An Enhanced Diabetic Retinopathy Classification Using ResNet-DenseNet Hybrid Model
An Enhanced Diabetic Retinopathy Classification Using ResNet-DenseNet Hybrid Model |
||
 |
 |
|
| © 2025 by IJETT Journal | ||
| Volume-73 Issue-8 |
||
| Year of Publication : 2025 | ||
| Author : Ashok Kumar Kavuru, Rajesh Kumar Patjoshi, Rakhee Panigrahi | ||
| DOI : 10.14445/22315381/IJETT-V73I8P123 | ||
How to Cite?
Ashok Kumar Kavuru, Rajesh Kumar Patjoshi, Rakhee Panigrahi,"An Enhanced Diabetic Retinopathy Classification Using ResNet-DenseNet Hybrid Model", International Journal of Engineering Trends and Technology, vol. 73, no. 8, pp.262-272, 2025. Crossref, https://doi.org/10.14445/22315381/IJETT-V73I8P123
Abstract
Diabetic Retinopathy is one of the blood vessel problems in the eye, which develops in people with uncontrolled diabetes, causing vision impairment worldwide. Early diagnosis of DR requires urgent attention because it helps to minimize the effects of this disease. Many deep learning methods have been proposed for classifying diabetic retinopathy, but developing models that are both effective and dependable requires a lot of knowledge and computing power. The proposed approach adopts ResNet and DenseNet deep learning architectures for the purpose of DR classification from retina fundus images. The enhancement of retinal fundus image features, including microaneurysms and hemorrhages during image preprocessing, is improved by using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Particle Swarm Optimization (PSO), which serves as a tool to perform hyperparameter tuning by optimizing learning rate together with dropout rate parameters to enhance the performance of the model. The combination methodology produces quicker model convergence with better generalization capabilities. The suggested model for finding diabetes retinopathy has a classification accuracy of 95.01% on the EyePACS dataset, which surpasses existing diagnostic systems. Hence, the proposed system is appropriate for large-scale DR screening.
Keywords
Diabetic Retinopathy, ResNet, DenseNet, CLAHE, Particle Swarm Optimization, Fundus images.
References
[1] James M. Breen et al., “501-P: Differences in Disease Severity and Treatment for Diabetes-Related Retinopathy between Type 1 and 2 Diabetes Mellitus,” Diabetes, vol. 72, no. 1, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Nicole Eichert et al., “60-OR: High Prevalence of Multiorgan Steatosis and Fibro Inflammation, Identified by Multiparametric Magnetic Resonance Imaging, in People with Type 2 Diabetes,” Diabetes, vol. 71, no. 1, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Moien Abdul Basith Khan et al., “Epidemiology of Type 2 Diabetes - Global Burden of Disease and Forecasted Trends,” Journal of Epidemiology and Global Health, vol. 10, no. 1, pp. 107-111, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Shi-Chue Hsing et al., “The Severity of Diabetic Retinopathy Is an Independent Factor for the Progression of Diabetic Nephropathy,” Journal of Clinical Medicine, vol. 10, no. 1, pp. 1-13, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Hamid Safi et al., “Early Detection of Diabetic Retinopathy,” Survey of Ophthalmology, vol. 63, no. 5, pp. 601-608, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Yijin Huang et al., “Identifying the Key Components in ResNet-50 for Diabetic Retinopathy Grading from Fundus Images: A Systematic Investigation,” arXiv Preprint, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Mohammad Z. Atwany, Abdulwahab H. Sahyoun, and Mohammad Yaqub, “Deep Learning Techniques for Diabetic Retinopathy Classification: A Survey,” IEEE Access, vol. 10, pp. 28642-28655, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Imran Qureshi, Jun Ma, and Qaisar Abbas, “Diabetic Retinopathy Detection and Stage Classification in Eye Fundus Images using Active Deep Learning,” Multimedia Tools and Applications, vol. 80, no. 8, pp. 11691-11721, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Sraddha Das et al., “Deep Learning Architecture based on Segmented Fundus Image Features for Classification of Diabetic Retinopathy,” Biomedical Signal Processing and Control, vol. 68, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Brahami Menaouer et al., “Diabetic Retinopathy Classification using Hybrid Deep Learning Approach,” SN Computer Science, vol. 3, no. 5, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Ling Dai et al., “A Deep Learning System for Detecting Diabetic Retinopathy Across the Disease Spectrum,” Nature Communications, vol. 12, no. 1, pp. 1-11, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Wejdan L. Alyoubi, Maysoon F. Abulkhair, and Wafaa M. Shalash, “Diabetic Retinopathy Fundus Image Classification and Lesions