Pith. sign in

REVIEW 4 major objections 6 minor 28 references

An All-in-one Approach for Accelerated Cardiac MRI Reconstruction

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proposes a stepwise Patch-GAN reconstruction that reconstructs highly undersampled multi-coil cardiac MRI k-space across contrasts, views, and trajectories, reporting average SSIM of 0.9907 on Task 1 and 0.9799 on Task 2 of the…

desk verdict Solid challenge-based cardiac MRI paper with a real all-in-one inconsistency in Task 1; Task 2's single-model universal result is the genuine contribution. read the letter →

arxiv 2411.10787 v1 pith:GYNXLNVI submitted 2024-11-16 eess.IV

Kian Anvari Hamedani , Narges Razizadeh , Shahabedin Nabavi , Mohsen Ebrahimi Moghaddam This is my paper · ORCID
classification eess.IV
keywords cardiacMRIreconstructionacceleratedPatch-GANk-spaceundersamplingcurriculumlearningmulti-contrastCMRCMRxRecon2024deep
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper aims to show that a single deep-learning model can reconstruct high-quality cardiac MRI from heavily undersampled k-space data across the multi-contrast, multi-view, multi-trajectory settings of the CMRxRecon2024 challenge. Its proposed stepwise Patch-GAN approach iteratively refines the k-space using an attention-based prompt UNet, sensitivity-map estimation, and a combined physical-plus-SSIM loss, trained with curriculum learning that starts at low acceleration factors and moves to higher ones. On the challenge tasks the authors report average SSIM of 0.9907 for Task 1 (uniform masks at 4x, 8x, 10x acceleration) and 0.9799 for Task 2 (mixed trajectories at up to 24x), above the compared PromptIR, E2E-Varnet, and PromptMR methods. If the claim holds, the practical payoff is markedly shorter cardiac scan times with image quality close to fully sampled reconstructions.

What carries the argument

The load-bearing mechanism is the stepwise k-space update $k^{(t+1)} = k^{(t)} - \eta^{(t)} M (k^{(t)} - k^{(0)}) + G_k$, where $M$ is the subsampling mask, $\eta^{(t)}$ is a learnable step size, and $G_k$ is a gradient-like term produced by an attention-based prompt UNet (APUNet) that receives adjacent k-spaces, a discriminative prompt, and a conjugate-symmetric sensitivity map. At each of the twelve steps, a physical loss (MSE on magnitude and phase of the central k-space) and an SSIM loss in image space are summed into a stepwise loss that is added to the generator's adversarial loss, mitigating vanishing gradients. The Patch-GAN discriminator receives zero-filled or reconstructed images concatenated with the undersampled input, which helps it learn subsampling artefacts. Curriculum learning orders training from 4x acceleration to higher factors.

What would settle it

Re-run all four methods on the same held-out CMRxRecon2024 test subjects with identical coil compression, undersampling masks, and preprocessing, then compare per-contrast NMSE, PSNR, and SSIM. If the proposed approach no longer exceeds PromptMR on average SSIM in both tasks, the central outperformance claim is refuted.

Watch

Extended reading notes

Core claim

The central discovery is that a single integrated Patch-GAN generator, without a separate video-restoration refinement stage, can serve as an all-in-one CMR reconstruction method. The model consumes adjacent undersampled multi-coil k-spaces, estimates sensitivity maps from auto-calibration lines, and refines the central k-space through twelve stepwise reconstructor modules. Each step compares the intermediate result against the ground-truth k-space through a physics-based loss (magnitude and phase MSE) and against the ground-truth image through SSIM, while a Patch-GAN discriminator encourages local realism. Curriculum learning is used to progress from easier low-acceleration reconstructions to harder high-acceleration ones. The authors report that this combination outperforms the compared methods on both tasks of CMRxRecon2024, with per-contrast SSIM values up to 0.9931.

Load-bearing premise

The load-bearing premise is that the comparison with PromptIR, E2E-Varnet, and PromptMR was run under identical data splits, preprocessing, masks, and evaluation protocol; the paper does not state whether the baseline numbers come from the challenge's official results or from the authors' own reimplementation, so the reported advantage could shrink under a strictly equal re-run.

