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: | Mai Ho |
| Institution: | University of New South Wales |
| Department: | Medicine and Healthy (Discipline of Psychiatry) |
| Country: | |
| Proposed Analysis: | 1) Objective: The primary objective of our proposed study is to develop a normative model to understand the neuroanatomical heterogeneity in Alzheimer's Disease (AD). We aim to analyze the ADNI dataset, comprising of neuroimaging data and associated clinical variables, to examine the individual variations in neuroanatomy and their relationship with AD progression and risk factors. 2) Methodology: Our study will implement the following step-by-step analysis: a) Data Preparation: We will extract the required data fields from the ADNI dataset, primarily focusing on MRI neuroimaging data, patient demographics (age, sex), clinical parameters and AD risk factors. b) Normative Model Construction: Utilizing neuroimaging data from healthy controls within the dataset, we will develop a normative model that represents the expected neuroanatomy as a function of age and sex. The model will be developed through machine learning algorithms, providing a multidimensional understanding of the healthy aging brain. c) Deviation Quantification: Next, we will quantify the deviation of each individual in the disease cohort from this normative model, creating a region-wise measure of neuroanatomical deviation. These deviations will be examined with respect to different clinical subtypes of Alzheimer's Disease, to identify neuroanatomical patterns that may associate with disease subtypes. d) Clinical Risk Factor Analysis: We will perform statistical analyses to investigate the association between the magnitude of these deviations and clinical risk factors for Alzheimer's Disease (e.g., ApoE4 status, family history, cardiovascular risk factors, etc.). e) Biomarker Evaluation: Finally, we aim to assess whether the identified neuroanatomical deviations have potential as biomarkers for stratifying cerebrovascular disease in AD patients. This will be done through a cross-validation approach, testing the predictive power of these deviations on an independent subset of the dataset. 3) Statistical Analysis: We will use appropriate statistical methods such as regression analyses, machine learning algorithms and survival analyses, to ascertain the associations between neuroanatomical deviations, clinical risk factors and disease outcomes. 4) Expected Outcomes: Through this analysis, we aim to advance our understanding of individual neuroanatomical variations in AD, and how these can inform risk prediction, disease progression, and potentially inform targeted therapeutic strategies. The results will potentially contribute to a personalized medicine approach in AD, where treatment and management strategies could be tailored according to the patient's unique neuroanatomical profile and associated clinical risks. |
| Additional Investigators |
