Skip to main content
. 2023 Mar 23;15(7):1932. doi: 10.3390/cancers15071932

Table A1.

Description on the type of radiomics features that are extracted from each region of interest from PET and CT images. More information on the type of features is present in PyRadiomics documentation (https://pyradiomics.readthedocs.io/en/latest/features.html (accessed on 20 March 2023)).

Radiomics Feature Description
Gray level co-occurrence matrix
(GLCM)		 Gray level co-occurrence matrix features describe the second order joint probability function of the voxel intensities. These features include measures such as contrast, correlation, energy, and homogeneity. In this study, 24 GLCM features were extracted using PyRadiomics.
Gray level difference matrix
(GLDM)		 Gray level difference matrix features describe the distribution of gray level differences within the ROI. These features include measures such as coarseness, contrast, and busyness. In this study, 14 GLDM features were extracted using
PyRadiomics.
Gray level run length matrix (GLRLM) Gray level run length matrix features describe the length of runs of consecutive voxels with the same gray level. These features include measures such as short-run emphasis, long-run emphasis, and run percentage. In this study, 16 GLRLM features were extracted using PyRadiomics.
Gray level size zone matrix (GLSZM) Gray level size zone matrix features describe the size of zones of consecutive voxels with the same gray level. These features include measures such as zone size, zone percentage, and zone entropy. In this study, 16 GLSZM features were extracted using PyRadiomics.
Neighbouring gray tone difference matrix (NGTDM) Neighbouring gray tone difference matrix features describe the distribution of voxel-level texture primitive patterns. These features include measures such as coarseness and contrast. In this study, 5 NGTDM features were extracted using PyRadiomics.
First order statistics First order statistics describe the distribution of voxel intensities within the ROI. These features include measures such as mean, median, skewness, and kurtosis. In this study, 18 first order features were extracted using PyRadiomics.
Shape-based (3D) Shape features describe the shape and size of the ROI. These features include measures such as volume, surface area, sphericity, and compactness. In this study, 14 shape features were extracted using PyRadiomics.