Exploring Dataset Variability in Diabetic Retinopathy Classification Using Transfer Learning Approaches

Kinjal Patni; Shruti Yagnik; Pratik Patel

doi:10.35882/jeeemi.v7i3.838

Kinjal Patni Department of Computer Engineering, Indus Institute of Technology and Engineering, Indus University, Ahmedabad, Gujarat, India
Shruti Yagnik Department of Computer Engineering, Indus Institute of Technology and Engineering, Indus University, Ahmedabad, Gujarat, India
Pratik Patel Department of Computer Science and Engineering, Parul Institute of Engineering and Technology, Parul University, Vadodara, Gujarat, India

DOI: https://doi.org/10.35882/jeeemi.v7i3.838

Keywords: Diabetic Retinopathy, Transfer Learning, Dataset Variability, Data Augmentation, Deep Learning Models

Abstract

Diabetic retinopathy (DR) stands as a primary international cause of vision impairment that needs effective and swift diagnostic services to protect eye structures from advancing deterioration. The variations of imaging data that appear between sources create major obstacles for achieving consistent performance from models. The elimination of performance fluctuation problems during DR classifications across two benchmark datasets EYE-PACS and APTOS is examined through systematic transfer learning analysis using different high-performing CNN architectures including VGG16, VGG19, ResNet50, Xception, InceptionV3, MobileNetV2, and InceptionResNetV2. The research evaluates how data heterogeneity affects and how augmentation approaches impact the accuracy while stabilizing robustness in deep learning models. The research provides new insights through its extensive investigation of generalization performance based on dataset changes which utilize modified data augmentation methods for retinal images. A collection of data transformations such as rotation, flipping, zooming and brightness modifications create simulated realistic scenarios to handle imbalanced data classes. Academic research involved CNN pre-training followed by transfer learning on both databases while researchers evaluated the models through both untreated source data and augmented image testing procedures. InceptionResNetV2 outperformed its counterparts with 96.2% accuracy and Xception delivered 95.7% accuracy in APTOS evaluation and both models scored 95.9% and 95.4% respectively on EYE-PACS testing. When augmentation was applied it increased the performance level by 3% to 5% across all running models. The experimental outcomes demonstrate how adequate variable training allows these models to recognize datasets regardless of their heterogeneity. This analysis confirms that combining reliable deep learning structures with purposeful data enhancement techniques substantially enhances DR diagnosis reliability to build scalable future diagnostic solutions for ophthalmology practice.

Downloads

Download data is not yet available.

Author Biographies

Kinjal Patni, Department of Computer Engineering, Indus Institute of Technology and Engineering, Indus University, Ahmedabad, Gujarat, India

Kinjal Patni, holding a Bachelor's and Master's degree in Computer Engineering, and currently pursuing a Ph.D. in the same field. With over eight years of academic and professional experience, I have consistently contributed to the domain of computer science and engineering through dedicated teaching, impactful research, and a strong commitment to innovation.

Shruti Yagnik, Department of Computer Engineering, Indus Institute of Technology and Engineering, Indus University, Ahmedabad, Gujarat, India

Dr. Shruti Yagnik, holding a Bachelor's degree in Computer Engineering, a Master's in Computer Science and Engineering (M. Tech), and a Ph.D. in Computer Engineering. With over more than 13 years of academic and professional experience, I have consistently contributed to the fields of computer science and engineering through teaching, research, and innovation.

Pratik Patel, Department of Computer Science and Engineering, Parul Institute of Engineering and Technology, Parul University, Vadodara, Gujarat, India

Dr. Pratik K. Patel, holding a Bachelor's degree in Computer Engineering, a Master's in Computer Science and Engineering (M. Tech), and a Ph.D. in Computer Engineering. With over 13 years of academic and professional experience, I have consistently contributed to the fields of computer science and engineering through teaching, research, and innovation.

References

S. Madhu, D. K. N. Bhargavi, M. V. S. Ramprasad, S. Gautam, and S. Bhavana, “Accurate diabetic retinopathy segmentation and classification model using gated recurrent unit with residual attention network,” Biomedical Signal Processing and Control, vol. 102, p. 107348, 2025, doi: 10.1016/j.bspc.2024.107348.

