Table 1.
Different U-Net models and their applications.
| Model | Application | Refs. |
|---|---|---|
| U-Net | Segmentation of neuronal structures in electron microscopic stacks | 18 |
| ResUNet | Semantic segmentation of remotely sensed data | 51 |
| Dense UNet | Semantic segmentation with a small number of samples | 52 |
| DUNet | Retinal vessel segmentation | 53 |
| Attention U-Net | Multi-class CT abdominal segmentation | 54 |
| UNet++ | Segmentation of cell nuclei, colon polyp, liver and lung nodule | 55 |
| MultiResUNet | Multimodal biomedical image segmentation. Used datasets are Fluorescence Microscopy, Electron Microscopy, Dermoscopy, Endoscopy and MRI | 56 |
| RAUNet | Semantic Segmentation of Cataract Surgical Instruments | 57 |
| Inception U-Net | Brain magnetic resonance imaging (MRI) tumor segmentation | 58 |
| U-Net GAN | Used to achieve detailed per-pixel feedback while sustaining the global coherence of synthesized images, by providing the global image feedback as well | 59 |
| V-Net | V-Net was first used to detect prostate segments from MRI volumes. | 60 |
| W-Net | W-Net is a W-shaped unsupervised image segmentation model. | 61 |