Pith. sign in

REVIEW 3 major objections 5 minor 37 references

Radiologist-in-the-Loop Self-Training for Generalizable CT Metal Artifact Reduction

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that putting a radiologist-aligned quality assessor inside a self-training loop lets CT metal artifact reduction generalize to real clinical images, with the assessor doubling as a no-ground-truth evaluation metric.

desk verdict A solid semi-supervised MAR method with a genuinely useful quality-filtering idea, but the clinical superiority claim rests on a circular CQA evaluation and needs independent radiologist confirmation. read the letter →

arxiv 2501.15610 v1 pith:IE3673SE submitted 2025-01-26 eess.IV cs.CV

Chenglong Ma , Zilong Li , Yuanlin Li , Jing Han , Junping Zhang , Yi Zhang , Jiannan Liu , Hongming Shan This is my paper · ORCID
classification eess.IVcs.CV
keywords CTmetalartifactreductionsemi-supervisedlearningself-trainingclinicalqualityassessmentpseudoground-truthselectiondomaingeneralizationradiologist-in-the-loopmedicalimage
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

RISE-MAR aims to solve a specific failure: MAR networks trained on simulated CT artifacts lose accuracy on real clinical scans because no paired ground truth exists there. The paper's central claim is that this domain gap can be closed by putting a radiologist-aligned clinical quality assessor (CQA) inside a self-training loop. The CQA, trained on radiologists' ten-point quality ratings, decides which pseudo ground-truths produced by a teacher MAR network are good enough to train a student network, and the student then refreshes the teacher by exponential moving average, producing more high-quality pseudo ground-truths over time. On clinical test sets from DeepLesion and a private dental dataset, RISE-MAR obtains the best CQA quality scores among the compared methods, and the paper argues this indicates better generalization to real metal artifacts. A sympathetic reader would care because the same CQA could serve as an automated evaluation metric where ground truth is unavailable.

What carries the argument

The load-bearing object is the clinical quality assessor (CQA), a transformer encoder with a quality head that maps a CT image to a ten-point clinical quality score, trained to match two radiologists' consensus ratings. The CQA is the gate in Eqs. (6)–(7): a pseudo ground-truth $\tilde{Y}_{\mathrm{cli}}$ from the teacher is kept only when its predicted quality $q_{\tilde{Y}}$ falls in $Q=[7,10]$; otherwise the sample is discarded to avoid confirmation bias. The second half of the mechanism is self-training: the student learns from verified pseudo pairs plus simulated pairs, and the teacher's parameters are updated by exponential moving average, so the teacher's clinical knowledge improves and more of its future predictions pass the CQA gate. This two-part loop—quality filter plus expanding pseudo ground-truth pool—is what transfers simulated-domain MAR knowledge to real clinical CT images.

What would settle it

Take a fresh set of real-metal-artifact CTs that were not used in CQA training, run RISE-MAR, SemiMAR, UDAMAR, and β-CycleGAN on them, and have independent radiologists blindly rank the outputs by diagnostic quality. If the CQA scores do not match the radiologists' rankings on these unseen outputs, the load-bearing claim that CQA provides radiologist-aligned generalization would be refuted.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that combining quality assurance and quantity assurance is what makes semi-supervised MAR generalize. Quality assurance is the CQA: a transformer-based multi-scale encoder with a quality head that outputs a quality score $\hat{q}=\sum_{k=1}^{10} k p^{(k)}$ on a ten-point Likert scale, trained with cross-entropy and supervised contrastive loss plus Diverse Quality Augmentation (MixUp across low-, moderate-, and high-quality images). Quantity assurance is the teacher–student loop: teacher $f_\phi$ pretrained on simulated pairs proposes pseudo ground-truths for unpaired clinical CTs; CQA accepts only those with quality in $Q=[7,10]$; the student $f_\theta$ trains on the resulting pseudo pairs together with simulated pairs; and EMA with decay $\eta=0.999$ updates the teacher. The reported outcome is that RISE-MAR achieves the highest CQA quality on clinical DeepLesion (8.55) and clinical Dental (7.30), with visual results showing streak removal and restoration of bony and soft-tissue structures. The paper concludes that radiologist feedback distilled into CQA, rather than a fixed prior network, is what prevents confirmation bias and expands the clinical training pool.

