ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation
Identifying polyps is challenging for automatic analysis of endoscopic images in computer-aided clinical support systems. Models based on convolutional networks (CNN), transformers, and their combinations have been proposed to segment polyps with promising results. However, those approaches have lim...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9845389/ |
| _version_ | 1811309561361989632 |
|---|---|
| author | Nguyen Thanh Duc Nguyen Thi Oanh Nguyen Thi Thuy Tran Minh Triet Viet Sang Dinh |
| author_facet | Nguyen Thanh Duc Nguyen Thi Oanh Nguyen Thi Thuy Tran Minh Triet Viet Sang Dinh |
| author_sort | Nguyen Thanh Duc |
| collection | DOAJ |
| description | Identifying polyps is challenging for automatic analysis of endoscopic images in computer-aided clinical support systems. Models based on convolutional networks (CNN), transformers, and their combinations have been proposed to segment polyps with promising results. However, those approaches have limitations either in modeling the local appearance of the polyps only or lack of multi-level feature representation for spatial dependency in the decoding process. This paper proposes a novel network, namely ColonFormer, to address these limitations. ColonFormer is an encoder-decoder architecture capable of modeling long-range semantic information at both encoder and decoder branches. The encoder is a lightweight architecture based on transformers for modeling global semantic relations at multi scales. The decoder is a hierarchical network structure designed for learning multi-level features to enrich feature representation. Besides, a refinement module is added with a new skip connection technique to refine the boundary of polyp objects in the global map for accurate segmentation. Extensive experiments have been conducted on five popular benchmark datasets for polyp segmentation, including Kvasir, CVC-Clinic DB, CVC-ColonDB, CVC-T, and ETIS-Larib. Experimental results show that our ColonFormer outperforms other state-of-the-art methods on all benchmark datasets. Our code is available at: <uri>https://github.com/ducnt9907/ColonFormer</uri>. |
| first_indexed | 2024-04-13T09:44:06Z |
| format | Article |
| id | doaj.art-bcfc0be113934763b1117262d46e9b57 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-13T09:44:06Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-bcfc0be113934763b1117262d46e9b572022-12-22T02:51:49ZengIEEEIEEE Access2169-35362022-01-0110805758058610.1109/ACCESS.2022.31952419845389ColonFormer: An Efficient Transformer Based Method for Colon Polyp SegmentationNguyen Thanh Duc0Nguyen Thi Oanh1https://orcid.org/0000-0002-6166-2011Nguyen Thi Thuy2https://orcid.org/0000-0002-9358-6201Tran Minh Triet3https://orcid.org/0000-0003-3046-3041Viet Sang Dinh4https://orcid.org/0000-0002-9254-1327School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, VietnamSchool of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, VietnamFaculty of Information Technology and Software Engineering Laboratory, University of Science, VNU-HCM, Ho Chi Minh, VietnamUniversity of Science, VNU-HCM, Ho Chi Minh, VietnamSchool of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, VietnamIdentifying polyps is challenging for automatic analysis of endoscopic images in computer-aided clinical support systems. Models based on convolutional networks (CNN), transformers, and their combinations have been proposed to segment polyps with promising results. However, those approaches have limitations either in modeling the local appearance of the polyps only or lack of multi-level feature representation for spatial dependency in the decoding process. This paper proposes a novel network, namely ColonFormer, to address these limitations. ColonFormer is an encoder-decoder architecture capable of modeling long-range semantic information at both encoder and decoder branches. The encoder is a lightweight architecture based on transformers for modeling global semantic relations at multi scales. The decoder is a hierarchical network structure designed for learning multi-level features to enrich feature representation. Besides, a refinement module is added with a new skip connection technique to refine the boundary of polyp objects in the global map for accurate segmentation. Extensive experiments have been conducted on five popular benchmark datasets for polyp segmentation, including Kvasir, CVC-Clinic DB, CVC-ColonDB, CVC-T, and ETIS-Larib. Experimental results show that our ColonFormer outperforms other state-of-the-art methods on all benchmark datasets. Our code is available at: <uri>https://github.com/ducnt9907/ColonFormer</uri>.https://ieeexplore.ieee.org/document/9845389/Polyp segmentationdeep learningencoder-decoder networkhierarchical multi-scale CNNcomputer-aided diagnosis |
| spellingShingle | Nguyen Thanh Duc Nguyen Thi Oanh Nguyen Thi Thuy Tran Minh Triet Viet Sang Dinh ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation IEEE Access Polyp segmentation deep learning encoder-decoder network hierarchical multi-scale CNN computer-aided diagnosis |
| title | ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation |
| title_full | ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation |
| title_fullStr | ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation |
| title_full_unstemmed | ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation |
| title_short | ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation |
| title_sort | colonformer an efficient transformer based method for colon polyp segmentation |
| topic | Polyp segmentation deep learning encoder-decoder network hierarchical multi-scale CNN computer-aided diagnosis |
| url | https://ieeexplore.ieee.org/document/9845389/ |
| work_keys_str_mv | AT nguyenthanhduc colonformeranefficienttransformerbasedmethodforcolonpolypsegmentation AT nguyenthioanh colonformeranefficienttransformerbasedmethodforcolonpolypsegmentation AT nguyenthithuy colonformeranefficienttransformerbasedmethodforcolonpolypsegmentation AT tranminhtriet colonformeranefficienttransformerbasedmethodforcolonpolypsegmentation AT vietsangdinh colonformeranefficienttransformerbasedmethodforcolonpolypsegmentation |