Transformer-based models for cardiovascular disease risk prediction: what is the potential?

Transformer-based models for cardiovascular disease risk prediction: what is the potential?

Accepted version

Peer-reviewed

Repository URI

https://www.repository.cam.ac.uk/handle/1810/407470

Repository DOI

https://doi.org/10.17863/CAM.133612

Files

Primary Accepted version (137.87 KB)

Type

Authors

Pennells, Lisa

https://orcid.org/0000-0002-8594-3061

Lannelongue, Loïc

https://orcid.org/0000-0002-9135-1345

Kaptoge, Stephen

https://orcid.org/0000-0002-1155-4872

Abstract

Graphical Abstract Summary of key features of transformer models in comparison with traditional statistical models For image description, please refer to the figure legend and surrounding text. Summary of key features of transformer models in comparison with traditional statistical models

Keywords

32 Biomedical and Clinical Sciences, 3201 Cardiovascular Medicine and Haematology, 3202 Clinical Sciences, Cardiovascular, Prevention, 2.5 Research design and methodologies (aetiology)

Journal Title

Eur Heart J

Journal ISSN

0195-668X
1522-9645

Publisher

Oxford University Press (OUP)

Publisher DOI

https://doi.org/10.1093/eurheartj/ehag641

Rights and licensing

Attribution 4.0 International

Except where otherwised noted, this item's license is described as Attribution 4.0 International

Collections

University of Cambridge Research Outputs (Articles and Conferences)