NC-PDNet: A Density-Compensated Unrolled Network for 2D and 3D Non-Cartesian MRI Reconstruction

• Published in: IEEE transactions on medical imaging 2022-07, Vol.41 (7), p.1625-1638
• Main Authors: Ramzi, Zaccharie, G R, Chaithya, Starck, Jean-Luc, Ciuciu, Philippe
• Format: Article
• Language: English
• Subjects: Ablation; Artificial Intelligence; Bioengineering; Cartesian coordinates; Computer Science; Computer Vision and Pattern Recognition; Deep learning; Density; Fourier transforms; Image Processing; Image reconstruction; Imaging; Iterative methods; Life Sciences; Magnetic resonance imaging; Medical Imaging; MRI; Neural networks; non-cartesian; Optimization; Sensitivity; Signal to noise ratio; Three-dimensional displays; unrolled networks
• ISSN: 0278-0062; 1558-254X
• DOI: 10.1109/TMI.2022.3144619

Deep Learning has become a very promising avenue for magnetic resonance image (MRI) reconstruction. In this work, we explore the potential of unrolled networks for non-Cartesian acquisition settings. We design the NC-PDNet (Non-Cartesian Primal Dual Netwok), the first density-compensated (DCp) unrolled neural network, and validate the need for its key components via an ablation study. Moreover, we conduct some generalizability experiments to test this network in out-of-distribution settings, for example training on knee data and validating on brain data. The results show that NC-PDNet outperforms baseline (U-Net, Deep image prior) models both visually and quantitatively in all settings. In particular, in the 2D multi-coil acquisition scenario, the NC-PDNet provides up to a 1.2 dB improvement in peak signal-to-noise ratio (PSNR) over baseline networks, while also allowing a gain of at least 1dB in PSNR in generalization settings. We provide the open-source implementation of NC-PDNet , and in particular the Non-uniform Fourier Transform in TensorFlow, tested on 2D multi-coil and 3D single-coil k-space data.