Optimizing Diabetic Retinopathy Classification Using EfficientNet-B3 with Data Augmentation and Oversampling

A A JE Veggy Priyangka; Tuga Mauritsius

Authors

A A JE Veggy Priyangka Bina Nusantara University
Tuga Mauritsius Bina Nusantara University

Keywords:

Diabetic retinopathy, Deep Learning, EfficientNet-B3, Data Augmentation, random oversampling

Abstract

Diabetic retinopathy (DR) is a leading cause of preventable blindness among diabetes patients. This study optimizes DR severity classification using EfficientNet-B3 with transfer learning combined with data handling strategies. Using the APTOS 2019 dataset containing 3,662 retinal fundus images across five severity classes, three experimental scenarios were evaluated: (1) baseline CNN, (2) CNN with data augmentation, and (3) CNN with data augmentation and random oversampling. Performance was measured using Quadratic Weighted Kappa (QWK), accuracy, precision, recall, F1-score, and ROC-AUC. Results demonstrate that Scenario III achieves the best performance with QWK of 0.8496 and accuracy of 77.00%, representing significant improvement over baseline (QWK: 0.4998) and augmentation-only models (QWK: 0.5728). The combination of data augmentation and random oversampling effectively addresses class imbalance in medical image datasets. This study provides empirical evidence on combining transfer learning with data balancing strategies for automated DR screening systems.

Author Biographies

A A JE Veggy Priyangka, Bina Nusantara University

Information Systems Management Department

Tuga Mauritsius, Bina Nusantara University

Information Systems Management Department

References

[1] International Diabetes Federation, “IDF Diabetes Atlas 11th Edition,” https://diabetesatlas.org/resources/idf-diabetes-atlas-2025/.

[2] P. Saeedi, I. Petersohn, P. Salpea, B. Malanda, S. Karuranga, N. Unwin, S. Colagiuri, L. Guariguata, A. A. Motala, K. Ogurtsova, J. E. Shaw, D. Bright, and R. Williams, “Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition,” Diabetes Res. Clin. Pract., vol. 157, p. 107843, Nov. 2019, doi: 10.1016/j.diabres.2019.107843.

[3] T. Y. Wong, C. M. G. Cheung, M. Larsen, S. Sharma, and R. Simó, “Diabetic retinopathy,” Nat. Rev. Dis. Primers, vol. 2, no. 1, p. 16012, Mar. 2016, doi: 10.1038/nrdp.2016.12.

[4] C. J. Flaxel, R. A. Adelman, S. T. Bailey, A. Fawzi, J. I. Lim, G. A. Vemulakonda, and G. Ying, “Diabetic Retinopathy Preferred Practice Pattern®,” Ophthalmology, vol. 127, no. 1, pp. P66–P145, Jan. 2020, doi: 10.1016/j.ophtha.2019.09.025.

[5] R. E. K. Man, E. K. Fenwick, C. Sabanayagam, L. Li, C. S. Tey, H. J. T. Soon, G. C. M. Cheung, G. S. W. Tan, T. Y. Wong, and E. L. Lamoureux, “Differential Impact of Unilateral and Bilateral Classifications of Diabetic Retinopathy and Diabetic Macular Edema on Vision-Related Quality of Life,” Investigative Opthalmology & Visual Science, vol. 57, no. 11, p. 4655, Sep. 2016, doi: 10.1167/iovs.16-20165.

[6] D. S. Fong, L. P. Aiello, F. L. Ferris, and R. Klein, “Diabetic Retinopathy,” Diabetes Care, vol. 27, no. 10, pp. 2540–2553, Oct. 2004, doi: 10.2337/diacare.27.10.2540.

[7] R. Klein, “The Epidemiology of Diabetic Retinopathy,” in Diabetic Retinopathy, Totowa, NJ: Humana Press, 2008, pp. 67–107. doi: 10.1007/978-1-59745-563-3_3.

[8] World Health Organization, “World report on vision,” https://www.who.int/publications/i/item/9789241516570.

[9] A. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. Thrun, “Dermatologist-level classification of skin cancer with deep neural networks,” Nature, vol. 542, no. 7639, pp. 115–118, Feb. 2017, doi: 10.1038/nature21056.

[10] H. C. Shin, H. R. Roth, M. Gao, L. Lu, Z. Xu, I. Nogues, J. Yao, D. Mollura, and R. M. Summers, “Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning,” IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1285–1298, May 2016, doi: 10.1109/TMI.2016.2528162.

[11] M. Tan and Q. Le, “Efficientnet: Rethinking model scaling for convolutional neural networks,” in International conference on machine learning, PMLR, 2019, pp. 6105–6114.

[12] L. Arora, S. K. Singh, S. Kumar, H. Gupta, W. Alhalabi, V. Arya, S. Bansal, K. T. Chui, and B. B. Gupta, “Ensemble deep learning and EfficientNet for accurate diagnosis of diabetic retinopathy,” Sci. Rep., vol. 14, no. 1, p. 30554, Dec. 2024, doi: 10.1038/s41598-024-81132-4.

[13] Y. Zhou, B. Wang, L. Huang, S. Cui, and L. Shao, “A Benchmark for Studying Diabetic Retinopathy: Segmentation, Grading, and Transferability,” IEEE Trans. Med. Imaging, vol. 40, no. 3, pp. 818–828, Mar. 2021, doi: 10.1109/TMI.2020.3037771.

