Pith. sign in

REVIEW 4 major objections 6 minor 39 references

Intensity augmentation for domain transfer of whole breast segmentation in MRI

T0 review · 4 major / 6 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 A 3D U-net trained on T1-weighted MRI, augmented by random style transfer or intensity remapping, segments breasts in T2-weighted and unseen-scanner MRI almost as well as a model trained on those domains themselves.

desk verdict Useful augmentation trick for breast MRI segmentation, but the main result is confounded by training effort and the independent set does not confirm the benefit. read the letter →

arxiv 1909.02642 v1 pith:4FR6SITE submitted 2019-09-05 eess.IV cs.CVcs.LGstat.ML

Linde S. Hesse , Grey Kuling , Mitko Veta , Anne L. Martel This is my paper · ORCID
classification eess.IVcs.CVcs.LGstat.ML
keywords breastMRIsegmentationdomainshiftdataaugmentationintensitystyletransferremapping3DU-NetDicesimilaritycoefficient
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 tries to establish that intensity-domain shift in MRI segmentation can be handled at training time, without any target-domain data, by heavily distorting image intensities while keeping shape intact. The authors train a 3D U-net for whole breast segmentation on T1-weighted scans and test it on T2-weighted scans from the same dataset and on an independent T1-weighted dataset from a different scanner and coil configuration. Adding either of two intensity augmentation strategies, neural style transfer or a new random intensity remapping, to geometric augmentation raises the Dice similarity coefficient from 0.71 to 0.90 on the T1-to-T2 transfer, close to the 0.92 same-domain baseline. On the independent public test set, geometric plus style augmentation reaches 0.89 after post-processing to remove unconnected thorax regions, close to the 0.92 inter-observer variability. If these numbers hold, a single model could be deployed across MRI protocols without retraining or collecting data from the target domain.

What carries the argument

The central mechanism is intensity augmentation at training time: two ways to randomly repaint a training volume while preserving its shape. Style transfer uses a pre-trained arbitrary style transfer network and replaces the style of each slice with a randomly sampled style embedding, combined with the image's own style at strength $\alpha=0.5$. Intensity remapping builds a per-volume curve by smoothing random noise with a moving average filter and adding a linear component, then maps every pixel through that curve. Both force the U-net to rely on breast shape rather than on scanner-specific intensity patterns, which is what lets the trained network carry over to T2-weighted and TWIST scans.

What would settle it

Re-score the intensity-augmented models on the independent test set without the connected-component post-processing that removes unconnected thorax regions; if the Dice falls below the geometric-augmentation-only baseline, then the claimed improvement on an 'unseen' dataset is partly an artifact of dataset-specific cleanup. A complementary check is to fix the segmentation protocol in advance and apply the same augmentation to a truly held-out site's T1-weighted scans, reporting Dice before any post-processing.

Watch

Extended reading notes

Core claim

Training a 3D U-net for whole breast segmentation on T1-weighted MRI, with geometric augmentation plus intensity augmentation, produces a model that segments T2-weighted scans from the same dataset and T1-weighted scans from an unseen scanner and coil almost as well as a model trained directly on those domains. With geometric augmentation alone the cross-domain Dice is 0.71; adding either style transfer or intensity remapping raises it to 0.90, versus a 0.92 same-domain baseline. On the independent public test set, geometric plus style augmentation achieves an average Dice of 0.89 after connected-component post-processing to remove unconnected thorax false positives, compared with inter-observer agreement of 0.92. Performance on the original T1 domain stays at 0.94, so the augmentation does not degrade performance on the source domain.

Load-bearing premise

The load-bearing premise is that the ground-truth segmentations and the added post-processing step for the independent test set measure breast-segmentation quality fairly across methods; if the post-processing favors the augmented models, the reported 0.89 Dice is not an unbiased estimate of cross-domain performance.

Editorial extensions

If this is right

  • A T1-trained segmentation model can be used on T2-weighted breast MRI with no target-domain data, removing the need for per-site retraining or domain-adaptation networks.
  • Performance on the source T1 domain stays at baseline (DSC 0.94), so the intensity augmentation adds cross-domain robustness without sacrificing accuracy on the training distribution.
  • The two augmentation variants differ practically: intensity remapping is fully 3D and takes about 0.6 seconds per scan, while style transfer is slice-wise and requires a GPU, yet both reach similar Dice on the T2 test set.
  • Because the augmentation is domain-agnostic, the same trained network transfers to an unseen dataset from a different scanner and coil configuration, scoring close to inter-observer variability.
  • The authors expect the same augmentation principle to transfer to other segmentation tasks where intensity distributions vary across imaging sites.

