Pith. sign in

REVIEW 4 major objections 5 minor 27 references

Learning from Sparse Point Labels for Dense Carcinosis Localization in Advanced Ovarian Cancer Assessment

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper shows that a heatmap-regression loss called Crag and Tail, trained on only about five sparse carcinosis point labels per frame, densely localizes carcinosis nodules in laparoscopic video and outperforms the MSE and Hill…

desk verdict A genuinely new sparse-label loss and clinical task, but the SOTA claim rests on one seed and six test videos. read the letter →

arxiv 2507.06643 v1 pith:BKXOPKCT submitted 2025-07-09 cs.CV cs.LG

Farahdiba Zarin , Riccardo Oliva , Vinkle Srivastav , Armine Vardazaryan , Andrea Rosati , Alice Zampolini Faustini , Giovanni Scambia , Anna Fagotti
show 2 more authors
Pietro Mascagni Nicolas Padoy
This is my paper · ORCID
classification cs.CVcs.LG
keywords sparseannotationskeypointlocalizationheatmapregressioncarcinosisovariancancerlaparoscopynoisylabelssurgicalAI
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper seeks to establish that dense localization of carcinosis nodules in laparoscopic video can be learned from only a few sparse point annotations per frame, without dense pixel-level labels. To do this it introduces a heatmap-regression task and a loss function, Crag and Tail, that adapts the multi-label Hill loss by down-weighting possible missed annotations while adding a term that reinforces true positives. On the authors' dataset, training on about five points per image with this loss reaches point-localization F1 of 54.59 and recall 46.76, beating the Hill and MSE baselines, and it also improves station-level classification used in Fagotti-score assessment. The result matters because full pixel-level annotation of a surgical image can take up to 90 minutes, whereas sparse points are quick for clinicians to provide; if the claim holds, sparse annotation could support automated ovarian-cancer assessment.

What carries the argument

The central object is the Crag and Tail loss, a dense heatmap-regression loss that combines a false-negative-suppressing negative term inherited from Hill loss with a new positive-reinforcement term. The loss is $$L_{\mathrm{CragAndTail}}=\sum_p \big(H\log(\hat{H}_+)+(H-\hat{H})^2\big)(1-\hat{H}_+)^\gamma + (1-H)\big(-(\$\lambda$-\hat{H}_-)\hat{H}_-^2\big),$$ with $\hat{H}_+=\sigma(\hat{H}-m)$ and $\hat{H}_-=\sigma(\hat{H})$, where $H$ is the sparse Gaussian-peak ground-truth heatmap and $\hat{H}$ the predicted heatmap. The first factor and the negative term jointly down-weight loss on pixels that are probably missed annotations, while the $(H-\hat{H})^2$ term prevents the focal positive term from under-weighting easy true positives. At inference, the predicted heatmap is processed by 5x5 non-maximum suppression followed by a threshold $t$ to obtain point predictions.

What would settle it

One concrete experiment would be to compute per-station recall on the 279 test frames for Crag and Tail($\lambda=1$). If stations that supplied few or no point labels during training show disproportionately low recall relative to the 46.76% average, the assumption that five sparse points generalize to all ~36 carcinosis instances would be disproved for those stations.

Watch

Extended reading notes

Core claim

The central discovery is that a model trained on only a few sparse point annotations per frame can densely localize all carcinosis instances if the loss is designed to treat unlabeled positives as false negatives and to keep true positives strongly weighted. The paper adapts the Hill loss, originally for multi-label classification with missing labels, to pixel-level heatmap regression and adds a positive-reinforcement term $(H-\hat{H})^2$ to obtain the Crag and Tail loss. On the authors' test set, at the tuned false-negative weight $\lambda=1$, this loss reaches point-localization F1 54.59 and recall 46.76, versus 52.06 and 44.01 for Hill and 18.99 and 15.83 for MSE, with similar or better station-level multilabel classification at default settings.

Load-bearing premise

The load-bearing premise is that the roughly five carcinosis points a surgeon marks per frame are representative of all 36 carcinosis instances actually present, so the model can learn to detect the unlabeled ones from these examples.

