Table 1.
Specification of Glaucoma Data Sets (Glaucomatous vs. Healthy).
Fig 1.
Some sample images of G1020 and ORIGA Dataset of the deployed model.
Fig 2.
The distribution of trial samples through k-Fold Cross-Validation in each iteration of the employed model.
Fig 3.
Glaucoma fundus image and its FDCT scale four coefficients of the proposed model.
Table 2.
Compilation of hyperparameter specifications for distinct classifiers.
Table 3.
Comparative analyses (%) of the deployed model based on PCA with LDA techniques.
Fig 4.
Accuracy with respect to number of Features using G1020 Dataset.
Fig 5.
Accuracy with respect to Number of Features using ORIGA Dataset.
Table 4.
The effectiveness of the deployed CAD model using the Retinal G1020 dataset result (%) in 10 × 5-fold approaches with GWO + ELM.
Table 5.
The effectiveness of the suggested CAD model using the Retinal G1020 dataset result (%) in 10 × 5-fold approaches with IMGWO+ELM.
Table 6.
The effectiveness of the suggested CAD model using the Retinal ORIGA dataset result (%) in 10 × 5-fold approaches with GWO + ELM.
Table 7.
The effectiveness of the suggested CAD model using the Retinal ORIGA dataset result (%) in 10 × 5-fold approaches with INGWO+ELM.
Fig 6.
Accuracy with respect to Number of Epochs on one run using G1020 dataset.
Fig 7.
Loss with respect to Number of Epochs on one run using G1020 dataset.
Fig 8.
Accuracy with respect to Number of Epochs on one run using ORIGA dataset.
Fig 9.
Loss with respect to Number of Epochs on one run using ORIGA dataset.
Fig 10.
Confusion matrix of the deployed scheme based on G1020 Dataset.
Fig 11.
Confusion matrix of the deployed scheme based on the ORIGA Dataset.
Table 8.
Comparative Analysis (%) employed scheme Glaucoma datasets with different specifications.
Table 9.
Performance analysis of Proposed model with existing CAD models with G1020 and ORIGA datasets.
Fig 12.
Performance Analysis on list of classifiers with the existing model using the G1020 Dataset.
Fig 13.
Performance Analysis on list of classifiers with the existing model using the ORIGA Dataset.
Fig 14.
Ablation study heatmap highlighting accuracy variation across model configurations for the G1020 and ORIGA datasets.
Fig 15.
The sequence of the suggested GlaucoXAI framework of the proposed model.
Fig 16.
Comparing list of explainable artificial intelligence (XAI) visualization methods for the classification of glaucoma.
Fig 17.
The Graph of the convergence factor with number of iteration of the employed model.