Disease identification from chest X-ray images using deep learning

Öz

Disease identification using deep learning has been an intensive area of research over the last years, and the models of which have been successfully employed in medical applications. Being of evidence from the studies conducted in recent years, the use of deep learning in medical sciences has gained more attention due to Covid-19 pandemic emerging in late 2019. In this work, ResNet50, Inceptionv3, VGG16, AlexNet, and a model designed as five-stages deep convolutional neural network cascaded with two fully connected layers, having 3.6 million parameters in total, has been used. A data set of 2500 images has been set up to consist of five classes labelled as Covid-19 (C), Bacterial Pneumonia (BZ), Viral Pneumonia (VZ), Lung Opacity (AO) and Normal (N), each of which has 500 entries randomly drawn from two different image data base. With the ResNet50 model, 95.73% accuracy, 0.9574 F1 score and 0.99672 AUC, with the Inceptionv3 model, 92.53% accuracy, 0.9251 F1 score and 0.99264 AUC, with the VGG16 model, 97.33% accuracy, 0.9734 F1 score and 0.9978 AUC, with the AlexNet model, 94.67% accuracy, 0.9487 F1 score and 0.99653 AUC, and finally with the designed model, 95.22% accuracy, 0.9521 F1 score and 0.99868 AUC value have been obtained.

Anahtar Kelimeler

• Chest X-ray images
• Covid-19
• pneumonia
• deep learning

DERİN ÖĞRENME İLE AKCİĞER X-RAY GÖRÜNTÜLERİNDEN HASTALIK TESPİTİ

Öz

Derin Öğrenme ile hastalık teşhisi son dönemlerde araştırmacıların üstünde yoğun şekilde çalıştıkları bir konu olup, yöntemleri pek çok sağlık alanında başarıyla uygulanmaktadır. Günümüzdeki birçok araştırmadan görüleceği üzere Derin Öğrenmenin tıp alanında kullanımı, 2019 yılının sonlarında ortaya çıkan ve pandemiye yol açan COVID-19 hastalığı ile daha da önem kazanmıştır. Bu çalışmada ResNet50, Inceptionv3, VGG16, AlexNet ve ayrıca 5 aşamalı evrişim ile 2 adet tam bağlantılı katman halinde tasarlanan 3,6 milyon parametreli bir model kullanılmıştır. İki ayrı veri setinden, görüntüler rastgele şekilde, her bir sınıftan 500 adet olacak şekilde, seçilip toplamda 2500 görüntü verisi kullanılarak, Covid19 (C), Bakteriyel Zatürre (BZ), Viral Zatürre (VZ), Akciğer Opaklığı (AO) ve Normal (N) olmak üzere, 5 sınıf içeren bir veri seti oluşturulmuştur. ResNet50 modeli ile %95,73 doğruluk, 0,9574 F1 skor ve 0,99672 AUC değeri, Inceptionv3 modeli ile %92,53 doğruluk, 0,9251 F1 skor ve 0,99264 AUC değeri, VGG16 modeli ile %97,33 doğruluk, 0,9734 F1 skor ve 0,9978 AUC değeri, AlexNet modeli ile %94,67 doğruluk, 0,9487 F1 skor ve 0,99653 AUC değeri, ve son olarak tasarlanan model ile %95,22 doğruluk, 0,9521 F1 skor ve 0,99868 AUC değeri elde edilmiştir.