Editorial extensions

If this is right

  • A single trained model can cover multiple contrasts, anatomical views, and k-space trajectories, so clinical deployment would not require retraining per imaging protocol.
  • The stepwise loss design lets a twelve-module generator train without vanishing gradients, pointing to a straightforward way to deepen reconstruction networks.
  • Curriculum learning from low to high acceleration factors contributes to the final quality, suggesting that easy-to-hard training is an effective ordering for undersampled reconstruction.
  • At acceleration factors up to 24x, the reported SSIM and PSNR values stay close to fully sampled quality, which would make substantially faster cardiac scans feasible.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not state whether the baseline numbers for PromptIR, E2E-Varnet, and PromptMR come from the challenge's official results or from the authors' own reimplementation; a standardized re-run on identical splits and preprocessing would clarify how much of the margin is architectural.
  • If the reported margins reproduce, the most direct clinical consequence is shorter breath-hold times and fewer motion artefacts in cardiac MRI, because 24x undersampling could become practical.
  • The ablation table decomposes gains per component, but it does not separately test the discriminator or curriculum on the unseen flow2D and BlackBlood contrasts; a targeted extension would quantify generalization beyond the training distribution.
  • The same stepwise Patch-GAN design could be tested on non-cardiac MR or on other non-Cartesian trajectories outside the challenge to see whether the all-in-one behaviour transfers.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper presents a stepwise reconstruction approach for accelerated multi-contrast cardiac MRI, built on a Patch-GAN architecture with an attention-based prompt UNet (APUNet) as the generator, a discriminator, curriculum learning, and a combined loss consisting of k-space physical (MSE of magnitude and phase) and image-domain SSIM terms computed at each reconstruction step. The method is evaluated on the CMRxRecon2024 challenge dataset for Task 1 (uniform undersampling at accelerations 4x, 8x, 10x) and Task 2 (multiple trajectories and accelerations up to 24x). The authors report average SSIM of 0.9907 (Task 1) and 0.9799 (Task 2), outperforming PromptIR, E2E-Varnet, and PromptMR, with ablation studies on cine contrasts supporting the proposed components. Code is made available on GitHub.

Significance. Reconstructing highly undersampled CMR k-space in a single model across contrasts, views, trajectories, and acceleration factors is a timely and valuable goal. The paper combines several plausible ingredients—stepwise unrolled updates, physical and structural losses per step, curriculum learning, and adversarial training—and reports consistent improvements over strong baselines across the contrast/view columns in Table 1. The public release of the code is a concrete strength that should facilitate verification. The significance is currently tempered by the evaluation protocol (three separate models for Task 1 vs. the 'all-in-one' claim), undisclosed baseline provenance, and the absence of statistical uncertainty; if these are addressed, the contribution would be a solid advance for the CMR reconstruction community.

