A prediction model can be evaluated by the consequences of using it for a decision. Decision curves examine that trade-off across a range of decision thresholds.
The original decision-curve method combines true-positive and false-positive consequences through a threshold probability reflecting their relative weight. It compares model-based strategies with alternatives across a threshold range and was illustrated with prostate cancer prediction models. Net benefit is a model-based decision measure with assumptions about the decision and its consequences. It does not itself demonstrate that using a model improves outcomes in a clinical trial.
Shares Analytical versus clinical validation, Evidence from measurement to patient benefit.
Shares Analytical versus clinical validation, Evidence from measurement to patient benefit.
Shares Analytical versus clinical validation, Evidence from measurement to patient benefit.
Shares Analytical versus clinical validation, Evidence from measurement to patient benefit.
Shares ROC-AUC, PR-AUC and time-dependent AUC, Calibration: reliability diagrams and the Brier score.
Shares Analytical versus clinical validation, Evidence from measurement to patient benefit.
Shares Calibration: reliability diagrams and the Brier score, Analytical versus clinical validation.
Every source this page cites, grouped by who published it. Grouping is read from the web address, not assigned by hand.
A source that has moved or gone is reported by the weekly link check. /audit/