Load-bearing premise

The framework assumes the CQA's learned ten-point quality score faithfully reproduces radiologist judgment on clinical images outside CQA's training distribution, and that using the same CQA both to select training pseudo ground-truths and to score final clinical test outputs does not bias the comparison.

Editorial extensions

If this is right

  • Semi-supervised MAR no longer needs a frozen pretrained prior: the teacher network's clinical knowledge is updated during training, so pseudo ground-truths improve in quality as training proceeds.
  • CQA provides a quantitative evaluation route for clinical MAR outputs where PSNR and SSIM cannot be computed, since no artifact-free ground truth exists for real metal artifact cases.
  • The ablation results imply the quality threshold matters: too loose a range admits low-quality pseudo ground-truths and degrades out-of-domain performance, while too strict a range starves the training pool.
  • Because the MAR backbone can be any effective learning-based network, the same radiologist-in-the-loop recipe can be attached to other MAR architectures, including those that consume linear-interpolation backprojected images.
  • The CQA quality scores on clinical test sets are the paper's headline evidence for generalization across anatomies (torso and dental).

Reading between the lines

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

  • We infer that the reported clinical quality gap is partly self-consistent by construction: the same CQA that filters training pseudo ground-truths also scores the final test outputs, so an independent blinded radiologist re-read of RISE-MAR and comparison outputs would be a stronger test of clinical benefit.
  • We infer that the CQA's blindness to subtle structural distortions in artifact-light regions, acknowledged in the paper's limitations, means the quality gate may admit clinically misleading pseudo ground-truths even when the numeric score is high.
  • We infer that a natural extension is to use the CQA score as a continuous reward (for example via reinforcement learning) rather than a binary gate, which the paper names as future work and which would let quality feedback shape the MAR network's updates directly.
  • We infer that the fixed quality range $Q=[7,10]$ may not transfer to new clinical domains: the optimal threshold depends on the distribution of pseudo ground-truth qualities, and the ablation shows both stricter and looser ranges hurt performance.
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 / 5 minor

Summary. The paper proposes RISE-MAR, a semi-supervised CT metal artifact reduction framework that combines a radiologist-aligned Clinical Quality Assessor (CQA) with teacher-student self-training. A teacher network pretrained on simulated data generates pseudo ground-truths for unpaired clinical images; the CQA filters these candidates to a predefined quality range (Q=[7,10]), and the student is trained jointly on simulated pairs and CQA-approved clinical pseudo pairs. The teacher is updated by EMA, which the authors argue progressively improves the quality and quantity of pseudo ground-truths. Experiments compare RISE-MAR with supervised, unsupervised, and semi-supervised baselines on simulated in-domain, simulated out-of-domain, and real clinical test sets. On the simulated out-of-domain CTPelvic1K set, RISE-MAR attains the best PSNR, SSIM, and CQA quality; on the clinical DeepLesion and Dental sets, RISE-MAR reports the best CQA quality scores. The paper includes ablations of the CQA training strategy, the quality threshold Q, and the self-training components.

Significance. If the clinical claim is accepted, RISE-MAR addresses a real problem: semi-supervised MAR methods with fixed priors suffer from confirmation bias and limited pseudo-label diversity. The use of a learned quality assessor to filter pseudo ground-truths is a reasonable and potentially impactful idea, and the proposed DQAug strategy for training the assessor is a useful contribution. The simulated out-of-domain experiments on CTPelvic1K provide credible independent evidence of generalization: RISE-MAR improves PSNR from 46.42 (SemiMAR) to 47.88 and SSIM from 99.10% to 99.25%, and the qualitative error maps support the claim. The code is publicly available, the paper reports ablations for the main design choices, and the evaluation uses multiple datasets with different anatomies. However, the headline clinical-superiority claim depends entirely on the CQA metric, and the same CQA model is entangled with the training objective. Because the manuscript itself concedes that CQA may miss structural distortions in low-artifact images, the clinical conclusion requires independent radiologist validation or a substantially weakened claim.

