Search

Objective Low-density lipoprotein cholesterol (LDL-C) has been commonly calculated by equations, but their performance has not been entirely satisfactory. This study aimed to develop a more accurate LDL-C prediction model using machine learning methods. Methods The study involved predicting directly measured LDL-C, using individual characteristics, lipid profiles, and other laboratory results as predictors. The models applied to predict LDL-C values were multiple regression, penalized regression, random forest, and XGBoost. Additionally, a novel 2-step prediction model was developed and introduced. The machine learning methods were evaluated against the Friedewald, Martin, and Sampson equations. Results The Friedewald, Martin, and Sampson equations had root mean squared error (RMSE) values of 12.112, 8.084, and 8.492, respectively, whereas the 2-step prediction model showed the highest accuracy, with an RMSE of 7.015. The LDL-C levels were also classified as a categorical variable according to the diagnostic criteria of the dyslipidemia treatment guideline, and concordance rates were calculated between the predictive values obtained from each method and the directly measured ones. The 2-step prediction model had the highest concordance rate (85.1%). Conclusion The machine learning method can calculate LDL-C more accurately than existing equations. The proposed 2-step prediction model, in particular, outperformed the other machine learning methods. [ABSTRACT FROM AUTHOR]

Background Lower skeletal muscle mitochondrial function is associated with future cognitive impairment and mobility decline, but the biological underpinnings for these associations are unclear. We examined metabolomic markers underlying skeletal muscle mitochondrial function, cognition and motor function. Methods We analysed data from 560 participants from the Baltimore Longitudinal Study of Aging (mean age: 68.4 years, 56% women, 28% Black) who had data on skeletal muscle oxidative capacity (post-exercise recovery rate of phosphocreatine, kPCr) via 31P magnetic resonance spectroscopy and targeted plasma metabolomics using LASSO model. We then examined which kPCr-related markers were also associated with cognition and motor function in a larger sample (n = 918, mean age: 69.4, 55% women, 27% Black). Results The LASSO model revealed 24 metabolites significantly predicting kPCr, with the top 5 being asymmetric dimethylarginine, lactic acid, lysophosphatidylcholine a C18:1, indoleacetic acid and triacylglyceride (17:1_34:3), also significant in multivariable linear regression. The kPCr metabolite score was associated with cognitive or motor function, with 2.5-minute usual gait speed showing the strongest association (r = 0.182). Five lipids (lysophosphatidylcholine a C18:1, phosphatidylcholine ae C42:3, cholesteryl ester 18:1, sphingomyelin C26:0, octadecenoic acid) and 2 amino acids (leucine, cystine) were associated with both cognitive and motor function measures. Conclusion Our findings add evidence to the hypothesis that mitochondrial function is implicated in the pathogenesis of cognitive and physical decline with aging and suggest that targeting specific metabolites may prevent cognitive and mobility decline through their effects on mitochondria. Future omics studies are warranted to confirm these findings and explore mechanisms underlying mitochondrial dysfunction in aging phenotypes. [ABSTRACT FROM AUTHOR]