Artificial intelligence in atrial fibrillation: emerging applications, research directions and ethical considerations
Atrial fibrillation (AF) is the most prevalent sustained arrhythmia and a major contributor to stroke and heart failure. Despite progress in management, challenges persist in early detection, risk stratification, and personalised treatment. Artificial intelligence (AI), especially machine learning (...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-06-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2025.1596574/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Atrial fibrillation (AF) is the most prevalent sustained arrhythmia and a major contributor to stroke and heart failure. Despite progress in management, challenges persist in early detection, risk stratification, and personalised treatment. Artificial intelligence (AI), especially machine learning (ML) and deep learning (DL), has emerged as a transformative tool in AF care. This scoping review examines the applications of AI across key domains: detection, risk prediction, treatment optimisation, and remote monitoring. AI-driven models enhance AF detection by analysing ECGs and wearable device data with high accuracy, enabling early identification of asymptomatic cases. By incorporating diverse clinical, imaging, and genomic data, predictive models outperform conventional risk scores in estimating stroke risk and disease progression. In treatment, AI assists in personalised anticoagulation decisions, catheter ablation planning, and optimising antiarrhythmic drug selection. Furthermore, AI-powered remote monitoring integrates wearable-derived insights with real-time decision support, improving patient engagement and adherence. Despite these advances, significant challenges persist, including algorithm transparency, bias, data integration, and regulatory hurdles. Explainable AI (XAI) is crucial to ensure clinician trust and facilitate implementation into clinical workflows. Future research should focus on large-scale validation, multi-modal data integration, and real-world AI deployment in AF management. AI has the potential to revolutionise AF care, shifting from reactive treatment to proactive, personalised management. Addressing current limitations through interdisciplinary collaboration will be key to realising AI's full potential in clinical practice and improving patient outcomes. |
|---|---|
| ISSN: | 2297-055X |