tetano
Editor, Senior Moderator
Sci Rep
. 2020 Nov 5;10(1):19196.
doi: 10.1038/s41598-020-76282-0.
Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography
Jun Chen[SUP] 1 [/SUP], Lianlian Wu[SUP] 2 3 4 [/SUP], Jun Zhang[SUP] 2 3 4 [/SUP], Liang Zhang[SUP] 1 [/SUP], Dexin Gong[SUP] 2 3 4 [/SUP], Yilin Zhao[SUP] 1 [/SUP], Qiuxiang Chen[SUP] 5 [/SUP], Shulan Huang[SUP] 5 [/SUP], Ming Yang[SUP] 5 [/SUP], Xiao Yang[SUP] 5 [/SUP], Shan Hu[SUP] 6 [/SUP], Yonggui Wang[SUP] 7 [/SUP], Xiao Hu[SUP] 6 [/SUP], Biqing Zheng[SUP] 6 [/SUP], Kuo Zhang[SUP] 6 [/SUP], Huiling Wu[SUP] 2 3 4 [/SUP], Zehua Dong[SUP] 2 3 4 [/SUP], Youming Xu[SUP] 2 3 4 [/SUP], Yijie Zhu[SUP] 2 3 4 [/SUP], Xi Chen[SUP] 2 3 4 [/SUP], Mengjiao Zhang[SUP] 2 [/SUP], Lilei Yu[SUP] 8 [/SUP], Fan Cheng[SUP] 9 [/SUP], Honggang Yu[SUP] 10 11 12 [/SUP]
Affiliations
Abstract
Computed tomography (CT) is the preferred imaging method for diagnosing 2019 novel coronavirus (COVID19) pneumonia. We aimed to construct a system based on deep learning for detecting COVID-19 pneumonia on high resolution CT. For model development and validation, 46,096 anonymous images from 106 admitted patients, including 51 patients of laboratory confirmed COVID-19 pneumonia and 55 control patients of other diseases in Renmin Hospital of Wuhan University were retrospectively collected. Twenty-seven prospective consecutive patients in Renmin Hospital of Wuhan University were collected to evaluate the efficiency of radiologists against 2019-CoV pneumonia with that of the model. An external test was conducted in Qianjiang Central Hospital to estimate the system's robustness. The model achieved a per-patient accuracy of 95.24% and a per-image accuracy of 98.85% in internal retrospective dataset. For 27 internal prospective patients, the system achieved a comparable performance to that of expert radiologist. In external dataset, it achieved an accuracy of 96%. With the assistance of the model, the reading time of radiologists was greatly decreased by 65%. The deep learning model showed a comparable performance with expert radiologist, and greatly improved the efficiency of radiologists in clinical practice.
. 2020 Nov 5;10(1):19196.
doi: 10.1038/s41598-020-76282-0.
Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography
Jun Chen[SUP] 1 [/SUP], Lianlian Wu[SUP] 2 3 4 [/SUP], Jun Zhang[SUP] 2 3 4 [/SUP], Liang Zhang[SUP] 1 [/SUP], Dexin Gong[SUP] 2 3 4 [/SUP], Yilin Zhao[SUP] 1 [/SUP], Qiuxiang Chen[SUP] 5 [/SUP], Shulan Huang[SUP] 5 [/SUP], Ming Yang[SUP] 5 [/SUP], Xiao Yang[SUP] 5 [/SUP], Shan Hu[SUP] 6 [/SUP], Yonggui Wang[SUP] 7 [/SUP], Xiao Hu[SUP] 6 [/SUP], Biqing Zheng[SUP] 6 [/SUP], Kuo Zhang[SUP] 6 [/SUP], Huiling Wu[SUP] 2 3 4 [/SUP], Zehua Dong[SUP] 2 3 4 [/SUP], Youming Xu[SUP] 2 3 4 [/SUP], Yijie Zhu[SUP] 2 3 4 [/SUP], Xi Chen[SUP] 2 3 4 [/SUP], Mengjiao Zhang[SUP] 2 [/SUP], Lilei Yu[SUP] 8 [/SUP], Fan Cheng[SUP] 9 [/SUP], Honggang Yu[SUP] 10 11 12 [/SUP]
Affiliations
- PMID: 33154542
- DOI: 10.1038/s41598-020-76282-0
Abstract
Computed tomography (CT) is the preferred imaging method for diagnosing 2019 novel coronavirus (COVID19) pneumonia. We aimed to construct a system based on deep learning for detecting COVID-19 pneumonia on high resolution CT. For model development and validation, 46,096 anonymous images from 106 admitted patients, including 51 patients of laboratory confirmed COVID-19 pneumonia and 55 control patients of other diseases in Renmin Hospital of Wuhan University were retrospectively collected. Twenty-seven prospective consecutive patients in Renmin Hospital of Wuhan University were collected to evaluate the efficiency of radiologists against 2019-CoV pneumonia with that of the model. An external test was conducted in Qianjiang Central Hospital to estimate the system's robustness. The model achieved a per-patient accuracy of 95.24% and a per-image accuracy of 98.85% in internal retrospective dataset. For 27 internal prospective patients, the system achieved a comparable performance to that of expert radiologist. In external dataset, it achieved an accuracy of 96%. With the assistance of the model, the reading time of radiologists was greatly decreased by 65%. The deep learning model showed a comparable performance with expert radiologist, and greatly improved the efficiency of radiologists in clinical practice.