Pith. sign in

REVIEW 2 major objections 4 minor 49 references

Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Mixup matches adversarial domain adaptation for knee MRI segmentation robustness

desk verdict A careful empirical comparison showing mixup and UDA improve cross-scanner knee cartilage segmentation, with mixup the simpler winner; the main caveat is annotation quality on the independent test set. read the letter →

arxiv 1908.04126 v3 pith:6PT6IC6U submitted 2019-08-12 eess.IV cs.CV

Egor Panfilov , Aleksei Tiulpin , Stefan Klein , Miika T. Nieminen , Simo Saarakkala This is my paper · ORCID
classification eess.IVcs.CV
keywords mixupunsuperviseddomainadaptationkneeMRIsegmentationcartilagerobustnessdeeplearningU-Netshift
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 asks whether deep-learning models that segment knee cartilage from MRI can be made more robust to new acquisition settings without expensive retraining. The authors compare two regularization techniques—mixup, a simple data-augmentation method, and adversarial unsupervised domain adaptation (UDA), which aligns feature distributions across labeled and unlabeled datasets—against a strong U-Net baseline. They find that for moderate changes in MRI protocols, both methods improve segmentation accuracy on an independent dataset, with mixup achieving gains comparable to UDA while being simpler and cheaper. The paper argues that mixup should be the preferred approach for improving robustness in medical image segmentation, since it requires no extra unlabeled data from the target domain and is less computationally demanding.

What carries the argument

The central objects are the U-Net baseline architecture, mixup augmentation, and an adversarial domain adaptation module. Mixup creates virtual training examples by interpolating between two images and their one-hot segmentation masks using a Beta-distributed parameter λ, forcing the model to learn smoother decision boundaries. The UDA approach uses a discriminator network that attempts to distinguish feature maps (from the output or penultimate encoder block) of the labeled source domain from those of the unlabeled target domain; the segmentation network is trained adversarially to fool the discriminator, aligning the representation spaces across domains. The paper evaluates two UDA variants: UDA1, which aligns only the output space, and UDA2, which additionally aligns an intermediate decoder feature via an ASPP module and a second discriminator.

What would settle it

To test whether the mixup gain is genuine acquisition robustness rather than annotation-bias alignment, one could re-annotate a subset of Dataset C with the same high-resolution protocol and expert panel used for Dataset A, then compute Dice scores on those clean labels. If the mixup model's advantage shrank or vanished on the refined annotations, the robustness claim would be weakened. Alternatively, one could test the model on a truly independent dataset from a different scanner and protocol with high-quality annotations; if the gain disappears, the 'moderate change' scope would be shown to be narrower than claimed.

Watch

Extended reading notes

Core claim

The central claim is that mixup, a regularization technique that trains on convex combinations of image pairs and their labels, can improve the robustness of a U-Net-based knee cartilage and meniscus segmentation model to new MRI acquisition settings. On an independent test set (Dataset C, acquired with the same scanner as the unlabeled training domain but in a separate study), the baseline model achieves femoral cartilage Dice scores of 0.791, while the best mixup setup (no weight decay) reaches 0.819, and the best UDA variant (UDA2) reaches 0.821. For tibial cartilage, mixup scores 0.802 and UDA1 scores 0.814, both below the UDA1's tibial score of 0.814 but above the baseline's 0.746. The paper also reports that mixup preserves performance on the source domain (Dataset A) better than UDA, which tends to degrade source-domain accuracy. The authors conclude that for moderate acquisition differences, both techniques improve generalization, but mixup is the simpler, less costly, and equally effective choice.

Load-bearing premise

The claim that mixup and UDA improve robustness to new MRI settings hinges on the assumption that Dataset C, though annotated by the same group and acquired with the same scanner as the unlabeled training data, is a representative 'unseen target' whose annotations are reliable enough to serve as ground truth; the paper itself notes Dataset C has lower resolution and annotation inaccuracies, so part of the measured gain could reflect smoother masks agreeing better with imperfect references rather than true acquisition invariance.

