Zhaoyu Hu1, Zhenhua Liu2, Yijie Dong2, Jianjian Liu3, Bin Huang4, Aihua Liu5, Jingjing Huang3, Xujuan Pu3, Xia Shi3, Jinhua Yu1, Yang Xiao6, Hui Zhang7, Jianqiao Zhou8. 1. Department of Electronic Engineering, Fudan University, Shanghai, 200433, China. 2. Department of Ultrasound, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, 200025, China. 3. Department of Ultrasound, Shanghai Public Health Clinical Center, Shanghai, 201508, China. 4. Department of Ultrasound, Xixi Hospital of Hangzhou, Hangzhou, 310023, China. 5. Department of Ultrasound, The Six Hospital of Wuhan, Affiliated Hospital of Jianghang University, Wuhan, 430015, China. 6. Institute of Biomedical and Health Engineering Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, 440305, China. yang.xiao@siat.ac.cn. 7. Department of Ultrasound, Shanghai Public Health Clinical Center, Shanghai, 201508, China. zhang.hui@zs-hospital.sh.cn. 8. Department of Ultrasound, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, 200025, China. zhousu30@126.com.
Abstract
BACKGROUND: Lung ultrasound (LUS) can be an important imaging tool for the diagnosis and assessment of lung involvement. Ultrasound sonograms have been confirmed to illustrate damage to a person's lungs, which means that the correct classification and scoring of a patient's sonogram can be used to assess lung involvement. METHODS: The purpose of this study was to establish a lung involvement assessment model based on deep learning. A novel multimodal channel and receptive field attention network combined with ResNeXt (MCRFNet) was proposed to classify sonograms, and the network can automatically fuse shallow features and determine the importance of different channels and respective fields. Finally, sonogram classes were transformed into scores to evaluate lung involvement from the initial diagnosis to rehabilitation. RESULTS AND CONCLUSION: Using multicenter and multimodal ultrasound data from 104 patients, the diagnostic model achieved 94.39% accuracy, 82.28% precision, 76.27% sensitivity, and 96.44% specificity. The lung involvement severity and the trend of COVID-19 pneumonia were evaluated quantitatively.
BACKGROUND: Lung ultrasound (LUS) can be an important imaging tool for the diagnosis and assessment of lung involvement. Ultrasound sonograms have been confirmed to illustrate damage to a person's lungs, which means that the correct classification and scoring of a patient's sonogram can be used to assess lung involvement. METHODS: The purpose of this study was to establish a lung involvement assessment model based on deep learning. A novel multimodal channel and receptive field attention network combined with ResNeXt (MCRFNet) was proposed to classify sonograms, and the network can automatically fuse shallow features and determine the importance of different channels and respective fields. Finally, sonogram classes were transformed into scores to evaluate lung involvement from the initial diagnosis to rehabilitation. RESULTS AND CONCLUSION: Using multicenter and multimodal ultrasound data from 104 patients, the diagnostic model achieved 94.39% accuracy, 82.28% precision, 76.27% sensitivity, and 96.44% specificity. The lung involvement severity and the trend of COVID-19 pneumonia were evaluated quantitatively.
Authors: Giovanni Volpicelli; Mahmoud Elbarbary; Michael Blaivas; Daniel A Lichtenstein; Gebhard Mathis; Andrew W Kirkpatrick; Lawrence Melniker; Luna Gargani; Vicki E Noble; Gabriele Via; Anthony Dean; James W Tsung; Gino Soldati; Roberto Copetti; Belaid Bouhemad; Angelika Reissig; Eustachio Agricola; Jean-Jacques Rouby; Charlotte Arbelot; Andrew Liteplo; Ashot Sargsyan; Fernando Silva; Richard Hoppmann; Raoul Breitkreutz; Armin Seibel; Luca Neri; Enrico Storti; Tomislav Petrovic Journal: Intensive Care Med Date: 2012-03-06 Impact factor: 17.440
Authors: Jean-Jacques Rouby; Charlotte Arbelot; Yuzhi Gao; Mao Zhang; Jie Lv; Youzhong An; Wang Chunyao; Du Bin; Carmen Silvia Valente Barbas; Felippe Leopoldo Dexheimer Neto; Fabiola Prior Caltabeloti; Emidio Lima; Andres Cebey; Sébastien Perbet; Jean-Michel Constantin Journal: Am J Respir Crit Care Med Date: 2018-08-01 Impact factor: 21.405
Authors: Adam Bernheim; Xueyan Mei; Mingqian Huang; Yang Yang; Zahi A Fayad; Ning Zhang; Kaiyue Diao; Bin Lin; Xiqi Zhu; Kunwei Li; Shaolin Li; Hong Shan; Adam Jacobi; Michael Chung Journal: Radiology Date: 2020-02-20 Impact factor: 11.105
Authors: Miguel José Francisco; Antonio Rahal; Fabio Augusto Cardillo Vieira; Paulo Savoia Dias da Silva; Marcelo Buarque de Gusmão Funari Journal: Einstein (Sao Paulo) Date: 2016 Jul-Sep
Authors: Jing Wang; Xiaofeng Yang; Boran Zhou; James J Sohn; Jun Zhou; Jesse T Jacob; Kristin A Higgins; Jeffrey D Bradley; Tian Liu Journal: J Imaging Date: 2022-03-05