Managing Diabetic Retinopathy with Deep Learning: A Data Centric Overview

2026-04-02

AI Summary

This study addresses the clinical reliability limitations of automated diabetic retinopathy (DR) detection, which stem from geographical biases, insufficient sample sizes, inconsistent annotations, and variable image quality in existing fundus datasets. For the first time, it provides a systematic review and comparative analysis of DR-related datasets from a data-centric perspective, categorizing them by scale, accessibility, and annotation type, and evaluating their suitability for tasks such as binary classification, severity grading, lesion localization, and multi-disease screening. The work identifies the widespread absence of standardized lesion-level annotations and longitudinal data, distills key characteristics of high-quality DR datasets, and offers concrete recommendations for constructing future datasets that enhance clinical reliability and model interpretability, thereby guiding the development of robust AI-assisted screening systems.

Abstract

Diabetic Retinopathy (DR) is a serious microvascular complication of diabetes, and one of the leading causes of vision loss worldwide. Although automated detection and grading, with Deep Learning (DL), can reduce the burden on ophthalmologists, it is constrained by the limited availability of high-quality datasets. Existing repositories often remain geographically narrow, contain limited samples, and exhibit inconsistent annotations or variable image quality; thereby, restricting their clinical reliability. This paper presents a comprehensive review and comparative analysis of fundus image datasets used in the management of DR. The study evaluates their usability across key tasks, including binary classification, severity grading, lesion localization, and multi-disease screening. It also categorizes the datasets by size, accessibility, and annotation type (such as image-level, lesion-level, and multi-disease). Finally, a recently published dataset is presented as a case study to illustrate broader challenges in dataset curation and usage. The review consolidates current knowledge while highlighting persistent gaps such as the lack of standardized lesion-level annotations and longitudinal data. It also outlines recommendations for future dataset development to support clinically reliable and explainable solutions in DR screening.

Problem

• Diabetic Retinopathy
• Deep Learning
• Fundus Image Datasets
• Data Quality
• Clinical Reliability

Innovation

• data-centric review
• diabetic retinopathy
• fundus image datasets
• lesion-level annotation
• dataset curation