MRI Core

MR Core: Summary/Abstract ADNI MR data has been used extensively in designing Alzheimer’s disease clinical trials, for observational research in normal brain aging and dementia, and for developing new image analysis methods, e.g., machine learning and cognitive/clinical prediction. During ADNI 3, 129 million ADNI MR downloads were provided to users. As in ADNI 2 and 3, the acquisitions for ADNI 4 will be performed at 3T. All sequences will be vendor product and FDA approved for clinical use. The ADNI 4 protocol will include 8 sequences: 3D T1 for morphometry; 3D FLAIR for structural cerebral vascular disease (CVD) ascertainment; a new whole brain sagittal 3D T2 to identify dilated peri vascular spaces; a new 2-tiered T2* gradient recalled echo (GRE) sequence to accommodate standard T2* images as well as the creation of susceptibility weighted imaging (SWI) and quantitative susceptibility mapping (QSM) at sites that have this capability; a 2-tiered diffusion MRI (dMRI) approach; perfusion arterial spin labeling (ASL) imaging at sites that can acquire 3D ASL; high resolution coronal T2 for hippocampal subfield measures; and task free functional MRI (TF-fMRI). The acquisition time will be under 1 hour to minimize participant burden. The objectives in ADNI 4 will continue to focus on providing curated images and numeric summary values from all MR modalities. A major shift in emphasis in ADNI 4 is toward recruitment that results in greater racial/ethnic, educational, and socio-economic diversity in the ADNI cohort. This is anticipated to result in greater pathologic heterogeneity, particularly a higher prevalence of CVD. To support this effort, we plan a total of six CVD-related measures in ADNI 4. This includes identification of infarctions, WMH volume, ASL perfusion, diffusion MRI, quantification of dilated peri vascular spaces, and cerebral micro bleeds. Specific aims of the MR core are: Aim 1) Perform data acquisition and quality control and distribute curated images for ADNI users. Aim 2) Provide quantitative measurements for all MR modalities and distribute these for ADNI users. Aim 3) Identify and deploy state of the art methods for post-acquisition data harmonization. While the ADNI core minimizes heterogeneity by standardizing acquisition parameters, some heterogeneity inevitably remains for a variety of reasons. To address this, a major new effort will be added in ADNI 4 to address MR data heterogeneity. Aim 4) Improve participant privacy protection by employing an optimized approach to deface images for each relevant MR modality and distribute defaced images for ADNI users. Aim 5) Optimize prediction of cognitive performance and clinical group membership cross sectionally using imaging and non-imaging data. Aim 6) Optimize predicting future change on functional/psychometric measures and progression from unimpaired to MCI, and MCI to dementia using imaging and non-imaging data. Aim 7) Identify optimum outcome metrics for clinical trials using imaging and non-imaging data.