Deep Learning-Based Automated Grading of Diabetic Retinopathy Using EfficientNet-B4 with Attention Mechanisms and Ensemble Fusion
.png)
.png)
Abstract
Diabetic retinopathy (DR) is the leading cause of preventable blindness among working-age adults worldwide, with India alone estimated to have over 11.9 million individuals with vision-threatening DR as of 2023. Early detection through systematic fundus screening can prevent up to 95% of severe vision loss; however, the global shortage of trained ophthalmologists — particularly acute in rural and semi-urban India — creates a critical bottleneck in DR screening programmes. Artificial intelligence-based automated DR grading presents a compelling solution, capable of screening large volumes of fundus images accurately and consistently without specialist involvement.
This study develops and evaluates a deep learning ensemble system for automated 5-class DR grading (No DR, Mild, Moderate, Severe, Proliferative DR) using EfficientNet-B4 as the primary backbone with dual attention mechanisms — Squeeze-and-Excitation (SE) networks for channel attention and Convolutional Block Attention Module (CBAM) for spatial attention. The ensemble integrates EfficientNet-B4, ResNet-50, and DenseNet-121 predictions through learned weighted averaging. Training and evaluation employ three publicly available benchmark datasets: APTOS 2019 Blindness Detection, IDRiD, and Messidor-2, comprising 7,842 fundus images. Focal loss (γ=2) and SMOTE oversampling address class imbalance. The proposed ensemble achieves macro-average AUC of 0.983, quadratic-weighted kappa of 0.921, sensitivity of 93.6%, and specificity of 96.8% across five DR grades. Gradient-weighted Class Activation Mapping (Grad-CAM) visualisations confirm that the model correctly attends to clinically relevant pathological features including microaneurysms, haemorrhages, hard exudates, and neovascularisation, providing ophthalmologist-interpretable explainability.
Keywords: diabetic retinopathy, deep learning, EfficientNet-B4, attention mechanism, ensemble learning, fundus image analysis, automated grading, Grad-CAM, focal loss, explainable AI, screening, ophthalmology, India
References
- [1] Gulshan, V., Peng, L., Coram, M., et al. (2016). Development and validation of a deep learning algorithm for detecting diabetic retinopathy. JAMA, 316(22), 2402–2410.
- [2] IDF. (2022). IDF Diabetes Atlas (10th ed.). International Diabetes Federation.
- [3] Selvaraju, R. R., Cogswell, M., Das, A., et al. (2017). Grad-CAM: Visual explanations from deep networks. Proceedings of ICCV 2017, 618–626.
- [4] Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for CNNs. Proceedings of ICML 2019, 6105–6114.
- [5] Ting, D. S. W., Cheung, C. Y. L., Lim, G., et al. (2017). Development and validation of a deep learning system for diabetic retinopathy in retinal fundus images. JAMA, 318(22), 2211–2223.
- [6] Woo, S., Park, J., Lee, J. Y., & Kweon, I. S. (2018). CBAM: Convolutional Block Attention Module. Proceedings of ECCV 2018, 3–19.
- [7] Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-Excitation Networks. Proceedings of CVPR 2018, 7132–7141.
- [8] Porwal, P., Pachade, S., Kamble, R., et al. (2018). Indian diabetic retinopathy image dataset (IDRiD). IEEE Dataport.
- [9] APTOS 2019 Blindness Detection. (2019). Kaggle Competition Dataset.
- [10] Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017). Focal loss for dense object detection. Proceedings of ICCV 2017, 2980–2988.
Explore Our Related Journals
Looking for the right journal for your next manuscript? Explore our international peer-reviewed journals covering engineering, management, computer science, artificial intelligence and multidisciplinary research.