gms | German Medical Science

64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

08. - 11.09.2019, Dortmund

Developing and implementing a health IT ontology for facilitating retrieval of health IT evaluation studies

Meeting Abstract

  • Verena Dornauer - Institute of Medical Informatics, UMIT – University for Health Sciences, Medical Informatics and Technology, Hall in Tirol, Austria
  • Maryam Ghalandari - Institute of Medical Informatics, Statistics and Epidemiology, University of Leipzig, Leipzig, Germany
  • Konrad Höffner - Institute of Medical Informatics, Statistics and Epidemiology, University of Leipzig, Leipzig, Germany
  • Franziska Jahn - Institute of Medical Informatics, Statistics and Epidemiology, University of Leipzig, Leipzig, Germany
  • Alfred Winter - Institute of Medical Informatics, Statistics and Epidemiology, University of Leipzig, Leipzig, Germany
  • Elske Ammenwerth - Institute of Medical Informatics, UMIT – University for Health Sciences, Medical Informatics and Technology, Hall in Tirol, Austria

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Dortmund, 08.-11.09.2019. Düsseldorf: German Medical Science GMS Publishing House; . DocAbstr. 94

doi: 10.3205/19gmds166, urn:nbn:de:0183-19gmds1665

Published:

©  Dornauer et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Introduction: Evidence-based health informatics needs easy access to published health IT evaluation studies. The indexing of health IT evaluation studies by using MeSH terms is not specific enough which makes retrieval difficult [1]. To solve this problem, we want to support the retrieval of health IT evaluation studies by using ontologically modelled key information from evaluation studies. The ontological approach allows semantically precise results when searching studies.

The aim of this presentation is to describe the development of the ontology and to present first results of a feasibility study.

Methods: We randomly selected 24 health IT evaluation studies and manually extracted key information such as DOI, PMID, first author, publishing journal, year published, study type, study methods and outcome criteria. We also extracted the type of intervention, its features, underlying software product, user group and the organizational unit where the study takes place for building the ontology. For example, from a sentences as “The electronic health record system is used by physicians”, we manually extracted “electronic health record system” as Application System and “physicians” as User Group. The extracted key elements were initially collected in an Excel File.

To classify the type of health IT intervention, we used an available classification of health IT applications [2]. For classifying the features of the health IT intervention, we used the EHR Functional Model of HL7 [3].

For the feasibility study, we modeled the ontology with Protégé [4] and used a locally hosted quadstore Open Link Virtuoso [5]. We queried the ontology with SPARQL [6].

Results: We are now able to search precisely for certain health IT evaluation studies. Some examples are: Which study is from a specific author or deals with specific outcome criteria (e.g. medication errors)? What type of application system is evaluated? Which features are investigated (e.g. interaction checking)? Who uses the application system in the study (physicians, nurses or other staff)? Which study type is used (e.g. RCT, quasi-experimental study, non-experimental study)? Which study methods are applied (e.g. questionnaires, document analysis)? These questions can all be answered easily by querying the ontology.

Discussion: The ontology-based retrieval of health IT evaluation studies allows retrieving semantically precise results concerning health IT evaluation studies. The ontology is flexible and can be extended if needed by adding further concepts. Another advantage over storing the studies in a SQL database is the possibility to model semantically meaningful relationships between concepts within the ontology.

Nevertheless, manual data extraction of studies for describing the studies is quite time consuming and not practicable. Therefore, at the moment, we explore (semi-)automatic methods of ontology population [7] considering rule-based and Machine Learning approaches. We will also implement a faceted search interface so that researcher and practitioners can use the application without having knowledge on Semantic Web technologies and SPARQL.

This feasibility study is part of the 3 years funded project “HITO: A Health IT Ontology”, funded by the Austrian Science Fund FWF (I 3726-N31) and the German DFG fund (WI 1605/11-1).

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.


References

1.
Dixon B, Zafar A, McGowan J. Development of a taxonomy for health information technology. Stud Health Technol Inform. 2007;129:616–20.
2.
Winter A, Haux R, Ammenwerth E, Brigl B, Hellrung N, Jahn F. Health Information Systems: Architecture and Strategies. 2nd Edition. London: Springer; 2011.
3.
ISO/HL7 10781 - Electronic Health Record System Functional Model, Release 2. 2014 [Accessed 2019 Jul 18]. Available from: http://wiki.hl7.org/images/8/83/EHRS_FM_R2_Final_ANEX_20140220_still_has_issues.pdf External link
4.
Musen MA; Protégé Team. The Protégé Project: A Look Back and a Look Forward. AI matters. 2015 Jun;1(4):4–12.
5.
OpenLink Software. OpenLink Virtuoso. 2019 [Accessed 2019 Mar 9]. Available from: https://virtuoso.openlinksw.com/ External link
6.
W3C. SPARQL Current Status. 2019 [Accessed 2019 Mar 9]. Available from: https://www.w3.org/standards/techs/sparql#w3c_all External link
7.
Maynard D, Bontcheva K, Augenstein I. Natural Language Processing for the Semantic Web. Morgan & Claypool; 2017.