Glaucoma Classification Using a NFNet-Based Deep Learning Model with a Customized Hybrid Attention Mechanism

TL;DR

This paper introduces a deep learning model with a custom attention mechanism for accurately detecting glaucoma from eye images.

Contribution

A novel hybrid attention mechanism combined with normalization-free ResNet architectures for improved glaucoma classification.

Findings

The model achieved 0.9394 accuracy on the LAG dataset, outperforming state-of-the-art ResNet variants.

On the combined dataset, the model showed 0.9193 accuracy, 0.9182 sensitivity, and 0.9202 specificity.

The hybrid attention mechanism significantly enhanced performance across multiple glaucoma datasets.

Abstract

Background/Objectives: Glaucoma is a leading cause of irreversible blindness worldwide, making accurate and efficient detection methods essential. One primary concern with glaucoma is that it often presents no early symptoms. Vision loss typically begins at the periphery and progresses unnoticed until it significantly affects central vision. Due to this gradual and usually silent progression, early detection through regular eye exams is vital for preventing permanent vision loss. Methods: In this study, we propose a hybrid attention mechanism that recalibrates feature maps from the feature extractor for glaucoma detection. We explored normalization-free ResNet (NF-ResNet) architectures to evaluate the proposed attention mechanism, specifically NF-ResNet-26, NF-ResNet-50, and NF-ResNet-101, in comparison to baseline state-of-the-art ResNet variants. Our approach was evaluated on three publicly available glaucoma datasets, LAG, EyePACS, and BrG, to differentiate between normal and glaucomatous from fundus images. Results: The experimental results demonstrate that our proposed hybrid attention module, combined with normalization-free architectures, significantly enhances performance compared to state-of-the-art ResNet variants. The proposed attention model based on the normalization-free ResNet-50 achieved an accuracy of 0.9394 on the LAG dataset, 0.9117 on the EyePACS dataset, and 0.9020 on the BrG dataset. When evaluated on the combined dataset, the model achieved an accuracy of 0.9193, sensitivity of 0.9182, and specificity of 0.9202. Conclusions: The results from these representative datasets for glaucoma detection highlight the exceptional performance of our attention module, establishing it as a highly competitive classification model in the field of glaucoma detection.

Figures