Predicting in-hospital mortality for MIMIC-III patients: A nomogram combined with SOFA score

Figure 1.. The pattern of missing values in the dataset (A) and the LASSO regression method was applied to filter variables and select predictors (B). LASSO = least absolute shrinkage and selection operator.

Figure 2.. Nomogram for predicting in-hospital mortality. To use the nomogram, an individual patient’s value is located on each variable axis, and a line is drawn upward to determine the number of points received for each variable value. The sum of these numbers is located on the Total Points axis, and a line is drawn downward to the risk of in-hospital mortality. BMI = body mass index, SOFA = sequential organ failure assessment.

Figure 3.. The calibration curves indicate the goodness-of-fit of the nomogram and ROC curves identifying the optimal cutoff value for predicting in-hospital mortality. (A) Calibration curve for the training set. (B) Calibration curve for the test set. The dotted line 45 degrees represents the ideal prediction, and the other dotted line represents the predictive performance of the nomogram. The closer the dotted line approaches the ideal prediction line, the better the predictive efficacy of the nomogram is. The solid line represents the internal validation curve based on Bootstrap sampling. (C) ROC curves for the training set. (D) ROC curves for the test set. AUC = the area under curve, ROC curve = receiver operating characteristic curve.

Figure 4.. The ROC curve of SOFA score for the training set and the test set. ROC curve = receiver operating characteristic curve, SOFA = sequential organ failure assessment.