S. Sarmah, M. Hazarika, P. Das, A. Satheesh, D. Barman, and A. Kumar, “Advancements in Diabetic Retinopathy Detection Using Deep Learning,” in Studies in Computational Intelligence, vol. 1182, Springer, 2025, pp. 91–117. doi: 10.1007/978-3-031-80813-5_7.

K. Chaturvedi, V. Bhandari, A. Kothari, A. Tekerek, R. Tiwari, and R. Shrivastava, “An Intelligent Approach to Analyze Severity Levels of Diabetic Retinopathy by Data Classification Using Transfer Learning,” in International Conference on Data Science and Big Data Analysis, 2025, pp. 1–16.

S. Kollem, B. Preethi, G. K. Kumar, R. S. Prasad, K. Praneeth, and S. Peddakrishna, “Hybrid Deep Learning Model with ResNet50 and SVM for Diabetic Retinopathy Classification,” in 3rd International Conference on Electronics and Renewable Systems, ICEARS 2025 - Proceedings, 2025, pp. 1605–1610. doi: 10.1109/ICEARS64219.2025.10940726.

X. Wei et al., “MSTNet: Multi-scale spatial-aware transformer with multi-instance learning for diabetic retinopathy classification,” Medical Image Analysis, vol. 102, p. 103511, 2025, doi: 10.1016/j.media.2025.103511.

A. Ikram and A. Imran, “ResViT FusionNet Model: An explainable AI-driven approach for automated grading of diabetic retinopathy in retinal images,” Computers in Biology and Medicine, vol. 186, p. 109656, 2025, doi: 10.1016/j.compbiomed.2025.109656.

S. Guefrachi, A. Echtioui, and H. Hamam, “Diabetic Retinopathy Detection Using Deep Learning Multistage Training Method,” Arabian Journal for Science and Engineering, vol. 50, no. 2, pp. 1079–1096, 2024, doi: 10.1007/s13369-024-09137-9.

M. Z. Atwany, A. H. Sahyoun, and M. Yaqub, “Deep Learning Techniques for Diabetic Retinopathy Classification: A Survey,” IEEE Access, vol. 10, pp. 28642–28655, 2022, doi: 10.1109/ACCESS.2022.3157632.

V. Thanikachalam, K. Kabilan, and S. K. Erramchetty, “Optimized deep CNN for detection and classification of diabetic retinopathy and diabetic macular edema,” BMC Medical Imaging, vol. 24, no. 1, pp. 1–17, 2024, doi: 10.1186/s12880-024-01406-1.

Y. Yang, Z. Cai, S. Qiu, and P. Xu, “A Novel Transformer Model with Multiple Instance Learning for Diabetic Retinopathy Classification,” IEEE Access, vol. 12, no. December 2023, pp. 6768–6776, 2024, doi: 10.1109/ACCESS.2024.3351473.

A. Ikram et al., “A Systematic Review on Fundus Image-Based Diabetic Retinopathy Detection and Grading: Current Status and Future Directions,” IEEE Access, vol. 12, no. June, pp. 96273–96303, 2024, doi: 10.1109/ACCESS.2024.3427394.

C. Liu, W. Wang, J. Lian, and W. Jiao, “Lesion classification and diabetic retinopathy grading by integrating softmax and pooling operators into vision transformer,” Frontiers in Public Health, vol. 12, no. 2, 2024, doi: 10.3389/fpubh.2024.1442114.

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,” IEEE Journal of Translational Engineering in Health and Medicine, vol. 11, no. January, pp. 341–350, 2023, doi: 10.1109/JTEHM.2023.3282104.

M. A. K. Raiaan et al., “A Lightweight Robust Deep Learning Model Gained High Accuracy in Classifying a Wide Range of Diabetic Retinopathy Images,” IEEE Access, vol. 11, no. April, pp. 42361–42388, 2023, doi: 10.1109/ACCESS.2023.3272228.

