Related Experiment Video
Updated: Jan 8, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
[Development and validation of a risk prediction model for cognitive impairment in rural elderly Chinese populations:
Fei Wang1, Weiran Li1, Xiang Shang1
1First Clinical Medical College, Anhui University of Chinese Medicine, Hefei 230061, China.
Objectives:
To develop and validate a risk prediction model for cognitive impairment in community-dwelling elderly individuals in China.
Methods:
This cross-sectional study was based on data from the 2011 China Health and Retirement Longitudinal Study (CHARLS), and the data of 2228 individuals aged ≥60 years were analyzed. The participants were randomly divided into a training set (n=1560) and an internal validation set (n=668) in a 7∶3 ratio. Thirty-eight candidate variables were collected, covering sociodemographic characteristics, lifestyle and behavioral habits, chronic disease history, physical function, and self-rated health status. Feature selection was performed using the least absolute shrinkage and selection operator (LASSO) regression, followed by multivariate logistic regression to identify independent risk factors for cognitive impairment. A nomogram was constructed based on these factors, its discrimination power and calibration were assessed using the receiver operating characteristic (ROC) curve and calibration plot, respectively, and its clinical utility was evaluated using decision curve analysis (DCA).
Results:
Age, years of education, alcohol consumption, systolic blood pressure, grip strength, and depressive symptoms were identified as independent predictors of cognitive impairment in Chinese elderly individuals. The area under the ROC curve of the constructed nomogram was 0.839 (95% CI: 0.814-0.864) in the training set and 0.840 (95% CI: 0.801-0.879) in the validation set, indicating good predictive performance of the model. The calibration plots demonstrated good agreement between the predicted and observed outcomes, and the DCA showed good clinical utility of the model.
Conclusions:
The nomogram developed in this study based on LASSO-selected predictors demonstrates high accuracy, discrimination power, and potential clinical applicability to facilitate early identification and intervention of cognitive impairment among rural elderly individuals in China.
