Loading
The C-index is the fraction of patient pairs in which the model ranked the patient who had the event sooner as higher risk; 0.5 is a coin toss and 1.0 is perfect ranking.
Harrell and colleagues introduced the concordance index as the probability that, for a random pair of patients, the one predicted to be at higher risk has the earlier event; Uno and colleagues gave a version that weights pairs by the inverse probability of censoring so that heavy censoring does not bias it upwards. A pooled pan-cancer C-index is inflated by tissue of origin (leukaemia patients die sooner than thyroid patients), so within-type or stratified C-indices are the honest measure. The C-index ranks; it says nothing about calibration.
Shares ROC-AUC, PR-AUC and time-dependent AUC, Tissue-of-origin signal in tumour data, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Censoring and events in survival data, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Univariate Cox scores and time-dependent metrics, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Tissue-of-origin signal in tumour data, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Calibration: reliability diagrams and the Brier score, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Tissue-of-origin signal in tumour data, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Tissue-of-origin signal in tumour data, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.
Shares Calibration: reliability diagrams and the Brier score, Cancer AI vocabulary (CanSim terms map) and the tag cansim-terms.