HAF-Net: Hybrid Attention Fusion Network for Diabetic Retinopathy Classification
Article Sidebar
Main Article Content
Abstract
Diabetic retinopathy (DR) is a serious diabetes complication that can cause blindness if not treated promptly. Accurate DR classification plays an important role in supporting diagnosis. To improve classification performance, this research proposed a hybrid attention fusion network (HAF-Net) for DR classification. The HAF-Net integrates a convolutional operator with a convolutional block attention module (CBAM) and a bottleneck attention module (BAM) due to their advantages in enhancing feature representation, focusing on both local and global contextual information. These strengths allow the network to improve its ability to distinguish between subtle differences and help gather important features in medical images. A residual layer is also added to compensate for feature loss during training. A hybrid fusion of multi-layer perceptron (MLP), long short-term memory (LSTM), and bidirectional LSTM (BiLSTM) features are employed to achieve DR classification. The HAF-Net model is tested on a retinal fundus image dataset published on Kaggle for DR classification research, covering 11 disease types. The proposed HAF-Net achieved a training accuracy of 99.35% with a training loss of 0.0129. The hybrid model-MLP achieved the highest validation accuracy of 76.41% and test accuracy of 80.17%, demonstrating the best generalization performance. The results indicate that the HAF-Net achieves competitive classification performance compared with previous studies and represents a promising approach for automated DR classification.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
M. Duggal et al., “Real-world evaluation of AI-driven diabetic retinopathy screening in public health settings: Validation and implementation study,” JMIR Medical Informatics, vol. 13, 2025.
K. Lee, G. H. Lee, S. E. Lee, J. M. Yang and K. Bae, “Glycemic control and retinal microvascular changes in type 2 diabetes mellitus patients without clinical retinopathy,” Diabetes & Metabolism Journal, vol. 48, no. 5, pp. 983–992, 2024.
Y. Zhang et al., “Serum disease-specific IgG Fc glycosylation as potential biomarkers for nonproliferative and proliferative diabetic retinopathy using mass spectrometry,” Molecular & Cellular Proteomics, no. 109555, 2023.
X. Ding et al., “Expanded field OCT angiography biomarkers for predicting clinically significant outcomes in non-proliferative diabetic retinopathy,” American Journal of Ophthalmology, vol. 270, pp. 216–226, 2025.
D. S. Pushparani, J. Varalakshmi, K. Roobini, P. Hamshapriya and A. Livitha, “Diabetic retinopathy—A review,” Current Diabetes Reviews, vol. 21, no. 7, pp. 43–55, 2025.
P. Ansari, N. Tabasumma, N. N. Snigdha, N. H. Siam, R. V. N. R. S. Panduru, S. Azam, J. M. A. Hannan and Y. H. A. Abdel-Wahab, “Diabetic retinopathy: An overview on mechanisms, pathophysiology and pharmacotherapy,” Diabetology, vol. 3, no. 1, pp. 159–175, 2022.
C. Sen, S. Doganay, and G. Ozcan, “New accurate deep learning model for diabetic retinopathy detection utilizing sequential pre-processing and transfer learning,” Biomed. Signal Process. Control, vol. 109, no. 108060, 2025.
P. Hariobulesu and F. Shaik, “Enhanced multigrade diabetic retinopathy detection and classification via ensembled deep learning model from retinal fundus images,” Expert Systems with Applications, vol. 285, no. 128116, 2025.
E.-M. Chakour, Z. Sadok, R. Kachouri, A. Mansouri, I. B. Andaloussi, M. Akil and A. Ahaitouf, “Mobile-based deep learning system for early detection of diabetic retinopathy,” IntelligenceBased Medicine, vol. 12, no. 100259, 2025.
S. Alanazi and R. Alanazi, “Enhancing diabetic retinopathy detection through federated convolutional neural networks: Exploring different stages of progression,” Alexandria Engineering Journal, vol. 120, pp. 215–228, 2025.
G. M. Ramadan, M. M. Adnan, Z. L. Naser, B. V. Kumar and K. Parthiban, “Diabetic Retinopathy Segmentation and Classification Using U-Net with Convolutional Neural Network,” 2023 3rd International Conference on Mobile Networks and Wireless Communications (ICMNWC), Tumkur, India, pp. 1-5, 2023.
A. M. A and S. S. S. Priya, “Detection and Classification of Diabetic Retinopathy Using Pretrained Deep Neural Networks,” 2023 International Conference on Innovations in Engineering and Technology (ICIET), Muvattupuzha, India, pp. 1-7, 2023.
F. Saeed, M. Hussain and H. A. Aboalsamh, “Automatic diabetic retinopathy diagnosis using adaptive fine-tuned convolutional neural network,” in IEEE Access, vol. 9, pp. 41344–41359, 2021.
G. Ali, A. Dastgir, M. W. Iqbal, M. Anwar and M. Faheem, “A Hybrid Convolutional Neural Network Model for Automatic Diabetic Retinopathy Classification From Fundus Images,” in IEEE Journal of Translational Engineering in Health and Medicine, vol. 11, pp. 341350, 2023
S. Y. Prasetyo, A. Wulandari, Z. N. Izdihar, G. Z. Nabiilah and H. A. Saputri, “Comparing convolutional neural network models with MobileNet-based in diabetic retinopathy classification,” in Proceedings International Workshop on Artificial Intelligence and Image Processing (IWAIIP), Yogyakarta, Indonesia, pp. 1–5, 2023.
