Table 6 Performance comparisons of subclass tumor using Brats2020 data.
From: Enhancing brain tumor segmentation in MRI images using the IC-net algorithm framework
Algorithms | DSC (WT) | DSC (TC) | DSC (ET) |
|---|---|---|---|
Self-calibrated attention U-Net50 | 0.905 | 0.821 | 0.781 |
Double attention U-Net33 | 0.8912 | 0.8427 | 0.7915 |
AD-Net22 | 0.872 | 0.823 | 0.803 |
dResU-Net24 | 0.8660 | 0.8004 | 0.8357 |
Attention-based CNN with U-Net44 | 0.90 | 0.86 | 0.83 |
Aggregation-and-Attention Network52 | 0.93 | 0.88 | 0.87 |
Znet50 | 0.839 | 0.762 | 0.746 |
Automated Multimodal53 | 0.840 | 0.780 | 0.760 |
Deep multi-task learning with multi-depth fusion54 | 0.860 | 0.772 | 0.700 |
Convolutional block attention - V-Net55 | 0.876 | 0.769 | 0.670 |
AGSE-VNet56 | 0.68 | 0.85 | 0.70 |
IC-Net | 0.998717 | 0.888930 | 0.866183 |
