Diabetic retinopathy detection models: Transition from conventional to deep learning approaches and a hybrid model

Öz

Diabetes is a chronic and common disease all over the world that has many side effects. One of the main side effects of this disease is diabetic retinopathy (DR), which if not diagnosed and treated in time, causes low vision and eventually blindness. Early diagnosis of this disease prevents the occurrence of complications of it. As artificial intelligence and deep learning technologies progress, various models for diagnosing medical images like fundus images in diabetic retinopathy became available. Due to the increase in the prevalence of DR, we propose a hybrid method with VGG-16 deep learning model and a hand-crafted feature extraction technique, namely Principal Component Analysis, for the classification of DR. To achieve this goal, two well-known datasets, namely APTOS 2019 and Messidor-2, are considered. On each of the datasets, two hand-crafted models, namely Principal Component Analysis (PCA) and Local Binary Patterns (LBP), and various deep learning models are implemented. In order to evaluate the models and compare them with each other and the state-of-the-art in this field, standard metrics such as accuracy are calculated. The results show that deep learning models are significantly superior to manual models regarding performance in diagnosing DR and its degrees. Moreover, the proposed hybrid method shows notable improvement in classification performance. Furthermore, the results obtained in this study have shown better performance than some other recent studies in the literature.

Anahtar Kelimeler

• Diabetic retinopathy
• Deep learning
• Medical image processing
• Convolutional neural network

Diyabetik retinopati algılama modelleri: Gelenekselden derin öğrenme yaklaşımlarına geçiş ve bir hibrit model

Öz

Diyabet, tüm dünyada yaygın ve kronik bir hastalıktır ve birçok yan etkiye sahiptir. Bu hastalığın başlıca yan etkilerinden biri, zamanında teşhis edilmez ve tedavi edilmezse düşük görmeye ve sonunda körlüğe neden olan diyabetik retinopatidir (DR). Bu hastalığın erken teşhisi, komplikasyonlarının ortaya çıkmasını önler. Yapay zekâ ve derin öğrenmeyle ilgili teknolojilerin ilerlemesiyle, diyabetik retinopatide fundus görüntüleri gibi tıbbi görüntüleri teşhis etmek için çeşitli modeller kullanılabilir hale gelmiştir. DR yaygınlığının artması nedeniyle, DR'nin sınıflandırılması için VGG-16 derin öğrenme modeli ve el-yapımı bir yöntem olan Principal Component Analysis ile hazırlanmış öznitelik çıkarma tekniklerini içeren hibrit bir yöntem öneriyoruz. Bu hedefe ulaşmak için, iyi bilinen iki veri kümesi, APTOS 2019 ve Messidor-2 dikkate alınmış ve her birinde, iyi bilinen elle hazırlanmıs¸ Principal Component Analysis (PCA) ve Local Binary Patterns (LBP) modelleri ve derin öğrenme modelleri uygulanmıştır. Modelleri birbirleriyle ve bu alanda yapılmış diğer çalışmalarla değerlendirmek için doğruluk gibi standart ölçütler hesaplanmıştır. Elde edilen sonuçlar, derin öğrenme modellerinin performans açısından DR ve derecelerini teşhis etmede geleneksel modellerden çok daha iyi olduğunu göstermiştir. Ayrıca, önerilen hibrit yöntem, sınıflandırma performansında dikkate değer bir iyileşme göstermektedir. Üstelik, bu çalışmada elde edilen sonuçlar, literatürdeki diğer bazı güncel çalışmalardan daha iyi performans göstermiştir.

Anahtar Kelimeler

• Diyabetik retinopati
• Derin öğrenme
• Tıbbi görüntü işleme
• Evrişimsel sinir ağı

Kaynakça

1. [1] Ting DS, Pasquale LR, Peng L, Campbell JP, Lee AY, Raman R, Tan GS, Schmetterer L, Keane PA, Wong TY. "Artificial intelligence and deep learning in ophthalmology". British Journal of Ophthalmology, 103(2), 167-175, 2019.
2. [2] Kaggle Ghosh S, Bhatia H. “APTOS 2019 Blindness Detection”. https://www.kaggle.com/c/aptos2019-blindness-detection/data (2019).
3. [3] Decencière E, Zhang X, Cazuguel G, Lay B, Cochene B, Trone C, Klein JC. “Feedback on a publicly distributed image database: the Messidor database”. Image Analysis & Stereology, 33, 231-234, 2014.
4. [4] Hotelling H. "Analysis of a complex of statistical variables into principal components". Journal of Educational Psychology, 24(6), 417-441, 1933.
5. [5] Ojala M, Pietikäinen M, Maenpaa T. "Multiresolution gray-scale and rotation invariant texture classification with local binary patterns". IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(7), 971-987, 2002.
6. [6] Krishna NV, Kotilingam KM, Premchand K, Iqbal MJ, Sudharshana KVS. "Diabetic retinopathy detection and classification using LBP and CNN". 2022 3rd International Conference for Emerging Technology (INCET), Belgaum, India, 27-29 May 2022.
7. [7] Simonyan K, Zisserman A. "Very Deep Convolutional Networks for Large-Scale Image Recognition". arXiv, 2014. https://arxiv.org/abs/1409.1556
8. [8] Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. "Rethinking the inception architecture for computer vision". Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, 27-30 June 2016.

