Towards a Traceable Clinical Guidelines Application: A Model Driven Approach

Resumen

Objectives: The goal of this research is to provide an overall framework to enable modelbased development of clinical guidelinebased decision support systems (GBDSSs). The automatically generated GBDSSs are aimed at providing guided support to the physician during the application of guidelines and automatically storing guideline application data for traceability purposes. Methods: The development process of a GBDSS for a guideline is based on modeldriven development (MDD) techniques which allow us to carry out such a process automatically, making development more agile and saving on human resource costs. We use UML Statecharts to represent the dynamics of guidelines and, based on this model, we use a MDD-based tool chain to generate the guide- line-dependent components of each GBDSS in an automatic way. In particular, as for the traceability capabilities of each GBDSS, MDD techniques are combined with database schema mappings for metadata manage-ment in order to automatically generate the GBDSS-persistent component as one of the main contributions of this paper. Results: The complete framework has been implemented as an Eclipse plug-in named GBDSSGenerator which, starting from the statechart representing a guideline, allows the development process to be carried out automatically by only selecting different menu options the plug-in provides. We have successfully validated our overall approach by generating the GBDSS for different types of clinical guidelines, even for laboratory guidelines. Conclusions: The proposed framework al - lows the development of clinical guidelinebased decision support systems in an automatic way making this process more agile and saving on human resource costs. © Schattauer 2010.