Wednesday, August 30, 2006

MIE2006 – open EHR

It is the final day of MIE2006 already. The session I am attending this morning consists of two papers on aspects of openEHR archetypes. The session is very well-attended considering it is the morning after the conference gala dinner.

The first speaker, Sebastian Garde, is speaking on 'Ubiquitous information for ubiquitous computing: expressing clinical data sets with openEHR archetypes'. Sebastian describes openEHR archetypes as intuitive, formal and flexible models of clinical content. He described the conversion of 260 items from the German clinical paediatric data set to 46 archetypes.

He said that one of the problems has been that clinical data sets do not necessarily follow basic design principles, and often do not define when and how often which data item has to be captured, whereas openEHR archetypes have built-in support for defining time series, etc. The website was introduced.

He concluded that archetypes can provide better structure for clinical data sets (not replace them); archetypes are more like a maximum than a minimum data set, having a much broader scope, and sees it as possible to transform existing clinical data sets into archetypes.

In the second talk, Thilo Schuler from Germany talked about 'Towards automatically generating graphical user interfaces from openEHR archetypes'. This is a proof of concept project that uses the clinical concepts described in the previous talk. Thilo said that a new overview paper explaining the basic principles of openEHR archetypes is available on the website. Three prototypes were built, using Mozilla XUL (XML user interface language), which allows for writing of Firefox browser extensions, and using Microsoft's C#. This was a very technically-oriented presentation that I think needed some detailed knowledge of the archetype concepts and of programming.

Technorati Tags: , ,


Post a Comment

<< Home