Premium
SU‐FF‐I‐39: Estimating Conversion Coefficient of KERMA Free in Air to Glandular Dose in Mammography: A Comparison Between BR12 Model and a Realistic Voxel Model
Author(s) -
Hoff G,
de Almeida C,
Drexler G
Publication year - 2006
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 - 10.1118/1.2240279
Subject(s) - kerma , voxel , mammography , monte carlo method , nuclear medicine , materials science , physics , collimator , optics , computational physics , dosimetry , mathematics , medicine , radiology , statistics , cancer , breast cancer
Purpose: To compare conversion coefficient of KERMA free in air to glandular dose, in mammography, simulated to BR12 model and a realistic breast voxel model. Method and Materials: We simulate the glandular dose (D g ) and KERMA free in air (K ai ), using the Monte Carlo program MCNP (version 4B) to estimate the conversion coefficient (c g ) of KERMA free in air at entrance skin in glandular dose. The computational universe generated to simulate a mammographic procedure mimics LORAD III mammographic equipment. The focal spot of molybdenum irradiates photons isotropically in a solid angle of 16.8°. The bucky is 0.6190 m far from de focal spot. Above the model there is a PMMA compress paddle 0.002 m thicker. The add filtration (30 μm Mo thicker and 25 μm Rh thicker) was located at 0.050 m far from the focal spot. Tow spectra were used in voxel model simulations: 28 kVp with Mo add filtration and 30 kVp with Rh add filtration. Results: The c g presented on Mo/Rh simulation was 1.5 times larger than the presented on Mo/Mo simulation. Comparing the voxel model to the BR12 model we have actually a super estimation on both simulated c g values: 3.4 times considering the simulation with Mo/Mo target/filter combination, and 2.4 times considering the simulation with Mo/Rh target/filter combination. Conclusion: The c g values show a decrease of 58.7% considering the Mo/Rh target/filter combination and a decrease of 70.2% considering the Mo/Mo target/filter combination, to the realistic breast model as comparative pattern. These variation on c g are probably caused by the definition of a non‐anthropomorphic model composed by an homogeneous distribution of tissues as pattern, that makes unviable the observation of the absorbed energy by each tissue; and because this model do not consider the position of glandular tissue in the real breast geometry.