Assessing Alzheimer's Disease Risk and Heterogeneity Using Multimodal Machine Learning Approaches
HSS Investigators: Honghuang Lin, Biqi Wang
Funding Agency: National Institute on Aging
Status: Ongoing
Project Overview: Alzheimer's Disease (AD) is the most common form of dementia characterized by progressive loss of cognitive function. Unfortunately, there are only very limited treatment options for AD and clinical interventions for AD have largely failed despite enormous efforts. We seek to develop multimodal machine learning models by leveraging the extensive collection of AD-related omics and phenotypical data recently generated from large-scale collaborative projects such as Alzheimer Disease Neuroimaging Initiative (ADNI), Accelerating Medicines Partnership-AD (AMP-AD) and the Alzherimer's Disease Sequencing Project (ADSP). Given the multifactorial nature of AD, the project will build an unsupervised machine learning framework to perform AD subtyping by harnessing rich informatuon across multiple spectrums of data, such as generic information, blood biomarkers, neuropsychological tests, and neuroimages. We will further develop expandable multimodal supervised machine learning models to assess AD risk from longitudinal follow-up of cognitively normal elders. The software packages developed from the current study will be shared through GitHub or CRAN for free access across the scientific community.