Hemorrhage semantic segmentation in fundus images for the diagnosis of diabetic retinopathy by using a convolutional neural network

Journal of Big Data

Table 2 Comparison of the results of the proposed method for retinal hemorrhage segmentation to the latest current methods in terms of accuracy, sensitivity and specificity

Authors and Refs.	Dataset	Sensitivity (%)	Specificity (%)	Accuracy (%)
Ten et al. [30]	CLEOPATRA	62.57	98.93	_
Badar et al. [34]	MESSIDOR	80.93	98.54	97.86
Orlando et al. [35]	DIARETDB1 and E-Ophta	48.83	_	_
Proposed	IDRID and DIARETDB1	80.49	99.68	98.68