tetano
Editor, Senior Moderator
Nat Biomed Eng
. 2021 Apr 15.
doi: 10.1038/s41551-021-00704-1. Online ahead of print.
A deep-learning pipeline for the diagnosis and discrimination of viral, non-viral and COVID-19 pneumonia from chest X-ray images
Guangyu Wang[SUP] #[/SUP][SUP] 1 [/SUP], Xiaohong Liu[SUP] #[/SUP][SUP] 2 [/SUP], Jun Shen[SUP] #[/SUP][SUP] 3 [/SUP], Chengdi Wang[SUP] #[/SUP][SUP] 4 [/SUP], Zhihuan Li[SUP] #[/SUP][SUP] 5 [/SUP], Linsen Ye[SUP] #[/SUP][SUP] 6 [/SUP], Xingwang Wu[SUP] #[/SUP][SUP] 7 [/SUP], Ting Chen[SUP] 8 [/SUP], Kai Wang[SUP] 2 [/SUP], Xuan Zhang[SUP] 2 [/SUP], Zhongguo Zhou[SUP] 9 [/SUP], Jian Yang[SUP] 10 [/SUP], Ye Sang[SUP] 10 [/SUP], Ruiyun Deng[SUP] 11 [/SUP], Wenhua Liang[SUP] 12 [/SUP], Tao Yu[SUP] 3 [/SUP], Ming Gao[SUP] 3 [/SUP], Jin Wang[SUP] 6 [/SUP], Zehong Yang[SUP] 3 [/SUP], Huimin Cai[SUP] 11 [/SUP], Guangming Lu[SUP] 13 [/SUP], Lingyan Zhang[SUP] 14 [/SUP], Lei Yang[SUP] 15 [/SUP], Wenqin Xu[SUP] 5 [/SUP], Winston Wang[SUP] 5 [/SUP], Andrea Olevera[SUP] 5 [/SUP], Ian Ziyar[SUP] 5 [/SUP], Charlotte Zhang[SUP] 11 [/SUP], Oulan Li[SUP] 11 [/SUP], Weihua Liao[SUP] 16 [/SUP], Jun Liu[SUP] 17 [/SUP], Wen Chen[SUP] 18 [/SUP], Wei Chen[SUP] 19 [/SUP], Jichan Shi[SUP] 20 [/SUP], Lianghong Zheng[SUP] 5 [/SUP], Longjiang Zhang[SUP] 13 [/SUP], Zhihan Yan[SUP] 19 [/SUP], Xiaoguang Zou[SUP] 21 [/SUP], Guiping Lin[SUP] 3 [/SUP], Guiqun Cao[SUP] 4 [/SUP], Laurance L Lau[SUP] 5 [/SUP], Long Mo[SUP] 16 [/SUP], Yong Liang[SUP] 5 [/SUP], Michael Roberts[SUP] 22 23 [/SUP], Evis Sala[SUP] 24 [/SUP], Carola-Bibiane Sch?nlieb[SUP] 23 [/SUP], Manson Fok[SUP] 5 [/SUP], Johnson Yiu-Nam Lau[SUP] 25 [/SUP], Tao Xu[SUP] 11 [/SUP], Jianxing He[SUP] 12 [/SUP], Kang Zhang[SUP] 26 27 [/SUP], Weimin Li[SUP] 28 [/SUP], Tianxin Lin[SUP] 29 [/SUP]
Affiliations
Abstract
Common lung diseases are first diagnosed using chest X-rays. Here, we show that a fully automated deep-learning pipeline for the standardization of chest X-ray images, for the visualization of lesions and for disease diagnosis can identify viral pneumonia caused by coronavirus disease 2019 (COVID-19) and assess its severity, and can also discriminate between viral pneumonia caused by COVID-19 and other types of pneumonia. The deep-learning system was developed using a heterogeneous multicentre dataset of 145,202 images, and tested retrospectively and prospectively with thousands of additional images across four patient cohorts and multiple countries. The system generalized across settings, discriminating between viral pneumonia, other types of pneumonia and the absence of disease with areas under the receiver operating characteristic curve (AUCs) of 0.94-0.98; between severe and non-severe COVID-19 with an AUC of 0.87; and between COVID-19 pneumonia and other viral or non-viral pneumonia with AUCs of 0.87-0.97. In an independent set of 440 chest X-rays, the system performed comparably to senior radiologists and improved the performance of junior radiologists. Automated deep-learning systems for the assessment of pneumonia could facilitate early intervention and provide support for clinical decision-making.
