{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T17:22:42Z","timestamp":1789752162989,"version":"4.0.1"},"reference-count":67,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2025,9,17]],"date-time":"2025-09-17T00:00:00Z","timestamp":1758067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background: Diabetic retinopathy (DR) is a leading cause of preventable vision impairment in individuals with diabetes. Early detection is essential, yet often hindered by subtle disease progression and reliance on manual expert screening. This study introduces an AI-based framework designed to achieve robust multiclass DR classification from retinal fundus images, addressing the challenges of early diagnosis and fine-grained lesion discrimination. Methods: The framework incorporates preprocessing steps such as pixel intensity normalization and geometric correction. A Hybrid Local-Global Retina Super-Resolution (HLG-RetinaSR) module is developed, combining deformable convolutional networks for local lesion enhancement with vision transformers for global contextual representation. Classification is performed using a hierarchical approach that integrates three models: a Convolutional Neural Network (CNN), DenseNet-121, and a custom multi-branch RefineNet-U architecture. Results: Experimental evaluation demonstrates that the combined HLG-RetinaSR and RefineNet-U approach consistently achieves precision, recall, F1-score, and accuracy values exceeding 99% across all DR severity levels. The system effectively emphasizes vascular abnormalities while suppressing background noise, surpassing existing state-of-the-art methods in accuracy and robustness. Conclusions: The proposed hybrid pipeline delivers a scalable, interpretable, and clinically relevant solution for DR screening. By improving diagnostic reliability and supporting early intervention, the system holds strong potential to assist ophthalmologists in reducing preventable vision loss.<\/jats:p>","DOI":"10.3390\/diagnostics15182355","type":"journal-article","created":{"date-parts":[[2025,9,17]],"date-time":"2025-09-17T05:59:51Z","timestamp":1758088791000},"page":"2355","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["A Multi-Model Image Enhancement and Tailored U-Net Architecture for Robust Diabetic Retinopathy Grading"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-5670-0716","authenticated-orcid":false,"given":"Archana","family":"Singh","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, Thapar Institute of Engineering and Technology, Patiala 147004, Punjab, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sushma","family":"Jain","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Thapar Institute of Engineering and Technology, Patiala 147004, Punjab, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6081-7555","authenticated-orcid":false,"given":"Vinay","family":"Arora","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Thapar Institute of Engineering and Technology, Patiala 147004, Punjab, India"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,9,17]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"20552076231194942","DOI":"10.1177\/20552076231194942","article-title":"Deep learning-enhanced diabetic retinopathy image classification","volume":"9","author":"Alwakid","year":"2023","journal-title":"Digit. Health"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"2540","DOI":"10.2337\/diacare.27.10.2540","article-title":"Diabetic retinopathy","volume":"27","author":"Fong","year":"2004","journal-title":"Diabetes Care"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Bhimavarapu, U., and Battineni, G. (2022). Deep learning for the detection and classification of diabetic retinopathy with an improved activation function. Healthcare, 11.","DOI":"10.3390\/healthcare11010097"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"663","DOI":"10.1007\/s11517-014-1167-5","article-title":"Computer-aided diabetic retinopathy detection using trace transforms on digital fundus images","volume":"52","author":"Ganesan","year":"2014","journal-title":"Med. Biol. Eng. Comput."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1677","DOI":"10.1016\/S0161-6420(03)00475-5","article-title":"Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales","volume":"110","author":"Wilkinson","year":"2003","journal-title":"Ophthalmology"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1016\/j.preteyeres.2015.08.001","article-title":"The progress in understanding and treatment of diabetic retinopathy","volume":"51","author":"Stitt","year":"2016","journal-title":"Prog. Retin. Eye Res."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"703","DOI":"10.1016\/j.oret.2023.03.003","article-title":"Performance of automated machine learning for diabetic retinopathy image classification from multi-field handheld retinal images","volume":"7","author":"Jacoba","year":"2023","journal-title":"Ophthalmol. Retin."