major comments (3)
  1. [Sec. III-E, Tables II and III, Eq. (6)-(7)] The clinical-superiority claim is evaluated exclusively with CQA quality scores on the clinical test sets, but the same CQA model is used in Eqs. (6)-(7) to filter the pseudo ground-truths that train RISE-MAR (Q=[7,10]). The student network is therefore explicitly shaped to satisfy this particular quality scorer, so the clinical rows of Tables II and III may reflect self-reference rather than improved clinical generalization. The Discussion (Sec. IV) concedes that CQA 'may not adequately detect structural distortions when artifacts are not prominent,' which is exactly the failure mode that would let a method inflate CQA scores while producing clinically misleading images. To support the abstract's claim of enhanced generalization on real clinical CT images, the authors need an independent evaluation of the final outputs—for example, blinded radiologist ratings on a holdout set of clinical images—rather than only the CQA scores produced by the model used in training.
  2. [Sec. II-A.1, Sec. III-A.4, Sec. III-E] The quality annotation protocol for paired images uses a piecewise rule based on absolute pixel errors in the region of interest, and the Discussion argues that CQA prioritizes clinical quality beyond PSNR/SSIM. If the radiologists' ratings are effectively derived from pixel-error thresholds, then the CQA may be approximating a distortion measure rather than a clinical acceptability measure, and the claimed distinction in Sec. IV is not established. The paper should report inter-rater agreement, the exact piecewise rule, and an independent radiologist validation of CQA scores on MAR outputs, especially given the acknowledged limitation that CQA may not detect structural distortions when artifacts are not prominent.
  3. [Sec. III-F.3, Table IV, Sec. III-E] The quality threshold Q=[7,10] is selected on the simulated CTPelvic1K set using CQA quality, PSNR, and SSIM, and then the same CQA is used as the primary clinical metric. This introduces a selection-bias interaction: the clinical comparison is not a neutral test of the method. At minimum, the authors should report the clinical CQA scores for the ablated configurations (e.g., Q=[1,10], Q=[9,10]) on the clinical test sets, and provide confidence intervals or significance tests for the differences in Tables II and III. Without such information, the magnitude of the clinical gains (e.g., 8.554 vs. 8.154 on DeepLesion) cannot be assessed.
minor comments (5)
  1. [Title and Sec. II heading] The phrase 'Self-T raining' in the title and in the section header contains an erroneous space; please fix.
  2. [Eq. (2)] Using p for both the one-hot quality label and the probability vector pψ in Eq. (2) is confusing; rename the label (e.g., y) to avoid ambiguity.
  3. [Fig. 6] The rows labeled with checkmarks, crosses, and λ values are not self-explanatory; please clarify the configuration of each row and the meaning of the color-coded bars in the caption.
  4. [Fig. 4] The caption refers to RISE-MAR, but the figure legend is not labeled; ensure the curve colors are defined in the caption so that the reader can map each curve to a method.
  5. [Sec. III-D] The 'Supervised [13]' baseline appears to cite the β-CycleGAN paper; if the supervised model is taken from that work its architecture should be made explicit, otherwise a separate citation for the supervised baseline is needed.

Circularity Check

1 steps flagged · score 4.0 of 10

Clinical superiority claim rests on CQA as both training filter and test metric; independent PSNR/SSIM supports out-of-domain generalization but not the clinical claim.

  1. fitted input called prediction [Sec. II-B.2, Eq. (6)-(7) and Sec. III-E, Tables II-III]
    ""When q eY falls outside a predefined quality range Q, it indicates that eY cli does not qualify as a pseudo ground-truth due to potential inaccuracies. Consequently, we exclude such cases from the training to prevent confirmation bias." ... "For evaluation, we provide CQA quality result of different MAR methods for quantitative comparison (in the last group of Tables II and III).""

    The same CQA model is used twice: as a selection filter on pseudo ground-truths in Eq. (7) (only accepted when CQA quality is in Q=[7,10]) and as the sole quantitative metric for the clinical test results in Sec. III-E. RISE-MAR is therefore trained to generate outputs that satisfy the CQA's notion of quality, and then those same outputs are scored with that very metric to claim clinical superiority. Since no independent radiologist assessment of the final MAR outputs is reported, the clinical-generalization claim is partly a consequence of optimizing toward the evaluation metric.

