Skip to main content

BIDS and the NeuroImaging Data Model (NIDM)

The following statement is designed to clarify the complementary nature of the Brain Imaging Data Structure(BIDS) and the NeuroImaging Data Model (NIDM). It is not designed to be a comprehensive review of either of these initiatives, but rather to highlight the synergy of the co-application of each of these emerging technologies. The context for this statement is the preservation and communication of neuroimaging data and the additional experimental data and analytics that are associated with the neuroimaging data.

BIDS is a standard prescribing a formal way to name and organize neuroimaging data and metadata in a file system that simplifies communication and collaboration between users, and enables easier data validation and software development through using consistent paths and naming for data files.

NIDM is a Semantic Web based metadata standard that helps capture and describe experimental data, analytic workflows and statistical results via the provenance of the data. NIDM uses consistent data descriptors in the form of machine accessible terminology, and a formal, extensible data model, which enables rich aggregation and query across datasets and domains.

BIDS is strict regarding file organization, naming, and file metadata but, in order to support wide adoption, permits substantial flexibility in the details of how other dataset metadata are described within the standard. NIDM formalizes the metadata description, permitting it to be ‘self-describing’, at the ‘cost’ of additional model complexity and required input from the user to define metadata and provenance details.

Within the overarching goal of supporting neuroimaging data reuse (through, e.g., publishing, sharing, and aggregation), both of these elements, data organization and data semantics, are critical. BIDS adds to ad hoc raw and derived brain imaging data the ability, for humans and machines, to use a consistent data structure;  NIDM adds additional capabilities to track provenance and disambiguate experimental details and data elements that, which while perhaps clear to the original data collector, may have substantial ambiguity to the eventual data user.

As a community, we recommend the integrated use of both of these standards, (BIDS plus NIDM can be defined as a “SemanticBIDS” representation), in order to both maximize the ‘ease of [re]use’ and ‘ease of sharing’ of neuroimaging data in support of greater research transparency. The BIDS and NIDM development communities will continue to work together to build tools for further synergies between these initiatives.

Signed, (Members of the BIDS and NIDM development communities, last updated 03:11EDT, 06-AUG-19)
Stefan Appelhoff, Center for Adaptive Rationality, Max Planck Institute for Human Development, Berlin, Germany
Julianna F. Bates, University of Massachusetts Medical School, Worcester, USA Satrajit Ghosh, Massachusetts Institute of Technology, Cambridge, USA
David B. Keator, University of California, Irvine, Irvine, USA
David N. Kennedy, University of Massachusetts Medical School, Worcester, USA
Russell Poldrack, Stanford University, Stanford, USA
Jean-Baptiste Poline, McGill University, Montreal, Canada
Jason Steffener, University of Ottawa, Ottawa, Canada
B. Nolan Nichols, Genentech, South San Francisco, USA Franklin Feingold, Stanford University, Stanford, USA Cyril Pernet, University of Edinburgh, Edinburgh, UK
Gustav Nilsonne, Karolinska Institute, Stockholm, Sweden
Camille Maumet, French National Institute for Research in Computer Science and Control (INRIA), Rennes, France
Guillaume Flandin, University College London, London, UK Rémi Gau, Université Catholique de Louvain, Louvain, Belgium Robert Oostenveld, Radboud University, Nijmegen, The Netherlands Elizabeth DuPre, McGill University, Montréal, Canada
Arnaud Delorme, University of California, San Diego, San Diego, USA Christopher J. Markiewicz, Stanford University, Stanford, USA
Natacha Perez, French National Institute for Research in Computer Science and Control (INRIA), Rennes, France
Karl G. Helmer, Massachusetts General Hospital, Harvard Medical School, Charlestown, USA Dorota Jarecka, Massachusetts Institute of Technology, Cambridge, USA
Jeffrey S. Grethe, University of California, San Diego, San Diego, USA Dianne Patterson, University of Arizona, Tucson, USA
Tibor Auer, University of Surrey, Guildford, UK
Hauke Bartsch, Haukeland University Hospital, University of Bergen, Bergen, Norway
Thomas E. Nichols, Oxford Big Data Institute, Li Ka Shing Centre for Health Information and
Discovery, Nuffield Department of Population Health, University of Oxford, Oxford, UK
Vince Calhoun, Tri-Institutional Center for Translational Research in Neuroimaging and Data Science
(TReNDS) [Georgia State University, Georgia Institute of Technology, Emory University],
Atlanta, GA, USA