Predicting micropapillary or solid pattern of lung adenocarcinoma with CT-based radiomics, conventional radiographic and clinical features

Table 3 radiomic features selected by LASSO regression and their coefficiencies

Coefficients	Feature family	Feature subtype
-0.01469447		Intercept
0.23464576	original	shape_Flatness
-0.07955321	square	glcm_InverseVariance
-0.23345119	square	glrlm_ShortRunLowGrayLevelEmphasis
0.11685419	exponential	firstorder_90Percentile
0.27726069		firstorder_Mean
0.06374112		firstorder_TotalEnergy
0.01856495		glszm_SizeZoneNonUniformityNormalized
-0.01994271	wavelet	HHH_firstorder_Kurtosis
0.17396774		HHH_glcm_Idmn
-0.13376102		HLH_glcm_Idn
0.01637181		HLL_firstorder_Skewness
0.23530962		HLL_glszm_ZoneEntropy
0.15230831		LHH_glcm_MCC
0.13845049		LHL_glszm_SmallAreaLowGrayLevelEmphasis
-0.13695941		LLH_glcm_MCC
0.09393939		LLH_glszm_GrayLevelNonUniformityNormalized
0.16229105		LLL_firstorder_90Percentile
0.36867058		LLL_firstorder_Kurtosis
0.40500583		LLL_glcm_JointEntropy

ISSN: 1465-993X