Return To Abstract Listing

Breast Imaging

E2840. Dose Delivered to Patients Between Tomosynthesis and Standard Compression Views in Mammography

Vincenti K1,  Szpakowski P1,  Barufaldi B2,  Levin D1,  Maidment A.2 1. Pennsylvania Hospital, Philadelphia, PA; 2. Hospital of the University of Pennsylvania, Philadelphia, PA

Address correspondence to K. Vincenti (

Objective: The purpose of this study was to evaluate the dosimetric implication of switching from full field digital mammography (FFDM) to digital breast tomosynthesis (DBT) using the Mammomat Inspiration System with tomosynthesis (Siemens Healthcare).

Materials and Methods: Tomosynthesis screening was initiated at our institution on December 1, 2015. As of March 15, 2016, 437 patients had been screened using the Mammomat Inspiration system. In the preceding year, 8485 patients were screened using traditional 2D digital mammography. The average glandular dose (AGD) was extracted by custom automated software. The software tracks radiation dose per image in terms of approximately 60 data elements including patient age, breast thickness, x-ray spectrum, procedure identifier, and view position. The dose per patient is obtained by summing over each image per breast and averaging between breasts. In total, the AGDs of 8922 patients were compared while accounting for type of view and breast thickness.

Results: Initial data in 8485 patients undergoing FFDM show that the mean AGD ± SD was 1.69 ± 0.71 and 1.94 ± 0.87 mGy for craniocaudal (CC) and mediolateral oblique (MLO) images, respectively. In 437 patients undergoing DBT, the AGD of the 3D component was 1.90 ± 0.62 and 2.08 ± 0.74 mGy for CC and MLO images, respectively. Average combined compressed breast thickness (CBT) was 49.9 mm for FFDM and 56.5 mm for DBT. Average radiation dose increased with CBT and decreased with age, consistent with expectations. Examining dose as a function of CBT showed that DBT had slightly lower dose for breast thickness below 50 mm, and FFDM had slightly lower dose for breast thickness above 50 mm. When considered in aggregate for the population examined, the 2D FFDM and 3D DBT dose did not substantially differ when corrected for CBT. However, given that the DBT study comprises 2D and 3D imaging, the total DBT dose was twice the FFDM dose. Comparison with an existing dataset of 15,465 patients screened using the Selenia Dimensions mammography system (Hologic Inc) demonstrated a 12% higher average radiation dose for the Mammomat DBT studies in comparison with Hologic DBT studies.

Conclusion: The average radiation dose for DBT is approximately twice that for FFDM in studies performed using the Mammomat System. DBT imaging roughly doubles the dose to the patient. If synthesized mammograms are approved for the Mammomat System, overall dose for DBT studies can be substantially reduced.