Anahtar Kelimeler

• Akciğer X-Ray görüntüleri
• Covid-19
• zatürre
• derin öğrenme

Kaynakça

1. Afshar P, Heidarian S, Naderkhani F, Oikonomou A, Plataniotis KN, Mohammadi A. (2020) “Covid-caps: A capsule network-based framework for identification of covid-19 cases from X-ray images”. Pattern Recognition Letters, Volume 138, 638–643. doi:10.1016/j.patec.2020.09.010
2. Ardila, D., Kiraly, A.P., Bharadwaj, S., Choi, B., Reicher, J.J., Peng, L., Tse, D., Etemadi, M., Ye, W., Corrado, G. (2019) “End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography”, Nature Medicine, 25(6), 954–961. doi:10.1038/s41591-019-0447-x
3. Arun Prakash, J., Asswin, C., Ravi, V., Sowmya, V. ve Soman, K. P. (2023) “Pediatric pneumonia diagnosis using stacked ensemble learning on multi-model deep CNN architectures”. Multimedia Tools and Applications, 82(14), 21311 -21351. doi:10.1007/s11042-022-13844-6
4. Bouchareb Y, Moradi Khaniabadi P, Al Kindi F, Al Dhuhli H, Shiri I, Zaidi H, Rahmim A. (2021) “Artificial intelligence-driven assessment of radiological images for COVID-19”, Comp. in Biology and Medicine, 136, 104665. doi: 10.1016/j.compbiomed.2021.104665.
5. Bozkurt, F. (2021) "Derin Öğrenme Tekniklerini Kullanarak Akciğer X-Ray Görüntülerinden COVID-19 Tespiti", Avrupa Bilim ve Teknoloji Dergisi, 24, 149-156. doi:10.31590/ejosat.898385
6. Brunese, L., Mercaldo, F., Reginelli, A., ve Santone, A. (2020) “Explainable deep learning for pulmonary disease and coronavirus COVID-19 detection from X-rays”, Computer Methods and Prog. in Biomedicine, 24(10), 1559-1567. doi:10.1038/s41591-018-0177-5
7. Condaragiu, S. ve Ciocoiu, I. B. (2021) “Evaluation of convolutional neural networks for COVID-19 detection from chest X-ray images”, 2021 International Symposium on Signals, Circuits and Systems ISSCS, 1-4. doi:10.1109/ISSCS52333.2021.9497418
8. Coudray, N., Ocampo, P.S., Sakellaropoulos, T., Narula, N., Snuderl, M., David, F., Moreira, A.L., Razavian, N. ve Tsirigos, A. (2018) “Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning”, Nature Medicine, doi:10.1038/s41591-018-0177-5