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Zhang, Z., Deng, C., and Paulus, Y.M. (2024). Advances in structural and functional retinal imaging and biomarkers for early detection of diabetic retinopathy. Biomedicines, 12.","DOI":"10.3390\/biomedicines12071405"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e35","DOI":"10.1016\/S2589-7500(19)30004-4","article-title":"Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: A clinical validation study","volume":"1","author":"Bellemo","year":"2019","journal-title":"Lancet Digit. Health"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"100377","DOI":"10.1016\/j.imu.2020.100377","article-title":"Diabetic retinopathy detection through deep learning techniques: A review","volume":"20","author":"Alyoubi","year":"2020","journal-title":"Inform. Med. Unlocked"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"584","DOI":"10.1038\/s41591-023-02702-z","article-title":"A deep learning system for predicting time to progression of diabetic retinopathy","volume":"30","author":"Dai","year":"2024","journal-title":"Nat. Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"5407","DOI":"10.1007\/s12652-020-01963-7","article-title":"Deep neural networks to predict diabetic retinopathy","volume":"14","author":"Gadekallu","year":"2023","journal-title":"J. Ambient. Intell. Humaniz. Comput."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"14779","DOI":"10.1007\/s00521-023-08454-8","article-title":"Deformable registration of multimodal retinal images using a weakly supervised deep learning approach","volume":"35","author":"Carmona","year":"2023","journal-title":"Neural Comput. Appl."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1007\/s13748-024-00325-0","article-title":"Self-supervised approach for diabetic retinopathy severity detection using vision transformer","volume":"13","author":"Ohri","year":"2024","journal-title":"Prog. Artif. Intell."},{"key":"ref_15","first-page":"736","article-title":"Histogram equalization techniques for enhancement of low radiance retinal images for early detection of diabetic retinopathy","volume":"22","author":"Singh","year":"2019","journal-title":"Eng. Sci. Technol. Int. J."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"355","DOI":"10.1016\/S0734-189X(87)80186-X","article-title":"Adaptive histogram equalization and its variations","volume":"39","author":"Pizer","year":"1987","journal-title":"Comput. Vis. Graph. Image Process."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1023\/B:VLSI.0000028532.53893.82","article-title":"Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement","volume":"38","author":"Reza","year":"2004","journal-title":"J. VLSI Signal Process. Syst. Signal Image Video Technol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"68","DOI":"10.1109\/30.754419","article-title":"Image enhancement based on equal area dualistic sub-image histogram equalization method","volume":"45","author":"Wang","year":"1999","journal-title":"IEEE Trans. Consum. Electron."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"757","DOI":"10.1109\/TCE.2007.381756","article-title":"Fast image\/video contrast enhancement based on weighted thresholded histogram equalization","volume":"53","author":"Wang","year":"2007","journal-title":"IEEE Trans. Consum. Electron."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1007\/s11760-016-0911-8","article-title":"Significance of incorporating chrominance information for effective color-to-grayscale image conversion","volume":"11","author":"Sowmya","year":"2017","journal-title":"Signal Image Video Process."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1109\/5254.708428","article-title":"Support vector machines","volume":"13","author":"Hearst","year":"1998","journal-title":"IEEE Intell. Syst. Their Appl."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1109\/TSMC.1985.6313426","article-title":"A fuzzy k-nearest neighbor algorithm","volume":"SMC-15","author":"Keller","year":"1985","journal-title":"IEEE Trans. Syst. Man. Cybern."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1023\/A:1022643204877","article-title":"Induction of decision trees","volume":"1","author":"Quinlan","year":"1986","journal-title":"Mach. Learn."},{"key":"ref_25","unstructured":"Zhang, H., and Su, J. (2004, January 20\u201324). Naive bayesian classifiers for ranking. Proceedings of the 15th European Conference on Machine Learning, Pisa, Italy."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1002\/wics.101","article-title":"Principal component analysis","volume":"2","author":"Abdi","year":"2010","journal-title":"Wiley Interdiscip. Rev. Comput. Stat."},{"key":"ref_27","first-page":"2694","article-title":"Enhancement of diabetic retinopathy imagery using contrast limited adaptive histogram equalization","volume":"2","author":"Shome","year":"2011","journal-title":"Int. J. Comput. Sci. Inf. Technol."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Liu, J., Zhou, X., Wan, Z., Yang, X., He, W., He, R., and Lin, Y. (2023). Multi-scale FPGA-based infrared image enhancement by using RGF and CLAHE. Sensors, 23.","DOI":"10.3390\/s23198101"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Raj, A., Shah, N.A., and Tiwari, A.K. (2022). A novel approach for fundus image enhancement. Biomed. Signal Process. Control, 71.","DOI":"10.1016\/j.bspc.2021.103208"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1109\/TMI.2019.2922960","article-title":"CT super-resolution GAN constrained by the identical, residual, and cycle learning ensemble (GAN-CIRCLE)","volume":"39","author":"You","year":"2019","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1211","DOI":"10.1049\/iet-ipr.2018.6212","article-title":"Fundus image quality assessment: Survey, challenges, and future scope","volume":"13","author":"Raj","year":"2019","journal-title":"IET Image Process."