Skip to content
Open access

A unified deep learning framework for multi-lesion detection in diabetic retinopathy using fundus images

Oct 2026 · Scientific Reports · Vol 16 · 0 citations · 60 references

TL;DR

The proposed unified multi-lesion detection system enables simultaneous detection of multiple DR lesions in a single-pass reducing computational overhead, while maintaining reasonable detection performance and demonstrating the potential of semantic segmentation models, particularly U-Net, in automated DR screening.

Abstract

Diabetic Retinopathy (DR), an ocular disease caused by diabetes, is a significant cause of vision impairment. Early and accurate detection of different types of retinal lesions, including Microaneurysms, Haemorrhages, Soft Exudates and Hard Exudates, is essential for DR diagnosis and prevention of associated vision loss. Conventional manual analysis of retinal fundus images is subjective and time-consuming, requiring specialized clinical expertise. Automated diagnostic systems can alleviate these issues and assist clinicians with objective analysis and improved outcomes. This work focuses on simultaneous detection of different lesion types in DR, through a unified framework. In particular, the U-Net architecture is employed for multi-lesion segmentation due to its ability to preserve fine structural details using its encoder-decoder design. Additionally, an Adaptive Batch Scaling (ABS) strategy is employed, wherein the batch size is dynamically adjusted across training epochs, based on the validation Dice score, to optimize pixel-level segmentation accuracy. The performance of U-Net is compared with other DL-based segmentation models such as Mask Region-based Convolutional Neural Network, YOLOv8, and Faster Region-based Convolutional Neural Network architectures, in terms of its ability to accurately segment and identify diverse DR lesions in retinal fundus images. Validation is performed on two different publicly available DR datasets, i.e., Indian Diabetic Retinopathy Image Dataset (IDRiD) and Diabetic Retinopathy Detection Dataset (DDR). Specifically, DDR dataset is used for training the model, which is tested separately on each of the two datasets – DDR and IDRiD, to assess intra-dataset and cross-dataset generalization. Performance metrics including accuracy, Dice score, Intersection over Union (IoU) and mean average precision (mAP), are used to evaluate the performance of the models. The findings reveal that the U-Net architecture demonstrates superior performance, achieving an accuracy of 0.925, Dice score of 0.771, IoU of 0.748 and mAP of 0.647. The results demonstrate the potential of semantic segmentation models, particularly U-Net, in automated DR screening. The proposed unified multi-lesion detection system enables simultaneous detection of multiple DR lesions in a single-pass reducing computational overhead, while maintaining reasonable detection performance. The system is envisaged to serve as a reliable tool in DR screening workflow, facilitating early diagnosis and timely intervention, thereby reducing the burden of DR-related vision worldwide.

Read PDF

Similar papers

Open access Sep 2026

Automated Multi-Class Retinal Disease Detection Using Deep Convolutional Neural Networks And Transfer Learning

An automated deep-learning framework for the multi-class classification of retinal fundus images into four categories: cataract, diabetic retinopathy, glaucoma, and normal is presented, demonstrating the potential of transfer-learning-based deep neural networks for scalable and accessible preliminary retinal disease sc...

Ankit Pandit and Aakash Patel · 0 citations
Open access Dec 2026

Transforming diabetic retinopathy detection with deep learning and handcrafted feature fusion

The task of early and accurate screening of diabetic retinopathy (DR) from fundus images is affected by the presence of class imbalance, small grade differences, and the lack of interpretability of deep learning models. This paper proposes HandEffiTrans-DR, a hybrid model that represents lesion statistics as semantic t...

G. Murali, K. Narayanan · 0 citations
Open access Sep 2026

Attention-Guided Residual Learning for Reliable Multi-Class Diabetic Retinopathy Severity Assessment

Diabetes is one of the most common metabolic diseases (DM) globally, accompanied by mild to severe secondary complications, including diabetic retinopathy (DR), which can damage the retina and lead to vision loss. DR detection is crucial, as early treatment of DR can effectively prevent vision loss in patients. Imag...

Confidence Chigozirim Olumba · 0 citations
Open access 2026

An Enhanced DenseNet-169 Framework for Automated Multi-Class Diabetic Retinopathy Classification Using Transfer Learning and SMOTE

Diabetic Retinopathy (DR) is the most common microvascular consequence of diabetes mellitus and the most common cause of permanent blindness in the working-age population worldwide. Early diagnosis and quick intervention are critical for preventing severe visual impairment, but traditional screening methods rely on ski...

S. A, A. Thangavel, D. V. et al. · 0 citations
Conference

Deep Learning for Diabetic Retinopathy with Color Fundus Image Analysis in AI for Medical Imaging

This work analyzes clinical photos of diabetic retinopathy (DR), a common and preventable cause of vision impairment, using deep-learning methods. We have analyzed the computational efficiency of several concatenated CNN models, including DenseNet201, ResNet50, VGG19, and MobileNet V2. ResNet50 has a training time of 1...

M. B. A. Miah · 0 citations
Open access Oct 2026

A Modified AlexNet-Based Deep Learning Framework for Multi-Disease Ophthalmic Classification and Severity Detection from Retinal Fundus Images

The study contributes an integrated fundus-image analysis pipeline that combines multi-scale deep feature learning with disease-specific severity assessment and provides a reproducible basis for further clinical validation.

S. Shinde · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.