This paper reports on a case-study of applying various publicly available resources (lexical, terminological and ontological) for medical recognition tasks, that is, for identifying medical entities in the analysis of clinical practice guideline texts. The paper provides a methodological support that systematises the entity recognition task in the medical domain. Preliminary analysis shows that many of the medical linguistic expressions describing goals and intentions in natural language are included in the current terminological resources. So, these resources can be used as a means of disambiguating and structuring this type of expressions, with the final aim of indexing guideline repositories for efficient searching.