Skip to main content

Table 2 Using CT and X-Ray images individually and combined, PSNR and SSIM calculated at different noise levels, i.e. specific noise (\(N_{sp}\)) and blind noise(\(N_{bl}\))

From: DMF-Net: a deep multi-level semantic fusion network for high-resolution chest CT and X-ray image de-noising

Trained

Image

Noise=15

Noise=20

Noise=25

  

PSNR

SSIM

PSNR

SSIM

PSNR

SSIM

Specific Noise

CT(No Aug)

29.33

0.8585

27.99

0.8230

26.95

0.7981

CT(Aug)

29.06

0.8592

27.95

0.8316

26.62

0.7902

X-Ray(No Aug)

30.94

0.8848

29.72

0.8560

28.72

0.8282

X-Ray(Aug)

31.03

0.8896

29.95

0.8637

29.19

0.8442

CT+X-Ray

30.24

0.8794

28.91

0.8445

28.32

0.8314

Blind Noise

CT(No Aug)

26.66

0.7973

27.10

0.8082

24.96

0.7444

CT(Aug)

28.23

0.8294

27.11

0.7934

26.57

0.7787

X-Ray(No Aug)

28.59

0.8445

27.72

0.8185

26.94

0.7926

X-Ray(Aug)

28.92

0.8525

28.01

0.8240

27.19

0.7954

CT+X-Ray

29.30

0.8582

28.37

0.8270

27.53

0.8020