Research Article
A Novel Medical Image Denoising Method Based on Conditional Generative Adversarial Network
Table 1
Average PSNR and SSIM of the denoised images by different denoising models on the JSRT dataset.
| Methods | PSNR (mean) | SSIM (mean) | JSRT | LIDC | JSRT | LIDC |
| Noise | 15.4436 | 18.3036 | 0.1170 | 0.1390 | WGAN+perceptual [27] | 16.3217 | 20.1425 | 0.7635 | 0.7930 | CGAN+perceptual | 12.5834 | 18.3698 | 0.4420 | 0.6326 | SRGAN [28] | 16.2250 | 21.4650 | 0.7779 | 0.8023 | CGAN+L1 [33] | 32.6180 | 34.1577 | 0.8945 | 0.8936 | Zhong et al. [14] | 32.9856 | 33.0452 | 0.8960 | 0.8847 | The proposed method | 33.2642 | 35.1086 | 0.9206 | 0.9328 |
|
|