Enhancing Big Data Analytics in Disease Risk Prediction and Biomarker Discovery
Approved Research ID: 96818
Approval date: April 14th 2023
This study will examine aging biomarkers using biological measures, brain imaging data, and genetic information to predict physical, mental, and behavioral health outcomes. Additionally, we will combine the summary aging biomarkers with psychological, social and environmental factors using machine learning models. It is also worth examining whether the relationship between these composite aging biomarkers and health outcomes depends on subject-level characteristics (e.g., sex, race). Our primary aim is to develop new machine learning models to estimate brain age based on neuroimaging data and examine the estimated brain age as an early biomarker for predicting health outcomes. Our secondary aim includes the development of new statistical models to quantify and test whether biological, neuroimaging and epigenetics measures of aging jointly predict health outcomes. The expected duration of this project is 3 years. Better understanding and quantification of the aging process can provide important insights into preventive strategies and early detection of chronic conditions. Our derived aging biomarkers can be used by other researchers, contributing to future innovation in aging studies. Furthermore, findings from this study will provide valuable guidelines on tailoring machine learning models for predicting disease onset and progression.