FFU-Net: Feature Fusion U-Net for Lesion Segmentation of Diabetic Retinopathy

Table 5

Comparative segmentation results of the proposed model against the state of the art on MA and HE.

		MA			HE
Methods	SEN	IOU	DICE	SEN	IOU	DICE
Dai et al.	0.5498	0.5237	0.6874	0.6895	0.6990	0.8228
Zhang et al.	0.4897	0.4723	0.6416	0.6418	0.6407	0.7810
Van Grinsven et al.	0.4832	0.4667	0.6364	0.6844	0.6761	0.8068
M-Net	0.5366	0.5097	0.6753	0.6872	0.6796	0.8093
FC-DenseNet	0.5521	0.5276	0.6908	0.6976	0.6960	0.8208
Sambyal et al.	0.5537	0.5438	0.7045	0.6998	0.7038	0.8261
FFU-Net	0.5933	0.5610	0.7188	0.7342	0.7365	0.8450
U-Net	0.4810	0.4490	0.6197	0.6366	0.6333	0.7755