Skip to main content
. 2021 Oct 29;10(21):5064. doi: 10.3390/jcm10215064

Table 3.

Comparison of the clinical, epidemiological, metabolic, and radiomics feature distributions between the two PET scanners.

SCANNER D690 vs. D-STE
Clinical-Epidemiological Features p-Value Second-Order Statistics p-Value
Age 0.659 GLCM entropy_log10 <0.001
Gender 0.659 GLCM entropy_log2 <0.001
Size 0.746 GLCM dissimilarity <0.001
“Conventional” PET features GLRLM SRE <0.001
SUVmax <0.001 GLRLM LRE <0.001
SUVmean <0.001 GLRLM LGRE 0.001
SUVlbm <0.001 GLRLM HGRE <0.001
SUVbsa <0.001 GLRLM SRLGE 0.002
L-L SUV R 0.003 GLRLM SRHGE <0.001
L-BP SUV R 0.009 GLRLM LRLGE <0.001
MTV <0.001 GLRLM LRHGE 0.006
TLG <0.001 GLRLM GLNU <0.001
First-order statistics GLRLM RLNU 0.052
Histo skewness 0.316 GLRLM RP <0.001
Histo kurtosis 0.758 NGLDM coarseness 0.799
Histo excess kurtosis 0.758 NGLDM contrast <0.001
Histo entropy_log10 <0.001 NGLDM busyness <0.001
Histo entropy_log2 <0.001 GLZLM SZE <0.001
Histo energy <0.001 GLZLM LZE <0.001
Shape volume_mL 0.917 GLZLM LGZE 0.003
Shape volume_vx 0.917 GLZLM HGZE <0.001
Shape sphericity 0.037 GLZLM SZLGE 0.877
Shape compacity 0.859 GLZLM SZHGE <0.001
Second-order statistics GLZLM LZLGE <0.001
GLCM homogeneity <0.001 GLZLM LZHGE 0.002
GLCM energy <0.001 GLZLM GLNU 0.803
GLCM contrast < 0.001 GLZLM ZLNU <0.001
GLCM correlation 0.052 GLZLM ZP < 0.001