An ontological framework for representing clinical knowledge in decision support systems

In the last decades, clinical evidence and expert consensus have been encoded into advanced Decision Sup- port Systems (DSSs) in order to promote a better integration into the clinical workflow and facilitate the automatic provi- sion of patient specific advice at the time and place where decisions are made. However, clinical knowledge, typically expressed as unstructured and free text guidelines, requires to be encoded into a computer interpretable form suitable for being interpreted and processed by DSSs. For this rea- son, this paper proposes an ontological framework, which en- ables the encoding of clinical guidelines from text to a formal representation, in order to allow querying, advanced reason- ing and management in a well defined and rigorous way. In particular, it jointly manages declarative and procedural as- pects of a standards based verifiable guideline model, named GLM-CDS (GuideLine Model for Clinical Decision Support), and expresses reasoning tasks that exploit such a represented knowledge in order to formalize integrity constraints, business rules and complex inference rules.