File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/TMI.2020.2966389
- Scopus: eid_2-s2.0-85085904736
- PMID: 31940526
- WOS: WOS:000544923000031
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: CT Male Pelvic Organ Segmentation via Hybrid Loss Network with Incomplete Annotation
Title | CT Male Pelvic Organ Segmentation via Hybrid Loss Network with Incomplete Annotation |
---|---|
Authors | |
Keywords | CT deep learning Image segmentation incomplete annotation male pelvic organ |
Issue Date | 2020 |
Citation | IEEE Transactions on Medical Imaging, 2020, v. 39, n. 6, p. 2151-2162 How to Cite? |
Abstract | Sufficient data with complete annotation is essential for training deep models to perform automatic and accurate segmentation of CT male pelvic organs, especially when such data is with great challenges such as low contrast and large shape variation. However, manual annotation is expensive in terms of both finance and human effort, which usually results in insufficient completely annotated data in real applications. To this end, we propose a novel deep framework to segment male pelvic organs in CT images with incomplete annotation delineated in a very user-friendly manner. Specifically, we design a hybrid loss network derived from both voxel classification and boundary regression, to jointly improve the organ segmentation performance in an iterative way. Moreover, we introduce a label completion strategy to complete the labels of the rich unannotated voxels and then embed them into the training data to enhance the model capability. To reduce the computation complexity and improve segmentation performance, we locate the pelvic region based on salient bone structures to focus on the candidate segmentation organs. Experimental results on a large planning CT pelvic organ dataset show that our proposed method with incomplete annotation achieves comparable segmentation performance to the state-of-the-art methods with complete annotation. Moreover, our proposed method requires much less effort of manual contouring from medical professionals such that an institutional specific model can be more easily established. |
Persistent Identifier | http://hdl.handle.net/10722/325478 |
ISSN | 2021 Impact Factor: 11.037 2020 SCImago Journal Rankings: 2.322 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wang, Shuai | - |
dc.contributor.author | Nie, Dong | - |
dc.contributor.author | Qu, Liangqiong | - |
dc.contributor.author | Shao, Yeqin | - |
dc.contributor.author | Lian, Jun | - |
dc.contributor.author | Wang, Qian | - |
dc.contributor.author | Shen, DInggang | - |
dc.date.accessioned | 2023-02-27T07:33:38Z | - |
dc.date.available | 2023-02-27T07:33:38Z | - |
dc.date.issued | 2020 | - |
dc.identifier.citation | IEEE Transactions on Medical Imaging, 2020, v. 39, n. 6, p. 2151-2162 | - |
dc.identifier.issn | 0278-0062 | - |
dc.identifier.uri | http://hdl.handle.net/10722/325478 | - |
dc.description.abstract | Sufficient data with complete annotation is essential for training deep models to perform automatic and accurate segmentation of CT male pelvic organs, especially when such data is with great challenges such as low contrast and large shape variation. However, manual annotation is expensive in terms of both finance and human effort, which usually results in insufficient completely annotated data in real applications. To this end, we propose a novel deep framework to segment male pelvic organs in CT images with incomplete annotation delineated in a very user-friendly manner. Specifically, we design a hybrid loss network derived from both voxel classification and boundary regression, to jointly improve the organ segmentation performance in an iterative way. Moreover, we introduce a label completion strategy to complete the labels of the rich unannotated voxels and then embed them into the training data to enhance the model capability. To reduce the computation complexity and improve segmentation performance, we locate the pelvic region based on salient bone structures to focus on the candidate segmentation organs. Experimental results on a large planning CT pelvic organ dataset show that our proposed method with incomplete annotation achieves comparable segmentation performance to the state-of-the-art methods with complete annotation. Moreover, our proposed method requires much less effort of manual contouring from medical professionals such that an institutional specific model can be more easily established. | - |
dc.language | eng | - |
dc.relation.ispartof | IEEE Transactions on Medical Imaging | - |
dc.subject | CT | - |
dc.subject | deep learning | - |
dc.subject | Image segmentation | - |
dc.subject | incomplete annotation | - |
dc.subject | male pelvic organ | - |
dc.title | CT Male Pelvic Organ Segmentation via Hybrid Loss Network with Incomplete Annotation | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/TMI.2020.2966389 | - |
dc.identifier.pmid | 31940526 | - |
dc.identifier.scopus | eid_2-s2.0-85085904736 | - |
dc.identifier.volume | 39 | - |
dc.identifier.issue | 6 | - |
dc.identifier.spage | 2151 | - |
dc.identifier.epage | 2162 | - |
dc.identifier.eissn | 1558-254X | - |
dc.identifier.isi | WOS:000544923000031 | - |