Diabetic Retinopathy Detection via PCA-MLP Using Independent GLCM and EfficientNetB3 Descriptors
DOI:
https://doi.org/10.31849/digitalzone.v17i1.33099Keywords:
Diabetic Retinopathy, Fundus Image, Feature Fusion, Multi-Layer Perceptron, Principal Component Analysis, EfficientNetB3Abstract
Diabetic Retinopathy (DR) is a microvascular complication of diabetes and a leading cause of preventable blindness. Manual diagnosis through fundus imagery is time-consuming and requires highly specialized expertise. This study proposes an automated DR detection system through a comparative multi-scenario framework that evaluates independent clinical and semantic descriptors using a Multi-Layer Perceptron (MLP) network. To prevent majority class bias, a random undersampling technique was applied to the MESSIDOR dataset, resulting in a balanced dataset of 700 fundus images (350 Grade 0 and 350 Grade 1 instances). The system implements a dual-stream preprocessing pipeline utilizing Ben Graham's method for illumination standardization and Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance clinical lesions. This study independently analyzes texture representations from the Gray Level Co-occurrence Matrix (GLCM) and deep semantic features from a pre-trained EfficientNetB3 model. To mitigate computational overhead and explicitly prevent data leakage, feature standardization and Principal Component Analysis (PCA) were strictly isolated and executed within the training phase of each fold during the 5-Fold Cross-Validation protocol. The evaluation demonstrated that the proposed comparative framework achieves robust computational efficiency alongside a promising diagnostic testing accuracy of 97.00%. While these empirical results indicate strong internal reliability with minimal prediction errors, further evaluation involving external validation datasets is recommended to fully substantiate its utility as a supportive clinical screening tool.
References
[1] M. M. Islam et al., “Predicting the risk of diabetic retinopathy using explainable machine learning algorithms,” Diabetes Metab. Syndr. Clin. Res. Rev., vol. 17, no. 12, 2023, https://doi.org/10.1016/j.dsx.2023.102919
[2] N. M. Makmur, F. Kwan, A. Dewi, and F. Indra, “Comparing Local Binary Pattern and Level Co-occurrence Matrix for Feature Extraction in Diabetic Retinopathy Classification,” Procedia Comput. Sci., vol. 227, pp. 355–363, 2023, https://doi.org/10.1016/j.procs.2023.10.534
[3] X. Tang, Y. Yang, L. Huang, and L. Qiu, “The Application of Texture Feature Analysis of Rectus Femoris Based on Local Binary Pattern (LBP) Combined With Gray-Level Co-Occurrence Matrix (GLCM) in Sarcopenia,” J. Ultrasound Med., pp. 2169–2179, 2022, https://doi.org/10.1002/jum.15896
[4] J. Engelmann, A. Villaplana-Velasco, A. Storkey, and M. O. Bernabeu, “Robust and Efficient Computation of Retinal Fractal Dimension Through Deep Approximation,” in International Workshop on Ophthalmic Medical Image Analysis, 2022, pp. 84–93. https://doi.org/10.1007/978-3-031-16525-2
[5] A. Shamsan, E. M. Senan, H. Salameh, and H. S. A. Shatnawi, “Predicting of diabetic retinopathy development stages of fundus images using deep learning based on combined features,” PLoS One, pp. 1–26, 2023, https://doi.org/10.1371/journal.pone.0289555
[6] P. Macsik, J. Pavlovicova, J. Goga, and S. Kajan, “Local Binary CNN for Diabetic Retinopathy Classification on Fundus Images,” Acta Polytech. Hungarica, vol. 19, no. 7, pp. 27–45, 2022, https://doi.org/10.12700/APH.19.7.2022.7.2
[7] S. G. Ali, X. Wang, L. Bi, Y. Jung, T. Chen, and H. Zhang, “Deep learning-based binocular system for automated diabetic retinopathy grading with prior clinical knowledge integration,” Vis. Comput., vol. 41, no. 8, pp. 5675–5688, 2025, https://doi.org/10.1007/s00371-024-03745-0
[8] N. M. Suhaimi et al., “Comparative Analysis of Superpixel and Gabor Methods for Exudate Feature Extraction in Diabetic Retinopathy Fundus Images,” in Proceedings of the 7th International Conference on Electrical, Control and Computer Engineering, 2024, vol. 2, pp. 123–136. https://doi.org/10.1007/978-981-97-3851-9
