What question did this study set out to answer?

The aim is to develop an automated method for classifying stages of diabetic retinopathy using deep learning.

April 3, 20260 citationsOpen Access

Automatic classification of diabetic retinopathy using convolutional neural networks: A deep learning-based ensemble approach

LPLuiz Henrique de Azevedo Pontes LCLucídio dos Anjos Formiga Cabral RCRicardo Azevedo Pontes de Carvalho

Key Points

The aim is to develop an automated method for classifying stages of diabetic retinopathy using deep learning.
Utilized twelve convolutional neural network architectures for classification.
Implemented an ensemble model combining EfficientNet-B3 and VGG-19.
Processed 5,842 retinal fundus images through normalization, CLAHE, and Gaussian filtering.
Employed data augmentation techniques to handle class imbalance.
Achieved a global accuracy of 80.1% for DR classification.
Obtained a quadratic weighted Kappa score of 0.78.
Sensitivity for advanced stages reached up to 87.3%, but only 21.0% for mild diabetic retinopathy.

Abstract

Diabetic retinopathy (DR) is one of the leading causes of preventable blindness. This study presentsan integrated approach for automatic classification of DR stages from retinal fundus images, using twelve convo-lutional neural network (CNN) architectures and an efficient ensemble (EfficientNet-B3 + VGG-19). A total of5,842 images were processed through a pipeline including CLAHE, Gaussian filtering, and normalization, alongwith data augmentation to address class imbalance. The ensemble outperformed individual models, achieving aglobal accuracy of 80.1% and a quadratic weighted Kappa of 0.78. Sensitivity was robust for advanced stages (upto 87.3%), but remained low for mild DR (21.0%), highlighting the ongoing challenge in early-stage detection.

Perguntar à IA

Bookmark

View Full Paper