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: | Zhenhong Deng |
| Institution: | Sun Yat-Sen memorial hospital, Sun Yat-Sen University |
| Department: | Department of Neurology |
| Country: | |
| Proposed Analysis: | People with mild cognitive impairment (MCI) have an increased risk of progressing to dementia. Roughly half of individuals with MCI develop dementia in the course of 3 years while the other half remain stable or revert to normal levels of cognition. It’s of great significance to identify individuals who are at high-risk of progression to dementia so that necessary interventions can be initiated as soon as possible.However, according the Practice guidelines for MCI from the American Academy of Neurology, to date, there are no biomarkers clearly shown to predict progression in patients with MCI. These biomarkers maybe not effective enough to recognizing individuals at high-risk of progression to dementia and therefore are not yet ready for clinical implementation. By utilizing ADNI’s dataset, we aim to explore risk factors and related biomarkers for conversion from MCI into dementia; establish and valid a prediction model for prognosis for individuals with MCI. We will select eligible participants from ADNI and the inclusion criteria are baseline diagnosis of MCI, availability of cognitive function assessments, MRI and/or gene data, and at least 6 months of follow-up. We plan to use Cox proportional hazards regression analysis to construct prognostic models based on dataset consist of clinical characteristics (including cognitive function assessment), CSF biomarkers, MRI measures and genotypes (SNPs). The primary clinical end point was any type of dementia. All patients who progressed to dementia are included in the analysis. Subsequently, we divide patients into AD-group (patients who progressed to dementia caused by AD) and non-AD group (patients who progressed to dementia not caused by AD) and repeat the analysis with conversion to either AD or other types of dementia as the clinical end point. First, we constructed a prognostic model only based on the clinical characteristics (demographic model). Second, we expand this model with MRI measures (MRI model) or genotypes (gene model), or CSF biomarkers (CSF model). Finally, we combined the previous models into 1 prognostic model containing clinical characteristics, CSF biomarkers, MRI measures and genotypes. The prognostic accuracy of these model will be estimated by the Harrell C statistic. Further, we intend to generate a nomogram and use calibration curves to assess whether actual outcomes approximately predicted outcomes for the nomogram and use time-dependent receiver operating characteristic (ROC) curves and areas under the curves (AUCs) to assess prognostic accuracy. |
| Additional Investigators |
