*2.6. Correlation Analysis and Statistical Analyses*

Statistical analyses were conducted using Matlab 2018b. Continuous variables were described using means and standard deviations, and categorical variables were summarized using frequencies. The normality distribution of continuous data was verified with a one-sample Kolmogorov-Smirnov Test. Pearson correlation was used to investigate the relationship between the structural measures and clinical scores. Age, gender, and educational level were regressed out before correlation analysis. The independent *t*-test were applied to perform the group comparisons for continuous demographic and clinical

variables. The categorical demographic was computed using a Chi-square test. Statistical significance was set at *p* < 0.05 (false discovery rate [FDR] corrected). Missing data were not included in all analysis.

#### **3. Results**
