A machine learning model based on routine blood-derived indices for early arterial stiffness prediction in the community - PubMed
4 hours ago
- #cardiovascular risk
- #arterial stiffness
- #machine learning
- Study developed a machine learning model for early arterial stiffness prediction using routine blood-derived indices.
- Increased arterial stiffness is a high-risk factor for cardiovascular diseases, necessitating early identification.
- 2948 community participants were enrolled in a cross-sectional study between June and December 2024.
- 24 blood-derived indices from metabolic, lipid, and inflammatory domains were analyzed.
- LASSO regression and logistic regression identified nine independent risk factors for arterial stiffness.
- Random forest model and SHAP analysis highlighted key predictors like age, blood pressure, and TyG-WHtR.
- The nomogram demonstrated excellent discrimination (AUC = 0.877) and good calibration.
- A web-based calculator was developed for individualized risk estimation.
- TyG-WHtR emerged as a key independent predictor of arterial stiffness.
- The model offers a practical tool for early community-based screening of arterial stiffness.