9. [9] Howard AG, Sandler M, Chu G, Chen L, Chen Y. "Searching for MobileNetV3." arXiv, 2019. https://arxiv.org/abs/1905.02244
10. [10] Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. "Densely connected convolutional networks". IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 21-26 July 2017.
11. [11] LeCun Y, Bottou L, Bengio Y, Haffner P. "Gradient-based learning applied to document recognition". Proceedings of the IEEE, 86(11), 2278-2324, 1998.
12. [12] Alyoubi WL, Abulkhair MF, Shalash WM. "Diabetic retinopathy fundus image classification and lesions localization system using deep learning". Sensors, 21(11), 3704, 2021.
13. [13] Lahmar C, Idri A. "On the value of deep learning for diagnosing diabetic retinopathy". Health and Technology, 12(1), 89-105, 2022.
14. [14] Gadekallu TR, Khare N, Bhattacharya S, Singh S, Maddikunta PKR, Ra IH, Alazab M. "Early detection of diabetic retinopathy using PCA-firefly based deep learning model". Electronics, 9(2), 274, 2020.
15. [15] Al-Antary MT, Arafa Y. "Multi-scale attention network for diabetic retinopathy classification". IEEE Access, 9, 54190-54200, 2021.
16. [16] Alahmadi MD. "Texture attention network for diabetic retinopathy classification". IEEE Access, 10, 55522-55532, 2022.
17. [17] Gadekallu TR, Khare N, Bhattacharya S, Singh S, Maddikunta PKR, Srivastava G. "Deep neural networks to predict diabetic retinopathy". Journal of Ambient Intelligence and Humanized Computing, 14(5), 5407-5420, 2023.
18. [18] Usman TM, Saheed YK, Ignace D, Nsang A. "Diabetic retinopathy detection using principal component analysis multi-label feature extraction and classification". International Journal of Cognitive Computing in Engineering, 4, 78-88, 2023.
19. [19] AbdelMaksoud E, Barakat S, Elmogy M. "A multi-label computer-aided diagnoses system for detecting and diagnosing diabetic retinopathy". 2019 14th International Conference on Computer Engineering and Systems (ICCES), Cairo, Egypt, 17-18 December 2019.
20. [20] Gayathri S, Gopi VP, Palanisamy P. "Diabetic retinopathy classification based on multipath CNN and machine learning classifiers". Physical and Engineering Sciences in Medicine, 44(3), 639-653, 2021.
21. [21] Gurcan OF, Beyca ÖF, Dogan O. "A comprehensive study of machine learning methods on diabetic retinopathy classification". International Journal of Computational Intelligence Systems, 14(1), 1132-1141, 2021.
22. [22] Adriman R, Muchtar K, Maulina N. "Performance evaluation of binary classification of diabetic retinopathy through deep learning techniques using texture feature". Procedia Computer Science, 179, 88-94, 2021.
23. [23] Bodapati JD. "Adaptive ensembling of multi-modal deep spatial representations for diabetic retinopathy diagnosis". Multimedia Tools and Applications, 83(26), 68467–68486, 2024.
24. [24] Macsik P, Pavlovicova J, Goga J, Kajan S. "Local binary CNN for diabetic retinopathy classification on fundus images". Acta Polytechnica Hungarica, 19(7), 27-45, 2022.
25. [25] Shorfuzzaman M, Hossain MS, El Saddik A. "An explainable deep learning ensemble model for robust diagnosis of diabetic retinopathy grading". ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 17(3s), 1-24, 2021.
26. [26] Mushtaq G, Siddiqui F. "Detection of diabetic retinopathy using deep learning methodology". IOP Conference Series: Materials Science and Engineering, 1070(1), 012049, 2021.
27. [27] Sikder N, Chowdhury MS, Arif ASM, Nahid AA. "Early blindness detection based on retinal images using ensemble learning". 2019 22nd International Conference on Computer and Information Technology (ICCIT), Dhaka, Bangladesh, 18-20 December 2019.
28. [28] Zhang G, Sun B, Zhang Z, Pan J, Yang W, Liu Y. "Multi-model domain adaptation for diabetic retinopathy classification". Frontiers in Physiology, 13, 918929, 2022.
29. [29] Riaz H, Park J, Choi H, Kim H, Kim J. "Deep and densely connected networks for classification of diabetic retinopathy". Diagnostics, 10(1), 24, 2020.
30. [30] Saxena G, Verma DK, Paraye A, Rajan A, Rawat A. "Improved and robust deep learning agent for preliminary detection of diabetic retinopathy using public datasets". Intelligence-Based Medicine, 3, 100022, 2020.
31. [31] Kallel F, Echtioui A. "Retinal fundus image classification for diabetic retinopathy using transfer learning technique". Signal, Image and Video Processing, 18(2), 1143-1153, 2024.
32. [32] Çelik Ertuğrul D, Bitirim Y, Yakoub Anber B. "Decision support system for diagnosing diabetic retinopathy from color fundus images". Journal of Imaging Science and Technology, 64(2), 20502-1-20502-10, 2020.
33. [33] De la Calleja J, Tecuapetla L, Auxilio Medina M, Bárcenas E, Urbina Nájera AB. “LBP and machine learning for diabetic retinopathy detection”. International Conference on Intelligent Data Engineering and Automated Learning. Cham: Springer International Publishing, 10-12 September 2014.
34. [34] Jena PK, Khuntia B, Palai C, Nayak M, Mishra TK, Mohanty SN. “A novel approach for diabetic retinopathy screening using asymmetric deep learning features”. Big Data and Cognitive Computing, 7(1), 25, 2023.
35. [35] Özmen T, Kuzu Ü, Koçyiğit Y, Sarnel H. "Early stage diabetes prediction by features selection with metaheuristic methods". Pamukkale University Journal of Engineering Sciences, 29(6), 596-606, 2023.
36. [36] Altan G. "Breast cancer diagnosis using deep belief networks on ROI images". Pamukkale University Journal of Engineering Sciences, 28(2), 286-291, 2022.
37. [37] Afaneh A, Noroozi F, Toygar Ö. “Recognition of identical twins using fusion of various facial feature extractors”. EURASIP Journal on Image and Video Processing, 2017(1), 81, 2017.
38. [38] Vaighan LM, Jabbarbabouei Z, Uyguroğlu F, Toygar Ö. “Exploring deep learning architectures for multiple apple leaf disease classification”. In International Conference on Advanced Engineering, Technology and Applications. Cham: Springer Nature Switzerland, 24-25 May 2024.
39. [39] Tsiknakis N, Theodoropoulos D, Manikis G, Ktistakis E, Boutsora O, Berto A, Marias K. “Deep learning for diabetic retinopathy detection and classification based on fundus images: A review”. Computers in Biology and Medicine,135, 104599, 2021.
40. [40] Abràmoff MD, Folk JC, Han DP, Walker JD, Williams DF, Russell SR, Niemeijer M. “Automated analysis of retinal images for detection of referable diabetic retinopathy”. JAMA ophthalmology, 131(3), 351-357, 2013.