Development of a Quality Control Protocol for Digital Breast Tomosynthesis Systems in the TOMMY Trial

Abstract

Seven Hologic Dimensions digital breast tomosynthesis (DBT) systems have been installed for the TOMMY trial, a UK based multi-centre trial comparing conventional 2D digital mammography with DBT. In the absence of established guidelines for DBT quality control, a specific protocol was developed and applied. Physics tests of 2D and DBT performance were conducted at baseline and are repeated every 6 months, and include dose to the European standard breast model, tomosynthesis contrast to noise ratio, geometric distortion, z-resolution and threshold contrast detail detection. In addition routine performance checks (daily, weekly and monthly) are conducted on each system and reviewed centrally. Doses delivered under automatic exposure control by the systems were found to be well matched, with a mean glandular dose for the standard breast model (53mm equivalent breast thickness) of 1.89 mGy (range 1.79 to 2.00 mGy) for DBT and 1.41 mGy (range 1.36 to 1.48 mGy) for 2D imaging. Detector performance and image quality measurements were also well matched. All the systems exceeded the achievable image quality standard in the European guidelines for conventional digital mammography.