R. B. Jayanthi Rajee, S. Mohamed Mansoor Roomi, V. Pooannamalai, and M. Parisa Begam, “A Transfer Learning Approach for Retinal Disease Classification,” Proceedings of 2023 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication, IConSCEPT 2023, pp. 1–6, 2023, doi: 10.1109/IConSCEPT57958.2023.10170532.

S. Nandhini, S. Sowbarnikkaa, J. Mageshwari, and C. Saraswathy, “An Automated Detection and Multi-stage classification of Diabetic Retinopathy using Convolutional Neural Networks,” ViTECoN 2023 - 2nd IEEE International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies, Proceedings, pp. 1–5, 2023, doi: 10.1109/ViTECoN58111.2023.10157960.

D. Raghu Raman, S. Nishanthi, and P. Babysha, “Diagnosis of Diabetic Retinopathy by using EfficientNet-B7 CNN Architecture in Deep Learning,” International Conference on Sustainable Computing and Smart Systems, ICSCSS 2023 - Proceedings, no. Icscss, pp. 430–435, 2023, doi: 10.1109/ICSCSS57650.2023.10169453.

T. Dharani, M. P. Prasamsa, B. L. Sirisha, J. B. Vivek, and B. H. Vardhan, “Diabetic Retinopathy classification through fundus images using Deep Learning,” 2023 11th International Symposium on Electronic Systems Devices and Computing, ESDC 2023, vol. 1, pp. 1–6, 2023, doi: 10.1109/ESDC56251.2023.10149877.

D. R. Dungrani, H. Rajesh Lotia, D. V. Parikh, A. S. Revathi, and K. Kavitha, “Detection and Classification of Diabetic Retinopathy using Deep Learning,” 5th Biennial International Conference on Nascent Technologies in Engineering, ICNTE 2023, no. Icnte, pp. 1–5, 2023, doi: 10.1109/ICNTE56631.2023.10146626.

S. Shah, S. Punjabi, S. Chavan, A. S. Revathi, and K. Kavitha, “A comparative study on Diabetic Retinopathy Detection and Classification,” 5th Biennial International Conference on Nascent Technologies in Engineering, ICNTE 2023, no. Icnte, pp. 1–6, 2023, doi: 10.1109/ICNTE56631.2023.10146636.

P. K. Das and S. Pumrin, “CNN Transfer Learning for Two Stage Classification of Diabetic Retinopathy using Fundus Images,” 8th International Conference on Digital Arts, Media and Technology and 6th ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering, ECTI DAMT and NCON 2023, pp. 443–447, 2023, doi: 10.1109/ECTIDAMTNCON57770.2023.10139437.

S. Ali and S. Raut, “Detection of Diabetic Retinopathy from fundus images using Resnet50,” 2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing, PCEMS 2023, pp. 1–5, 2023, doi: 10.1109/PCEMS58491.2023.10136073.

G. Kalyani, B. Janakiramaiah, A. Karuna, and L. V. N. Prasad, “Diabetic retinopathy detection and classification using capsule networks,” Complex and Intelligent Systems, vol. 9, no. 3, pp. 2651–2664, 2023, doi: 10.1007/s40747-021-00318-9.

M. V. Krishna and B. S. Rao, “Detection and Diagnosis of Diabetic Retinopathy Using Transfer Learning Approach,” International Journal of Intelligent Engineering and Systems, vol. 16, no. 3, pp. 62–74, 2023, doi: 10.22266/ijies2023.0630.05.

S. C. Pravin, S. P. K. Sabapathy, S. Selvakumar, S. Jayaraman, and S. V. Subramani, “An Efficient DenseNet for Diabetic Retinopathy Screening,” International Journal of Engineering and Technology Innovation, vol. 13, no. 2, pp. 125–136, 2023, doi: 10.46604/IJETI.2023.10045.

D. Das, S. K. Biswas, and S. Bandyopadhyay, “Detection of Diabetic Retinopathy using Convolutional Neural Networks for Feature Extraction and Classification (DRFEC),” Multimedia Tools and Applications, vol. 82, no. 19, pp. 29943–30001, 2023, doi: 10.1007/s11042-022-14165-4.

