Prediction of incident atrial fibrillation from retinal fundus images using a multimodal foundation model
4 affiliations
- 1Department of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, National Clinical Research Center for Eye Diseases, Shanghai Key Laboratory of Ocular Fundus Diseases, Shanghai Engineering Center for Visual Science and Photomedicine, Shanghai Engineering Center for Precise Diagnosis and Treatment of Eye Diseases, Shanghai, PR China
- 2Institute of Translational Medicine, National Center for Translational Medicine, Shanghai Jiao Tong University, Shanghai, PR China
- 3Department of Endocrinology and Metabolism, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
- 4School of Computer Science, Faculty of Engineering, The University of Sydney, Sydney, NSW Australia
Abstract
Atrial fibrillation (AF), a common cardiac arrhythmia, presents significant challenges for early detection and management due to its asymptomatic and paroxysmal characteristics. In this study, we introduce the RetiAF score, a multimodal foundation-model-based biomarker derived from retinal fundus images for early detection of AF. We identified that the RetiAF score demonstrated a robust performance across multi-ethnic, multi-center datasets, achieving AUROCs of 0.8610 and 0.8019 on the UK Biobank (UKBB) development and internal testing datasets, and an AUROC of 0.7803 on an external dataset acquired in Shanghai, China. In addition, we identified that the RetiAF score consistently outperformed the traditional risk scores such as CHARGE-AF (AUROC: 0.7553) and C2HEST (AUROC: 0.7246) for the UKBB internal testing dataset. Multivariable logistic regression and propensity score analyses further demonstrated that the RetiAF score was independently associated with AF risk (p < 0.001). When stratified by higher C2HEST scores (≥3), the RetiAF score achieved an AUROC of 0.9619, highlighting its potential for identifying high-risk patients before the clinical onset of AF. The multimodal hybrid version of RetiAF (Hybrid_RetiAF) score, which incorporated clinical features (e.g., Age, BMI, etc) into the deep learning model, further enhanced predictive performance and achieved AUROCs of 0.8924 and 0.8381 on the UKBB Cohorts. As a secondary exploratory analysis, we evaluated whether RetiAF-derived scores were associated with chronic ischemic heart disease in UKBB, suggesting shared cardio-retinal risk information. These findings underscore the potentials of non-invasive retinal imaging as a scalable tool for AF and cardiovascular risk assessment, offering a promising alternative for large-scale screenings and personalized interventions.
Related
- Echocardiography-based multimodal models for thromboembolic risk prediction in atrial fibrillation: a narrative review
- Impact of an AI algorithm for multi-day prediction of incident atrial fibrillation on clinical decision-making: PROVISION-AF study
- Retinal Arterial Wall Reflectivity and Incident Cardiovascular Disease
- Predictive Value of P-Wave Dispersion Combined with Left Atrial Reservoir Strain for Incident Atrial Fibrillation in Patients with Hypertrophic Cardiomyopathy: A Retrospective Cohort Study.
- Non-alcoholic Fatty Liver Disease/Metabolic Dysfunction-Associated Steatotic Liver Disease and the Risk of Incident Atrial Fibrillation: An Updated Systematic Review and Meta-Analysis of Cohort Studies
BibTeX
@article{xu2026prediction,
title = {{Prediction of incident atrial fibrillation from retinal fundus images using a multimodal foundation model}},
author = {Yupeng Xu and Yige Peng and Yan Jiang and Jingtong Xie and Meng Xiang and Qian Zhu and Jin Wei and Ruoyan Cui and Xun Xu and Yuhang Ma and Yufan Wang and Jinman Kim and Lei Bi and Kun Liu},
journal = {NPJ Digital Medicine},
year = {2026},
month = jul,
doi = {10.1038/s41746-026-02969-9},
url = {https://doi.org/10.1038/s41746-026-02969-9},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Prediction of incident atrial fibrillation from retinal fundus images using a multimodal foundation model
AU - Yupeng Xu
AU - Yige Peng
AU - Yan Jiang
AU - Jingtong Xie
AU - Meng Xiang
AU - Qian Zhu
AU - Jin Wei
AU - Ruoyan Cui
AU - Xun Xu
AU - Yuhang Ma
AU - Yufan Wang
AU - Jinman Kim
AU - Lei Bi
AU - Kun Liu
PY - 2026
DA - 2026/07/08
JO - NPJ Digital Medicine
DO - 10.1038/s41746-026-02969-9
UR - https://doi.org/10.1038/s41746-026-02969-9
AB - Atrial fibrillation (AF), a common cardiac arrhythmia, presents significant challenges for early detection and management due to its asymptomatic and paroxysmal characteristics. In this study, we introduce the RetiAF score, a multimodal foundation-model-based biomarker derived from retinal fundus images for early detection of AF. We identified that the RetiAF score demonstrated a robust performance across multi-ethnic, multi-center datasets, achieving AUROCs of 0.8610 and 0.8019 on the UK Biobank (UKBB) development and internal testing datasets, and an AUROC of 0.7803 on an external dataset acquired in Shanghai, China. In addition, we identified that the RetiAF score consistently outperformed the traditional risk scores such as CHARGE-AF (AUROC: 0.7553) and C2HEST (AUROC: 0.7246) for the UKBB internal testing dataset. Multivariable logistic regression and propensity score analyses further demonstrated that the RetiAF score was independently associated with AF risk (p < 0.001). When stratified by higher C2HEST scores (≥3), the RetiAF score achieved an AUROC of 0.9619, highlighting its potential for identifying high-risk patients before the clinical onset of AF. The multimodal hybrid version of RetiAF (Hybrid_RetiAF) score, which incorporated clinical features (e.g., Age, BMI, etc) into the deep learning model, further enhanced predictive performance and achieved AUROCs of 0.8924 and 0.8381 on the UKBB Cohorts. As a secondary exploratory analysis, we evaluated whether RetiAF-derived scores were associated with chronic ischemic heart disease in UKBB, suggesting shared cardio-retinal risk information. These findings underscore the potentials of non-invasive retinal imaging as a scalable tool for AF and cardiovascular risk assessment, offering a promising alternative for large-scale screenings and personalized interventions.
SN - 2398-6352
ER -