Architecture and services for formalising and evaluating care actions from computer-interpretable guidelines

Abstract

Listen

Translate article

Computer-interpretable guidelines (CIGs) can be understood as an alternative to deal with the various limitations of paper-based clinical guidelines and to put forward new possibilities for healthcare improvements. The effort spent in evaluating and adapting recommendations described in CIGs can be reduced if formalisation and modularisation of domain knowledge are employed. The approach presented in this paper considers that the content of CIGs and the adaptation rules are two distinct elements that need to be associated at run-time. In a first step, we propose to formalise the recommendations that are presented to healthcare professionals, named here ‘care actions’, in order to produce the set of elements that need to be adapted. In a second step, we exploit the medical background, defining and applying adaptation rules to generate a personalised treatment plan for the patient. The approach was applied to 21 CIGs to identify the potential and limitations of the implemented prototypes.