors without 6DOF couch, one has to introduce extra margin for target which is far away from the treatment iso-center. Six DOF can avoid extra CTV-PTV margin. The residual rotational errors studied for the couch were less than 0.2 degree. This accuracy for the 6DOF system is sufficient for SRS applications even for single isocenter multiple tumor technique and extra CTV-PTV margin might be not necessary for some cases.