no code implementations • 6 Oct 2023 • Florent Sureau, Mahdi Latreche, Marion Savanier, Claude Comtat
In this work, we investigate hybrid PET reconstruction algorithms based on coupling a model-based variational reconstruction and the application of a separately learnt Deep Neural Network operator (DNN) in an ADMM Plug and Play framework.
no code implementations • 27 Sep 2022 • Marion Savanier, Emilie Chouzenoux, Jean-Christophe Pesquet, Cyril Riddell
We unfold the Dual Block coordinate Forward-Backward (DBFB) algorithm, embedded in an iterative reweighted scheme, allowing the learning of key parameters in a supervised manner.