Table 2.

AUC, accuracy, sensitivity and specificity related to the exudate detection for E-OPHTHA exudates database taking into account different input vectors to the SVM classifier. In particular, they are composed by texture and shape descriptors.

	Accuracy	Sensitivity	Specificity	AUC
LBPV	0.8205 ± 0.0320	0.7389 ± 0.1034	0.8331 ± 0.0534	0.8684 ± 0.0435
LBPV-${\phi }_B$	0.8369 ± 0.0336	0.7603 ± 0.0897	0.8481 ± 0.0519	0.8834 ± 0.0397
LBPV-${\gamma }_B{\phi }_B$	0.8445 ± 0.0276	0.7657 ± 0.0894	0.8561 ± 0.0449	0.8872 ± 0.0384
LBPV-${\gamma }_B{\gamma }_L$	0.8447 ± 0.0228	0.7496 ± 0.1079	0.8591 ± 0.0432	0.8803 ± 0.0409
LBPV-${\gamma }_B{\gamma }_L{\phi }_B{\phi }_L$	0.8533 ± 0.0245	0.7721 ± 0.0857	0.8651 ± 0.0399	0.8948 ± 0.0351