Maximizing the Spatial Resolution of the MatriXX Device for Radiosurgery Plan QA
K Tierney*, C Coffey, M Morales, G Ding, Vanderbilt University, Nashville, TNSU-E-T-45 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: The MatriXX array detector array has been shown to provide inadequate sampling and absolute dose measurements for small, highly modulated fields. These devices generally fail due to spatial frequency contributions exceeding the sampling rate as defined by the Nyquist criteria and maximum point doses falling between chamber locations. This study attempts to demonstrate the feasibility of using an extended source detector distance (ESDD) measurement combined with array shifting to maximize MatriXX resolution for the quality assurance of radiosurgery fields.
Methods: Measurements were taken using a Novalis TX with HD-MLCs at an extended 202 cm SDD with 3.8 mm shifts in the radial and transverse directions. A VBA code was developed to sum the four shifted images into a single dose plane. A line pair pattern with decreasing MLC openings from 1.25 to 0.25 cm was used to quantify the MatriXX resolution using conventional and shifted ESDD techniques. Next, the quality assurance of an intensity modulated radiosurgery (IMRS) plan was measured conventionally with the MatriXX and portal imaging device, and then compared to the shifted ESDD technique using a 3%, 3 mm gamma analysis and point dose comparison.
Results: The ESDD technique increased the effective sampling frequency from 1.3 to 5.6 cm⁻¹ allowing for the accurate representation of frequency contributions up to 2 cm⁻¹ corresponding to dose fluctuations from single HD-MLC openings. The gamma passing rate of the ESDD measured IMRS fields averaged 95.4% compared to 86.4% for conventional SDD and 99.5% for portal dosimetry. Absolute point doses differences improved on average from 4.3% and 3.3%, for conventional SDD and portal respectively, to 2.7% for the ESSD technique.
Conclusion: This study demonstrates that the resolution of the MatriXX can be improved by a factor of four with extended SDD and array shifting techniques allowing for improved QA of radiosurgery plans.