Premium
SU‐E‐T‐165: Systematic Evaluation of Uncertainties Associated with GAFCHROMIC EBT2 Film Dosimetry for 6MV Photon Beams
Author(s) -
Kim J,
Kim S,
Shaikh M,
Li H,
Huang Y,
Wen N,
GlideHurst C,
Jin J,
Nurushev T,
Chetty I J
Publication year - 2011
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 - 10.1118/1.3612115
Subject(s) - dosimetry , imaging phantom , scanner , nuclear medicine , optics , irradiation , physics , materials science , radiation , medicine , nuclear physics
Method and Materials:Eight packets of films were exposed to 13.5cm ×13.5cm, 6MV radiation fields in a solid water phantom. Dose levels of 1.1, 3.2, 5.3, 7.4, and 9.5 Gy were delivered to five films in each packet. Films were scanned both before and after irradiation using an Epson flat‐bed scanner (24hr wait‐time for post‐irradiation coloration). Corresponding 2D dose distributions were measured with a detector‐array (MatriXX). Point dose comparisons were performed with an ion chamber. Digitized film images were registered to the 2D dose distribution to generate a correction map that compensated the scanner non‐uniform response as a function of dose. Optical density (OD) and net optical density (NetOD) values were calculated for all images. Dose response curves were established using mean values of a central 0.5cm × 0.5cm region‐of‐interest (ROI). Images were converted to dose, and error uncertainties (1SD) were measured in the central 8cm × 8cm ROI. Results: The overall dosimetric uncertainties (1SD) of the NetOD approach were 2.2%, 1.9%, and 3.5% for red, green, and blue channels, respectively. The corresponding uncertainties of OD were 2.7%, 3.1%, and 8.3%, respectively. For low dose range (<3 Gy), the green channel revealed higher uncertainty (SDgreen= 3.3%) than the red channel (SDred=2.6%). However, for high doses (3∼9 Gy), the green channel showed less variability (SDgreen=1.6%, SDred=2.9%). Minimum SDred and SDgreen were 1.6% at 5.3Gy and 1.3% at 7.4 Gy, respectively. Scanner non‐uniformity correction mitigated the irregular response of scanner detector elements observed initially. Conclusion: NetOD may be a more useful metric for benchmarking EBT2 than OD. We demonstrated that the lowest dose uncertainties were achieved using the red channel for low dose range, while the green channel was preferred for higher doses. Scanner non‐uniformity correction is necessary for higher precision dosimetry.