File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Transformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation

TitleTransformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation
Authors
KeywordsLiver segmentation
self-ensembling
optic disk (OD) segmentation
semisupervised learning
skin lesion segmentation
Issue Date2021
Citation
IEEE Transactions on Neural Networks and Learning Systems, 2021, v. 32, n. 2, p. 523-534 How to Cite?
AbstractA common shortfall of supervised deep learning for medical imaging is the lack of labeled data, which is often expensive and time consuming to collect. This article presents a new semisupervised method for medical image segmentation, where the network is optimized by a weighted combination of a common supervised loss only for the labeled inputs and a regularization loss for both the labeled and unlabeled data. To utilize the unlabeled data, our method encourages consistent predictions of the network-in-training for the same input under different perturbations. With the semisupervised segmentation tasks, we introduce a transformation-consistent strategy in the self-ensembling model to enhance the regularization effect for pixel-level predictions. To further improve the regularization effects, we extend the transformation in a more generalized form including scaling and optimize the consistency loss with a teacher model, which is an averaging of the student model weights. We extensively validated the proposed semisupervised method on three typical yet challenging medical image segmentation tasks: 1) skin lesion segmentation from dermoscopy images in the International Skin Imaging Collaboration (ISIC) 2017 data set; 2) optic disk (OD) segmentation from fundus images in the Retinal Fundus Glaucoma Challenge (REFUGE) data set; and 3) liver segmentation from volumetric CT scans in the Liver Tumor Segmentation Challenge (LiTS) data set. Compared with state-of-the-art, our method shows superior performance on the challenging 2-D/3-D medical images, demonstrating the effectiveness of our semisupervised method for medical image segmentation.
Persistent Identifierhttp://hdl.handle.net/10722/299489
ISSN
2023 Impact Factor: 10.2
2023 SCImago Journal Rankings: 4.170
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLi, Xiaomeng-
dc.contributor.authorYu, Lequan-
dc.contributor.authorChen, Hao-
dc.contributor.authorFu, Chi Wing-
dc.contributor.authorXing, Lei-
dc.contributor.authorHeng, Pheng Ann-
dc.date.accessioned2021-05-21T03:34:31Z-
dc.date.available2021-05-21T03:34:31Z-
dc.date.issued2021-
dc.identifier.citationIEEE Transactions on Neural Networks and Learning Systems, 2021, v. 32, n. 2, p. 523-534-
dc.identifier.issn2162-237X-
dc.identifier.urihttp://hdl.handle.net/10722/299489-
dc.description.abstractA common shortfall of supervised deep learning for medical imaging is the lack of labeled data, which is often expensive and time consuming to collect. This article presents a new semisupervised method for medical image segmentation, where the network is optimized by a weighted combination of a common supervised loss only for the labeled inputs and a regularization loss for both the labeled and unlabeled data. To utilize the unlabeled data, our method encourages consistent predictions of the network-in-training for the same input under different perturbations. With the semisupervised segmentation tasks, we introduce a transformation-consistent strategy in the self-ensembling model to enhance the regularization effect for pixel-level predictions. To further improve the regularization effects, we extend the transformation in a more generalized form including scaling and optimize the consistency loss with a teacher model, which is an averaging of the student model weights. We extensively validated the proposed semisupervised method on three typical yet challenging medical image segmentation tasks: 1) skin lesion segmentation from dermoscopy images in the International Skin Imaging Collaboration (ISIC) 2017 data set; 2) optic disk (OD) segmentation from fundus images in the Retinal Fundus Glaucoma Challenge (REFUGE) data set; and 3) liver segmentation from volumetric CT scans in the Liver Tumor Segmentation Challenge (LiTS) data set. Compared with state-of-the-art, our method shows superior performance on the challenging 2-D/3-D medical images, demonstrating the effectiveness of our semisupervised method for medical image segmentation.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Neural Networks and Learning Systems-
dc.subjectLiver segmentation-
dc.subjectself-ensembling-
dc.subjectoptic disk (OD) segmentation-
dc.subjectsemisupervised learning-
dc.subjectskin lesion segmentation-
dc.titleTransformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TNNLS.2020.2995319-
dc.identifier.pmid32479407-
dc.identifier.scopuseid_2-s2.0-85100701697-
dc.identifier.volume32-
dc.identifier.issue2-
dc.identifier.spage523-
dc.identifier.epage534-
dc.identifier.eissn2162-2388-
dc.identifier.isiWOS:000616310400006-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats