Table 2 The list of the detailed features used in the study.
From: The importance of feature aggregation in radiomics: a head and neck cancer study
Family | Feature | Quantitative feature |
|---|---|---|
Filter-based | Laplacian of Gaussian Gabor Sobel | Sigma \(= 2\hbox {mm}\), radius \(= 4\hbox {mm}\) Sigma \(= 11/3\), freq. \(= 0.4\), radius \(= 4 \,\hbox {mm}\) Kernel size = 3 × 3 × 3 |
Grey-level texture matrices | GLRLM Radius \(= 2 \,\hbox {mm}\) Angles \(=\) Half of all directions (3D), symmetrical Discretization \(= 64\) grey levels | ShortRunEmphasis LongRunEmphasis GreyLevelNonuniformity RunLengthNonuniformity LowGreyLevelRunEmphasis HighGreyLevelRunEmphasis ShortRunLowGreyLevelEmphasis ShortRunHighGreylevelEmphasis LongRunLowGreyLevelEmphasis LongRunHighGreyLevelEmphasis |
GLCM Radius \(= 2 \,\hbox {mm}\) Angles \(=\) Half of all directions (3D), symmetrical Discretization \(= 64\) grey levels | Energy InverseDifferenceMoment Entropy HaralickCorrelation ClusterShade ClusterProminence Inertia Correlation |
