COVID-19 CT ground-glass opacity segmentation based on attention mechanism threshold

作者全名："Rao, Yunbo; Lv, Qingsong; Zeng, Shaoning; Yi, Yuling; Huang, Cheng; Gao, Yun; Cheng, Zhanglin; Sun, Jihong"

作者地址："[Rao, Yunbo; Lv, Qingsong; Yi, Yuling] Univ Elect Sci & Technol China, Sch Informat & Software Engn, Chengdu 611731, Peoples R China; [Zeng, Shaoning] Univ Elect Sci & Technol China, Yangtze Delta Reg Inst Huzhou, Huzhou 313000, Peoples R China; [Huang, Cheng] Chongqing Med Univ, Clin Coll 5, Chongqing 402177, Peoples R China; [Gao, Yun] Chongqing Univ Posts & Telecommun, Chongqing 400065, Peoples R China; [Cheng, Zhanglin] Adv Technol Chinese Acad Sci, Shenzhen 610042, Peoples R China; [Sun, Jihong] Zhejiang Univ, Sir Run Run Shaw Hosp, Sch Med, Hangzhou 310014, Peoples R China"

通信作者："Zeng, SN (通讯作者)，Univ Elect Sci & Technol China, Yangtze Delta Reg Inst Huzhou, Huzhou 313000, Peoples R China."

来源：BIOMEDICAL SIGNAL PROCESSING AND CONTROL

ESI学科分类：ENGINEERING

WOS号：WOS:000898953000007

JCR分区：Q1

影响因子：4.9

年份：2023

卷号：81

期号：　

开始页：　

结束页：　

文献类型：Article

关键词：COVID-19; GGO segmentation; Attention mechanism; Adaptive threshold; Pneumonia; CT image

摘要："The ground glass opacity (GGO) of the lung is one of the essential features of COVID-19. The GGO in computed tomography (CT) images has various features and low-intensity contrast between the GGO and edge structures. These problems pose significant challenges for segmenting the GGO. To tackle these problems, we propose a new threshold method for accurate segmentation of GGO. Specifically, we offer a framework for adjusting the threshold parameters according to the image contrast. Three functions include Attention mechanism threshold, Contour equalization, and Lung segmentation (ACL). The lung is divided into three areas using the attention mechanism threshold. Further, the segmentation parameters of the attention mechanism thresholds of the three parts are adaptively adjusted according to the image contrast. Only the segmentation regions restricted by the lung segmentation results are retained. Extensive experiments on four COVID datasets show that ACL can segment GGO images at low contrast well. Compared with the state-of-the-art methods, the similarity Dice of the ACL segmentation results is improved by 8.9%, the average symmetry surface distance ASD is reduced by 23%, and the required computational power FLOP s are only 0.09% of those of deep learning models. For GGO segmentation, ACL is more lightweight, and the accuracy is higher. Code will be released at https://github.com/Lqs-github/ACL."

基金机构："Science and Technology Project of Sichuan, China; National Natural Science Foundation of China; [2022ZHCG0033]; [2021YFG0314]; [2020YFG0459]; [U19A2078]"

基金资助正文："Acknowledgements This research was supported by the Science and Technology Project of Sichuan, China (Grant NOs. 2022ZHCG0033, 2021YFG0314, 2020YFG0459) , and the National Natural Science Foundation of China (Grant NO. U19A2078) ."