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: | Meng Wang |
| Institution: | University of Calgary |
| Department: | Cumming School of Medicine |
| Country: | |
| Proposed Analysis: | This PhD project aims to develop more accurate dementia risk assessment tools that integrate expert clinician opinions with patient data to provide patients with personalized dementia risk estimates. The specific aims include (1) evaluate the predictive accuracy of tree-based machine learning models for predicting dementia risk using patient data, (2) elicit and quantify clinician expert knowledge about dementia risk factors and dementia risk, (3) develop and validate clinically useful risk indices that integrate clinician expert opinions with patient data to predict dementia risk. Therefore, This PhD project will use a combination of expert elicitation methodology, data, and statistical/ machine learning algorithms to achieve these objectives. Data from the Alzheimer’s disease (AD) Neuroimaging Initiative (ADNI) and PROspective Registry of Persons with Memory SyMPToms (PROMPT) registry, which are two registries of individuals with memory symptoms, are proposed to train and validate the models developed in this study. Potential risk predictors will include demographic and social-economic factors, cognitive assessments, neuroimaging, biomarkers, comorbidities, and behavior risk factors. Bayesian survival trees and random survival forest regression will be used to train and develop these dementia risk prediction models using the ADNI cohort. Expert elicitation methodology will be used to quantify and elicit expert opinions into subjective probabilities, which will be used as prior distributions for the tree-based ML models. Computer simulations and bootstrapping will be conducted to evaluate the predictive performance of these models based on calibration and discrimination. External validation of the models will be implemented in the PROMPT registry. |
| Additional Investigators | |
| Investigator's Name: | Tolu Sajobi |
| Proposed Analysis: | Meng (my PhD student) will be conducting analysis for developing a better risk assessment tool for dementia, using the combination Bayesian Statistics and Machine learning techniques. |
| Investigator's Name: | Eric Smith |
| Proposed Analysis: | Meng (my PhD student) will be conducting analysis for developing a better risk assessment tool of dementia, using the combination Bayesian Statistics and Machine learning techniques. |