Reading between the lines

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

  • If the effect holds generally, random intensity remapping could become a standard cheap regularizer applied alongside geometric augmentation in any MRI segmentation pipeline, since it needs no pretrained network or target data.
  • A natural extension is to test the same shape-preserving intensity distortion on other anatomies or modality pairs (e.g., CT to MRI, or T1 to post-contrast T1), where the assumption that shape carries the segmentation remains plausible.
  • The reported score on the independent test set depends on a dataset-specific post-processing step that removes unconnected thorax segments; a stricter evaluation would apply a pre-registered protocol to a new dataset with no such cleanup. This caveat is my inference, not a claim made in the paper.
  • Since style transfer and intensity remapping performed equivalently on the T1-to-T2 task, the simpler method is probably the better deployment choice unless a specific failure mode argues otherwise.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper addresses intensity domain shift in whole breast MRI segmentation. The authors train a 3D U-Net on T1-weighted Sunnybrook scans and test on T2-weighted Sunnybrook scans, applying two intensity augmentation strategies—style transfer and a novel intensity remapping—in addition to geometric augmentation. They report that intensity augmentation raises Dice similarity coefficient (DSC) from 0.71 to 0.90 on the T2 test set, close to the T2-trained baseline of 0.92, and that the best model reaches 0.89 on a publicly available QIN-Breast T1-weighted TWIST dataset, close to inter-observer variability of 0.92. They also perform a Friedman test with Dunn's multiple comparison correction. The paper claims that intensity augmentation is a simple and effective method for overcoming intensity domain shift without needing target-domain data.

Significance. If the central comparison were rigorously controlled, the result would be practically valuable: intensity augmentation is easy to implement, requires no target-domain data, and is substantially simpler than GAN-based or domain-adversarial adaptation. The paper also provides a useful comparison of two intensity augmentation variants and makes its evaluation on a public dataset possible by providing details of its own manual ground truth generation. The main strengths are the clear experimental setup for the T1-to-T2 transfer, the explicit description of the proposed intensity remapping, and the inclusion of statistical testing. However, the central evidence is weakened by a training-effort confound and by the QIN-Breast evaluation being less independent than claimed. If the authors can add the missing controlled comparisons and reconcile the reported baselines and evaluation protocol, the contribution would be credible; in its current form the central claim is not yet established.

major comments (4)
  1. [Section III-B-d and Section IV] The headline T1-to-T2 comparison in Table IV is confounded by training effort. Section III-B-d states that the intensity-augmentation experiments pre-generate two style-transferred and two intensity-remapped volumes per scan and train with a 1:2 ratio of original to augmented volumes, whereas geometric augmentation is applied online once per original volume. Since Section IV fixes all networks at 200 epochs, an epoch in the intensity-augmentation experiments passes over roughly three times as many samples as an epoch in the geometric-only condition, and therefore performs roughly three times as many SGD updates. The observed 0.19 DSC increase could be caused by additional training data or optimization steps rather than by intensity invariance. The authors should add a matched-iteration baseline (e.g., train the geometric-only network for proportionally more epochs) or a control that adds the same amount of extra data with non-intensity augmentation, to isolate the effect of intensity augmentation.
  2. [Section III-E, Section III-F, and Appendix A] The QIN-Breast evaluation does not provide independent validation of the intensity-augmentation claim. The ground truth was created in-house by the authors with boundary rules different from the Sunnybrook protocol (Section III-E), and the post-processing in Section III-F was implemented after observing thorax false positives on this dataset. After this post-processing, Appendix A Table I(b) shows that none of the intensity-augmentation strategies is statistically significantly different from geometric augmentation alone. Thus the reported 0.89 DSC cannot distinguish intensity augmentation from geometric augmentation. At minimum, the post-processing rule should be pre-specified and justified as part of the method before evaluation, and the QIN-Breast results should be reported with the raw and post-processed statistics together so that the effect of post hoc selection is transparent.
  3. [Abstract and Section V] There is an internal inconsistency in the reported baseline gain. The abstract states the improvement is from a DSC of 0.71 to 0.90, while Section V states 'from a DSC of 0.62 to 0.90 by applying intensity remapping and geometric augmentation.' Table IV lists the no-augmentation mean as 0.63 and the geometric-only mean as 0.71. The authors should state clearly which baseline is being used and correct the discrepancy; the 0.62 value is not present in Table IV.
  4. [Section IV] The evaluation protocol selects the single best of five networks per augmentation strategy based on the T2 validation set, then reports the test DSCs and significance tests for that selected network. This ignores between-run variability, and because the validation set is small (8 volumes) and comes from the same domain as the test set, the best-of-five selection can inflate the reported T2 test results. The authors should report results for all five runs, or use a selection procedure whose variance is accounted for (e.g., reporting mean performance over runs or performing the statistical test across all runs).
minor comments (6)
  1. [Section I] The sentence 'The first step in the analysis of breast MR images is to obtain a a breast segmentation' contains a duplicated article 'a a'; it should be 'obtain a breast segmentation.'
  2. [Section III-A] The abbreviation 'ReLu' should be 'ReLU' for consistency with standard usage.
  3. [Section III-E] The phrase 'every 3 th-5th slice' is unclear; it should be rewritten as 'every third to fifth slice.'
  4. [Section III-F] The section heading appears as 'F . QIN-Breast Post-processing' with an extra space before the period; this is a formatting artifact that should be corrected.
  5. [Section III-B-c] The 'linear component size' parameter is introduced but its role is not fully explained; a sentence describing how the linear component is added to the smoothed random curve and how its magnitude affects the remapping would improve reproducibility.
  6. [Section IV] The text reports a DSC value of 0.62 in the discussion but Table IV lists 0.63 for the no-augmentation condition; please verify and align all reported numbers.

