The evaluation of the effect of data balancing over the classification performances of ensemble of networks for the diabetic retinopathy

Abstract

Diabetic retinopathy (DR) is a retinal condition that occurs due to diabetes mellitus and might lead to blindness. Early identification and treatment are crucial to slow down or prevent vision loss and degeneration. However, categorizing DR into several levels of severity remains a challenging problem due to the complexity of the disease. The Diabetic Retinopathy Grading System divides retinal pictures into five severity categories: No DR, Mild Non-Proliferative Diabetic Retinopathy (NPDR), Moderate NPDR, Severe NPDR, and Proliferative Diabetic Retinopathy. In this study, three deep learning models, namely ResNet50, Densenet201, and InceptionV3, were utilized for the classification of the APTOS 2019 diabetic retinopathy image dataset. For the individual experiments of the models, transfer learning with fine-tuning and layer freezing was applied. Additionally, a decision-level fusion idea using soft voting was implemented across the three pre-trained models. The maximum accuracy achieved for the classification of the original imbalanced dataset was 85% with the fusion idea. To further improve the classification performance, a balancing technique based on oversampling with augmentation operations was applied to the original APTOS 2019 dataset. The proposed approach, which involves the idea of soft voting-based fusion across models along with data balancing, improved the classification performance and achieved an accuracy of 90%.

Keywords

• Data Balancing
• Diabetic Retinopathy
• Ensemble of CNNs
• Soft Voting

References

1. REFERENCES
2. [1] Mohamed Q, Gillies MC, Wong TY. Management of diabetic retinopathy: a systematic review. JAMA 2007;298:902916. [CrossRef]
3. [2] Akram MU, Khalid S, Tariq A, Khan SA, Azam F. Detection and classification of retinal lesions for grading of diabetic retinopathy. Comput Biol Med 2007;45:161171. [CrossRef]
4. [3] Shorten C, Khoshgoftaar TM, Furht B. Deep learning applications for COVID-19. J Big Data 2021;8:154. [CrossRef]c
5. [4] Bodapati JD, Veeranjaneyulu N. Feature extraction and classification using deep convolutional neural networks. J Cyber Secur Mobil 2019;261276. [CrossRef]
6. [5] Li Y, Shen L. Skin lesion analysis towards melanoma detection using deep learning network. Sensors (Basel) 2018;18:556. [CrossRef]
7. [6] Minaee S, Kalchbrenner N, Cambria E, Nikzad N, Chenaghlu M, Gao J. Deep learning-based text classification: a comprehensive review. ACM Comput Surv 2021;54:140.
8. [7] Kussul N, Lavreniuk M, Skakun S, Shelestov A. Deep learning classification of land cover and crop types using remote sensing data. IEEE Geosci Remote Sens Lett 2017;14:778782. [CrossRef]

