Arrhythmias & electrophysiologyCohort study
Prediction of New-Onset Atrial Fibrillation in Patients with Hypertrophic Cardiomyopathy Using Machine Learning
- Prospective cohort study
- multicentre
- prediction model development and external validation
Authors' conclusion
Our ML model showed superior performance compared to the conventional HCM-AF score for the prediction of new-onset AF in patients with HCM.
Study context
- Population
- Patients with hypertrophic cardiomyopathy (HCM) without prior atrial fibrillation (AF)128 (12%) developed new-onset AF
- Participants
- 1,069 patients
- Exposure
- Machine learning (ML) model vs HCM-AF score for predicting new-onset AF
- Outcome
- Prediction of new-onset AF (AUC)
- Follow-up
- Median 2.1 years
Main finding
As reported in the abstract
New-onset AF prediction (test set)
| Group | AUC | Sensitivity | Specificity |
|---|---|---|---|
| ML model | 0.84 (95% CI 0.77-0.91) | 0.82 (95% CI 0.65-0.93) | 0.76 (95% CI 0.71-0.81) |
| HCM-AF score | 0.64 (95% CI 0.54-0.73) | 0.67 (95% CI 0.52-0.88) | 0.57 (95% CI 0.41-0.70) |
- ML model vs HCM-AF score: P < 0.001 — DeLong's test
Observational data: associations, not effects.
Original publication
Prediction of New-Onset Atrial Fibrillation in Patients with Hypertrophic Cardiomyopathy Using Machine Learning. Eur J Heart Fail. Published 18 December 2024. doi:10.1002/ejhf.3546
Summarised from the published abstract; every number is checked against it