major comments (4)
  1. [Section 3, Table 1] Task 1 results are averages of three separately trained models, one for each acceleration factor (4, 8, and 10x), as stated in the text: 'three different models for three acceleration factors of 4, 8, and 10x have been trained.' The abstract and contributions, however, claim an 'all-in-one' approach. This inconsistency is load-bearing: the superior Task 1 SSIM of 0.9907 is not achieved by a single universal model but by an oracle selection among three specialized models. Unless the baselines were likewise allowed one model per acceleration factor (which is not stated), the comparison is inequitable. The authors should either train and report a single model for Task 1 or explicitly state that the baselines were also evaluated with per-acceleration models.
  2. [Section 3, Table 1] The source of the baseline numbers (PromptIR, E2E-Varnet, PromptMR) is not disclosed. It is not stated whether these numbers are taken from the CMRxRecon2024 official leaderboard or recomputed by the authors under the same data splits, masks, coil compression, and preprocessing pipeline. Without this information, the reported margins cannot be verified as method-level comparisons rather than artifacts of protocol differences. Please specify the provenance of every baseline entry and, if feasible, include a detailed reproducibility statement.
  3. [Section 2.2, Eq. (2) and Algorithm 1] The paper identifies Gk in Eq. (2) as 'the gradient of the sparsity regularization term R(x)' (text after Eq. (2)). In Algorithm 1 (lines 12–13), Gk is computed as FFT(ISS), where ISS is derived from the APUNet output; no constraint or derivation links this network output to ∇R(x). This unsupported identification underlies the unrolled-optimization interpretation of Eq. (2). Please either provide a formal argument or empirical validation that the learned module approximates the regularizer gradient, or revise the text to describe Gk as a learned k-space correction without the gradient claim.
  4. [Section 3, Tables 1 and 2] No error bars, confidence intervals, or significance tests are provided. Many ablation differences are within a few thousandths of SSIM (e.g., Table 2, cine SAX: proposed 0.9903 vs. without physical loss 0.9892; cine LAX: proposed 0.9893 vs. without curriculum learning 0.9845). Without subject-level variance, these differences could be within noise. Report mean ± std over the evaluation subjects and, ideally, paired statistical tests for the main comparison and the ablation study.
minor comments (6)
  1. [Table 1] The first number in each triplet (e.g., '2.6') is not labeled; state explicitly that it denotes NMSE in percent (or as a ratio) to avoid ambiguity.
  2. [Abstract and Table 1] The abstract reports SSIM as 99.07 and 97.99, while Table 1 gives 0.9907 and 0.9799; use a consistent scale across the manuscript.
  3. [Section 2.2 and Algorithm 1] The word 'centeral' appears multiple times (e.g., in Section 2.2 and Algorithm 1); it should be 'central'.
  4. [Section 2.1] The description of CMRxRecon2024 says Flow2D and BlackBlood images are used exclusively for validation as unseen data, but the paper does not report results for these contrasts in Table 1; clarify which contrasts are included in the official evaluation and why these two are omitted.
  5. [Introduction] The citation of the authors' own prior works [13,14] for breath-holding artifacts seems unrelated; consider citing a dedicated CMR motion-artifact reference instead.
  6. [Figure 4] Figure 4 is referenced in the text but not discussed; please add a brief interpretation of the qualitative differences shown in the reconstruction example.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; reported results are external benchmark comparisons with incidental self-citations only.

full rationale

The paper's central claim is an empirical performance comparison against external baselines (PromptIR, E2E-Varnet, PromptMR) on the CMRxRecon2024 challenge dataset. The reconstruction method is a Patch-GAN stepwise architecture, and its reported SSIM/PSNR/NMSE values are evaluated on challenge data, not derived from a fitted parameter or from any prior result by the authors. The self-citations [13,14] appear in the introduction in support of the generic statement that patients are asked to hold their breath to minimize motion artefacts; this is background knowledge and is not load-bearing for the reconstruction claim. No uniqueness theorem is imported, no ansatz is smuggled via self-citation, and there is no equation in which an output is defined in terms of an input in a way that makes the reported performance true by construction. The skeptical concern about Task 1 reporting the average of three acceleration-specific models while claiming an all-in-one model is a matter of experimental protocol consistency, not circularity: it questions the fairness of the comparison and the accuracy of the 'all-in-one' label, but it does not make the result reducible to its inputs by definition. The derivation chain is therefore self-contained with respect to circularity, and the paper's evaluation is anchored to an external benchmark.

Assumptions & free parameters 9 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard MRI reconstruction assumptions, a borrowed unrolling scheme, and the ad hoc identification of a network output with a regularizer gradient. The many hand-chosen hyperparameters, including an unreported lambda, are free parameters not derived from theory.

