Edge-enhancement cascaded network for lung lobe segmentation based on CT images
In order to reduce postoperative complications, it is required that the puncture needle should not pass through the lung lobe without tumor as far as possible in lung biopsy surgery. Therefore, it is necessary to accurately segment the lung lobe on the lung CT images. This paper proposed an automati...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-03-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2023.1098756/full |
| _version_ | 1827999279384363008 |
|---|---|
| author | Nan Bao Ye Yuan Ye Yuan Qingyao Luo Qiankun Li Qiankun Li Li-Bo Zhang |
| author_facet | Nan Bao Ye Yuan Ye Yuan Qingyao Luo Qiankun Li Qiankun Li Li-Bo Zhang |
| author_sort | Nan Bao |
| collection | DOAJ |
| description | In order to reduce postoperative complications, it is required that the puncture needle should not pass through the lung lobe without tumor as far as possible in lung biopsy surgery. Therefore, it is necessary to accurately segment the lung lobe on the lung CT images. This paper proposed an automatic lung lobe segmentation method on lung CT images. Considering the boundary of the lung lobe is difficult to be identified, our lung lobe segmentation network is designed to be a multi-stage cascade network based on edge enhancement. In the first stage, the anatomical features of the lung lobe are extracted based on the generative adversarial network (GAN), and the lung lobe boundary is Gaussian smoothed to generate the boundary response map. In the second stage, the CT images and the boundary response map are used as input, and the dense connection blocks are used to realize deep feature extraction, and finally five lung lobes are segmented. The experiments indicated that the average value of Dice coefficient is 0.9741, which meets the clinical needs. |
| first_indexed | 2024-04-10T05:57:18Z |
| format | Article |
| id | doaj.art-67636a98cac548f6aac1dba263acbc9b |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-04-10T05:57:18Z |
| publishDate | 2023-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-67636a98cac548f6aac1dba263acbc9b2023-03-03T11:41:14ZengFrontiers Media S.A.Frontiers in Physics2296-424X2023-03-011110.3389/fphy.2023.10987561098756Edge-enhancement cascaded network for lung lobe segmentation based on CT imagesNan Bao0Ye Yuan1Ye Yuan2Qingyao Luo3Qiankun Li4Qiankun Li5Li-Bo Zhang6College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, ChinaCollege of Computer Science and Engineering, Northeastern University, Shenyang, ChinaKey Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, ChinaCollege of Medicine and Biological Information Engineering, Northeastern University, Shenyang, ChinaInstitute of Intelligent Machines, Chinese Academy of Sciences, Hefei, ChinaDepartment of Automation, University of Science and Technology of China, Hefei, ChinaDepartment of Radiology, General Hospital of the Northern Theater of the Chinese People’s Liberation Army, Shenyang, ChinaIn order to reduce postoperative complications, it is required that the puncture needle should not pass through the lung lobe without tumor as far as possible in lung biopsy surgery. Therefore, it is necessary to accurately segment the lung lobe on the lung CT images. This paper proposed an automatic lung lobe segmentation method on lung CT images. Considering the boundary of the lung lobe is difficult to be identified, our lung lobe segmentation network is designed to be a multi-stage cascade network based on edge enhancement. In the first stage, the anatomical features of the lung lobe are extracted based on the generative adversarial network (GAN), and the lung lobe boundary is Gaussian smoothed to generate the boundary response map. In the second stage, the CT images and the boundary response map are used as input, and the dense connection blocks are used to realize deep feature extraction, and finally five lung lobes are segmented. The experiments indicated that the average value of Dice coefficient is 0.9741, which meets the clinical needs.https://www.frontiersin.org/articles/10.3389/fphy.2023.1098756/fulllung lobe segmentationCT imagesmulti-stage cascaded networkedge enhancementboundary response map |
| spellingShingle | Nan Bao Ye Yuan Ye Yuan Qingyao Luo Qiankun Li Qiankun Li Li-Bo Zhang Edge-enhancement cascaded network for lung lobe segmentation based on CT images Frontiers in Physics lung lobe segmentation CT images multi-stage cascaded network edge enhancement boundary response map |
| title | Edge-enhancement cascaded network for lung lobe segmentation based on CT images |
| title_full | Edge-enhancement cascaded network for lung lobe segmentation based on CT images |
| title_fullStr | Edge-enhancement cascaded network for lung lobe segmentation based on CT images |
| title_full_unstemmed | Edge-enhancement cascaded network for lung lobe segmentation based on CT images |
| title_short | Edge-enhancement cascaded network for lung lobe segmentation based on CT images |
| title_sort | edge enhancement cascaded network for lung lobe segmentation based on ct images |
| topic | lung lobe segmentation CT images multi-stage cascaded network edge enhancement boundary response map |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2023.1098756/full |
| work_keys_str_mv | AT nanbao edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT yeyuan edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT yeyuan edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT qingyaoluo edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT qiankunli edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT qiankunli edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages AT libozhang edgeenhancementcascadednetworkforlunglobesegmentationbasedonctimages |