Spss statistics program version 19

Statistical analyses were performed using the SPSS statistics program version 19 (IBM Corporation: New York, NY, USA). The chi-squared test was applied to compare nonparametric frequency of categorical variables between the two groups. A two-sided p value of < 0.05 indicated statistical significance. Receiver operating characteristic (ROC) curve was used to evaluate the performance of machine learning models and precision–recall curve (PRC) was suitable for evaluating the performance of models when the data distribution was imbalanced. The efficacy of our prediction model was presented as sensitivity, specificity, F1-score, accuracy as follows: Sensitivity = [True Positives (CRKP)]/[True Positives (CRKP) + False Negatives]; Specificity = [True Negatives (CSKP)]/[True Negatives (CSKP) + False Positives]; Accuracy = [True Positives (CRKP) + True negative (CSKP)]/[True Positives (CRKP) + True Negatives (CSKP) + False Positives + False Negatives].