Table 10 Performance of different models on the HRF dataset. Significant values are in bold.
From: An effective vessel segmentation method using SLOA-HGC
Model | Miou | Dice | ACC | HD | FPS |
|---|---|---|---|---|---|
UNet | 0.703 ± 0.005 | 0.626 ± 0.011 | 0.952 ± 0.002 | 767.217 ± 5.064 | 5.394 ± 0.267 |
SA-UNet | 0.722 ± 0.020 | 0.658 ± 0.026 | 0.955 ± 0.008 | 802.238 ± 7.555 | 4.681 ± 0.389 |
UNET++ | 0.727 ± 0.018 | 0.666 ± 0.017 | 0.956 ± 0.007 | 576.168 ± 8.069 | 3.682 ± 0.264 |
RIMNet | 0.707 ± 0.010 | 0.631 ± 0.017 | 0.954 ± 0.002 | 462.498 ± 5.349 | 4.832 ± 0.209 |
DeepLabV3 | 0.718 ± 0.019 | 0.657 ± 0.012 | 0.950 ± 0.005 | 614.745 ± 8.894 | 4.007 ± 0.205 |
BTS-DSN | 0.717 ± 0.016 | 0.649 ± 0.028 | 0.955 ± 0.001 | 658.788 ± 9.313 | 4.598 ± 0.501 |
U2Net | 0.735 ± 0.011 | 0.679 ± 0.023 | 0.958 ± 0.007 | 567.485 ± 8.611 | 2.575 ± 0.469 |
SLOA-HGC | 0.741 ± 0.024 | 0.689 ± 0.031 | 0.958 ± 0.007 | 501.277 ± 7.102 | 3.264 ± 0.308 |
