Fundus Görüntülerinden Derin Öğrenme Teknikleri ile Glokom Hastalığının Tespiti

Öz

Glokom optik siniri etkileyen ve erken teşhis edilmediği durumlarda kısmi ya da kalıcı körlüğe neden olan bir retina hastalığıdır. Zamanla görme kaybına neden olan glokomun teşhisi için doktorlar fundus görüntülerini kullanmaktadır. Glokomun etken teşhisi oldukça önemlidir. Bu çalışmada, fundus görüntülerinden glokom tespiti için Evrişimli Sinir Ağları (ESA) modellerinden olan AlexNet, ResNet-18, VGG16, SqueezeNet ve GoogleNet kullanılmıştır. Kullanılan mimariler için elde edilen sonuçlar doğruluk, duyarlılık, özgüllük ve f1-ölçütü değerleri olmak üzere farklı performans metriklerine göre değerlendirilmiştir. Sonuçlara göre test veri kümesinde en iyi duyarlılık değeri %97.96 ile VGG16 tarafından elde edildiği, özgüllük, doğruluk ve f1-ölçütü için en iyi değerlerin ise sırasıyla %98.97, %97.98 ve %98 ile GoogleNet olduğu tespit edilmiştir.

Anahtar Kelimeler

• Evrişimsel Sinir Ağları
• Fundus görüntüleri
• Glokom
• Yapay Zekâ

Detection of Glaucoma Disease with Deep Learning Techniques from Fundus Images

Abstract

Glaucoma is a retinal disease that affects the optic nerve and causes partial or permanent blindness if not diagnosed early. To diagnose glaucoma, which causes vision loss over time, doctors use fundus images. The causative diagnosis of glaucoma is very important. In this study, Convolutional Neural Networks (CNN) models AlexNet, ResNet-18, VGG16, SqueezeNet, and GoogleNet were used for glaucoma detection from fundus images. The results obtained for the architectures used were evaluated according to different performance metrics such as accuracy, sensitivity, specificity, and f1-criterion values. According to the results, it was determined that the best sensitivity value in the test dataset was obtained by VGG16 with 97.96%, and the best values for specificity, accuracy, and f1-criterion were GoogleNet with 98.97%, 97.98%, and 98%, respectively.

Keywords

• Evrişimsel Sinir Ağları
• Fundus görüntüleri
• Glokom
• Yapay Zekâ

References

1. Ahmad, S., Ansari, S. U., Haider, U., Javed, K., Rahman, J. U., & Anwar, S. (2022). Confusion matrix-based modularity induction into pretrained CNN. Multimedia Tools and Applications, 1-27.
2. Alghamdi, H. S., Tang, H. L., Waheeb, S. A., & Peto, T. (2016, October). Automatic optic disc abnormality detection in fundus images: A deep learning approach. In Ophthalmic Medical Image Analysis International Workshop (Vol. 3, No. 2016). University of Iowa.
3. Almazroa, A., Alodhayb, S., Burman, R., Sun, W., Raahemifar, K., & Lakshminarayanan, V. (2015, October). Optic cup segmentation based on extracting blood vessel kinks and cup thresholding using Type-II fuzzy approach. In 2015 2nd International Conference on Opto-Electronics and Applied Optics (IEM OPTRONIX) (pp. 1-3). IEEE.
4. Alsulami, F., Alseleahbi, H., Alsaedi, R., Almaghdawi, R., Alafif, T., Ikram, M., ... & WeTeach, W. HiGANCNN: A Hybrid Generative Adversarial Network and Convolutional Neural Network for Glaucoma Detection.
5. Carrillo, J., Bautista, L., Villamizar, J., Rueda, J., & Sanchez, M. (2019, April). Glaucoma detection using fundus images of the eye. In 2019 XXII Symposium on Image, Signal Processing and Artificial Vision (STSIVA) (pp. 1-4). IEEE.
6. Chen, X., Xu, Y., Wong, D. W. K., Wong, T. Y., & Liu, J. (2015, August). Glaucoma detection based on deep convolutional neural network. In 2015 37th annual international conference of the IEEE engineering in medicine and biology society (EMBC) (pp. 715-718). IEEE.
7. Cho, H., Hwang, Y. H., Chung, J. K., Lee, K. B., Park, J. S., Kim, H. G., & Jeong, J. H. (2021). Deep learning ensemble method for classifying glaucoma stages using fundus photographs and convolutional neural networks. Current eye research, 46(10), 1516-1524.
8. Clifton, L., Clifton, D. A., Pimentel, M. A., Watkinson, P. J., & Tarassenko, L. (2012). Gaussian processes for personalized e-health monitoring with wearable sensors. IEEE Transactions on Biomedical Engineering, 60(1), 193-197.