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"927","DOI":"10.1007\/s13246-020-00890-3","article-title":"Automated classification of diabetic retinopathy through reliable feature selection","volume":"43","author":"Gayathri","year":"2020","journal-title":"Phys. Eng. Sci. Med."},{"key":"ref_33","first-page":"173","article-title":"Image classification using convolutional neural networks","volume":"16","author":"Hasan","year":"2024","journal-title":"Int. J. Mech. Eng. Res. Technol."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Ikram, A., and Imran, A. (2025). ResViT FusionNet Model: An explainable AI-driven approach for automated grading of diabetic retinopathy in retinal images. Comput. Biol. Med., 186.","DOI":"10.1016\/j.compbiomed.2025.109656"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"100303","DOI":"10.1016\/j.health.2024.100303","article-title":"An advanced deep neural network for fundus image analysis and enhancing diabetic retinopathy detection","volume":"5","author":"Shamrat","year":"2024","journal-title":"Healthc. Anal."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Venkaiahppalaswamy, B., Reddy, P.P., and Batha, S. (2023). Hybrid deep learning approaches for the detection of diabetic retinopathy using optimized wavelet based model. Biomed. Signal Process. Control, 79.","DOI":"10.1016\/j.bspc.2022.104146"},{"key":"ref_37","first-page":"4561","article-title":"Diabetic Retinopathy Detection: A Hybrid Intelligent Approach","volume":"80","author":"Rahman","year":"2024","journal-title":"Comput. Mater. Contin."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"123418","DOI":"10.1016\/j.eswa.2024.123418","article-title":"Enhancing diabetic retinopathy detection through preprocessing and feature extraction with MGA-CSG algorithm","volume":"249","author":"Navaneethan","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"149700","DOI":"10.1109\/ACCESS.2024.3475550","article-title":"Developing a transparent diagnosis model for diabetic retinopathy using explainable AI","volume":"12","author":"Shahzad","year":"2024","journal-title":"IEEE Access"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"54190","DOI":"10.1109\/ACCESS.2021.3070685","article-title":"Multi-scale attention network for diabetic retinopathy classification","volume":"9","author":"Arafa","year":"2021","journal-title":"IEEE Access"},{"key":"ref_41","first-page":"12","article-title":"Classification of diabetic retinopathy disease using convolutional neural network","volume":"6","author":"Minarno","year":"2022","journal-title":"JOIV Int. J. Inform. Vis."},{"key":"ref_42","first-page":"606","article-title":"Classification of Dermoscopic Images Using CNN-SVM","volume":"8","author":"Minarno","year":"2024","journal-title":"JOIV Int. J. Inform. Vis."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1109\/JPROC.2020.3004555","article-title":"A comprehensive survey on transfer learning","volume":"109","author":"Zhuang","year":"2020","journal-title":"Proc. IEEE"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Yuan, D., Liu, Y., Xu, Z., Zhan, Y., Chen, J., and Lukasiewicz, T. (2023). Painless and accurate medical image analysis using deep reinforcement learning with task-oriented homogenized automatic pre-processing. Comput. Biol. Med., 153.","DOI":"10.1016\/j.compbiomed.2022.106487"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Salvi, M., Acharya, U.R., Molinari, F., and Meiburger, K.M. (2021). The impact of pre-and post-image processing techniques on deep learning frameworks: A comprehensive review for digital pathology image analysis. Comput. Biol. Med., 128.","DOI":"10.1016\/j.compbiomed.2020.104129"},{"key":"ref_46","first-page":"30","article-title":"Improving retinal image quality using the contrast stretching, histogram equalization, and CLAHE methods with median filters","volume":"14","author":"Ningsih","year":"2020","journal-title":"Int. J. Image, Graph. Signal Process."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Goceri, E. (2023). Evaluation of denoising techniques to remove speckle and Gaussian noise from dermoscopy images. Comput. Biol. Med., 152.","DOI":"10.1016\/j.compbiomed.2022.106474"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"152261","DOI":"10.1109\/ACCESS.2021.3125791","article-title":"Hybrid CNN-SVD based prominent feature extraction and selection for grading diabetic retinopathy using extreme learning machine algorithm","volume":"9","author":"Nahiduzzaman","year":"2021","journal-title":"IEEE Access"},{"key":"ref_49","unstructured":"Dong, C., Loy, C.C., and Tang, X. (2016, January 11\u201314). Accelerating the super-resolution convolutional neural network. Proceedings of the Computer Vision\u2013ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part II 14."},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Lim, B., Son, S., Kim, H., Nah, S., and Mu Lee, K. (2017, January 21\u201326). Enhanced deep residual networks for single image super-resolution. