In this deliverable, we describe the knowledge engineering process used to formalize the clinical practice guidelines implemented in CAPABLE. More precisely the following steps will be illustrated: Choice of the guidelines - selecting the most compatible guidelines for the system. Text processing mechanism - detecting recommendations and translating them into a set of computer-interpretable rules. Analysis of each recommendation - detecting what raw data type was involved, which vocabulary is mostly suitable for coding, and what abstractions need to be performed on this raw data. This allows for achieving clearer representation, at the same time producing reusable pieces of knowledge. The detected raw data also represents the minimum data set needed by the CAPABLE DSS (Decision Support System), so this step has been fundamental for the development of other CAPABLE components. Translation of each recommendation into a set of computer-interpretable rules, organization of rules into a logical flow, and their representation using the Composer tool, i.e., the authoring tool developed by one of the project partners. In this step, adaptations have been implemented to allow smooth guideline implementation in different healthcare settings. Validation of the represented knowledge by simulating realistic clinical scenarios (oncologists have been involved in this phase). Particular attention has been put to multimorbidity management, since running guidelines for different pathologies may lead to contradictory recommendations, which must be resolved Handover to the implementation team and refinement iterations with the team itself. The deliverable will then provide the actual description of the formalized guidelines, through a set of flowcharts, and links to their representation in Composer.

This deliverable is a part of a project receiving funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 875052