full rationale

The paper has genuine independent evidence for out-of-domain generalization on simulated data, where RISE-MAR achieves strong PSNR and SSIM on the simulated CTPelvic1K test set. The central clinical claim, however, is circular in a partial and important sense: the CQA is used both to admit pseudo ground-truths into training and to score the clinical test outputs, so RISE-MAR is explicitly shaped to satisfy CQA while CQA is the only quantitative yardstick on real clinical images. The Discussion itself flags that CQA 'may not adequately detect structural distortions when artifacts are not prominent,' and no blinded radiologist validation of the final outputs is provided. This limits the clinical-superiority claim to a metric-entanglement caveat rather than a fully independent validation. The paper's self-citations, such as the reconstruction network [25] and contrastive learning [27], are not load-bearing in the same way: they support architecture and loss choices, not the clinical generalization claim itself. There is no imported uniqueness theorem and no ansatz smuggled by citation. Overall, the core MAR mechanism is not circular, but the clinical evaluation is, so a moderate score is warranted.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The framework rests on radiologist annotations distilled into CQA and on simulated artifacts standing in for real clinical corruption. No new physical entities are introduced. The free parameters are few, but the quality threshold is selected on the target out-of-domain benchmark and then reused for clinical evaluation, which is the main circularity concern.

free parameters (3)
  • Quality threshold range Q = [7, 10]
    Pseudo ground-truths are kept only if CQA quality score lies in this range. It is chosen by ablation on the simulated CTPelvic1K out-of-domain set (Table IV) and then applied to all clinical experiments.
  • lambda_scl = 0.01
    Balancing factor for the supervised contrastive loss in Eq. (4). The paper reports it as the best of 0, 0.001, 0.01, and 0.1 in Fig. 6.
  • EMA decay eta = 0.999
    Decay rate for teacher network updates in Eq. (9). It is a standard large EMA value but is not ablated in the paper.
assumptions (4)
  • domain assumption Radiologist ten-point Likert quality scores are a valid, transferable measure of clinical CT image quality.
    Sec. II-A.1 defines clinical quality and uses it as all CQA labels. If these scores do not reflect diagnostic usability, both the training filter and the clinical test metric lose meaning.
  • domain assumption The CQA model generalizes from its annotated training set to unseen MAR outputs, including the final outputs of RISE-MAR.
    Eq. (6)-(7) use CQA to accept pseudo ground-truths and Sec. III-E uses CQA to score methods. No independent radiologist check on final outputs is reported.
  • domain assumption Simulated artifacts following Zhang et al. [30] are a sufficient proxy for real clinical artifacts for pretraining and for training CQA on 25,000 simulated pairs.
    All paired data and the simulated part of the clinical quality assessment dataset are generated synthetically. The domain gap is exactly the problem the paper tries to solve, so the simulation is a load-bearing proxy.
  • ad hoc to paper DQAug moderate-quality images from under-trained MAR networks plus MixUp cover the quality spectrum relevant for clinical MAR outputs.
    Sec. II-A.2 assumes this augmentation makes CQA robust to subtle distortions. The paper does not quantitatively characterize whether these synthetic moderate-quality images match distortions produced by real MAR methods.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Radiologist-in-the-Loop Self-Training for Generalizable CT Metal Artifact Reduction." pith.science (2026). https://pith.science/paper/IE3673SE

@misc{pith2026250115610,
  author       = {Pith},
  title        = {Pith review of: Radiologist-in-the-Loop Self-Training for Generalizable CT Metal Artifact Reduction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IE3673SE}},
  note         = {Machine review of arXiv:2501.15610}
}
read the original abstract

Metal artifacts in computed tomography (CT) images can significantly degrade image quality and impede accurate diagnosis. Supervised metal artifact reduction (MAR) methods, trained using simulated datasets, often struggle to perform well on real clinical CT images due to a substantial domain gap. Although state-of-the-art semi-supervised methods use pseudo ground-truths generated by a prior network to mitigate this issue, their reliance on a fixed prior limits both the quality and quantity of these pseudo ground-truths, introducing confirmation bias and reducing clinical applicability. To address these limitations, we propose a novel Radiologist-In-the-loop SElf-training framework for MAR, termed RISE-MAR, which can integrate radiologists' feedback into the semi-supervised learning process, progressively improving the quality and quantity of pseudo ground-truths for enhanced generalization on real clinical CT images. For quality assurance, we introduce a clinical quality assessor model that emulates radiologist evaluations, effectively selecting high-quality pseudo ground-truths for semi-supervised training. For quantity assurance, our self-training framework iteratively generates additional high-quality pseudo ground-truths, expanding the clinical dataset and further improving model generalization. Extensive experimental results on multiple clinical datasets demonstrate the superior generalization performance of our RISE-MAR over state-of-the-art methods, advancing the development of MAR models for practical application. Code is available at https://github.com/Masaaki-75/rise-mar.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 34 canonical work pages

