Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images
Cerebrovascular diseases are a widespread threat to human health. The accurate extraction of cerebral vessel structures is of paramount importance in the diagnosis and treatment of cerebrovascular diseases. However, the complexity of cerebral vessel structures and the low imaging contrast present si...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-06-01
|
| Series: | EngMedicine |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2950489924000046 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841545922366930944 |
|---|---|
| author | Tao Han Junchen Xiong Tingyi Lin Tao An Cheng Wang Jianjun Zhu Zhongliang Li Ligong Lu Yi Zhang Gao-Jun Teng |
| author_facet | Tao Han Junchen Xiong Tingyi Lin Tao An Cheng Wang Jianjun Zhu Zhongliang Li Ligong Lu Yi Zhang Gao-Jun Teng |
| author_sort | Tao Han |
| collection | DOAJ |
| description | Cerebrovascular diseases are a widespread threat to human health. The accurate extraction of cerebral vessel structures is of paramount importance in the diagnosis and treatment of cerebrovascular diseases. However, the complexity of cerebral vessel structures and the low imaging contrast present significant challenges for vessel segmentation. Therefore, we propose a Multiscale Attention Network based on topological learning to extract vessel structures from angiographic images. This method employs a Multiscale Squeeze Attention (MSA) module for channel-wise attention learning, extracting multiscale attention feature maps from angiographic images. To maintain the topological connectivity of vessel segmentation, we introduced the clDice loss function to enforce skeleton connectivity of vessel segmentation. We conducted an experimental analysis of the proposed method using a publicly available cerebral vessel dataset. The results demonstrated that the proposed method achieved a sensitivity score of 0.8507 and a dice score of 0.8669 for cerebrovascular segmentation, enabling accurate and complete extraction of vascular structures. The proposed method was extended to coronary angiography images. The results show that the proposed method can accurately extract coronary structures, proving its broad applicability to other vascular segmentation tasks. |
| format | Article |
| id | doaj-art-98659e167beb4b33a9b2433fdc4330af |
| institution | Kabale University |
| issn | 2950-4899 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | EngMedicine |
| spelling | doaj-art-98659e167beb4b33a9b2433fdc4330af2025-01-11T06:42:26ZengElsevierEngMedicine2950-48992024-06-0111100004Multiscale attention network via topology learning for cerebral vessel segmentation in angiography imagesTao Han0Junchen Xiong1Tingyi Lin2Tao An3Cheng Wang4Jianjun Zhu5Zhongliang Li6Ligong Lu7Yi Zhang8Gao-Jun Teng9Hanglok-Tech Co., Ltd., Hengqin, Guangdong, China; Center of Interventional Radiology and Vascular Surgery, Department of Radiology, Zhongda Hospital, Medical School, Southeast University, Nanjing, ChinaHanglok-Tech Co., Ltd., Hengqin, Guangdong, ChinaHanglok-Tech Co., Ltd., Hengqin, Guangdong, ChinaZhuhai People's Hospital, Zhuhai, Guangdong, ChinaHanglok-Tech Co., Ltd., Hengqin, Guangdong, ChinaHanglok-Tech Co., Ltd., Hengqin, Guangdong, China; Center of Interventional Radiology and Vascular Surgery, Department of Radiology, Zhongda Hospital, Medical School, Southeast University, Nanjing, ChinaZhuhai People's Hospital, Zhuhai, Guangdong, ChinaZhuhai People's Hospital, Zhuhai, Guangdong, ChinaCenter of Interventional Radiology and Vascular Surgery, Department of Radiology, Zhongda Hospital, Medical School, Southeast University, Nanjing, China; Corresponding author.Center of Interventional Radiology and Vascular Surgery, Department of Radiology, Zhongda Hospital, Medical School, Southeast University, Nanjing, China; Corresponding author.Cerebrovascular diseases are a widespread threat to human health. The accurate extraction of cerebral vessel structures is of paramount importance in the diagnosis and treatment of cerebrovascular diseases. However, the complexity of cerebral vessel structures and the low imaging contrast present significant challenges for vessel segmentation. Therefore, we propose a Multiscale Attention Network based on topological learning to extract vessel structures from angiographic images. This method employs a Multiscale Squeeze Attention (MSA) module for channel-wise attention learning, extracting multiscale attention feature maps from angiographic images. To maintain the topological connectivity of vessel segmentation, we introduced the clDice loss function to enforce skeleton connectivity of vessel segmentation. We conducted an experimental analysis of the proposed method using a publicly available cerebral vessel dataset. The results demonstrated that the proposed method achieved a sensitivity score of 0.8507 and a dice score of 0.8669 for cerebrovascular segmentation, enabling accurate and complete extraction of vascular structures. The proposed method was extended to coronary angiography images. The results show that the proposed method can accurately extract coronary structures, proving its broad applicability to other vascular segmentation tasks.http://www.sciencedirect.com/science/article/pii/S2950489924000046Cerebrovascular diseaseDigital subtraction angiographyVessel segmentationDeep learning |
| spellingShingle | Tao Han Junchen Xiong Tingyi Lin Tao An Cheng Wang Jianjun Zhu Zhongliang Li Ligong Lu Yi Zhang Gao-Jun Teng Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images EngMedicine Cerebrovascular disease Digital subtraction angiography Vessel segmentation Deep learning |
| title | Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| title_full | Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| title_fullStr | Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| title_full_unstemmed | Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| title_short | Multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| title_sort | multiscale attention network via topology learning for cerebral vessel segmentation in angiography images |
| topic | Cerebrovascular disease Digital subtraction angiography Vessel segmentation Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2950489924000046 |
| work_keys_str_mv | AT taohan multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT junchenxiong multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT tingyilin multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT taoan multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT chengwang multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT jianjunzhu multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT zhongliangli multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT ligonglu multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT yizhang multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages AT gaojunteng multiscaleattentionnetworkviatopologylearningforcerebralvesselsegmentationinangiographyimages |