Pith. sign in

REVIEW 3 major objections 8 minor 1 cited by

Continuous K-space Recovery Network with Image Guidance for Fast MRI Reconstruction

T0 review · 3 major / 8 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 IGKR-Net reconstructs undersampled MRI k-space continuously, using image-guided implicit neural representation and progressive training, and reports the best metrics on four public datasets.

desk verdict A sensible INR-based k-space recovery network with consistent benchmark gains, but the 'significant' claim outruns the statistics and one table is mislabeled. read the letter →

arxiv 2411.11282 v2 pith:H5HDDN2S submitted 2024-11-18 eess.IV cs.AIcs.CV

Yucong Meng , Zhiwei Yang , Minghong Duan , Yonghong Shi , Zhijian Song This is my paper · ORCID
classification eess.IVcs.AIcs.CV MSC 68T0768U1092C55
keywords MRIreconstructionk-spacerecoveryimplicitneuralrepresentationimage-domainguidancetransformermulti-stagetrainingundersampledfast
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

This paper claims that the right way to reconstruct undersampled MRI is to recover the missing k-space values directly, by modeling k-space as a continuous implicit neural representation. The proposed network, IGKR-Net, encodes sampled k-values together with their coordinates, queries unsampled coordinates with a transformer decoder, and uses image-domain information plus a progressive low-to-high resolution training schedule to guide and stabilize the recovery. If the claim holds, accelerated MRI could produce denser, higher-fidelity images from the same acquisition time, or matching image quality from less data. The paper reports consistent metric improvements over prior methods on single-coil and multi-coil datasets at sampling ratios from 10% to 40%.

What carries the argument

The central object is an implicit neural representation (INR) of k-space: a neural function that maps any k-space coordinate to a k-value, learned by a transformer encoder that turns sampled k-values plus their coordinates into a continuous latent space, and a transformer decoder that queries unsampled coordinates from that space. The architecture carries the argument by making k-space recovery a continuous regression problem rather than a discrete inpainting problem. Two implicit transformer stages, LRIT and HRIT, reconstruct low-resolution then high-resolution k-space; the image-domain guidance module (IDGM) injects features from the low-quality image into the k-space recovery; the tri-attention refinement module (TARM) sharpens the final image; and a four-stage loss schedule trains the modules progressively.

What would settle it

Run IGKR-Net on prospectively undersampled single- or multi-coil data from a scanner using one of the tested trajectories and compare PSNR, SSIM, and NMSE against ReconFormer and SwinGAN; if the advantage seen on retrospectively masked data disappears or reverses, the claim that the method generalizes to real accelerated MRI is refuted.

Watch

Extended reading notes

Core claim

The paper argues that undersampled k-space recovery should be framed as continuous coordinate-to-value regression rather than discrete image-to-image restoration. IGKR-Net encodes the sampled k-values and their coordinates into a latent space using transformer attention, then decodes any unsampled coordinate into a k-value; image-domain guidance from the corrupted reconstruction and a tri-attention refinement stage improve the result, while a multi-stage low-to-high-resolution training schedule avoids the over-smoothing caused by filling a dense spectrum directly from sparse samples. The authors report that this design achieves the best PSNR, SSIM, NMSE, and LPIPS among the compared methods on CC359, fastMRI, IXI, and SKM-TEA under 1D Cartesian, 2D Gaussian, and 2D random undersampling masks.

Load-bearing premise

The benchmark rests on retrospectively masking already-sampled k-space with 1D Cartesian and 2D Gaussian masks, and the reported gains could collapse if real accelerated acquisition introduces noise and trajectory effects those masks do not reproduce.

Editorial extensions

If this is right

  • At 20% and 40% sampling with 1D Cartesian and 2D Gaussian masks, IGKR-Net reports higher PSNR/SSIM and lower NMSE/LPIPS than the compared baselines on CC359, fastMRI, and IXI.
  • On multi-coil SKM-TEA at 25% and 12.5% sampling, it reports the best PSNR, SSIM, and NMSE among the methods compared.
  • Ablation results show that removing the low-to-high-resolution progressive strategy drops PSNR from 33.06 to 31.65 on CC359 with a 20% 1D Cartesian mask, indicating the staged recovery scheme is load-bearing.
  • Removing either the image-domain guidance module or the tri-attention refinement module degrades the reported metrics, so both components contribute to the final performance.
  • The reported k-space error maps indicate that the method better preserves high-frequency spectral content than the compared dual-domain baselines, which the authors link to sharper edges in the reconstructed images.

