mlsanityData to 14 Oct 2026

Robust Physics-based Deep MRI Reconstruction Via Diffusion Purification

arXiv 2309.05794 ↗First posted 11 Sep 2023eess.IVcited by 4

On the map: Diffusion models → Medical imaging and segmentation

Abstract (arXiv)

Deep learning (DL) techniques have been extensively employed in magnetic resonance imaging (MRI) reconstruction, delivering notable performance enhancements over traditional non-DL methods. Nonetheless, recent studies have identified vulnerabilities in these models during testing, namely, their susceptibility to (textiti) worst-case measurement perturbations and to (textitii) variations in training/testing settings like acceleration factors and k-space sampling locations. This paper addresses the robustness challenges by leveraging diffusion models. In particular, we present a robustification strategy that improves the resilience of DL-based MRI reconstruction methods by utilizing pretrained diffusion models as noise purifiers. In contrast to conventional robustification methods for DL-based MRI reconstruction, such as adversarial training (AT), our proposed approach eliminates the need to tackle a minimax optimization problem. It only necessitates fine-tuning on purified examples. Our experimental results highlight the efficacy of our approach in mitigating the aforementioned instabilities when compared to leading robustification approaches for deep MRI reconstruction, including AT and randomized smoothing.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 42 of its 69 references that are in the map. First posted on arXiv 11 Sep 2023.

Builds on

42 of 69 references.

Closest ancestors · 6

Same direction, newest first

  1. 2023Annealed Score-Based Diffusion Model for MR Motion Artifact Reduction8 Jan
  2. 2022Diffusion Models for Medical Image Analysis: A Comprehensive Survey14 Nov
  3. 2022Adaptive Diffusion Priors for Accelerated MRI Reconstruction12 Jul
  4. 2022Towards performant and reliable undersampled MR reconstruction via diffusion model sampling8 Mar
  5. 2022Measurement-conditioned Denoising Diffusion Probabilistic Model for Under-sampled Medical Image Reconstruction5 Mar
  6. 2021Score-based diffusion models for accelerated MRI8 Oct

Other references · 36

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2020Denoising Diffusion Implicit Modelscited by 9,781
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 3,045
  8. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  9. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,451
  10. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
  11. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  12. 2020Fast is better than free: Revisiting adversarial trainingcited by 727
24 more
  1. 2022Diffusion Models for Adversarial Purificationcited by 476
  2. 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 470
  3. 2017MoDL: Model Based Deep Learning Architecture for Inverse Problemscited by 387
  4. 2020Rethinking Pre-training and Self-trainingcited by 352
  5. 2021Domain Adaptation for Medical Image Analysis: A Surveycited by 352
  6. 2020Deep Learning Techniques for Inverse Problems in Imagingcited by 291
  7. 2017A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstructioncited by 232
  8. 2019On instabilities of deep learning in image reconstruction - Does AI come at a cost?cited by 212
  9. 2023Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistencycited by 182
  10. 2023Decomposed Diffusion Sampler for Accelerating Large-Scale Inverse Problemscited by 132
  11. 2021Deep Equilibrium Architectures for Inverse Problems in Imagingcited by 127
  12. 2017Framing U-Net via Deep Convolutional Framelets: Application to Sparse-view CTcited by 88
  13. 2021Deep Dense Multi-scale Network for Snow Removal Using Semantic and Geometric Priorscited by 82
  14. 2023GridFormer: Residual Dense Transformer with Grid Structure for Image Restoration in Adverse Weather Conditionscited by 62
  15. 2019Image Reconstruction: From Sparsity to Data-adaptive Methods and Machine Learningcited by 60
  16. 2018Deep Residual Learning for Accelerated MRI using Magnitude and Phase Networkscited by 53
  17. 2022Physics-Driven Deep Learning for Computational Magnetic Resonance Imagingcited by 50
  18. 2021Recurrent Variational Network: A Deep Learning Inverse Problem Solver applied to the task of Accelerated MRI Reconstructioncited by 34
  19. 2017ADMM-Net: A Deep Learning Approach for Compressive Sensing MRIcited by 28
  20. 2019LORAKI: Autocalibrated Recurrent Neural Networks for Autoregressive MRI Reconstruction in k-Spacecited by 28
  21. 2020Domain Shift in Computer Vision models for MRI data analysis: An Overviewcited by 16
  22. 2022On the Robustness of deep learning-based MRI Reconstruction to image transformationscited by 7
  23. 2021On Instabilities of Conventional Multi-Coil MRI Reconstruction to Small Adverserial Perturbationscited by 4
  24. 2023SMUG: Towards robust MRI reconstruction by smoothed unrollingcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 1; 2026: 1.

4 papers.

  1. 2024Exploring Low-Dimensional Subspaces in Diffusion Models for Controllable Image Editingcited by 15
  2. 2026Taming Score-Based Denoisers in ADMM: A Convergent Plug-and-Play Frameworkcited by 2
  3. 2024Decoupled Data Consistency with Diffusion Purification for Image Restorationcited by 1
  4. 2025Training-Free Adversarial Robustness in Computational MRIcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Resolution-Robust 3D MRI Reconstruction with 2D Diffusion Priors: Diverse-Resolution Training Outperforms Interpolation
  2. 2025Conditional Denoising Diffusion Model-Based Robust MR Image Reconstruction from Highly Undersampled Data
  3. 2024On Instabilities of Unsupervised Denoising Diffusion Models in Magnetic Resonance Imaging Reconstruction
  4. 2023Learning to Reconstruct Accelerated MRI Through K-space Cold Diffusion without Noise
  5. 2024Robust multi-coil MRI reconstruction via self-supervised denoising
  6. 2022Physics-informed Deep Diffusion MRI Reconstruction with Synthetic Data: Break Training Data Bottleneck in Artificial Intelligence
  7. 2024Sampling-Pattern-Agnostic MRI Reconstruction through Adaptive Consistency Enforcement with Diffusion Model
  8. 2023Diffusion Modeling with Domain-conditioned Prior Guidance for Accelerated MRI and qMRI Reconstruction