Diagnosing Diabetic Retinopathy in OCTA Images Based on Multilevel Information Fusion Using a Deep Learning Framework

Table 1

The segmentation accuracy and IOU for every class using our model.

Class	Accuracy	IOU
Background	93.2%	92.6%
Vessels	93.8%	54.1%
FAZ	92.3%	84.6%
Average	93.1%	77.1%