An XAI approach for COVID-19 detection using transfer learning with X-ray images
The coronavirus disease (COVID-19) has continued to cause severe challenges during this unprecedented time, affecting every part of daily life in terms of health, economics, and social development. There is an increasing demand for chest X-ray (CXR) scans, as pneumonia is the primary and vital compl...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-04-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023023447 |
| _version_ | 1797836947591266304 |
|---|---|
| author | Salih Sarp Ferhat Ozgur Catak Murat Kuzlu Umit Cali Huseyin Kusetogullari Yanxiao Zhao Gungor Ates Ozgur Guler |
| author_facet | Salih Sarp Ferhat Ozgur Catak Murat Kuzlu Umit Cali Huseyin Kusetogullari Yanxiao Zhao Gungor Ates Ozgur Guler |
| author_sort | Salih Sarp |
| collection | DOAJ |
| description | The coronavirus disease (COVID-19) has continued to cause severe challenges during this unprecedented time, affecting every part of daily life in terms of health, economics, and social development. There is an increasing demand for chest X-ray (CXR) scans, as pneumonia is the primary and vital complication of COVID-19. CXR is widely used as a screening tool for lung-related diseases due to its simple and relatively inexpensive application. However, these scans require expert radiologists to interpret the results for clinical decisions, i.e., diagnosis, treatment, and prognosis. The digitalization of various sectors, including healthcare, has accelerated during the pandemic, with the use and importance of Artificial Intelligence (AI) dramatically increasing. This paper proposes a model using an Explainable Artificial Intelligence (XAI) technique to detect and interpret COVID-19 positive CXR images. We further analyze the impact of COVID-19 positive CXR images using heatmaps. The proposed model leverages transfer learning and data augmentation techniques for faster and more adequate model training. Lung segmentation is applied to enhance the model performance further. We conducted a pre-trained network comparison with the highest classification performance (F1-Score: 98%) using the ResNet model. |
| first_indexed | 2024-04-09T15:18:11Z |
| format | Article |
| id | doaj.art-58a4d103dde641d6b25b3bffbbdef885 |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-04-09T15:18:11Z |
| publishDate | 2023-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-58a4d103dde641d6b25b3bffbbdef8852023-04-29T14:55:28ZengElsevierHeliyon2405-84402023-04-0194e15137An XAI approach for COVID-19 detection using transfer learning with X-ray imagesSalih Sarp0Ferhat Ozgur Catak1Murat Kuzlu2Umit Cali3Huseyin Kusetogullari4Yanxiao Zhao5Gungor Ates6Ozgur Guler7Electrical & Computer Engineering, Virginia Commonwealth University, Richmond, VA, USADepartment of Electrical Engineering & Computer Science, University of Stavanger, Rogaland, NorwayOld Dominion University, Batten College of Engineering & Technology, Norfolk, VA, USADepartment of Electric Power Engineering, Norwegian University of Science and Technology, Trondheim, Norway; Corresponding author.Department of Computer Science, Blekinge Institute of Technology, Karlskrona, SwedenElectrical & Computer Engineering, Virginia Commonwealth University, Richmond, VA, USADepartment of Pulmonary Medicine, Private Genesis Hospital, Diyarbakir, TurkeyeKare, Inc Fairfax, VA, USAThe coronavirus disease (COVID-19) has continued to cause severe challenges during this unprecedented time, affecting every part of daily life in terms of health, economics, and social development. There is an increasing demand for chest X-ray (CXR) scans, as pneumonia is the primary and vital complication of COVID-19. CXR is widely used as a screening tool for lung-related diseases due to its simple and relatively inexpensive application. However, these scans require expert radiologists to interpret the results for clinical decisions, i.e., diagnosis, treatment, and prognosis. The digitalization of various sectors, including healthcare, has accelerated during the pandemic, with the use and importance of Artificial Intelligence (AI) dramatically increasing. This paper proposes a model using an Explainable Artificial Intelligence (XAI) technique to detect and interpret COVID-19 positive CXR images. We further analyze the impact of COVID-19 positive CXR images using heatmaps. The proposed model leverages transfer learning and data augmentation techniques for faster and more adequate model training. Lung segmentation is applied to enhance the model performance further. We conducted a pre-trained network comparison with the highest classification performance (F1-Score: 98%) using the ResNet model.http://www.sciencedirect.com/science/article/pii/S2405844023023447COVID-19Explainable artificial intelligenceTransfer learning |
| spellingShingle | Salih Sarp Ferhat Ozgur Catak Murat Kuzlu Umit Cali Huseyin Kusetogullari Yanxiao Zhao Gungor Ates Ozgur Guler An XAI approach for COVID-19 detection using transfer learning with X-ray images Heliyon COVID-19 Explainable artificial intelligence Transfer learning |
| title | An XAI approach for COVID-19 detection using transfer learning with X-ray images |
| title_full | An XAI approach for COVID-19 detection using transfer learning with X-ray images |
| title_fullStr | An XAI approach for COVID-19 detection using transfer learning with X-ray images |
| title_full_unstemmed | An XAI approach for COVID-19 detection using transfer learning with X-ray images |
| title_short | An XAI approach for COVID-19 detection using transfer learning with X-ray images |
| title_sort | xai approach for covid 19 detection using transfer learning with x ray images |
| topic | COVID-19 Explainable artificial intelligence Transfer learning |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023023447 |
| work_keys_str_mv | AT salihsarp anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT ferhatozgurcatak anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT muratkuzlu anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT umitcali anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT huseyinkusetogullari anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT yanxiaozhao anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT gungorates anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT ozgurguler anxaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT salihsarp xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT ferhatozgurcatak xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT muratkuzlu xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT umitcali xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT huseyinkusetogullari xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT yanxiaozhao xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT gungorates xaiapproachforcovid19detectionusingtransferlearningwithxrayimages AT ozgurguler xaiapproachforcovid19detectionusingtransferlearningwithxrayimages |