[Retracted] Establishment of a Risk Prediction Model for Pulmonary Infection in Patients with Advanced Cancer

Figure 3

The calibration curve of nomogram for predicting developing pulmonary infection in the training ((a) ) and testing ((b) ) dataset. The calibration focused on the accuracy of the absolute risk prediction of the model (i.e., the consistency between the probability of serious postoperative severity predicted by the model and that actually observed). The -axis represents the actual rate of developing pulmonary infection. The -axis represents the predicted probability of developing pulmonary infection. For a nomogram with better calibration, the scatter points should be arranged along a 45° diagonal line. The Hosmer-Lemeshow goodness-of-fit test is often used to compare whether significant differences exist between the prediction probability and the actual occurrence, with indicating no statistically significant difference, and the calibration of the model was good.

(a)

(b)