Abstract
INTRODUCTION: The older adult are at high risk of sarcopenia, making early identification and scientific intervention crucial for healthy aging. METHODS: This study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), including a cohort of 2,717 middle-aged and older adult participants. Ten machine learning algorithms, such as CatBoost, XGBoost, and NGBoost, were used to construct predictive models. RESULTS: Among these algorithms, the XGBoost model performed the best, with an ROC-AUC of 0.7, and was selected as the final predictive model for sarcopenia risk. SHAP technology was used to visualize the prediction results, enhancing the interpretability of the model, and the system was built on a web platform. DISCUSSION: The system provides the probability of sarcopenia onset within 4 years based on input variables and identifies critical influencing factors. This facilitates understanding and use by medical professionals. The system supports early identification and scientific intervention for sarcopenia in the older adult, offering significant clinical value and application potential.