Fig. 3: Relationship between the class ratio and evaluation results. | npj Digital Medicine

Fig. 3: Relationship between the class ratio and evaluation results.

From: The Framing of machine learning risk prediction models illustrated by evaluation of sepsis in general wards

Fig. 3: Relationship between the class ratio and evaluation results.

The figure shows the mean values (circle or diamond) and individual cross-validation values (dots) for the best within-class MLRP model for each framing structures. a AUPRC (gray) and AUROC (red) discrimination results; (b) ACE; (c) The stratified Brier score for the positive class; (d) The stratified Brier score for negative class. AUPRC area under the precision recall curve, AUROC area under the receiver operating characteristics curve, ACE average calibration error.

Back to article page