Development and Validation of a Simplified Martin-Hopkins LDL-C Equation Using Machine Learning

Can a simplified formula-based alternative to the Martin-Hopkins low-density lipoprotein cholesterol (LDL-C) equation be developed using machine learning that is sufficiently accurate for clinical care?In this cross-sectional study, almost 5 million patients were randomly assigned to a training set and a test set, and a new equation was developed using multivariate adaptive regression splines. The new equation performed similarly to the original Martin-Hopkins equation, while outperforming the Friedewald, Sampson–National Institutes of Health (NIH), and Modified Sampson-NIH equations; accuracy was validated against preparative ultracentrifugation in 2 external datasets.Results suggest that the new equation provided LDL-C accuracy comparable with that of the original Martin-Hopkins equation and may streamline implementation.

Corresponding Author: Seth S. Martin, MD, MHS, Johns Hopkins Hospital, 600 N Wolfe St, Blalock 536, Baltimore, MD 21287 (smart100@jhmi.edu).

Link to the article in your story

We encourage you to link out to this article in your your story using the link below. It includes an access token that will give free access to the article for your readers up to one year after publication. (The link will be live after the article publishes and embargo is lifted.)

Please see the article for additional information, including full author list, author contributions and affiliations, conflict of interest and financial disclosures, and funding and support.

Need more information? Contact us.