Editorial extensions

If this is right

  • At $\lambda=1$, Crag and Tail raises point-localization recall by 2.75 points over Hill (46.76 vs. 44.01) while keeping precision nearly equal (73.79 vs. 73.89).
  • The ablation that sets the positive-reinforcement term to zero retains only the negative loss and drops F1 to 11.89, and removing the original positive Hill term drops F1 to 11.98, so both positive terms are necessary for the reported performance.
  • Setting $\lambda=1$ rather than the default 1.5 improves F1 from 39.84 to 54.59, showing that the false-negative suppression strength is a key tunable for this sparse-label regime.
  • On the station-level multilabel task, Crag and Tail at default $\lambda$ reaches F1 81.84 versus Hill's 80.06 and MSE's 48.22, suggesting the dense localization gains carry over to the clinically relevant Fagotti-score-assessment proxy.
  • The 0.5 Masked MSE baseline, which naively halves the loss on assumed false negatives, improves over MSE from F1 18.99 to 29.70, confirming that false-negative handling is the main driver of the sparse-label difficulty.

Reading between the lines

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

  • A natural extension is per-station recall evaluation: because the training set averages five points over six anatomical stations, stations that contributed few or no labeled points may be under-recovered, and the paper's reported recall gain may not be uniform across stations.
  • The same loss could transfer to other sparse-instance detection tasks where annotators mark only a subset of instances, such as counting cell nuclei or detecting polyps, but the optimal $\lambda$ would need retuning to the missing-label rate.
  • Comparing Crag and Tail against strategic point placement, such as one point on each of five stations rather than five points clustered on one station, would show whether the loss or the annotation distribution drives the observed gain.
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 / 5 minor

Summary. The paper introduces a new task of dense carcinosis keypoint localization in laparoscopic video frames for advanced ovarian cancer assessment, where only a few sparse point annotations are available per training image. To address the sparse-label setting, the authors adapt the Hill loss, originally designed for multi-label classification with missing labels, to dense heatmap regression, and propose a new loss function named Crag and Tail loss that adds a positive reinforcement term (H - H_hat)^2 to the Hill loss. The method is evaluated on a newly collected dataset of 30 videos, comparing against MSE and Hill baselines, with metrics for point localization and multilabel station classification. The paper reports that Crag and Tail loss achieves the highest recall and F1 for point localization, and supports the claim with an ablation study of loss components.

Significance. If the reported results hold, the work is practically significant because it reduces annotation effort for dense localization tasks in surgical imaging, and it introduces a clinically relevant new task. The adaptation of Hill loss to dense heatmap regression is a reasonable idea, and the ablation study provides some insight into loss design. However, the empirical support is currently limited: the evaluation is based on a single training run, a test set of only 6 videos, and comparisons that involve different inference thresholds and tuned hyperparameters, so the claimed state-of-the-art improvement is not yet established. No code or trained models are released, which limits reproducibility.

major comments (4)
  1. [Table 1, Section 3] The central claim of state-of-the-art performance rests on a single training run per loss evaluated on 6 test videos (279 frames) with no confidence intervals, error bars, or significance tests. At the tuned lambda=1 setting, Crag and Tail improves point localization F1 over Hill by 2.53 points (54.59 vs 52.06) but has slightly lower precision (73.79 vs 73.89) and lower multilabel F1 (77.88 vs 78.03). Given the strong within-video correlation of the test frames, these differences are within plausible sampling noise. Please provide results across multiple training seeds or cross-validation folds, report variance, and use a paired statistical test across videos (e.g., Wilcoxon signed-rank) to support the superiority claim.
  2. [Section 3, Implementation Details] The inference procedure selects the top k=30 predicted points per image, while Section 3 states that test images contain on average 36 carcinosis instances. This imposes an upper bound on recall of roughly 30/36 = 83% even with perfect localization, and the cap may affect different losses unevenly. The authors should either increase k to a value above the maximum number of instances in any test image (e.g., 50 or 100) and report results, or explicitly justify the choice of k=30 and analyze its influence on the reported recall and F1.
  3. [Section 2.2, Eq. (2)] The proposed Crag and Tail loss mixes unnormalized raw network outputs H_hat in the reinforcement term (H - H_hat)^2 with sigmoid-transformed values H_hat+ and H_hat- in the other terms. HRNet outputs are unbounded, so the squared term may have a different scale and could dominate the loss or lead to training instability. Please clarify whether this mixed use is intentional, report the typical range of raw outputs during training, or use a consistent transformation (e.g., (H - H_hat+)^2) so that all terms operate on comparable scales.
  4. [Section 4, Table 1; Section 3, Implementation Details] The comparison between losses is confounded by method-specific inference thresholds (t=0.2 for MSE, t=0.3 for all others) and by hyperparameter tuning: Crag and Tail uses lambda=1 while Hill uses lambda=1.5 in the main comparison, and the paper states that tuned lambda=1 results for both are provided without describing the tuning grid, the validation metric, or the selection procedure. This makes it difficult to attribute the observed differences to the loss function rather than to threshold or hyperparameter choices. Please report the tuning protocol and show results across a range of thresholds (e.g., precision-recall curves) or use a common threshold selection procedure for all methods.
minor comments (5)
  1. [Eq. (1), Eq. (2)] The log( H_hat+ ) and log( H_hat ) terms should include a small epsilon inside the logarithm to avoid numerical instability when the predicted probability is zero.
  2. [Section 2.1] The notation Y* in R^{2 x n} uses n for the number of keypoints, but later in Section 2.2 the text uses n=0 and n>0 to indicate whether any carcinosis points exist; please use distinct symbols or clarify the overloading.
  3. [Section 3] The sentence 'For a frame-level indication of whether the station contains carcinosis or not, 2D keypoint annotations are used for 1-5 prominent cancerous keypoints inside the masks' is unclear about how the keypoints relate to the frame-level labels; please rephrase to explain the annotation protocol.
  4. [Table 2] The ablation rows such as 'Hlog(H_hat+) = 0' and 'Only reinforce term in pos' are not self-explanatory; please include the exact modified loss equations or reference the components of Eq. (2) that are removed or changed.
  5. [Figure 3] The caption for the qualitative results does not describe the content; it should specify which methods are compared, what the markers indicate, and what the reader should observe.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the loss comparison is an empirical evaluation on held-out test data, and no prediction reduces by construction to a fitted quantity.

full rationale

The paper's central claim is an empirical ranking of loss functions for sparse-point heatmap regression. The proposed Crag and Tail loss is explicitly defined in Eq. (2) as a hand-designed modification of the external Hill loss [26], adding a positive-reinforcement term (H - H_hat)^2. No equation in the derivation chain is fitted to the test-set outcomes that are later reported. Hyperparameters such as lambda, gamma, and m are selected on the validation split and reported with default and tuned values, and the metrics in Table 1 are computed on held-out test videos (279 frames). The evaluation using dense carcinosis masks is independent of the sparse point labels used for training, so the reported F1 and recall values are measurements rather than tautological outputs of the loss definition. The only self-citations to the authors' prior work are contextual background references (surgical phase recognition, surgical data science, action triplets) and are not load-bearing for the loss derivation or the empirical comparison. The absence of error bars, single training runs, and the small number of test videos are legitimate concerns about statistical robustness and reproducibility, but they are correctness risks rather than circularity: a noisy or underpowered measurement is still a measurement. No uniqueness theorem, imported ansatz, or fitted-input-renamed-as-prediction step appears in the paper. The claim that Crag and Tail outperforms Hill and MSE is therefore not forced by construction or by self-citation, and the derivation chain is self-contained as an empirical study.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The central claim rests on standard heatmap regression assumptions plus a domain assumption about the consistency of carcinosis appearance. No new physical entities are introduced. The loss hyperparameters λ, γ, m, and t are tuned on validation data and affect the reported results.

free parameters (6)
  • lambda = 1.5 default, 1 after tuning
    Controls down-weighting of false negatives; tuned on validation and affects the negative loss term (Tables 1 and 2).
  • gamma = 2
    Focal weight for semi-hard positives; ablation shows performance drops when changed.
  • m = 1
    Margin for rescaling positive samples; ablation shows sensitivity.
  • delta = not reported
    Spread of the Gaussian heatmap in Section 2.2; affects target heatmap density but the value is not given.
  • t = 0.2 for MSE, 0.3 for others
    Inference threshold after non-maximum suppression; tuned per loss.
  • k = 30
    Number of top points kept at inference after NMS.
assumptions (3)
  • standard math Gaussian heatmaps centered at annotated keypoints are a valid regression target for keypoint localization.
    Equation in Section 2.2 defines the ground truth as summed Gaussians; this is standard practice in pose estimation.
  • domain assumption Unannotated pixels outside sparse keypoints are potential false negatives, not true negatives.
    Section 2.2 states this assumption explicitly and it justifies the loss design; if false, the down-weighting of background is unnecessary.
  • domain assumption The visual signature of carcinosis is consistent across patients and video frames, so sparse labels generalize to dense locations.
    Implicit in the task formulation and dataset split; without it, training on five points per frame cannot localize 36 instances per test frame.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning from Sparse Point Labels for Dense Carcinosis Localization in Advanced Ovarian Cancer Assessment." pith.science (2026). https://pith.science/paper/BKXOPKCT

