Nausikaa Geeraert MSc, DiplPhys, Presenter: Employee, General Electric Company

Remy Klausz DiplEng, Abstract Co-Author: Employee, General Electric Company

Lionel Desponds PhD, Abstract Co-Author: Employee, General Electric Company

Serge L. Muller PhD, Abstract Co-Author: Employee, General Electric Company

Isabelle Bloch, Abstract Co-Author: Nothing to Disclose

Hilde Bosmans PhD, Abstract Co-Author: Co-founder, Qaelum NV Research Grant, Siemens AG

Originally introduced by Hammerstein in 1979 for comparing doses delivered with different radiographic techniques to the average breast in mammography, Average Glandular Dose (AGD) is the dosimetry quantity generally recommended for collective risk assessment. Conversion factors from Entrance Surface Air Kerma (ESAK) to AGD were computed by different authors using Monte Carlo computation on reference breast models of homogeneous glandularity. We want to investigate more in detail AGD applicability to individual risk assessment and in particular the influence of the breast glandular quantity and distribution.

Four semi-cilindrical phantoms are defined. Phantom 1 is the reference breast model used by Dance (1990), i.e. 8 cm radius, 45 mm thick, 10 mm skin, 50% homogeneous glandularity intra-skin. Phantom 2 differs from 1 by a doubled surface (11.3 cm radius). Phantom 3 glandularity is 0%. Phantom 4 has 50% average intra-skin glandularity, but contains a 75cm² x 3.5cm region with 59% glandularity embedded in 0% glandularity resulting in a surfacic glandularity (SG) of 75%, a peak glandularity (PG) as used by the automatic exposure control of 46%, a thickness-based glandularity (TG) from tables in Dance (2000) of 41% and an image-based VG of 34% including skin. AGDs are compared for the same exposure (0.45 mm Al Half-Value-Layer) resulting in an AGD of 1mGy for phantom 1.

AGDs are 1mGy (phantoms 1 and 2), 1.2mGy (phantom 3). AGD for phantom 3 is the highest in spite there is no glandular tissue at risk. AGD for phantom 2 is the same as phantom 1; however the quantity of glandular tissue at risk is doubled. AGD of phantom 4 depends on the selected glandularity: 0.90 mGy (SG), 1.03 mGy (PG), 1.05 mGy (TG) and 1.08 mGy (VG).

As shown by our phantom studies, AGD does not reflect the individual risk. For AGD computation the glandular content is not correctly described by a single quantity. Recently developed mammographic image-based glandularity computation methods could be used to improve the AGD computation. Moreover access to the description of glandular quantity and distribution will make the practical development of metrics for individual risk assessment possible.

Demonstrate limitations of the current use of AGD and propose improvements to individual radiation risk assessment using image-based volumetric glandularity computations.

Geeraert, N, Klausz, R, Desponds, L, Muller, S, Bloch, I, Bosmans, H, Impact of Breast Glandular Description on Average Glandular Dose and Radiation Risk Assessment in Mammography.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13044182.html