Personalizing Mammographic Dosimetry Using Multilayered Anatomy-Based Breast Models

A methodology for patient-oriented calculations of mean glandular dose (MGD) is introduced in this study. The method takes into consideration the influence of the glandular tissue distribution in the MGD. The glandular tissue information was estimated from conventional mammography images using breast density assessment software followed by the Mammography-Image Based (MIB) method presented in this work. The corresponding dose conversion coefficients (DgN−HLB) were determined using a Heterogeneously-Layered Breast (HLB) geometry. The effect of the glandular tissue distribution on the MGD was studied using a set of HLB models and their corresponding homogeneous model. DgN−HLB values were between 48% lower and 24% larger than the value calculated using a homogeneous glandular tissue distribution, despite the current methods predicting the same coefficient for all glandular tissue distributions. The proposed methods were applied to a group of patients. For the cases analyzed, the variation in MGD was as large as 14.8% for a highly heterogeneous dense breast.