Fig. 5From: Machine learning algorithm to predict mortality in patients undergoing continuous renal replacement therapyComparisons of in-hospital mortality prediction models such as random forest, APACHE II, SOFA, and MOSAIC in the test set. a Receiver operating characteristic curves of random forest, APACHE II, SOFA, and MOSAIC. The bar graph indicates the median value of the AUC in the model. The error bar indicates the range. b Decision curve analysis of random forest, APACHE II, SOFA, and MOSAIC for in-hospital mortality prediction. *P < 0.05. APACHE, Acute Physiology and Chronic Health Evaluation; SOFA, Sequential Organ Failure Assessment; MOSAIC, Mortality Scoring system for AKI with CRRTBack to article page