Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Feb;44(1):48-58.
doi: 10.1016/j.jbi.2010.05.001. Epub 2010 May 10.

Toward an Ontology-Based Framework for Clinical Research Databases

Affiliations
Free PMC article

Toward an Ontology-Based Framework for Clinical Research Databases

Y Megan Kong et al. J Biomed Inform. .
Free PMC article

Abstract

Clinical research includes a wide range of study designs from focused observational studies to complex interventional studies with multiple study arms, treatment and assessment events, and specimen procurement procedures. Participant characteristics from case report forms need to be integrated with molecular characteristics from mechanistic experiments on procured specimens. In order to capture and manage this diverse array of data, we have developed the Ontology-Based eXtensible data model (OBX) to serve as a framework for clinical research data in the Immunology Database and Analysis Portal (ImmPort). By designing OBX around the logical structure of the Basic Formal Ontology (BFO) and the Ontology for Biomedical Investigations (OBI), we have found that a relatively simple conceptual model can represent the relatively complex domain of clinical research. In addition, the common framework provided by BFO makes it straightforward to develop data dictionaries based on reference and application ontologies from the OBO Foundry.

Figures

Figure 1
Figure 1
OBX Conceptual Model Representation. 1A. The schematic representation of the OBX core conceptual model showing high level concepts of Study Design, Study, Independent Continuant (Object), Event, Dependent Continuant (Quality, Value, or Result) and Context (for each event). 1B. UML model for the Independent Continuant Object. Human Subject, Population (grouping of human subject), Animal Subject, Biological Sample (from human or animal subject), Compound, Site, Instrument, and software are all specific objects modeled in OBX. A complete UML representation of the resulting OBX Conceptual Model with cardinality restrictions can be found at http://pathcuric1.swmed.edu/Research/scheuermann/OBX.html.
Figure 2
Figure 2
Biomaterial Transformation in OBX Model. UML model of the OBX Biomaterial Transformation class (2A) and a comparison of the Intervention domain of the CDISC-SDTM with the biomaterial transformation component of the OBX model (2B). The arrow illustrates that the three Intervention classes - Concomitant Medications, Exposure and Substance Use - in CDISC_STDM map to one Substance Merge class in OBX.
Figure 3
Figure 3
Assay in OBX Model. UML model of the OBX Assay class (2A) and a comparison of the Findings domain of the CDISC-SDTM with the Assay component of the OBX model (2B). PG, pharmacokinetics; MB, microbiology; QS, questionnaire; InclExcl, inclusion/exclusion.
Figure 4
Figure 4
Data Transformation in OBX Model. UML model of the OBX Data Transformation class (4A) and a comparison of the Findings and the Events domains of the CDISC-SDTM with the Assay and the Data Transformation component of the OBX model (2B). PG, pharmacokinetics; MB, microbiology; QS, questionnaire; InclExcl, inclusion/exclusion; AE, adverse event; Disp, disposition; Devi, deviation.
Figure 5
Figure 5
Composite Process in OBX Model. OBX captures the composite process (sequence of event) through a panel of event (Panel of Events) and the order of the events (Sequence of Events).
Figure 6
Figure 6
Study Design in OBX Model. OBX captures the study type, inclusion exclusion criteria, study descriptions, location of the study, study Arm, study period, visit and the planned events for each visit.

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback