Medcalc v 20

Values were described as medians with interquartile ranges (IQRs) unless otherwise indicated. The Wilcoxon test was used to compare the time of each variable. To compare the differences in variables between two groups (i.e., patients with and without increased TAR after STTA), the Mann–Whitney U test or Chi–square test was used as appropriate. Statistical analyses were performed using MedCalc v.20.027 software (MedCalc Software, Ostend, Belgium). P-values < 0.05 were considered significant.

Patient characteristics were retrieved retrospectively from our prospective database: Age, body mass index (BMI), use of neoadjuvant treatment and ECOG performance score. Patients of the powered group were propensity score matched 1:1 to patients in the manual group using nearest neighbor matching for the covariates listed above.
Univariate and multivariate linear regression models (continuous variables) and logistic regression models (binary variables) were used to determine differences in the incidence of postoperative leakage, Clavien–Dindo complications, length of stay and 90-day mortality after propensity score matching. A P-value of <0.05 was set as the threshold for statistical significance. Statistical analyses were performed in MEDCALC (v.20.027).

SPSS 22.0 software (IBM, Armonk, NY) and MedCalc v. 20.0 (MedCalc Software, Ostend, Belgium) were used for the statistical analysis. Measurement data were represented as mean ± standard deviation. The t-test for independent samples (normally distributed and homoscedastic data) and the Mann–Whitney U test (skewed or heteroscedastic data) were used to compare each parameter between the pathologic types, WHO grades (G1-2 vs G3), pT stages (early vs late), and pN stages (N0 vs N1-2). Differences with a p < 0.05 were considered statistically significant. The receiver operating characteristic (ROC) curve of each parameter was plotted, the area under the ROC curve (AUC) was calculated; the best diagnostic threshold for each parameter was determined based on the maximum Youden’s index (Youden’s index = sensitivity + specificity—1), and the diagnostic power of T1 and ADC values in identifying MC and AC, and the grade and stage of rectal adenocarcinoma were evaluated.

SPSS 22.0 software (IBM, Armonk, NY) was used for statistical analysis. The Kolmogorov-Smirnov test was performed for analyzing normality. Data conforming to the normal distribution were expressed as mean ± standard deviation (SD). The intraclass correlation coefficient (ICC) was used to evaluate the interobserver consistency of the measured parameters. ICC values of less than 0.40, 0.41–0.75, and greater than 0.75 were considered to indicate poor, fair, and good agreement, respectively. The t-test for independent samples was used to compare APT SI, D, D*, f and ADC parameters between pathological types (MC vs. AC), WHO grades (low- vs. high-grade), pT stages (pT1-2 vs. pT3-4), pN stages (pN1-2 vs. pN0), perineural invasion (positive vs. negative), lymphovascular invasion (positive vs. negative), and EMVI statuses (positive vs. negative). For parameters with significant differences between groups, the receiver operating characteristic (ROC) curve was used to analyze their diagnostic efficacy using the software of MedCalc v. 20.0 (MedCalc Software, Ostend, Belgium). DeLong test was used to compare the differences of area under ROC curves (AUCs). The forward model of binary logistic regression was applied for parameter fusion. Differences with P<0.05 were considered statistically significant.