Early Diagnosis of Retinal Blood Vessel Damage via Deep Learning-Powered Collective Intelligence Models

Table 5

Comparison of TDCN models with imagenet models and literature.

Model	Accuracy	AUC ROC	Cohen’s kappa
Inception	73.2	0.91	0.738
Xception	74.8	0.87	0.772
Resnet50	73.8	0.89	0.776
Shaban.et al. [33]	88	0.930	0.910
S. Kassani et al. [34]	83.09	0.950	0.892
Taufiqurrahman et al. [35]	85	0.820	0.925
Bodapati et al. [36]	84.31	0.970	0.758
TDCN-ACO	78.4	0.907	0.795
TDCN-PSO	90.3	0.956	0.967