Editorial extensions

If this is right

  • If mixup's gains hold across other medical imaging tasks, it offers a cheap, annotation-free robustness boost that can be applied to any existing segmentation pipeline.
  • The finding that mixup with no weight decay outperforms mixup with weight decay suggests that over-regularization can be counterproductive, pointing to a practical hyperparameter choice.
  • The consistent improvements across KL grades (osteoarthritis severity stages) indicate that the robustness gains are not confined to a particular disease severity, but benefit the full spectrum.
  • The slice-wise analysis shows that improvements are concentrated in weight-bearing areas of the knee, which are the most clinically relevant, suggesting that the method's benefits align with clinical priorities.

Reading between the lines

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

  • The paper's comparison of mixup and UDA is only at one moderate domain shift; it is plausible that for larger scanner/protocol differences, UDA's explicit domain alignment would provide a larger advantage than mixup, which only relies on generic input smoothing.
  • The reported mixup gain might be partly due to it acting as a shape regularizer that produces smoother, more conservative segmentations that incidentally agree better with the lower-quality lower-resolution annotations in Dataset C; a controlled study with high-quality reference labels would separate true acquisition invariance from annotation-bias compensation.
  • The slice-wise error maps suggest that a targeted loss weighting of the weight-bearing regions could amplify the robustness benefits further, a testable extension the authors did not explore.
  • The authors' recommendation to prefer mixup over UDA assumes that unlabeled target-domain data is not available; when such data is plentiful and the domain shift is severe, UDA may still be the better investment, so the practical choice should be data-dependent.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. This manuscript addresses the robustness of deep-learning-based segmentation of knee cartilage and menisci in MRI across acquisition settings. The authors propose a 2D U-Net baseline and evaluate two regularization strategies: input mixup and adversarial unsupervised domain adaptation (UDA) based on structured-output alignment. Their validation setup uses three datasets: OAI-derived Dataset A (annotated, source), Dataset B (unannotated, same scanner/protocol as the target), and independent Dataset C (annotated, target). Models are trained with 5-fold cross-validation, and performance is measured by volumetric Dice on the Dataset A test subset and on Dataset C, with KL-grade stratified analysis and slice-wise anatomical localization. The main findings are that both mixup (especially without weight decay) and UDA improve Dice on Dataset C relative to the baseline, with UDA2 slightly better for femoral cartilage and mixup without weight decay better for tibial cartilage, while mixup is simpler and cheaper. The paper concludes that mixup should be preferred for moderate acquisition shifts and releases code and pre-trained models.

Significance. The study is valuable for the osteoarthritis imaging community: it is, to my knowledge, the first end-to-end comparison of mixup and adversarial UDA for knee MRI segmentation, uses a genuinely independent test set (Dataset C), cross-validates the source data, reports tissue- and location-specific results, and makes code and models publicly available. The finding that a simple regularization technique (mixup) can match a substantially more complex UDA pipeline is practically useful. However, the robustness claim rests on Dataset C as ground truth, and the manuscript itself notes that Dataset C has lower image and annotation resolution and reference inaccuracies. Because the reported gains are a few Dice points, the risk that part of the improvement reflects smoother predictions agreeing better with coarse labels, rather than acquisition invariance, is a real threat to the headline conclusion. The evaluation design otherwise supports the core observation that both methods beat the baseline on Dataset C.