Reading between the lines

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

  • Editorial inference: because the decoder queries arbitrary coordinates, the same continuous k-space function could in principle be evaluated on non-Cartesian or variable-density trajectories beyond the 1D Cartesian and 2D Gaussian masks tested; that extension is not a paper claim.
  • Editorial inference: the reported gains come from retrospectively masked fully sampled k-space, so a reader should not assume identical clinical benefit until the method is tested on prospectively undersampled scanner data.
  • Editorial inference: the low-to-high-resolution staged querying recipe is general, and could be transferred to other sparse-measurement inverse problems such as non-Cartesian MRI or radio interferometry, though the paper does not test those settings.
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

3 major / 8 minor

Summary. The paper proposes IGKR-Net, a deep network for accelerated MRI reconstruction that recovers undersampled k-space using an implicit neural representation (INR) perspective. The architecture combines a low-resolution implicit transformer (LRIT), an image-domain guidance module (IDGM), a high-resolution implicit transformer (HRIT), and a tri-attention refinement module (TARM), trained with a four-stage progressive loss schedule. Experiments on the CC359, fastMRI, IXI, and SKM-TEA datasets compare the method against ZF, ResNet, UNet, SwinMR, RefineGAN, SwinGAN, ReconFormer, KIKI-Net, and D5C5 under 1D Cartesian and 2D Gaussian/random undersampling at multiple sampling ratios, reporting PSNR, SSIM, NMSE, and LPIPS. The central claim is that IGKR-Net consistently achieves the best performance with significant improvements over previous methods.

Significance. If the empirical claim holds, the paper offers a useful architectural contribution: an INR-based encoder–decoder for continuous k-space querying with image-domain guidance and progressive training, supported by a reasonable amount of experimental breadth across single-coil and multi-coil datasets, multiple masks, and several sampling ratios. The paper also reports an efficiency comparison and ablation studies for each module, and states that code will be released. However, the load-bearing empirical claim of 'significant improvements' is not backed by any statistical significance analysis, and one table has a dataset-label inconsistency; these issues currently leave the central performance claim conditional rather than established.

major comments (3)
  1. [Abstract; §4.2.1; Tables 1–4] The claim that IGKR-Net 'consistently achieves the best performance with significant improvements' is not supported by the reported statistics. On fastMRI (Table 2) the PSNR margins over the strongest baseline ReconFormer are 0.02, 0.04, 0.11, and 0.05 dB for the four mask/ratio settings, while the per-slice standard deviations are about 2.5–2.9 dB; on SKM-TEA (Table 4) the margins are 0.08 and 0.15 dB with no error bars or test statistics at all. No paired significance test, confidence interval, or per-subject variance estimate is reported in Tables 1–5. The consistent small advantage may be real, but the present evidence does not establish the word 'significant'; please add paired tests (e.g., Wilcoxon signed-rank or paired bootstrap over slices or volumes) or temper the wording accordingly.
  2. [§4.2.3; Table 5] Table 5 is captioned 'multi-coil SKM-TEA', but §4.2.3 states that this experiment is conducted using the CC359 dataset, and the 20% and 40% rows match the CC359 results in Table 1. This internal inconsistency undermines the mask-robustness evidence. Please correct the caption or the section text, and verify that the table reports the dataset the authors intended.
  3. [§3.2.5, Eq. (19)] Equation (19) prints the same expression in both branches of the piecewise definition of L_i. As written, L_2, L_3, and L_4 compare the intermediate reconstructions against the low-resolution targets K_lr and I_lr rather than the full-resolution K and I that the preceding sentence states. Because the multi-stage loss schedule in Algorithm 1 is load-bearing for the claimed benefit of progressive training, this is not purely cosmetic; please correct the equation so the i=2,3,4 branch uses K and I.
minor comments (8)
  1. [§3.1.1] The heading 'MRI Reconstrucion' has a typo; it should be 'MRI Reconstruction'.
  2. [§3.2.5; Algorithm 1] The text at the end of §3.2.5 refers to 'IGIT-Net' instead of 'IGKR-Net', and Algorithm 1 names the image guidance module 'DIFM' while the architecture section consistently calls it 'IDGM'; please unify the names.
  3. [§3.2.3, Eqs. (14)–(15)] Equations (14) and (15) reference 'the Encoder in Figure.4(a)' and 'the Decoder shown in Figure.4(b)', but the encoder and decoder are depicted in Figure 3(a) and Figure 3(b); the figure references should be corrected.
  4. [§4.1.1] The CC359 dataset is cited to Warfield et al. (2004), which describes the STAPLE segmentation algorithm rather than the Calgary–Campinas dataset; the authors should cite the actual CC359 dataset paper. Similarly, the IXI dataset is cited to Orhaug and Forssell (2021), which is an article on information extraction from images and not the IXI neuroimaging dataset; a proper dataset reference is needed.
  5. [§4.1.3] The statement that all baselines are 'retrained using their default parameter settings' should be clarified: it is unclear whether the baselines were retrained on the same training/validation splits as IGKR-Net and whether any hyperparameter tuning was performed, which is relevant for fair comparison.
  6. [§4.2.3] There are typos in §4.2.3: 'despiting' should be 'despite' and 'we can fine' should be 'we can find'; the sentence about RefineGAN outperforming by '3.66dB and 0.21dB' is correct numerically but should be split for readability.
  7. [§4.3.3, Table 7] The claim of an 'optimal balance' between parameters and computational complexity is not quantified; Table 7 shows that IGKR-Net has about 12 times the parameters of ReconFormer (13.89M vs 1.14M) while using fewer FLOPs, so a simple trade-off statement would be more precise.
  8. [Abstract] The abstract lists CC359, fastMRI, and IXI datasets, while the rest of the paper also validates on SKM-TEA; please make the abstract consistent with the full set of datasets.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central claim is an empirical architecture evaluation on held-out data, with self-citations appearing only in background sections.