free parameters (9)
  • Number of reconstructor modules = 12
    Chosen by hand; not motivated by analysis. Affects capacity and reconstruction quality.
  • Learning rate = 0.002
    AdamW optimizer hyperparameter, standard tuning.
  • Weight decay = 0.1
    AdamW hyperparameter.
  • Gradient clipping value = 0.1
    Chosen by hand.
  • Number of auto-calibration lines = 16
    Chosen by hand; affects sensitivity map estimation.
  • Adjacent k-space length = 5
    Number of neighboring slices/phases used as input; chosen by hand.
  • Stepwise loss adjustment coefficient lambda = Not reported
    Equation (3) defines lambda but the paper never reports its value, making exact replication difficult.
  • Number of training epochs = 12
    Chosen by hand; curriculum learning typically needs more epochs, but the paper does not discuss this.
  • Learning rate scheduler step size and gamma = 11 and 0.1
    Hyperparameters chosen by hand.
assumptions (5)
  • domain assumption The CS optimization problem in Eq (1) with sparsity regularization is the correct formulation for accelerated MRI reconstruction.
    Standard compressed sensing model for MRI; accepted in the field.
  • domain assumption The iterative update in Eq (2) from E2E-Varnet [21] converges to a solution of Eq (1) when Gk is the gradient of the regularizer.
    Borrowed from prior work; the paper does not prove convergence for its architecture.
  • ad hoc to paper The APUNet output Gk approximates the gradient of the sparsity regularizer R(x).
    Algorithm 1 computes Gk as a network output; no explicit R(x) is defined, so the claim that Gk is a gradient is assumed.
  • domain assumption The sensitivity map estimator, an APUNet, produces accurate coil sensitivity maps from auto-calibration lines.
    Learning-based sensitivity estimation is common; accuracy is not analyzed.
  • domain assumption The CMRxRecon2024 benchmark and its evaluation protocol are valid measures of reconstruction quality.
    The dataset and metrics are community standard.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An All-in-one Approach for Accelerated Cardiac MRI Reconstruction." pith.science (2026). https://pith.science/paper/GYNXLNVI

@misc{pith2026241110787,
  author       = {Pith},
  title        = {Pith review of: An All-in-one Approach for Accelerated Cardiac MRI Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GYNXLNVI}},
  note         = {Machine review of arXiv:2411.10787}
}
read the original abstract

Cardiovascular magnetic resonance (CMR) imaging is the gold standard for diagnosing several heart diseases due to its non-invasive nature and proper contrast. MR imaging is time-consuming because of signal acquisition and image formation issues. Prolonging the imaging process can result in the appearance of artefacts in the final image, which can affect the diagnosis. It is possible to speed up CMR imaging using image reconstruction based on deep learning. For this purpose, the high-quality clinical interpretable images can be reconstructed by acquiring highly undersampled k-space data, that is only partially filled, and using a deep learning model. In this study, we proposed a stepwise reconstruction approach based on the Patch-GAN structure for highly undersampled k-space data compatible with the multi-contrast nature, various anatomical views and trajectories of CMR imaging. The proposed approach was validated using the CMRxRecon2024 challenge dataset and outperformed previous studies. The structural similarity index measure (SSIM) values for the first and second tasks of the challenge are 99.07 and 97.99, respectively. This approach can accelerate CMR imaging to obtain high-quality images, more accurate diagnosis and a pleasant patient experience.

Figures

Figures reproduced from arXiv: 2411.10787 by the authors.

Figure 1
Figure 1. The overview of the proposed approach. t. In [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. The structures of the APUNet and the discriminator networks. the generator to achieve more efficient performance in the face of unseen data. Given Inputreal and Inputfake, the discriminator network predicts P redreal and P redfake, then uses binary cross-entropy (BCE) [19] loss to approximate the discriminator loss LDisc, which is used to learn the discriminator network. This process is repeated t times until the fi… view at source ↗
Figure 3
Figure 3. An example of several k-space subsampling masks and corresponding images. with a learning rate of 0.002, weight decay of 0.1, and a gradient clipping value of 0.1. The learning rate was scheduled to be adjusted by a scheduler with a step size of 11 and a gamma of 0.1. The generator network consists of 12 reconstructor modules, and we used 16 auto-calibration lines for the undersampled k-spaces with an adjacent k-spa… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: An example of qualitative comparison of images reconstructed from subsampled k-space using a uniform sampling pattern with a 10x acceleration factor, reconstructed by the proposed approach and ground truth images. cine CMR images of the first task of the CMRxRecon2024 …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 18 canonical work pages

  1. [1]

    Heart106(3), 174–181 (2020)

    Arnold, J.R., McCann, G.P.: Cardiovascular magnetic resonance: applications and practical considerations for the general cardiologist. Heart106(3), 174–181 (2020)

    work page 2020

  2. [2]

    In: Pro- ceedings of the 26th annual international conference on machine learning

