. 2022 Oct 14;13(11):5813–5835. doi: 10.1364/BOE.472176

Table 3. Generalization comparisons of our DACNN and other advanced methods on EyePACS dataset. “-” denotes no results reported in corresponding works. The best results were shown in bold. ^a .

Method	AUC	Acc	Kappa
AFN [61]	-	-	0.859
LAT [9]	-	-	0 . 884
CABNet [12]	-	0.862	0.868
DeepDR [25]	0.937	-	-
Our DACNN	0.961	0.865	0.881

^{^a}

AUC: the area under the receiver operating characteristic (ROC) curve; Acc: accuracy

Table 3. Generalization comparisons of our DACNN and other advanced methods on EyePACS dataset. “-” denotes no results reported in corresponding works. The best results were shown in bold. a .