Standardized Centralized Alzheimer’s
& Related Dementias Neuroimaging
Images are uploaded by ADRCs to a portal in the Laboratory of Neuroimaging (LONI) at the University of Southern California where they are de-identified, and then processed for quality assurance and harmonization at the PET and MRI laboratories at the University of Michigan and Mayo Clinic. Following this, images are analyzed by the PET laboratory at UC Berkeley, and the MRI laboratories at Mayo Clinic, and UC Davis. Harmonized images are available through queries at NACC (with data stored at LONI), as are numerical summaries. Results will be returned to the ADRCs via the NACC database.
Current SCAN protocols are focused on widely accepted and widely applied biomarkers, generally, but not exclusively related to characterizing research participants in the A/T/N framework. SCAN goals are to assist the ADRCs in the pursuit of collaborative research, but we do not want to stifle innovative new approaches to imaging data. Future directions will be based on the needs and goals of the ADRC program, and may include other MRI techniques or PET tracers as best practices for their acquisition and analysis become available, and depending on the interest of the NIA Alzheimer’s Disease Research Centers.
Prospective image files that meet SCAN acquisition protocols will be included in this project, but existing images and prospective images that don’t meet protocol will still be accepted by NACC.
For many years, researchers at the NIA-funded Alzheimer’s Disease Research Centers have been collecting different types of PET and MRI images on their research participants. These data, which we call “legacy data” were collected using different acquisition methods. Because of this, image data could not be simply combined across Centers, resulting in lost opportunities for scientific collaboration. Some legacy data is currently stored on the National Alzheimer’s Coordinating Center (NACC) website. However, for the immediate future, SCAN will not focus on this legacy data, but on prospectively acquired data.
The goal of SCAN will be to promote standardization of PET and MR image acquisition so that images can be combined across multiple centers. These prospectively acquired images will be subject to standardized quality control procedures, and then uploaded to NACC where they will be appropriately archived, labeled, and made available to qualified investigators. In addition, the SCAN team will develop numerical summaries of the PET and MR data that will be available on the NACC website and linked to the ADRC participants through their ID. Appropriate images for inclusion in the SCAN workflow are images acquired prospectively on NACC participants, following SCAN protocols regardless of the funding source.
SCAN has a set of required acquisition schemes for PET and MR data. This includes required PET protocols for acquiring amyloid, tau, and FDG PET scans, and required MRI sequences for structural T1 and FLAIR scans. Additional MR sequences are optional. Analyzed data available to investigators will include regional measures of amyloid, tau, and FDG tracer uptake, brain volumes, white matter hyperintensities, cerebral infarction, and, depending on the availability of advanced MR sequences, fractional anisotropy, cerebral blood flow, and functional connectivity.
Timing of SCAN rollout
SCAN methods are currently being developed and instituted. Below is a rough timeline of the anticipated workflow
- Finalizing technical manuals, procedures for data uploading
- Begin image uploads
- Begin image QC and analysis, upload documentation
- Expand website FAQs, help desk
MRI and PET acquisition protocols
All longitudinal scans must be performed on the same MRI scanner.
Each MRI scanner must be certified and approved by SCAN. MR scanners that are now certified for ADNI 3 can be certified for SCAN using a phantom scan. MR scanners that have not been certified for ADNI 3 (or SCAN) must be certified using a human volunteer scan. This must be permissible using a local IRB – i.e. SCAN does not plan to create a central IRB nor will SCAN manage local IRB administrative proceedures.
The Standard MRI protocol (using SCAN-specific parameters) is required. Your remaining scan time can be used for your site specific sequences.
There are 2 options for SCAN MR – each site can choose which they prefer.Option 1: Site only required to do 2 primary SCAN sequences: T1 & FLAIR
- The first 10 minutes of exam belongs to NIA/SCAN — this is when the SCAN T1 & FLAIR would be done
- Remainder of exam is unique to site
- SCAN would only collect and analyze T1 and FLAIR
All longitudinal scans must be performed on the same PET scanner.
Each PET scanner must be certified and approved by SCAN. ADNI approved PET scanners will not need to be re-certified.
The following tracers and imaging protocols are all compatible with SCAN:
|Tracer||Suggested Dose (mCi)||Acquisition start-stop|
|PIB||15||40–60 or 50–70|
|MK6240||5||70–90 or 90–110|
|PI2620 (non-AD)||5||0–60 or 60–90|
To be SCAN compliant, all image data (MRI and PET) will be uploaded to LONI accompanied by the required meta-data. Documentation will be posted here when available.
- SCAN will troubleshoot prospectively with ADRCs to advise on making data SCAN compliant.
- De-face decision pending.
- Only SCAN approved data should be uploaded to LONI, however if non-approved data are uploaded (with or without approved data) it will be quarantined.
- Fate of non-SCAN compliant data uploaded to LONI is currently uncertain — for future decision.
- “Legacy data” already at NACC: decision pending for management/curation/analysis.
SCAN is a multi-institutional project that was funded as a U24 grant (AG067418) by the National Institute on Aging in May 2020 with the goal of standardizing the acquisition, curation, and analysis of PET and MR images acquired through the NIA Alzheimer’s Disease Research Centers Program. The Principal Investigators are William Jagust (UC Berkeley) and Clifford Jack (Mayo Clinic). Other institutions participating in the leadership of SCAN include UC Davis (Charles DeCarli), Lawrence Berkeley National Laboratory (Suzanne Baker), and the University of Michigan (Bob Koeppe). Cerise Elliott is the NIA program official, and John Hsiao is the NIA scientific officer.