major comments (2)
  1. [Section 4, Tables 2 and 3; Figure 5] The central claim that mixup and UDA improve robustness to MRI acquisition changes is entangled with a label-quality confound. The authors themselves attribute the baseline drop on Dataset C partly to 'lower original resolution of images and annotations' and Figure 5 highlights 'inaccuracies of the reference segmentations.' The observed gains are modest (FC 0.791 to 0.819 and TC 0.746 to 0.802 for mixup-WD; FC 0.821 and TC 0.799 for UDA2), so even a small bias in the reference standard could account for a substantial share of the improvement. Because regularized models tend to produce smoother, less over-segmented masks, better Dice on coarse labels is not by itself evidence of acquisition invariance. I ask for an additional analysis: a subset of Dataset C with refined or consensus annotations, boundary-based metrics such as surface Dice or Hausdorff distance together with volume-bias estimates, or a control experiment on downsampled or noisy Dataset A images with intact high-quality labels. Until such an analysis is provided, the conclusion should be framed as robustness to annotation style and resolution as well as to acquisition settings.
  2. [Section 3.6 and Section 4 (mixup - WD)] The paper's preferred variant, mixup without weight decay, appears to have been introduced after observing that mixup with weight decay underfits on Dataset A ('we hypothesized that avoiding the use of weight decay could address the underfitting'). The manuscript does not describe a pre-specified model-selection protocol or a separate validation set for choosing this variant. If the decision to remove weight decay was informed by performance on the Dataset A test subset or on Dataset C, the reported improvement is optimistically biased by test-set selection. Please clarify the selection procedure and, ideally, report the performance of the mixup variant with and without weight decay on a held-out validation fold that was not used for any design decision. The same transparency is needed for the choice between UDA1 and UDA2 and for the auxiliary loss weights.
minor comments (4)
  1. [Section 3.3, Eq. (5)] The manuscript does not state whether the mixup coefficient lambda is resampled per minibatch or per sample, and whether the same lambda is used to weight the two loss terms. Please clarify this for reproducibility.
  2. [Table 3 and Figure 4] The statistical testing section is underspecified. Please state whether the Wilcoxon signed-rank test was applied to the full test set per tissue and/or to each KL subgroup, report exact p-values for the headline comparisons, and note that subgroup sample sizes (11 to 16 scans) limit statistical power. Also report the number of bootstrap resamples used for the 95% confidence intervals in Figure 4.
  3. [Section 3.5 and Table 2] Dataset C contains annotations only for femoral and tibial cartilage, so the target-domain results do not cover patellar cartilage or menisci. The abstract and conclusions should scope the robustness claims to the tissues actually evaluated on Dataset C, and the phrase 'clinically important areas of the knee joint' should distinguish tissues evaluated on the target from those evaluated only on Dataset A.
  4. [Table 1] The comparison to previously published methods is difficult to interpret because different methods use different splits, some use 3D or multi-stage pipelines, and the table mixes medial/lateral sub-tissues with aggregate scores. The caveat in the text is appropriate; a short statement that this is an indicative rather than head-to-head comparison would help readers avoid overinterpreting the row.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the robustness claim is an empirical, held-out evaluation with fixed hyperparameters, not a derivation from fitted inputs.

full rationale

This paper makes no parametric derivation whose conclusion reduces to its inputs. The central claim is that mixup and adversarial UDA improve robustness of knee cartilage segmentation under moderate acquisition changes, and this is supported by measured Dice scores on Dataset C, which was held out entirely from training and validation. Hyperparameters (mixup alpha = 0.7, weight decay 5e-5, adversarial loss weights from Tsai et al.) are fixed before evaluation and are not fitted to Dataset C; the reported Dice values are measured outcomes, not consequences of the hyperparameter values by construction. The only overlapping-author citation is [37] for the U-Net filter configuration, a minor architectural precedent that is not load-bearing for the robustness conclusion. The paper's own admission that Dataset C has lower original resolution and imperfect reference annotations is a validity threat to interpreting the gains as acquisition robustness rather than label-quality effects, but that is a correctness/confounding concern, not circularity: it does not make the result definitionally equivalent to an input. The evaluation is self-contained against external benchmarks (Table 1) and the code is released, so there is no fitted parameter renamed as a prediction and no self-citation chain forcing the result.

