Related Experiment Video
Updated: Aug 11, 2026
08:10
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
Dynamic Risk Assessment for the Development of Persistent Atrial Fibrillation Using Statistical and Machine Learning
Alexei Nakonechnyi1, Shaul Geliaks1, Ilan Goldenberg1
1University of Rochester Medical Center, Rochester, New York, USA.
Summary
Cardiac implantable electronic devices (CIEDs) can predict persistent atrial fibrillation (AF) progression. Machine learning models analyzing AF burden from CIEDs enable accurate risk stratification for timely clinical management.
Area of Science:
- Cardiology
- Biomedical Engineering
- Data Science
Background:
- Cardiac implantable electronic devices (CIEDs) detect atrial fibrillation (AF), but predicting progression to persistent AF is challenging.
- Longitudinal CIED data offers potential for dynamic risk stratification using statistical and machine learning (ML) methods.
Purpose of the Study:
- To develop a risk stratification model for predicting persistent AF progression.
- Utilize clinical data and CIED-derived AF burden over a 6-month window.
Main Methods:
- Analysis of continuous CIED data from 1985 patients without prior persistent AF.
- Summarized AF burden and clinical variables using overlapping 6-month rolling windows.
- Employed a gradient-boosted decision tree model (XGBoost) for prediction.
Main Results:
- Higher AF burden strongly correlated with persistent AF progression (max HR 8.66, p < 0.001).
- Patients with >8 hours/day AF burden had a 63% progression rate.
- The ML model achieved high predictive performance (sensitivity 99.4%, specificity 95.7%).
Conclusions:
- CIED-detected AF burden is a significant predictor of progression to persistent AF.
- ML analysis of 6-month CIED data provides accurate, real-time risk stratification.
- This approach supports earlier and more targeted clinical management of AF.
