|
|
|
|
LEADER |
03324nam a2200637Ia 4500 |
001 |
10.1109-JBHI.2021.3104629 |
008 |
220427s2021 CNT 000 0 und d |
020 |
|
|
|a 21682194 (ISSN)
|
245 |
1 |
0 |
|a COVID-19 Screening in Chest X-Ray Images Using Lung Region Priors
|
260 |
|
0 |
|b Institute of Electrical and Electronics Engineers Inc.
|c 2021
|
856 |
|
|
|z View Fulltext in Publisher
|u https://doi.org/10.1109/JBHI.2021.3104629
|
520 |
3 |
|
|a Early screening of COVID-19 is essential for pandemic control, and thus to relieve stress on the health care system. Lung segmentation from chest X-ray (CXR) is a promising method for early diagnoses of pulmonary diseases. Recently, deep learning has achieved great success in supervised lung segmentation. However, how to effectively utilize the lung region in screening COVID-19 still remains a challenge due to domain shift and lack of manual pixel-level annotations. We hereby propose a multi-appearance COVID-19 screening framework by using lung region priors derived from CXR images. Firstly, we propose a multi-scale adversarial domain adaptation network (MS-AdaNet) to boost the cross-domain lung segmentation task as the prior knowledge to the classification network. Then, we construct a multi-appearance network (MA-Net), which is composed of three sub-networks to realize multi-appearance feature extraction and fusion using lung region priors. At last, we can obtain prediction results from normal, viral pneumonia, and COVID-19 using the proposed MA-Net. We extend the proposed MS-AdaNet for lung segmentation task on three different public CXR datasets. The results suggest that the MS-AdaNet outperforms contrastive methods in cross-domain lung segmentation. Moreover, experiments reveal that the proposed MA-Net achieves accuracy of 98.83\% and F1-score of 98.71\% on COVID-19 screening. The results indicate that the proposed MA-Net can obtain significant performance on COVID-19 screening. © 2013 IEEE.
|
650 |
0 |
4 |
|a Article
|
650 |
0 |
4 |
|a Biological organs
|
650 |
0 |
4 |
|a chest X-ray image
|
650 |
0 |
4 |
|a Chest X-ray image
|
650 |
0 |
4 |
|a Classification networks
|
650 |
0 |
4 |
|a convolutional neural network
|
650 |
0 |
4 |
|a coronavirus disease 2019
|
650 |
0 |
4 |
|a COVID-19
|
650 |
0 |
4 |
|a COVID-19
|
650 |
0 |
4 |
|a Cross-domain
|
650 |
0 |
4 |
|a Deep learning
|
650 |
0 |
4 |
|a Deep Learning
|
650 |
0 |
4 |
|a Diagnosis
|
650 |
0 |
4 |
|a diagnostic imaging
|
650 |
0 |
4 |
|a Domain adaptation
|
650 |
0 |
4 |
|a Early diagnosis
|
650 |
0 |
4 |
|a Health-care system
|
650 |
0 |
4 |
|a human
|
650 |
0 |
4 |
|a Humans
|
650 |
0 |
4 |
|a image segmentation
|
650 |
0 |
4 |
|a learning
|
650 |
0 |
4 |
|a learning algorithm
|
650 |
0 |
4 |
|a lung
|
650 |
0 |
4 |
|a Lung
|
650 |
0 |
4 |
|a lung disease
|
650 |
0 |
4 |
|a lung segmentation
|
650 |
0 |
4 |
|a Lung segmentation
|
650 |
0 |
4 |
|a machine learning
|
650 |
0 |
4 |
|a mathematical model
|
650 |
0 |
4 |
|a multi-appearance
|
650 |
0 |
4 |
|a pneumonia
|
650 |
0 |
4 |
|a Prior knowledge
|
650 |
0 |
4 |
|a SARS-CoV-2
|
650 |
0 |
4 |
|a thorax radiography
|
650 |
0 |
4 |
|a training
|
650 |
0 |
4 |
|a tuberculosis
|
650 |
0 |
4 |
|a Unsupervised domain adaptation
|
650 |
0 |
4 |
|a X ray
|
650 |
0 |
4 |
|a X-Rays
|
700 |
1 |
|
|a An, J.
|e author
|
700 |
1 |
|
|a Cai, Q.
|e author
|
700 |
1 |
|
|a Gao, Z.
|e author
|
700 |
1 |
|
|a Qu, Z.
|e author
|
773 |
|
|
|t IEEE Journal of Biomedical and Health Informatics
|