False positive
When an AI tool flags something as a problem that turns out not to be one. A high false positive rate means the tool cries wolf often, which can wear down the clinicians acting on its alerts.
A deterioration alert goes live on a 30-bed ward and pages the charge nurse about ten times a shift, and on review eight of those ten patients were stable and needed nothing. Within a few weeks staff acknowledge alerts from the corridor without going to the bedside, which means the two real ones get the same quick dismissal. Before go-live, ask for the expected number of false alerts per nurse per shift at the chosen threshold, and during the pilot track how often alerts are acknowledged without a bedside assessment.
Terms like this come up in real clinical scenarios across the HelloAI courses: bite-sized modules with verifiable certificates. An account takes one minute, no password needed.
Sign in →