Extended Data Table 1 Model performance for predicting AKI within the full range of possible prediction windows from 6 to 72 h

From: A clinically applicable approach to continuous prediction of future acute kidney injury

  1. a, b, With shorter time windows, closer to the actual onset of AKI, the model achieves a higher area under the receiver operating characteristic curve (ROC AUC; a) but a lower area under the precision–recall curve (PR AUC; b). This stems from different numbers of positive steps within windows of different length. These differences affect both the model precision and the false-positive rate. When making predictions across shorter time windows there is more uncertainty in the exact time of the AKI onset owing to minor physiological fluctuations, and this results in a lower precision being needed to achieve high sensitivity. Bootstrap pivotal 95% confidence intervals are calculated using n = 200 bootstrap samples.