SCRD-Net: A Deep Convolutional Neural Network Model for Glaucoma Detection in Retina Tomography

Table 1

Comparison of our model with state-of-the-art network models.

Method		ACC	AUC	0.05-SE	0.15-SE	SE
Traditional	MRA [7]	—	—	—	0.650	—
	GPS [8]	—	0.800	0.420	0.610	—
	SAS [9]	—	0.890	0.590	0.770	—
Deep learning	Googlenet [51]	0.860	0.925	0.694	0.857	0.846
	Resnet [54]	0.846	0.895	0.635	0.835	0.862
	Densnet [55]	0.857	0.925	0.590	0.895	0.851
	Vgg-16 [56]	0.859	0.913	0.698	0.888	0.808
	Alexnet [57]	0.843	0.896	0.644	0.813	0.879
	Efficientnet [60]	0.858	0.929	0.612	0.924	0.898
Ours	SCRD-Net	0.884	0.940	0.783	0.912	0.910