. 2021 Apr 15.
doi: 10.1038/s41551-021-00704-1. Online ahead of print.
A deep-learning pipeline for the diagnosis and discrimination of viral, non-viral and COVID-19 pneumonia from chest X-ray images
Guangyu Wang[SUP] #[/SUP][SUP] 1 [/SUP], Xiaohong Liu[SUP] #[/SUP][SUP] 2 [/SUP], Jun Shen[SUP] #[/SUP][SUP] 3 [/SUP], Chengdi Wang[SUP] #[/SUP][SUP] 4 [/SUP], Zhihuan Li[SUP] #[/SUP][SUP] 5 [/SUP], Linsen Ye[SUP] #[/SUP][SUP] 6 [/SUP], Xingwang Wu[SUP] #[/SUP][SUP] 7 [/SUP], Ting Chen[SUP] 8 [/SUP], Kai Wang[SUP] 2 [/SUP], Xuan Zhang[SUP] 2 [/SUP], Zhongguo Zhou[SUP] 9 [/SUP], Jian Yang[SUP] 10 [/SUP], Ye Sang[SUP] 10 [/SUP], Ruiyun Deng[SUP] 11 [/SUP], Wenhua Liang[SUP] 12 [/SUP], Tao Yu[SUP] 3 [/SUP], Ming Gao[SUP] 3 [/SUP], Jin Wang[SUP] 6 [/SUP], Zehong Yang[SUP] 3 [/SUP], Huimin Cai[SUP] 11 [/SUP], Guangming Lu[SUP] 13 [/SUP], Lingyan Zhang[SUP] 14 [/SUP], Lei Yang[SUP] 15 [/SUP], Wenqin Xu[SUP] 5 [/SUP], Winston Wang[SUP] 5 [/SUP], Andrea Olevera[SUP] 5 [/SUP], Ian Ziyar[SUP] 5 [/SUP], Charlotte Zhang[SUP] 11 [/SUP], Oulan Li[SUP] 11 [/SUP], Weihua Liao[SUP] 16 [/SUP], Jun Liu[SUP] 17 [/SUP], Wen Chen[SUP] 18 [/SUP], Wei Chen[SUP] 19 [/SUP], Jichan Shi[SUP] 20 [/SUP], Lianghong Zheng[SUP] 5 [/SUP], Longjiang Zhang[SUP] 13 [/SUP], Zhihan Yan[SUP] 19 [/SUP], Xiaoguang Zou[SUP] 21 [/SUP], Guiping Lin[SUP] 3 [/SUP], Guiqun Cao[SUP] 4 [/SUP], Laurance L Lau[SUP] 5 [/SUP], Long Mo[SUP] 16 [/SUP], Yong Liang[SUP] 5 [/SUP], Michael Roberts[SUP] 22 23 [/SUP], Evis Sala[SUP] 24 [/SUP], Carola-Bibiane Sch?nlieb[SUP] 23 [/SUP], Manson Fok[SUP] 5 [/SUP], Johnson Yiu-Nam Lau[SUP] 25 [/SUP], Tao Xu[SUP] 11 [/SUP], Jianxing He[SUP] 12 [/SUP], Kang Zhang[SUP] 26 27 [/SUP], Weimin Li[SUP] 28 [/SUP], Tianxin Lin[SUP] 29 [/SUP]
Affiliations
- PMID: 33859385
- DOI: 10.1038/s41551-021-00704-1
Abstract
Common lung diseases are first diagnosed using chest X-rays. Here, we show that a fully automated deep-learning pipeline for the standardization of chest X-ray images, for the visualization of lesions and for disease diagnosis can identify viral pneumonia caused by coronavirus disease 2019 (COVID-19) and assess its severity, and can also discriminate between viral pneumonia caused by COVID-19 and other types of pneumonia. The deep-learning system was developed using a heterogeneous multicentre dataset of 145,202 images, and tested retrospectively and prospectively with thousands of additional images across four patient cohorts and multiple countries. The system generalized across settings, discriminating between viral pneumonia, other types of pneumonia and the absence of disease with areas under the receiver operating characteristic curve (AUCs) of 0.94-0.98; between severe and non-severe COVID-19 with an AUC of 0.87; and between COVID-19 pneumonia and other viral or non-viral pneumonia with AUCs of 0.87-0.97. In an independent set of 440 chest X-rays, the system performed comparably to senior radiologists and improved the performance of junior radiologists. Automated deep-learning systems for the assessment of pneumonia could facilitate early intervention and provide support for clinical decision-making.