PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation
The structural morphology of mesenteric artery vessels is of significant importance for the diagnosis and treatment of colorectal cancer. However, developing automated vessel segmentation methods for this purpose remains challenging. Existing convolution-based segmentation methods have limitations i...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-12-01
|
| Series: | Frontiers in Physiology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphys.2023.1308987/full |
| _version_ | 1827589734571966464 |
|---|---|
| author | Kun Zhang Kun Zhang Kun Zhang Peixia Xu Meirong Wang Pengcheng Lin Danny Crookes Bosheng He Bosheng He Bosheng He Liang Hua |
| author_facet | Kun Zhang Kun Zhang Kun Zhang Peixia Xu Meirong Wang Pengcheng Lin Danny Crookes Bosheng He Bosheng He Bosheng He Liang Hua |
| author_sort | Kun Zhang |
| collection | DOAJ |
| description | The structural morphology of mesenteric artery vessels is of significant importance for the diagnosis and treatment of colorectal cancer. However, developing automated vessel segmentation methods for this purpose remains challenging. Existing convolution-based segmentation methods have limitations in capturing long-range dependencies, while transformer-based models require large datasets, making them less suitable for tasks with limited training samples. Moreover, over-segmentation, mis-segmentation, and vessel discontinuity are common challenges in vessel segmentation tasks. To address these issues, we propose a parallel encoding architecture that combines transformers and convolutions to retain the advantages of both approaches. The model effectively learns position deviations and enhances robustness for small-scale datasets. Additionally, we introduce a vessel edge capture module to improve vessel continuity and topology. Extensive experimental results demonstrate the improved performance of our model, with Dice Similarity Coefficient and Average Hausdorff Distance scores of 81.64% and 7.7428, respectively. |
| first_indexed | 2024-03-09T01:06:07Z |
| format | Article |
| id | doaj.art-dca5c2cac3434195b5c2250db4ef4b5d |
| institution | Directory Open Access Journal |
| issn | 1664-042X |
| language | English |
| last_indexed | 2024-03-09T01:06:07Z |
| publishDate | 2023-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physiology |
| spelling | doaj.art-dca5c2cac3434195b5c2250db4ef4b5d2023-12-11T10:02:03ZengFrontiers Media S.A.Frontiers in Physiology1664-042X2023-12-011410.3389/fphys.2023.13089871308987PE-Net: a parallel framework for 3D inferior mesenteric artery segmentationKun Zhang0Kun Zhang1Kun Zhang2Peixia Xu3Meirong Wang4Pengcheng Lin5Danny Crookes6Bosheng He7Bosheng He8Bosheng He9Liang Hua10School of Electrical Engineering, Nantong University, Nantong, Jiangsu, ChinaNantong Key Laboratory of Intelligent Control and Intelligent Computing, Nantong Institute of Technology, Nantong, Jiangsu, ChinaNantong Key Laboratory of Intelligent Medicine Innovation and Transformation, Affiliated Hospital 2 of Nantong University, Nantong, Jiangsu, ChinaSchool of Electrical Engineering, Nantong University, Nantong, Jiangsu, ChinaDepartment of Radiology, Affiliated Hospital 2 of Nantong University, Nantong, Jiangsu, ChinaSchool of Electrical Engineering, Nantong University, Nantong, Jiangsu, ChinaSchool of Electronics, Electrical Engineering and Computer Science, Queen’s University Belfast, Belfast, United KingdomNantong Key Laboratory of Intelligent Medicine Innovation and Transformation, Affiliated Hospital 2 of Nantong University, Nantong, Jiangsu, ChinaDepartment of Radiology, Affiliated Hospital 2 of Nantong University, Nantong, Jiangsu, ChinaClinical Medicine Research Center, Affiliated Hospital 2 of Nantong University, Nantong, Jiangsu, ChinaSchool of Electrical Engineering, Nantong University, Nantong, Jiangsu, ChinaThe structural morphology of mesenteric artery vessels is of significant importance for the diagnosis and treatment of colorectal cancer. However, developing automated vessel segmentation methods for this purpose remains challenging. Existing convolution-based segmentation methods have limitations in capturing long-range dependencies, while transformer-based models require large datasets, making them less suitable for tasks with limited training samples. Moreover, over-segmentation, mis-segmentation, and vessel discontinuity are common challenges in vessel segmentation tasks. To address these issues, we propose a parallel encoding architecture that combines transformers and convolutions to retain the advantages of both approaches. The model effectively learns position deviations and enhances robustness for small-scale datasets. Additionally, we introduce a vessel edge capture module to improve vessel continuity and topology. Extensive experimental results demonstrate the improved performance of our model, with Dice Similarity Coefficient and Average Hausdorff Distance scores of 81.64% and 7.7428, respectively.https://www.frontiersin.org/articles/10.3389/fphys.2023.1308987/fullvessel volumetransformeraxial attentionedge featureparallel encoding |
| spellingShingle | Kun Zhang Kun Zhang Kun Zhang Peixia Xu Meirong Wang Pengcheng Lin Danny Crookes Bosheng He Bosheng He Bosheng He Liang Hua PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation Frontiers in Physiology vessel volume transformer axial attention edge feature parallel encoding |
| title | PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation |
| title_full | PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation |
| title_fullStr | PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation |
| title_full_unstemmed | PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation |
| title_short | PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation |
| title_sort | pe net a parallel framework for 3d inferior mesenteric artery segmentation |
| topic | vessel volume transformer axial attention edge feature parallel encoding |
| url | https://www.frontiersin.org/articles/10.3389/fphys.2023.1308987/full |
| work_keys_str_mv | AT kunzhang penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT kunzhang penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT kunzhang penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT peixiaxu penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT meirongwang penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT pengchenglin penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT dannycrookes penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT boshenghe penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT boshenghe penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT boshenghe penetaparallelframeworkfor3dinferiormesentericarterysegmentation AT lianghua penetaparallelframeworkfor3dinferiormesentericarterysegmentation |