Author: Jingying Lin, Stevan Pecic, Strahinja Stojadinovic 👨🔬
Affiliation: University of Belgrade, The University of Texas Southwestern Medical Center 🌍
Purpose: This study aims to determine the magnitude of polarization-induced dosimetric error for EBT4 films at different scanning orientations, and to provide guidance on selecting appropriate regions of interest (ROIs) for small-field dosimetry when measuring relative output factors (ROFs) using Leksell Gamma Knife (LGK) systems.
Methods: EBT4 films were positioned at the center plane of LGK spherical dosimetry phantom and irradiated in 15-second increments using a 16 mm collimator to generate a calibration range of [0.7, 10.5] Gy. Subsequently, a separate set of films was irradiated for 2 minutes using 16-, 8- and 4-mm collimators to determine ROFs. EBT4 films were digitized using an EPSON Expression 12000XL scanner at resolutions ranging from 150dpi to 1200dpi in both landscape and portrait orientations. Two calibration curves, generated for portrait and landscape scanning orientations, were used to compute the known absorbed doses in the calibration films and quantify polarization-induced dosimetric errors. Furthermore, full-width half-maximum (FWHM) and penumbra measurements were compared against reference LGP values for all scan resolutions. Finally, LGP ROFs were calculated for a range of ROIs defined relative to the beams' FWHM.
Results: Portrait-scanned films evaluated with the landscape calibration resulted, on average, in a (17.8±4.4)% dose difference, while the converse yielded a (-16.2±4.4)% polarization-induced dose error. FWHM and penumbra measurements for all collimators were within ±0.5 mm relative to LGK reference values, without significant differences across scanning resolutions. The TG-178 recommendation of selecting a "small ROI" for film-based ROF determination was clarified as ¼ FWHM ROI to achieve accuracy within ±1% of LGP values.
Conclusion: Accurate EBT4 film-based dosimetry for Gamma Knife radiosurgery necessitates consistent scanning orientation and appropriate ROI selection. An ROI larger than ¼ FWHM yields underestimated LGP ROFs, analogous to the volume-averaging effect when using a detector that is too large for a small radiation field.