Circularity Check

1 steps flagged · score 3.0 of 10

The core augmentation comparison is an empirical measurement, not a definitional derivation; the only genuine circularity is that the QIN-Breast 'independent test' score uses post-processing chosen after inspecting that same test set.

  1. fitted input called prediction [Section III-C and Section III-F/IV (QIN-Breast evaluation)]
    "Section III-C: 'The QIN-Breast dataset is used as a completely unseen test set and was only used for final testing.' Section III-F: 'Initial resulting segmentations on the QIN-Breast dataset showed segmented regions in the thorax which were not connected to the breast segmentation. In order to focus on the performance of the segmentation of the breast itself, we decided to implement post-processing for this dataset to remove those regions from the segmentations.'"

    The post-processing rule (2D connected-component selection with a 100-pixel threshold plus 3D largest-component retention) was added after the authors inspected QIN-Breast outputs, yet the same dataset is described as a completely unseen final test set. The reported QIN-Breast DSC of 0.89 is therefore not produced by a fixed, pre-specified pipeline; it is the score of a pipeline whose post-processing was fitted to the test set being evaluated. This makes the 'independent test set' claim partially circular by construction, even though the central T1-to-T2 comparison is not definitionally circular.

full rationale

The paper's central T1-to-T2 comparison (DSC increase from 0.71 to 0.90 with intensity augmentation) is an empirical measurement, not a derivation, and no equation or fitted parameter forces that gain by construction. The augmentation ratio and remapping parameters are stated but not defined in terms of the test DSC, so there is no self-definitional step. The use of the authors' prior work [10] for architecture depth, dropout, and Sunnybrook ground truth is normal prior-work reuse and does not carry the novel claim. The main validity concern in the T1-to-T2 comparison is a confound: intensity-augmented runs use two pre-generated augmented volumes per original at a 1:2 ratio while geometric-only runs see one online augmented volume per epoch, and all runs train for 200 epochs, so the comparison is not matched in data volume or optimizer updates. However, a confound is an experimental control issue, not a circular derivation. The one genuine circular element is the QIN-Breast evaluation: the dataset is called a completely unseen final test set, yet the post-processing rule used to produce the reported 0.89 was designed after inspecting QIN-Breast segmentations. That makes the 'independent test' score partially test-set-derived, though not forced by construction. Overall score 3.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method relies on several hand-picked augmentation parameters and domain assumptions. No new physical entities are introduced. The main circularity concern is the QIN-Breast post-processing and in-house ground truth, which affect the 'independent test' claim.

