Report on SNePS and RTS

Electronic health record-keeping systems in the Referent Tracking paradigm use globally unique identifiers to make explicit reference to all relevant particular entities, and structure data by referring to universals and relations defined in external, commonly-used ontologies. SNePS is a general-purpose Knowledge Representation, Reasoning and Acting system with a logical language interface. We have been exploring the use of SNePS as a reasoning system that can combine (1) instance data stored in a Referent Tracking system with (2) facts from domain ontologies such as FMA, and (3) rules from a domain-independent theory such as BFO-core. This paper outlines and discusses the implementation of three individual components we have developed to interface SNePS with tools from each of these three. Together these components make up a proof-of-concept implementation of a SNePSbased ontology reasoning system for use with referent tracking. 1 Referent Tracking System Referent tracking (RT) is intended to improve electronic health recordkeeping by allowing unique reference to particular things in the world (e.g. patients, their body parts, disorders), universals (e.g. Human Being, Leg, Fracture), and the relationships that hold between and among these (e.g. ‘x is a human’, ‘y is a left leg’, ‘y is a part of x’, ‘Leg is a type of Body Part’, and so on). Each particular entity to which the system can refer has assigned to it a globally unique IUI (Instance Unique Identifer) This IUI is then used whenever reference must be made to the particular. References to particulars appear in the assertions that are stored in the system. Several different types of assertions are possible. These include: that a particular instantiates a given universal; that a particular stands in some relation to one or more other particulars; that a particular is associated with a given concept code; and that a particular has assigned to it a name. These assertions also refer to universals and relations that are defined in external ontologies such as the Foundational Model of Anatomy (FMA) [Rosse and Mejino, 2003]. Similarly, assertions that associate concept codes with particulars can and do use codes from systems such as SNOMED CT [The International Health Terminology Standards Development Organisation, 2008]. The software that is the subject of this paper uses the referent tracking prototype implementation discussed in [Shahid Manzoor and Rudnicki, 2007]. We refer to this system in the rest of this paper as The ∗This work is supported by the grant “UB Task Force for ontology-based IT support for large scale field studies in Psychiatry” from the John R. Oishei Foundation. The Foundation’s mission is to enhance the quality of life for Buffalo area residents by supporting education, healthcare, scientific research and the cultural, social, civic and other charitable needs of the community. The Foundation was established in 1940 by John R. Oishei, founder of Trico Products Corporation.