Existing Data

Go back to the main page

Existing Data

Will the researcher use other datasets in the study (e.g., ABCD has high restrictions on re-release options, but HCP has more open re-release options).

Best Practices:

  • Familiarize yourself with sharing, privacy, and security requirements of any existing data sources that you will use.

Things to Avoid:

  • Data of questionable quality
  • Excessive variance in acquisition/analysis method across data

Value Set Definitions: 

  • No: No other datasets will be used
  • Yes: but the data has no re-release restrictions
  • Yes: and the data has re-release restrictions

Value of Use Case Example:

Maybe - Given that Jordan may not be able to obtain enough subjects to test her hypothesis, she should consider whether she can use an existing dataset.

Discussion of Use Case:

Jordan will need to determine whether incorporation of other data sets will either necessitate the use of Cloud resources (i.e., that is the only pathway for use), or whether they put any limits on the use of Cloud resources (e.g., privacy or security constraints).

Join the discussion
Comments

See Also:

Data Size
Data Complexity/Scope
    • Raw Data Repositories
    • Processed data archives:
      • NeuroVault: NeuroVault is a public repository of unthresholded statistical maps, parcellations, and atlases of the brain. It complements the raw data stores listed above by providing a host for the raw derived (but unthresholded) statistical maps that accompany published (usually static) thresholded example images. This greatly facilitates metaanalysis from the complete brain space in contrast to foci of activation-based metaanalysis that the typically published (thresholded) tabular results support. 
      • ReproLake: The ReproLake is the ReproNim publically accessible neuroimaging metadata store. It hosts metadata that facilitates search and discovery of information based upon experiment and acquisition details, analysis results, processing workflows, etc. 
      • Preprocessed Connectomes Project: The goal of the Preprocessed Connectomes Project is to systematically preprocess the data from the 1000 Functional Connectomes Project (FCP) and International Neuroimaging Data-sharing Initiative (INDI) and openly share the results. This effort greatly reduces the redundancy of the preprocessing that typically would have to be performed by each investigator accessing a particular dataset.