R. Shalini and S. Sasikala, “Artificial Intelligence Approach for Diabetic Retinopathy Severity Detection,” Informatica (Slovenia), vol. 46, no. 8, pp. 195–204, 2022, doi: 10.31449/inf.v46i8.4425.

M. Ragab, A. S. A. M. Al-Ghamdi, B. Fakieh, H. Choudhry, R. F. Mansour, and D. Koundal, “Prediction of Diabetes through Retinal Images Using Deep Neural Network,” Computational Intelligence and Neuroscience, vol. 2022, 2022, doi: 10.1155/2022/7887908.

P. Pragathi and A. Nagaraja Rao, “An effective integrated machine learning approach for detecting diabetic retinopathy,” Open Computer Science, vol. 12, no. 1, pp. 83–91, 2022, doi: 10.1515/comp-2020-0222.

K. Nirmala, K. Saruladha, and K. Dekeba, “Investigations of CNN for Medical Image Analysis for Illness Prediction,” Computational Intelligence and Neuroscience, vol. 2022, 2022, doi: 10.1155/2022/7968200.

B. Sumathy et al., “Prediction of diabetic Retinopathy using Health Records with Machine Learning Classifiers and data Science,” International Journal of Reliable and Quality E-Healthcare, vol. 11, no. 2, pp. 1–16, 2022, doi: 10.4018/IJRQEH.299959.

S. Gupta, S. Thakur, and A. Gupta, “Optimized Feature Selection Approach for Smartphone Based Diabetic Retinopathy Detection,” Proceedings of 2nd International Conference on Innovative Practices in Technology and Management, ICIPTM 2022, pp. 350–355, 2022, doi: 10.1109/ICIPTM54933.2022.9754021.

R. Yasashvini, V. Raja Sarobin M, R. Panjanathan, S. Graceline Jasmine, and L. Jani Anbarasi, “Diabetic Retinopathy Classification Using CNN and Hybrid Deep Convolutional Neural Networks,” Symmetry, vol. 14, no. 9, 2022, doi: 10.3390/sym14091932.

D. Nagpal, S. N. Panda, M. Malarvel, P. A. Pattanaik, and M. Zubair Khan, “A review of diabetic retinopathy: Datasets, approaches, evaluation metrics and future trends,” Journal of King Saud University - Computer and Information Sciences, vol. 34, no. 9, pp. 7138–7152, 2022, doi: 10.1016/j.jksuci.2021.06.006.

G. Kumar, S. Chatterjee, and C. Chattopadhyay, “DRISTI: a hybrid deep neural network for diabetic retinopathy diagnosis,” Signal, Image and Video Processing, vol. 15, no. 8, pp. 1679–1686, 2021, doi: 10.1007/s11760-021-01904-7.

L. Akshita, H. Singhal, I. Dwivedi, and P. Ghuli, “Diabetic retinopathy classification using deep convolutional neural network,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 24, no. 1, pp. 208–216, 2021, doi: 10.11591/ijeecs.v24.i1.pp208-216.

A. Sugeno, Y. Ishikawa, T. Ohshima, and R. Muramatsu, “Simple methods for the lesion detection and severity grading of diabetic retinopathy by image processing and transfer learning,” Computers in Biology and Medicine, vol. 137, no. August, p. 104795, 2021, doi: 10.1016/j.compbiomed.2021.104795.

M. Z. Atwany, A. H. Sahyoun, and M. Yaqub, “Deep Learning Techniques for Diabetic Retinopathy Classification: A Survey,” IEEE Access, vol. 10, pp. 28642–28655, 2022, doi: 10.1109/ACCESS.2022.3157632.

Kaggle, "APTOS 2019 Blindness Detection," [Online]. Available: https://www.kaggle.com/competitions/APTOS2019-blindness-detection/data. [Accessed: Jan. 26, 2025].

Kaggle, "EyePACS Diabetic Retinopathy Detection," [Online]. Available: https://www.kaggle.com/competitions/diabetic-retinopathy-detection. [Accessed: Jan. 26, 2025].