This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Difference between revisions of "RIMBAA: UMCG"

From HL7Wiki
Jump to navigation Jump to search
Line 3: Line 3:
 
The UMCG approach is the MR-MO-MS path. Some RMIMs with some CMETs make up the Information Model on witch the services are based, it is also used as internal models in the applications. Existing (legacy) systems are mapped using the AR-MO-MS path.
 
The UMCG approach is the MR-MO-MS path. Some RMIMs with some CMETs make up the Information Model on witch the services are based, it is also used as internal models in the applications. Existing (legacy) systems are mapped using the AR-MO-MS path.
  
[[Image:UMCGmatrix.png]]
+
[[Image:Umcg tech matrix.jpg]]
  
 
Voor de MR-MO stuk gebruiken we ORM.
 
Voor de MR-MO stuk gebruiken we ORM.

Revision as of 13:43, 12 January 2009

This page uses terminology as defined in this whitepaper.

The UMCG approach is the MR-MO-MS path. Some RMIMs with some CMETs make up the Information Model on witch the services are based, it is also used as internal models in the applications. Existing (legacy) systems are mapped using the AR-MO-MS path.

Umcg tech matrix.jpg

Voor de MR-MO stuk gebruiken we ORM.

  • voor ORM gebruiken we de Java Persistance API
  • maar ook in .Net hebben we zoiets (ik weet niet meer hoe die heet, geen bekende)

Voor MO-MS gebruiken we webservices, JAX-WS (Java6 implementatie van WebServices) en die gebruikt weer intern JAXB (Java API for XML Binding).

Voor AR-MO gebruiken we nu custom code. De AR onsluiten we via XML DataSets en die mappen we weer op de MO. Voor AR-MO willen we wel graag een standaard tool gebruiken zoals Progress DataXtend Semantic Integrator (zie ook de presentatie!) of BEA AquaLogic Data Services (heb ik inmiddels ook presentatie van).

User Interfaces

UMCG generates user interfaces based on (template) R-MIMs:

  1. Starting point for the development process are R-MIMs, a template (LIM) derived/specialization of the clinical sattement model.
  2. These v3 templates are imported into "InfoPath Template Parts", which subsequently are subject to the following:
    • Tweak v3 datatype support and other XML schema issues. InfoPath itself isn't v3 data type aware.
    • Add look & feel - the HL7 models itself have no L&F,
    • Associate coded attributes with value sets for testing. Based on CTS. Using this the ComboBox/List/Checkbox lists can be displayed using the appropriate codes. Code tables exist for departments, specialties, LOINC, and Apelon (SNOMED).
    • Add WebService calls to pre-populate elements that already have a value from the clinical data repository.
  3. The "InfoPath Template Parts" are subsequently used as building blocks to create a form (a user interface).

The relationship between [as well as internal to] the "InfoPath Template Parts" (act relationships in the original v3 model), are not visible to the end user. The layout of the document implicitely documents (informs) the user about the relationships.