Fig. 6

Interpretation of the RF with a local interpretable model explainer in four cases. Two living patients (cases 1 and 3) and two deceased patients (cases 2 and 4) are illustrated. Features with a blue bar favor the outcome, and those with a red bar contradict the outcome. The x-axis shows how much each feature adds or subtracts from the final probability value for the patient (i.e., a feature with a weight of 0.3 is equivalent to a 30% change in the probability of the outcome). RF random forest, GCS glasgow coma scale, ISS injury severity score, BE base excess, BMI body mass index