Title | BCS-Net: Boundary, Context, and Semantic for Automatic COVID-19 Lung Infection Segmentation From CT Images |
Authors | Cong, Runmin Yang, Haowei Jiang, Qiuping Gao, Wei Li, Haisheng Wang, Cong Zhao, Yao Kwong, Sam |
Affiliation | Beijing Jiaotong Univ, Inst Informat Sci, Beijing 100044, Peoples R China Beijing Technol & Business Univ, Beijing Key Lab Big Data Technol Food Safety, Beijing 100048, Peoples R China City Univ Hong Kong, Dept Comp Sci, Hong Kong, Peoples R China Beijing Key Lab Adv Informat Sci & Network Techno, Beijing 100044, Peoples R China Ningbo Univ, Sch Informat Sci & Engn, Ningbo 315211, Peoples R China Peking Univ, Shenzhen Grad Sch, Sch Elect & Comp Engn, Shenzhen 518055, Peoples R China Peng Cheng Lab, Shenzhen 518055, Peoples R China Huawei Technol, Distributed & Parallel Software Lab, Shenzhen 518129, Peoples R China City Univ Hong Kong, Shenzhen Res Inst, Shenzhen 51800, Peoples R China |
Keywords | NETWORK |
Issue Date | 2022 |
Publisher | IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT |
Abstract | The spread of COVID-19 has brought a huge disaster to the world, and the automatic segmentation of infection regions can help doctors to make diagnosis quickly and reduce workload. However, there are several challenges for the accurate and complete segmentation, such as the scattered infection area distribution, complex background noises, and blurred segmentation boundaries. To this end, in this article, we propose a novel network for automatic COVID-19 lung infection segmentation from computed tomography (CT) images, named BCS-Net, which considers the boundary, context, and semantic attributes. The BCS-Net follows an encoder-decoder architecture, and more designs focus on the decoder stage that includes three progressively boundary-context-semantic reconstruction (BCSR) blocks. In each BCSR block, the attention-guided global context (AGGC) module is designed to learn the most valuable encoder features for decoder by highlighting the important spatial and boundary locations and modeling the global context dependence. Besides, a semantic guidance (SG) unit generates the SG map to refine the decoder features by aggregating multiscale high-level features at the intermediate resolution. Extensive experiments demonstrate that our proposed framework outperforms the existing competitors both qualitatively and quantitatively. |
URI | http://hdl.handle.net/20.500.11897/652471 |
ISSN | 0018-9456 |
DOI | 10.1109/TIM.2022.3196430 |
Indexed | SCI(E) |
Appears in Collections: | 深圳研究生院待认领 |