A Study of Continuous Regression for Diabetic Retinopathy Grading with Intra- and Cross-Dataset Evaluation

• Mateus Bastos Magalhães Mar UEA
• Diana Maria Da Câmara Gorayeb UEA

DOI: https://doi.org/10.5753/eniac.2026.31398

Resumo

Diabetic Retinopathy (DR) severity follows an ordinal progression, although it is often treated as a multiclass classification problem. This study presents an empirical comparison of binary, multiclass, and continuous regression-based formulations using EyePACS and APTOS 2019, under intra and cross-dataset settings. Results show that the linear formulation achieves higher Quadratic Weighted Kappa than softmax in the evaluated scenarios, while dataset characteristics substantially affect performance. The study provides a comparative analysis of these formulations under a unified experimental protocol, rather than proposing a new ordinal learning method.

Referências

Shearer, C. The New Blueprint for Data Mining. Journal of Data Warehousing, v. 5, n. 4, p. 13-22, 2000.

Patni, K., Yagnik, S., and Patel, P. (2025). Exploring dataset variability in diabetic retinopathy classification using transfer learning approaches. Journal of Electronics, Electromedical Engineering, and Medical Informatics, 7(3):763–777.

Chopra, M., Sparrenberg, L., Berger, A., Khanna, S., Terheyden, J. H., and Sifa, R. (2025). From retinal pixels to patients: Evolution of deep learning research in diabetic retinopathy screening. arXiv preprint arXiv:2511.11065.

Zhang, G.-H., Zhuo, G.-P., Zhang, Z.-X., Sun, B., Yang, W.-H., and Zhang, S.-C. (2024). Diabetic retinopathy identification based on multi-source-free domain adaptation. International Journal of Ophthalmology, 17(7):1193–1204.

Sugeno, A., Ishikawa, Y., Ohshima, T., and Muramatsu, R. (2021). Simple methods for the lesion detection and severity grading of diabetic retinopathy by image processing and transfer learning. Computers in Biology and Medicine, 137:104795.

Islam, M. R., Abdulrazak, L. F., Nahiduzzaman, M., Goni, M. O. F., Anower, M. S., Ahsan, M., Haider, J., and Kowalski, M. (2022). Applying supervised contrastive learning for the detection of diabetic retinopathy and its severity levels from fundus images. Computers in Biology and Medicine, 146:105602.

Saxena, G., Verma, D. K., Paraye, A., Rajan, A., and Rawat, A. (2020). Improved and robust deep learning agent for preliminary detection of diabetic retinopathy using public datasets. Intelligence-Based Medicine, 3–4:100022.

Katada, Y., Ozawa, N., Masayoshi, K., Ofuji, Y., Tsubota, K., and Kurihara, T. (2020). Automatic screening for diabetic retinopathy in interracial fundus images using artificial intelligence. Intelligence-Based Medicine, 3–4:100024.

Pak, A., Ziyaden, A., Tukeshev, K., Jaxylykova, A., and Abdullina, D. (2020). Comparative analysis of deep learning methods of detection of diabetic retinopathy. Cogent Engineering, 7(1):1805144.

Chong, D. D., Das, N., and Singh, R. P. (2024). Diabetic retinopathy: Screening, prevention, and treatment. Cleveland Clinic Journal of Medicine, 91(8):503–510.

Seo, H., Park, S.-J., and Song, M. (2025). Diabetic retinopathy (DR): Mechanisms, current therapies, and emerging strategies. Cells, 14(5):376.

Ansari, P., Tabasumma, N., Snigdha, N. N., Siam, N. H., Panduru, R. V. N. R. S., Azam, S., Hannan, J. M. A., and Abdel-Wahab, Y. H. A. (2022). Diabetic retinopathy: An overview on mechanisms, pathophysiology and pharmacotherapy. Diabetology, 3:159–175.

Yang, Z., Tan, T.-E., Shao, Y., Wong, T. Y., and Li, X. R. (2022). Classification of diabetic retinopathy: Past, present and future. Frontiers in Endocrinology, 13:1079217.

Mohanty, C., Mahapatra, S., Acharya, B., Kokkoras, F., Gerogiannis, V. C., Karamitsos, I., and Kanavos, A. (2023). Using deep learning architectures for detection and classification of diabetic retinopathy. Sensors, 23:5726.

Rom, Y., Aviv, R., Cohen, G. Y., Friedman, Y. E., Ianchulev, T., and Dvey-Aharon, Z. (2024). Diabetes detection from non-diabetic retinopathy fundus images using deep learning methodology. Heliyon, 10:e36592.

Bhoopal, S., Rao, M., and Krishnappa, C. H. (2024). Enhanced diabetic retinopathy detection and classification using fundus images with ResNet50 and CLAHE-GAN. Indonesian Journal of Electrical Engineering and Computer Science, 35(1):366–377.