Receiver operating characteristic (ROC) curve

Case contributed by Matt A. Morgan

This graph demonstrates a theoretically perfect test (green "curve") with an area under the curve of 1.0. The grey line demonstrates a worthless test, with a 50/50 chance of being accurate in detection.

This graph compares different modalities, tests, or readers given a certain disease and patient population. The green curve represents a "better" test with a lower false positive rate at a given level of sensitivity (or a higher sensitivity at a given false positive rate). The green curve has a larger area under the curve (AUR). The grey line represents a worthless test that is no better than chance.

Derivations of sensitivity and specificity

How to use cases

You can use Radiopaedia cases in a variety of ways to help you learn and teach.

Creating your own cases is easy.