Premium
SU‐F‐T‐476: Performance of the AS1200 EPID for Periodic Photon Quality Assurance
Author(s) -
DeMarco J,
Moran J,
Barnes M,
Greer P,
Kim G,
Fraass B,
Yang W,
McKenzie Boehnke E
Publication year - 2016
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4956661
Subject(s) - linear particle accelerator , quality assurance , dosimetry , monitor unit , optics , flatness (cosmology) , dicom , standard deviation , truebeam , image guided radiation therapy , physics , photon , particle accelerator , nuclear medicine , beam (structure) , computer science , medical imaging , mathematics , artificial intelligence , statistics , engineering , medicine , operations management , external quality assessment , cosmology , quantum mechanics
Purpose: To assess the dosimetric performance of a new amorphous silicon flat‐panel electronic portal imaging device (EPID) suitable for high‐intensity, flattening‐filter‐free delivery mode. Methods: An EPID‐based QA suite was created with automation to periodically monitor photon central‐axis output and two‐dimensional beam profile constancy as a function of gantry angle and dose‐rate. A Varian TrueBeamTM linear accelerator installed with Developer Mode was used to customize and deliver XML script routines for the QA suite using the dosimetry mode image acquisition for an aS1200 EPID. Automatic post‐processing software was developed to analyze the resulting DICOM images. Results: The EPID was used to monitor photon beam output constancy (central‐axis), flatness, and symmetry over a period of 10 months for four photon beam energies (6x, 15x, 6xFFF, and 10xFFF). EPID results were consistent to those measured with a standard daily QA check device. At the four cardinal gantry angles, the standard deviation of the EPID central‐axis output was <0.5%. Likewise, EPID measurements were independent for the wide range of dose rates (including up to 2400 mu/min for 10xFFF) studied with a standard deviation of <0.8% relative to the nominal dose rate for each energy. Also, profile constancy and field size measurements showed good agreement with the reference acquisition of 0° gantry angle and nominal dose rate. XML script files were also tested for MU linearity and picket‐fence delivery. Using Developer Mode, the test suite was delivered in <60 minutes for all 4 photon energies with 4 dose rates per energy and 5 picket‐fence acquisitions. Conclusion: Dosimetry image acquisition using a new EPID was found to be accurate for standard and high‐intensity photon beams over a broad range of dose rates over 10 months. Developer Mode provided an efficient platform to customize the EPID acquisitions by using custom script files which significantly reduced the time. This work was funded in part by Varian Medical Systems.