Association of lower extremity peripheral arterial disease with quantitative muscle features from computed tomography angiography

Insights into Imaging

Table 3 Multivariable analysis results of the two logistic regression models for peripheral arterial disease prediction

Variables in logistic regression model	Coefficient^a	OR (95% CI)	p value
*LRM-I*
histogram_10Percentile	-1.29	0.28 (0.13, 0.61)	0.001
glcm_Correlation	NR	NR	NR
gldm_DependenceNonUniformityNormalized	-0.67	0.51 (0.27, 0.98)	0.04
glszm_GrayLevelNonUniformityNormalized	NR	NR	NR
glszm_SmallAreaLowGrayLevelEmphasis	NR	NR	NR
Constant^b	-0.85	0.43 (NR, NR)	0.02
*LRM-II*
CTA runoff score	1.18	3.27 (1.42, 7.53)	0.006
histogram_10Percentile	-1.11	0.33 (0.14, 0.77)	0.01
glcm_Correlation	NR	NR	NR
gldm_DependenceNonUniformityNormalized	-0.68	0.51 (0.25, 1.03)	0.06
glszm_GrayLevelNonUniformityNormalized	NR	NR	NR
glszm_SmallAreaLowGrayLevelEmphasis	NR	NR	NR
Constant^b	-0.90	0.41 (NR, NR)	0.03

Abbreviations: OR Odds ratio, CI Confidence interval, LRM Logistic regression model, GLCM Gray level co-occurrence matrix, GLDM Gray level dependence matrix, GLSZM Gray level size zone matrix, CTA Computed tomography angiography, NR Not reported
^aCoefficient of variables in logistic regression equation
^bIntercept in logistic regression equation