EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network
According to medical research, colorectal polyps are considered typical precancerous lesions, making colonoscopic polyp images crucial for the early diagnosis of rectal cancer. However, variations in polyp size and shape, texture inconsistencies, and boundary ambiguities often pose significant chall...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10870365/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823857119135268864 |
|---|---|
| author | Xuhui Guan Jiwang Zhou Jian Chen Xiaodan Xu Yizhang Jiang Kaijian Xia |
| author_facet | Xuhui Guan Jiwang Zhou Jian Chen Xiaodan Xu Yizhang Jiang Kaijian Xia |
| author_sort | Xuhui Guan |
| collection | DOAJ |
| description | According to medical research, colorectal polyps are considered typical precancerous lesions, making colonoscopic polyp images crucial for the early diagnosis of rectal cancer. However, variations in polyp size and shape, texture inconsistencies, and boundary ambiguities often pose significant challenges for polyp segmentation. To address these issues, we propose a multi-scale feature fusion network based on edge enhancement. Specifically, we utilize multi-scale feature fusion in each feature layer, where the extracted features are fused. From these extracted features, we generate a global mapping graph as a bootstrap region. Additionally, we introduce Spatial Channel Convolution (SCEConv) and Reverse Gated Channel Transformer (RGCT) to incorporate boundary information into the segmentation network. This approach enhances the layered features and produces a more refined segmentation map. Extensive qualitative and quantitative experiments on five benchmark datasets and two private datasets demonstrate that the EMFF-Net proposed in this paper significantly improves segmentation accuracy across six metrics. This represents a clear advantage over traditional CNNs and existing SOTA techniques, Especially, we achieved 81% mDice and 74% mIoU on the CVC-ColonDB dataset. |
| format | Article |
| id | doaj-art-2fa417030cf640c2bb07e060b791ab90 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-2fa417030cf640c2bb07e060b791ab902025-02-12T00:02:23ZengIEEEIEEE Access2169-35362025-01-0113255982561110.1109/ACCESS.2025.353850410870365EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion NetworkXuhui Guan0Jiwang Zhou1Jian Chen2https://orcid.org/0009-0001-9930-878XXiaodan Xu3Yizhang Jiang4https://orcid.org/0000-0002-4558-9803Kaijian Xia5https://orcid.org/0000-0002-1650-9982School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu, ChinaULisboa School, Shanghai University, Shanghai, ChinaChangshu Key Laboratory of Medical Artificial Intelligence and Big Data, Suzhou, Jiangsu, ChinaChangshu Key Laboratory of Medical Artificial Intelligence and Big Data, Suzhou, Jiangsu, ChinaSchool of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu, ChinaChangshu Key Laboratory of Medical Artificial Intelligence and Big Data, Suzhou, Jiangsu, ChinaAccording to medical research, colorectal polyps are considered typical precancerous lesions, making colonoscopic polyp images crucial for the early diagnosis of rectal cancer. However, variations in polyp size and shape, texture inconsistencies, and boundary ambiguities often pose significant challenges for polyp segmentation. To address these issues, we propose a multi-scale feature fusion network based on edge enhancement. Specifically, we utilize multi-scale feature fusion in each feature layer, where the extracted features are fused. From these extracted features, we generate a global mapping graph as a bootstrap region. Additionally, we introduce Spatial Channel Convolution (SCEConv) and Reverse Gated Channel Transformer (RGCT) to incorporate boundary information into the segmentation network. This approach enhances the layered features and produces a more refined segmentation map. Extensive qualitative and quantitative experiments on five benchmark datasets and two private datasets demonstrate that the EMFF-Net proposed in this paper significantly improves segmentation accuracy across six metrics. This represents a clear advantage over traditional CNNs and existing SOTA techniques, Especially, we achieved 81% mDice and 74% mIoU on the CVC-ColonDB dataset.https://ieeexplore.ieee.org/document/10870365/Polyp segmentationedge-enhancementmulti-scale feature fusionboundary information |
| spellingShingle | Xuhui Guan Jiwang Zhou Jian Chen Xiaodan Xu Yizhang Jiang Kaijian Xia EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network IEEE Access Polyp segmentation edge-enhancement multi-scale feature fusion boundary information |
| title | EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network |
| title_full | EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network |
| title_fullStr | EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network |
| title_full_unstemmed | EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network |
| title_short | EMFF-Net: Edge-Enhancement Multi-Scale Feature Fusion Network |
| title_sort | emff net edge enhancement multi scale feature fusion network |
| topic | Polyp segmentation edge-enhancement multi-scale feature fusion boundary information |
| url | https://ieeexplore.ieee.org/document/10870365/ |
| work_keys_str_mv | AT xuhuiguan emffnetedgeenhancementmultiscalefeaturefusionnetwork AT jiwangzhou emffnetedgeenhancementmultiscalefeaturefusionnetwork AT jianchen emffnetedgeenhancementmultiscalefeaturefusionnetwork AT xiaodanxu emffnetedgeenhancementmultiscalefeaturefusionnetwork AT yizhangjiang emffnetedgeenhancementmultiscalefeaturefusionnetwork AT kaijianxia emffnetedgeenhancementmultiscalefeaturefusionnetwork |