This is the Term Resolution page for NIDM-Experiment (NIDM-E), meaning that each term in NIDM-E has a URI that can be referred to when the term is used in a semantic web application. For example, https://incf-nidash.github.io/nidm-experiment/index.html#blooddraw points to the entry on this page for the term nidm:blooddraw, in the nidm namespace.

NIDM-E is a controlled terminology that can be used to describe neuroscience-related experiments. While NIDM-E was initially constructed to describe MRI-based neuroimaging studies, it has expanded to include both general terms needed to describe experiments and studies and domain-specific terms for non-MRI data.

While NIDM-E should currently be considered a "terminology" it is built from collections of terms from existing ontologies and augmented only when a needed term either does not exist in an ontology or the existing term has a different meaning or usage from that relevant for NIDM-E. The basis of NIDM-E is PROV-O, an W3C provenance ontology that has as its base elements, Entities, Activities, and Agents. Other terms have been imported from general ontologies as diverse and the Semanticscience Integrated Ontology (SIO) and the Information Artifact Ontology (IAO), and from domain-specific ontologies such as Ontoneurolog. NIDM-E also contains collections of terms that are used in the Brain Imaging Data Structure (BIDS) and DICOM standards.

Almost all terms in NIDM-E contain either definitions, or in the case of the BIDS terms, descriptions (which are not formal definitions). There are a few exceptions from ontologies in which most imported terms have definitions, but a very few do not. Terms imported into NIDM-E keep the definition from the parent ontology. Non-imported terms, those not found in other ontologies and created specifically for NIDM-E, all have definitions in the form of "X is a Y which Zs", where X is the term, Y is the genus and Z are the differentia (Seppala, Ruttenburg, Smith. Ci.Inf., Brasília, DF, v.46 n.1, p.73-88, jan./abr. 2017).

The terms in NIDM-E are broken into several categories: classes, datatype properties, object properties, annotation properties, and named individuals. Terms considered to be classes are general terms that often describe what an experimental variable is "about", e.g., "age". Datatype properties connect numerical values to Entities, while object properties connect Entities and annotation properties are used to attach a note to an Entity that describes something about the Entity.

NIDM-E should also be considered "under construction". As noted above, the initial terms in NIDM-E were related to MRI-based neuroimaging, but NIDM-E has been constructed such that "hooks" exist that can be used to add terms needed to annotate terms from other modalities. NIDM-E was built by selecting terms needed to annotate real datasets and has expanded as new datasets are considered. As such, NIDM-E is a "practical", rather than "theoretical" terminology and so terms may be missing that may be useful, but were not needed for the datasets upon which NIDM-E was built to describe.

Suggestions for new terms, edits to term parentage, and term definitions are welcome. These can be suggested using the Issues function in GitHub here. Click on "Issues" -> "New issue" and a list of issue-type templates will become available. These can be used to make suggestions and track the discussion and resolution of each issue.

A use case for NIDM-E is shown below to illustrate how the MRI acquisition activity might be modeled.