Artificial intelligence-based phenotyping of atrial fibrillation through generative topographic mapping: a prospective Murcia AF Project III cohort study

Soler-Espejo, E, Chen, Y, Ramos-Bratos, MP, Ortega-Martorell, S ORCID: 0000-0001-9927-3209, Olier, I ORCID: 0000-0002-5679-7501, Rivera-Caravaca, JM, Marin, F, Roldan, V and Lip, GYH ORCID: 0000-0002-7566-1626 Artificial intelligence-based phenotyping of atrial fibrillation through generative topographic mapping: a prospective Murcia AF Project III cohort study. Journal of Medical Internet Research. ISSN 1439-4456 (Accepted)

Text Soler-Espejo et al. GTM AF paper - JMIR.pdf - Accepted Version Access Restricted Available under License Creative Commons Attribution. Download (1MB)

Abstract

Background: The clinical heterogeneity of atrial fibrillation (AF) challenges current classifications and risk scores, limiting real-world applicability. Artificial intelligence (AI)-driven methods may enhance phenotyping and risk stratification.

Objective: To apply a Generative Topographic Mapping (GTM)-based clustering approach to a large, real-world prospective AF cohort in order to identify clinically relevant phenotypes and assess their associations with clinical outcomes.

Methods: We conducted a prospective observational cohort study including consecutive adult outpatients with newly diagnosed AF who initiated oral anticoagulation between January 2016 and November 2021. Cardiometabolic risk was modelled as a multidimensional construct integrating cardiovascular and metabolic comorbidities. GTM was applied to project high-dimensional clinical data into a low-dimensional latent space, enabling probabilistic patient representation and visualisation of phenotypic structure. Unsupervised hierarchical clustering using Ward’s method was performed on the latent space to identify AF phenotypes, and patients were assigned to clusters based on their highest posterior probability. Clinical outcomes, including thromboembolic events, major bleeding, major adverse cardiovascular events (MACE), cardiovascular death, and all-cause death, were assessed over a maximum follow-up of 2 years. Non-fatal outcomes were analysed using Fine–Gray competing-risk models and reported as sub-distribution hazard ratios (sHRs), whereas cardiovascular and all-cause death were analysed using adjusted Cox proportional hazards models and reported as adjusted hazard ratios (aHRs).

Results: Among 3,259 AF patients (median age 77 [IQR 70-83] years; 52.8% female; mean follow-up 1.81 years), four phenotypes emerged: (1) older with highest cardiometabolic burden; (2) older with lower cardiometabolic risk; (3) comparatively younger with intermediate-high cardiometabolic risk; and (4) comparatively younger with intermediate cardiometabolic risk. Compared to Phenotype 1, Phenotype 4 demonstrated lower risks of thromboembolic events (sHR 0.70; 95% confidence interval [CI] 0.50–0.98), major bleeding (sHR 0.61; 95% CI 0.42–0.89), and MACE (sHR 0.67; 95% CI 0.47–0.94). Regarding cardiovascular and all-cause death, all phenotypes demonstrated a lower risk compared to Phenotype 1: Phenotype 2 (aHR 0.52; 95% CI 0.33–0.85; and aHR 0.66; 95% CI 0.50–0.87, respectively), Phenotype 3 (aHR 0.57; 95% CI 0.32–0.99; and aHR 0.44; 95% CI 0.30–0.65, respectively), and Phenotype 4 (aHR 0.49; 95% CI 0.33–0.72; and aHR 0.56; 95% CI 0.44–0.71, respectively). However, these associations were attenuated after further adjustment for age, sex, and major comorbidities, with Phenotype 4 retaining a significant association with lower cardiovascular death and Phenotypes 3 and 4 retaining a significant association with lower all-cause death.

Conclusion: GTM-based clustering analysis identified prognostically distinct AF phenotypes, highlighting the potential of machine learning approaches to support personalised AF care. Further external validation is needed to establish the generalisability of these findings.

Item Type:	Article
Uncontrolled Keywords:	Artificial Intelligence and Digital Technologies Research Institute (AIDT); 08 Information and Computing Sciences; 11 Medical and Health Sciences; 17 Psychology and Cognitive Sciences; Medical Informatics; 4203 Health services and systems
Subjects:	Q Science > QA Mathematics > QA76 Computer software R Medicine > RD Surgery
Divisions:	Computer Science and Mathematics
Publisher:	JMIR Publications
Date of acceptance:	19 June 2026
Date Deposited:	24 Jun 2026 09:44
Last Modified:	24 Jun 2026 09:44
URI:	https://researchonline.ljmu.ac.uk/id/eprint/28889

View Item