TY - JOUR
T1 - Dual-Sampling Attention Network for Diagnosis of COVID-19 from Community Acquired Pneumonia
AU - Ouyang, Xi
AU - Huo, Jiayu
AU - Xia, Liming
AU - Shan, Fei
AU - Liu, Jun
AU - Mo, Zhanhao
AU - Yan, Fuhua
AU - Ding, Zhongxiang
AU - Yang, Qi
AU - Song, Bin
AU - Shi, Feng
AU - Yuan, Huan
AU - Wei, Ying
AU - Cao, Xiaohuan
AU - Gao, Yaozong
AU - Wu, Dijia
AU - Wang, Qian
AU - Shen, Dinggang
N1 - Funding Information:
Manuscript received April 28, 2020; revised May 12, 2020; accepted May 12, 2020. Date of publication May 18, 2020; date of current version July 30, 2020. This work was supported in part by the Wuhan Science and Technology Program under Grant 2018060401011326, in part by the Hubei Provincial Novel Pneumonia Emergency Science and Technology Project under Grant 2020FCA021, in part by the Huazhong University of Science and Technology Novel Coronavirus Pneumonia Emergency Science and Technology Project under Grant 2020kfyXGYJ014, in part by the Novel Coronavirus Special Research Foundation of the Shanghai Municipal Science and Technology Commission under Grant 20441900600, in part by the Key Emergency Project of Pneumonia Epidemic of Novel Coronavirus Infection under Grant 2020sk3006, in part by the Emergency Project of Prevention and Control for COVID-19 of Central South University under Grant 60260005, in part by the National Natural Science Foundation of China under Grant 81871337 and Grant 6204100022, in part by the National Key Research and Development Program of China under Grant 2018YFC0116400, and in part by the Science and Technology Commission of Shanghai Municipality (STCSM) under Grant 19QC1400600 and Grant 17411953300. (Xi Ouyang, Jiayu Huo, Liming Xia, Fei Shan, Jun Liu, Zhanhao Mo, Fuhua Yan, Zhongxiang Ding, Qi Yang, and Bin Song contributed equally to this work.) (Corresponding authors: Qian Wang; Dinggang Shen.) Please see the Acknowledgment section of this article for the author affiliations.
Funding Information:
This work was supported in part by the Wuhan Science and Technology Program under Grant 2018060401011326, in part by the Hubei Provincial Novel Pneumonia Emergency Science and Technology Project under Grant 2020FCA021, in part by the Huazhong University of Science and Technology Novel Coronavirus Pneumonia Emergency Science and Technology Project under Grant 2020kfyXGYJ014, in part by the Novel Coronavirus Special Research Foundation of the Shanghai Municipal Science and Technology Commission under Grant 20441900600, in part by the Key Emergency Project of Pneumonia Epidemic of Novel Coronavirus Infection under Grant 2020sk3006, in part by the Emergency Project of Prevention and Control for COVID-19 of Central South University under Grant 60260005, in part by the National Natural Science Foundation of China under Grant 81871337 and Grant 6204100022, in part by the National Key Research and Development Program of China under Grant 2018YFC0116400, and in part by the Science and Technology Commission of Shanghai Municipality (STCSM) under Grant 19QC1400600 and Grant 17411953300.
Publisher Copyright:
© 1982-2012 IEEE.
PY - 2020/8
Y1 - 2020/8
N2 - The coronavirus disease (COVID-19) is rapidly spreading all over the world, and has infected more than 1,436,000 people in more than 200 countries and territories as of April 9, 2020. Detecting COVID-19 at early stage is essential to deliver proper healthcare to the patients and also to protect the uninfected population. To this end, we develop a dual-sampling attention network to automatically diagnose COVID-19 from the community acquired pneumonia (CAP) in chest computed tomography (CT). In particular, we propose a novel online attention module with a 3D convolutional network (CNN) to focus on the infection regions in lungs when making decisions of diagnoses. Note that there exists imbalanced distribution of the sizes of the infection regions between COVID-19 and CAP, partially due to fast progress of COVID-19 after symptom onset. Therefore, we develop a dual-sampling strategy to mitigate the imbalanced learning. Our method is evaluated (to our best knowledge) upon the largest multi-center CT data for COVID-19 from 8 hospitals. In the training-validation stage, we collect 2186 CT scans from 1588 patients for a 5-fold cross-validation. In the testing stage, we employ another independent large-scale testing dataset including 2796 CT scans from 2057 patients. Results show that our algorithm can identify the COVID-19 images with the area under the receiver operating characteristic curve (AUC) value of 0.944, accuracy of 87.5%, sensitivity of 86.9%, specificity of 90.1%, and F1-score of 82.0%. With this performance, the proposed algorithm could potentially aid radiologists with COVID-19 diagnosis from CAP, especially in the early stage of the COVID-19 outbreak.
AB - The coronavirus disease (COVID-19) is rapidly spreading all over the world, and has infected more than 1,436,000 people in more than 200 countries and territories as of April 9, 2020. Detecting COVID-19 at early stage is essential to deliver proper healthcare to the patients and also to protect the uninfected population. To this end, we develop a dual-sampling attention network to automatically diagnose COVID-19 from the community acquired pneumonia (CAP) in chest computed tomography (CT). In particular, we propose a novel online attention module with a 3D convolutional network (CNN) to focus on the infection regions in lungs when making decisions of diagnoses. Note that there exists imbalanced distribution of the sizes of the infection regions between COVID-19 and CAP, partially due to fast progress of COVID-19 after symptom onset. Therefore, we develop a dual-sampling strategy to mitigate the imbalanced learning. Our method is evaluated (to our best knowledge) upon the largest multi-center CT data for COVID-19 from 8 hospitals. In the training-validation stage, we collect 2186 CT scans from 1588 patients for a 5-fold cross-validation. In the testing stage, we employ another independent large-scale testing dataset including 2796 CT scans from 2057 patients. Results show that our algorithm can identify the COVID-19 images with the area under the receiver operating characteristic curve (AUC) value of 0.944, accuracy of 87.5%, sensitivity of 86.9%, specificity of 90.1%, and F1-score of 82.0%. With this performance, the proposed algorithm could potentially aid radiologists with COVID-19 diagnosis from CAP, especially in the early stage of the COVID-19 outbreak.
KW - COVID-19 Diagnosis
KW - Dual Sampling Strategy
KW - Explainability
KW - Imbalanced Distribution
KW - Online Attention
UR - http://www.scopus.com/inward/record.url?scp=85088891863&partnerID=8YFLogxK
U2 - 10.1109/TMI.2020.2995508
DO - 10.1109/TMI.2020.2995508
M3 - Article
C2 - 32730212
AN - SCOPUS:85088891863
SN - 0278-0062
VL - 39
SP - 2595
EP - 2605
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 8
M1 - 9095328
ER -