A data-driven method for automatic regularization selection in a hybrid DL-SENSE reconstruction
Zahra Hosseini1, Thorsten Feiweier2, John Conklin3, Stephan Kannengiesser2, Marcel Dominik Nickel2, Min Lang3, Azadeh Tabari3, Augusto Lio Concalves Filho3, Wei-Ching Lo4, Maria Gabriela Figueiro Longo3, Michael Lev3, Pamela Schaefer3, Otto Rapalino3, Susie Huang3, Stephen Cauley5, and Bryan Clifford4
1MR R&D Collaboration, Siemens Medical Solutions USA, Atlanta, GA, United States, 2Siemens Healthcare GmbH, Erlangen, Germany, 3Department of Radiology, Massachusetts General Hospital, Boston, MA, United States, 4MR R&D Collaboration, Siemens Medical Solutions USA, Boston, MA, United States, 5Department of Radiology, A. A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States
The integration of deep learning priors into regularized CG-SENSE reconstructions enables high quality MR images to be generated from noisy, undersampled data. The regularization parameter in these methods can be tuned to control the level of denoising, allowing a network to generalize to novel SNR conditions without retraining. However, manual tuning of the regularization parameter can be time consuming. This work presents a data-driven method for automatic regularization selection using commonly acquired noise calibration data. Results indicate the method generalizes across clinically relevant imaging scenarios and provides diagnostically equivalent image quality to that obtained by manual parameter tuning.
This abstract and the presentation materials are available to members only;
a login is required.