In other words, the statistical tests they used are appropriate for
parametric statistics (means) but not appropriate for non-parametric statistics (medians).
Parametric statistics are based on the theory of a normal distribution.
3)
parametric statistics are robust to modest violations of normality (non-equality of variances, samples from non-normally distributed populations) and thus can be used with non-normal distributions, as long as the normality violations are not excessive.
All data were subjected to homogeneity (Hartley) and normality (Shapiro-Wilk) tests to verify the assumptions of
parametric statistics. Data were subjected to analysis of variance (ANOVA), and means were compared by the Tukey test at 5% probability, using the software SISVAR 5.4 (Ferreira 2011).
The processing steps are briefly explained below as follows: (1) estimate and write: the images were bias-corrected and segmented into GM, WM, and CSF; (2) DARTEL create template: a customized template was created for our study; (3) DARTEL existing template: once the study-specific template was created from the above step, the remaining subjects were registered nonlinearly to this template using DARTEL existing template module; (4) normalize to MNI space; (5) smooth: after normalizing and registering all subjects to MNI space, the resulting images were modulated (without including affine component) and smoothed using a full-width half-maximum (FWHM) of 8 mm; and (6)
parametric statistics: the smoothed images were used for statistical inference.
Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were log transformed for
parametric statistics. Post-hoc analysis was done by the Bonferroni correction.