Skip to main content

Table 3 Top 10 radiomic features with highest gain value in 3D-analysis patch-based model

From: Detection of pancreatic cancer with two- and three-dimensional radiomic analysis in a nationwide population-based real-world dataset

Feature

Gain value

GLCM: correlation (wavelet-LLL)

54.27

GLCM: IMC2 (original)

36.62

GLCM: IMC2 (wavelet-LLL)

24.27

First order: mean (log-sigma-3–0-mm-3D)

17.12

GLDM: small dependence emphasis (wavelet-HHH)

13.34

GLDM: gray level variance (wavelet-HHH)

11.97

First order: minimum (wavelet-LLL)

11.80

GLCM: inverse variance (wavelet-LHH)

10.43

GLCM: IMC2 (wavelet-HHL)

10.14

GLCM: inverse variance (wavelet-HHH)

9.67

  1. GLCM Gray Level Co-occurrence Matrix, IMC Informational Measure of Correlation, GLDM: Gray Level Dependence Matrix
Back to article page

BMC Cancer

ISSN: 1471-2407

Contact us