[14] M. A. Mazurowski, P. A. Habas, J. M. Zurada, J. Y. Lo, J. A. Baker, and G. D. Tourassi, “Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance,” Neural Networks, vol. 21, no. 2–3, pp. 427–436, Mar. 2008, doi: 10.1016/j.neunet.2007.12.031.

[15] B. Bektaş Güneş, B. Bulut, V. Kalın, and R. Khazhin, “Classification of Eye Disease from Fundus Images Using EfficientNet,” Artificial Intelligence Theory and Applications, vol. 2, no. 1, pp. 1–7, 2022, [Online]. Available: https://dergipark.org.tr/en/pub/aita/article/1134144

[16] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic Minority Over-sampling Technique,” Journal of Artificial Intelligence Research, vol. 16, pp. 321–357, Jun. 2002, doi: 10.1613/jair.953.

[17] V. Gulshan, L. Peng, M. Coram, M. C. Stumpe, D. Wu, A. Narayanaswamy, S. Venugopalan, K. Widner, T. Madams, J. Cuadros, R. Kim, R. Raman, P. C. Nelson, J. L. Mega, and D. R. Webster, “Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs,” JAMA, vol. 316, no. 22, p. 2402, Dec. 2016, doi: 10.1001/jama.2016.17216.

[18] H. Pratt, F. Coenen, D. M. Broadbent, S. P. Harding, and Y. Zheng, “Convolutional Neural Networks for Diabetic Retinopathy,” Procedia Comput. Sci., vol. 90, pp. 200–205, 2016, doi: 10.1016/j.procs.2016.07.014.

[19] R. Gargeya and T. Leng, “Automated Identification of Diabetic Retinopathy Using Deep Learning,” Ophthalmology, vol. 124, no. 7, pp. 962–969, Jul. 2017, doi: 10.1016/j.ophtha.2017.02.008.

[20] B. Tymchenko, P. Marchenko, and D. Spodarets, “Deep Learning Approach to Diabetic Retinopathy Detection,” Mar. 2020, [Online]. Available: http://arxiv.org/abs/2003.02261

[21] J. M. Johnson and T. M. Khoshgoftaar, “Survey on deep learning with class imbalance,” J. Big Data, vol. 6, no. 1, p. 27, Dec. 2019, doi: 10.1186/s40537-019-0192-5.

[22] C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, p. 60, Dec. 2019, doi: 10.1186/s40537-019-0197-0.

[23] H. He and E. A. Garcia, “Learning from Imbalanced Data,” IEEE Trans. Knowl. Data Eng., vol. 21, no. 9, pp. 1263–1284, Sep. 2009, doi: 10.1109/TKDE.2008.239.

[24] M. Buda, A. Maki, and M. A. Mazurowski, “A systematic study of the class imbalance problem in convolutional neural networks,” Neural Networks, vol. 106, pp. 249–259, Oct. 2018, doi: 10.1016/j.neunet.2018.07.011.

[25] R. Wirth and J. Hipp, "CRISP-DM: Towards a standard process model for data mining," in Proc. 4th Int. Conf. Practical Applications of Knowledge Discovery and Data Mining, Manchester, U.K., 2000, pp. 29–39.

[26] Asian Pacific Tele-Ophthalmology Society, “APTOS 2019 Blindness Detection,” https://www.kaggle.com/c/aptos2019-blindness-detection.

[27] J. R. Landis and G. G. Koch, “The Measurement of Observer Agreement for Categorical Data,” Biometrics, vol. 33, no. 1, p. 159, Mar. 1977, doi: 10.2307/2529310.

[28] G. Quellec, K. Charrière, Y. Boudi, B. Cochener, and M. Lamard, “Deep image mining for diabetic retinopathy screening,” Med. Image Anal., vol. 39, pp. 178–193, Jul. 2017, doi: 10.1016/j.media.2017.04.012.

[29] F. Arcadu, F. Benmansour, A. Maunz, J. Willis, Z. Haskova, and M. Prunotto, “Deep learning algorithm predicts diabetic retinopathy progression in individual patients,” NPJ Digit. Med., vol. 2, no. 1, p. 92, Sep. 2019, doi: 10.1038/s41746-019-0172-3.

[30] J. D. Bodapati, V. Naralasetti, S. N. Shareef, S. Hakak, M. Bilal, P. K. R. Maddikunta, and O. Jo, “Blended Multi-Modal Deep ConvNet Features for Diabetic Retinopathy Severity Prediction,” Electronics (Basel)., vol. 9, no. 6, p. 914, May 2020, doi: 10.3390/electronics9060914.

[31] C. J. Hellín, A. A. Olmedo, A. Valledor, J. Gómez, M. López-Benítez, and A. Tayebi, “Unraveling the Impact of Class Imbalance on Deep-Learning Models for Medical Image Classification,” Applied Sciences, vol. 14, no. 8, p. 3419, Apr. 2024, doi: 10.3390/app14083419.

[32] K. Ohri, M. Kumar, and D. Sukheja, “Self-supervised approach for diabetic retinopathy severity detection using vision transformer,” Progress in Artificial Intelligence, vol. 13, no. 3, pp. 165–183, Sep. 2024, doi: 10.1007/s13748-024-00325-0.