Skip to main content
. 2020 Jan 20;41(8):2104–2120. doi: 10.1002/hbm.24933

Table 1.

Segmentation performance for the new algorithmic implementations and the standard Morel method applied on the Dataset II template

Classifiers % global error % TP (nuclei)
3‐NN + CS 7.0 74.8
Parzen + CS 6.9 88.4
Morel atlas 13.3 69.7

Note: The proposed implementations outperformed standard Morel segmentation on both performance metrics: Global error and true positive (TP) rate. The bold values indicate the best results as lowest global error and highest TP.