Workload and transmission data for the installation of a digital breast tomosynthesis system

Digital breast tomosynthesis (DBT) differs from conventional mammography in target/filter, kVp range, and imaging geometry. The aim of this study was to assess the breast input exposure of a DBT system by completing a workload survey of DBT installations, and to determine the parameters α, β, and γ in the Archer equation for the primary radiation generated by the clinical workload distributions.

The authors conducted a retrospective survey of the x-ray breast imaging performed between September 2011 and September 2012 in three clinical DBT rooms equipped with Selenia Dimensions systems (Hologic Inc., Bedford, MA). A total of 343 examinations were analyzed to calculate the workload (mA-minute) and the primary air kerma at 1 m from the source (K(1)). Transmission curves were calculated for the primary radiation generated by the workload distributions of the DBT rooms, and were fitted to the Archer equation.

There were large variations in patient volume and workload in the three examination rooms. In all these rooms, the average tube voltage (kVp) was about 31, the average K(1) per patient was 16-21 mGy, and the average mA-minute per patient was 1.4-2.2 times higher than that of the mammography room described in NCRP Report No. 147. Most DBT screening examinations consisted of four two-dimensional mammographic views plus four tomosynthesis scans; the numbers of views acquired in diagnostic examinations varied widely. Tomosynthesis scans contributed about 30% of total mA-minute and about 50% of K(1). For the primary radiation generated by the clinical workload distributions, α was similar to that of 40-45 kVp W/Al (target/filter), and α + β was similar to that of 30 kVp W/Al.

The workload (mA-minute and K(1)) distributions of mammographic examinations with DBT differ from conventional mammography. A field survey of patient volume and x-ray tube usage is important for the shielding design of DBT facilities. The clinical workloads and transmission data presented in this paper may be useful for clinical medical physicists.