Table 5.

Differences between post-processing methods (PPMs): Detailed analysis

PPM-pair	P	Slope	Intercept	Outliers	CV
(1) ADC_1 vs. ADC_2	<0.0001	n.s.	n.s.	4	10.2
(2) ADC_1 vs. ADC_3	<0.0001	n.s.	n.s.	4	10.4
(3) ADC_1 vs. ADC_4	<0.0001	n.s.	n.s.	1	3.2
(4) ADC_2 vs. ADC_3	0.003	n.s.	n.s.	1	1.1
(5) ADC_2 vs. ADC_4	<0.0001	0.12	-0.11	3	9
(6) ADC_3 vs. ADC_4	<0.0001	0.13	-0.1	3	9.1

Note P value according to Wilcoxon signed-rank test

If there was a significant (P < 0.05) correlation between x (PPM_A) and y (PPM_A minus PPM_B), the slope and intercept of the corresponding regression equation are given. These results indicate a proportional error within PPM-pairs (5) and (6); see also Fig. 5. Accordingly, the difference between ADC_2/ADC_3 and ADC_4 increased with rising ADC-levels

ADC apparent diffusion coefficient, Outliers number of ADC-values beyond +/-1.96 * SD, CV coefficient of variation (%)