Research Article
R2AU-Net: Attention Recurrent Residual Convolutional Neural Network for Multimodal Medical Image Segmentation
Table 1
Performance comparison of R2AU-Net and other networks on the ISIC dataset.
| Methods | F1-score | Sensitivity | Specificity | Accuracy | AUC |
| SegNet [19] | 0.7092 | 0.8056 | 0.8147 | 0.8121 | 0.8101 | U-Net [3] | 0.6954 | 0.7160 | 0.8636 | 0.8216 | 0.7898 | Attention U-Net [20] | 0.7541 | 0.7627 | 0.8966 | 0.8585 | 0.8297 | RU-Net [21] | 0.679 | 0.792 | 0.928 | 0.880 | 0.8374 | R2U-Net [21] | 0.691 | 0.726 | 0.971 | 0.904 | 0.8565 | Attention Res U-Net | 0.8545 | 0.8363 | 0.9519 | 0.9190 | 0.8941 | R2AU-Net | 0.8660 | 0.8214 | 0.9699 | 0.9277 | 0.8957 |
|
|