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: | Chang Su |
| Institution: | Weill Cornell Medicine |
| Department: | Department of Population Health Sciences |
| Country: | |
| Proposed Analysis: | Alzheimer’s disease (AD) is associated with diverse clinical manifestations and the pathology is unclear. Identification of AD subtypes may help to address the heterogeneity and provide appropriate treatment to patients. Previous studies mainly identified the AD subtypes based on the information at baseline, which may not capture the disease progression well. To address this, we propose to design a novel AD subtyping approach based on the longitudinal information of patients. In particular, we plan to extract the demographics, genetic information, neuroimage, longitudinal clinical assessments, and biospecimen data of patients for analysis. By integrating the multimodal data, we propose to mine the patterns of AD progression according to the longitudinal information such as motor and non-motor assessments, and biospecimen test results. Accordingly, we will design a clustering model that can identify AD subtypes with meaningful progression characteristics. More importantly, we expect to find meaningful markers for estimating the disease progression of a new patient at baseline. To this end, we will propose to design a prediction model for the progression subtypes based on the genetic information, neuroimages, and biospecimen test results at baseline. We have experience in developing computational models in neurodegenerative disease, e.g., Parkinson’s disease (PD) research. Specifically, 1) we have developed a novel model to identify phenotypic associations in PD using the Parkinson Disease Progression Marker Initiative (PPMI) dataset [1]. 2) We developed a novel PD subtyping model based on multimodality fusion using BioFIND dataset and drafted a paper ready to submit. 3) Our survey paper on mining genetic data in PD using machine learning has been published in the journal npj Parkinson’s Disease [2]. We believe that such experience in PD will allow us to develop robust computational models to improve AD. In this context, we expect to apply to access the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. We propose to develop computational models to disentangle the heterogeneity of AD, using the rich multi-modal information from ADNI. References: [1] Pan W, Su C., Chen K., Henchcliffe C, Wang F. Learning Phenotypic Associations for Parkinson's Disease with Longitudinal Clinical Records. medRxiv. 2020 Jan 1. [2] Su C., Tong J. & Wang F. Mining genetic and transcriptomic data using machine learning approaches in Parkinson’s disease. npj Parkinsons Dis. 6, 24 (2020). https://doi.org/10.1038/s41531-020-00127-w |
| Additional Investigators |
