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: | Sofia Broomé |
| Institution: | Therapanacea |
| Department: | Machine learning R&D |
| Country: | |
| Proposed Analysis: | We are planning to predict clinical status, ADAS-Cog and ventricular volume from the multi-modal input variables found in the ADNI dataset. Our goal is to benchmark several methods for disease progression prediction in Alzheimer’s disease, including statistical, classical machine learning methods as well as deep learning methods. Regarding the deep methods, we will in particular focus on recurrent neural networks and attention-based (Transformer) models in order to model the temporal dimension of the data. We will use the ADNI dataset mainly in the framework of the TADPOLE challenge, i.e. use the D1-D3 splits, but also evaluate splits including newer ADNI data which was not part of TADPOLE to see if this may improve the baselines from the challenge run in 2019. |
| Additional Investigators | |
| Investigator's Name: | Julia Gachot |
| Proposed Analysis: | We are planning to predict clinical status, ADAS-Cog and ventricular volume from the multi-modal input variables found in the ADNI dataset. Our goal is to benchmark several methods for disease progression prediction in Alzheimer’s disease, including statistical, classical machine learning methods as well as deep learning methods. Regarding the deep methods, we will in particular focus on recurrent neural networks and attention-based (Transformer) models in order to model the temporal dimension of the data. We will use the ADNI dataset mainly in the framework of the TADPOLE challenge, i.e. use the D1-D3 splits, but also evaluate splits including newer ADNI data which was not part of TADPOLE to see if this may improve the baselines from the challenge run in 2019. |
| Investigator's Name: | Audrey Duran |
| Proposed Analysis: | We are planning to predict clinical status, ADAS-Cog and ventricular volume from the multi-modal input variables found in the ADNI dataset. Our goal is to benchmark several methods for disease progression prediction in Alzheimer’s disease, including statistical, classical machine learning methods as well as deep learning methods. Regarding the deep methods, we will in particular focus on recurrent neural networks and attention-based (Transformer) models in order to model the temporal dimension of the data. We will use the ADNI dataset mainly in the framework of the TADPOLE challenge, i.e. use the D1-D3 splits, but also evaluate splits including newer ADNI data which was not part of TADPOLE to see if this may improve the baselines from the challenge run in 2019. |
| Investigator's Name: | Enrica Cavedo |
| Proposed Analysis: | We are planning to predict clinical status, ADAS-Cog and ventricular volume from the multi-modal input variables found in the ADNI dataset. Our goal is to benchmark several methods for disease progression prediction in Alzheimer’s disease, including statistical, classical machine learning methods as well as deep learning methods. Regarding the deep methods, we will in particular focus on recurrent neural networks and attention-based (Transformer) models in order to model the temporal dimension of the data. We will use the ADNI dataset mainly in the framework of the TADPOLE challenge, i.e. use the D1-D3 splits, but also evaluate splits including newer ADNI data which was not part of TADPOLE to see if this may improve the baselines from the challenge. |
