Figure 11
From: NetMets: software for quantifying and visualizing errors in biological network segmentation
Segmentation of cerebellar climbing fibers from the DIADEM Challenge data set [17]. (a) Minimum intensity projection of the raw image data set. (b) The ground-truth model with geometric error rendered using a diverging color map. (c) The test-case model shows detected edges rendered in different colors. Undetected edges are gray. (d-g) A close-up of the indicated regions (arrows) showing geometric error for the ground-truth (left) and test cases (right). (h-i) Edge mapping between the ground-truth and test case. Undetected nodes are shown in red. (j-k) Edge-mapping provides an estimate of fiber correspondence in regions that are incorrectly connected. The result of the DIADEM Metric for this data set is M = 0.315.