TY - JOUR
T1 - Deep convolution neural networks to differentiate between COVID-19 and other pulmonary abnormalities on chest radiographs
T2 - Evaluation using internal and external datasets
AU - Cho, Yongwon
AU - Hwang, Sung Ho
AU - Oh, Yu Whan
AU - Ham, Byung Joo
AU - Kim, Min Ju
AU - Park, Beomjin
N1 - Funding Information:
This study was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education (2018R1D1A1A02085358 and 2020R1I1A1A01071600) and by a grant from Korea University Anam Hospital, Seoul, Republic of Korea (Grant Nos. K1809771 and O2000221). The authors would like to thank Editage ( www.editage.co.kr ) for English language editing.
Funding Information:
Korea University Anam Hospital, Grant/Award Numbers: K1809771, O2000221; National Research Foundation of Korea, The Ministry of Education, Grant/Award Numbers: 2018R1D1A1A02085358, 2020R1I1A1A01071600 Funding information
Publisher Copyright:
© 2021 Wiley Periodicals LLC.
PY - 2021/9
Y1 - 2021/9
N2 - We aimed to evaluate the performance of convolutional neural networks (CNNs) in the classification of coronavirus disease 2019 (COVID-19) disease using normal, pneumonia, and COVID-19 chest radiographs (CXRs). First, we collected 9194 CXRs from open datasets and 58 from the Korea University Anam Hospital (KUAH). The number of normal, pneumonia, and COVID-19 CXRs were 4580, 3884, and 730, respectively. The CXRs obtained from the open dataset were randomly assigned to the training, tuning, and test sets in a 70:10:20 ratio. For external validation, the KUAH (20 normal, 20 pneumonia, and 18 COVID-19) dataset, verified by radiologists using computed tomography, was used. Subsequently, transfer learning was conducted using DenseNet169, InceptionResNetV2, and Xception to identify COVID-19 using open datasets (internal) and the KUAH dataset (external) with histogram matching. Gradient-weighted class activation mapping was used for the visualization of abnormal patterns in CXRs. The average AUC and accuracy of the multiscale and mixed-COVID-19Net using three CNNs over five folds were (0.99 ± 0.01 and 92.94% ± 0.45%), (0.99 ± 0.01 and 93.12% ± 0.23%), and (0.99 ± 0.01 and 93.57% ± 0.29%), respectively, using the open datasets (internal). Furthermore, these values were (0.75 and 74.14%), (0.72 and 68.97%), and (0.77 and 68.97%), respectively, for the best model among the fivefold cross-validation with the KUAH dataset (external) using domain adaptation. The various state-of-the-art models trained on open datasets show satisfactory performance for clinical interpretation. Furthermore, the domain adaptation for external datasets was found to be important for detecting COVID-19 as well as other diseases.
AB - We aimed to evaluate the performance of convolutional neural networks (CNNs) in the classification of coronavirus disease 2019 (COVID-19) disease using normal, pneumonia, and COVID-19 chest radiographs (CXRs). First, we collected 9194 CXRs from open datasets and 58 from the Korea University Anam Hospital (KUAH). The number of normal, pneumonia, and COVID-19 CXRs were 4580, 3884, and 730, respectively. The CXRs obtained from the open dataset were randomly assigned to the training, tuning, and test sets in a 70:10:20 ratio. For external validation, the KUAH (20 normal, 20 pneumonia, and 18 COVID-19) dataset, verified by radiologists using computed tomography, was used. Subsequently, transfer learning was conducted using DenseNet169, InceptionResNetV2, and Xception to identify COVID-19 using open datasets (internal) and the KUAH dataset (external) with histogram matching. Gradient-weighted class activation mapping was used for the visualization of abnormal patterns in CXRs. The average AUC and accuracy of the multiscale and mixed-COVID-19Net using three CNNs over five folds were (0.99 ± 0.01 and 92.94% ± 0.45%), (0.99 ± 0.01 and 93.12% ± 0.23%), and (0.99 ± 0.01 and 93.57% ± 0.29%), respectively, using the open datasets (internal). Furthermore, these values were (0.75 and 74.14%), (0.72 and 68.97%), and (0.77 and 68.97%), respectively, for the best model among the fivefold cross-validation with the KUAH dataset (external) using domain adaptation. The various state-of-the-art models trained on open datasets show satisfactory performance for clinical interpretation. Furthermore, the domain adaptation for external datasets was found to be important for detecting COVID-19 as well as other diseases.
KW - COVID-19
KW - chest radiography
KW - computer-aided diagnosis (CAD)
KW - deep learning
KW - lung diseases
UR - http://www.scopus.com/inward/record.url?scp=85105620409&partnerID=8YFLogxK
U2 - 10.1002/ima.22595
DO - 10.1002/ima.22595
M3 - Article
AN - SCOPUS:85105620409
VL - 31
SP - 1087
EP - 1104
JO - International Journal of Imaging Systems and Technology
JF - International Journal of Imaging Systems and Technology
SN - 0899-9457
IS - 3
ER -