Pulse.

a daily field guide to health research that matters

◆ Console

‹ Thu · 29 Oct 2026
Promising but preliminary

Harnessing Clinical Characteristics to Identify High-Risk of Drug-induced QTc Prolongation in Hospitalized Patients

Machine learning identified hospitalized patients at high risk for dangerous heart-rhythm changes with reasonable accuracy.

This retrospective machine-learning development study with external validation examined Harnessing Clinical Characteristics to Identify High-Risk of Drug-induced QTc Prolongation in Hospitalized Patients. The best model reached an external-validation AUC of 0.864, although recall fell to 0.650.

What the study was

Study design
Retrospective machine-learning development study with external validation
Population
Hospitalized patients with paired electrocardiograms
Sample size
2,451 development participants and 150 external-validation participants
Category
Diagnostics
Maturity
Validated
Journal
Clinical medicine (London, England)

Why it surfaced

Retrospective machine-learning development study with external validation; The best model reached an external-validation AUC of 0.864, although recall fell to 0.650. Score kept conservative because only the PubMed abstract and metadata were reviewed.

A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.