Table 5 Impact of different attention mechanisms on the model.
From: Enhanced digital pathology image recognition via multi-attention mechanisms: the MACC-Net approach
CA | SA | PA | Accuracy | IOU | DSC |
---|---|---|---|---|---|
0.961 | 0.656 | 0.722 | |||
√ | 0.965 | 0.667 | 0.759 | ||
√ | 0.966 | 0.677 | 0.768 | ||
√ | 0.957 | 0.657 | 0.755 | ||
√ | √ | 0.975 | 0.688 | 0.812 | |
√ | √ | 0.977 | 0.694 | 0.811 | |
√ | √ | 0.977 | 0.691 | 0.807 | |
√ | √ | √ | 0.981 | 0.728 | 0.847 |