Table 1.
Summary of related work.
Fig 1.
CNN, FCN, and FCEDN architecture.
Fig 2.
KELM configuration.
Fig 3.
Genetic algorithm flowchart.
Fig 4.
Proposed methodology.
Fig 5.
Sample images from the datasets.
Table 2.
A Detailed description of the dataset.
Table 3.
Description of the experimental augmented dataset.
Table 4.
Configuration of FCEDN.
Table 5.
Comparative overview of parameter configurations for FCEDN, enhanced with G-GWO and G-GWO-KELM optimizations.
Table 6.
Performance evaluation matrices.
Fig 6.
Segmentation performance matrices comparison.
Fig 7.
Input, pre-processing, ground truth, and the predicted segmentation results were obtained by the proposed model for some sample images.
Table 7.
Different classifiers Performance evaluation using IDRiD dataset.
Fig 8.
ACC, SEN, and SPC evaluation using IDRiD dataset.
Fig 9.
MCC, ER, and F1-Score evaluation using IDRiD dataset.
Fig 10.
Confusion matrix comparison of four algorithms on IDRiD dataset for diabetic retinopathy, showcasing GGWO-KELM’s exceptional performance.
Table 8.
Different classifiers performance evaluation using DR-HSGIS dataset.
Fig 11.
Accuracy evaluation using DR-HAGIS dataset.
Fig 12.
Sensitivity evaluation using DR-HAGIS dataset.
Fig 13.
Error Rate evaluation using DR-HAGIS dataset.
Fig 14.
Specificity evaluation using DR-HAGIS dataset.
Fig 15.
MCC evaluation using DR-HAGIS dataset.
Fig 16.
F-1 Score valuation using DR-HAGIS dataset.
Fig 17.
Comparison of confusion matrices for different classifiers across three diseases: Diabetic retinopathy, DME, and Glaucoma.
Table 9.
Different classifiers performance evaluation using OIDR dataset.
Fig 18.
Accuracy evaluation using OIDR dataset.
Fig 19.
Sensitivity evaluation using OIDR dataset.
Fig 20.
Specificity evaluation using OIDR dataset.
Fig 21.
MCC evaluation using OIDR dataset.
Fig 22.
Error Rate evaluation using OIDR dataset.
Fig 23.
F-1 Score evaluation using OIDR dataset.
Fig 24.
Comparison of confusion matrices for different classifiers using OIDR dataset.
Table 10.
Comparitiva analysis.