Skip to main content

Table 2 Features selected for validation classification task

From: Prediction of high infiltration levels in pituitary adenoma using MRI-based radiomics and machine learning

Radiomic feature Radiomic class Filter Mask
Maximum2DDiameterSlice shape original original
LowGrayLevelZoneEmphasis glszm log-sigma-5-0 original
Skewness firstorder wavelet-LLH original
Maximum firstorder wavelet-LHH original
Minimum firstorder wavelet-LHH original
ZoneEntropy glszm wavelet-LHH original
MCC glcm wavelet-HLL original
Minimum firstorder wavelet-HLH original
Kurtosis firstorder wavelet-HHL original
SmallDependenceLowGrayLevelEmphasis gldm wavelet-HHL original
Mean firstorder wavelet-HHH original
Flatness shape original egde
DependenceEntropy gldm log-sigma-1-0 egde
GrayLevelNonUniformity glszm log-sigma-4-0 egde
Busyness ngtdm wavelet-LHH egde
LargeDependenceHighGrayLevelEmphasis gldm wavelet-HLL egde
SizeZoneNonUniformityNormalized glszm wavelet-HHH egde
Contrast ngtdm wavelet-HHH egde
SmallDependenceLowGrayLevelEmphasis gldm wavelet-HHH egde
  1. Features selected for validation classification task