Methods Inf Med 2017; 56(02): 180-187
DOI: 10.3414/ME15-02-0013
Paper
Schattauer GmbH

The MADE Reference Information Model for Interoperable Pervasive Telemedicine Systems [*]

Nick L. S. Fung
1   Biomedical Signals and Systems Group, University of Twente, Enschede, The Netherlands
,
Valerie M. Jones
1   Biomedical Signals and Systems Group, University of Twente, Enschede, The Netherlands
,
Hermie J. Hermens
1   Biomedical Signals and Systems Group, University of Twente, Enschede, The Netherlands
2   Roessingh Research and Development, Enschede, The Netherlands
› Institutsangaben

Funding: The MobiGuide project (http://www.mobiguide-project.eu/) has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 287811.
Weitere Informationen

Publikationsverlauf

received: 19. Februar 2016

accepted: 10. Januar 2017

Publikationsdatum:
25. Januar 2018 (online)

Preview

Summary

Objectives: The main objective is to develop and validate a reference information model (RIM) to support semantic interoperability of pervasive telemedicine systems. The RIM is one component within a larger, computer-interpretable "MADE language" developed by the authors in the context of the MobiGuide project. To validate our RIM, we applied it to a clinical guideline for patients with gestational diabetes mellitus (GDM).

Methods: The RIM is derived from a generic data flow model of disease management which comprises a network of four types of concurrent processes: Monitoring (M), Analysis (A), Decision (D) and Effectuation (E). This resulting MADE RIM, which was specified using the formal Vienna Development Method (VDM), includes six main, high-level data types representing measurements, observations, abstractions, action plans, action instructions and control instructions.

Results: The authors applied the MADE RIM to the complete GDM guideline and derived from it a domain information model (DIM) comprising 61 archetypes, specifically 1 measurement, 8 observation, 10 abstraction, 18 action plan, 3 action instruction and 21 control instruction archetypes. It was observed that there are six generic patterns for transforming different guideline elements into MADE archetypes, although a direct mapping does not exist in some cases. Most notable examples are notifications to the patient and/or clinician as well as decision conditions which pertain to specific stages in the therapy.

Conclusions: The results provide evidence that the MADE RIM is suitable for modelling clinical data in the design of pervasive tele-medicine systems. Together with the other components of the MADE language, the MADE RIM supports development of pervasive telemedicine systems that are interoperable and independent of particular clinical applications.

* Supplementary material published on our website https://doi.org/10.3414/ME15-02-0013