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: | Zoe Kourtzi |
Institution: | University of Cambridge |
Department: | Psychology |
Country: | |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate pre-symptomatic prediction of Alzheimer's disease. |
Additional Investigators | |
Investigator's Name: | Avraam Papadopoulos |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Nadia Smith |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Ignacio Partarrieu |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Spencer Thomas |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Jenny Venton |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Matt Hall |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | May Yong |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Shujun Wang |
Proposed Analysis: | Machine learning approaches for imputing longitudinal data as part of the main project |
Investigator's Name: | Yuanxi Li |
Proposed Analysis: | Given the difficulties in acquiring PET scans combined with the challenge of accurate prediction of Alzheimer's disease subtypes, we are centring our analysis on two pillars; first, using ADNI cohorts to develop an image translation technique via normalising flow to accurately reconstruct PET images using T1w scans. Second, developing graph based semi-supervised learning techniques to accurately predict the disease subtypes while integrating phenotypical, demographics and cognitive variables. We are aiming to combine these two techniques to ultimately deliver a less invasive, quicker and more accurate presymptomatic prediction of Alzheimer's disease. |
Investigator's Name: | Michael Burkhart |
Proposed Analysis: | To apply machine learning methods on trajectory modeling of dementia progression. |