full rationale

The claimed contribution is an empirical architecture evaluation, not a derivation that presupposes its own conclusion. IGKR-Net is trained on retrospectively undersampled k-space pairs generated by Eq. (1) and tested on held-out CC359, fastMRI, IXI, and SKM-TEA volumes with fully sampled reference data; Tables 1 through 5 report metrics on validation and test splits that were not used for training. No fitted parameter is relabeled as a prediction: the only fitted object is the network's weight set, and the predicted quantities are k-space and image outputs on unseen slices. The INR formulation in Eqs. (3)-(5) is a modeling choice rather than a self-referential definition; it does not assume the reconstruction quality it is meant to establish. Self-citations such as Meng et al. 2024 and Yang et al. 2024a,b,c appear in the introduction and related work as background context and are not load-bearing for the benchmark claims. No uniqueness theorem, external prior result, or ansatz is imported from the authors' own prior work to force the architecture. The lack of paired significance tests and the Table 5 caption inconsistency are correctness and evidence concerns, not circular reasoning. The evaluation is therefore self-contained against external held-out data, and the circularity score is 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities and fits no physical constants. Its central claim rests on supervised training with fully sampled k-space, simulated undersampling masks, and standard Fourier transforms. The main hand-chosen elements are the training schedule, mask design, and equal-weight loss combination. The largest unvalidated premise is that simulated masks faithfully mimic real accelerated acquisition.

free parameters (3)
  • Multi-stage training epoch schedule = (0, 20, 60, 100, 200)
    Chosen by hand in Section 4.1.3; the ablation in Table 6 shows that removing progressive staging lowers PSNR by 1.41 dB, so the schedule materially affects the result.
  • Undersampling mask central region = 8% of k-space lines for 1D Cartesian; 16% of k-space points for 2D Gaussian
    Set in Section 4.1.2 following FastMRI convention; this controls task difficulty and is part of the evaluation setup, not a physical constant.
  • Equal weighting of stage losses = 1.0 for each summed stage loss in Equation (20)
    No loss-weight tuning is reported; all four losses are added with equal weight, which is a hand-chosen simplification.
assumptions (5)
  • domain assumption Fully sampled k-space K is available as training target and is treated as ground truth.
    Equation (19) and Algorithm 1 supervise every stage against K/I or downsampled versions, so the benchmark claim depends on the reference k-space being accurate.
  • domain assumption Retrospective masks applied to fully sampled k-space approximate real accelerated acquisition.
    Section 4.1.2 generates all undersampled inputs with Cartesian or Gaussian masks; no prospectively undersampled data is used, so transfer to real acceleration protocols is assumed.
  • standard math FFT/IFFT correctly model the linear MRI encoding relationship between k-space and image.
    The method relies on the Fourier pair in Equations (9), (12), (16), and (18).
  • domain assumption Validation and test splits from the same datasets represent generalization to new scans.
    The superiority conclusion assumes that held-out volumes from CC359, fastMRI, IXI, and SKM-TEA predict performance on unseen patients; no external scanner or prospective cohort is tested.
  • domain assumption PSNR, SSIM, NMSE, and LPIPS reflect reconstruction utility.
    The paper declares superiority using these metrics without a reader study or clinical endpoint.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Continuous K-space Recovery Network with Image Guidance for Fast MRI Reconstruction." pith.science (2026). https://pith.science/paper/H5HDDN2S

