Room: AAPM ePoster Library
To measure the offset between radiation and mechanical fields at non-isocentric locations for multi–target single-isocenter frameless stereotactic radiosurgery (SRS) in a single set-up using an in-house phantom.
A phantom (5 ball-bearings (BB) embedded in epoxy resin) was designed and manufactured to mimic multi-target single-isocenter SRS geometry. One BB was placed at the center of the phantom while the remaining four were located at 3.6, 5.4, 7.7 and 9.9cm away from the central BB. Winston-Lutz (W-L) images were acquired for each BB with the electronic portal imaging device (EPID) and analyzed with the RIT software (Radiological Imaging Technology). A 1x1cm2 MLC field was generated to target each BB at various gantry, collimator, and couch angles. First, standard W-L images were acquired for the central BB and the daily setup errors reported by RIT were removed to accurately position the phantom at the radiation isocenter. Next, W-L images were acquired at the off-center locations using 8 beams per BB. Measurements were repeated 8 times by two users. The 2D offset between the center of the BB and the center of the radiation field was measured for each image.
The average offset for each BB increases with the distance from the isocenter. Over the approximately 10 cm range tested, the average offset (mean ± s) increases from 0.22 ± 0.03 mm to 0.69 ± 0.10mm. The maximum offset measured were 0.70, 1.06, 1.27 and 1.49mm at distances 3.6, 5.4, 7.7 and 9.9cm respectively. For each BB, the 1 mm tolerance is exceeded for only one of the 8 beams, when the couch is rotated. Consistently, our daily W-L results show a maximum offset of 0.9 mm occurs at couch 45°.
An in-house phantom was successfully used to characterize the machine mechanical performance for multi–target single-isocenter frameless SRS.
Brain, Linear Accelerator, Quality Assurance