Development and external validation of a machine learning model for cardiovascular risk prediction in individuals with chronic lung disease: Evidence from CHARLS and ELSA - Summary - MDSpire
Advertisement
Development and external validation of a machine learning model for cardiovascular risk prediction in individuals with chronic lung disease: Evidence from CHARLS and ELSA
To develop machine learning-based cardiovascular disease (CVD) risk prediction models specifically for patients with chronic lung diseases (CLD) using data from CHARLS and validate the models with ELSA data.
Key Findings:
XGBoost was identified as the optimal prediction model for CVD risk in the CLD population.
The model demonstrated robust generalizability when validated with external data from ELSA.
Key predictors included demographics, lifestyle factors, and health status.
Interpretation:
The study highlights the potential of machine learning in enhancing CVD risk prediction specifically for patients with chronic lung disorders.
Limitations:
The reliance on self-reported diagnoses for CLD and CVD may introduce reporting bias.
The study's findings may not be generalizable to populations outside of China and the UK.
Conclusion:
The developed machine learning model provides a promising approach for CVD risk assessment in patients with CLD, facilitating early identification and intervention.
