The practical clinical role of machine learning models with different algorithms in predicting prostate cancer local recurrence after radical prostatectomy

Table 4 Selected features of radiomics model and combined model

Radiomic model	Combined model
T2WI_log-sigma-1-mm-3D_ngtdm_Contrast	T2WI_log-sigma-1-mm-3D_ngtdm_Contrast
ADC_exponential_glrlm_LongRunLowGrayLevelEmphasis	ADC_exponential_glrlm_LongRunLowGrayLevelEmphasis
ADC_exponential_glszm_SmallAreaEmphasis	ADC_exponential_glszm_SmallAreaEmphasis
DWI_exponential_glszm_SmallAreaLowGrayLevelEmphasis	DWI_exponential_glszm_Zone%
DWI_exponential_glszm_Zone%	DWI_gradient_glrlm_LongRunHighGrayLevelEmphasis
DWI_gradient_glrlm_LongRunHighGrayLevelEmphasis	DWI_gradient_glrlm_LongRunLowGrayLevelEmphasis
DWI_gradient_glrlm_LongRunLowGrayLevelEmphasis	DWI_square_glrlm_RunVariance
DWI_square_glrlm_RunVariance	DCE_lbp-3D-m1_firstorder_RootMeanSquared
DWI_wavelet-HHL_firstorder_Mean	DCE_lbp-3D-m2_firstorder_Median
DCE_lbp-3D-m1_firstorder_RootMeanSquared	DCE_wavelet-LHL_firstorder_Median
DCE_lbp-3D-m2_firstorder_Median	PI-RR
DCE_lbp-3D-m2_firstorder_Skewness
DCE_wavelet-LHL_firstorder_Mean
DCE_wavelet-LHL_firstorder_Median

T2WI = T2-weighted imaging; DWI = diffusion-weighted imaging; ADC = apparent diffusion coefficient; DCE = dynamic contrast-enhanced; PI-RR = Prostate Imaging for Recurrence Reporting

ISSN: 1470-7330