    Bengio, Y., Louradour, J., Collobert, R., Weston, J.: Curriculum learning. In: Pro- ceedings of the 26th annual international conference on machine learning. pp. 41–48 (2009)

    work page 2009

  3. [3]

    iterative image reconstruction using a total variation constraint

    Block, K.T., Uecker, M., Frahm, J.: Undersampled radial mri with multiple coils. iterative image reconstruction using a total variation constraint. Magnetic Reso- nance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine57(6), 1086–1098 (2007)

    work page 2007

  4. [4]

    Bydder,M.,Robson,M.D.:Partialfourierpartiallyparallelimaging.MagneticRes- onance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine53(6), 1393–1401 (2005)

    work page 2005

  5. [5]

    IEEE Transactions on information theory 52(4), 1289–1306 (2006)

    Donoho, D.L.: Compressed sensing. IEEE Transactions on information theory 52(4), 1289–1306 (2006)

    work page 2006

  6. [6]

    claustro

    Enders, J., Zimmermann, E., Rief, M., Martus, P., Klingebiel, R., Asbach, P., Klessen, C., Diederichs, G., Bengner, T., Teichgräber, U., et al.: Reduction of claustrophobia during magnetic resonance imaging: methods and design of the" claustro" randomized controlled trial. BMC medical imaging11, 1–15 (2011) All-in-one CMRI Reconstruction 11

    work page 2011

  7. [7]

    Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine47(6), 1202–1210 (2002)

    Griswold, M.A., Jakob, P.M., Heidemann, R.M., Nittka, M., Jellus, V., Wang, J., Kiefer, B., Haase, A.: Generalized autocalibrating partially parallel acquisitions (grappa). Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine47(6), 1202–1210 (2002)

    work page 2002

  8. [8]

    Magnetic resonance in medicine79(6), 3055–3071 (2018)

    Hammernik, K., Klatzer, T., Kobler, E., Recht, M.P., Sodickson, D.K., Pock, T., Knoll, F.: Learning a variational network for reconstruction of accelerated mri data. Magnetic resonance in medicine79(6), 3055–3071 (2018)

    work page 2018

Show all 28 references
  1. [9]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Isola, P., Zhu, J.Y., Zhou, T., Efros, A.A.: Image-to-image translation with condi- tional adversarial networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1125–1134 (2017)

    2017

  2. [10]

    arXiv preprint arXiv:1711.05101 (2017)

    Loshchilov, I., Hutter, F.: Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101 (2017)

    2017 arXiv

  3. [11]

    Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine58(6), 1182–1195 (2007)

    Lustig, M., Donoho, D., Pauly, J.M.: Sparse mri: The application of compressed sensing for rapid mr imaging. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine58(6), 1182–1195 (2007)

    2007

  4. [12]

    arXiv preprint arXiv:2404.01082 (2024)

    Lyu, J., Qin, C., Wang, S., Wang, F., Li, Y., Wang, Z., Guo, K., Ouyang, C., Tänzer, M., Liu, M., et al.: The state-of-the-art in cardiac mri reconstruction: Results of the cmrxrecon challenge in miccai 2023. arXiv preprint arXiv:2404.01082 (2024)

    2024 arXiv

  5. [13]

    arXiv preprint arXiv:2409.00375 (2024)

    Nabavi, S., Hamedani, K.A., Moghaddam, M.E., Abin, A.A., Frangi, A.F.: Sta- tistical distance-guided unsupervised domain adaptation for automated multi- class cardiovascular magnetic resonance image quality assessment. arXiv preprint arXiv:2409.00375 (2024)

    2024 arXiv

  6. [14]

    Computer Methods and Programs in Biomedicine242, 107770 (2023)

    Nabavi, S., Simchi, H., Moghaddam, M.E., Abin, A.A., Frangi, A.F.: A generalised deep meta-learning model for automated quality control of cardiovascular magnetic resonance images. Computer Methods and Programs in Biomedicine242, 107770 (2023)

    2023

  7. [15]

    Advances in Neural Information Processing Systems 36 (2024)

    Potlapalli, V., Zamir, S.W., Khan, S.H., Shahbaz Khan, F.: Promptir: Prompt- ing for all-in-one image restoration. Advances in Neural Information Processing Systems 36 (2024)

    2024

  8. [16]

    Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine42(5), 952–962 (1999)

    Pruessmann, K.P., Weiger, M., Scheidegger, M.B., Boesiger, P.: Sense: sensitivity encoding for fast mri. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine42(5), 952–962 (1999)

