Table 2 Number of radiomics features extracted for different feature classes and image filters.

From: Improving prognostic performance in resectable pancreatic ductal adenocarcinoma using radiomics and deep learning features fusion in CT images

Filter/features

First-order

GLCM

GLDM

GLRLM

GLSZM

NGTDM

Shape

Total

Exponential

16

0

11

12

7

0

0

46

Gradient

18

23

14

16

16

5

0

92

lbp

56

0

44

48

28

0

0

176

Logarithm

18

23

14

16

16

5

0

92

Original

18

23

14

16

16

5

12

104

Square

18

23

14

16

16

4

0

91

Squareroot

18

23

14

16

16

5

0

92

Wavelet

144

184

112

128

128

39

0

735

Total

306

299

237

268

243

63

12

1428

  1. GLCM grey level co-occurrence matrix, GLDM grey level difference matrix, GLRLM grey level run length matrix, GLSZM gray level size zone, NGTDM neighboring gray tone difference matrix.
Back to article page