Room: Exhibit Hall
Purpose: The purpose of this study is to evaluate the susceptibility of the Gamma Knife frame to distortions as a function of applied torque to the head pins. Distortions were quantified by the mean and maximum values of the fiducial position error during the frame localization process in Gamma Plan.
Methods: A Leksell G frame was attached to a head Rando Phantom using aluminum pins. Utilizing a calibrated torque wrench, 30 cN-m torque was applied to each pin and a CT was acquired with the localizer mounted. The torque applied to the pins was subsequently increased by 10 cN-m and a new CT was acquired for analysis. The process was repeated until 120 cN-m torque was reached. Overall, 8 CT scans were acquired corresponding to 30, 40, 50, 60, 75, 90, 105, and 120 cN-m. Using Gamma Plan, the CT dataset for each setting were defined as stereotactic references and fiducial position mean and maximum errors were determined. This stereotactic coordinate system was determined using 3 different window/level settings and averaged.
Results: Results show that for the investigated geometry, average fiducial position mean error for the 8 CT scans was 0.3 mm (STD = 0.04 mm) while the average maximum error was 0.61 mm (STD = 0.09 mm). Figure 1 shows a plot of the mean and maximum error as a function of the torque applied. This value is similar to the mean and maximum error measured on a no-phantom, frame-only CT.
Conclusion: A validation study was performed to investigate the effect of various applied pin torque to distortions of the Leksell G frame. For the setup investigated, no correlation was determined between the fiducial position mean and maximum errors when torque value of 30 cN-m to 120 cN-m were applied.