Development of an interpretable machine learning model-based online tool for risk prediction of anxiety symptoms in Chinese older adults - PubMed
5 days ago
- #elderly-health
- #machine-learning
- #anxiety-symptoms
- Development of an online tool using interpretable machine learning (ML) models to predict anxiety symptoms (AS) in Chinese older adults.
- Study utilized data from the Chinese Longitudinal Healthy Longevity Survey (CLHLS) 2018, with 38 predictor variables based on the health ecology model.
- Five ML methods (Random Forest, XGBoost, LightGBM, Decision Tree, Logistic Regression) were used, with Random Forest performing the best (AUC: 0.790).
- Key risk factors identified include sleep quality, self-reported health, depressive symptoms, vegetable intake, ventilation method, and economic situation.
- Prevalence of AS among 9535 participants was 11.89%.
- Online tool available at: https://jiqixueximoxingyuce.shinyapps.io/anxiety_prediction_tool/.
- Aims to enable early identification of high-risk groups and precise interventions for healthy aging.