free parameters (5)
  • style transfer weight alpha = 0.5
    Set to 0.5 based on Jackson et al. [20], controls the blend of random style and original image style.
  • moving average window size for intensity remapping = 20
    Chosen by visual inspection of remapped images (Section III-B.c).
  • linear component size for intensity remapping = 0.5
    Chosen by visual inspection; determines how much original intensity ordering is preserved.
  • augmentation ratio (original to augmented) = 1:2
    Determined by preliminary experiments (Section III-B.d).
  • geometric augmentation ranges = scaling 0.8-1.2, rotation -5 to 5 deg, translation 10 mm in-plane / 5 mm slice
    Chosen by visual inspection of possible variation (Section III-B.a).
assumptions (4)
  • domain assumption Distorting intensities while preserving shape forces the network to learn shape features rather than texture or intensity features.
    This is the central hypothesis motivating the method, based on Geirhos et al. [30] and Jackson et al. [20]. It is not proven in this paper.
  • domain assumption The 3D U-Net with the specified architecture is an adequate segmentation model for breast MRI.
    The network is taken from prior work [10] and is not re-validated here.
  • domain assumption The Sunnybrook T1-to-T2 transfer is representative of intensity domain shift in breast MRI.
    Used to test the method; results may not generalize to other protocols.
  • domain assumption The manually generated ground truth for QIN-Breast, with different boundary rules than Sunnybrook, is an appropriate reference for evaluating segmentation performance.
    Ground truth was created by the authors with manual cutting points and lateral boundary rules, which differ from the Sunnybrook protocol.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Intensity augmentation for domain transfer of whole breast segmentation in MRI." pith.science (2026). https://pith.science/paper/4FR6SITE

@misc{pith2026190902642,
  author       = {Pith},
  title        = {Pith review of: Intensity augmentation for domain transfer of whole breast segmentation in MRI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4FR6SITE}},
  note         = {Machine review of arXiv:1909.02642}
}
read the original abstract

The segmentation of the breast from the chest wall is an important first step in the analysis of breast magnetic resonance images. 3D U-nets have been shown to obtain high segmentation accuracy and appear to generalize well when trained on one scanner type and tested on another scanner, provided that a very similar T1-weighted MR protocol is used. There has, however, been little work addressing the problem of domain adaptation when image intensities or patient orientation differ markedly between the training set and an unseen test set. To overcome the domain shift we propose to apply extensive intensity augmentation in addition to geometric augmentation during training. We explored both style transfer and a novel intensity remapping approach as intensity augmentation strategies. For our experiments, we trained a 3D U-net on T1-weighted scans and tested on T2-weighted scans. By applying intensity augmentation we increased segmentation performance from a DSC of 0.71 to 0.90. This performance is very close to the baseline performance of training and testing on T2-weighted scans (0.92). Furthermore, we applied our network to an independent test set made up of publicly available scans acquired using a T1-weighted TWIST sequence and a different coil configuration. On this dataset we obtained a performance of 0.89, close to the inter-observer variability of the ground truth segmentations (0.92). Our results show that using intensity augmentation in addition to geometric augmentation is a suitable method to overcome the intensity domain shift and we expect it to be useful for a wide range of segmentation tasks.

Figures

Figures reproduced from arXiv: 1909.02642 by the authors.

Figure 1
Figure 1. Examples of the intensity augmentation methods. The image at the left is the original scan, the top row shows the style [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the style transfer pipeline. The input scan volume is converted to 3-channel RGB volume and analyzed [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Overview of the intensity remapping pipeline. The intensities of the input image are transformed to new intensities [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Example slices of the two datasets. The images shown were resampled to the axial direction and to the same size and [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Overlay of two breast volumes of similar size from [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Box plots of the DSC of the different augmentations methods on the Sunnybrook T2 test set and the QIN-Breast dataset [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 7
Figure 7. Figure 7: Sample output segmentations with different augmentation strategies. The green solid line represent the ground truth [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 26 canonical work pages

  1. [1]

    Estimating the global cancer incidence and mortality in 2018: Globocan sources and methods,

    J. Ferlay, M. Colombet, I. Soerjomataram, C. Mathers, D. Parkin, M. Piñeros, A. Znaor, and F. Bray, “Estimating the global cancer incidence and mortality in 2018: Globocan sources and methods,” International journal of cancer , vol. 144, no. 8, pp. 1941–1953, 2019

    work page 2018

  2. [2]

    Prospective study of breast cancer incidence in women with a brca1 or brca2 mutation under surveillance with and without magnetic resonance imaging

