# ROC-AUC, PR-AUC and time-dependent AUC

Source: https://onco.cc/terms/roc-auc/  
OnCo record `roc-auc` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

ROC-AUC is the chance a classifier scores a random positive above a random negative; PR-AUC focuses on the positives and is the better summary when they are rare; time-dependent AUC applies the idea to survival at a chosen horizon.

## Summary

A receiver operating characteristic curve plots a binary classifier's performance across thresholds and is standard in assessing diagnostic tests (Wikipedia); the area under it summarises ranking quality. Precision and recall measure the retrieved positives (Wikipedia), and the area under the precision-recall curve is sensitive to class imbalance where ROC-AUC is not, which matters when responders or rare subtypes are one in ten. Time-dependent AUC asks the same ranking question about who has had the event by a given time, complementing the C-index.

## Fields

- Kind: Term
- Last checked: 2026-09-24
- Also known as: ROC-AUC; AUROC; ROC curve; area under the ROC curve; PR-AUC; AUPRC; precision-recall curve; average precision; time-dependent AUC; AUC(t)
- Tags: cansim-terms

## Notes

- Listed in the CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme; CanSim page path /terms/roc-auc-pr-auc.

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Receiver_operating_characteristic
- Wikipedia: precision and recall: https://en.wikipedia.org/wiki/Precision_and_recall
- Wikipedia: https://en.wikipedia.org/wiki/Receiver_operating_characteristic

## Connected records

- terms: [Accuracy, macro-F1 and confusion matrices](https://onco.cc/terms/accuracy-f1/), [Biomarker](https://onco.cc/terms/biomarker/), [C-index (concordance index), Harrell's and Uno's](https://onco.cc/terms/concordance-index/), [Calibration: reliability diagrams and the Brier score](https://onco.cc/terms/calibration/), [Cancer AI vocabulary (CanSim terms map)](https://onco.cc/terms/cancer-ai-vocabulary/)

---
JSON: https://onco.cc/api/v1/entities/roc-auc.json