Deep convolution neural networks to differentiate between COVID-19 and other pulmonary abnormalities on chest radiographs: Evaluation using internal and external datasets
- Authors
- Cho Yongwon; Hwang Sung Ho; Oh Yu-Whan; Ham Byung-Joo; Kim Min Ju; Park Beom Jin
- Issue Date
- Sep-2021
- Publisher
- John Wiley & Sons Inc.
- Keywords
- chest radiography; computer-aided diagnosis (CAD); COVID-19; deep learning; lung diseases
- Citation
- International Journal of Imaging Systems and Technology, v.31, no.3, pp 1087 - 1104
- Pages
- 18
- Indexed
- SCIE
SCOPUS
- Journal Title
- International Journal of Imaging Systems and Technology
- Volume
- 31
- Number
- 3
- Start Page
- 1087
- End Page
- 1104
- URI
- https://scholarworks.korea.ac.kr/kumedicine/handle/2021.sw.kumedicine/64103
- DOI
- 10.1002/ima.22595
- ISSN
- 0899-9457
1098-1098
- Abstract
- 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.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - 2. Clinical Science > Department of Psychiatry > 1. Journal Articles
- 2. Clinical Science > Department of Radiology > 1. Journal Articles
- 4. Research institute > Institute of Human Behavior and Genetics > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.