@misc{pith2026241111282,
  author       = {Pith},
  title        = {Pith review of: Continuous K-space Recovery Network with Image Guidance for Fast MRI Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H5HDDN2S}},
  note         = {Machine review of arXiv:2411.11282}
}
read the original abstract

Magnetic resonance imaging (MRI) is a crucial tool for clinical diagnosis while facing the challenge of long scanning time. To reduce the acquisition time, fast MRI reconstruction aims to restore high-quality images from the undersampled k-space. Existing methods typically train deep learning models to map the undersampled data to artifact-free MRI images. However, these studies often overlook the unique properties of k-space and directly apply general networks designed for image processing to k-space recovery, leaving the precise learning of k-space largely underexplored. In this work, we propose a continuous k-space recovery network from a new perspective of implicit neural representation with image domain guidance, which boosts the performance of MRI reconstruction. Specifically, (1) an implicit neural representation based encoder-decoder structure is customized to continuously query unsampled k-values. (2) an image guidance module is designed to mine the semantic information from the low-quality MRI images to further guide the k-space recovery. (3) a multi-stage training strategy is proposed to recover dense k-space progressively. Extensive experiments conducted on CC359, fastMRI, and IXI datasets demonstrate the effectiveness of our method and its superiority over other competitors.

Figures

Figures reproduced from arXiv: 2411.11282 by the authors.

Figure 1
Figure 1. Our main idea. (a) Previous studies typically employ general networks to reconstruct the undersampled data both in the k-space and image domain. (b) We customize a continuous k-space recovery network from a new perspective of implicit neural representation with image domain guidance, thereby enhancing the performance of MRI reconstruction. Essentially, the aliasing artifacts arise because the ac￾celerated undersampl… view at source ↗
Figure 2
Figure 2. The architecture of the our IGKR-Net. Given the undersampled k-space, we first encode the coordinates 𝐶𝑠 and k-values 𝐾𝑠 of the sampled points, and the resulting features are added and then fed into LRIT. In LRIT, we use the LR coordinates to query and get the recovered results, i.e., 𝐾̂ 1 and 𝐼̂ 1 . Next, we send 𝐼̂ 1 into IDGM and get 𝐾̂ 2 and 𝐼̂ 2 with the guidance of the low-quality image 𝐼𝑠 . Then, the k-values… view at source ↗
Figure 3
Figure 3. Our transformer based Encoder (a) and Decoder (b). 3.1.3. Continuous K-space Recovery Network via INR As described in equation (1), it is common to undersam￾ple the k-space for accelerated MRI. The correct recovery of the unsampled k-values, i.e., 𝐾𝑢𝑠, is crucial for high￾quality MRI reconstruction. Inspired by the continuous im￾age modeling of INR, we customize a k-space recovery net. Specifically, 𝐾𝑢𝑠 can be queri… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: (a) The proposed IDGM consists of two stages, i.e., shallow fusion stage and deep fusion stage. (b) The proposed TARM consists of three branches, i.e., PA, CA, and SA. Specifically, given a set of sampled k-values 𝐾𝑠 and its coordinates 𝐶𝑠 , a tokenization procedure is…
Figure 5
Figure 5. Figure 5: Visual comparison of different methods under various undersampling masks. Red boxes illustrate the enlarged views in detail. Yellow ellipses highlight the performance differences in the results of anatomical structures among various methods. 4.1.3. Baselines and Traini…
Figure 6
Figure 6. Figure 6: Results of the mask experiments. The X-axis denotes the sampling ratio, while the Y-axis corresponds to various metrics [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 8
Figure 8. Figure 8: Visual comparison results with k-space error maps [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 7
Figure 7. Figure 7: Visual comparison results of the mask experiments. directly up-sampling the low-resolution outputs obtained by LRIT as the reconstruction results, resulted in a severe decrease of (PSNR, SSIM) by (4.50, 0.1872), respectively. This suggests that using the dense k-space …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. DH-Mamba: Exploring Dual-domain Hierarchical State Space Models for MRI Reconstruction

    eess.IV 2025-01 conditional novelty 6.0 of 10

    DH-Mamba is a dual-domain hierarchical Mamba network that uses circular k-space scanning and local diversity enhancement to outperform prior MRI reconstruction methods on three public datasets.

