NIDM-Terms: community-based terminology management for improved neuroimaging dataset descriptions and query

Abstract

The biomedical research community is motivated to share and reuse data from studies and projects by funding agencies and publishers. Effectively combining and reusing neuroimaging data from publicly available datasets, requires the capability to query across datasets in order to identify cohorts that match both neuroimaging and clinical/behavioral data criteria. Critical barriers to operationalizing such queries include, in part, the broad use of undefined study variables with limited or no annotations that make it difficult to understand the data available without significant interaction with the original authors. Using the Brain Imaging Data Structure (BIDS) to organize neuroimaging data has made querying across studies for specific image types possible at scale. However, in BIDS, beyond file naming and tightly controlled imaging directory structures, there are very few constraints on ancillary variable naming/meaning or experiment-specific metadata. In this work, we present NIDM-Terms, a set of user-friendly terminology management tools and associated software to better manage individual lab terminologies and help with annotating BIDS datasets. Using these tools to annotate BIDS data with a Neuroimaging Data Model (NIDM) semantic web representation, enables queries across datasets to identify cohorts with specific neuroimaging and clinical/behavioral measurements. This manuscript describes the overall informatics structures and demonstrates the use of tools to annotate BIDS datasets to perform integrated cross-cohort queries.

Keywords: NIDM; annotation; dataset; neuroimaging; query.

FIGURE 1

A schematic of a acquisition object “T2.nii” that was generated from the Session “Visit_3” of participant “1f3g2k6” annotated with the protocol used and “hadImageUsageType” and “hadImageContrastType.”

FIGURE 2

This figure illustrates the various functionalities the NIDM-Terms User Interface (UI) supports including: browse terms (A), suggest terms (B), export terms (C), and add new communities (D).

FIGURE 3

Stage 1 of the neurobagel annotator workflow. Left: a list of pre-defined categories (common data elements) is shown, each associated with a specific color. Right: the column names of a loaded demographic.tsv file with optional descriptions are displayed. The user now selects each category by clicking on it (e.g., “Diagnosis” in yellow), and then associates the category with each column that contains information about this category by clicking on it (e.g., “group,” “group_dx,” and “number_comorbid_dx”). An existing association is reflected by the column being highlighted in the color of the category.

FIGURE 4

Stage 2 of the neurobagel annotator workflow. The user is now asked to annotate the values inside each column that has been linked to one of the predefined categories. Left: each category associated with at least one column is represented with a colored button. By clicking on each button, the user can annotate the values in the associated columns. Right: the annotation view for each category (here “Sex” in blue) contains specific elements such as an explanation (collapsed here), an overview of the associated column names (here “sex”), and an overview of the unique values in the associated column (bottom). The user can map each unique value to a pre-defined list of controlled terms (here with a drop-down menu) or indicate that the value reflects a “missing value” (e.g., a data entry error or a truly missing response).

FIGURE 5

An illustration of the NIDM-Terms workflow.