Gamma Analysis of Normalized and Un-Normalized Dose Distributions
S Stojadinovic*, O Luo, Q Bao, A Pompos, X Gu, T Solberg, UT Southwestern Medical Center, DALLAS, TXSU-E-T-386 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall
Purpose: The gamma index method, as currently implemented in all commercial QA software, calls for selection of a normalization point to evaluate agreement between two dose distributions. The implication of this is that there is an infinite number of possible solutions! Which one to pick? A unique and more relevant solution is obtained only if no normalization point is used.
Methods and Materials: The set of test cases suggested by the AAPM TG119 were planned using Pinnacle 8.0m and delivered on a Varian 21EX linac for 6 and 18 MV photons. The recommended point and planar dose measurements were obtained using a Pinpoint ion chamber, EDR2 film and MatriXX. The gamma index method using typical 3%, 3 mm criteria with and without a normalization point was used to assess the agreement between calculated and delivered planar dose distributions. The analysis was extended to a set of data for clinically treated patients.
Results: The comparison with the TG119 benchmark data showed that all point dose and planar measurements for 6 MV were within the published range. Similar results, although without published data to compare with, were obtained for 18 MV as well. For all complex tests, the percentage of points passing the gamma criteria of 3%, 3 mm was (95.8±1.6)% and (95.6±1.0)% for 6 MV and 18 MV, respectively. Without a normalization point, however, the same gamma analysis fell to (20.7±6.7)% and (13.9±4.0)% for 6 MV and 18 MV, respectively. The clinical data set showed the same trend, with the gamma passing rate declining from (98.9±0.7)% to (33.4±13.1)%.
Conclusion: The gamma index method provides a unique answer for gamma passing rate only without normalizing dose distributions to any particular point. The common gamma criteria of 3%, 3 mm, however, is a very poor metric in that case.