Diyabetik Retinopati Teşhisi için Fundus Görüntülerinin Derin Öğrenme Tabanlı Sınıflandırılması

Year 2021, , 156 - 167, 01.12.2021

Yusuf Bahri Özçelik , Aytaç Altan

https://doi.org/10.31590/ejosat.1011806

Cited By: 9

Abstract

Günümüzde en yaygın körlük nedenlerinden biri olan Diyabetik Retinopati (DR), gözün retina ağ tabakasında yer alan kan damarlarında diyabete bağlı olarak oluşan hasarlanmalardır. Hastaların görme yetisini kaybetmemesi için DR’nin erken teşhis ve tedavisi hayati önem taşımaktadır. Bu çalışmada, DR’nin erken teşhis ve tedavisi için fundus görüntüleri kullanılarak derin öğrenme tabanlı bir model geliştirilmiştir. Geliştirilen model iki aşamadan oluşmaktadır. İlk aşamada, modelin aşırı öğrenmesinin engellenebilmesi için fundus görüntülerine iki boyutlu sinyal işleme teknikleri uygulanmıştır. İkinci aşamada, derin öğrenme tekniklerinden Evrişimli Sinir Ağı (ESA) ve transfer öğrenmesi yöntemleri kullanılarak sınıflandırma modeli oluşturulmuştur. Modelin eğitiminde 5100 fundus görüntü verisi kullanılmıştır. Elde edilen model sağlıklı (DR yok), hafif Non-Proliferatif DR (NPDR), orta NPDR, şiddetli NPDR ve Proliferatif DR (PDR) gibi 5 sınıfı içeren 900 fundus görüntü verisi üzerinde test edilmiştir. Modelin sağlamlığı 10-kat çapraz doğrulama yöntemi kullanılarak doğrulanmıştır. Önerilen modelin sınıflandırma performansı %97.8 olarak ölçülmüştür. Ayrıca, modelin sınıflandırma performansı literatürde yer alan üç model ile kıyaslanmıştır. Elde edilen sonuçlar, önerilen modelin, DR’yi teşhis etmek için çok etkili ve başarılı olduğunu göstermektedir.

Keywords

Diyabetik retinopati, Derin öğrenme, Fundus, Sınıflandırma, Evrişimli sinir ağı, Transfer öğrenme

Deep Learning-based Classification of Fundus Images for the Diagnosis of Diabetic Retinopathy

Abstract

Diabetic Retinopathy (DR), one of the most common causes of blindness today, is damage to the blood vessels in the retinal mesh layer of the eye due to diabetes. Early diagnosis and treatment of DR is vital so that patients do not lose their sight. In this study, a deep learning-based model is developed using fundus images for the early diagnosis and treatment of DR. The developed model consists of two stages. In the first stage, two-dimensional signal processing techniques are applied to the fundus images to prevent overfitting of the model. In the second stage, the classification model is created by using deep learning techniques, Convolutional Neural Network (CNN) and transfer learning methods. 5100 fundus image data is used in the training of the model. The validity of the obtained model is tested on 900 fundus image data containing 5 classes such as No DR, mild Non-Proliferative DR (NPDR), moderate NPDR, severe NPDR and Proliferative DR (PDR). The robustness of the model is verified using the 10-fold cross validation method. The classification performance of the proposed model is measured as 97.8%. Moreover, the classification performance of the model is compared with the three models in the literature. The obtained results show that the proposed model is very effective and successful for diagnosing DR.

Keywords

Diabetic retinopathy, Deep learning, Fundus, Classification, Convolutional neural network, Transfer learning