Reference graph

Works this paper leans on

57 extracted references · 39 canonical work pages · cited by 1 Pith paper

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in ":" * " " * FUNCTION f...

  2. [2]

    , author Ekinci, D.A

    author Aghabiglou, A. , author Ekinci, D.A. , author Eksioglu, E.M. , author Toreyin, B.U. , year 2024 . title Mr image reconstruction using iterative up and downsampling network . journal Expert Systems with Applications volume 237 , pages 121590

    work page 2024

  3. [3]

    , author Moeller, S

    author Ak c akaya, M. , author Moeller, S. , author Weing \"a rtner, S. , author U g urbil, K. , year 2019 . title Scan-specific robust artificial-neural-networks for k-space interpolation (raki) reconstruction: Database-free deep learning for fast imaging . journal Magnetic resonance in medicine volume 81 , pages 439--453

    work page 2019

  4. [4]

    , author Zhao, W

    author Chen, H. , author Zhao, W. , author Xu, T. , author Shi, G. , author Zhou, S. , author Liu, P. , author Li, J. , year 2024 . title Spectral-wise implicit neural representation for hyperspectral image reconstruction . journal IEEE Transactions on Circuits and Systems for Video Technology volume 34 , pages 3714--3727 . :10.1109/TCSVT.2023.3318366

    arXiv 2024

  5. [5]

    , author Liu, S

    author Chen, Y. , author Liu, S. , author Wang, X. , year 2021 . title Learning continuous image representation with local implicit image function , in: booktitle Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. pages 8628--8638

    work page 2021

  6. [6]

    , author Chen, Y

    author Chen, Z. , author Chen, Y. , author Liu, J. , author Xu, X. , author Goel, V. , author Wang, Z. , author Wang, X. , year 2022 . title Videoinr: Learning video implicit neural representation for continuous space-time super-resolution , in: booktitle Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. pages ...

    work page 2022

  7. [7]

    , author Zhang, H

    author Chen, Z. , author Zhang, H. , year 2019 . title Learning implicit fields for generative shape modeling , in: booktitle Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. pages 5939--5948

    work page 2019

  8. [8]

    , author Schmidt, A.M

    author Desai, A.D. , author Schmidt, A.M. , author Rubin, E.B. , author Sandino, C.M. , author Black, M.S. , author Mazzoli, V. , author Stevens, K.J. , author Boutin, R. , author R \'e , C. , author Gold, G.E. , et al., year 2022 . title Skm-tea: A dataset for accelerated mri reconstruction with dense image labels for quantitative clinical evaluation . j...

    arXiv 2022

Show all 57 references
  1. [9]

    , year 2020

    author Doneva, M. , year 2020 . title Mathematical models for magnetic resonance imaging reconstruction: An overview of the approaches, problems, and future research areas . journal IEEE Signal Processing Magazine volume 37 , pages 24--32 . :10.1109/MSP.2019.2936964

    2020

  2. [10]

    , author Jun, Y

    author Eo, T. , author Jun, Y. , author Kim, T. , author Jang, J. , author Lee, H.J. , author Hwang, D. , year 2018 . title Kiki-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images . journal Magnetic resonance in medicine v...

    2018

  3. [11]

    , author Yan, Y

    author Feng, C.M. , author Yan, Y. , author Chen, G. , author Xu, Y. , author Hu, Y. , author Shao, L. , author et al , year 2022 . title Multi-modal transformer for accelerated mr imaging . journal IEEE Transactions on Medical Imaging

    2022

  4. [12]

    , author Yan, Y

    author Feng, C.M. , author Yan, Y. , author Fu, H. , author Xu, Y. , year 2021 . title Task transformer network for joint mri reconstruction and super-resolution , in: booktitle Medical Image Computing and Computer Assisted Intervention--MICCAI 2021: 24th International Confere...