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

The central claim rests on standard deep-learning practice plus three domain assumptions: the reference annotations are trustworthy enough for Dice evaluation, Dataset C represents the target acquisition domain for UDA, and the chosen test statistics are appropriate. The no-invented-entity ledger is clean; the main burden is the private-dataset dependence.

free parameters (4)
  • mixup alpha = 0.7
    Augmentation strength in Eq. 1; set within the typical recommended range without sensitivity analysis.
  • weight decay constant = 5e-5 (removed in recommended mixup - WD setting)
    Initially used for all settings; dropped for mixup after observing underfitting, making the recommended configuration partially dependent on test performance.
  • UDA loss weights (gamma_segm, gamma_adv) = 1 and 1e-3 (UDA1); auxiliary 1e-1 and 2e-4 (UDA2)
    Transferred from the original UDA paper [41]; set manually to prioritize the segmentation loss.
  • Network architecture choices (initial filters, depth, upsampling) = 24 filters, depth 6, bilinear upsampling
    Selected by 'extensive experimental search' on validation data (Section 3.2), not derived from first principles.
assumptions (4)
  • domain assumption Manual segmentations are usable as ground truth for evaluation.
    The paper notes annotation inaccuracies and lower resolution in Dataset C (Section 4, Figure 5), yet treats them as reference for Dice computation.
  • domain assumption Dataset C is a valid proxy for the target acquisition domain of Dataset B because both use the same scanner and protocol.
    This underpins the UDA scenario where unlabeled B is aligned to A; if B and C differ substantially from the intended test distribution, the UDA result is not generalizable.
  • standard math Existing deep learning training assumptions (U-Net architecture, Adam, cross-entropy loss) are appropriate.
    Standard practice; not the focus of this paper.
  • standard math The Wilcoxon signed-rank test assumes paired, i.i.d. scan-level observations across methods.
    Used for significance claims in Table 3; multiple comparisons were not adjusted.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation." pith.science (2026). https://pith.science/paper/6PT6IC6U

@misc{pith2026190804126,
  author       = {Pith},
  title        = {Pith review of: Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6PT6IC6U}},
  note         = {Machine review of arXiv:1908.04126}
}
read the original abstract

Degeneration of articular cartilage (AC) is actively studied in knee osteoarthritis (OA) research via magnetic resonance imaging (MRI). Segmentation of AC tissues from MRI data is an essential step in quantification of their damage. Deep learning (DL) based methods have shown potential in this realm and are the current state-of-the-art, however, their robustness to heterogeneity of MRI acquisition settings remains an open problem. In this study, we investigated two modern regularization techniques -- mixup and adversarial unsupervised domain adaptation (UDA) -- to improve the robustness of DL-based knee cartilage segmentation to new MRI acquisition settings. Our validation setup included two datasets produced by different MRI scanners and using distinct data acquisition protocols. We assessed the robustness of automatic segmentation by comparing mixup and UDA approaches to a strong baseline method at different OA severity stages and, additionally, in relation to anatomical locations. Our results showed that for moderate changes in knee MRI data acquisition settings both approaches may provide notable improvements in the robustness, which are consistent for all stages of the disease and affect the clinically important areas of the knee joint. However, mixup may be considered as a recommended approach, since it is more computationally efficient and does not require additional data from the target acquisition setup.

Figures

Figures reproduced from arXiv: 1908.04126 by the authors.

Figure 2
Figure 2. Schematic view of our approaches – without [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. DESS MRI scan (a) and the annotations of knee carti￾lage and meniscal tissues (b), both rescaled to isotropic resolution. White lines in (a) indicate the orientation of sagittal slices. different scanner and a distinct imaging protocol (see exam￾ples in [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Distributions of the planar DSCs computed slice-wise [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Example images of tibiofemoral contact zones from Datasets A and C, respective annotations, and the segmentation masks [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 42 canonical work pages

  1. [1]

    O. M. Abdulaal et al. 3T MRI of the knee with opti- mised isotropic 3d sequences: Accurate delineation of intra- articular pathology without prolonged acquisition times. Eu- ropean radiology, 27(11):4563–4570, 2017. 1

    work page 2017

  2. [2]

    Altahawi and N

    F. Altahawi and N. Subhas. 3d MRI in musculoskeletal imag- ing: Current and future applications. Current Radiology Re- ports, 6(8):27, 2018. 1

    work page 2018

  3. [3]

