We propose invertible Learned Primal-Dual as a method for tomographic image reconstruction. This is a learned iterative method based on the Learned Primal-Dual neural network architecture, which incorporates ideas from invertible neural networks. The invertibility significantly reduces the GPU memory foot...
The learned primal-dual network structure was used in this study, where the input and output of the network consisted of both low- and high-energy data. The network was trained on 30 patients who went through normal-dose chest DECT scans with simulated noises inserted into the raw data. It...
We propose the Learned Primal-Dual algorithm for tomographic reconstruction. The algorithm accounts for a (possibly non-linear) forward operator in a deep neural network by unrolling a proximal primal-dual optimization method, but where the proximal operators have been replaced with convolutional neural...
We note that this is related to the primal-dual method discussed in [2], where the dual operator is learned in the measurement space. Our learned gradient descent network is visualized in Fig. 2, and may be intuitively explained as follows. We initialize the MPI by feeding t...
Improved image denoising via self-supervised Weickert operator learning and plug-and-play learned Primal Dual Neurocomputing Volume 620, 1 March 2025, Page 129267 Purchase options CorporateFor R&D professionals working in corporate organizations. Academic and personalFor academic or personal use only. Loo...
In the last decade, primal-dual algorithms have become popular due to their ability to employ non-smooth regularisation, which is used to overcome the limited sampling problem in photoacoustic tomography. The algorithm performs updates in both the image domain (primal) and the data domain (dual)...
We reformulate the L1–L1 minimization into an augmented Lagrangian scheme through adding a new auxiliary variable, additionally the dictionary is updated by simply adding the multiplication of dual and primal variables. Experimental results demonstrate that the new proposed method can obtain very ...
We reformulate the L1-L1 minimization into an augmented Lagrangian scheme through adding a new auxiliary variable, additionally the dictionary is updated by simply adding the multiplication of dual and primal variables. Experimental results demonstrate that the new proposed method can obtain very ...
We have adapted, implemented and trained the Learned Primal Dual algorithm suggested by Adler and Öktem and evaluated its performance in reconstructing projection data from our PET scanner. Learned Primal Dual reconstructions are compared to Maximum Likelihood Expectation Maximisation (MLEM) reconstructions...
We have adapted, implemented and trained the Learned Primal Dual algorithm suggested by Adler and Öktem and evaluated its performance in reconstructing projection data from our PET scanner. Learned Primal Dual reconstructions are compared to Maximum Likelihood Expectation Maximisation (MLEM) reconstructions...