  1. [1]

    CT dose reduction and dose management tools: overview of available options,

    C. H. McCollough, M. R. Bruesewitz, and J. M. Kofler Jr, “CT dose reduction and dose management tools: overview of available options,” Radiographics, vol. 26, no. 2, pp. 503–512, 2006

    work page 2006

  2. [2]

    Metal artifact reduction in CT: where are we after four decades?

    L. Gjesteby, B. De Man, Y . Jin, H. Paganetti, J. Verburg, D. Giantsoudi, and G. Wang, “Metal artifact reduction in CT: where are we after four decades?” IEEE Access, vol. 4, pp. 5826–5849, 2016

    work page 2016

  3. [3]

    Tolerance levels for quality assurance of electron density values generated from CT in radiotherapy treatment planning,

    W. Kilby, J. Sage, and V . Rabett, “Tolerance levels for quality assurance of electron density values generated from CT in radiotherapy treatment planning,” Phys. Med. Biol. , vol. 47, no. 9, p. 1485, 2002

    work page 2002

  4. [4]

    Reduction of CT artifacts caused by metallic implants

    W. A. Kalender, R. Hebel, and J. Ebersberger, “Reduction of CT artifacts caused by metallic implants.” Radiology, vol. 164, no. 2, pp. 576–577, 1987

    work page 1987

  5. [5]

    Nor- malized metal artifact reduction (NMAR) in computed tomography,

    E. Meyer, R. Raupach, M. Lell, B. Schmidt, and M. Kachelrieß, “Nor- malized metal artifact reduction (NMAR) in computed tomography,” Med. Phys., vol. 37, no. 10, pp. 5482–5493, 2010

    work page 2010

  6. [6]

    Deep learning for tomographic image reconstruction,

    G. Wang, J. C. Ye, and B. De Man, “Deep learning for tomographic image reconstruction,” Nat. Mach. Intell. , vol. 2, no. 12, pp. 737–748, 2020

    2020

  7. [7]

    DuDoNet: Dual domain network for CT metal artifact reduction,

    W.-A. Lin, H. Liao, C. Peng, X. Sun, J. Zhang, J. Luo, R. Chellappa, and S. K. Zhou, “DuDoNet: Dual domain network for CT metal artifact reduction,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. , 2019, pp. 10 512–10 521

    work page 2019

  8. [8]

    Dual domain diffusion guidance for 3D CBCT metal artifact reduction,

    Y . Choi, D. Kwon, and S. J. Baek, “Dual domain diffusion guidance for 3D CBCT metal artifact reduction,” in IEEE Winter Conf. Appl. Comput. Vis., 2024, pp. 7965–7974

    work page 2024

Show all 37 references
  1. [9]

    Quad-Net: Quad-domain network for CT metal artifact reduction,

    Z. Li, Q. Gao, Y . Wu, C. Niu, J. Zhang, M. Wang, G. Wang, and H. Shan, “Quad-Net: Quad-domain network for CT metal artifact reduction,” IEEE Trans. Med. Imaging , vol. 43, no. 5, pp. 1866–1879, 2024

    2024

  2. [10]

    ADN: Artifact disentan- glement network for unsupervised metal artifact reduction,

    H. Liao, W.-A. Lin, S. K. Zhou, and J. Luo, “ADN: Artifact disentan- glement network for unsupervised metal artifact reduction,” IEEE Trans. Med. Imaging, vol. 39, no. 3, pp. 634–643, 2019

    2019

  3. [11]

    Dense transformer based enhanced coding network for unsupervised metal artifact reduction,

    W. Xie and M. B. Blaschko, “Dense transformer based enhanced coding network for unsupervised metal artifact reduction,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. , 2023, pp. 77–86

    2023

  4. [12]

    Investigation of domain gap problem in several deep-learning-based CT metal artefact reduction methods,