    2021

  5. [13]

    , author Boesiger, P

    author Gamper, U. , author Boesiger, P. , author Kozerke, S. , year 2008 . title Compressed sensing in dynamic mri . journal Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine volume 59 , pages 365--373

    2008

  6. [14]

    , author Vandergheynst, P

    author Grattarola, D. , author Vandergheynst, P. , year 2022 . title Generalised implicit neural representations . journal Advances in Neural Information Processing Systems volume 35 , pages 30446--30458

    2022

  7. [15]

    , author Jakob, P.M

    author Griswold, M.A. , author Jakob, P.M. , author Heidemann, R.M. , author Nittka, M. , author Jellus, V. , author Wang, J. , author Kiefer, B. , author Haase, A. , year 2002 . title Generalized autocalibrating partially parallel acquisitions (grappa). journal Magnetic reson...

    2002

  8. [16]

    , author Mei, Y

    author Guo, P. , author Mei, Y. , author Zhou, J. , author Jiang, S. , author Patel, V.M. , year 2023 . title Reconformer: Accelerated mri reconstruction using recurrent transformer . journal IEEE transactions on medical imaging

    2023

  9. [17]

    author Haldar, J.P. , . title Autocalibrated loraks for fast constrained mri reconstruction , in: booktitle 2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI)

    2015

  10. [18]

    , author Klatzer, T

    author Hammernik, K. , author Klatzer, T. , author Kobler, E. , author Recht, M.P. , author Sodickson, D.K. , author Pock, T. , author Knoll, F. , year 2018 . title Learning a variational network for reconstruction of accelerated mri data . journal Magnetic resonance in medici...

    2018

  11. [19]

    , author Zhang, X

    author He, K. , author Zhang, X. , author Sun, J. , year 2016 . title Deep residual learning for image recognition , in: booktitle Proceedings of the IEEE conference on computer vision and pattern recognition , pp. pages 770--778

    2016

  12. [20]

    , author Fang, Y

    author Huang, J. , author Fang, Y. , author Wu, Y. , author Wu, H. , author Gao, Z. , author Li, Y. , author et al , year 2022 . title Swin transformer for fast mri . journal Neurocomputing volume 493 , pages 281--304

    2022

  13. [21]

    , author Kim, H.P

    author Hyun, C.M. , author Kim, H.P. , author Lee, S.M. , author Lee, S. , author Seo, J.K. , year 2018 . title Deep learning for undersampled mri reconstruction . journal Physics in Medicine & Biology volume 63 , pages 135007

    2018

  14. [22]

    title A deep cascade of convolutional neural networks for dynamic mr image reconstruction

    author Jo , author Schlemper , author Jose , author Caballero , author Joseph , author Hajnal , author et al , year 2017 . title A deep cascade of convolutional neural networks for dynamic mr image reconstruction. journal IEEE Transactions on Medical Imaging

    2017

  15. [23]

    , author Yoo, J

    author Lee, D. , author Yoo, J. , author Tak, S. , author Ye, J.C. , year 2018 . title Deep residual learning for accelerated mri using magnitude and phase networks . journal IEEE Transactions on Biomedical Engineering volume 65 , pages 1985--1995 . :10.1109/TBME.2018.2821699

    2018

  16. [24]

    , author Dai, T

    author Li, H. , author Dai, T. , author Li, Y. , author Zou, X. , author Xia, S.T. , year 2022 . title Adaptive local implicit image function for arbitrary-scale super-resolution , in: booktitle 2022 IEEE International Conference on Image Processing (ICIP) , organization IEEE ...

    2022

  17. [25]

    , author Firkat, E

    author Li, J. , author Firkat, E. , author Zhu, J. , author Zhu, B. , author Zhu, J. , author Hamdulla, A. , year 2025 . title Hi-slam: Hierarchical implicit neural representation for slam . journal Expert Systems with Applications volume 271 , pages 126487 . https://www.scien...

    2025

  18. [26]

    , author Donoho, D

    author Lustig, M. , author Donoho, D. , author Pauly, J.M. , year 2007 . title Sparse mri: The application of compressed sensing for rapid mr imaging . journal Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine v...

    2007

  19. [27]

    , author Pauly, J.M

    author Lustig, M. , author Pauly, J.M. , year 2010 . title Spirit: iterative self-consistent parallel imaging reconstruction from arbitrary k-space . journal Magnetic resonance in medicine volume 64 , pages 457--471

    2010

  20. [28]

    , author Gong, E

    author Mardani, M. , author Gong, E. , author Cheng, J.Y. , author Vasanawala, S.S. , author Zaharchuk, G. , author Xing, L. , author et al , year 2019 . title Deep generative adversarial neural networks for compressive sensing mri . journal IEEE Transactions on Medical Imagin...

