File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Local and global structure-aware entropy regularized mean teacher model for 3d left atrium segmentation

TitleLocal and global structure-aware entropy regularized mean teacher model for 3d left atrium segmentation
Authors
KeywordsStructural consistency
Self-ensembling
Entropy minimization
Segmentation
Issue Date2020
PublisherSpringer.
Citation
23rd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2020), Lima, Peru, 4-8 October 2020. In Martel, AL, Abolmaesumi, P, Stoyanov, D, et al. (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part I, p. 562-571. Cham, Switzerland: Springer, 2020 How to Cite?
AbstractEmerging self-ensembling methods have achieved promising semi-supervised segmentation performances on medical images through forcing consistent predictions of unannotated data under different perturbations. However, the consistency only penalizes on independent pixel-level predictions, making structure-level information of predictions not exploited in the learning procedure. In view of this, we propose a novel structure-aware entropy regularized mean teacher model to address the above limitation. Specifically, we firstly introduce the entropy minimization principle to the student network, thereby adjusting itself to produce high-confident predictions of unannotated images. Based on this, we design a local structural consistency loss to encourage the consistency of inter-voxel similarities within the same local region of predictions from teacher and student networks. To further capture local structural dependencies, we enforce the global structural consistency by matching the weighted self-information maps between two networks. In this way, our model can minimize the prediction uncertainty of unannotated images, and more importantly that it can capture local and global structural information and their complementarity. We evaluate the proposed method on a publicly available 3D left atrium MR image dataset. Experimental results demonstrate that our method achieves outstanding segmentation performances than the state-of-the-art approaches in scenes with limited annotated images.
Persistent Identifierhttp://hdl.handle.net/10722/299475
ISBN
ISSN
2023 SCImago Journal Rankings: 0.606
Series/Report no.Lecture Notes in Computer Science ; 12261

 

DC FieldValueLanguage
dc.contributor.authorHang, Wenlong-
dc.contributor.authorFeng, Wei-
dc.contributor.authorLiang, Shuang-
dc.contributor.authorYu, Lequan-
dc.contributor.authorWang, Qiong-
dc.contributor.authorChoi, Kup Sze-
dc.contributor.authorQin, Jing-
dc.date.accessioned2021-05-21T03:34:29Z-
dc.date.available2021-05-21T03:34:29Z-
dc.date.issued2020-
dc.identifier.citation23rd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2020), Lima, Peru, 4-8 October 2020. In Martel, AL, Abolmaesumi, P, Stoyanov, D, et al. (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part I, p. 562-571. Cham, Switzerland: Springer, 2020-
dc.identifier.isbn9783030597092-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/299475-
dc.description.abstractEmerging self-ensembling methods have achieved promising semi-supervised segmentation performances on medical images through forcing consistent predictions of unannotated data under different perturbations. However, the consistency only penalizes on independent pixel-level predictions, making structure-level information of predictions not exploited in the learning procedure. In view of this, we propose a novel structure-aware entropy regularized mean teacher model to address the above limitation. Specifically, we firstly introduce the entropy minimization principle to the student network, thereby adjusting itself to produce high-confident predictions of unannotated images. Based on this, we design a local structural consistency loss to encourage the consistency of inter-voxel similarities within the same local region of predictions from teacher and student networks. To further capture local structural dependencies, we enforce the global structural consistency by matching the weighted self-information maps between two networks. In this way, our model can minimize the prediction uncertainty of unannotated images, and more importantly that it can capture local and global structural information and their complementarity. We evaluate the proposed method on a publicly available 3D left atrium MR image dataset. Experimental results demonstrate that our method achieves outstanding segmentation performances than the state-of-the-art approaches in scenes with limited annotated images.-
dc.languageeng-
dc.publisherSpringer.-
dc.relation.ispartofMedical Image Computing and Computer Assisted Intervention – MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part I-
dc.relation.ispartofseriesLecture Notes in Computer Science ; 12261-
dc.subjectStructural consistency-
dc.subjectSelf-ensembling-
dc.subjectEntropy minimization-
dc.subjectSegmentation-
dc.titleLocal and global structure-aware entropy regularized mean teacher model for 3d left atrium segmentation-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-030-59710-8_55-
dc.identifier.scopuseid_2-s2.0-85093091507-
dc.identifier.spage562-
dc.identifier.epage571-
dc.identifier.eissn1611-3349-
dc.publisher.placeCham, Switzerland-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats