Table 1.

AUC, accuracy, sensitivity and specificity related to the exudate detection for E-OPHTHA exudates database taking into account different morphological input vectors to the SVM classifier.

	Accuracy	Sensitivity	Specificity	AUC
${\gamma }_B$	0.6405 ± 0.0572	0.6137 ± 0.1378	0.6489 ± 0.0813	0.6701 ± 0.0617
${\gamma }_L$	0.6684 ± 0.0353	0.6009 ± 0.1449	0.6759 ± 0.0521	0.6907 ± 0.0601
${\phi }_B$	0.7014 ± 0.0247	0.6396 ± 0.1420	0.7090 ± 0.0426	0.7337 ± 0.0777
${\phi }_L$	0.5794 ± 0.0588	0.5105 ± 0.0887	0.5914 ± 0.0823	0.5783 ± 0.0366
${\gamma }_B{\phi }_B$	0.7554 ± 0.0161	0.6592 ± 0.1179	0.7698 ± 0.0292	0.7877 ± 0.0524
${\gamma }_B{\phi }_L$	0.6550 ± 0.0528	0.6223 ± 0.1243	0.6636 ± 0.0745	0.6946 ± 0.0546
${\gamma }_B{\gamma }_L$	0.7321 ± 0.0208	0.6394 ± 0.0839	0.7465 ± 0.0271	0.7543 ± 0.0461
${\gamma }_L{\phi }_B$	0.7112 ± 0.0262	0.6446 ± 0.1344	0.7194 ± 0.0451	0.7482 ± 0.0684
${\gamma }_L{\phi }_L$	0.6685 ± 0.0320	0.5935 ± 0.1363	0.6774 ± 0.0499	0.6882 ± 0.0564
${\phi }_B{\phi }_L$	0.7014 ± 0.0256	0.6346 ± 0.1376	0.7099 ± 0.0432	0.7349 ± 0.0749
${\gamma }_B{\gamma }_L{\phi }_B{\phi }_L$	0.7620 ± 0.0165	0.6648 ± 0.1124	0.7762 ± 0.0300	0.7924 ± 0.0493