Fig. 1

Relative importance of included variables. Tree depth is average depth at which variable is first used in a tree split, assuming that the most discriminatory variables will split the dataset earlier in trees at lower depths. Variable importance (VIMP) is a measure of how changes in the variable impact prediction error. The bigger the VIMP, the more impact the variable has on prediction. Variables having smaller depth are more discriminatory, and those with bigger importance have a greater impact on prediction. BMI body mass index, BUN blood urea nitrogen, DBP diastolic blood pressure, FiO2 fraction of inspired oxygen, Hb hemoglobin, HCT hematocrit, PaCO2 partial pressure of carbon dioxide, PaO2 partial pressure of oxygen, PLT platelet count, POC glucose point of care, RASS Richmond Agitation-Sedation Scale, SBP systolic blood pressure, SpO2 peripheral capillary oxygen saturation, WBC white blood cell count