Discretely-constrained deep network for weakly supervised segmentation

Peng, Jizong; Kervadec, Hoel; Dolz, José; Ayed, Ismail Ben; Pedersoli, Marco; Desrosiers, Christian

doi:10.1016/j.neunet.2020.07.011

Cited by 37 publications

(17 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The incorporation of prior knowledge in deep learning methods is attracting more and more attention for improving performance of deep learning models. Some prior knowledge which act as constrained item have been proposed to integrate into loss functions [44], [45]. The participation of proper prior knowledge often bring performance improvement for models.…”

Section: Discussionmentioning

confidence: 99%

Self-Ensembling Co-Training Framework for Semi-Supervised COVID-19 CT Segmentation

Dou

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

The coronavirus disease 2019 (COVID-19) has become a severe worldwide health emergency and is spreading at a rapid rate. Segmentation of COVID lesions from computed tomography (CT) scans is of great importance for supervising disease progression and further clinical treatment. As labeling COVID-19 CT scans is laborintensive and time-consuming, it is essential to develop a segmentation method based on limited labeled data to conduct this task. In this paper, we propose a self-ensembled co-training framework, which is trained by limited labeled data and large-scale unlabeled data, to automatically extract COVID lesions from CT scans. Specifically, to enrich the diversity of unsupervised information, we build a co-training framework consisting of two collaborative models, in which the two models teach each other during training by using their respective predicted pseudo-labels of unlabeled data. Moreover, to alleviate the adverse impacts of noisy pseudo-labels for each model, we propose a selfensembling strategy to perform consistency regularization Manuscript

show abstract

Section: Discussionmentioning

confidence: 99%

Self-Ensembling Co-Training Framework for Semi-Supervised COVID-19 CT Segmentation

Dou

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

show abstract

“…Nevertheless, their formulation is limited to linear constraints. More recently, inequality constraints have been tackled by augmenting the learning objective with a penalty-based function, e.g., L 2 penalty, which can be imposed within a continuous optimization framework [5,18,19], or in the discrete domain [28]. Despite these methods have demonstrated remarkable performance in weakly supervised segmentation, they require that prior knowledge, exact or approximate, is given.…”

Section: Related Workmentioning

confidence: 99%

Looking at the whole picture: constrained unsupervised anomaly segmentation

Silva-Rodríguez¹,

Naranjo²,

Dolz³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Current unsupervised anomaly localization approaches rely on generative models to learn the distribution of normal images, which is later used to identify potential anomalous regions derived from errors on the reconstructed images. However, a main limitation of nearly all prior literature is the need of employing anomalous images to set a class-specific threshold to locate the anomalies. This limits their usability in realistic scenarios, where only normal data is typically accessible. Despite this major drawback, only a handful of works have addressed this limitation, by integrating supervision on attention maps during training. In this work, we propose a novel formulation that does not require accessing images with abnormalities to define the threshold. Furthermore, and in contrast to very recent work, the proposed constraint is formulated in a more principled manner, leveraging well-known knowledge in constrained optimization. In particular, the equality constraint on the attention maps in prior work is replaced by an inequality constraint, which allows more flexibility. In addition, to address the limitations of penalty-based functions we employ an extension of the popular log-barrier methods to handle the constraint. Comprehensive experiments on the popular BRATS'19 dataset demonstrate that the proposed approach substantially outperforms relevant literature, establishing new state-of-the-art results for unsupervised lesion segmentation. IntroductionUnder the supervised learning paradigm, deep learning models have achieved astonishing performance in a wide range of applications. Nevertheless, a main limitation of these models is the large amount of labeled data required for training. Obtaining such curated labeled datasets is a cumbersome process prone to subjectivity, which makes access to sufficient training data difficult in practice. This problem is further magnified in the context of medical image segmentation, where labeling involves assigning a category to each image pixel

show abstract

“…Additionally, following the literature in weakly supervised segmentation, we can obtain the corresponding image class activation maps (CAMs 1 ) from samples in D, resulting in the set 12 represents the max-normalized CAM of the i th sample and modality m k .…”

Section: Let Us Denote a Set Of N Training Images Asmentioning

confidence: 99%

“…[6], bounding boxes [7], [8] or global target information, such as object size [9]- [12]. A common strategy is to use imagelevel labels to derive pixel-wise class activation maps (CAMs) [13], which serve to identify object regions in the image.…”

Section: Introductionmentioning

confidence: 99%

Weakly supervised segmentation with cross-modality equivariant constraints

Patel¹,

Dolz²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Weakly supervised learning has emerged as an appealing alternative to alleviate the need for large labeled datasets in semantic segmentation. Most current approaches exploit class activation maps (CAMs), which can be generated from image-level annotations. Nevertheless, resulting maps have been demonstrated to be highly discriminant, failing to serve as optimal proxy pixel-level labels. We present a novel learning strategy that leverages self-supervision in a multi-modal image scenario to significantly enhance original CAMs. In particular, the proposed method is based on two observations. First, the learning of fully-supervised segmentation networks implicitly imposes equivariance by means of data augmentation, whereas this implicit constraint disappears on CAMs generated with image tags. And second, the commonalities between image modalities can be employed as an efficient self-supervisory signal, correcting the inconsistency shown by CAMs obtained across multiple modalities. To effectively train our model, we integrate a novel loss function that includes a within-modality and a cross-modality equivariant term to explicitly impose these constraints during training. In addition, we add a KL-divergence on the class prediction distributions to facilitate the information exchange between modalities, which, combined with the equivariant regularizers further improves the performance of our model. Exhaustive experiments on the popular multi-modal BRATS dataset demonstrate that our approach outperforms relevant recent literature under the same learning conditions.

show abstract

Discretely-constrained deep network for weakly supervised segmentation

Cited by 37 publications

References 25 publications

Self-Ensembling Co-Training Framework for Semi-Supervised COVID-19 CT Segmentation

Self-Ensembling Co-Training Framework for Semi-Supervised COVID-19 CT Segmentation

Looking at the whole picture: constrained unsupervised anomaly segmentation

Weakly supervised segmentation with cross-modality equivariant constraints

Contact Info

Product

Resources

About