Table 3 Quantitative analysis of image quality and performance of deep-learning models in the reconstructed images from different methods with dataset No. 2 in Group 2
From: Artifact suppression for breast specimen imaging in micro CBCT using deep learning
| Â | SSIM | CNR | PSNR |
|---|---|---|---|
Image from FBP | 0.601 | 2.290 | 23.654 |
Image from IR | 0.777 | 5.710 | 28.208 |
Image from the sinogram with linear interpolation | 0.789 | 5.210 | 29.669 |
Image from denoise with ResU-Net | 0.637 | 5.520 | 24.640 |
Image from the sinogram with U-Net | 0.856 | 7.360 | 29.182 |
Image from the synthesized sinogram with the modified U-Net and ResU-Net models in the proposed method | 0.866 | 8.330 | 30.612 |