mlsanityData to 7 Oct 2026

Self-Supervised Adversarial Diffusion Models for Fast MRI Reconstruction

arXiv 2406.15656 ↗First posted 21 Jun 2024eess.IV cs.CVcited by 9

On the map: Diffusion models → Medical imaging and segmentation

Abstract (arXiv)

Purpose: To propose a self-supervised deep learning-based compressed sensing MRI (DL-based CS-MRI) method named "Adaptive Self-Supervised Consistency Guided Diffusion Model (ASSCGD)" to accelerate data acquisition without requiring fully sampled datasets. Materials and Methods: We used the fastMRI multi-coil brain axial T2-weighted (T2-w) dataset from 1,376 cases and single-coil brain quantitative magnetization prepared 2 rapid acquisition gradient echoes (MP2RAGE) T1 maps from 318 cases to train and test our model. Robustness against domain shift was evaluated using two out-of-distribution (OOD) datasets: multi-coil brain axial postcontrast T1 -weighted (T1c) dataset from 50 cases and axial T1-weighted (T1-w) dataset from 50 patients. Data were retrospectively subsampled at acceleration rates R in {2x, 4x, 8x}. ASSCGD partitions a random sampling pattern into two disjoint sets, ensuring data consistency during training. We compared our method with ReconFormer Transformer and SS-MRI, assessing performance using normalized mean squared error (NMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Statistical tests included one-way analysis of variance (ANOVA) and multi-comparison Tukey's Honesty Significant Difference (HSD) tests. Results: ASSCGD preserved fine structures and brain abnormalities visually better than comparative methods at R = 8x for both multi-coil and single-coil datasets. It achieved the lowest NMSE at R in {4x, 8x}, and the highest PSNR and SSIM values at all acceleration rates for the multi-coil dataset. Similar trends were observed for the single-coil dataset, though SSIM values were comparable to ReconFormer at R in {2x, 8x}. These results were further confirmed by the voxel-wise correlation scatter plots. OOD results showed significant (p << 10^-5 ) improvements in undersampled image quality after reconstruction.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 19 of its 46 references that are in the map. First posted on arXiv 21 Jun 2024.

Builds on

19 of 46 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023CDiffMR: Can We Replace the Gaussian Noise with K-Space Undersampling for Fast MRI?25 Jun
  2. 2022High-Frequency Space Diffusion Models for Accelerated MRI10 Aug
  3. 2022Adaptive Diffusion Priors for Accelerated MRI Reconstruction12 Jul
  4. 2022Bayesian MRI Reconstruction with Joint Uncertainty Estimation using Diffusion Models3 Feb

Other references · 15

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 464
  5. 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 340
  6. 2022Understanding Diffusion Models: A Unified Perspectivecited by 337
  7. 2017A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstructioncited by 231
  8. 2022ReconFormer: Accelerated MRI Reconstruction Using Recurrent Transformercited by 52
  9. 2024Tutorial on Diffusion Models for Imaging and Visioncited by 49
  10. 2017MR image reconstruction using deep density priorscited by 37
  11. 2022PyTorch Image Quality: Metrics for Image Quality Assessmentcited by 34
  12. 2019MRI Reconstruction Using Deep Bayesian Estimationcited by 28
3 more
  1. 2021Self-Supervised Learning for MRI Reconstruction with a Parallel Network Training Frameworkcited by 26
  2. 2019Self-supervised learning of inverse problem solvers in medical imagingcited by 8
  3. 2022Accelerated respiratory-resolved 4D-MRI with separable spatio-temporal neural networkscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 8; 2026: 1.

9 papers.

  1. 2025Triad: Vision Foundation Model for 3D Magnetic Resonance Imagingcited by 23
  2. 2025MRI super-resolution reconstruction using efficient diffusion probabilistic model with residual shiftingcited by 10
  3. 2025A Physics-Informed Deep Learning Model for MRI Brain Motion Correctioncited by 4
  4. 2025Systematic Review and Meta-analysis of AI-driven MRI Motion Artifact Detection and Correctioncited by 3
  5. 2025Res-MoCoDiff: Residual-guided diffusion models for motion artifact correction in brain MRIcited by 1
  6. 2025Efficient Vision Mamba for MRI Super-Resolution via Hybrid Selective Scanningcited by 1
  7. 2025Dual-domain Multi-path Self-supervised Diffusion Model for Accelerated MRI Reconstructioncited by 0
  8. 2025An Efficient 3D Latent Diffusion Model for T1-contrast Enhanced MRI Generationcited by 0
  9. 2026MRI super-resolution in ten sampling steps using a diffusion bridge modelcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Self-Supervised MRI Reconstruction with Unrolled Diffusion Models
  2. 2024Robust multi-coil MRI reconstruction via self-supervised denoising
  3. 2023Robust Physics-based Deep MRI Reconstruction Via Diffusion Purification
  4. 2024Sampling-Pattern-Agnostic MRI Reconstruction through Adaptive Consistency Enforcement with Diffusion Model
  5. 2025A Self-supervised Diffusion Bridge for MRI Reconstruction
  6. 2026Phase-map synthesis from magnitude-only MR images using conditional score-based diffusion models with application in training of accelerated MRI reconstruction models
  7. 2025Measurement Score-Based MRI Reconstruction with Automatic Coil Sensitivity Estimation
  8. 2024Fast MRI for All: Bridging Access Gaps by Training without Raw Data