References

• APTOS (2019). Blindness detection. URL: https://www.kaggle. com/c/aptos2019-blindness-detection.
• Aiello, L. M. (2003). Perspectives on diabetic retinopathy. American Journal of Ophthalmology, 136(1), 122-135.
• Antcliff, R. J., Stanford, M. R., Chauhan, D. S., Graham, E. M., Spalton, D. J., Shilling, J. S., & Marshall, J. (2000). Comparison between optical coherence tomography and fundus fluorescein angiography for the detection of cystoid macular edema in patients with uveitis. Ophthalmology, 107(3), 593-599.
• Chakraborty, S., Jana, G. C., Kumari, D., & Swetapadma, A. (2020). An improved method using supervised learning technique for diabetic retinopathy detection. International Journal of Information Technology, 12(2), 473-477.
• Chan, T. H., Jia, K., Gao, S., Lu, J., Zeng, Z., & Ma, Y. (2015). PCANet: A simple deep learning baseline for image classification. IEEE Transactions on Image Processing, 24(12), 5017-5032.
• Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1251-1258.
• Deepa, V., Kumar, C. S., & Andrews, S. S. (2021). Fusing dual‐tree quaternion wavelet transform and local mesh based features for grading of diabetic retinopathy using extreme learning machine classifier. International Journal of Imaging Systems and Technology, 31, 1625-1637.
• Dhakal, A., Bastola, L. P., & Shakya, S. (2019). Detection and classification of diabetic retinopathy using adaptive boosting and artificial neural network. International Journal of Advanced Research and Publications, 3(8), 191-196.
• Gayathri, S., Gopi, V. P., & Palanisamy, P. (2020). A lightweight CNN for Diabetic Retinopathy classification from fundus images. Biomedical Signal Processing and Control, 62, 102115, 1-11.
• Grossniklaus, H. E., Geisert, E. E., & Nickerson, J. M. (2015). Introduction to the retina. Progress in Molecular Biology and Translational Science, 134, 383-396.
• Han, D., Liu, Q., & Fan, W. (2018). A new image classification method using CNN transfer learning and web data augmentation. Expert Systems with Applications, 95, 43-56.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770-778.
• Hood, D. C., Raza, A. S., de Moraes, C. G. V., Liebmann, J. M., & Ritch, R. (2013). Glaucomatous damage of the macula. Progress in Retinal and Eye Research, 32, 1-21.
• Hossin, M., & Sulaiman, M. N. (2015). A review on evaluation metrics for data classification evaluations. International Journal of Data Mining & Knowledge Management Process, 5(2), 1-11.
• Indolia, S., Goswami, A. K., Mishra, S. P., & Asopa, P. (2018). Conceptual understanding of convolutional neural network-a deep learning approach. Procedia Computer Science, 132, 679-688.
• İnan, S. (2014). Retina anatomisi. Kocatepe Tıp Dergisi, 15(3), 355-359.
• Khan, S., Islam, N., Jan, Z., Din, I. U., & Rodrigues, J. J. C. (2019). A novel deep learning based framework for the detection and classification of breast cancer using transfer learning. Pattern Recognition Letters, 125, 1-6.
• Khan, M. A., Balgi, A. P., Chaithra, C., & Kumar, P. (2020). Diabetic retinopathy detection by image processing algorithms and machine learning technique. JNNCE Journal of Engineering & Management, 4(1), 8-16.
• Klein Kobrin, B. E. (2007). Overview of epidemiologic studies of diabetic retinopathy. Ophthalmic Epidemiology, 14(4), 179-183.
• Kramer, C. K., Rodrigues, T. C., Canani, L. H., Gross, J. L., & Azevedo, M. J. (2011). Diabetic retinopathy predicts all-cause mortality and cardiovascular events in both type 1 and 2 diabetes: meta-analysis of observational studies. Diabetes Care, 34(5), 1238-1244.
• La Cour, M., & Friis, J. (2002). Macular holes: classification, epidemiology, natural history and treatment. Acta Ophthalmologica Scandinavica, 80(6), 579-587.
• Math, L., & Fatima, R. (2021). Adaptive machine learning classification for diabetic retinopathy. Multimedia Tools and Applications, 80(4), 5173-5186.
• Nazir, T., Irtaza, A., Shabbir, Z., Javed, A., Akram, U., & Mahmood, M. T. (2019). Diabetic retinopathy detection through novel tetragonal local octa patterns and extreme learning machines. Artificial Intelligence in Medicine, 99, 101695, 1-10.
• Nida, N., Irtaza, A., Javed, A., Yousaf, M. H., & Mahmood, M. T. (2019). Melanoma lesion detection and segmentation using deep region based convolutional neural network and fuzzy C-means clustering. International Journal of Medical Informatics, 124, 37-48.
• Qummar, S., Khan, F. G., Shah, S., Khan, A., Shamshirband, S., Rehman, Z. U., Iftikhar, A. K, & Jadoon, W. (2019). A deep learning ensemble approach for diabetic retinopathy detection. IEEE Access, 7, 150530-150539.
• Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A. C., & Fei-Fei, L. (2015). Imagenet large scale visual recognition challenge. International Journal of Computer Vision, 115(3), 211-252.
• Saeedi, P., Petersohn, I., Salpea, P., Malanda, B., Karuranga, S., Unwin, N., Colagiuri, S., Guariguata, L., Motala, A. A., Ogurtsova, K., Shaw, J. E., Bright, D., Williams, R., & IDF Diabetes Atlas Committee. (2019). Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the international diabetes federation diabetes atlas. Diabetes Research and Clinical Practice, 157, 107843, 1-10.
• Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 1-14.
• Stratton, I. M., Kohner, E. M., Aldington, S. J., Turner, R. C., Holman, R. R., Manley, S. E., & Matthews, D. R. (2001). UKPDS 50: risk factors for incidence and progression of retinopathy in Type II diabetes over 6 years from diagnosis. Diabetologia, 44(2), 156-163.
• Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016). Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2818-2826.
• Şentürk, Z. K. (2020). Artificial neural networks based decision support system for the detection of diabetic retinopathy. Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 24(2), 424-431.
• Vijayan, T., Sangeetha, M., Kumaravel, A., & Karthik, B. (2020). Gabor filter and machine learning based diabetic retinopathy analysis and detection. Microprocessors and Microsystems, 103353, 1-8.
• Willoughby, C. E., Ponzin, D., Ferrari, S., Lobo, A., Landau, K., & Omidi, Y. (2010). Anatomy and physiology of the human eye: effects of mucopolysaccharidoses disease on structure and function–a review. Clinical & Experimental Ophthalmology, 38, 2-11.