    1999

  9. [17]

    IEEE transactions on medical imaging38(1), 280–290 (2018)

    Qin, C., Schlemper, J., Caballero, J., Price, A.N., Hajnal, J.V., Rueckert, D.: Con- volutional recurrent neural networks for dynamic mr image reconstruction. IEEE transactions on medical imaging38(1), 280–290 (2018)

    2018

  10. [18]

    Journal of Cardiovascular Magnetic Resonance 24(1), 13 (2022)

    Raman, S.V., Markl, M., Patel, A.R., Bryant, J., Allen, B.D., Plein, S., Seiberlich, N.: 30-minute cmr for common clinical indications: Â aâ society for cardiovascular magnetic resonance white paper. Journal of Cardiovascular Magnetic Resonance 24(1), 13 (2022)

    2022

  11. [19]

    The Journal of Machine Learning Research 11, 2387–2422 (2010)

    Reid, M.D., Williamson, R.C.: Composite binary losses. The Journal of Machine Learning Research 11, 2387–2422 (2010)

    2010

  12. [20]

    IEEE transactions on Medical Imaging37(2), 491–503 (2017)

    Schlemper, J., Caballero, J., Hajnal, J.V., Price, A.N., Rueckert, D.: A deep cas- cade of convolutional neural networks for dynamic mr image reconstruction. IEEE transactions on Medical Imaging37(2), 491–503 (2017)

    2017

  13. [21]

    In: Medical Image Computing and Computer Assisted Intervention– MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part II 23

    Sriram, A., Zbontar, J., Murrell, T., Defazio, A., Zitnick, C.L., Yakubova, N., Knoll, F., Johnson, P.: End-to-end variational networks for accelerated mri re- construction. In: Medical Image Computing and Computer Assisted Intervention– MICCAI 2020: 23rd International Confere...

    2020

  14. [22]

    Heart 105(22), 1755–1762 (2019)

    Vasquez, M., Nagel, E.: Clinical indications for cardiovascular magnetic resonance. Heart 105(22), 1755–1762 (2019)

    2019

  15. [23]

    Scientific Data11(1), 687 (2024)

    Wang, C., Lyu, J., Wang, S., Qin, C., Guo, K., Zhang, X., Yu, X., Li, Y., Wang, F., Jin, J., et al.: Cmrxrecon: A publicly available k-space dataset and benchmark to advance deep learning for cardiac mri. Scientific Data11(1), 687 (2024)

    2024

  16. [24]

    IEEE transactions on image processing 13(4), 600–612 (2004)

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing 13(4), 600–612 (2004)

    2004

  17. [25]

    arXiv preprint arXiv:2406.19043 (2024)

    Wang, Z., Wang, F., Qin, C., Lyu, J., Cheng, O., Wang, S., Li, Y., Yu, M., Zhang, H., Guo, K., et al.: Cmrxrecon2024: A multi-modality, multi-view k-space dataset boosting universal machine learning for accelerated cardiac mri. arXiv preprint arXiv:2406.19043 (2024)

    2024 arXiv

  18. [26]

    In: International Workshop on Statistical Atlases and Computational Models of the Heart

    Xin, B., Ye, M., Axel, L., Metaxas, D.N.: Fill the k-space and refine the image: Prompting for dynamic and multi-contrast mri reconstruction. In: International Workshop on Statistical Atlases and Computational Models of the Heart. pp. 261–

  19. [27]

    Yang, G., Yu, S., Dong, H., Slabaugh, G., Dragotti, P.L., Ye, X., Liu, F., Arridge, S., Keegan, J., Guo, Y., et al.: Dagan: deep de-aliasing generative adversarial net- worksforfastcompressedsensingmrireconstruction.IEEEtransactionsonmedical imaging 37(6), 1310–1321 (2017)

    2017

  20. [28]

    In: International Workshop on Statistical Atlases and Computa- tional Models of the Heart

    Yiasemis, G., Moriakov, N., Sonke, J.J., Teuwen, J.: Deep cardiac mri reconstruc- tion with admm. In: International Workshop on Statistical Atlases and Computa- tional Models of the Heart. pp. 479–490. Springer (2023)

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.