    M. Du, K. Liang, Y . Liu, and Y . Xing, “Investigation of domain gap problem in several deep-learning-based CT metal artefact reduction methods,” arXiv preprint arXiv:2111.12983 , 2021

    2021 arXiv

  5. [13]

    Unsupervised CT metal artifact learning us- ing attention-guided β-CycleGAN,

    J. Lee, J. Gu, and J. C. Ye, “Unsupervised CT metal artifact learning us- ing attention-guided β-CycleGAN,” IEEE Trans. Med. Imaging, vol. 40, no. 12, pp. 3932–3944, 2021

    2021

  6. [14]

    Building a bridge: Close the domain gap in CT metal artifact reduction,

    T. Wang, H. Yu, Y . Liu, H. Sun, and Y . Zhang, “Building a bridge: Close the domain gap in CT metal artifact reduction,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. , 2023, pp. 206–216

    2023

  7. [15]

    Unsupervised CT metal artifact reduction by plugging diffusion priors in dual domains,

    X. Liu, Y . Xie, S. Diao, S. Tan, and X. Liang, “Unsupervised CT metal artifact reduction by plugging diffusion priors in dual domains,” IEEE Trans. Med. Imaging, 2024

    2024

  8. [16]

    IDOL-Net: An interactive dual-domain parallel network for CT metal artifact reduction,

    T. Wang, Z. Lu, Z. Yang, W. Xia, M. Hou, H. Sun, Y . Liu, H. Chen, J. Zhou, and Y . Zhang, “IDOL-Net: An interactive dual-domain parallel network for CT metal artifact reduction,” IEEE Trans. Radiat. Plasma. Med. Sci., vol. 6, no. 8, pp. 874–885, 2022

    2022

  9. [17]

    Wasserstein generative ad- versarial networks,

    M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein generative ad- versarial networks,” in Proc. Int. Conf. Mach. Learn. , vol. 70, 2017, pp. 214–223

    2017

  10. [18]

    The unusual effectiveness of averaging in GAN training,

    Y . Yazıcı, C.-S. Foo, S. Winkler, K.-H. Yap, G. Piliouras, and V . Chan- drasekhar, “The unusual effectiveness of averaging in GAN training,” in Int. Conf. Learn. Represent. , 2019

    2019

  11. [19]

    U-DuDoNet: Unpaired dual- domain network for CT metal artifact reduction,

    Y . Lyu, J. Fu, C. Peng, and S. K. Zhou, “U-DuDoNet: Unpaired dual- domain network for CT metal artifact reduction,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. , 2021, pp. 296–306

    2021

  12. [20]

    Deep- learning-based metal artefact reduction with unsupervised domain adap- tation regularization for practical CT images,

    M. Du, K. Liang, L. Zhang, H. Gao, Y . Liu, and Y . Xing, “Deep- learning-based metal artefact reduction with unsupervised domain adap- tation regularization for practical CT images,” IEEE Trans. Med. Imag- ing, vol. 42, no. 8, pp. 2133–2145, 2023

    2023

  13. [21]

    SemiMAR: Semi-supervised learning for CT metal artifact reduction,

    T. Wang, H. Yu, Z. Wang, H. Chen, Y . Liu, J. Lu, and Y . Zhang, “SemiMAR: Semi-supervised learning for CT metal artifact reduction,” IEEE J. Biomed. Health Inform. , 2023

    2023

  14. [22]

    Domain-adversarial training of neural networks,

    Y . Ganin, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Lavi- olette, M. March, and V . Lempitsky, “Domain-adversarial training of neural networks,” J. Mach. Learn. Res., vol. 17, no. 59, pp. 1–35, 2016

    2016

  15. [23]

    Pseudo-labeling and confirmation bias in deep semi-supervised learn- ing,

    E. Arazo, D. Ortego, P. Albert, N. E. O’Connor, and K. McGuinness, “Pseudo-labeling and confirmation bias in deep semi-supervised learn- ing,” in Proc. Int. Jt. Conf. Neural Netw. , 2020, pp. 1–8

    2020

  16. [24]

    mixup: Beyond empirical risk minimization,

    H. Zhang, M. Cisse, Y . Dauphin N., and L.-P. David, “mixup: Beyond empirical risk minimization,” in Int. Conf. Learn. Represent. , 2018

