Table 1 Comparison of predictive indicators of individual models.
From: Enhanced digital pathology image recognition via multi-attention mechanisms: the MACC-Net approach
Model | Accuracy | Precision | Recall | F1-score | IOU | DSC |
---|---|---|---|---|---|---|
U-Net41 | 0.968 | 0.881 | 0.889 | 0.783 | 0.651 | 0.783 |
Att-Unet42 | 0.976 | 0.841 | 0.901 | 0.825 | 0.707 | 0.825 |
SegNet43 | 0.968 | 0.898 | 0.887 | 0.783 | 0.652 | 0.783 |
UTNet44 | 0.971 | 0.823 | 0.835 | 0.781 | 0.649 | 0.781 |
CE-Net45 | 0.945 | 0.741 | 0.832 | 0.739 | 0.662 | 0.739 |
TransUnet46 | 0.973 | 0.817 | 0.929 | 0.809 | 0.684 | 0.809 |
Swin-Unet47 | 0.976 | 0.856 | 0.923 | 0.822 | 0.702 | 0.822 |
ResUnet48 | 0.972 | 0.878 | 0.916 | 0.826 | 0.709 | 0.826 |
ChannelNet49 | 0.979 | 0.901 | 0.921 | 0.841 | 0.718 | 0.841 |
MACC-Net (our) | 0.981 | 0.907 | 0.931 | 0.847 | 0.728 | 0.847 |