Towards structured sharing of raw and derived neuroimaging data across existing resources

摘要

Data sharing efforts increasingly contribute to the acceleration ofscientific discovery. Neuroimaging data is accumulating in distributeddomain-specific databases and there is currently no integrated access mechanismnor an accepted format for the critically important meta-data that is necessaryfor making use of the combined, available neuroimaging data. In thismanuscript, we present work from the Derived Data Working Group, an open-accessgroup sponsored by the Biomedical Informatics Research Network (BIRN) and theInternational Neuroimaging Coordinating Facility (INCF) focused on practicaltools for distributed access to neuroimaging data. The working group developsmodels and tools facilitating the structured interchange of neuroimagingmeta-data and is making progress towards a unified set of tools for such dataand meta-data exchange. We report on the key components required for integratedaccess to raw and derived neuroimaging data as well as associated meta-data andprovenance across neuroimaging resources. The components include (1) astructured terminology that provides semantic context to data, (2) a formaldata model for neuroimaging with robust tracking of data provenance, (3) a webservice-based application programming interface (API) that provides aconsistent mechanism to access and query the data model, and (4) a provenancelibrary that can be used for the extraction of provenance data by imageanalysts and imaging software developers. We believe that the framework and setof tools outlined in this manuscript have great potential for solving many ofthe issues the neuroimaging community faces when sharing raw and derivedneuroimaging data across the various existing database systems for the purposeof accelerating scientific discovery.