An investigation into the impact of anatomical variation upon mean glandular dose produced within a standard breast

Abstract

By using the EGS4 Monte Carlo computer program the normalised mean glandular dose (MGD) was calculated for a breast model that is intended to reflect the composition of an "average" breast. The reliability of the calculation was established by comparing the predictions with previously published values for a breast model. The breast model used was then altered in order to reflect the possible extremes in glandular distribution that could occur within a compressed breast. These results show that the MGD could vary by up to a factor of four depending upon where the majority of glandular tissue is located within the breast. The impact of this variation in MGD upon risk assessment within mammography is then discussed.