# Image Reconstruction for MRI using Deep CNN Priors Trained without Groundtruth

arXiv: https://arxiv.org/abs/2204.04771 · HTML: https://mlsanity.com/paper/2204.04771/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Apr 2022.

## Built on by (most cited first, 4 of 4)

- [A Self-Supervised Approach to Reconstruction in Sparse X-Ray Computed Tomography](https://mlsanity.com/paper/2211.00002/) (2022)
- [Efficient Model-Based Deep Learning via Network Pruning and Fine-Tuning](https://mlsanity.com/paper/2311.02003/) (2023)
- [Self-supervised Deep Hyperspectral Inpainting with the Plug and Play and Deep Image Prior Models](https://mlsanity.com/paper/2501.08195/) (2025)
- [Regularization by Neural Style Transfer for MRI Field-Transfer Reconstruction with Limited Data](https://mlsanity.com/paper/2308.10968/) (2023)

## Abstract (arXiv)

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.