Table 7.

The comparisons between the customized EyeNet, ResNet50 [40], Inception V3 [44], VGG19 [45], the proposed E-DenseNet BC with different depths and weights on EyePACS dataset due to ACC, SEN, SPE, DSC, QKS, calculation time (T) in minutes (m) performance measures

Model	ACC (%)	SEN (%)	SPE (%)	DSC (%)	QKS	T(m)
A customized EyeNet	95.5	95.7	73	95	0.90	22
ResNet50 [40]	79.2	83	53	86.7	0.78	43
Inception V3 [44]	72.6	76.7	61	82	0.65	55
VGG19 [45]	82.3	87	49	80.6	0.69	45
E-DenseNet BC-169	86	90	55	93.3	0.89	8
E-DenseNet BC-169-ImageNet	83	87	61	91.6	0.86	7
E-DenseNet BC-201-ImageNet	90.6	94.3	53	95	0.92	9
E-DenseNet BC-121-ImageNet	88.1	93	68	93	0.89	5
E-DenseNet BC-121	96.8	98.3	72	98.3	0.97	5