DIR‐DBTnet: Deep iterative reconstruction network for three‐dimensional digital breast tomosynthesis imaging

Abstract

The goal of this study is to develop a three-dimensional (3D) iterative reconstruction framework based on the deep learning (DL) technique to improve the digital breast tomosynthesis (DBT) imaging performance. Methods: In this work, the DIR-DBTnet is developed for DBT image reconstruction by mapping the conventional iterative reconstruction (IR) algorithm to the deep neural network. By design, the DIR-DBTnet learns and optimizes the regularizer and the iteration parameters automatically during the network training with a large amount of simulated DBT data. Numerical, experimental, and clinical data are used to evaluate its performance. Quantitative metrics such as the artifact spread function (ASF), breast density, and the signal difference to noise ratio (SDNR) are measured to assess the image quality. Results: Results show that the proposed DIR-DBTnet is able to reduce the in-plane shadow artifacts and the out-of-plane signal leaking artifacts compared to the filtered backprojection (FBP) and the total variation (TV)-based IR methods. Quantitatively, the full width half maximum (FWHM) of the measured ASF from the clinical data is 27.1% and 23.0% smaller than those obtained with the FBP and TV methods, while the SDNR is increased by 194.5% and 21.8%, respectively. In addition, the breast density obtained from the DIR-DBTnet network is more accurate and consistent with the ground truth. Conclusions: In conclusion, a deep iterative reconstruction network, DIR-DBTnet, has been proposed for 3D DBT image reconstruction. Both qualitative and quantitative analyses of the numerical, experimental, and clinical results demonstrate that the DIR-DBTnet has superior DBT imaging performance than the conventional algorithms.

Low-dose CT imaging performance of U-Net for varied spatial resolutions

“…A numerical 3D binary phantoms 5,6 made of two different materials, PMMA and adipose, are simulated. 7 The energy-dependent attenuation coefficients of these materials are obtained from the NIST (National Institute of Standards and Technology, USA).…”

New reconstruction methodology for chest tomosynthesis based on deep learning

“…(1) a post-processing step, after reconstruction, 4 (2) a pre-processing of the projection data, 5 or (3) within an IR algorithm, in the so-called unrolled methods. 6,7 Nevertheless, despite further improving image quality over traditional IR, none of the proposed methods show true tomographic results.…”

VVBP-tensor-based deep learning framework for high-attenuation artifact reduction in digital breast tomosynthesis

“…They can be divided into three types, including image domain based methods, 2, 3 projection domain based methods 4,5 and deep iterative reconstruction methods. [6][7][8] First, the image domain based methods establish models based on the characteristics of noise and artifacts in DBT image volume, and reduce the corresponding noise and artifacts. For example, Mota et al used blind deconvolution and total variation (TV) minimization to reduce out-of-plane artifacts.…”

“…9 For example, Su et al proposed a DIR-DBTnet to improve the DBT imaging performance, which maps the ADMM optimization algorithm to the DNN model. 8 However, the domain transformation from projection domain to image domain is hard to be accurately realized by the DNN model and easily introduces new artifacts in reconstructed DBT images. 10 In summary, it is still a challenge to effectively reduce high-attenuation artifacts and preserve texture appearance of background region in DBT images.…”