Table 1.

Comparison of results between manual segmentation and the proposed method on the DRIVE, STARE, HRF and CHASE_DB1 datasets

Dataset	Method	Se	Sp	Acc	AUC
DRIVE	2nd expert	0.7760	0.9725	0.9473	-
	Proposed Method	0.8179	0.9879	0.9846	0.9874
STARE	2nd expert	0.8719	0.9388	0.9353	-
	Proposed Method	0.8751	0.9894	0.9872	30.9941
HRF	2nd expert	0.8010	0.8011	0.9650	-
	Proposed Method	0.8025	0.9854	0.9856	0.9882
CHASE_DB1	2nd expert	0.7686	0.9779	0.9560	-
	Proposed Method	0.7972	0.7972	0.9889	0.9874