Key result
Deep learning on 12-lead ECGs identifies genotype-positive LQTS with ~0.90 AUC.
Why the study?
Congenital LQTS is rare and disease carriers may lack obvious ECG abnormalities, leading to underdiagnosis by general cardiologists.
Does a deep learning model trained on 12-lead ECGs improve the identification of congenital long QT syndrome compared to conventional QTc measurement and expert evaluation?
Observational (n=12,770)
Yes
Does a deep learning model trained on 12-lead ECGs improve the identification of congenital long QT syndrome compared to conventional QTc measurement and expert evaluation?
Deep learning models applied to 12-lead ECGs can accurately identify congenital long QT syndrome, outperforming conventional QTc measurements and identifying the onset of the QRS complex as a novel informative feature.
No takes yet. Share an insight, caveat, or question.
May aid ECG-based LQTS detection; leaves open prospective validation before clinical adoption.
Aufiero et al. (2022) conducted an observational in Congenital long QT syndrome (LQTS) (n=12,770). Deep learning (1DCNN) model vs. Conventional QTc measurement and expert cardiologist evaluation was evaluated on Detection of genotype-positive LQTS patients (AUC). A deep learning model trained on 12-lead ECGs identified genotype-positive long QT syndrome patients with an average AUC of 0.90, 0.92, and 0.89 for LQTS 1, 2, and 3, respectively.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: