Entropy-guided contrastive learning for semi-supervised medical image segmentation.

In this paper, we present a comprehensive review of recently proposed semi-supervised learning methods for medical image segmentation and summarized both the technical novelties and empirical results. ... Semi-supervised learning has emerged as an appealing strategy and been widely applied to medical image segmentation tasks to train deep models with limited annotations. ... ANALYSIS OF EMPIRICAL RESULTS FOR SEMI-SUPERVISED MEDICAL IMAGE SEGMENTATION Common Evaluation Metrics for Medical Image Segmentation For medical image segmentation tasks, Dice Similarity Coefficient ...

arXiv:2207.14191v3 fatcat:gva2fzpos5efxfbod5kb4axm5a

Multiple Versions

Semi-supervised learning has been widely applied to medical image segmentation tasks since it alleviates the heavy burden of acquiring expert-examined annotations and takes the advantage of unlabeled data ... Medical image segmentation is a fundamental and critical step in many clinical approaches. ... In this paper, we present a novel uncertainty-guided mutual consistency learning framework for semi-supervised medical image segmentation. ...

arXiv:2112.02508v2 fatcat:dnppas5hhvgl3ejmrvg245wmj4

Multiple Versions

Experimental results on two benchmark medical segmentation datasets reveal our method's effectiveness in improving segmentation performance when few annotated images are available. ... However, it is not trivial to build reasonable pairs for a segmentation task in an unsupervised way. ... Among these self-supervised methods, contrastive learning has become a prevailing strategy for pre-training medical image segmentation models (Chaitanya et al., 2020; Zeng et al., 2021; Peng et al., 2021b ...

arXiv:2202.02371v2 fatcat:sup2l4iahjdl3dajxcytmmmwsq

Open Access Multiple Versions

In this paper, we present a novel semi-supervised learning method, Dual-Decoder Consistency via Pseudo-Labels Guided Data Augmentation (DCPA), for medical image segmentation. ... Comprehensive comparisons with state-of-the-art semi-supervised medical image segmentation methods were conducted under typical scenarios, utilizing 10% and 20% labeled data, as well as in the extreme ... Related work Supervised medical image segmentation In recent years, the superior learning capability of deep learning has inspired a fruitful of supervised learning based methods for medical image segmentation ...

arXiv:2308.16573v3 fatcat:jha47s2ymvdopp6jjl4xq4d74a

Multiple Versions

branch, and guides the entropy (student-confidence) term to avoid trivial solutions. ... In conjunction with a standard cross-entropy over the labeled pixels, our novel formulation integrates two important terms: (i) a Shannon entropy loss defined over the less-supervised images, which encourages ... Motivated by these observations, we propose a novel formulation for learning with mixed supervision in medical image segmentation. ...

arXiv:2012.08051v1 fatcat:kyduresnynakdkronhjzpkv3mu

Open Access

In this paper, we aim to boost the performance of semi-supervised learning for medical image segmentation with limited labels using a self-ensembling contrastive learning technique. ... Deep learning has demonstrated significant improvements in medical image segmentation using a sufficiently large amount of training data with manual labels. ... In a nutshell, we present a novel self-ensembling contrastive learning architecture for semi-supervised medical image segmentation, combining the imagelevel supervised loss and feature-level contrastive ...

arXiv:2105.12924v2 fatcat:sadhwoifbbbn7otbe4j6urtqji

Open Access Multiple Versions

However, multiple target objects (with different semantic meanings) may exist in a single image, which poses a problem for adapting traditional contrastive learning methods from prevalent 'image-level ... Compared with current state-of-the-art training strategies, our proposed pipeline yields a substantial improvement of 5.53% and 6.09% on Dice score for both medical image segmentation cohorts respectively ... In this work, we propose a semantic-aware attention-guided contrastive learning (AGCL) framework to advance multiobject medical image segmentation with contrastive learning. ...

arXiv:2106.01596v2 fatcat:cm7xh4qrarewjj5gegpudldzve

Multiple Versions

Despite achieving promising results in a breadth of medical image segmentation tasks, deep neural networks require large training datasets with pixel-wise annotations. ... , as well as recent semi-supervised approaches. ... Semi-supervised segmentation in medical images Semi-supervised learning is closely related to the proposed methodology. ...

arXiv:2109.10902v3 fatcat:ywbwpn3ckja7dapeqsnnq6veqi

Multiple Versions

To alleviate this burden, we propose a novel learning scenario, Exemplar Learning (EL), to explore automated learning processes for medical image segmentation with a single annotated image example. ... This innovative learning task is particularly suitable for medical image segmentation, where all categories of organs can be presented in one single image and annotated all at once. ... Related Work Semi-Supervised Medical Image Segmentation. ...

