Deriving novel atrial fibrillation phenotypes using a tree-based artificial intelligence-enhanced electrocardiography approach
Atrial fibrillation (AF) is classically categorised by arrhythmia duration, but these subtypes have limitations in capturing mechanistic and prognostic diversity. A variational autoencoder, trained on >1.1M ECGs, extracted representative features, filtered for an AF cohort of 20,291 unique patients. These features were input into an unsupervised tree-based clustering method to map AF heterogeneity as a tree structure and identify phenogroups. Five phenogroups stratified by future disease risk were identified: (1) higher-risk AF; (2) highest-risk AF with heart failure (HF); (3) average paroxysmal AF; (4) lower-risk paroxysmal AF; and (5) higher-risk paroxysmal AF. The tree trajectory positioned individuals based on shared traits, emphasising explainability. Paroxysmal phenogroups 4 and 5 differed in risk and ventricular structure, with phenogroup 5 exhibiting more adverse features. Mixed AF phenogroup 2 reflected advanced AF with greater HF burden and mortality risk. This AI-ECG framework augments AF subtypes with a risk-based dimension, supporting personalised care.
Authors
- Ibrahim Antoun (ORCID: https://orcid.org/0000-0002-4374-7476)
- G. Andre Ng
- Riyaz Somani (ORCID: https://orcid.org/0000-0001-5162-3644)
- Joseph Barker (ORCID: https://orcid.org/0000-0002-5483-9608)
- Konstantinos Patlatzoglou (ORCID: https://orcid.org/0000-0002-5888-8490)
- Libor Pastika (ORCID: https://orcid.org/0000-0001-6892-6553)
- Daniel Kramer (ORCID: https://orcid.org/0000-0002-1284-4871)
- Arunashis Sau (ORCID: https://orcid.org/0000-0002-0204-7078)
- Declan P. O’Regan (ORCID: https://orcid.org/0000-0002-0691-0270)
- Lara Curran (ORCID: https://orcid.org/0000-0003-1260-3741)
- Fu Siong Ng (ORCID: https://orcid.org/0000-0002-8681-4368)
- Jonathan W. Waks (ORCID: https://orcid.org/0000-0001-5560-5638)
- Mehak Gurnani (ORCID: https://orcid.org/0009-0003-9228-2518)
- Nicholas S. Peters (ORCID: https://orcid.org/0000-0002-3581-8078)
- Boroumand Zeidaabadi (ORCID: https://orcid.org/0009-0004-0025-9896)
- Paolo Inglese (ORCID: https://orcid.org/0000-0001-6179-9643)
Institutions
- Beth Israel Deaconess Medical Center (US)
- Harvard University (US)
- Imperial College Healthcare NHS Trust (GB)
- Chelsea and Westminster Hospital NHS Foundation Trust (GB)
- Glenfield Hospital (GB)
- University Hospitals of Leicester NHS Trust (GB)
- Lung Institute (US)
- NIHR Leicester Cardiovascular Biomedical Research Unit (GB)
- MRC London Institute of Medical Sciences (GB)
- Deaconess Hospital (US)
- NIHR Leicester Biomedical Research Centre (GB)
- Imperial College London (GB)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2025-12-04
- DOI
- https://doi.org/10.1038/s41746-025-02159-z
- Citations
- 1
- Primary Topic
- Atrial Fibrillation Management and Outcomes
- Type
- article
- Field-Weighted Citation Impact
- 1.39
Funders
- British Heart Foundation
- Medical Research Council