    E. Warner, K. Hill, P. Causer, D. Plewes, R. Jong, M. Yaffe, W. D. Foulkes, P. Ghadirian, H. Lynch, F. Couch, J. Wong, F. Wright, P. Sun, and S. a. Narod, “Prospective study of breast cancer incidence in women with a brca1 or brca2 mutation under surveillance with and without magnetic resonance imaging.” Journal of clinical oncology , vol. 29, no. 13, pp....

    work page 2011

  3. [3]

    Effectiveness of screening with annual magnetic resonance imaging and mammography: results of the initial screen from the ontario high risk breast screening program,

    A. M. Chiarelli, M. V . Prummel, D. Muradali, V . Majpruz, M. Horgan, J. C. Carroll, A. Eisen, W. S. Meschino, R. S. Shumak, E. Warner et al., “Effectiveness of screening with annual magnetic resonance imaging and mammography: results of the initial screen from the ontario high risk breast screening program,” Journal of clinical oncology , vol. 32, no. 21...

    work page 2014

  4. [4]

    Efficacy of mri and mammography for breast- cancer screening in women with a familial or genetic predisposition,

    M. Kriege, C. T. Brekelmans, C. Boetes, P. E. Besnard, H. M. Zon- derland, I. M. Obdeijn, R. A. Manoliu, T. Kok, H. Peterse, M. M. Tilanus-Linthorst et al., “Efficacy of mri and mammography for breast- cancer screening in women with a familial or genetic predisposition,” New England Journal of Medicine , vol. 351, no. 5, pp. 427–437, 2004

    work page 2004

  5. [5]

    Background parenchymal enhancement at breast mr imaging and breast cancer risk,

    V . King, J. D. Brooks, J. L. Bernstein, A. S. Reiner, M. C. Pike, and E. A. Morris, “Background parenchymal enhancement at breast mr imaging and breast cancer risk,” Radiology, vol. 260, no. 1, pp. 50–60, 2011. 10

    work page 2011

  6. [6]

    The associa- tion of background parenchymal enhancement at breast mri with breast cancer: A systematic review and meta-analysis,

    C. M. Thompson, I. Mallawaarachchi, D. K. Dwivedi, A. P. Ayyappan, N. K. Shokar, R. Lakshmanaswamy, and A. K. Dwivedi, “The associa- tion of background parenchymal enhancement at breast mri with breast cancer: A systematic review and meta-analysis,” Radiology, vol. 292, no. 3, pp. 552–561, sep 2019

    work page 2019

  7. [7]

    S. Wu, S. P. Weinstein, M. J. DeLeo, E. F. Conant, J. Chen, S. M. Domchek, and D. Kontos, “Quantitative assessment of background parenchymal enhancement in breast MRI predicts response to risk- reducing salpingo-oophorectomy: preliminary evaluation in a cohort of BRCA1/2 mutation carriers,” Breast Cancer Research , vol. 17, no. 1, p. 67, 2015

    work page 2015

  8. [8]

    Mammographic density, mri background parenchymal enhancement and breast cancer risk,

    M. C. Pike and C. L. Pearce, “Mammographic density, mri background parenchymal enhancement and breast cancer risk,” Annals of Oncology , vol. 24, no. SUPPL.B, pp. viii37–viii41, 2013

    work page 2013

Show all 39 references
  1. [9]

    A survey on deep learning in medical image analysis,

    G. Litjens, T. Kooi, B. E. Bejnordi, A. A. A. Setio, F. Ciompi, M. Ghafoorian, J. A. Van Der Laak, B. Van Ginneken, and C. I. Sánchez, “A survey on deep learning in medical image analysis,” Medical image analysis, vol. 42, pp. 60–88, 2017

    2017

  2. [10]

    An investiga- tion of the effect of fat suppression and dimensionality on the accuracy of breast mri segmentation using u-nets,

    H. Fashandi, G. Kuling, Y . Lu, H. Wu, and A. L. Martel, “An investiga- tion of the effect of fat suppression and dimensionality on the accuracy of breast mri segmentation using u-nets,” Medical physics, vol. 46, no. 3, pp. 1230–1244, 2019

    2019

  3. [11]

    Automatic Breast and Fibroglandular Tissue Segmentation in Breast MRI Using Deep Learning by a Fully-Convolutional Residual Neural Network U- Net,

    Y . Zhang, J.-H. Chen, K.-T. Chang, V . Y . Park, M. J. Kim, S. Chan, P. Chang, D. Chow, A. Luk, T. Kwong, and M.-Y . Su, “Automatic Breast and Fibroglandular Tissue Segmentation in Breast MRI Using Deep Learning by a Fully-Convolutional Residual Neural Network U- Net,” Academ...

