Figure legends:
FIGURE 1. Instructions for enrolling in the training cohort and validation cohort cases.
FIGURE 2. A: Lasso regression coefficient path diagram; B: Lasso regression cross-validation curve. Three predictors with non-zero coefficients were obtained by Lasso regression (screening lambda by 10-fold cross-validation, based on lambda.1se, i.e., the maximum lambda corresponding to an error mean within one standard deviation of the minimum): age, CRP natural log-transformed value (InCRP), Monocytes percentage (%Mon).
FIGURE 3. A: Nomogram of the pneumonia risk model; B-G: ROC curves (Bootstrap=500 times), calibration curves, and DCA curves of the pneumonia risk model in the training and validation cohorts. The ROC curves show good discrimination of the pneumonia risk model in both the training and validation cohorts. The calibration curves showed that the pneumonia risk model has good calibration accuracy. The decision curve analysis showed that the pneumonia risk model has high clinical value in predicting the probability of pneumonia in SARS-CoV-2 infected patients.
FIGURE 4. Comparison of the models in the whole study cohort.
A. Receiver operator characteristic curves of the models are presented to compare their discriminatory accuracy for predicting pneumonia risk. P values show the AUC for the pneumonia risk model versus the AUCs for predictors incorporated in the model alone; the predictive ability of the predictors in the model individually and the overall predictive power of the pneumonia risk model are contrasted via IDI.
B. Decision curve analyses comparing the net benefit of the nomogram of the pneumonia risk model versus the other variables incorporated in the nomogram alone are shown. AUC: area under the curve; CI: confidence interval; IDI: Integrated discrimination improvement.
FIGURE 5. Correlation between the predicted probability of pneumonia risk and pneumonia grade(A: Training cohort; B: Validation cohort). A positive linear correlation was found between the predicted pneumonia probability of the pneumonia risk model and pneumonia grade using GAM.