    Ambellan, A

    F. Ambellan, A. Tack, M. Ehlke, and S. Zachow. Automated segmentation of knee bone and cartilage combining statis- tical shape knowledge and convolutional neural networks: Data from the osteoarthritis initiative. Medical Image Anal- ysis, 52(2):109–118, 2019. 1, 2, 5, 6, 7

    work page 2019

  4. [4]

    Chaitanya, N

    K. Chaitanya, N. Karani, C. Baumgartner, and E. Konukoglu. Semi-supervised and task-driven data augmentation. arXiv e-prints, page arXiv:1902.05396, Feb 2019. 2

    arXiv 1902

  5. [5]

    Chaitanya, N

    K. Chaitanya, N. Karani, C. F. Baumgartner, A. Becker, O. Donati, and E. Konukoglu. Semi-supervised and task- driven data augmentation. In International Conference on Information Processing in Medical Imaging , pages 29–41. Springer, 2019. 2

    work page 2019

  6. [6]

    A. S. Chaudhari et al. Utility of deep learning super- resolution in the context of osteoarthritis MRI biomarkers. Journal of Magnetic Resonance Imaging, 2019. 1, 6, 7

    work page 2019

  7. [7]

    C. Chen, Q. Dou, H. Chen, and P.-A. Heng. Semantic-aware generative adversarial nets for unsupervised domain adapta- tion in chest x-ray segmentation. In International Workshop on Machine Learning in Medical Imaging , pages 143–151. Springer, 2018. 2

    work page 2018

  8. [8]

    Chen et al

    L.-C. Chen et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE transactions on pattern analysis and machine intelligence, 40(4):834–848, 2017. 4

    work page 2017

Show all 49 references
  1. [9]

    Chopra, S

    S. Chopra, S. Balakrishnan, and R. Gopalan. Dlid: Deep learning for domain adaptation by interpolating between do- mains. In ICML workshop on challenges in representation learning, 2013. 2

    2013

  2. [10]

    E. B. Dam, M. Lillholm, J. Marques, and M. Nielsen. Au- tomatic segmentation of high- and low-field knee MRIs us- ing knee image quantification with data from the osteoarthri- tis initiative. Journal of Medical Imaging, 2(2):024001, apr

  3. [11]

    Daume III and D

    H. Daume III and D. Marcu. Domain adaptation for statis- tical classifiers. Journal of artificial Intelligence research , 26:101–126, 2006. 2

    2006

  4. [12]

    A. D. Desai, G. E. Gold, B. A. Hargreaves, and A. S. Chaud- hari. Technical considerations for semantic segmentation in MRI using convolutional neural networks. arXiv preprint arXiv:1902.01977, 2019. 1, 2, 6

    1902 arXiv

  5. [13]

    Dou et al

    Q. Dou et al. Unsupervised cross-modality domain adap- tation of convnets for biomedical image segmentations with adversarial loss. arXiv preprint arXiv:1804.10916, 2018. 2

    2018 arXiv

  6. [14]

    Eaton-Rosen, F

    Z. Eaton-Rosen, F. Bragman, S. Ourselin, and M. J. Cardoso. Improving data augmentation for medical image segmenta- tion. MIDL 2018 submission at https://openreview. net/forum?id=rkBBChjiG, 2018. 2

    2018

  7. [15]

    C. A. Emery, J. L. Whittaker, A. Mahmoudian, L. S. Lohmander, E. M. Roos, K. L. Bennell, C. M. Toomey, R. A. Reimer, D. Thompson, J. L. Ronsky, et al. Establishing out- come measures in early knee osteoarthritis. Nature Reviews Rheumatology, page 1, 2019. 1

    2019

  8. [16]

    Esteva, B

    A. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. Thrun. Dermatologist-level classifi- cation of skin cancer with deep neural networks. Nature, 542(7639):115, 2017. 1

    2017

  9. [17]

    Esteva, A

    A. Esteva, A. Robicquet, B. Ramsundar, V . Kuleshov, M. De- Pristo, K. Chou, C. Cui, G. Corrado, S. Thrun, and J. Dean. A guide to deep learning in healthcare. Nature medicine, 25(1):24, 2019. 1

    2019

  10. [18]

    Ganin et al

    Y . Ganin et al. Domain-adversarial training of neural networks. The Journal of Machine Learning Research , 17(1):2096–2030, 2016. 2

    2016

  11. [19]

    H. Guo, Y . Mao, and R. Zhang. Mixup as locally linear out- of-manifold regularization. In Proceedings of the AAAI Con- ference on Artificial Intelligence , volume 33, pages 3714– 3722, 2019. 2

    2019

  12. [20]

    Karani, K

    N. Karani, K. Chaitanya, C. Baumgartner, and E. Konukoglu. A lifelong learning approach to brain MR segmentation across scanners and protocols. In International Conference on Medical Image Computing and Computer-Assisted Inter- vention, pages 476–484. Springer, 2018. 1, 2

    2018

  13. [21]

    Kellgren and J

    J. Kellgren and J. Lawrence. Radiological assessment of osteo-arthrosis. Annals of the rheumatic diseases, 16(4):494,

  14. [22]

    D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014. 5

    2014 arXiv

  15. [23]

    Kuka ˇcka, V

    J. Kuka ˇcka, V . Golkov, and D. Cremers. Regulariza- tion for deep learning: A taxonomy. arXiv preprint arXiv:1710.10686, 2017. 2

    2017 arXiv

  16. [24]

    Kumar, A

    D. Kumar, A. Gandhamal, S. Talbar, and A. F. M. Hani. Knee articular cartilage segmentation from MR images: A review. ACM Computing Surveys (CSUR), 51(5):97, 2018. 2

    2018

  17. [25]

    F. Liu. Susan: segment unannotated image structure us- ing adversarial network. Magnetic resonance in medicine , 81(5):3330–3345, 2019. 2

    2019

  18. [26]

    Liu et al

    F. Liu et al. Deep convolutional neural network and 3d de- formable approach for tissue segmentation in musculoskele- tal magnetic resonance imaging. Magnetic Resonance in Medicine, 79(4):2379–2391, 2018. 2

    2018

  19. [27]

    M. T. Nieminen, V . Casula, M. T. Nevalainen, and S. S. Saarakkala. Osteoarthritis year in review 2018: imaging.Os- teoarthritis and cartilage, 2018. 1

    2018

  20. [28]

    Norman, V

    B. Norman, V . Pedoia, and S. Majumdar. Use of 2d u-net convolutional neural networks for automated cartilage and meniscus segmentation of knee mr imaging data to determine relaxometry and morphometry. Radiology, 288(1):177–185,

  21. [29]

    Paszke et al

    A. Paszke et al. Automatic differentiation in pytorch. In NIPS-W, 2017. 5

    2017

  22. [30]

    Peterfy, E

    C. Peterfy, E. Schneider, and M. Nevitt. The osteoarthritis initiative: report on the design rationale for the magnetic res- onance imaging protocol for the knee. Osteoarthritis and cartilage, 16(12):1433–1441, 2008. 2

    2008

  23. [31]

    Podlipsk ´a et al

    J. Podlipsk ´a et al. Comparison of diagnostic performance of semi-quantitative knee ultrasound and knee radiography with MRI: Oulu knee osteoarthritis study. Scientific reports, 6:22365, 2016. 4

    2016

  24. [32]

    Ronneberger, P

    O. Ronneberger, P. Fischer, and T. Brox. U-net: Convo- lutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, pages 234–241, 2015. 3

    2015

  25. [33]

    M. D. Ryzhkov. Knee cartilage segmentation algorithms: a critical literature review. Master’s thesis, Utrecht University,

  26. [34]

    Stammberger et al

    T. Stammberger et al. Interobserver reproducibility of quantitative cartilage measurements: comparison of b-spline snakes and manual segmentation. Magnetic resonance imag- ing, 17(7):1033–1042, 1999. 1

    1999

  27. [35]

    A. Tack, A. Mukhopadhyay, and S. Zachow. Knee menisci segmentation using convolutional neural networks: Data from the osteoarthritis initiative. Osteoarthritis and Carti- lage, 26(5):680 – 688, 2018. 1, 5, 6, 7

    2018

  28. [36]

    Tack and S

    A. Tack and S. Zachow. Accurate automated volumetry of cartilage of the knee using convolutional neural networks: Data from the osteoarthritis initiative. In IEEE 16th Interna- tional Symposium on Biomedical Imaging (ISBI 2019), pages 40 – 43, 2019. 1, 5, 6, 7

    2019

  29. [37]

    Tiulpin et al

    A. Tiulpin et al. Deep-learning for tidemark segmentation in human osteochondral tissues imaged with micro-computed tomography. arXiv preprint arXiv:1907.05089, 2019. 3

    1907 arXiv

  30. [38]

    Tiulpin, S

    A. Tiulpin, S. Klein, S. Bierma-Zeinstra, J. Thevenot, E. Rahtu, J. van Meurs, E. H. Oei, and S. Saarakkala. Multi- modal machine learning-based knee osteoarthritis progres- sion prediction from plain radiographs and clinical data. arXiv preprint arXiv:1904.06236, 2019. 1

    1904 arXiv

  31. [39]

    Tiulpin and S

    A. Tiulpin and S. Saarakkala. Automatic grading of indi- vidual knee osteoarthritis features in plain radiographs us- ing deep convolutional neural networks. arXiv preprint arXiv:1907.08020, 2019. 1

    1907 arXiv

  32. [40]

    Tiulpin, J

    A. Tiulpin, J. Thevenot, E. Rahtu, P. Lehenkari, and S. Saarakkala. Automatic knee osteoarthritis diagnosis from plain radiographs: A deep learning-based approach. Scien- tific reports, 8(1):1727, 2018. 1

    2018

  33. [41]

    Tsai et al

    Y .-H. Tsai et al. Learning to adapt structured output space for semantic segmentation. In Proceedings of the IEEE Con- ference on Computer Vision and Pattern Recognition, pages 7472–7481, 2018. 3, 4

    2018

  34. [42]

    Tzeng, J

    E. Tzeng, J. Hoffman, N. Zhang, K. Saenko, and T. Darrell. Deep domain confusion: Maximizing for domain invariance. arXiv preprint arXiv:1412.3474, 2014. 2

    2014 arXiv

  35. [43]

    W. E. Van Spil et al. Osteoarthritis phenotypes and novel therapeutic targets. Biochemical pharmacology, 2019. 1

    2019

  36. [44]

    Verma et al

    V . Verma et al. Manifold mixup: Better represen- tations by interpolating hidden states. arXiv preprint arXiv:1806.05236, 2018. 5

    2018 arXiv

  37. [45]

    J. Wang, M. Knol, A. Tiulpin, F. Dubost, M. De Bruijne, M. Vernooij, H. Adams, M. A. Ikram, W. Niessen, and G. Roshchupkin. Grey matter age prediction as a biomarker for risk of dementia: A population-based study. BioRxiv, page 518506, 2019. 1

    2019

  38. [46]

    Xu and M

    Z. Xu and M. Niethammer. Deepatlas: Joint semi-supervised learning of image registration and segmentation. arXiv preprint arXiv:1904.08465, 2019. 2

    1904 arXiv

  39. [47]

    Yaguchi, F

    Y . Yaguchi, F. Shiratani, and H. Iwaki. Mixfeat: Mix feature in latent space learns discriminative space. ICLR 2019 sub- mission at https://openreview.net/forum?id= HygT9oRqFX, 2019. 2

    2019

  40. [48]

    Zhang, M

    H. Zhang, M. Cisse, Y . N. Dauphin, and D. Lopez-Paz. mixup: Beyond empirical risk minimization. International Conference on Learning Representations, 2018. 2

    2018

  41. [2015]

    https://dspace.library.uu.nl/handle/ 1874/308831. 2

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.