File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/TMI.2023.3294824
- Scopus: eid_2-s2.0-85164687654
- PMID: 37436855
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: RECIST-Induced Reliable Learning: Geometry-Driven Label Propagation for Universal Lesion Segmentation
Title | RECIST-Induced Reliable Learning: Geometry-Driven Label Propagation for Universal Lesion Segmentation |
---|---|
Authors | |
Keywords | RECIST semi-supervised learning Universal lesion segmentation weakly-supervised learning |
Issue Date | 12-Jul-2023 |
Publisher | Institute of Electrical and Electronics Engineers |
Citation | IEEE Transactions on Medical Imaging, 2023, v. 43, n. 1, p. 149-161 How to Cite? |
Abstract | Automatic universal lesion segmentation (ULS) from Computed Tomography (CT) images can ease the burden of radiologists and provide a more accurate assessment than the current Response Evaluation Criteria In Solid Tumors (RECIST) guideline measurement. However, this task is underdeveloped due to the absence of large-scale pixel-wise labeled data. This paper presents a weakly-supervised learning framework to utilize the large-scale existing lesion databases in hospital Picture Archiving and Communication Systems (PACS) for ULS. Unlike previous methods to construct pseudo surrogate masks for fully supervised training through shallow interactive segmentation techniques, we propose to unearth the implicit information from RECIST annotations and thus design a unified RECIST-induced reliable learning (RiRL) framework. Particularly, we introduce a novel label generation procedure and an on-the-fly soft label propagation strategy to avoid noisy training and poor generalization problems. The former, named RECIST-induced geometric labeling, uses clinical characteristics of RECIST to preliminarily and reliably propagate the label. With the labeling process, a trimap divides the lesion slices into three regions, including certain foreground, background, and unclear regions, which consequently enables a strong and reliable supervision signal on a wide region. A topological knowledge-driven graph is built to conduct the on-the-fly label propagation for the optimal segmentation boundary to further optimize the segmentation boundary. Experimental results on a public benchmark dataset demonstrate that the proposed method surpasses the SOTA RECIST-based ULS methods by a large margin. Our approach surpasses SOTA approaches over 2.0%, 1.5%, 1.4%, and 1.6% Dice with ResNet101, ResNet50, HRNet, and ResNest50 backbones. |
Persistent Identifier | http://hdl.handle.net/10722/345766 |
ISSN | 2023 Impact Factor: 8.9 2023 SCImago Journal Rankings: 3.703 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Zhou, Lianyu | - |
dc.contributor.author | Yu, Lequan | - |
dc.contributor.author | Wang, Liansheng | - |
dc.date.accessioned | 2024-08-28T07:40:34Z | - |
dc.date.available | 2024-08-28T07:40:34Z | - |
dc.date.issued | 2023-07-12 | - |
dc.identifier.citation | IEEE Transactions on Medical Imaging, 2023, v. 43, n. 1, p. 149-161 | - |
dc.identifier.issn | 0278-0062 | - |
dc.identifier.uri | http://hdl.handle.net/10722/345766 | - |
dc.description.abstract | Automatic universal lesion segmentation (ULS) from Computed Tomography (CT) images can ease the burden of radiologists and provide a more accurate assessment than the current Response Evaluation Criteria In Solid Tumors (RECIST) guideline measurement. However, this task is underdeveloped due to the absence of large-scale pixel-wise labeled data. This paper presents a weakly-supervised learning framework to utilize the large-scale existing lesion databases in hospital Picture Archiving and Communication Systems (PACS) for ULS. Unlike previous methods to construct pseudo surrogate masks for fully supervised training through shallow interactive segmentation techniques, we propose to unearth the implicit information from RECIST annotations and thus design a unified RECIST-induced reliable learning (RiRL) framework. Particularly, we introduce a novel label generation procedure and an on-the-fly soft label propagation strategy to avoid noisy training and poor generalization problems. The former, named RECIST-induced geometric labeling, uses clinical characteristics of RECIST to preliminarily and reliably propagate the label. With the labeling process, a trimap divides the lesion slices into three regions, including certain foreground, background, and unclear regions, which consequently enables a strong and reliable supervision signal on a wide region. A topological knowledge-driven graph is built to conduct the on-the-fly label propagation for the optimal segmentation boundary to further optimize the segmentation boundary. Experimental results on a public benchmark dataset demonstrate that the proposed method surpasses the SOTA RECIST-based ULS methods by a large margin. Our approach surpasses SOTA approaches over 2.0%, 1.5%, 1.4%, and 1.6% Dice with ResNet101, ResNet50, HRNet, and ResNest50 backbones. | - |
dc.language | eng | - |
dc.publisher | Institute of Electrical and Electronics Engineers | - |
dc.relation.ispartof | IEEE Transactions on Medical Imaging | - |
dc.rights | This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. | - |
dc.subject | RECIST | - |
dc.subject | semi-supervised learning | - |
dc.subject | Universal lesion segmentation | - |
dc.subject | weakly-supervised learning | - |
dc.title | RECIST-Induced Reliable Learning: Geometry-Driven Label Propagation for Universal Lesion Segmentation | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/TMI.2023.3294824 | - |
dc.identifier.pmid | 37436855 | - |
dc.identifier.scopus | eid_2-s2.0-85164687654 | - |
dc.identifier.volume | 43 | - |
dc.identifier.issue | 1 | - |
dc.identifier.spage | 149 | - |
dc.identifier.epage | 161 | - |
dc.identifier.eissn | 1558-254X | - |
dc.identifier.issnl | 0278-0062 | - |