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: | Ayodeji Ijishakin |
| Institution: | University College London |
| Department: | Computer Science |
| Country: | |
| Proposed Analysis: | I will be using data from ADNI in-order to train a state of the art deep generative modelling called a Diffusion model. This model will be able to generate high-resolution brains which look like those of individuals with Alzheimer's disease as well as healthy controls and individuals who have suffered traumatic Brian injury. This generative model will be multi-task, as it will have a generative component and well as a classification component. The utility of a model of this sort is that the classification is based on the similarity of individual brains to a 'prototypical' version of the disease which is learnt by the model. The idea is by learning prototypes you can infer in a data-driven way variants of the diseases which are detectable in MRI. You can also use this approach to understand why neural networks classify one brain image as one disease type over another, because the final prediction is a linear combination of the similarity scores. |
| Additional Investigators |