    2019

  4. [12]

    Generative adversarial nets,

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversarial nets,” in Advances in neural information processing systems , 2014, pp. 2672– 2680

    2014

  5. [13]

    A comprehensive survey on domain adaptation for visual applications,

    G. Csurka, “A comprehensive survey on domain adaptation for visual applications,” in Domain Adaptation in Computer Vision Applications . Springer, 2017, pp. 1–35

    2017

  6. [14]

    Transfer learning for domain adaptation in mri: Application in brain lesion segmentation,

    M. Ghafoorian, A. Mehrtash, T. Kapur, N. Karssemeijer, E. Marchiori, M. Pesteie, C. R. Guttmann, F.-E. de Leeuw, C. M. Tempany, B. van Ginneken et al. , “Transfer learning for domain adaptation in mri: Application in brain lesion segmentation,” in International Conference on M...

    2017

  7. [15]

    Convolutional neural networks for medical image analysis: Full training or fine tuning?

    N. Tajbakhsh, J. Y . Shin, S. R. Gurudu, R. T. Hurst, C. B. Kendall, M. B. Gotway, and J. Liang, “Convolutional neural networks for medical image analysis: Full training or fine tuning?” IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1299–1312, 2016

    2016

  8. [16]

    Imagenet classification with deep convolutional neural networks,

    A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Advances in neural infor- mation processing systems , 2012, pp. 1097–1105

    2012

  9. [17]

    Domain randomization for transferring deep neural networks from sim- ulation to the real world,

    J. Tobin, R. Fong, A. Ray, J. Schneider, W. Zaremba, and P. Abbeel, “Domain randomization for transferring deep neural networks from sim- ulation to the real world,” in 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2017, pp. 23–30

    2017

  10. [18]

    Arbitrary style transfer in real-time with adaptive instance normalization,

    X. Huang and S. Belongie, “Arbitrary style transfer in real-time with adaptive instance normalization,” in Proceedings of the IEEE Interna- tional Conference on Computer Vision , 2017, pp. 1501–1510

    2017

  11. [19]

    Image style transfer using convolutional neural networks,

    L. A. Gatys, A. S. Ecker, and M. Bethge, “Image style transfer using convolutional neural networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 2414–2423

    2016

  12. [20]

    Style augmentation: Data augmentation via style random- ization,

    P. T. Jackson, A. Atapour-Abarghouei, S. Bonner, T. Breckon, and B. Obara, “Style augmentation: Data augmentation via style random- ization,” arXiv preprint arXiv:1809.05375 , 2018

    2018 arXiv

  13. [21]

    Domain-adversarial training of neural networks,

    Y . Ganin, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Lavio- lette, M. Marchand, and V . Lempitsky, “Domain-adversarial training of neural networks,” The Journal of Machine Learning Research , vol. 17, no. 1, pp. 2096–2030, 2016

    2016

  14. [22]

    Adversarial discrim- inative domain adaptation,

    E. Tzeng, J. Hoffman, K. Saenko, and T. Darrell, “Adversarial discrim- inative domain adaptation,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 7167–7176

    2017

  15. [23]

    Domain-adversarial neural networks to address the appearance variability of histopathology images,

    M. W. Lafarge, J. P. Pluim, K. A. Eppenhof, P. Moeskops, and M. Veta, “Domain-adversarial neural networks to address the appearance variability of histopathology images,” in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support . Springe...

    2017

  16. [24]

    Differential data augmentation techniques for medical imaging classification tasks,

    Z. Hussain, F. Gimenez, D. Yi, and D. Rubin, “Differential data augmentation techniques for medical imaging classification tasks,” in AMIA Annual Symposium Proceedings , vol. 2017. American Medical Informatics Association, 2017, p. 979

    2017

  17. [25]

    Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology,

    D. Tellez, G. Litjens, P. Bandi, W. Bulten, J.-M. Bokhorst, F. Ciompi, and J. van der Laak, “Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology,” arXiv preprint arXiv:1902.06543 , 2019

    1902 arXiv

  18. [26]

    Detecting cancer metastases on gigapixel pathology images,

    Y . Liu, K. Gadepalli, M. Norouzi, G. E. Dahl, T. Kohlberger, A. Boyko, S. Venugopalan, A. Timofeev, P. Q. Nelson, G. S. Corrado et al. , “Detecting cancer metastases on gigapixel pathology images,” arXiv preprint arXiv:1703.02442, 2017