    2019

  21. [29]

    , author Yang, Z

    author Meng, Y. , author Yang, Z. , author Shi, Y. , author Song, Z. , year 2024 . title Boosting vit-based mri reconstruction from the perspectives of frequency modulation, spatial purification, and scale diversification . journal arXiv preprint arXiv:2412.10776

    2024 arXiv

  22. [30]

    , author Srinivasan, P.P

    author Mildenhall, B. , author Srinivasan, P.P. , author Tancik, M. , author Barron, J.T. , author Ramamoorthi, R. , author Ng, R. , year 2021 . title Nerf: Representing scenes as neural radiance fields for view synthesis . journal Communications of the ACM volume 65 , pages 99--106

    2021

  23. [31]

    , author Oliver, R

    author Minhas, A.S. , author Oliver, R. , year 2022 . title Magnetic resonance imaging basics . journal Electrical Properties of Tissues: Quantitative Magnetic Resonance Mapping , pages 47--82

    2022

  24. [32]

    , author Chung, T

    author Moore, M.M. , author Chung, T. , year 2017 . title Review of key concepts in magnetic resonance physics . journal Pediatric radiology volume 47 , pages 497--506

    2017

  25. [33]

    , author Nag, S

    author Nitski, O. , author Nag, S. , author McIntosh, C. , author Wang, B. , year 2020 . title Cdf-net: Cross-domain fusion network for accelerated mri reconstruction , in: editor Martel, A.L. , editor Abolmaesumi, P. , editor Stoyanov, D. , editor Mateus, D. , editor Zuluaga,...

    2020

  26. [34]

    , author Forssell, G

    author Orhaug, T. , author Forssell, G. , year 2021 . title Information extraction from images , in: booktitle Outer Space-A New Dimension of the Arms Race . publisher Routledge , pp. pages 215--227

    2021

  27. [35]

    , author Weiger, M

    author Pruessmann, K.P. , author Weiger, M. , author Scheidegger, M.B. , author Boesiger, P. , year 1999 . title Sense: sensitivity encoding for fast mri . journal Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medici...

    1999

  28. [36]

    , author Jeong, W.K

    author Quan, T.M. , author Jeong, W.K. , year 2018 . title Compressed sensing mri reconstruction using a generative adversarial network with a cyclic loss . journal IEEE transactions on medical imaging volume 37 , pages 1488--1497

    2018

  29. [37]

    , author Baratin, A

    author Rahaman, N. , author Baratin, A. , author Arpit, D. , author Draxler, F. , author Lin, M. , author Hamprecht, F. , author Bengio, Y. , author Courville, A. , year 2019 . title On the spectral bias of neural networks , in: booktitle International Conference on Machine Le...

    2019

  30. [38]

    , author Xia, W

    author Ran, M. , author Xia, W. , author Huang, Y. , author Lu, Z. , author Bao, P. , author Liu, Y. , author Sun, H. , author Zhou, J. , author Zhang, Y. , year 2020 . title Md-recon-net: a parallel dual-domain convolutional neural network for compressed sensing mri . journal...

    2020

  31. [39]

    , author Bresler, Y

    author Ravishankar, S. , author Bresler, Y. , year 2011 . title Mr image reconstruction from highly undersampled k-space data by dictionary learning . journal IEEE Transactions on Medical Imaging volume 30 , pages 1028--1041 . :10.1109/TMI.2010.2090538

    2011

  32. [40]

    , author Hendee, W.R

    author Scherzinger, A.L. , author Hendee, W.R. , year 1985 . title Basic principles of magnetic resonance imaging—an update . journal Western journal of medicine volume 143 , pages 782

    1985

  33. [41]

    , author Caballero, J

    author Schlemper, J. , author Caballero, J. , author Hajnal, J.V. , author Price, A.N. , author Rueckert, D. , year 2018 . title A deep cascade of convolutional neural networks for dynamic mr image reconstruction . journal IEEE transactions on Medical Imaging volume 37 , pages...

