Development and Validation of a Simplified Martin-Hopkins Low-density Lipoprotein Cholesterol (LDL-C) Equation Using Machine Learning
Performance · Accuracy vs ultracentrifugation-measured LDL-C (test set)Result — go to the resultMedian difference −0.5 mg/dLAuthors' conclusion
Study results suggest that the new machine learning-based LDL-C equation provided results comparable with those of the original Martin-Hopkins method while simplifying implementation to a single equation.
Study context
- Population
- Adult and pediatric patients with complete lipid panel data, Very Large Database of Lipids (Vertical Auto Profile ultracentrifugation, 2015–2019)
- Participants
- 4,939,528 participants
- Comparison
- LDL-C-MH-MARS vs LDL-C-MH
- Outcome
- Accuracy of estimated vs ultracentrifugation-measured LDL-C (bias, RMSE, concordance by guideline-based categories)
Methodological detailscross-sectional, machine learning, randomly assigned training and test sets, external validation
- Cross-sectional
- Machine learning
- Randomly assigned training and test sets
- External validation
Performance
As reported in the abstract
Accuracy vs ultracentrifugation-measured LDL-C (test set)
Median difference
−0.5 mg/dL
By group
LDL-C-MH-MARS
- Median bias (IQR)
- −0.1 (−2.1 to 1.8) mg/dL
- Root mean square error (RMSE)
- 4.7 mg/dL
- Correctly classified
- 89.7%
LDL-C-MH
- RMSE
- 4.9 mg/dL
- Correctly classified
- 89.6%
LDL-C-S
- RMSE
- 5.8 mg/dL
- Correctly classified
- 86.3%
LDL-C-MS
- RMSE
- 6.0 mg/dL
- Correctly classified
- 84.7%
LDL-C-F
- RMSE
- 7.2 mg/dL
- Correctly classified
- 83.1%
Median bias (IQR)
LDL-C-MH-MARS−0.1 (−2.1 to 1.8) mg/dL
LDL-C-MHNot reported in the abstract
LDL-C-SNot reported in the abstract
LDL-C-MSNot reported in the abstract
LDL-C-FNot reported in the abstract
Root mean square error (RMSE)
LDL-C-MH-MARS4.7 mg/dL
LDL-C-MH4.9 mg/dL
LDL-C-S5.8 mg/dL
LDL-C-MS6.0 mg/dL
LDL-C-F7.2 mg/dL
Correctly classified
LDL-C-MH-MARS89.7%
LDL-C-MH89.6%
LDL-C-S86.3%
LDL-C-MS84.7%
LDL-C-F83.1%
Further findings
| Outcome | Values |
|---|---|
| Underestimation when classifying LDL-C <70 mg/dL, TG 200 to 399 mg/dL | LDL-C-MH-MARS 16% · LDL-C-MH 17% · LDL-C-MS 28% · LDL-C-S 39% · LDL-C-F 60% |
External validation (Mayo Clinic; FOURIER)
- Similar patterns; LDL-C-MH-MARS and LDL-C-MH highest accuracy
Original publication
Development and Validation of a Simplified Martin-Hopkins LDL-C Equation Using Machine Learning. JAMA Cardiol. Published 1 September 2026. doi:10.
Summarised from the published abstract; every number is checked against it