Self-paced and self-consistent co-training for semi-supervised image segmentation

Wang, Ping; Peng, Jizong; Pedersoli, Marco; Zhou, Yuanfeng; Zhang, Caiming; Desrosiers, Christian

doi:10.1016/j.media.2021.102146

Cited by 50 publications

(13 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To improve the reliability of pseudo labels, one can adopt an iterative training procedure, which distills previously learned knowledge into a neural network with equal or larger capacity to boost model performance on label estimation (Xie et al, 2019b;Zoph et al, 2020). Also, it is interesting to introduce a pseudo label assessment module to select high quality pseudo labels for more effective uncertainty-aware consistency regularization (Xia et al, 2020a,b;Liu and Tan, 2021;Wang et al, 2021a). We will explore these extensions in future work.…”

Section: Supervised Learning With Extremely Low Data Settingsmentioning

confidence: 99%

Enhancing MR Image Segmentation with Realistic Adversarial Data Augmentation

Chen¹,

Chen²,

Ouyang³

et al. 2021

Preprint

View full text Add to dashboard Cite

The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose an adversarial data augmentation approach to improve the efficiency in utilizing training data and to enlarge the dataset via simulated but realistic transformations. Specifically, we present a generic task-driven learning framework, which jointly optimizes a data augmentation model and a segmentation network during training, generating informative examples to enhance network generalizability for the downstream task. The data augmentation model utilizes a set of photometric and geometric image transformations and chains them to simulate realistic complex imaging variations that could exist in magnetic resonance (MR) imaging. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications.

show abstract

Section: Supervised Learning With Extremely Low Data Settingsmentioning

confidence: 99%

Enhancing MR Image Segmentation with Realistic Adversarial Data Augmentation

Chen¹,

Chen²,

Ouyang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, the teacher-student strategy has been employed in model compression [1], to distil knowledge from multi-modal to monomodal segmentation networks [16], or in domain adaptation [40]. Semi-supervised segmentation has also benefited from teacher-student architectures [10,34,39]. In these approaches, however, the segmentation loss evaluating the consistency between the teacher and student models is computed on the unannotated data.…”

Section: Related Workmentioning

confidence: 99%

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

Dolz¹,

Desrosiers²,

Ayed³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Deep neural networks have achieved promising results in a breadth of medical image segmentation tasks. Nevertheless, they require large training datasets with pixel-wise segmentations, which are expensive to obtain in practice. Mixed supervision could mitigate this difficulty, with a small fraction of the data containing complete pixel-wise annotations, while the rest being less supervised, e.g., only a handful of pixels are labeled. In this work, we propose a dual-branch architecture, where the upper branch (teacher) receives strong annotations, while the bottom one (student) is driven by limited supervision and guided by the upper branch. 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 confident student predictions at the bottom branch; and (ii) a Kullback-Leibler (KL) divergence, which transfers the knowledge from the predictions generated by the strongly supervised branch to the less-supervised branch, and guides the entropy (student-confidence) term to avoid trivial solutions. Very interestingly, we show that the synergy between the entropy and KL divergence yields substantial improvements in performances. Furthermore, we discuss an interesting link between Shannon-entropy minimization and standard pseudo-mask generation, and argue that the former should be preferred over the latter for leveraging information from unlabeled pixels. Through a series of quantitative and qualitative experiments, we show the effectiveness of the proposed formulation in segmenting the left-ventricle endocardium in MRI images. We demonstrate that our method significantly outperforms other strategies to tackle semantic segmentation within a mixed-supervision framework. More interestingly, and in line with recent observations in classification, we show that the branch trained with reduced supervision and guided by the top branch largely outperforms the latter.

show abstract

“…To overcome the challenge of multi-stage optimization, one potential end-to-end training solution is collaborative training (co-training). As a common method in semi-supervised learning that jointly uses labeled and unlabeled data to improve the generalization of the model through collaboration among multiple learners [23], it has been broadly applied to image classification [24], target recognition [25] and image segmentation [26]. For medical image analysis, co-training has been adopted for semi-supervised leaning [27,28] or consistency learning in different views [25,29].…”

Section: Introductionmentioning

confidence: 99%

Label Propagation for Annotation-Efficient Nuclei Segmentation from Pathology Images

Lin¹,

Qu²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Nuclei segmentation is a crucial task for whole slide image analysis in digital pathology. Generally, the segmentation performance of fully-supervised learning heavily depends on the amount and quality of the annotated data. However, it is timeconsuming and expensive for professional pathologists to provide accurate pixel-level ground truth, while it is much easier to get coarse labels such as point annotations. In this paper, we propose a weakly-supervised learning method for nuclei segmentation that only requires point annotations for training. The proposed method achieves label propagation in a coarse-to-fine manner as follows. First, coarse pixel-level labels are derived from the point annotations based on the Voronoi diagram and the k-means clustering method to avoid overfitting. Second, a co-training strategy with an exponential moving average method is designed to refine the incomplete supervision of the coarse labels. Third, a self-supervised visual representation learning method is tailored for nuclei segmentation of pathology images that transforms the hematoxylin component images into the H&E stained images to gain better understanding of the relationship between the nuclei and cytoplasm. We comprehensively evaluate the proposed method using two public datasets. Both visual and quantitative results demonstrate the superiority of our method to the state-ofthe-art methods, and its competitive performance compared to the fully-supervised methods. The source codes for implementing the experiments will be released after acceptance.

show abstract

Self-paced and self-consistent co-training for semi-supervised image segmentation

Cited by 50 publications

References 33 publications

Enhancing MR Image Segmentation with Realistic Adversarial Data Augmentation

Enhancing MR Image Segmentation with Realistic Adversarial Data Augmentation

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

Label Propagation for Annotation-Efficient Nuclei Segmentation from Pathology Images

Contact Info

Product

Resources

About