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Honolulu, HI, USA.","DOI":"10.1109\/CVPRW.2017.151"},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"e2023MS003774","DOI":"10.1029\/2023MS003774","article-title":"A machine learning augmented data assimilation method for high-resolution observations","volume":"16","author":"Howard","year":"2024","journal-title":"J. Adv. Model. Earth Syst."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"1254","DOI":"10.1109\/LGRS.2018.2830403","article-title":"Deformable convolutional neural networks for hyperspectral image classification","volume":"15","author":"Zhu","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"1057","DOI":"10.1109\/TIP.2020.3042059","article-title":"Light field image super-resolution using deformable convolution","volume":"30","author":"Wang","year":"2020","journal-title":"IEEE Trans. Image Process."},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1007\/s00371-018-1503-0","article-title":"Image classification by combining local and global features","volume":"35","author":"Kabbai","year":"2019","journal-title":"Vis. Comput."},{"key":"ref_55","unstructured":"Aitken, A., Ledig, C., Theis, L., Caballero, J., Wang, Z., and Shi, W. (2017). Checkerboard artifact free sub-pixel convolution: A note on sub-pixel convolution, resize convolution and convolution resize. arXiv."},{"key":"ref_56","doi-asserted-by":"crossref","unstructured":"Gulati, S., Guleria, K., and Goyal, N. (2023, January 6\u20138). Classification of diabetic retinopathy using pre-trained deep learning model-DenseNet 121. Proceedings of the 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India.","DOI":"10.1109\/ICCCNT56998.2023.10308181"},{"key":"ref_57","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5\u20139). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention\u2014MICCAI 2015: 18th International Conference, Munich, Germany. Proceedings, Part III 18.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_58","unstructured":"Kaggle (2025, April 07). APTOS 2019 Blindness Detection. Available online: https:\/\/www.kaggle.com\/competitions\/aptos2019-blindness-detection."},{"key":"ref_59","unstructured":"(2025, April 07). MESSIDOR-2: Diabetic Retinopathy Images. Available online: http:\/\/www.adcis.net\/en\/third-party\/messidor2\/."},{"key":"ref_60","unstructured":"(2025, April 07). DDR: Diabetic Retinopathy Dataset. Available online: https:\/\/www.kaggle.com\/datasets\/mariaherrerot\/ddrdataset."},{"key":"ref_61","doi-asserted-by":"crossref","unstructured":"Islam, M.R., Abdulrazak, L.F., Nahiduzzaman, M., Goni, M.O.F., Anower, M.S., Ahsan, M., Haider, J., and Kowalski, M. (2022). Applying supervised contrastive learning for the detection of diabetic retinopathy and its severity levels from fundus images. Comput. Biol. Med., 146.","DOI":"10.1016\/j.compbiomed.2022.105602"},{"key":"ref_62","doi-asserted-by":"crossref","first-page":"16797","DOI":"10.1007\/s10489-022-04295-5","article-title":"Joint grading of diabetic retinopathy and diabetic macular edema using an adaptive attention block and semisupervised learning","volume":"53","author":"Guo","year":"2023","journal-title":"Appl. Intell."},{"key":"ref_63","doi-asserted-by":"crossref","first-page":"1083","DOI":"10.1007\/s11042-023-15120-7","article-title":"Self-adaptive stacking ensemble approach with attention based deep neural network models for diabetic retinopathy severity prediction","volume":"83","author":"Bodapati","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref_64","doi-asserted-by":"crossref","first-page":"1739","DOI":"10.1007\/s10278-023-00813-0","article-title":"Diabetic retinopathy prediction based on wavelet decomposition and modified capsule network","volume":"36","author":"Oulhadj","year":"2023","journal-title":"J. Digit. Imaging"},{"key":"ref_65","doi-asserted-by":"crossref","unstructured":"Han, Z., Yang, B., Deng, S., Li, Z., and Tong, Z. (2023). Category weighted network and relation weighted label for diabetic retinopathy screening. Comput. Biol. Med., 152.","DOI":"10.1016\/j.compbiomed.2022.106408"},{"key":"ref_66","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1109\/TMI.2020.3023463","article-title":"CABNet: Category attention block for imbalanced diabetic retinopathy grading","volume":"40","author":"He","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_67","doi-asserted-by":"crossref","unstructured":"Wang, X., Xu, M., Zhang, J., Jiang, L., and Li, L. (2021, January 2\u20139). Deep multi-task learning for diabetic retinopathy grading in fundus images. Proceedings of the AAAI Conference on Artificial Intelligence, Virtual.","DOI":"10.1609\/aaai.v35i4.16388"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/15\/18\/2355\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:46:46Z","timestamp":1760035606000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/15\/18\/2355"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,17]]},"references-count":67,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2025,9]]}},"alternative-id":["diagnostics15182355"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics15182355","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,17]]}}}