Improving arrhythmic risk prediction using cardiac magnetic resonance within deep learning in ischemic heart disease
Sen, Ahmet, Jones, Richard E., Morgan, Holly, Zaidi, Hassan, Halliday, Brian P., Hammersley, Daniel J., Chiribiri, Amedeo, Perera, Divaka, Prasad, Sanjay K., Bishop, Martin J. npj Cardiovascular Health
· Springer Nature
· 2026-07-05
· DOI: 10.1038/s44325-026-00142-5
Аннотация
Sudden arrhythmic death remains a major clinical risk in ischemic heart disease (IHD), underscoring the need for improved risk stratification. Late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) provides measures of scar burden and heterogeneity, but its incremental prognostic value beyond conventional markers such as left ventricular ejection fraction remains uncertain. We analysed two independent IHD cohorts (Dataset 1: n = 399, 54 events; National Research Ethics Service approvals 07/H0708/83 and 09/H0504/104+5; Dataset 2: n = 424, 50 events; derived from the prospectively registered REVIVED-BCIS2 trial, ISRCTN45979711, registered 20 November 2012)using clinical and LGE-CMR-derived variables to evaluate the contribution of LGE-CMR features, and compare machine learning-based survival modelling approaches. A brute-force feature-selection strategy identified optimal predictor subsets for Cox proportional hazards models, Random Survival Forests, and DeepSurv, evaluated using cross-cohort and pooled validation strategies. Scar entropy consistently emerged as a strong predictor of major arrhythmic events. Non-linear approaches outperformed Cox regression, with DeepSurv demonstrating superior generalization across cohorts and Random Survival Forests showing robust performance in pooled analyses. These findings support scar heterogeneity as an important prognostic marker and suggest that machine-learning survival models may improve arrhythmic risk prediction in patients with IHD.
Sen, Ahmet, Jones, Richard E., Morgan, Holly et al. Improving arrhythmic risk prediction using cardiac magnetic resonance within deep learning in ischemic heart disease. // npj Cardiovascular Health. – 2026. – DOI: 10.1038/s44325-026-00142-5. – URL: https://www.nature.com/articles/s44325-026-00142-5
Sen, Ahmet, Jones, Richard E, Morgan, Holly, Zaidi, Hassan, Halliday, Brian P, Hammersley, Daniel J, Chiribiri, Amedeo, Perera, Divaka, Prasad, Sanjay K, Bishop, & Martin J (2026). Improving arrhythmic risk prediction using cardiac magnetic resonance within deep learning in ischemic heart disease. npj Cardiovascular Health. https://doi.org/10.1038/s44325-026-00142-5
Sen, et al. “Improving arrhythmic risk prediction using cardiac magnetic resonance within deep learning in ischemic heart disease.” npj Cardiovascular Health, 2026. doi:10.1038/s44325-026-00142-5.
@article{Sen2026,
title = {Improving arrhythmic risk prediction using cardiac magnetic resonance within deep learning in ischemic heart disease},
author = {Sen and Ahmet and Jones and E., Richard and Morgan and Holly and Zaidi and Hassan and Halliday and P., Brian and Hammersley and J., Daniel and Chiribiri and Amedeo and Perera and Divaka and Prasad and K., Sanjay and Bishop and J., Martin},
journal = {npj Cardiovascular Health},
year = {2026},
doi = {10.1038/s44325-026-00142-5},
url = {https://www.nature.com/articles/s44325-026-00142-5}
}
Fibroblast growth factor 21 (FGF21) is a key regulator in metabolic diseases. Serum FGF21 levels are significantly elevated in patients with hyperuricemia and correlate positively with uric acid. In hyperuricemic mice, FGF21 is...
Artificial intelligence (AI) in cardiology is rapidly expanding, but enthusiasm for anticipated benefits should be balanced by a more critical inspection of ethical concerns. In this state-of-the-art review, we discuss the four pillars...
Acute stroke alerts are frequently activated for non-cerebrovascular conditions, creating false positives that strain resources and increase diagnostic uncertainty. We developed machine learning models integrating large-language models,...
Circadian imbalance may contribute to hypertension, but its association with incident hypertension, genetic susceptibility, and molecular pathways remains unclear. This prospective study included 107,270 UK Biobank participants free of...