Fig. 7
From: Nuclei instance segmentation from histopathology images using Bayesian dropout based deep learning
Uncertainty quantification estimated by the BayesNuSeg on the test set of the PanNuke dataset. The first column contains the ground truth nuclei (red contours) and the mean of MC dropout sampling predictions (blue contours), overlaid on the original image. The color bar next to the model uncertainty represents the uncertainty in range [0, 1], where 0 denotes lower and 1 highly uncertain predictions