Classifying Dementia and Alzheimer's Disease Participants Using Machine Learning Methods Based on Comorbidities and Lifestyle Factors

This study highlights the potential of machine learning methods, such as XGBoost, to accurately classify participants with diagnosed dementia, Alzheimer's disease, or MCI against a healthy population. By leveraging these features and identifying key predictors through feature importance testing, machine learning methods enhance the precision of participant classification. Additionally, analyzing large publicly available datasets supports the development of future risk prediction models for undiagnosed individuals, ultimately facilitating earlier interventions and improved patient outcomes in dementia care and research.