Premium
Confidence limit variation for a single IMRT system following the TG119 protocol
Author(s) -
Gordon J. D.,
Krafft S. P.,
Jang S.,
SmithRaymond L.,
Stevie M. Y.,
Hamilton R. 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 - 10.1118/1.3555298
Subject(s) - quality assurance , confidence interval , ionization chamber , imaging phantom , robustness (evolution) , statistics , reproducibility , computer science , nuclear medicine , medical physics , mathematics , medicine , physics , ionization , ion , biochemistry , external quality assessment , chemistry , pathology , quantum mechanics , gene
Purpose: To evaluate the robustness of TG119‐based quality assurance metrics for an IMRT system. Methods: Four planners constructed treatment plans for the five IMRT test cases described in TG119. All plans were delivered to a 30 cm × 30 cm × 15 cm solid water phantom in one treatment session in order to minimize session‐dependent variation from phantom setup, film quality, machine performance, etc. Composite measurements utilized film and an ionization chamber. Per‐field measurements were collected using a diode array device at an effective depth of 5 cm. All data collected were analyzed using the TG119 specifications to determine the confidence limit values for each planner separately and then compared. Results: The mean variance of ion chamber measurements for each planner was within 1.7% of the planned dose. The resulting confidence limits were 3.13%, 1.98%, 3.65%, and 4.39%. Confidence limit values determined by composite film analysis were 8.06%, 13.4%, 9.30%, and 16.5%. Confidence limits from per‐field measurements were 1.55%, 0.00%, 0.00%, and 2.89%. Conclusions: For a single IMRT system, the accuracy assessment provided by TG119‐based quality assurance metrics showed significant variations in the confidence limits between planners across all composite and per‐field evaluations. This observed variation is likely due to the different levels of modulation between each planner's set of plans. Performing the TG119 evaluation using plans produced by a single planner may not provide an adequate estimation of IMRT system accuracy.