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: | Hao Yuan Bai |
| Institution: | McGill University |
| Department: | Computer Science |
| Country: | |
| Proposed Analysis: | Recent years have seen the rise of a network-based theory of pathological progression in neurodegenerative diseases. This theory posits that misfolded proteins induce cytotoxic changes in neurons and that these misfolded proteins propagate along neuronal pathways rather than diffuse through the brain parenchyma. In the case of Alzheimer’s disease (AD), two different misfolded proteins, ß-amyloid and tau, have been implicated as the toxic agents. Initially, it was thought that ß-amyloid was the principal agent of cell death and tissue atrophy associated with AD but, in the last decade, evidence has emerged that suggest that tau has a more direct role while ß-amyloid may still have an accelerant role in the pathological cascade. We developed an Epidemic Spreading Model (ESM, Iturria-Medina et al., 2014) that models the propagation of ß-amyloid along white-matter pathways, using PET data to measure ß- amyloid concentration and MRI/DTI data to measure cortical anatomy and white matter pathways. This model revealed the fact that the accumulation of ß-amyloid was a result of “underclearance” by a compromised neurovascular network rather than over-production by the cellular apparatus, a finding with profound ramifications for the design of therapeutic intervention for AD. We subsequently applied the ESM model to tau propagation, using a large (> 1000 patients) multi-centre PET cohort. Since the dataset was cross-sectional, we first used the Subtype and Stage Inference (SuStaIn) method (Young et al., 2018) to identify the spatiotemporal pattern of ß-amyloid propagation for four AD subtypes in the cohort. ESM was then applied to each of the sub-types in turn. The sub-types revealed clear differences in ß-amyloid propagation that matched well with the clinical presentation of the subjects within each sub- type. The above mechanistic characterization of disease sub-type allows for a stratified intervention protocol for each sub-type and, in future, the potential for personalized intervention. However, there remains significant work to be done. Notably, the current ESM model assumes that all nodes (i.e. brain regions) and spokes (white matter connections) in the network are identical. We now propose to extend the model to allow for heterogeneity in nodal (chemical, genetic) and spoke (direction, neurotransmitter type) attributes. This project will (i) extend the ESM framework to allow for these extra dimensions and will use existing tau cohort to explore the impact of these refinements on model accuracy. |
| Additional Investigators |
