Haupt-Inhalte der Seite
Exchange Semantics with the Terminology Server
Communication between partners (for example, computer systems) relies on several requirements which must be fulfilled in order to get them to understand each other:
- A connection via a physical medium
- A protocol for exchanging messages
- A grammar, that is a syntax for structuring information
- A common terminology
A common terminology defined by a controlled vocabulary is one of the key assets for semantic interoperability between partners. Semantic interoperability implies that both partners know the meaning, the concept, behind a given term used in a message and interpret and understand it in the same way. For example the marital status of a person can be denoted by the character 'm' which stands for the concept 'married'.
But how does a partner know what a term actually means? They know what it means because one of them or an independent organization has defined the term along with the concept it encodes and provides a dictionary to look it up. There are several organizations like the ISO and DIN which standardize such terms and group them into whole catalogues. For example the concept 'married' is encoded in the Marital Status catalogue of the HL7 organization. Prominent catalogues in the medical domain are the International Classification of Diseases (ICD) and the Systematized Nomenclature of Medicine - Clinical Terms (SNOMED CT).
ICW believes in standardization and has therefore developed a new Java-based web application capable of providing terms and catalogues. This product, which leverages the eHealth Framework (eHF), is called the Terminology Server.
Its core functionalities are:
- Management of standardized terms and synonyms
- Search of catalogues with flexible parameterization
- Support for localization requirements such as multiple languages
- Refinement of selections like post filtering of search results
- Support of views on catalogues
In the current release, the main usage scenario addresses the management of classifications. This includes the allocation of release independent catalogues (like the ICD-10), the search for specific entries, and the retrieval of the encoded concepts. These objectives are realized using common and widespread technologies, like Java, Spring, Hibernate and Apache Axis. The eHF provides a solid foundation for the integration of those technologies along with functionality for the data handling of catalogues. The Terminology Server defines XML dialects to handle classifications for maintenance. Once imported into the underlying database, the catalogues can be accessed through either a Java API or by Web Services using SOAP.
In future releases, a web service interface using ReST will extend the availability of integration alternatives and the interfaces themselves will be extended with new service signatures. In addition, aspects concerning the navigation and structuring of catalogues, like a table of contents and tree views, will be added and enhanced subsequently. But more importantly, further development efforts concentrate on mapping and relations between entries in catalogues. These semantic relationships will enable a better transformation of the message contents and therefore strengthen the interoperability of computer systems at the level of a meaningful information exchange.
