Fig. 7From: Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentationEstimates of the clinical parameters in both networks. The x-axis indicates the value for the gold standard, and the y-axis indicates the error computed as the difference between the network estimates and the gold standard estimates. Each point corresponds to one image in the dataset (U-net in red, SW-net in green, and Topcon in blue). The mean value of the error for each set is drawn with a discontinuous line. a Cell density (ECD). b Polymegethism (CV). c Pleomorphism (HEX)Back to article page