Once healthcare-specific models have been captured formally (i.e., in a metamodel-based language), the application of model transformation, analysis and code generation techniques is rather straightforward. Unfortunately, in many healthcare settings valuable domain knowledge is hidden in unstructured text (e.g., in a research paper or a national report on clinical guidelines). This motivates the need for tools to annotate such texts with metadata. Such tools can be prototyped easily for one type of healthcare artifacts (e.g., for clinical guidelines or care pathways) and one purpose (e.g., for workflow management or decision support) but it is a research challenge to build a robust and generic (i.e., metamodel-independent) tool for this important type of model extraction support. This paper desribes our ongoing work to building such a tool on top of a state-of-the-art MDE platform.