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: | Alex Doran |
| Institution: | McGill University |
| Department: | Bioengineering |
| Country: | |
| Proposed Analysis: | I hope to model local and global dynamics correlating with individual cognition in Alzheimer's disease using a customized version of the "The Virtual Brain" dynamics model with parameters fit to resting state functional connectivity MRI as described in the paper Differentiation of Alzheimer's disease based on local and global parameters in personalized Virtual Brain models. I would first use a GP-based BO online learning approach to optimize stimulation patterns to minimize spiking patterns indicative of Alzheimer's such as hyperactivity in the hippocampus and impaired deactivation in the default mode of network during memory encoding. I would hierarchically build response models of more complex stimulation patterns starting with simple stimulation mechanisms. I plan to assess each biomarker or combination of biomarkers as the optimized cost function by measuring the effective dimensionality and discrimination task completion. This stimulation online learning approach would make use of the chaotic RNNs trained through stochastic gradient descent. Additionally, I will test the generalizability of this approach with higher dimensional inputs representing various patient states and hopefully improve generalizability by considering additional tasks and fitting model parameters to active state data. I also plan to explore the utility of long-term dynamics in a neurodegenerative model and how the DBS stimulation algorithm could preserve the initially measured functionality. Of course, I will focus this description with additional research. |
| Additional Investigators |
