Table 3.

Multiple Linear Regression Tables

Regression		CMA Reconstruction DWI-Derived Parameter
Model	Coefficient	ADC	D	F
1	Intercept	2.262 ± 0.190	1.271 ± 0.061	21.132 ± 2.316
	STE	−0.037 ± 0.052, P = 0.479	−0.037 ± 0.017, P = 0.032*	1.262 ± 0.636, P = 0.051.
	INF	−0.024 ± 0.090, P = 0.792	0.010 ± 0.029, P = 0.718	−1.319 ± 1.099, P = 0.234
	BAL	−0.179 ± 0.091, P = 0.052	−0.021 ± 0.029, P = 0.465	−2.182 ± 1.107, P = 0.052.
	FIB	−0.008 ± 0.047, P = 0.866	0.000 ± 0.015, P = 0.993	−0.529 ± 0.570, P = 0.356
2	Intercept	2.345 ± 0.267	1.207 ± 0.084	22.570 ± 3.277
	STE	−0.057 ± 0.051, P = 0.271	−0.043 ± 0.016, P = 0.0099**	0.935 ± 0.628, P = 0.140
	NASH	−0.178 ± 0.145, P = 0.223	0.037 ± 0.046, P = 0.415	−2.925 ± 1.774, P = 0.103
	FIB	−0.056 ± 0.037, P = 0.136	−0.008 ± 0.012, P = 0.504	−1.284 ± 0.457, = 0.006**
Regression		BLN Reconstruction DWI-Derived Parameter
Model	Coefficient	ADC	D	F
1	Intercept	1.890 ± 0.170	0.944 ± 0.045	20.971 ± 2.521
	STE	−0.057 ± 0.047, P = 0.226	−0.059 ± 0.012, P = 7.6e-06***	1.160 ± 0.693, P = 0.098.
	INF	−0.025 ± 0.081, P = 0.761	−0.008 ± 0.021, P = 0.718	−0.678 ± 1.197, P = 0.573
	BAL	−0.088 ± 0.081, P= 0.282	0.026 ± 0.021, P = 0.221	−1.208 ± 1.204, P = 0.319
	FIB	−0.024 ± 0.042, P= 0.561	0.013 ± 0.011, P = 0.256	−1.342 ± 0.621, P = 0.033*
2	Intercept	1.966 ± 0.236	0.879 ± 0.061	22.659 ± 3.488
	STE	−0.066 ± 0.045, P = 0.149	−0.061 ± 0.012, P = 1.48e-06***	1.047 ± 0.669, P = 0.121
	NASH	−0.117 ± 0.128, P = 0.360	0.059 ± 0.033, P = 0.080.	−2.229 ± 1.888, P = 0.241
	FIB	−0.048 ± 0.033, P = 0.146	0.014 ± 0.008, P = 0.103	−1.692 ± 0.487, P = 0.0008***

The coefficient ± standard error and significance of the effects of each histologic feature of NAFLD are reported above for both reconstruction methods and both models. The P-value range is specified by the number of adjacent asterisks, as follows:

^***0–0.001,

^**0.001–0.01,

^*0.01–0.05.