Monte Carlo simulation of conversion coefficients for dosimetry in mammography

Abstract

The mean glandular dose (MGD) is considered the most adequate dosimetric quantity to assess the risks associated with mammography, since glandular tissue is the most radiosensitive one. However, the MGD cannot be measured directly due to the heterogeneous distribution of breast tissues. The well-established method for determine this quantity is based on the product of incident air kerma (K_air) in the breast by conversion coefficients, called normalized glandular dose (DgN) and calculated by the Monte Carlo (MC) method. However, the DgN coefficients have a wide variation with the characteristics of the breast and the incident spectrum, leading to the necessity of using complex tables. In this work, the correlation between MGD and other dosimetric quantities (mean dose in the whole breast, mean dose in the skin and mean dose in the homogeneous tissue) was studied, using the code MC PENELOPE (v. 2018) + PenEasy (v. 2020), in order to determine new conversion coefficients in mammography. The results of this study showed that, in the energy range of interest in mammography (15-25 keV), the new conversion coefficients obtained from mean dose measurements in the whole breast or in the homogeneous tissue are approximately constant for each breast characteristic, while the new conversion coefficients of the mean dose measured in the skin showed a high variability in this range. Thus, these conversion coefficients may be useful for estimating the MGD from the experimental dose assessment based on dose distribution in phantoms, and can be used as an alternative/complementary to the DgN coefficients.