Localization System Using Deep Learning,” Sensors, vol. 21, no. 11, pp. 1-22, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Laxmi Math, and Ruksar Fatima, “Adaptive Machine Learning Classification for Diabetic Retinopathy,” Multimedia Tools and Applications, vol. 80, no. 4, pp. 5173-5186, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[14] S.G. Kobat et al., “Automated Diabetic Retinopathy Detection Using Horizontal and Vertical Patch Division-Based Pre-Trained DenseNET with Digital Fundus Images,” Diagnostics, vol. 12, no. 8, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Borys Tymchenko, Philip Marchenko, and Dmitry Spodarets, “Deep Learning Approach to Diabetic Retinopathy Detection,” arXiv Preprint, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Niloy Sikder et al., “Severity Classification of Diabetic Retinopathy using an Ensemble Learning Algorithm through Analyzing Retinal Images,” Symmetry, vol. 13, no. 4, pp. 1-26, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Tahira Nazir et al., “Detection of Diabetic Eye Disease from Retinal Images Using a Deep Learning Based CenterNet Model,” Sensors, vol. 21, no. 16, pp. 1-18, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Mohamed Chetoui, and Moulay A. Akhloufi, “Explainable Diabetic Retinopathy using EfficientNET,” 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada, pp. 1966-1969, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Gazala Mushtaq, and Farheen Siddiqui, “Detection of Diabetic Retinopathy using Deep Learning Methodology,” IOP Conference Series: Materials Science and Engineering, Tamil Nadu, India, vol. 1070, no. 1, pp. 1-13, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[20] San-Li Yi et al., “Diabetic Retinopathy Diagnosis Based on RA-EfficientNet,” Applied Science, vol. 11, no. 22, pp. 1-14, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Sandeep Madarapu, Samit Ari, and Kamalakanta Mahapatra, “A Multi-Resolution Convolutional Attention Network for Efficient Diabetic Retinopathy Classification,” Computers and Electrical Engineering, vol. 117, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Samia Akhtar et al., “A Classification Framework for Diabetic Retinopathy Detection Using Transfer Learning,” 2024 2nd International Conference on Cyber Resilience (ICCR), Dubai, United Arab Emirates, pp. 1-5, 2024.
[CrossRef] [Publisher Link]
[23] Diabetic Retinopathy Detection, 2025. [Online]. Available: https://www.kaggle.com/competitions/diabetic-retinopathy-detection
[24] Yuki Yoshimi et al., “Image Preprocessing with Contrast-Limited Adaptive Histogram Equalization Improves the Segmentation Performance of Deep Learning for the Articular Disk of the Temporomandibular Joint on Magnetic Resonance Images,” Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, vol. 138, no. 1, pp. 128-141, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Brett Koonce, Convolutional Neural Networks with Swift for Tensorflow, 1st ed., Apress Berkeley, CA, pp. 63-72, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Marwa M. Emam et al., “Breast Cancer Diagnosis using Optimized Deep Convolutional Neural Network based on Transfer Learning Technique and Improved Coati Optimization Algorithm,” Expert Systems with Applications, vol. 255, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Yuhan Bai, “RELU-Function and Derived Function Review,” 2022 International Conference on Science and Technology Ethics and Human Future (STEHF 2022), vol. 144, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Meetu Jain et al., “An Overview of Variants and Advancements of PSO Algorithm,” Applied Science, vol. 12, no. 17, pp. 1-21, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Juan Terven et al., “A Comprehensive Survey of Loss Functions and Metrics in Deep Learning,” Artificial Intelligence Review, vol. 58, no. 7, pp. 1-172, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Kele Xu, Dawei Feng, and Haibo Mi, “Deep Convolutional Neural Network-Based Early Automated Detection of Diabetic Retinopathy Using Fundus Image,” Molecules, vol. 22, no. 12, pp. 1-7, 2017
[CrossRef] [Google Scholar] [Publisher Link]
[31] Shu-I Pao et al., “Detection of Diabetic Retinopathy using Bichannel Convolutional Neural Network,” Journal of Ophthalmology, vol. 2020, no. 1, pp. 1-7, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[32] Borys Tymchenko, Philip Marchenko, and Dmitry Spodarets, “Deep Learning Approach to Diabetic Retinopathy Detection,” arXiv Preprint, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[33] Along He et al., “CABNet: Category Attention Block for Imbalanced Diabetic Retinopathy Grading,” IEEE Transactions on Medical Imaging, vol. 40, no. 1, pp. 143-153, 2021.
[CrossRef] [Google Scholar] [Publisher Link]