Room: Exhibit Hall | Forum 5
Purpose: To both compare fiducial and CBCT localization and to investigate the effect of changing various scan parameters (FOV, slice thickness, and CTDI) when using CBCT for stereotactic localization on the Gamma Knife Icon.
Methods: With the Icon, image-guided Gamma Knife radiosurgery is now possible. For frame-based patients, two methods of localization are possible: fiducial or CBCT. Utilizing the Elekta solid water dosimetry phantom, CT images were acquired while varying FOV (250 and 600 mm) and slice thickness (0.8 and 2.0 mm) parameters from the clinically used baseline of (1mm slice thickness and 300 mm FOV). These images were imported into GammaPlan and fiducial localization was performed. A test radiosurgery plan was developed, and the phantom was set up on the Gamma Knife Icon for â€œtreatment.â€? Three CBCTs of each CTDI setting (2.5 and 6.3) were acquired and registered with the fiducial reference images for comparison of stereotactic reference methods and scan parameter effects. Additionally, each registration was performed with and without the optional region of interest (ROI) used to exclude areas of CBCT artifact from the registrations.
Results: Registration comparisons between CBCT and fiducially localized CT yielded average 3D translational differences of 0.86, 0.74, 1.06, 0.96, and 0.60 mm for baseline, 0.8 mm slice, 2mm slice, 250 mm FOV, and 600 mm FOV, respectively. CTDI of CBCT and small FOV were not found to be statistically significant parameters while the large slice thickness and large FOV were found to be statistically significant in all situations. Small slice thickness was barely significant for all scans, but not when separated by CTDI of the CBCT.
Conclusion: For this phantom, there is a difference between CBCT and fiducial localization for the baseline scan parameters. Additionally, large slice thicknesses and FOV should be avoided while ROI use may help with registration accuracy.