    2017 arXiv

  19. [27]

    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,” inEuropean conference on computer vision. Springer, 2016, pp. 694–711

    2016

  20. [28]

    Texture networks: Feed-forward synthesis of textures and stylized images

    D. Ulyanov, V . Lebedev, A. Vedaldi, and V . S. Lempitsky, “Texture networks: Feed-forward synthesis of textures and stylized images.” in ICML, vol. 1, no. 2, 2016, p. 4

    2016

  21. [29]

    Exploring the structure of a real-time, arbitrary neural artistic stylization network,

    G. Ghiasi, H. Lee, M. Kudlur, V . Dumoulin, and J. Shlens, “Exploring the structure of a real-time, arbitrary neural artistic stylization network,” arXiv preprint arXiv:1705.06830 , 2017

    2017 arXiv

  22. [30]

    Imagenet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness,

    R. Geirhos, P. Rubisch, C. Michaelis, M. Bethge, F. A. Wichmann, and W. Brendel, “Imagenet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness,” arXiv preprint arXiv:1811.12231, 2018

    2018 arXiv

  23. [31]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in International Conference on Medical image computing and computer-assisted intervention . Springer, 2015, pp. 234–241

    2015

  24. [32]

    3d u-net: learning dense volumetric segmentation from sparse anno- tation,

    Ö. Çiçek, A. Abdulkadir, S. S. Lienkamp, T. Brox, and O. Ronneberger, “3d u-net: learning dense volumetric segmentation from sparse anno- tation,” in International conference on medical image computing and computer-assisted intervention . Springer, 2016, pp. 424–432

    2016

  25. [33]

    TensorFlow: Large-scale machine learning on heterogeneous systems,

    M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G. S. Corrado, A. Davis, J. Dean, M. Devin, S. Ghemawat, I. Goodfellow, A. Harp, G. Irving, M. Isard, Y . Jia, R. Jozefowicz, L. Kaiser, M. Kudlur, J. Levenberg, D. Mané, R. Monga, S. Moore, D. Murray, C. Olah, M. ...

    2015

  26. [34]

    Chollet et al

    F. Chollet et al. , “Keras,” https://keras.io, 2015

    2015

  27. [35]

    SciPy: Open source scientific tools for Python,

    E. Jones, T. Oliphant, P. Peterson et al. , “SciPy: Open source scientific tools for Python,” 2001–, [Online; accessed 04-09-2019]. [Online]. Available: http://www.scipy.org/

    2001

  28. [36]

    Variations of dynamic contrast-enhanced magnetic resonance imaging in evaluation of breast cancer therapy response: a multicenter data analysis challenge,

    W. Huang, A. Tudorica, S. Chui, and K. Kemmer, “Variations of dynamic contrast-enhanced magnetic resonance imaging in evaluation of breast cancer therapy response: a multicenter data analysis challenge,” 2014. [Online]. Available: http://doi.org/10.7937/K9/TCIA. 2014.A2N1IXOX

    2014 doi

  29. [37]

    Breast segmentation in mri using poisson surface reconstruction initialized with random forest edge detection,

    A. L. Martel, C. Gallego-Ortiz, and Y . Lu, “Breast segmentation in mri using poisson surface reconstruction initialized with random forest edge detection,” in Medical Imaging 2016: Image Processing , vol. 9784. International Society for Optics and Photonics, 2016, p. 97841B

    2016

  30. [38]

    User-guided 3d active contour segmentation of anatomical structures: significantly improved efficiency and reliability,

    P. A. Yushkevich, J. Piven, H. C. Hazlett, R. G. Smith, S. Ho, J. C. Gee, and G. Gerig, “User-guided 3d active contour segmentation of anatomical structures: significantly improved efficiency and reliability,” Neuroimage, vol. 31, no. 3, pp. 1116–1128, 2006

    2006

  31. [39]

    Simultaneous truth and performance level estimation (staple): an algorithm for the validation of image segmentation,

    S. K. Warfield, K. H. Zou, and W. M. Wells, “Simultaneous truth and performance level estimation (staple): an algorithm for the validation of image segmentation,” IEEE transactions on medical imaging , vol. 23, no. 7, p. 903, 2004. 11 APPENDIX A STATISTICAL RESULTS TABLE I: Sta...

    2004

Pith tools

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