MRIO: The Magnetic Resonance Imaging Acquisition and Analysis Ontology

Abstract

Objective: Magnetic resonance imaging of the brain is a useful tool in both the clinic and research settings, aiding in the diagnosis and treatments of neurological disease and expanding our knowledge of the brain. However, there are many challenges inherent in managing and analyzing MRI data, due in large part to the heterogeneity of data acquisition.

Materials and methods: To address this, we have developed MRIO, the Magnetic Resonance Imaging Acquisition and Analysis Ontology.

Results: MRIO provides well-reasoned classes and logical axioms for the acquisition of several MRI acquisition types and well-known, peer-reviewed analysis software, facilitating the use of MRI data. These classes provide a common language for the neuroimaging research process and help standardize the organization and analysis of MRI data for reproducible datasets. We also provide queries for automated assignment of analyses for given MRI types.

Discussion: MRIO aids researchers in managing neuroimaging studies by helping organize and annotate MRI data and integrating with existing standards such as Digital Imaging and Communications in Medicine and the Brain Imaging Data Structure, enhancing reproducibility and interoperability. MRIO was constructed according to Open Biomedical Ontologies Foundry principals and has contributed several terms to the Ontology for Biomedical Investigations to help bridge neuroimaging data to other domains.

Conclusion: MRIO addresses the need for a "common language" for MRI that can help manage the neuroimaging research, by enabling researchers to identify appropriate analyses for sets of scans and facilitating data organization and reporting.

Keywords: Biomedical; Translational Research; biological ontologies; magnetic resonance imaging; neuroimaging; neuroinformatics.

Figure 1.

Magnetic resonance imaging assay: Our contributed terms and edits to the magnetic resonance imaging assay term from OBI (OBI:0002985). We added axioms for the pulse sequence protocol used in MRI acquisition, as well as an evaluant role used to tie back to the raw magnetic resonance image data set (OBI:0003354) output via its relation to the magnetic resonance imaging participant (OBI:0003329).

Figure 2.

MRI acquisition types: The acquisition of MRI data from a scanner is represented in MRIO in two steps: the imaging assay that describes the process and the image data set that describes the data itself. A) Different MRI acquisition types are defined according to their acquisition parameters, each with the own logical and semantic definitions. We group similar MRI acquisition types together (i.e. T1 weighted MRI is commonly used to refer to T1 weighted images acquired using spin echo (SE) or gradient echo (GRE) sequences). B) Reconstructed MRI are often the output of specific MR imaging assay sequences, while some are produced by other MRI analyses and data transformations. These classes represent the data that is used clinically or analyzed in the research setting.

Figure 3.

Magnetic resonance image data set analysis: The process of analyzing MRI data follows the format of OBI’s data transformation, applying algorithms to reconstructed magnetic resonance image data sets to produce some information content entity. The output could be a specific value or even a new, transformed magnetic resonance image data set.

Figure 4.

MRI analysis terms: MRIO contains groups of common neuroimaging analyses types, with specific analyses included as child classes. Specific MRI analyses are grouped by similar processes or the regions of the brain they analyze.

Figure 5.

DICOM to BIDS transformation: BIDS specifies directory structures and naming conventions for storing MRI data on a file system, making BIDS datasets easier for a human to read and work with than raw DICOMs. MRIO can serve as a middle layer between DICOM and BIDS datasets, with terms that map standardized DICOM tags to the naming conventions specified by BIDS. Additionally, analysis terms from MRIO – including the name of the analysis and kind of data produced – can be used to annotate analyses of BIDS data sets, which BIDS call “derivates.”