9. [8] Adnan MM, Rahim MSM, Rehman A, Mehmood Z, Saba T, Naqvi RA. Automatic image annotation based on deep learning models: a systematic review and future challenges. IEEE Access 2021;9:5025550256. [CrossRef]
10. [9] Amin J, Sharif M, Anjum MA, Raza M, Bukhari SAC. Convolutional neural network with batch normalization for glioma and stroke lesion detection using MRI. Cogn Syst Res 2020;59:304311. [CrossRef]
11. [10] Gayathri S, Gopi VP, Palanisamy P. Automated classification of diabetic retinopathy through reliable feature selection. Phys Eng Sci Med. 2020;43:927945. [CrossRef]
12. [11] Gayathri S, Krishna AK, Gopi VP, Palanisamy P. Automated binary and multiclass classification of diabetic retinopathy using Haralick and multiresolution features. IEEE Access 2020;8:5749757504.
13. [12] Pratt H, Coenen F, Broadbent DM, Harding SP, Zheng Y. Convolutional neural networks for diabetic retinopathy. Procedia Comput Sci 2016;90:200205. [CrossRef]
14. [13] Macsik P, Pavlovicova J, Goga J, Kajan S. Local binary CNN for diabetic retinopathy classification on fundus images. Acta Polytech Hung 2022;19:2931. [CrossRef]
15. [14] Sarki R, Michalska S, Ahmed K, Wang H, Zhang Y. Automatic detection of diabetic eye disease through deep learning using fundus images. bioRxiv 2020;8:763136. [CrossRef]
16. [15] Das D, Biswas SK, Bandyopadhyay S. Detection of diabetic retinopathy using convolutional neural networks for feature extraction and classification (DRFEC). Multimed Tools Appl 2022;82:2994330001. [CrossRef]
17. [16] Raja SMV, Panjanathan R. Diabetic retinopathy classification using CNN and hybrid deep convolutional neural networks. Symmetry (Basel) 2020;14:1932. [CrossRef]
18. [17] Patel R, Chaware A. Transfer learning with fine-tuned MobileNetV2 for diabetic retinopathy. 2020 Int Conf Emerg Technol (INCET). 2020;14. [CrossRef]
19. [18] Alyoubi WL, Abulkhair MF, Shalash WM. Diabetic retinopathy fundus image classification and lesions localization system using deep learning. Sensors (Basel) 2021;21:3704. [CrossRef]
20. [19] Bodapati JD, Shaik NS, Naralasetti V. Composite deep neural network with gated-attention mechanism for diabetic retinopathy severity classification. J Ambient Intell Humaniz Comput 2021;12:98259839. [CrossRef]
21. [20] Agus EM, Mochammad HCM, Yufis A, Fitri B, Hanung AN, Zaidah I. Classification of diabetic retinopathy disease using convolutional neural network. Int J Inform Vis 2022;6:1218. [CrossRef]
22. [21] Khalifa NEM, Loey M, Taha MHN, Mohamed HNET. Deep transfer learning models for medical diabetic retinopathy detection. Acta Inform Med 2019;27:327. [CrossRef]
23. [22] Sikder N, Masud M, Bairagi AK, Arif ASM, Nahid AA, Alhumyani HA. Severity classification of diabetic retinopathy using an ensemble learning algorithm through analyzing retinal images. Symmetry (Basel) 2021;13:670. [CrossRef]
24. [23] Rodriguez JD, Perez A, Lozano JA. Sensitivity analysis of k-fold cross validation in prediction error estimation. IEEE Trans Pattern Anal Mach Intell 2009;32:569575. [CrossRef]
25. [24] Perez L, Wang J. The effectiveness of data augmentation in image classification using deep learning. arXiv 2017;1712.04621.
26. [25] Yüzkat M, Ilhan HO, Aydin N. Multi-model CNN fusion for sperm morphology analysis. Comput Biol Med 2021;137:104790. [CrossRef]
27. [26] Kieu LM, Ou Y, Truong LT, Cai CA. Class-specific soft voting framework for customer booking prediction in on-demand transport. Transp Res C Emerg Technol 2020;114:377390. [CrossRef]
28. [27] Wang L, Schaefer A. Diagnosing diabetic retinopathy from images of the eye fundus. CS230. Stanford. Edu.
29. [28] Pak A, Ziyaden A, Tukeshev K, Jaxylykova A, Abdullina D. Comparative analysis of deep learning methods of detection of diabetic retinopathy. Cogent Eng 2020;7:1805144. [CrossRef]
30. [29] Li Y, Hsu JS, Bari N, Qiu X, Viswanathan M, Shi W, et al. Interpretable evaluation of diabetic retinopathy grade regarding eye color fundus images. 2022 IEEE Int Conf Biomed Eng Informat (BIBE) 2022;1116. [CrossRef]
31. [30] Lazuardi RN, Abiwinanda N, Suryawan TH, Hanif M, Handayani A. Automatic diabetic retinopathy classification with EfficientNet. 2020 IEEE Reg 10 Conf (TENCON). 2020;10:756760. [CrossRef]
32. [31] Oulhadj M, Riffi J, Chaimae K, Mahraz AM, Ahmed B, Yahyaouy A, et al. Diabetic retinopathy prediction based on deep learning and deformable registration. Multimed Tools Appl 2022;81:2870928727. [CrossRef]