English

Image Reconstruction for MRI using Deep CNN Priors Trained without Groundtruth

Image and Video Processing 2022-04-12 v1

Abstract

We propose a new plug-and-play priors (PnP) based MR image reconstruction method that systematically enforces data consistency while also exploiting deep-learning priors. Our prior is specified through a convolutional neural network (CNN) trained without any artifact-free ground truth to remove undersampling artifacts from MR images. The results on reconstructing free-breathing MRI data into ten respiratory phases show that the method can form high-quality 4D images from severely undersampled measurements corresponding to acquisitions of about 1 and 2 minutes in length. The results also highlight the competitive performance of the method compared to several popular alternatives, including the TGV regularization and traditional UNet3D.

Keywords

Cite

@article{arxiv.2204.04771,
  title  = {Image Reconstruction for MRI using Deep CNN Priors Trained without Groundtruth},
  author = {Weijie Gan and Cihat Eldeniz and Jiaming Liu and Sihao Chen and Hongyu An and Ulugbek S. Kamilov},
  journal= {arXiv preprint arXiv:2204.04771},
  year   = {2022}
}
R2 v1 2026-06-24T10:43:50.096Z