mlsanityData to 7 Oct 2026

Dual-Domain Self-Supervised Learning for Accelerated Non-Cartesian MRI Reconstruction

arXiv 2302.09244 ↗First posted 18 Feb 2023eess.IV cs.CV cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

While enabling accelerated acquisition and improved reconstruction accuracy, current deep MRI reconstruction networks are typically supervised, require fully sampled data, and are limited to Cartesian sampling patterns. These factors limit their practical adoption as fully-sampled MRI is prohibitively time-consuming to acquire clinically. Further, non-Cartesian sampling patterns are particularly desirable as they are more amenable to acceleration and show improved motion robustness. To this end, we present a fully self-supervised approach for accelerated non-Cartesian MRI reconstruction which leverages self-supervision in both k-space and image domains. In training, the undersampled data are split into disjoint k-space domain partitions. For the k-space self-supervision, we train a network to reconstruct the input undersampled data from both the disjoint partitions and from itself. For the image-level self-supervision, we enforce appearance consistency obtained from the original undersampled data and the two partitions. Experimental results on our simulated multi-coil non-Cartesian MRI dataset demonstrate that DDSS can generate high-quality reconstruction that approaches the accuracy of the fully supervised reconstruction, outperforming previous baseline methods. Finally, DDSS is shown to scale to highly challenging real-world clinical MRI reconstruction acquired on a portable low-field (0.064 T) MRI scanner with no data available for supervised training while demonstrating improved image quality as compared to traditional reconstruction, as determined by a radiologist study.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 26 of its 63 references that are in the map. First posted on arXiv 18 Feb 2023.

Builds on

26 of 63 references.

References · 26

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 464
  4. 2017MoDL: Model Based Deep Learning Architecture for Inverse Problemscited by 384
  5. 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 340
  6. 2017A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstructioncited by 231
  7. 2017Image reconstruction by domain transform manifold learningcited by 231
  8. 2020End-to-End Variational Networks for Accelerated MRI Reconstructioncited by 201
  9. 2021ResViT: Residual vision transformers for multi-modal medical image synthesiscited by 177
  10. 2017Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstructioncited by 105
  11. 2017Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Losscited by 82
  12. 2021Unsupervised MRI Reconstruction via Zero-Shot Learned Adversarial Transformerscited by 76
14 more
  1. 2020DuDoRNet: Learning a Dual-Domain Recurrent Network for Fast MRI Reconstruction with Deep T1 Priorcited by 72
  2. 2019Reducing Uncertainty in Undersampled MRI Reconstruction with Active Acquisitioncited by 55
  3. 2021Zero-Shot Self-Supervised Learning for MRI Reconstructioncited by 53
  4. 2017A Transfer-Learning Approach for Accelerated MRI using Deep Neural Networkscited by 43
  5. 2022DSFormer: A Dual-domain Self-supervised Transformer for Accelerated Multi-contrast MRI Reconstructioncited by 30
  6. 2021Over-and-Under Complete Convolutional RNN for MRI Reconstructioncited by 29
  7. 2021Self-Supervised Learning for MRI Reconstruction with a Parallel Network Training Frameworkcited by 26
  8. 2020Unsupervised MRI Reconstruction with Generative Adversarial Networkscited by 24
  9. 2020Neural Network-based Reconstruction in Compressed Sensing MRI Without Fully-sampled Training Datacited by 14
  10. 2021Dual-Octave Convolution for Accelerated Parallel MR Image Reconstructioncited by 12
  11. 2020CG-SENSE revisited: Results from the first ISMRM reproducibility challengecited by 7
  12. 2019SEGAN: Structure-Enhanced Generative Adversarial Network for Compressed Sensing MRI Reconstructioncited by 6
  13. 2021Density Compensated Unrolled Networks for Non-Cartesian MRI Reconstructioncited by 4
  14. 2023Physics-informed self-supervised deep learning reconstruction for accelerated first-pass perfusion cardiac MRIcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 0; 2025: 4; 2026: 2.

13 papers.

Sort:

By area:

  1. 2023Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learningcited by 34
  2. 2023Spatial and Modal Optimal Transport for Fast Cross-Modal MRI Reconstructioncited by 5
  3. 2023Score-based Diffusion Models With Self-supervised Learning For Accelerated 3D Multi-contrast Cardiac Magnetic Resonance Imagingcited by 4
  4. 2023Joint Supervised and Self-supervised Learning for MRI Reconstructioncited by 4
  5. 2025Benchmarking Self-Supervised Learning Methods for Accelerated MRI Reconstructioncited by 4
  6. 2025Re-Visible Dual-Domain Self-Supervised Deep Unfolding Network for MRI Reconstructioncited by 1
  7. 2023Single Image Compressed Sensing MRI via a Self-Supervised Deep Denoising Approachcited by 0
  8. 2023Deep-learning-based acceleration of MRI for radiotherapy planning of pediatric patients with brain tumorscited by 0
  9. 2023Dual-Domain Multi-Contrast MRI Reconstruction with Synthesis-based Fusion Networkcited by 0
  10. 2025Dual-domain Multi-path Self-supervised Diffusion Model for Accelerated MRI Reconstructioncited by 0
  11. 2025Self-Supervised Joint Reconstruction and Denoising of T2-Weighted PROPELLER MRI of the Lungs at 0.55Tcited by 0
  12. 2026NexOP: Joint Optimization of NEX-Aware k-space Sampling and Image Reconstruction for Low-Field MRIcited by 0
1 more
  1. 2026MoE-dqINR: A Unified Mixture-of-Experts Implicit Neural Representation Framework for Scan-Specific Dynamic and Quantitative MRI Reconstructioncited by 0