Monte Carlo calculation of monoenergetic and polyenergetic DgN coefficients for mean glandular dose estimates in mammography using a homogeneous breast model

We computed normalized glandular dose (DgN) coefficients for mean glandular dose estimates in contemporary 2D mammography units, taking into account a homogeneous model for the breast which reflects recent literature reports. We developed a Monte Carlo code based on the simulation toolkit GEANT4 ver. 10.00. The breast was modelled as a cylinder with a semi-cylindrical section with a radius of 10 cm, enveloped in a 1.45 mm thick skin layer, as found out in recent reports in the analysis of breast computed tomography clinical scans. The compressed breast thickness was between 3 cm and 8 cm. The DgN coefficients were calculated for monoenergetic x-ray beams between 4.25 keV and 49.25 keV and were fitted with polynomial curves. Polyenergetic DgN coefficients were then computed for spectra obtained for various anode/filter combinations as adopted in routine clinical practice: Mo/Mo 30 m (25-40 kV), Mo/Rh 25 m (25-40 kV), Rh/Rh 25 m (25-40 kV), W/Ag 50 m (26-34 kV), W/Al 500 m (26-38 kV), W/Al 700 m (28-40 kV) and W/Rh 50 m (24-35 kV). Monoenergetic DgN curve fit coefficients and polyenergetic DgNp coefficients were released for research and clinical work. Polyenergetic DgNp coefficients were 6% higher than those provided in the recent literature, on average. The differences range between -18% and 30%; up to 50% of the computed coefficients differed by less than 10%. The dataset of DgN coefficients are provided as tables for varying glandular fraction by mass and compressed breast thickness. Moreover, a computer code has been developed for generating user specific coefficients DgNp for user defined x-ray spectra up to 49 kV, calculated by spectral weighting from the dataset of monoenergetic DgN coefficients.