Figure 9: A model using both preoperative and intraoperative data outperformed a model using preoperative data alone in predicting postoperative hospital mortality.

A: The postoperative model had greater area under the receiver operating characteristic curve (0.93 vs. 0.87). B: The postoperative model had greater area under the precision-recall curve (0.21 vs. 0.15). The postoperative model reclassified cases that did (C) and did not (D) feature in-hospital mortality. Red dots are patients at high-risk for mortality according to the postoperative model; green dots are patients at low-risk. C, D: The postoperative model correctly reclassified 11.2% of all cases. Gray areas represent regions for which predictive discrimination or precision are ≤0.2, precluding reasonable clinical application.