Research Article
[Retracted] Antialiasing Attention Spatial Convolution Model for Skin Lesion Segmentation with Applications in the Medical IoT
Table 4
Performance evaluation of the state-of-the-art segmentation methods in the PH2 dataset.
| Model name | Jaccard index | Accuracy | Recall | Precision | F1 |
| Unet [36] | 0.79 | 0.91 | 0.89 | 0.89 | 0.87 | Attention Unet [37] | 0.80 | 0.92 | 0.90 | 0.89 | 0.88 | Unet++ [38] | 0.79 | 0.91 | 0.88 | 0.90 | 0.87 | Recurrent U-Net [39] | 0.80 | 0.92 | 0.93 | 0.87 | 0.88 | Mask-RCNN [25] | 0.84 | 0.94 | 0.93 | — | — | DeepLab V3+ [26] | 0.81 | 0.92 | 0.94 | — | — | Proposed model | 0.95 | 0.96 | 0.95 | 0.93 | 0.94 |
|
|