9. Dey, A., & Bandyopadhyay, S. K. (2016). Automated glaucoma detection using support vector machine classification method. British Journal of Medicine and Medical Research, 11(12), 1.
10. Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. nature, 542(7639), 115-118.
11. Gheisari, S., Shariflou, S., Phu, J., Kennedy, P. J., Agar, A., Kalloniatis, M., & Golzan, S. M. (2021). A combined convolutional and recurrent neural network for enhanced glaucoma detection. Scientific reports, 11(1), 1-11.
12. Gómez-Ríos, A., Tabik, S., Luengo, J., Shihavuddin, A. S. M., Krawczyk, B., & Herrera, F. (2019). Towards highly accurate coral texture images classification using deep convolutional neural networks and data augmentation. Expert Systems with Applications, 118, 315-328.
13. Hatanaka, Y., Noudo, A., Muramatsu, C., Sawada, A., Hara, T., Yamamoto, T., & Fujita, H. (2010, March). Vertical cup-to-disc ratio measurement for diagnosis of glaucoma on fundus images. In Medical Imaging 2010: Computer-Aided Diagnosis (Vol. 7624, pp. 945-952). SPIE.
14. Hemelings, R., Elen, B., Barbosa-Breda, J., Blaschko, M. B., De Boever, P., & Stalmans, I. (2021). Deep learning on fundus images detects glaucoma beyond the optic disc. Scientific Reports, 11(1), 1-12.
15. Krizhevsky, A.; Sutskever, I.; Hinton, G.E. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems; Curran Associates, Inc.: New York, NY, USA, 2012;pp. 1097–1105.
16. Nayak, D. R., Das, D., Majhi, B., Bhandary, S. V., & Acharya, U. R. (2021). ECNet: An evolutionary convolutional network for automated glaucoma detection using fundus images. Biomedical Signal Processing and Control, 67, 102559.
17. Orlando, J. I., Prokofyeva, E., del Fresno, M., & Blaschko, M. B. (2017, January). Convolutional neural network transfer for automated glaucoma identification. In 12th international symposium on medical information processing and analysis (Vol. 10160, pp. 241-250). SPIE.
18. Özbay, E. (2022). An active deep learning method for diabetic retinopathy detection in segmented fundus images using artificial bee colony algorithm. Artificial Intelligence Review, 1-28.
19. Qassim H, Verma A, Feinzimer D. Compressed residual-VGG16 CNN model for big data places image recognition. 8th Annual Computing and Communication Workshop and Conference. Las Vegas: IEEE;2018. p. 169-175
20. Sevastopolsky, A. (2017). Optic disc and cup segmentation methods for glaucoma detection with modification of U-Net convolutional neural network. Pattern Recognition and Image Analysis, 27(3), 618-624.
21. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9).
22. Tham, Y. C., Li, X., Wong, T. Y., Quigley, H. A., Aung, T., & Cheng, C. Y. (2014). Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology, 121(11), 2081-2090.
23. Uddin, M., Tammimies, K., Pellecchia, G., Alipanahi, B., Hu, P., Wang, Z., ... & Scherer, S. W. (2014). Brain-expressed exons under purifying selection are enriched for de novo mutations in autism spectrum disorder. Nature genetics, 46(7), 742-747.
24. Weinreb, R. N., Aung, T., & Medeiros, F. A. (2014). The pathophysiology and treatment of glaucoma: a review. Jama, 311(18), 1901-1911.
25. Zavan, F. H. D. B., Bellon, O. R., Silva, L., & Medioni, G. G. (2019). Benchmarking parts based face processing in-the-wild for gender recognition and head pose estimation. Pattern Recognition Letters, 123, 104-110