Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Dual Attention-aided DenseNet-121 for Classification of Glaucoma from Fundus Images

Soham Chakraborty, Ayush Roy, Payel Pramanik, Daria Valenkova, Ram Sarkar

TL;DR

An attention-aided DenseNet-121 for classifying normal and glaucomatous eyes from fundus images is presented, which has shown superior results than state-of-the-art models.

Abstract

Deep learning and computer vision methods are nowadays predominantly used in the field of ophthalmology. In this paper, we present an attention-aided DenseNet-121 for classifying normal and glaucomatous eyes from fundus images. It involves the convolutional block attention module to highlight relevant spatial and channel features extracted by DenseNet-121. The channel recalibration module further enriches the features by utilizing edge information along with the statistical features of the spatial dimension. For the experiments, two standard datasets, namely RIM-ONE and ACRIMA, have been used. Our method has shown superior results than state-of-the-art models. An ablation study has also been conducted to show the effectiveness of each of the components. The code of the proposed work is available at: https://github.com/Soham2004GitHub/DADGC.

A Dual Attention-aided DenseNet-121 for Classification of Glaucoma from Fundus Images

TL;DR

An attention-aided DenseNet-121 for classifying normal and glaucomatous eyes from fundus images is presented, which has shown superior results than state-of-the-art models.

Abstract

Deep learning and computer vision methods are nowadays predominantly used in the field of ophthalmology. In this paper, we present an attention-aided DenseNet-121 for classifying normal and glaucomatous eyes from fundus images. It involves the convolutional block attention module to highlight relevant spatial and channel features extracted by DenseNet-121. The channel recalibration module further enriches the features by utilizing edge information along with the statistical features of the spatial dimension. For the experiments, two standard datasets, namely RIM-ONE and ACRIMA, have been used. Our method has shown superior results than state-of-the-art models. An ablation study has also been conducted to show the effectiveness of each of the components. The code of the proposed work is available at: https://github.com/Soham2004GitHub/DADGC.
Paper Structure (10 sections, 4 figures, 4 tables)

This paper contains 10 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Convolutional Block Attention Module (CBAM)
  5. Channel Recalibration Module (CRM)
  6. Results
  7. Experimental Setup
  8. Ablation Study
  9. Comparison with SOTA
  10. Conclusion and Future scope

Figures (4)

  • Figure 1: Block diagram of the proposed glaucoma classification model.
  • Figure 2: Channel Recalibration Module.
  • Figure 3: Heatmap of $F_{CRM}$ for normal and glaucoma fundus images.
  • Figure 4: Confusion matrices of the proposed model for ACRIMA and RIM-ONE datasets for the best fold among the 5 folds. 'G' and 'N' indicate 'Glaucoma' and 'Normal' classes respectively.