Table 1
Comparative results of segmentation models.
| Model | DSC | mAP | AUC ROC | MSSIM |
|---|---|---|---|---|
| UNet | 0.9680 | 0.9949 | 0.9960 | 0.1313 |
| UNet++ | 0.9690 | 0.9953 | 0.9960 | 0.1356 |
| SwinUNet | 0.9637 | 0.9939 | 0.9950 | 0.1409 |
| PE-UNet-I | 0.9649 | 0.9950 | 0.9960 | 0.1418 |
| PE-UNet-D | 0.9685 | 0.9952 | 0.9962 | 0.1315 |
| PE-UNet-L | 0.9746 | 0.9970 | 0.9976 | 0.1380 |
Notes. Comparative results of segmentation models with respect to four metrics: DSC and MSSIM which are threshold dependent, and mAP and AUC ROC which are integrated over the threshold range. All results are averaged over 5 folds. The best result for every metric is indicated in bold, significance tests for the DSC metric are provided in Table 2.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.