Key result
Time-aware AI models using temporal ECG sequences predict sudden cardiac arrest with a 0.78 C-index.
Why the study?
Predicting sudden cardiac arrest from the ECG is complicated by dynamic ECG remodeling over time, which traditional machine learning models trained on ECG snapshots fail to capture.
Does an AI-transformer model trained on temporal ECG feature sequences predict sudden cardiac arrest in at-risk patients?
Cohort (n=2,837)
Does an AI-transformer model trained on temporal ECG feature sequences predict sudden cardiac arrest in at-risk patients?
Effect estimate: C-index 0.78
Time-aware AI-transformer models using repeated 12-lead ECGs can dynamically predict sudden cardiac arrest up to one year prior to the event, with RR interval being the strongest long-term predictor.
May enable dynamic sudden cardiac arrest risk stratification from serial ECGs; leaves open prospective validation before clinical adoption.
Background: Prediction of sudden cardiac arrest (SCA) from the ECG is of great importance, but is complicated by the uncertainty introduced by dynamic remodeling of the ECG over time. Traditional machine learning models trained on ECG snapshots have not captured this dynamic remodeling. Hypothesis: Artificial intelligence (AI)- models trained on temporal ECG feature sequences will reveal dynamic biomarkers of cardiac arrest. . Methods: Our registry consisted of 108,704 12-lead ECG recordings from 2,837 patients (age=61 ± 14 years, 34% female, 46% white) who experienced cardiac arrest within 1 year. Routine ECG features (e.g. P-duration, QRS-duration, QT-interval) were indexed by “time to arrest”. Training, test and validation cohorts were created using patient-stratified splitting, and training data was augmented to make the transformer model robust against noisy timestamps and data sparsity. We calculated the importance of features dynamically over time using Harrell’s Concordance Index (C-Index) and global SHAP values at multiple prediction horizons (30-360 days). (Fig 1) Results: In the training cohort, SCA prediction performance peaked at 60 days before arrest (C-index = 0.89) and tapered monotonically thereafter. In the hold-out test cohort, SCA prediction peaked at 120 days (C-index = 0.78) and fell only slightly to 0.76 at 365 days (Fig 2). Features most important for long-term prediction were the RR-interval (importance ~50%), with an emergence of QT-interval, P-duration and PR-interval as additional co-predictors at timepoints closer to the SCA event (cumulative importance ~ 32%) (Fig 3). Conclusion: ECG-features’ vary in their importance to predicting SCA over a one-year period. Using time-aware AI-transformer models, RR interval was the strongest predictor from 1 year preceding the event, with other features emerging as the event draws closer. This study shows the importance of using repeated 12-lead ECGs in at-risk patients, and combining them into dynamic-predictive indices.
No takes yet. Share an insight, caveat, or question.
Bandyopadhyay et al. (2025) conducted a cohort in Sudden cardiac arrest (n=2,837). Temporal ECG feature sequences analyzed by AI-transformer models was evaluated on Sudden cardiac arrest prediction performance (C-index) (C-index 0.78). Time-aware AI-transformer models using temporal ECG feature sequences predicted sudden cardiac arrest with a peak C-index of 0.78 at 120 days before the event in the hold-out test cohort.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: