Key result
The Weill Cornell machine learning model demonstrated a higher correlation with directly measured LDL-C compared to the Friedewald formula and Martin-Hopkins equation.
Why the study?
Contemporary equations for LDL-C estimation have limited accuracy in certain scenarios, such as high triglycerides and very low LDL-C.
Does a machine learning model improve the accuracy of LDL-C estimation compared to the Friedewald formula and Martin-Hopkins equation in patients undergoing lipid profiling?
Cross-Sectional (n=10,936)
No
Does a machine learning model improve the accuracy of LDL-C estimation compared to the Friedewald formula and Martin-Hopkins equation in patients undergoing lipid profiling?
A novel machine learning model provides more accurate estimation of LDL-C than the Friedewald and Martin-Hopkins equations, particularly in patients with very low LDL-C or high triglycerides.
Answers are grounded in this paper and cite their sources.
May support ML LDL-C estimation in lipid profiling; hypothesis-generating and requires prospective validation before practice change.
BACKGROUND: Low-density lipoprotein cholesterol (LDL-C) is a target for cardiovascular prevention. Contemporary equations for LDL-C estimation have limited accuracy in certain scenarios (high triglycerides [TG], very low LDL-C). OBJECTIVES: We derived a novel method for LDL-C estimation from the standard lipid profile using a machine learning (ML) approach utilizing random forests (the Weill Cornell model). We compared its correlation to direct LDL-C with the Friedewald and Martin-Hopkins equations for LDL-C estimation. METHODS: The study cohort comprised a convenience sample of standard lipid profile measurements (with the directly measured components of total cholesterol [TC], high-density lipoprotein cholesterol [HDL-C], and TG) as well as chemical-based direct LDL-C performed on the same day at the New York-Presbyterian Hospital/Weill Cornell Medicine (NYP-WCM). Subsequently, an ML algorithm was used to construct a model for LDL-C estimation. Results are reported on the held-out test set, with correlation coefficients and absolute residuals used to assess model performance. RESULTS: Between 2005 and 2019, there were 17,500 lipid profiles performed on 10,936 unique individuals (4,456 females; 40.8%) aged 1 to 103. Correlation coefficients between estimated and measured LDL-C values were 0.982 for the Weill Cornell model, compared to 0.950 for Friedewald and 0.962 for the Martin-Hopkins method. The Weill Cornell model was consistently better across subgroups stratified by LDL-C and TG values, including TG >500 and LDL-C <70. CONCLUSIONS: An ML model was found to have a better correlation with direct LDL-C than either the Friedewald formula or Martin-Hopkins equation, including in the setting of elevated TG and very low LDL-C.
No takes yet. Share an insight, caveat, or question.
Singh et al. (2020) conducted a cross-sectional in Hyperlipidemia (n=10,936). Weill Cornell machine learning model for LDL-C estimation vs. Friedewald formula and Martin-Hopkins equation was evaluated on Correlation coefficient between estimated and directly measured LDL-C (p=<0.001). The Weill Cornell machine learning model demonstrated a higher correlation with directly measured LDL-C compared to the Friedewald formula and Martin-Hopkins equation.
Synapse has enriched 10 related papers. Consider them for comparative context: