Key result
Among 22,325 potential VT alarms identified by a search algorithm in 858 ICU patients, expert adjudication classified 53.62% as true VT, 29.05% as false, and 17.33% as unresolved.
Why the study?
False VT alarms are common during in-hospital ECG monitoring, with prior research attributing the majority to algorithm deficiencies.
Observational (n=5,320)
The creation of a large, human-annotated VT database highlights that nearly 30% of ICU VT alarms are false and provides a gold standard dataset for developing and testing new automated VT algorithms.
No takes yet. Share an insight, caveat, or question.
High false VT alarm burden may fuel ICU alarm fatigue; leaves open rigorous validation of new detection algorithms on this annotated dataset.
Pelter et al. (2023) conducted an observational in Ventricular tachycardia alarms (n=5,320). New VT algorithm and expert annotation was evaluated on Adjudication of potential VT alarms as true, false, or unresolved. Among 22,325 potential VT alarms identified by a search algorithm in 858 ICU patients, expert adjudication classified 53.62% as true VT, 29.05% as false, and 17.33% as unresolved.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: