AUTHOR=Liu Haihong , Zhang Xiaolei , Liu Haining , Chong Sheau Tsuey 

TITLE=Using Machine Learning to Predict Cognitive Impairment Among Middle-Aged and Older Chinese: A Longitudinal Study

JOURNAL=International Journal of Public Health

VOLUME=Volume 68 - 2023

YEAR=2023

URL=https://www.ssph-journal.org/journals/international-journal-of-public-health/articles/10.3389/ijph.2023.1605322

DOI=10.3389/ijph.2023.1605322

ISSN=1661-8564

ABSTRACT=Objective: To explore the predictive effect of machine learning on cognitive impairment, and to explore its important factors affecting cognitive impairment. 
Methods: A total of 2,326 middle-aged and elderly people completed questionnaire, and physical examination evaluation at baseline, Year 2, and Year 4 follow-ups. This study employed a random forest model and logistic regression to predict the cognitive impairment of Year 2 and Year 4 longitudinally. Through the Year 4 cross-sectional data, the same method was applied to establish a prediction model and verify the accuracy of the longitudinal prediction of cognitive impairment. 
Results: Random forest models showed high accuracy for all outcomes at Year 2, Year 4, and cross-sectional Year 4 [AUC = 0.81, 0.79, 0.80] compared to logistic regression [AUC=0.61, 0.62, 0.70]. Baseline physical examination (e.g., BMI), biomarkers (e.g., cholesterol), functioning (e.g., functional limitations), demography (e.g., age), and emotional status (e.g., depression) characteristics emerged as the top ten important predictors in predicting cognitive impairment. 
Conclusions: ML algorithms could enhance the prediction of cognitive impairment among the middle-aged and older Chinese for four years and identify essential risk markers.