Microaneurysm detection in fundus images using a two-step convolutional neural network

Table 5 Final score (CPM)

Dataset	Method	CPM	\(F_{AUC}\)
Retinopathy Online Challenge	Proposed method	0.461	0.660
	Dashtbozorg [44]	0.471	0.484
	Chudzik [9]	0.298	–
	Budak [9]	0.221	–
	Javidi [9]	0.261	–
	B Wu’s [39]	0.202	0.302
	Valladolid [42]	0.322	–
	Waikato group [41]	0.206	0.273
	Latim [25]	0.381	0.489
	OkMedical [10]	0.357	0.430
	Fujita Lab [18]	0.310	0.378
E-Optha-MA	Proposed method	0.471	0.637
	Dashtbozorg [44]	0.510	0.575
	Chudzik [9]	0.431	–
	Budak [9]	0.431	–
	B Wu’s [39]	0.431	0.386

The quantities given in italic form for “Retinopathy Online Challenge dataset” and “E-Ophtha-MA dataset” represent the best results
Competetion measure (CPM) of Retinopathy Online Challenge at different points

ISSN: 1475-925X