NC-PDNet: A density-compensated unrolled network for 2D and 3D non-Cartesian MRI reconstruction

Abstract

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.

Details

• Type of work: Journal article
• Venue: IEEE Transactions on Medical Imaging
• Published: 18 January 2022
• Volume: 41
• Pages: 1625—1638
• DOI: 10.1109/tmi.2022.3144619
• Authors: Zaccharie Ramzi, Chaithya Giliyar Radhakrishna, Jean-Luc Starck, and Philippe Ciuciu
• Cited by: 63 times (OpenAlex)
• Data sources: ref.bib, OpenAlex

Links

• Publisher (DOI)

BibTeX

@article{ramzi2022nc,
  title={NC-PDNet: A density-compensated unrolled network for 2D and 3D non-Cartesian MRI reconstruction},
  author={Ramzi, Zaccharie and Chaithya, GR and Starck, Jean-Luc and Ciuciu, Philippe},
  journal={IEEE Transactions on Medical Imaging},
  volume={41},
  number={7},
  pages={1625--1638},
  year={2022},
  publisher={IEEE}
}