    2018

  34. [42]

    , author Martel, J

    author Sitzmann, V. , author Martel, J. , author Bergman, A. , author Lindell, D. , author Wetzstein, G. , year 2020 . title Implicit neural representations with periodic activation functions . journal Advances in neural information processing systems volume 33 , pages 7462--7473

    2020

  35. [43]

    , author Cameron, D

    author Vassiliou, V.S. , author Cameron, D. , author Prasad, S.K. , author Gatehouse, P.D. , year 2018 . title Magnetic resonance imaging: Physics basics for the cardiologist . journal JRSM cardiovascular disease volume 7 , pages 2048004018772237

    2018

  36. [44]

    , author Kumar, K.M

    author Velayudham, A. , author Kumar, K.M. , author Priya, M.K. , year 2025 . title Improving brain mri denoising using convolutional autoencoder and sparse representations . journal Expert Systems with Applications volume 263 , pages 125711

    2025

  37. [45]

    , author Ran, M

    author Wang, Z. , author Ran, M. , author Yang, Z. , author Yu, H. , author Jin, J. , author Wang, T. , author Lu, J. , author Zhang, Y. , year 2024 . title Generalizable mri motion correction via compressed sensing equivariant imaging prior . journal IEEE Transactions on Circ...

    2024

  38. [46]

    , author Zou, K.H

    author Warfield, S.K. , author Zou, K.H. , author Wells, W.M. , year 2004 . title Simultaneous truth and performance level estimation (staple): an algorithm for the validation of image segmentation . journal IEEE transactions on medical imaging volume 23 , pages 903--921

    2004

  39. [47]

    , author Li, Y

    author Wu, Q. , author Li, Y. , author Sun, Y. , author Zhou, Y. , author Wei, H. , year 2022 . title An arbitrary scale super-resolution approach for 3d mr images via implicit neural representation . journal IEEE Journal of Biomedical and Health Informatics

    2022

  40. [48]

    , author Wang, Z

    author Xu, X. , author Wang, Z. , author Shi, H. , year 2021 . title Ultrasr: Spatial encoding is a missing key for implicit image function-based arbitrary-scale super-resolution . journal arXiv preprint arXiv:2103.12716

    2021 arXiv

  41. [49]

    , author Yu, S

    author Yang, G. , author Yu, S. , author Dong, H. , author Slabaugh, G. , author Dragotti, P.L. , author Ye, X. , author et al , year 2017 . title Dagan: Deep de-aliasing generative adversarial networks for fast compressed sensing mri reconstruction . journal IEEE transactions...

    2017

  42. [50]

    , author Fu, K

    author Yang, Z. , author Fu, K. , author Duan, M. , author Qu, L. , author Wang, S. , author Song, Z. , year 2024 a. title Separate and conquer: Decoupling co-occurrence via decomposition and representation for weakly supervised semantic segmentation , in: booktitle Proceeding...

    2024

  43. [51]

    , author Meng, Y

    author Yang, Z. , author Meng, Y. , author Fu, K. , author Wang, S. , author Song, Z. , year 2024 b. title More: Class patch attention needs regularization for weakly supervised semantic segmentation . journal arXiv preprint arXiv:2412.11076

    2024 arXiv

  44. [52]

    , author Meng, Y

    author Yang, Z. , author Meng, Y. , author Fu, K. , author Wang, S. , author Song, Z. , year 2024 c. title Tackling ambiguity from perspective of uncertainty inference and affinity diversification for weakly supervised semantic segmentation . https://arxiv.org/abs/2404.08195, ...

    2024 arXiv

  45. [53]

    , author Knoll, F

    author Zbontar, J. , author Knoll, F. , author Sriram, A. , author Murrell, T. , author Huang, Z. , author Muckley, M.J. , year 2018 . title fastmri: An open dataset and benchmarks for accelerated mri . journal arXiv preprint arXiv:1811.08839

    2018 arXiv

  46. [54]

    , author Yang, T

    author Zhao, X. , author Yang, T. , author Zhang, X. , year 2023 . title Swingan: A dual-domain swin transformer-based generative adversarial network for mri reconstruction . journal Computers in Biology and Medicine volume 153 , pages 106513

    2023

  47. [55]

    , author Li, Z

    author Zheng, X. , author Li, Z. , author Liu, D. , author Zhou, X. , author Shan, C. , year 2024 . title Spatial attention-guided light field salient object detection network with implicit neural representation . journal IEEE Transactions on Circuits and Systems for Video Tec...

    2024

  48. [56]

    , author Zhou, S.K

    author Zhou, B. , author Zhou, S.K. , year 2020 . title Dudornet: Learning a dual-domain recurrent network for fast mri reconstruction with deep t1 prior , in: booktitle 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. pages 4272--4281 . :10.110...

    2020

  49. [57]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

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