DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection
Currently, many deep learning models are being used to classify COVID‐19 and normal cases from chest X‐rays. However, the available data (X‐rays) for COVID‐19 is limited to train a robust deep‐learning model. Researchers have used data augmentation techniques to tackle this issue by increasing the n...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Wiley-Blackwell Publishing
2021
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/32820/1/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection.pdf https://eprints.ums.edu.my/id/eprint/32820/2/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection%20_ABSTRACT.pdf |
| _version_ | 1825714763986894848 |
|---|---|
| author | Pir Masoom Shah Hamid Ullah Rahim Ullah Dilawar Shah Yulin Wang Saif ul Islam Abdullah Gani Rodrigues, Joel J. P. C. |
| author_facet | Pir Masoom Shah Hamid Ullah Rahim Ullah Dilawar Shah Yulin Wang Saif ul Islam Abdullah Gani Rodrigues, Joel J. P. C. |
| author_sort | Pir Masoom Shah |
| collection | UMS |
| description | Currently, many deep learning models are being used to classify COVID‐19 and normal cases from chest X‐rays. However, the available data (X‐rays) for COVID‐19 is limited to train a robust deep‐learning model. Researchers have used data augmentation techniques to tackle this issue by increasing the numbers of samples through flipping, translation, and rotation. However, by adopting this strategy, the model compromises for the learning of high‐dimensional features for a given problem. Hence, there are high chances of overfitting. In this paper, we used deep‐convolutional generative adversarial networks algorithm to address this issue, which generates synthetic images for all the classes (Normal, Pneumonia, and COVID‐19). To validate whether the generated images are accurate, we used the k‐mean clustering technique with three clusters (Normal, Pneumonia, and COVID‐19). We only selected the X‐ray images classified in the correct clusters for training. In this way, we formed a synthetic dataset with three classes. The generated dataset was then fed to The EfficientNetB4 for training. The experiments achieved promising results of 95% in terms of area under the curve (AUC). To validate that our network has learned discriminated features associated with lung in the X‐rays, we used the Grad‐CAM technique to visualize the underlying pattern, which leads the network to its final decision. |
| first_indexed | 2024-03-06T03:16:39Z |
| format | Article |
| id | ums.eprints-32820 |
| institution | Universiti Malaysia Sabah |
| language | English English |
| last_indexed | 2024-03-06T03:16:39Z |
| publishDate | 2021 |
| publisher | Wiley-Blackwell Publishing |
| record_format | dspace |
| spelling | ums.eprints-328202022-06-16T07:33:41Z https://eprints.ums.edu.my/id/eprint/32820/ DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection Pir Masoom Shah Hamid Ullah Rahim Ullah Dilawar Shah Yulin Wang Saif ul Islam Abdullah Gani Rodrigues, Joel J. P. C. RC581-951 Specialties of internal medicine Currently, many deep learning models are being used to classify COVID‐19 and normal cases from chest X‐rays. However, the available data (X‐rays) for COVID‐19 is limited to train a robust deep‐learning model. Researchers have used data augmentation techniques to tackle this issue by increasing the numbers of samples through flipping, translation, and rotation. However, by adopting this strategy, the model compromises for the learning of high‐dimensional features for a given problem. Hence, there are high chances of overfitting. In this paper, we used deep‐convolutional generative adversarial networks algorithm to address this issue, which generates synthetic images for all the classes (Normal, Pneumonia, and COVID‐19). To validate whether the generated images are accurate, we used the k‐mean clustering technique with three clusters (Normal, Pneumonia, and COVID‐19). We only selected the X‐ray images classified in the correct clusters for training. In this way, we formed a synthetic dataset with three classes. The generated dataset was then fed to The EfficientNetB4 for training. The experiments achieved promising results of 95% in terms of area under the curve (AUC). To validate that our network has learned discriminated features associated with lung in the X‐rays, we used the Grad‐CAM technique to visualize the underlying pattern, which leads the network to its final decision. Wiley-Blackwell Publishing 2021 Article PeerReviewed text en https://eprints.ums.edu.my/id/eprint/32820/1/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection.pdf text en https://eprints.ums.edu.my/id/eprint/32820/2/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection%20_ABSTRACT.pdf Pir Masoom Shah and Hamid Ullah and Rahim Ullah and Dilawar Shah and Yulin Wang and Saif ul Islam and Abdullah Gani and Rodrigues, Joel J. P. C. (2021) DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection. Expert Systems, 39. pp. 1-13. https://onlinelibrary.wiley.com/doi/epdf/10.1111/exsy.12823 https://doi.org/10.1111/exsy.12823 https://doi.org/10.1111/exsy.12823 |
| spellingShingle | RC581-951 Specialties of internal medicine Pir Masoom Shah Hamid Ullah Rahim Ullah Dilawar Shah Yulin Wang Saif ul Islam Abdullah Gani Rodrigues, Joel J. P. C. DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title | DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title_full | DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title_fullStr | DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title_full_unstemmed | DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title_short | DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection |
| title_sort | dc gan based synthetic x ray images augmentation for increasing the performance of efficientnet for covid 19 detection |
| topic | RC581-951 Specialties of internal medicine |
| url | https://eprints.ums.edu.my/id/eprint/32820/1/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection.pdf https://eprints.ums.edu.my/id/eprint/32820/2/DC-GAN-based%20synthetic%20X-ray%20images%20augmentation%20for%20increasing%20the%20performance%20of%20EfficientNet%20for%20COVID-19%20detection%20_ABSTRACT.pdf |
| work_keys_str_mv | AT pirmasoomshah dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT hamidullah dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT rahimullah dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT dilawarshah dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT yulinwang dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT saifulislam dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT abdullahgani dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection AT rodriguesjoeljpc dcganbasedsyntheticxrayimagesaugmentationforincreasingtheperformanceofefficientnetforcovid19detection |