aceA comprehensive review on automated diabetic retinopathy detection and classification using fundus image
Abstract: Diabetic Retinopathy (DR) is a vision-threatening complication in diabetic patients. It harms retinal vessels and may lead to blindness. Detection at an early stage and its classification can prevent the risk of vision loss. However, fundus image-based manual screening of DR is a time-consuming and complex process. In recent years, many automated techniques for DR detection have been developed to screen and diagnose the disease condition at an early stage. These techniques are explored using the keywords diabeti…
"Fundus - diabetic retinopathy" by Shaofeng Hao, Changyan Liu, Na Li, Yanrong Wu, Dongdong Li, Qingyue Gao, Ziyou Yuan, Guanyan Li, Huilin Li, Jianzhou Yang, and Shengfu Fan. is licensed under CC BY 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/.
On 2026-08-17, a review in Graefe's Archive summarized automated diabetic retinopathy detection and classification using fundus images, surveying ML, DL and hybrid methods, their datasets, pre-processing and metrics, and noting newer ensemble, transformer and attention approaches.
Early automated screening matters because diabetic retinopathy damages retinal vessels and can cause blindness, so earlier detection could reduce vision loss, but the review indicates that current systems still contend with overfitting, complexity, imbalance and feature extraction issues that leave accuracy and robustness uncertain.
- Diabetic retinopathy harms retinal vessels and may lead to blindness if not detected early.
- Review searched Google Scholar, PubMed, Medline, IEEE Explore, and Science Direct using diabetic retinopathy, fundus image, ophthalmology with ML and DL keywords.
- Analyzed methodologies, datasets, pre-processing steps, and performance evaluation metrics across ML, DL, and hybrid approaches.
- Discussed recent ensemble learning, transformer-based techniques, and attention mechanisms for detection and classification.
Automated fundus-image analysis for diabetic retinopathy enables earlier screening and classification that can prevent vision loss in diabetic patients.
The rundown
The review systematically gathered prior research via Google Scholar, PubMed, Medline, IEEE Explore and Science Direct and examined ML, DL and hybrid pipelines for DR detection and classification from fundus images, including pre-processing and evaluation metrics.
It highlights recent directions in ensemble learning, transformer-based techniques and attention mechanisms, while noting persistent challenges of overfitting, model complexity, class imbalance and deep feature extraction that affect accuracy and robustness.
Sources
- Peer-reviewedGraefe's Archive for Clinical and Experimental Ophthalmology2026-08-17
How should this claim be treated?
The debate