[9] I. Ahmad, V. Prakash, S. Manoj, and M. Gore, “Segmented Fractal and Central Symmetric LBP Based Texture Features for the Detection of Diabetic Retinopathy Using SVM,” SN Comput. Sci., vol. 5, pp. 1–14, 2024, https://doi.org/10.1007/s42979-024-02996-x
[10] J. Du, B. Zou, P. Ouyang, and R. Zhao, “Retinal microaneurysm detection based on transformation splicing and multi-context ensemble learning,” Biomed. Signal Process. Control, vol. 74, no. November 2021, 2022, https://doi.org/10.1016/j.bspc.2022.103536
[11] S. Gupta, S. Thakur, and A. Gupta, Comparative study of different machine learning models for automatic diabetic retinopathy detection using fundus image, vol. 83, no. 12. Springer US, 2024. https://doi.org/10.1007/s11042-023-16813-9
[12] N. Gundluru et al., “Enhancement of Detection of Diabetic Retinopathy Using Harris Hawks Optimization with Deep Learning Model,” Hindawi Comput. Intell. Neurosci., vol. 2022, 2022, https://doi.org/10.1155/2022/8512469
[13] R. Kumari, S. Kumar, and S. Godara, “An Efficient Diabetic Retinopathy Detection and Classification System Using LRKSA-CNN and,” Int. J. Adv. Comput. Sci. Appl., vol. 15, no. 12, pp. 613–627, 2025, https://doi.org/10.14569/IJACSA.2024.0151263
[14] U. Ishtiaq, E. Rahayu, M. Faizal, and Z. Ishtiaque, “A Hybrid Technique for Diabetic Retinopathy Detection Based on Ensemble-Optimized CNN and Texture Features,” Diagnostics MDPI, vol. 13, pp. 1–21, 2023, https://doi.org/10.3390/diagnostics13101816
[15] H. N. Pham et al., “Automated Grading in Diabetic Retinopathy Using Image Processing and Modified EfficientNet,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 12496 LNAI, pp. 505–515, 2020, https://doi.org/10.1007/978-3-030-63007-2_39
[16] A. Elmoufidi and H. Ammoun, “Diabetic Retinopathy Prevention Using EfficientNetB3 Architecture and Fundus Photography,” SN Comput. Sci., vol. 4, no. 1, pp. 1–9, 2023, https://doi.org/10.1007/s42979-022-01482-6
[17] U. Pamula et al., “Optimizing diabetic retinopathy detection with electric fish algorithm and bilinear convolutional networks,” Sci. Rep., pp. 1–19, 2025, https://doi.org/10.1038/s41598-025-99228-w
[18] M. A. I. Mahmood, N. Aktar, and M. F. Kader, “A hybrid approach for diagnosing diabetic retinopathy from fundus image exploiting deep features,” Heliyon, vol. 9, no. 9, pp. 1–14, 2023, https://doi.org/10.1016/j.heliyon.2023.e19625
[19] A. Gupta and V. Kumar, “A Deep Learning Fusion Approach for Automatic Diagnosis and Grading of Diabetic Retinopathy,” 2023 14th Int. Conf. Comput. Commun. Netw. Technol., pp. 1–6, 2023, https://doi.org/10.1109/ICCCNT56998.2023.10306810
[20] A. Karsaz, “Diabetic retinopathy screening using improved support vector domain description : a clinical study,” Soft Comput., vol. 26, no. 19, pp. 10085–10101, 2022, https://doi.org/10.1007/s00500-022-07387-z
[21] R. K. Patel and M. Kashyap, “Automated screening of glaucoma stages from retinal fundus images using BPS and LBP based GLCM features,” Int. J. Imaging Syst. Technol., vol. 33, no. 1, pp. 246–261, 2022, https://doi.org/10.1002/ima.22797
[22] S. Gupta, “Microaneurysm detection by multiple feature subset extraction and selection based on SVM-weights and Genetic Algorithm-Neural Network,” IEEE ICACCS, pp. 129–134, 2021, https://doi.org/10.1109/ICACCS51430.2021.9441746
[23] S. Gayathri, V. P. Gopi, and P. Palanisamy, “Automated classification of diabetic retinopathy through reliable feature selection,” Phys. Eng. Sci. Med., vol. 43, no. 3, pp. 927–945, 2020, https://doi.org/10.1007/s13246-020-00890-3.
[24] P. Macsik, J. Pavlovicova, S. Kajan, and J. Goga, “Image preprocessing‐based ensemble deep learning classification of diabetic,” IET Image Process, vol. 18, pp. 807–822, 2024, https://doi.org/10.1049/ipr2.12987
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Loneli Costaner, Lisnawita, Guntoro
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