@misc{pith2026250706643,
  author       = {Pith},
  title        = {Pith review of: Learning from Sparse Point Labels for Dense Carcinosis Localization in Advanced Ovarian Cancer Assessment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BKXOPKCT}},
  note         = {Machine review of arXiv:2507.06643}
}
read the original abstract

Learning from sparse labels is a challenge commonplace in the medical domain. This is due to numerous factors, such as annotation cost, and is especially true for newly introduced tasks. When dense pixel-level annotations are needed, this becomes even more unfeasible. However, being able to learn from just a few annotations at the pixel-level, while extremely difficult and underutilized, can drive progress in studies where perfect annotations are not immediately available. This work tackles the challenge of learning the dense prediction task of keypoint localization from a few point annotations in the context of 2d carcinosis keypoint localization from laparoscopic video frames for diagnostic planning of advanced ovarian cancer patients. To enable this, we formulate the problem as a sparse heatmap regression from a few point annotations per image and propose a new loss function, called Crag and Tail loss, for efficient learning. Our proposed loss function effectively leverages positive sparse labels while minimizing the impact of false negatives or missed annotations. Through an extensive ablation study, we demonstrate the effectiveness of our approach in achieving accurate dense localization of carcinosis keypoints, highlighting its potential to advance research in scenarios where dense annotations are challenging to obtain.

Figures

Figures reproduced from arXiv: 2507.06643 by the authors.

Figure 1
Figure 1. Carcinosis appears as protruding nodules on the organ surfaces. Green points are sparse point labels from annotators. Red arrows points to more possible instances. 1 Introduction Recent breakthroughs in surgical computer vision are driving the development of next-generation AI-assisted systems for the operating room [14]. The field has undergone significant evolution, transitioning from coarse-grained surgical workf… view at source ↗
Figure 2
Figure 2. Summarization of type of labels and loss components. False negatives (FN) with probabilities too similar to true positives (TP) have a low negative label weight, reducing their effect. The positive label term, since it focuses on hard-mining for posi￾tives, reduces the weight of easy positives. The reinforcement term doesn’t remove the hard-mining for difficult TP samples, but rather adds a higher effect of TPs. a m… view at source ↗
Figure 3
Figure 3. Qualitative results of the loss performance compared. LCragAndT ail succeeds in localizing more carcinosis points. The traditional LMSE struggles to learn from sparse point labels. For reference, the sparse point labels given by the clinicians are shown. within the Plan Investissements d’Avenir by the ANR under reference ANR- 10-IAHU￾02 (IHU Strasbourg). This work was granted access to the servers/HPC resources man￾… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 19 canonical work pages

  1. [1]

    arXiv preprint arXiv:1902.06426 (2019)

    Allan, M., Shvets, A., Kurmann, T., Zhang, Z., Duggal, R., Su, Y.H., Rieke, N., Laina, I., Kalavakonda, N., Bodenstedt, S., et al.: 2017 robotic instrument segmen- tation challenge. arXiv preprint arXiv:1902.06426 (2019)

    arXiv 2019

  2. [2]

    Jour- nal of the National Comprehensive Cancer Network19(2), 191–226 (2021)

    Armstrong, D.K., Alvarez, R.D., Bakkum-Gamez, J.N., Barroilhet, L., Behbakht, K., Berchuck, A., Chen, L.m., Cristea, M., DeRosa, M., Eisenhauer, E.L., et al.: Ovarian cancer, version 2.2020, nccn clinical practice guidelines in oncology. Jour- nal of the National Comprehensive Cancer Network19(2), 191–226 (2021)

    work page 2021

  3. [3]