arXiv:2204.01713v2 fatcat:wisei5sognenfc4nvshd6xc2ju

Multiple Versions

High-quality pseudo labels are essential for semi-supervised semantic segmentation. ... To address these issues, we propose an Uncertainty-guided Collaborative Mean-Teacher (UCMT) for semi-supervised semantic segmentation with the high-confidence pseudo labels. ... Conclusion We present an uncertainty-guided collaborative mean-teacher for semi-supervised medical image segmentation. ...

arXiv:2301.04465v1 fatcat:xbqmoslqh5bffj4x7o54hmlgwy

Open Access

Deep learning-based semi-supervised learning (SSL) algorithms have led to promising results in medical images segmentation and can alleviate doctors' expensive annotations by leveraging unlabeled data. ... Meanwhile, our framework outperforms the state-of-the-art semi-supervised learning methods. ... Yechong Huang for constructive discussions, suggestion and manuscript proofread and also thank the organization teams of MICCAI 2018 left atrial segmentation challenge, the National Institutes of Health ...

doi:10.1609/aaai.v35i10.17066 fatcat:hyoiiacznfdm3gssmwdj7uyvqi

Contrastive learning (CL) is a form of self-supervised learning and has been widely used for various tasks. ... Different from widely studied instance-level contrastive learning, pixel-wise contrastive learning mainly helps with pixel-wise tasks such as medical landmark detection. ... Besides, some researchers introduce pixel-wise contrastive learning into supervised [36] or semi-supervised learning [1-3, 15, 16, 23, 35, 37, 39, 45, 46, 51, 52, 54] and succeed to improve their models ...

arXiv:2211.07118v1 fatcat:usd6d24xbrec5alckfqb3seenm

This is a very frequent DA scenario in medical imaging, for instance, when the source and target images come from different clinical sites. ... Our formulation is based on minimizing a label-free entropy loss defined over target-domain data, which we further guide with a domain invariant prior on the segmentation regions. ... This is the underlying motivation for entropy minimization, which was first introduced in semi-supervised [5] and unsupervised [13] learning. ...

arXiv:2005.03697v1 fatcat:37fikv6cgbc3ljdsfxstg27roq

Multiple Versions

Semi-supervised segmentation tackles the scarcity of annotations by leveraging unlabeled data with a small amount of labeled data. ... Such a reconstructed segmentation mask aids in estimating the pixel-level uncertainty guiding the segmentation network. ... Acknowledgments: This research work was partly funded by the Canada Research Chair on Shape Analysis in Medical Imaging, the Natural Sciences and Engineering Research Council of Canada (NSERC), and the ...

arXiv:2203.05682v1 fatcat:fcfcg7salvfx3b64shxwu7e4di

Deep learning-based semi-supervised learning (SSL) algorithms have led to promising results in medical images segmentation and can alleviate doctors' expensive annotations by leveraging unlabeled data. ... Meanwhile, our framework outperforms the state-of-the-art semi-supervised medical image segmentation methods. Code is available at: https://github.com/Luoxd1996/DTC ... Yechong Huang for constructive discussions, suggestion and manuscript proofread and also thank the organization teams of MICCAI 2018 left atrial segmentation challenge, the National Institutes of Health ...

arXiv:2009.04448v2 fatcat:x7bsdzugcvfdjbyrb4uqxyobr4

Multiple Versions

Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Uncertainty-Guided Mutual Consistency Learning for Semi-Supervised Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Boundary-aware Information Maximization for Self-supervised Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Dual-Decoder Consistency via Pseudo-Labels Guided Data Augmentation for Semi-Supervised Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Teach me to segment with mixed supervision: Confident students become masters [article]

Preserved Fulltext

Self-Ensembling Contrastive Learning for Semi-Supervised Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Semantic-Aware Contrastive Learning for Multi-object Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Mixed-supervised segmentation: Confidence maximization helps knowledge distillation [article]

Preserved Fulltext

Other Versions

Exemplar Learning for Medical Image Segmentation [article]

Preserved Fulltext

Other Versions

Co-training with High-Confidence Pseudo Labels for Semi-supervised Medical Image Segmentation [article]

Preserved Fulltext

Semi-supervised Medical Image Segmentation through Dual-task Consistency

Preserved Fulltext

Information-guided pixel augmentation for pixel-wise contrastive learning [article]

Preserved Fulltext

Source-Relaxed Domain Adaptation for Image Segmentation [article]

Preserved Fulltext

Other Versions

Leveraging Labeling Representations in Uncertainty-based Semi-supervised Segmentation [article]

Preserved Fulltext

Semi-supervised Medical Image Segmentation through Dual-task Consistency [article]

Preserved Fulltext

Other Versions