Related Experiment Video
Updated: Apr 29, 2026
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
Stroke prediction in elderly patients with atrial fibrillation using machine learning combined clinical and left
Hao Huang1, Yan Xiong1, Yuan Yao2
1Department of Cardiology, Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu, China.
Insights
Predicting stroke risk in elderly patients with atrial fibrillation (AF) is crucial. A Support Vector Machine (SVM) model using clinical and cardiac imaging data shows high accuracy for identifying high-risk individuals.
Area of Science:
- Cardiology
- Medical Imaging
- Data Science
Background:
- Atrial fibrillation (AF) is a major cause of ischemic stroke.
- Accurate stroke risk prediction in elderly AF patients is essential.
- Identifying distinct risk phenotypes is key for effective management.
Purpose of the Study:
- To identify clinical and LAA radiomic phenotypes in elderly AF patients.
- To analyze the correlation between these phenotypes and stroke risk.
- To develop and evaluate machine learning models for stroke risk prediction.
Main Methods:
- Prospective cohort study of 419 elderly AF patients (≥65 years).
- K-means clustering for phenotype identification (clinical and LAA radiomics).
- Machine learning models (LR, NB, SVM, RF, XGBoost) developed and assessed using AUROC, Hosmer-Lemeshow, and DCA.
Main Results:
- Three clinical phenotypes identified: cardiac enlargement/dysfunction, normal, and metabolic/coagulation abnormalities.
- Elevated stroke risk observed in cardiac enlargement/dysfunction and metabolic/coagulation groups.
- SVM model demonstrated superior predictive performance with an AUROC of 0.858.
Conclusions:
- A stroke-risk prediction model for elderly AF patients using SVM shows strong efficacy.
- The model effectively integrates clinical and LAA radiomic data for risk stratification.
- This approach can aid in personalized stroke prevention strategies for AF patients.
Background:
Atrial fibrillation (AF) is one of the primary etiologies for ischemic stroke, and it is of paramount importance to delineate the risk phenotypes among elderly AF patients and to investigate more efficacious models for predicting stroke risk.
Methods:
This single-center prospective cohort study collected clinical data and cardiac computed tomography angiography (CTA) images from elderly AF patients. The clinical phenotypes and left atrial appendage (LAA) radiomic phenotypes of elderly AF patients were identified through K-means clustering. The independent correlations between these phenotypes and stroke risk were subsequently analyzed. Machine learning algorithms-Logistic Regression, Naive Bayes, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting-were selected to develop a predictive model for stroke risk in this patient cohort. The model was assessed using the Area Under the Receiver Operating Characteristic Curve, Hosmer-Lemeshow tests, and Decision Curve Analysis.
Results:
A total of 419 elderly AF patients (≥ 65 years old) were included. K-means clustering identified three clinical phenotypes: Group A (cardiac enlargement/dysfunction), Group B (normal phenotype), and Group C (metabolic/coagulation abnormalities). Stroke incidence was highest in Group A (19.3%) and Group C (14.5%) versus Group B (3.3%). Similarly, LAA radiomic phenotypes revealed elevated stroke risk in patients with enlarged LAA structure (Group B: 20.0%) and complex LAA morphology (Group C: 14.0%) compared to normal LAA (Group A: 2.9%). Among the five machine learning models, the SVM model achieved superior prediction performance (AUROC: 0.858 [95% CI: 0.830-0.887]).
Conclusion:
The stroke-risk prediction model for elderly AF patients constructed based on the SVM algorithm has strong predictive efficacy.
