Skip to main content

Table 1 Performance obtained using the proposed method

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

Performance Sen Spe Acc Dice IoU Prec LT
Avg (%) 80.49 99.68 98.68 86.51 76.61 99.98 0.0038
  1. Sen sensitivity, Spe specificity, Acc accuracy, IoU Intersection over Union, Prec precision, LT loss for training data, Avg average