An Efficient Retinal Segmentation-Based Deep Learning Framework for Disease Prediction

Table 1

DRIVE dataset performance with various architecture.

Network	U-Net [28]	RU-Net [4]	Dense U-net [18]	Proposed EDLFDPRS
Accuracy	95.08	95.65	95.53	97.32
score	81.87	81.55	81.71	83.85
Recall	82.65	84.62	84.81	82.56
Precision	85.14	84.72	83.94	85.19
Sensitivity	82.65	83.74	79.86	83.96
Specificity	97.97	97.83	97.36	98.68
AUC	96.81	97.84	97.82	98.79