Table 4.

An overview of features extracted from MRI and feature selection methods.

References	Original feature set	Dimension	Extraction tool	Selection method
Zhang et al. (28)	First-order features, GLCM, GLZLM, NGLDM, GLRLM, gender, age	40+2	LifeX	Distance Correlation, RF, Lasso, XGBoost, and GBDT
Ma et al. (31)	First-order statistics, shape, GLCM, GLRLM, GLSZM, NGTDM, GLDM, Wavelet features	1,874	MATLAB	Lasso
Gutta et al. (41)	First-order feature, shape, GLCM, GLRLM, GLSZM, NGTDM, GLDM	1,284	PyRadiomics	Importance score from gradient boosting algorithm
Le et al. (56)	Intensity, image derivative, geodesic, texture, posterior probability maps	704	Cancer Imaging Phenomics Toolkit	F-score evaluation criterion, recursive feature elimination
Al-Saffar and Yildirim (40)	GLCM, intensity	40		MI evaluation criterion, SVD
Kandemirli et al. (45)	Intensity, shape, GLCM, GLRLM, GLSZM, GLDM	3,255	Pyradiomics	Intraclass correlation coefficient, XGBoost's inherent feature selection and additional feature selection method
Gao et al. (57)	First order features, shape, GLCM, GLRLM, GLSZM	1,421	PyRadiomics	Chi2 verification, Seaborn library, inherent feature selection of RF
Chen et al. (32)	Local feature, intensity, shape, texture and wavelet features	1,091	MATLAB	Intraclass correlation coefficients, feature scores of RF, forward search strategy

gray-level co-occurrence matrix; GLZLM, gray-level zone length matrix; NGLDM, neighborhood gray-level dependence matrix; GLRLM, gray-level run length matrix; GLSZM, gray-level size zone matrix; NGTDM, neighboring gray tone difference matrix; GLDM, gray-level dependence matrix.