(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose: To investigate the source of consistent -up to 2mm longitudinal and lateral- MLC shift observations for 10 Varian (6 C-series and 4 TrueBeam) linear accelerators VMAT weekly MLC QA using an electronic portal imaging device (EPID).
Method: Picket fence test images of VMAT vs. static field with the same MLC sequences are routinely compared for weekly MLC QA at our institution. Our test method is as follows: first, the MLC picket fence pattern is acquired with the static gantry at four cardinal angles and collimator angles at 0º and 90º. Second, the same MLC pattern is acquired with VMAT delivery with gantry speed in two ranges (0.5-1.8 º/s and 1.8-4.8 º/s), to find out if the geometric difference is from MLC or the EPID. As a stationary reference point, a front pointer is placed on the couch and aligned to the rotation isocenter, and its shadow projected on all acquired MLC picket fence pattern.
Results: EPID sag errors on MLC patterns mostly appear on our Trilogy machines. All 4 TrueBeam machines pass the daily machine performance check (MPC), which uses EPID within a 1 mm error. For errors related to EPID sag, different gantry speed showed the same differences, which suggests it is gantry speed independent. Different images from different collimator or gantry angles indicate that the mismatch is from the EPID motion for different gantry angles. The difference was up to 1 mm.
Conclusions: The EPID motion of up to 1mm for different gantry angles during VMAT is not easily detected even with the daily MPC. Considering this finding, the meaning of the Gamma passing rate of EPID-based VMAT verification system needs to be revisited.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
