TABLE 2.

Eleven selected radiomics features for predicting the molecular subgroups of medulloblastoma patients.

No.	Selected features	Type	Sequence	Transform	pFDR
f1	Median	Intensity	Fluid-attenuated inversion recovery (FLAIR)	Wavelet. HHH	<0.001
f2	Gray-level co-occurrence matrix (GLCM) cluster shade	Texture	FLAIR	Wavelet. HHL	<0.001
f3	Mean	Intensity	FLAIR	Wavelet. HLL	<0.001
f4	Root mean squared	Intensity	FLAIR	Wavelet. LHL	<0.001
f5	Gray-level size zone matrix small area low gray level emphasis	Texture	T1	Wavelet. HHH	<0.001
f6	Median	Intensity	T1c	Wavelet. LHL	<0.001
f7	Skewness	Intensity	T1c	Wavelet. HLL	<0.001
f8	Maximum	Intensity	Apparent diffusion coefficient	Wavelet. LLH	<0.001
f9	GLCM inverse variance	Texture	T1	Wavelet. HLH	<0.001
f10	Skewness	Intensity	FLAIR	Wavelet. LHH	<0.001
f11	Skewness	Intensity	FLAIR	Wavelet. LHL	<0.001