Complete heart block (CHB) is a life-threatening arrhythmia, yet current electrocardiography (ECG)-based methods for risk stratification have limited predictive power. A new study has detailed the development and validation of a novel artificial intelligence-enhanced ECG (AI-ECG) model that can accurately predict the risk of incident CHB, significantly outperforming traditional risk markers.¹

The AI-ECG Risk Estimator for CHB (AIRE-CHB) is a deep learning model based on a residual convolutional neural network. It was trained to analyse standard 12-lead ECGs to identify subtle, subclinical patterns indicative of future CHB risk that are not captured by conventional interpretation methods.

This prognostic cohort study involved developing the AIRE-CHB model using data from the Beth Israel Deaconess Medical Center (BIDMC) in the US and externally validating it in the UK Biobank (UKB) volunteer cohort. The development cohort included 1,163,401 ECGs from 189,539 patients. The primary outcome was a new diagnosis of CHB occurring more than 31 days after the index ECG.

In the development cohort, the AIRE-CHB model predicted incident CHB with a C index of 0.836. The area under the receiver operating characteristics curve (AUROC) for predicting CHB within one year was 0.889 (95% CI, 0.863–0.916). This performance was substantially better than the traditional guideline-based marker, the presence of bifascicular block, which had an AUROC of 0.594 (95% CI, 0.567–0.620).

Patients in the highest-risk quartile, as identified by the model, had an adjusted hazard ratio (aHR) of 11.6 (95% CI, 7.62–17.7; p<0.001) for developing CHB compared with those in the lowest-risk quartile.

The model’s high performance was confirmed in the external validation cohort of 50,641 ECGs from the UKB, where it achieved a C index of 0.936 (95% CI, 0.900–0.972). The aHR for the high-risk versus low-risk group in this cohort was 7.17 (95% CI, 1.67–30.81; p<0.001).

This study presents a novel deep learning model capable of identifying individuals at high risk of developing CHB from a standard ECG. The authors concluded that the model, “AIRE-CHB could be used in diverse settings to aid in decision-making for individuals with syncope or at risk of high-grade atrioventricular block.”¹ This tool could potentially refine clinical pathways by identifying patients who may benefit from enhanced rhythm monitoring or consideration for empirical pacemaker implantation.

References

1. Sau A, Zhang H, Barker J, et al. Artificial Intelligence–Enhanced Electrocardiography for Complete Heart Block Risk Stratification. JAMA Cardiol. Published online August 20, 2025. https://doi.org/10.1001/jamacardio.2025.2522

Disclaimer: The information presented in this article is for educational purposes only. Any quotes included reflect the opinions of the individual quoted, and do not necessarily reflect the views of the publisher. The publisher does not guarantee the accuracy or completeness of the content and accepts no responsibility for any errors, or any consequences arising from its use.