Room: Exhibit Hall | Forum 3
Purpose: To demonstrate the use and validation of the dosimeter-specific correction factors provided by the newly published IAEA TRS 483 code of practice in small field dosimetry for MLC and stereotactic cone commissioning.
Methods: C-arm gantry linear accelerator small field output factors (OF) for MLC opening size from 5 mm to 3 cm with multiple back up jaw sizes, and stereotactic cones were acquired at 3 institutions with a total of six commercial small field dosimeters. Detector and field size specific corrections factors in Table 26 of the IAEA TRS 483 report were utilized to correct for all readings. Correction factors corresponding to the equivalent square field size were used for each measurement. For field sizes that did not have correction factors tabulated in the IAEA report, interpolation through polynomial fit was performed. The uncorrected raw data and the corrected OF were compared among the utilized detectors at each institution.
Results: For the MLC small field OF measurements, the correction factors have brought the maximum absolute differences in readings substantially closer (2.23% to 0.19% for institution 1 and 4.22% to 1.2% for institution 2). Excellent agreement was shown between all corrected readings and measurements from dosimeters that do not require correction factors such as plastic scintillator detector and Gafchromic film. However, for the measurements corresponding to the smallest MLC square field of 5 mm, the corrected OF values were consistently below the vendor-suggested historical range for both institutions, regardless of detector type. For cone OF, the correction factors resulted in improved maximum absolute difference from 3.5% to 1.3 % for institution 1 (2 detectors), and from 11% to 2% for institution 3 (3 detectors).
Conclusion: The study results motivate change in practice and further clinical implementation of the IAEA TRS 483 code of practice for small field dosimetry characterization.
Funding Support, Disclosures, and Conflict of Interest: This work is supported by BrainLab