Room: Exhibit Hall | Forum 5
Purpose: To evaluate the accuracy of surface guided patient setup compared to traditional 3 point tattoo setup.
Methods: 59 Patients, 608 alignments were performed using 3 tattoos aligned with room lasers, and 6DOF surface guidance (SG) offsets were calculated using the planning CT body contour as reference (AlignRT). SG equivalent shifts along the three principal directions were calculated from the original recorded 6DOF translational and rotational offsets using rigid body motion equations. Target localization imaging (kV/MV ports, kV 2D-2D match, kV-CBCT or Clarity ultrasound) was performed and IGRT shifts were recorded. SG vs tattoo shifts to treatment position based on IGRT were compared to determine setup accuracy of the two methods for treatment sites including breast, chest/abdomen, and pelvic regions.
Results: The averaged shifts of tattoo to treatment position was 0.38Â±0.32cm (vertical), 0.34Â±0.35cm (longitudinal), and 0.29Â±0.26cm (lateral); the averaged shifts of SG to treatment position was 0.48Â±0.41cm (vertical), 0.65Â±0.61cm (longitudinal), and 0.33Â±0.32cm (lateral). The overall shifts of tattoo-based and SG-based patient setup were comparable within 3 mm, and tattoo-based setup has slightly less shifts. The averaged difference in IGRT shifts between tattoo-based and SG-based setups was 1 mm in the vertical direction and 3 mm in the longitudinal direction. Large shifts (2 to 4.7 cm) of AlignRT along longitudinal direction were seen in 14 cases (2.3%), and the maximum tattoo shifts for these cases was 0.7 cm. Large rotations (over 3 degree, 18.6%) recorded by AlignRT were observed in this study.
Conclusion: The very small average differences between the two setups confirm the general equivalence of tattoo-based and SG-based patient setup. However, the very large AlignRT shifts seen in some cases shows the importance of improving the workflow of SG-based patient setup, such as reducing patient skin covering and choosing more characteristic skin surface as ROI.
Not Applicable / None Entered.