-We discuss the applicability of using the OBI assay paradigm for representing patient questionnaires, neuropsychological tests, and neurological exams, as well to annotate data generated from these assessments. We conclude that the specification for OBI 'assay' employs a broad enough notion of evaluation to allow for these uses. However, it would be preferable to introduce subclasses of OBI 'planned process' or OBI 'assay' that explicitly addresses these types of use cases and provides clear groupings for general types of assays.
The Ontology for Biomedical Investigations (OBI) is an
integrated ontology for the description of biological and
clinical investigations [
by excited Neon atoms. Count of geese flying over a house.”
Neuropsychological tests are used to assess cognitive
domains such as attention, visual-spatial ability, memory,
executive function, and language comprehension and
expression. In addition to representing the structure of these
neuropsychological tests, it is crucial to capture the cognitive
processes and functions that they evaluate as well as the data
they produce. The neuropsychological Testing Ontology (NPT)
utilizes OBI’s assay paradigm to represent these tests [
been developed to represent both clinical measures and patient
reported outcomes (PRO) associated with the New York State
Multiple Sclerosis Consortium (NYSMSC) patient data
registry [
perception of their own functional abilities and affective states. A difficulty in using the assay framework has been in reconciling what qualifies as a physical examination and subsequent evaluation. An output of a survey in which a patient is asked to make a judgment about his or her perceived limitation in a particular limb or visual acuity may indeed qualify in this case as a sort of post-hoc physical exam which allows the evaluant to also be the evaluator. The OBI ‘selfreported handedness assessment’ supports the application of ‘assay’ to cases where a patient self-evaluates outside the context of a direct physical exam.
obtain basic demographic data fit within OBI’s account of assays. A patient responding to questions such as date of birth, marital status, insurance provider, etc. pushes one to reconsider what is being evaluated, especially since no physical examination is involved.
A related project is the development of an ontology-based
medical history module to extend a legacy clinical information
management system. This module collects, structures, and
stores data using OBO Foundry ontologies and semantic web
technology. Part of this work involves the development of an
ontological model for health history questionnaires, each
consisting of a series of questions to be answered by the patient
during a medical history interview session. While many
question answers that make up a patient's clinical history are
clearly about the patient’s body or are the result of some
physical examination of the patient, others do not seem to fit
the OBI assay framework. Family history questions are
problematic in this regard. So are questions about the existence
of a previous diagnosis, such as “Has a doctor ever told you
that you had a myocardial infarction or heart attack?” [
As it is currently defined, OBI ‘assay’ allows for a broad interpretation of what it means to physically examine or evaluate a patient. While neuropsychological tests and clinical exams can be made to fit within the assay framework, modification is required. Subjects being asked to evaluate aspects of their own bodily functioning or cognitive and affective status provides another challenge for understanding and implementing OBI ‘assay’, yet this ontological class can still provide a plausible solution. However, questionnaires, demographic information, and factual tests with no interpretive or summary outputs go beyond what can be accomplished using OBI ‘assay’. As a result, they raise interesting questions about what modifications or additions to OBI are required.
Our poster details the discussed uses of OBI ‘assay’ and summarizes the difficulties encountered. We offer alternatives and suggest the inclusion of a general set of assay and planned process types which will aid in recognizing distinctions between the various assessment strategies. Our hope is that this work will promote development in OBI and assist others who are using the assay paradigm in OBI. The Ontology for Biomedical Investigations (OBI) is an integrated ontology for the description of biological and clinical investigations. It represents the design, types of analyses and assays performed, specifications, and data generated during an investigation. Thus, it provides classes such as assay, plan specification, and measurement datum. An assay is a planned process which produces information about an evaluant. Examples of assays include: assay the wavelength of light emitted by excited neon atoms” and “count the number of geese flying over a house. Subclasses of OBI assay include laboratoryspecific examples, such as sequencing assay and metabolite profiling . However, other types include performing a clinical assessment , age measurement assay, and handedness assay . Several projects at the University at Buffalo seek to represent and annotate data generated from different types of questionnaires, forms, and tests. Each of these provide a use case that broadens the current application of OBI assay in one or more ways, possibly stretching its applicability.
OBI Assay! The class ‘assay’ is central to OBI’s purpose and utility. The paradigm for representing assays involves several key components that relate to the assay class, as shown below. The evaluant role specifies the mode of participation in the assay for the entity under study. The measurement data item represents information derived from executing an assay. The assay objective specifies the goal of the assay. Each of these is essential to representing and differentiating subtypes of assay. OBI ‘assay’ has many subclasses. Among these, the ‘analyte assay’ classes represents “classic” laboratory assays in which a substance with an analyte role is detected in a mixture, which bears the evaluant role. Other OBI assays omit naming of the analyte and its role, but follow a similar design pattern, where the evaluant role is reserved for the entity under study.
A key question is whether the material entity bearing the evaluant role can be a sentient creature, a person, who may be assayed via observation or direct questioning to yield information that is about non-material aspects of that person. A precedent for this in OBI is the ‘handedness assay’ and its subclasses, which represent assays about the handedness of a person.