Attention Driven Deep Neural Network for Automated Diabetic Retinopathy Detection

This paper presented an attention-driven deep neural network model for automated diabetic retinopathy detection using retinal fundus images. The proposed approach integrates Convolutional Neural Networks (CNN) with an attention mechanism to enhance feature extraction by focusing on critical retinal lesion regions such as microaneurysms, hemorrhages, and exudates. This targeted feature learning significantly improves detection accuracy, sensitivity, and overall efficiency compared to conventional CNN-based methods. The experimental results demonstrate that the proposed model achieves high classification performance and effectively distinguishes between different stages of diabetic retinopathy. The incorporation of the attention mechanism enables the system to prioritize clinically relevant regions, thereby improving diagnostic reliability. This approach highlights the potential of deep learning-based systems in supporting early diagnosis and assistingophthalmologists in automated screening applications. In future work, the proposed system can be further enhanced by utilizing larger and more diverse datasets to improve model robustness and generalization capability. The development of real-time diabetic retinopathy detection systems can also be explored for practical clinical deployment. Furthermore, integrating the model with mobile and cloud-based healthcare applications can increase accessibility, especially in remote and resource-limited regions. Improving model interpretability through explainable AI techniques will enhance clinical trust and facilitate wider adoption in real-world medical diagnosis. Additionally, incorporating advanced architectures and hybrid models may further improve detection performance and efficiency

1] J. Gulshan et al., “Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs,” IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1266–1274, 2016. [2] H. Pratt, F. Coenen, D. Broadbent, S. Harding, and Y. Zheng, “Convolutional neural networks for diabetic retinopathy,” in Proc. IEEE Int. Conf. Image Process., 2016, pp. 1–5. [3] D. S. W. Ting et al., “Development and validation of a deep learning system for diabetic retinopathy and related eye diseases,” Lancet Digit. Health, vol. 1, no. 2, pp. e106–e114, 2019. [4] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2016, pp. 770–778. [5] A. Krizhevsky, I. Sutskever, and G. Hinton, “ImageNet classification with deep convolutional neural networks,” in Adv. Neural Inf. Process. Syst., 2012, pp. 1097–1105. [6] C. Szegedy et al., “Going deeper with convolutions,” in Proc. IEEE CVPR, 2015, pp. 1–9. [7] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint, 2014. [8] M. Tan and Q. Le, “EfficientNet: Rethinking model scaling for CNN,” in Proc. ICML, 2019, pp. 6105–6114. [9] A. Vaswani et al., “Attention is all you need,” in Adv. Neural Inf. Process. Syst., 2017, pp. 5998–6008. [10] S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “CBAM: Convolutional block attention module,” in Proc. ECCV, 2018, pp. 3–19.