Institute of Computing Technology, Chinese Academy IR
| Contour attention network for cerebrovascular segmentation from TOF-MRA volumetric images | |
| Yang, Chaozhi1; Zhang, Haiyan2; Chi, Dianwei3; Li, Yachuan1; Xiao, Qian1; Bai, Yun1; Li, Zongmin1,4,7; Li, Hongyi5; Li, Hua6 | |
| 2023-09-06 | |
| 发表期刊 | MEDICAL PHYSICS
 |
| ISSN | 0094-2405 |
| 页码 | 12 |
| 摘要 | BackgroundCerebrovascular segmentation is a crucial step in the computer-assisted diagnosis of cerebrovascular pathologies. However, accurate extraction of cerebral vessels from time-of-flight magnetic resonance angiography (TOF-MRA) data is still challenging due to the complex topology and slender shape.PurposeThe existing deep learning-based approaches pay more attention to the skeleton and ignore the contour, which limits the segmentation performance of the cerebrovascular structure. We aim to weight the contour of brain vessels in shallow features when concatenating with deep features. It helps to obtain more accurate cerebrovascular details and narrows the semantic gap between multilevel features.MethodsThis work proposes a novel framework for priority extraction of contours in cerebrovascular structures. We first design a neighborhood-based algorithm to generate the ground truth of the cerebrovascular contour from original annotations, which can introduce useful shape information for the segmentation network. Moreover, We propose an encoder-dual decoder-based contour attention network (CA-Net), which consists of the dilated asymmetry convolution block (DACB) and the Contour Attention Module (CAM). The ancillary decoder uses the DACB to obtain cerebrovascular contour features under the supervision of contour annotations. The CAM transforms these features into a spatial attention map to increase the weight of the contour voxels in main decoder to better restored the vessel contour details.ResultsThe CA-Net is thoroughly validated using two publicly available datasets, and the experimental results demonstrate that our network outperforms the competitors for cerebrovascular segmentation. We achieved the average dice similarity coefficient (DSC$DSC$) of 68.15 and 99.92% in natural and synthetic datasets. Our method segments cerebrovascular structures with better completeness.ConclusionsWe propose a new framework containing contour annotation generation and cerebrovascular segmentation network that better captures the tiny vessels and improve vessel connectivity. |
| 关键词 | 3D U-Net cerebrovascular segmentation contour attention network one-voxel-thick contour generation |
| DOI | 10.1002/mp.16720 |
| 收录类别 | SCI |
| 语种 | 英语 |
| 资助项目 | This work is partly supported by National Key Ramp;amp;D Program of China (Grant No. 2019YFF0301800), National Natural Science Foundation of China (Grant Nos. 61379106, 61806199), the General Research Projects of Beijing Educations Committee in China (Gra[61379106] ; National Key Ramp;amp;D Program of China[61806199] ; National Key Ramp;amp;D Program of China[KM201910005013] ; National Natural Science Foundation of China[ZR2015FM011] ; General Research Projects of Beijing Educations Committee in China ; Shandong Provincial Natural Science Foundation ; [2019YFF0301800] |
| WOS研究方向 | Radiology, Nuclear Medicine & Medical Imaging |
| WOS类目 | Radiology, Nuclear Medicine & Medical Imaging |
| WOS记录号 | WOS:001059200900001 |
| 出版者 | WILEY |
| 引用统计 | |
| 文献类型 | 期刊论文 |
| 条目标识符 | http://119.78.100.204/handle/2XEOYT63/21384 |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Li, Zongmin |
| 作者单位 | 1.China Univ Petr EastChina, Coll Comp Sci & Technol, Qingdao, Peoples R China 2.Weihai Chest Hosp, Weihai, Peoples R China 3.Yantai Inst Technol, Sch Artificial Intelligence, Yantai, Peoples R China 4.China Univ Petr, Shengli Coll, Dongying, Peoples R China 5.Chinese Acad Med Sci, Beijing Hosp, Inst Geriatr Med, Natl Ctr Gerontol, Beijing, Peoples R China 6.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing, Peoples R China 7.China Univ Petr EastChina, Coll Comp Sci & Technol, Qingdao 266580, Peoples R China |
| 推荐引用方式 GB/T 7714 | Yang, Chaozhi,Zhang, Haiyan,Chi, Dianwei,et al. Contour attention network for cerebrovascular segmentation from TOF-MRA volumetric images[J]. MEDICAL PHYSICS,2023:12. |
| APA | Yang, Chaozhi.,Zhang, Haiyan.,Chi, Dianwei.,Li, Yachuan.,Xiao, Qian.,...&Li, Hua.(2023).Contour attention network for cerebrovascular segmentation from TOF-MRA volumetric images.MEDICAL PHYSICS,12. |
| MLA | Yang, Chaozhi,et al."Contour attention network for cerebrovascular segmentation from TOF-MRA volumetric images".MEDICAL PHYSICS (2023):12. |
| 条目包含的文件 | 条目无相关文件。 | |||||
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论