Analysis of the features related to healthy aging
Approved Research ID: 86776
Approval date: August 25th 2022
The aim of this research is to establish the machine learning based predictive models to predict elderlies with subjective cognitive decline or mild cognitive impairment at high risk of future neurodegenerative progression through MRI and/or patient genotyping. Imaging features may serve as surrogates for certain outcome phenotypes. Genetic signatures can thus be linked to imaging features.
Specifically, the longitudinal MRI and genetic signatures of patients will be statistical analyzed to reveal the clinical risks of neurodegenerative effects in the individual level. Then, we propose to use the machine learning-based algorithm to model the longitudinal disease progression of the patients. Finally, the constructed machine learning-based predictive models can provide individualized diagnostic and therapeutic strategies in each patient. Combining genetic profiles and imaging features also has powerful synergistic potential in terms of risk stratification and precision medicine, proving opportunities for early, targeted intervention and prognostication.