    2018

  17. [25]

    Prompted contextual transformer for incomplete-view CT reconstruction,

    C. Ma, Z. Li, J. He, J. Zhang, Y . Zhang, and H. Shan, “Prompted contextual transformer for incomplete-view CT reconstruction,” arXiv preprint arXiv:2312.07846, 2023

    2023 arXiv

  18. [26]

    Swin Transformer: Hierarchical vision transformer using shifted win- dows,

    Z. Liu, Y . Lin, Y . Cao, H. Hu, Y . Wei, Z. Zhang, S. Lin, and B. Guo, “Swin Transformer: Hierarchical vision transformer using shifted win- dows,” in Proc. IEEE/CVF Int. Conf. Comput. Vis. , 2021, pp. 9992– 10 002

    2021

  19. [27]

    Learning to distill global representation for sparse-view CT,

    Z. Li, C. Ma, J. Chen, J. Zhang, and H. Shan, “Learning to distill global representation for sparse-view CT,” in Proc. IEEE/CVF Int. Conf. Comput. Vis., 2023, pp. 21 196–21 207

    2023

  20. [28]

    DeepLesion: automated mining of large-scale lesion annotations and universal lesion detection with deep learning,

    K. Yan, X. Wang, L. Lu, and R. M. Summers, “DeepLesion: automated mining of large-scale lesion annotations and universal lesion detection with deep learning,” J. Med. Imaging , vol. 5, no. 3, p. 036501, 2018

    2018

  21. [29]

    Deep learning to segment pelvic bones: large-scale CT datasets and baseline models,

    P. Liu, H. Han, Y . Du, H. Zhu, Y . Li, F. Gu, H. Xiao, J. Li, C. Zhao, L. Xiao et al., “Deep learning to segment pelvic bones: large-scale CT datasets and baseline models,” Int. J. Comput. Assist. Radiol. Surg. , vol. 16, pp. 749–756, 2021

    2021

  22. [30]

    Convolutional neural network based metal artifact reduction in x-ray computed tomography,

    Y . Zhang and H. Yu, “Convolutional neural network based metal artifact reduction in x-ray computed tomography,” IEEE Trans. Med. imaging , vol. 37, no. 6, pp. 1370–1381, 2018

    2018

  23. [31]

    Unpaired image-to- image translation using cycle-consistent adversarial networks,

    J.-Y . Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to- image translation using cycle-consistent adversarial networks,” in Proc. IEEE/CVF Int. Conf. Comput. Vis. , 2017, pp. 2223–2232

    2017

  24. [32]

    Image quality assessment: from error visibility to structural similarity,

    Z. Wang, A. Bovik, H. Sheikh, and E. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, 2004

    2004

  25. [33]

    IQA- GPT: Computed tomography image quality assessment with vision- language and ChatGPT models,

    Z. Chen, B. Hu, C. Niu, T. Chen, Y . Li, H. Shan, and G. Wang, “IQA- GPT: Computed tomography image quality assessment with vision- language and ChatGPT models,” Vis. Comput. Ind. Biomed. Art , vol. 7, no. 20, p. 036501, 2024

    2024

  26. [34]

    Rich human feedback for text-to-image generation,

    Y . Liang, J. He, G. Li, P. Li, A. Klimovskiy, N. Carolan, J. Sun, J. Pont- Tuset, S. Young, F. Yang, J. Ke, K. D. Dvijotham, K. M. Collins, Y . Luo, Y . Li, K. J. Kohlhoff, D. Ramachandran, and V . Navalpakkam, “Rich human feedback for text-to-image generation,” in Proc. IEEE...

    2024

  27. [35]

    Very deep convolutional networks for large-scale image recognition,

    K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” in Int. Conf. Learn. Represent. , 2015

    2015

  28. [36]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark et al., “Learning transferable visual models from natural language supervision,” in Int. Conf. Mach. Learn., 2021, pp. 8748–8763

    2021

  29. [37]

    Perceptual losses for real-time style transfer and super-resolution,

    J. Johnson, A. Alahi, and L. Fei-Fei, “Perceptual losses for real-time style transfer and super-resolution,” in Proc. Eur. Conf. Comput. Vis. , 2016, pp. 694–711

    2016

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.