A. M. Pour, H. Seyedarabi, S. H. A. Jahromi and A. Javadzadeh, “Automatic detection and monitoring of diabetic retinopathy using efficient convolutional neural networks and contrast limited adaptive histogram equalization,” in IEEE Access, vol. 8, pp. 136668–136673, 2020.
X. Luo et al., “A Lesion-Fusion Neural Network for Multi-View Diabetic Retinopathy Grading,” in IEEE Journal of Biomedical and Health Informatics, vol. 29, no. 5, pp. 3184-3193, May 2025.
Y. Niu, L. Gu, Y. Zhao and F. Lu, “Explainable Diabetic Retinopathy Detection and Retinal Image Generation,” in IEEE Journal of Biomedical and Health Informatics, vol. 26, no. 1, pp. 44-55, Jan. 2022.
S. Madarapu, S. Ari and K. Mahapatra, “C2xFNet: Cascaded Dense Block With Twofold Cross-Feature Enhancement Module for Diabetic Retinopathy Grading,” in IEEE Transactions on Instrumentation and Measurement, vol. 74, pp. 1-10, 2025.
H. Mustafa, S. F. Ali, M. Bilal and M. S. Hanif, “Multi-stream deep neural network for diabetic retinopathy severity classification under a boosting framework,” in IEEE Access, vol. 10, pp. 113172–113183, 2022.
R. Rashid, W. Aslam, A. Mehmood, D. L. R. Vargas, I. D. L. T. Diez and I. Ashraf, “A Detectability Analysis of Retinitis Pigmetosa Using Novel SE-ResNet Based Deep Learning Model and Color Fundus Images,” in IEEE Access, vol. 12, pp. 28297-28309, 2024.
M. M. Farag, M. Fouad and A. T. Abdel-Hamid, “Automatic Severity Classification of Diabetic Retinopathy Based on DenseNet and Convolutional Block Attention Module,” in IEEE Access, vol. 10, pp. 38299-38308, 2022.
M. U. Yaseen, M. M. Nasralla, F. Aslam, S. S. Ali and S. B. A. Khattak, “A Novel Approach Based on Multi-Level Bottleneck Attention Modules Using Self-Guided Dropblock for Person Re-Identification,” in IEEE Access, vol. 10, pp. 123160-123176, 2022.
K. Bharatheedasan, T. Maity, L. A. Kumaraswamidhas and M. Durairaj, “Enhanced fault diagnosis and remaining useful life prediction of rolling bearings using a hybrid multilayer perceptron and LSTM network model,” Alexandria Engineering Journal, vol. 115, pp. 355–369, 2025.
A. R. Zangene, O. Williams Samuel, A. Abbasi, K. Nazarpour, A. A. McEwan and G. Li, “An Attention-based Bidirectional LSTM Model for Continuous Cross-Subject Estimation of Knee Joint Angle during Running from sEMG Signals,” 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Sydney, Australia, pp. 14, 2023.
J. Huang, M. Lin, J. Li and G. Liu, “Based on MLP and BiLSTM battery capacity forecast method research,” 2024 4th International Conference on Intelligent Power and Systems (ICIPS), Yichang, China, pp. 407-411, 2024.
K. S. Sanjay Nithish, “Retinal fundus images,” Kaggle, 2023. [Online]. Available: https://www.kaggle. com/datasets/kssanjaynithish03/retinal-fundus-images. [Accessed: Jul. 2, 2025].
C. Wan, J. Zhao, X. Hong, W. Yang, and S. Zhang, “HCSP-Net: A novel model of age-related macular degeneration classification based on color fundus photography,” Computers, Materials and Continua, vol. 79, no. 1, pp. 391–407, 2024.
M. M. Hemal and S. Saha, “Explainable deep learning-based meta-classifier approach for multi-label classification of retinal diseases,” Array, vol. 26, Art. no. 199402, 2025.
W. Nazih, A. O. Aseeri, O. Y. Atallah, and S. ElSappagh, “Vision transformer model for predicting the severity of diabetic retinopathy in fundus photography-based retina images,” in IEEE Access, vol. 11, pp. 117546–117561, 2023.
A. Jabbar, H. B. Liaqat, A. Akram, M. U. Sana, I. D. Azp´ıroz, and I. De La Torre Diez, “A lesion-based diabetic retinopathy detection through hybrid deep learning model,” in IEEE Access, vol. 12, pp. 40019–40036, 2024.
A. J. Kadhim, H. Seyedarabi, R. Afrouzian, and F. S. Hasan, “Diabetic retinopathy classification using hybrid color-based CLAHE and blood vessel in deep convolution neural network,” in IEEE Access, vol. 12, pp. 194750–194761, 2024.
W. K. Wong, F. H. Juwono, and C. Apriono, “Diabetic retinopathy detection and grading: A transfer learning approach using simultaneous parameter optimization and feature-weighted ECOC ensemble,” in IEEE Access, vol. 11, pp. 83004–83016, 2023.
A. Raza, S. Bin Musa, A. S. Bin Khalid, M. M. Alam, M. M. Su’ud, and F. Noor, “Enhancing medical image classification through PSO-optimized dual deterministic approach and robust transfer learning,”in IEEE Access, vol. 12, pp. 177144–177159, 2024.
M. Shafiq, Q. Fan, F. H. Alghamedy, and W. J. Obidallah, “DualEye-FeatureNet: A dual-stream feature transfer framework for multimodal ophthalmic image classification,” in IEEE Access, vol. 12, pp. 143985–144008, 2024.
