Cardiovascular disease risk prediction using automated machine learning: A prospective study of 423,604 UK Biobank participants.

Alaa AM, Bolton T, Di Angelantonio E, Rudd JHF, van der Schaar M. · 2019

Read Cardiovascular disease risk prediction using automated machine learning: A prospective study of 423,604 UK Biobank participants. by Alaa AM, Bolton T, Di Angelantonio E, Rudd JHF, van der Schaar M. (2019) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc6519796.txt.

Description

Alaa and co-authors test automated machine learning for heart disease risk prediction. Their prospective study appeared in PLoS ONE. It uses UK Biobank data on 423,604 people free of cardiovascular disease at baseline. Their tool, AutoPrognosis, builds and tunes model pipelines on its own.

The model drew on 473 variables. It reached an AUC of 0.774 for risk prediction. The Framingham score reached 0.724, and a Cox model with all variables reached 0.758. Of 4,801 cases within five years, the new model caught 368 more than Framingham. Novel predictors included usual walking pace and self-rated health. The authors stress that added data beat added model complexity.

Catalog record

CREATORAlaa AM, Bolton T, Di Angelantonio E, Rudd JHF, van der Schaar M.
YEAR2019
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC6519796

Browse the full library · About the Tower of Babel