Skip to main content

Table 1 Quantitative results of pulmonary nodule segmentation in which highlighted values represent the best result of each metric

From: HR-MPF: high-resolution representation network with multi-scale progressive fusion for pulmonary nodule segmentation and classification

Methods Acc SE Prec MIoU DSC
ENet 0.9895 0.8787 0.9032 0.8960 0.8908
SegNet 0.9873 0.8914 0.8852 0.8906 0.8859
PSPNet 0.9888 0.8742 0.8936 0.8900 0.8838
DeeplabV3+ 0.9923 0.9168 0.9241 0.9222 0.9204
UNet++ 0.9918 0.9150 0.9285 0.9263 0.9227
Fast SCNN 0.9859 0.8130 0.8978 0.8605 0.8477
DFANet 0.9874 0.8208 0.9115 0.8716 0.8612
FANet 0.9901 0.9012 0.9127 0.9049 0.8934
SPNet 0.9931 0.9165 0.9117 0.9106 0.9096
Ours 0.9937 0.9377 0.9427 0.9378 0.9373