9. Demir, F. B. ve Yılmaz, E. (2021) “X-ray görüntülerinden COVID-19 tespiti için derin öğrenme temelli bir yaklaşım”, European Journal of Science and Technology, Special Issue 32, 627-632, doi:10.31590/ejosat.1039522
10. Esteva, A., Kuprel, B., Novoa, R.A., Ko, J., Swetter, S.M., Blau, H.M. ve Thrun, S. (2017) “Dermatologist-level classification of skin cancer with deep neural networks”, Nature, 542, 115–118 (2017). doi:10.1038/nature21056
11. Goodfellow, I., Bengio, Y. Ve Courville, A. (2016) Deep Learning, The MIT Press
12. He, K., Zhang, X., Ren, S. ve Sun, J. (2015) “Deep residual learning for image recognition”, Computer Vision and Pattern Recognition, doi:10.48550/arXiv.1512.03385v1 
13. Ibrahim A. U., Ozsoz, M, Serte, S., Al-Turjman F. ve Yakoi P. S. (2021) “Pneumonia classification using deep learning from chest X-ray images during COVID-19”, Cognitive Computation, 1-13. doi:10.1007/s12559-020-09787-5.
14. Ismael, A. M. ve Şengür, A. (2021). “Deep learning approaches for COVID-19 detection based on chest X-ray images”, Expert Sys. with App, 164. doi: 10.1016/j.eswa.2020.114054
15. Khan A. I., Shah, J. L.ve Bhat, M. M. (2020) “Coronet: A deep neural network for detection and diagnosis of covid-19 from chest X-ray images”, Computer Methods and Programs in Biomedicine, 196, 105581. doi:10.1016/j.cmpb.2020.105581
16. Krizhevsky, A., Sutskever I. ve Hinton, G. E. (2012) “ImageNet classification with deep convolutional neural networks” Communications of the ACM, 60, 84 - 90. doi:10.1145/3065386
17. Kumar, S. (2022), “Covid19-Pneumonia-Normal Chest X-ray Images”, Mendeley Data, V1, doi: 10.17632/dvntn9yhd2.1
18. Liu, C. ve Niu, S. (2024) “Automated Fruit Sorting in Smart Agriculture System: Analysis of Deep Learning-based Algorithms”, International Journal of Advanced Computer Science and Applications, (IJACSA), Vol. 15, No. 1, doi:10.14569/IJACSA.2024.0150183
19. Loey, M., Smarandache, F. Ve Khalifa, N. E. (2020) “Within the Lack of Chest COVID-19 X-ray Dataset: A Novel Detection Model Based on GAN and Deep Transfer Learning. Symmetry, 12, 651. doi:10.3390/sym12040651
20. MathWorks, Erişim Adresi: https://www.mathworks.com/help/deeplearning/ref/, (Erişim Tarihi: 02.12.2024)
21. Mehrparvar, F. , https://www.kaggle.com/datasets/fatemehmehrparvar/lung-disease
22. Nayak, S. R., Nayak, D. R., Sinha, U., Arora, V. ve Pachori R. B. (2021) “Application of deep learning techniques for detection of COVID-19 cases using chest X-ray images: A comprehensive study”, Biomedical Signal Processing and Control, 64:102365, doi:10.1016/j.bspc.2020.102365
23. Putzu, L., Piras, L., Giacinto, G. (2020) “Convolutional neural networks for relevance feedbackin content based image retrieval”, Multimedia Tools and Applications, 79:26995–27021, doi:10.1007/s11042-020-09292-9.
24. Rahadian, R., ve Suyanto, S. (2019).” Deep residual neural network for age classification with face image”, 2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI), 21-24. doi:10.1109/ISRITI48646.2019.9034664
25. Sevli O. (2022) “A deep learning-based approach for diagnosing COVID-19 on chest x-ray images, and a test study with clinical experts”, Computational Intelligence, (38), 1-25. doi: 10.1111/coin.12526.
26. Simonyan, K. ve Zisserman, . (2015) “Very deep convolutional networks for large-scale image recognition”, The 3rd International Conference on Learning Representations (ICLR2015). doi:10.48550/arXiv.1409.1556
27. Singh, K. K., Siddhartha, M., ve Singh, A. (2020). “Diagnosis of Coronavirus Disease (COVID-19) from Chest X-Ray images using modified XceptionNet”, Romanian Journal of Information Science and Technology, 23(657), 91 -115.
28. Suzuki, K. (2017) “Overview of deep learning in medical imaging”. Radiological Physics and Technology. 10(3), 257-273. doi:10.1007/s12194-017-0406-5
29. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V. ve Rabinovich A. (2015) “Going deeper with convolutions”. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, 7-12 June 2015, 1-9. doi:10.1109/CVPR.2015.7298594
30. Ullah, Z., Usman, M., Latif, S. ve Gwak, J. (2023) “Densely attention mechanism based network for COVID-19 detection in chest X-rays”, Scientific Reports, 13, 261. doi:10.1038/s41598-022-27266-9
31. Wang, L., Lin, Z. Q. ve Wong, A. (2020) “COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images,” Scientific Reports, 10(1). 19549
32. WHO, (2020) World Health Organization: Use of Chest Imaging in Covid-19. 2020. Erişim Adresi: https://www.who.int/publications/i/item/use-of-chest-imaging-in-covid-19 (Erişim Tarihi: 07.01.2025)
33. WHO, (2021), World Health Orginazation. (2021, August 28). Coronavirus (COVID-19) Dashboard Erişim Adresi: http://covid19.who.int. (Erişim Tarihi: 20.12.2024)
34. Wikipedia, (2024), COVID-19 pandemisi, .Erişim Adresi: https://tr.wikipedia.org/wiki/COVID-19_pandemisi (Erişim Tarihi: 20.12.2024)