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: | Lewis Shaw |
| Institution: | Leeds Beckett University |
| Department: | Computer Science and Engineering |
| Country: | |
| Proposed Analysis: | Proposed is a non-commercial, academic study of machine learning classification for dementia diagnosis predictions. As a doctorate student a Leeds Beckett University, data use will adhere to the internal ethics of the institution. This research will be a direct succession to a previous study using the publicly available Oasis-2 dataset. A variety of supervised machine learning algorithms (including random forest decision tree, extreme gradient boosted decision tree and neural network) will be used to assess the viability of AI as a diagnostic aid. Furthermore, the longitudinal aspect of the available ADNI datasets will allow assessment of prognosis predictions, based on very early symptoms of dementia that machine learning may detect in the available variables. This aspect of dementia research in particular is important to the advancement of diagnostics, and where successful, in a real-world application will allow patients to receive necessary care/support as soon as possible. Finally, machine learning model interpretability techniques will be thoroughly explored to understand how the dataset features are being used by each algorithm, and to propose that AI decisions can be made understandable to clinicians without technical knowledge. Ultimately, where my previous research has faced limitations due to dataset size, I believe that the comparative size and dimensions of the ADNI datasets are likely to facilitate marginal improvements in the performance of the machine learning models I have constructed. |
| Additional Investigators |
