There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
Principal Investigator | |
Principal Investigator's Name: | Kan Li |
Institution: | Merck & Co. |
Department: | BARDS |
Country: | |
Proposed Analysis: | Impairment caused by Alzheimer’s disease (AD) affects multiple domains (e.g., cognition, behavior, and quality of life) and progresses heterogeneously in time and across domains and individuals. The heterogeneous nature and unknown pathogenic mechanisms of AD make it impossible to use a single outcome to reliably reflect disease severity and progression. Consequently, AD studies, e.g., ADNI, collect a multitude of longitudinal outcomes including clinical variables (e.g., neuropsychological, functional, and behavioral assessments) and neuroimaging data (e.g., magnetic resonance imaging, MRI), which are predictive of AD progression. Moreover, rich information of genetic markers such as single-nucleotide polymorphisms (SNPs) are often available from genome-wide association studies (GWAS). The growing public threat of AD has raised the urgency to utilize these multi-modal (clinical, neuroimaging, and genetic) data to develop robust prediction models that identify individuals at greatest risk of future cognitive decline and AD onset, with the ultimate goal of providing targeted disease-modifying therapeutic intervention. We propose to use the rich multi-modal data to develop novel AD prediction models which provide accurate personalized predictions of disease progression and dynamically update the predictions based on new subject-specific data. This model development is important to target those at high risk of AD, as well as personalize their management, prognosis and treatment selection. The overall objectives of this project are to: (1) build an increasingly more sophisticated class of dynamic prediction models for personalized prediction of future outcome trajectories and risks of target events, using the multivariate longitudinal clinical outcomes; (2) generalize the prediction models via high-dimensional vertex-based morphology analysis of the MRI data; (3) improve the predictive performance by utilizing the GWAS data; (4) develop and standardize the proposed prediction models via professional software development and web deployment. The ADNI data is the best to fit our purpose. The investigator has published a series of papers in medical and statistical journals using ADNI dataset. |
Additional Investigators |