Digital breast tomosynthesis (DBT) has the potential to en-
hance breast cancer detection by reducing the confounding e ect of su-
perimposed tissue associated with conventional mammography. In addi-
tion the increased volumetric information should enable temporal datasets
to be more accurately compared, a task that radiologists routinely apply
to conventional mammograms to detect the changes associated with ma-
lignancy. In this paper we address the problem of comparing DBT data
by combining reconstruction of a pair of temporal volumes with their reg-
istration. Using a simple test object, and DBT simulations from in vivo
breast compressions imaged using MRI, we demonstrate that this com-
bined reconstruction and registration approach produces improvements
in both the reconstructed volumes and the estimated transformation pa-
rameters when compared to performing the tasks sequentially.