Table 2 The performance of the proposed Vanilla U-Net model
From: Convolutional neural network for automated mass segmentation in mammography
Model | Training | Validation | Test | ||
|---|---|---|---|---|---|
| Â | Dice | Loss | Dice | Loss | Dice |
-Proposed Vanilla U-Net, with Aug. | 0.972 | 0.016 | 0.942 | 0.024 | 0.951 |
-Proposed Vanilla U-Net, no Aug. | 0.910 | 0.064 | 0.842 | 0.144 | 0.922 |