NIDM — Neuroimaging Data Model

Overview

NIDM (Neuroimaging Data Model) is a W3C PROV-based semantic data model for representing neuroimaging experiments, statistical results, and analysis provenance. Developed within the ReproNim ecosystem and the NIDASH working group under INCF, NIDM provides a machine-readable, graph-based representation of the neuroimaging research workflow, from acquisition parameters through to analysis pipelines and statistical results. It comprises two main components: NIDM-Experiment, which describes experimental design and dataset metadata using ontology terms, and NIDM-Results, which packages neuroimaging statistical maps with the full provenance of how they were produced. NIDM-Results is natively supported by SPM and FSL, and is accepted by NeuroVault for open sharing of statistical maps with methodological context. NIDM extends BIDS at the semantic layer: where BIDS organises the file structure, NIDM describes the meaning of the data and its provenance, using PROV-O as its formal foundation.

Connections

• Extends: BIDS (semantic provenance layer)
• Built on: PROV-O (W3C Provenance Ontology)
• Endorsed by: INCF (via NIDASH working group)
• Vocabularies: Cognitive Atlas (task terms), UBERON (anatomy)

Resources

• http://nidm.nidash.org
• https://doi.org/10.1038/sdata.2016.10 (Maumet et al. 2016, Scientific Data)
• https://github.com/incf-nidash/nidm-specs