Fig. 4From: Application of convolutional neural networks towards nuclei segmentation in localization-based super-resolution fluorescence microscopy imagesSegmentation examples for a colon tissue image (H3K9me3) depicting a high-grade dysplasia. Original test image (a) is segmented using (b) Mask R-CNN, (c) ANCIS and (d) StarDist network models. Training and testing were done on colon tissue data. Red arrows indicate False Positives, yellow arrows point out False Negatives, and pink and orange arrows indicate over- and under-segmentations, respectivelyBack to article page