Deep learning reconstruction of digital breast tomosynthesis images for accurate breast density and patient-specific radiation dose estimation
Authors:
Jonas Teuwen,
Nikita Moriakov,
Christian Fedon,
Marco Caballo,
Ingrid Reiser,
Pedrag Bakic,
Eloy García,
Oliver Diaz,
Koen Michielsen,
Ioannis Sechopoulos
Abstract:
The two-dimensional nature of mammography makes estimation of the overall breast density challenging, and estimation of the true patient-specific radiation dose impossible. Digital breast tomosynthesis (DBT), a pseudo-3D technique, is now commonly used in breast cancer screening and diagnostics. Still, the severely limited 3rd dimension information in DBT has not been used, until now, to estimate…
▽ More
The two-dimensional nature of mammography makes estimation of the overall breast density challenging, and estimation of the true patient-specific radiation dose impossible. Digital breast tomosynthesis (DBT), a pseudo-3D technique, is now commonly used in breast cancer screening and diagnostics. Still, the severely limited 3rd dimension information in DBT has not been used, until now, to estimate the true breast density or the patient-specific dose. This study proposes a reconstruction algorithm for DBT based on deep learning specifically optimized for these tasks. The algorithm, which we name DBToR, is based on unrolling a proximal-dual optimization method. The proximal operators are replaced with convolutional neural networks and prior knowledge is included in the model. This extends previous work on a deep learning-based reconstruction model by providing both the primal and the dual blocks with breast thickness information, which is available in DBT. Training and testing of the model were performed using virtual patient phantoms from two different sources. Reconstruction performance, and accuracy in estimation of breast density and radiation dose, were estimated, showing high accuracy (density <+/-3%; dose <+/-20%) without bias, significantly improving on the current state-of-the-art. This work also lays the groundwork for develo** a deep learning-based reconstruction algorithm for the task of image interpretation by radiologists.
△ Less
Submitted 29 March, 2021; v1 submitted 11 June, 2020;
originally announced June 2020.
