Informaticopia

Tuesday, October 25, 2005

Architecture for Knowledge-Based and Federated Search of Online Clinical Evidence

Architecture for Knowledge-Based and Federated Search of Online Clinical Evidence

This interesting paper has just been published in the Journal of Medical Internet Research.

The authors Enrico Coiera, Martin Walther, Ken Nguyen and Nigel H Lovell from the University of New South Wales, Sydney, Australia, clearly identify some of issues for clinicians in keeping up with the volumes of relevant literature and difficulties in searching across multiple databases for relevant information.

Their approach uses federated meta search tools, with a variety of filters (e.g. disease processes) into a system they have called "Quick Clinical (QC) which guides the user through the different profiles and keywords into queries which can be sent to a variety of different resources and their individual syntax in an effort to identify key resources in the shortest time possible. The interface and architecture which underly QC are described and their integration into Unified Query Language (UQL) and Unified Response Language (URL) using XML outlined.

I was slightly confused by their explanation of how "specific wrappers translate a UQL query into the native query language and format of the source", but I'm sure this will make sense to the more technically advanced. Some initial technical and user evaluation data is presented, but I think we will have to wait to see whether the product is scaleable and whether it is more widely adopted before decisions can be made about it's possible utility in clinical practice.


Technorati Tags:

0 Comments:

Post a Comment

<< Home