    In: Medical Image Computing and Computer-Assisted Intervention–MICCAI 2008: 11th International Confer- ence, New York, NY, USA, September 6-10, 2008, Proceedings, Part II 11

    Blum, T., Padoy, N., Feußner, H., Navab, N.: Modeling and online recognition of surgical phases using hidden markov models. In: Medical Image Computing and Computer-Assisted Intervention–MICCAI 2008: 11th International Confer- ence, New York, NY, USA, September 6-10, 2008, Proceedings, Part II 11. pp. 627–635. Springer (2008)

    work page 2008

  4. [4]

    In: CVPR (2020)

    Cheng, B., Xiao, B., Wang, J., Shi, H., Huang, T.S., Zhang, L.: Higherhrnet: Scale- aware representation learning for bottom-up human pose estimation. In: CVPR (2020)

    2020

  5. [5]

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

    Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., Schiele, B.: The cityscapes dataset for semantic urban scene understanding. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 3213–3223 (2016)

    2016

  6. [6]

    Zarin et al

    Fagotti, A., Ferrandina, G., Fanfani, F., Ercoli, A., Lorusso, D., Rossi, M., Scam- bia, G.: A laparoscopy-based score to predict surgical outcome in patients with 10 F. Zarin et al. advanced ovarian carcinoma: a pilot study. Annals of surgical oncology13, 1156– 1161 (2006)

    work page 2006

  7. [7]

    Acta obstetricia et gy- necologica Scandinavica 90(10), 1126–1131 (2011)

    Fagotti, A., Vizzielli, G., Costantini, B., Lecca, A., Gallotta, V., Gagliardi, M.L., Scambia, G., Fanfani, F.: Learning curve and pitfalls of a laparoscopic score to describe peritoneal carcinosis in advanced ovarian cancer. Acta obstetricia et gy- necologica Scandinavica 90(10), 1126–1131 (2011)

    work page 2011

  8. [8]

    American journal of obstetrics and gynecology209(5), 462–e1 (2013)

    Fagotti, A., Vizzielli, G., De Iaco, P., Surico, D., Buda, A., Mandato, V.D., Petruzzelli, F., Ghezzi, F., Garzarelli, S., Mereu, L., et al.: A multicentric trial (olympia–mito 13) on the accuracy of laparoscopy to assess peritoneal spread in ovarian cancer. American journal of obstetrics and gynecology209(5), 462–e1 (2013)

    work page 2013

Show all 27 references
  1. [9]

    International Journal of Gynecologic Cancer31(1) (2021)

    Fleming, N.D., Westin, S.N., Meyer, L.A., Shafer, A., Rauh-Hain, J.A., Onstad, M., Cobb, L., Bevers, M., Fellman, B.M., Burzawa, J., et al.: Correlation of sur- geon radiology assessment with laparoscopic disease site scoring in patients with advanced ovarian cancer. Internati...

    2021

  2. [10]

    Gynecologic oncology161(1), 56–62 (2021)

    Harrison, R.F., Cantor, S.B., Sun, C.C., Villanueva, M., Westin, S.N., Fleming, N.D., Toumazis, I., Sood, A.K., Lu, K.H., Meyer, L.A.: Cost-effectiveness of la- paroscopic disease assessment in patients with newly diagnosed advanced ovarian cancer. Gynecologic oncology161(1), ...

    2021

  3. [11]

    Computer Methods and Programs in Biomedicine186, 105201 (2020)

    Hervella, Á.S., Rouco, J., Novo, J., Penedo, M.G., Ortega, M.: Deep multi-instance heatmap regression for the detection of retinal vessel crossings and bifurcations in eye fundus images. Computer Methods and Programs in Biomedicine186, 105201 (2020)

    2020

  4. [12]

    In: CVPR 2011

    Johnson, S., Everingham, M.: Learning effective human pose estimation from in- accurate annotation. In: CVPR 2011. pp. 1465–1472. IEEE (2011)

    2011

  5. [13]

    In: European Conference on Computer Vision

    Liu, Q., Zhang, Y., Bai, S., Yuille, A.: Explicit occlusion reasoning for multi-person 3d human pose estimation. In: European Conference on Computer Vision. pp. 497–

  6. [14]

    Maier-Hein, L., Eisenmann, M., Sarikaya, D., März, K., Collins, T., Malpani, A., Fallert, J., Feussner, H., Giannarou, S., Mascagni, P., et al.: Surgical data science– fromconceptstowardclinicaltranslation.Medicalimageanalysis 76,102306(2022)

    2022

  7. [15]

    Mirikharaji,Z.,Yan,Y.,Hamarneh,G.:Learningtosegmentskinlesionsfromnoisy annotations. In: Domain Adaptation and Representation Transfer and Medical Image Learning with Less Labels and Imperfect Data: First MICCAI Workshop, DART 2019, and First International Workshop, MIL3ID 201...

