Skip to main content

Table 4 Performance of several nuclei segmentation methods on the multiple organ H&E stained histopathology image dataset (MOSID)

From: An automatic nuclei segmentation method based on deep convolutional neural networks for histopathology images

MethodsPrecisionRecallF1-ScoreADCAJI
CP [35]N/AN/A0.4050.5970.123
Fiji [36]N/AN/A0.6650.6490.273
CNN2 [13]N/AN/A0.7540.6930.348
CNN3 [20]N/AN/A0.8270.7620.508
NB [22]0.8360.8940.8520.809N/A
NucSeg0.913 ±0.0030.821 ±0.0040.861 ±0.0010.812 ±0.0010.669 ±0.001
NucSeg-N0.897 ±0.0040.813 ±0.0040.849 ±0.0020.805 ±0.0020.649 ±0.004
NucSeg-P0.909 ±0.0020.777 ±0.0040.835 ±0.0020.809 ±0.0010.664 ±0.002
NucSeg-NP0.899 ±0.0040.777 ±0.0050.828 ±0.0020.701 ±0.0020.647 ±0.005