Arrhythmias & electrophysiology
Machine learning for prediction of ventricular arrhythmia episodes from intracardiac electrograms of automatic implantable cardioverter-defibrillators
Authors' conclusion
In this study, applying ML to ICD-acquired IEGMs predicted impending ventricular tachycardia or ventricular fibrillation events seconds before they occurred, whereas midterm to long-term predictions were not successful.
Study context
- Population
- Patients with Biotronik implantable cardioverter-defibrillator (ICDs) in the CERTITUDE registry, 2010–2020
- Participants
- 13,516 patients · 301,647 IEGM recordings
- Exposure
- Short-range vs Mid-range vs Long-range
- Outcome
- Prediction of impending VA (sustained VT or VF) episodes by ML-processed IEGMs
Methodological detailsregistry-based, machine learning (convolutional neural network, ResNet architecture)
- Registry-based
- Machine learning (convolutional neural network, ResNet architecture)
Main finding
As reported in the abstract
Prediction of sustained VT/VF
| Group | AUC |
|---|---|
| Short-range | 0.83 (95% CI 0.79-0.87) |
Short-range
AUC0.83 (95% CI 0.79-0.87)
Observational data: associations, not effects.
Further findings
Sustained VT or VF episodes
- 27,845 episodes in 4467 patients
- 33.0%
Original publication
Machine learning for prediction of ventricular arrhythmia episodes from intracardiac electrograms of automatic implantable cardioverter-defibrillators. Heart Rhythm. Published 24 May 2024. doi:10.
Summarised from the published abstract; every number is checked against it