    2019

  8. [16]

    Medical Image Analysis78, 102433 (2022)

    Nwoye, C.I., Yu, T., Gonzalez, C., Seeliger, B., Mascagni, P., Mutter, D., Marescaux, J., Padoy, N.: Rendezvous: Attention mechanisms for the recognition of surgical action triplets in endoscopic videos. Medical Image Analysis78, 102433 (2022)

    2022

  9. [17]

    In: International conference on medical image computing and computer-assisted intervention

    Payer, C., Štern, D., Bischof, H., Urschler, M.: Regressing heatmaps for multiple landmark localization using cnns. In: International conference on medical image computing and computer-assisted intervention. pp. 230–238. Springer (2016)

    2016

  10. [18]

    In: International conference on machine learning

    Ren, M., Zeng, W., Yang, B., Urtasun, R.: Learning to reweight examples for robust deep learning. In: International conference on machine learning. pp. 4334–

  11. [19]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Rivoir, D., Pfeiffer, M., Docea, R., Kolbinger, F., Riediger, C., Weitz, J., Speidel, S.: Long-term temporally consistent unpaired video translation from simulated surgical 3d data. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 3343–3353 (202...

    2021

  12. [20]

    CA: a cancer journal for clinicians 68(1), 7–30 (2018)

    Siegel, R.L., Miller, K.D., Jemal, A.: Cancer statistics, 2018. CA: a cancer journal for clinicians 68(1), 7–30 (2018)

    2018

  13. [21]

    In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition

    Simon, T., Joo, H., Matthews, I., Sheikh, Y.: Hand keypoint detection in single images using multiview bootstrapping. In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. pp. 1145–1153 (2017)

    2017

  14. [22]

    IEEE transactions on neural networks and learning systems 34(11), 8135–8153 (2022)

    Song, H., Kim, M., Park, D., Shin, Y., Lee, J.G.: Learning from noisy labels with deep neural networks: A survey. IEEE transactions on neural networks and learning systems 34(11), 8135–8153 (2022)

    2022

  15. [23]

    Seminars in Oncology Nursing 35(2), 151–156 (2019)

    Stewart, C., Ralyea, C., Lockwood, S.: Ovarian cancer: An in- tegrated review. Seminars in Oncology Nursing 35(2), 151–156 (2019). https://doi.org/https://doi.org/10.1016/j.soncn.2019.02.001, https://www.sciencedirect.com/science/article/pii/S0749208119300129, gyne- cology Oncology

    2019 doi

  16. [24]

    Sun, K., Xiao, B., Liu, D., Wang, J.: Deep high-resolution representation learning for human pose estimation (2019), https://arxiv.org/abs/1902.09212

    2019 arXiv

  17. [25]

    IEEE Transactions on Image Process- ing 31, 623–635 (2022)

    Yi, R., Huang, Y., Guan, Q., Pu, M., Zhang, R.: Learning from pixel-level label noise: A new perspective for semi-supervised se- mantic segmentation. IEEE Transactions on Image Process- ing 31, 623–635 (2022). https://doi.org/10.1109/tip.2021.3134142, http://dx.doi.org/10.1109...

    2022

  18. [26]

    Zhang, Y., Cheng, Y., Huang, X., Wen, F., Feng, R., Li, Y., Guo, Y.: Simple and robust loss design for multi-label learning with missing labels (2021)

    2021

  19. [27]

    In: Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, Oc- tober 13–17, 2019, Proceedings, Part VI 22

    Zhu, H., Shi, J., Wu, J.: Pick-and-learn: Automatic quality evaluation for noisy- labeled image segmentation. In: Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, Oc- tober 13–17, 2019, Proceedings, Part VI...

    2019

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.