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

Engineering a data model for distributed research networks in Oncology based on FHIR

Meeting Abstract

  • Jori Kern - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Heidelberg, Germany
  • Martin Boeker - Institut für Medizinische Biometrie und Statistik, Medizinische Fakultät und Universitätsklinikum, Universität Freiburg, Freiburg, Germany
  • Daniel Brucker - German Cancer Consortium (DKTK), partner site Frankfurt, Frankfurt, Germany
  • Birgit Dlugosch - Clinical Cancer Registry of Lower Saxony, Hannover, Germany
  • Petra Duhm-Harbeck - IT Center for Clinical Research, Lübeck, Germany
  • Lars Ebert - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Cäcilia Engels - Charité – University Medicine Berlin, German Biobank Node, Berlin, Germany
  • Claudia Funke - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Thomas Ganslandt - Medizinische Fakultät Mannheim der Universität Heidelberg, Mannheim, Germany
  • Tobias Hartz - Klinisches Krebsregister Niedersachsen, Hannover, Germany
  • Gabriele Husmann - Klinikum der Johann Wolfgang Goethe-Universität, Frankfurt am Main, Germany
  • David Juárez - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Alexander Kiel - Leipzig Research Center for Civilization Diseases, University Leipzig, Leipzig, Germany
  • Björn Kroll - IT Center for Clinical Research, Lübeck, Germany
  • Peter Kuhn - Universitätsklinikum Ulm, Ulm, Germany
  • Mohamed Lambarki - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Stefan Palm - West German Cancer Center, University Hospital Essen, German Cancer Consortium, Partner Site University Hospital Essen, Essen, Germany
  • Esther Erika Schmidt - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Fabian Siegel - Heinrich-Lanz-Center for Digital Health; Mannheim University Medicine; Ruprecht-Karls-University Heidelberg, Mannheim, Germany
  • Dennis Spiegel - Kairos GmbH, Berlin, Germany
  • Christian Stephan - Kairos GmbH, Berlin, Germany
  • Deniz Tas - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany
  • Barbara Uhl - University Hospital Frankfurt, German Cancer Consortium (DKTK), partner site Frankfurt/Mainz;, Frankfurt, Germany; German Cancer Research Center (DKFZ), Heidelberg, Germany
  • Mike Wähnert - Kairos GmbH, Berlin, Germany
  • Martin Lablans - Federated Information Systems, German Cancer Research Center (DKFZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany

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; 2019. DocAbstr. 307

doi: 10.3205/19gmds026, urn:nbn:de:0183-19gmds0266

Veröffentlicht: 6. September 2019

© 2019 Kern et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe http://creativecommons.org/licenses/by/4.0/.


Gliederung

Text

Introduction: Originally designed in 2012 in the German Cancer Consortium [1], the decentral component of the Clinical Communication IT infrastructure (the so called “bridgehead”) uses a well-proven but proprietary interface. With the increased number of these bridgeheads in 22 university hospitals as part of multiple research networks and initiatives, we expect interoperability problems hampering the use of the already harmonized data. Firstly, any data transformation among differing dataset definitions can induce an unintentional loss of information. Secondly, each network uses their own bridgehead instances, which lead to needless data duplication. Thirdly, apart from the data model, the interfaces have to be adapted. Thus, we aimed to provide a modular dataset definition that is usable by multiple research networks and eliminates the need for data transformation or duplication while simultaneously representing the required complexity of different research areas. This effort is synergistic with the ongoing definition of a modular core dataset within the German Medical Informatics Initiative, which envisages an oncology module that has, however, not yet been specified in detail [2].

Methods: As a starting point, we chose to design the ADT-inspired [3] German Cancer Consortium’s data model in the well-established HL7 FHIR [4] data exchange format. To reflect real-world practices in tumor documentation and handling of biosamples, the model has been drafted as a joint effort of an ad-hoc working group of tumor documentation experts, clinicians, biobankers and software vendors. Technical questions and limitations of FHIR have been discussed at the Interoperability Forum in Berlin. Currently, we are implementing a prototype of this data model in a HAPI-based implementation [5] and CentraXX [6].

Results: We have identified a set of FHIR resources to model the patient with one or more underlying cases, which each contain a single tumor diagnosis and their respective therapies, including surgery, radiation therapy and medication statements. Data regarding samples is connected to the patient and, via a service request, also directly to the case. Disease progression information including a timestamp are stored as progresses and are part of the cases. As this does not provide enough detail to portray cancer specific information we have connected a tumor resource, defined by topographical and histological properties, to the diagnosis. Other entity specific information e.g. on cardiac, pulmonary or other diseases can also be connected to the diagnosis, which is a flexible way to model common data as well as highly specific entity-related data.

Discussion: A data model, both close to the clinical reality and to the complex challenges of scientific re-use of data, might provide ways for cooperation between different research networks not feasible before. To keep their maintainability, models should be defined using the FHIR base profiles by HL7 Germany [7] adapted for research-specific use cases. This will also improve interoperability, including the seamless integration of this dataset into the overarching effort to establish a modular core dataset within the MII. Further discussion of this first version of the data model is essential for its refinement to meet the criteria of other potential partners.

The authors declare that they have no competing interests.

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


References

1.
Lablans M, Schmidt EE, Ückert F. An architecture for translational cancer research as exemplified by the German cancer consortium. JCO clinical cancer informatics. 2018 Feb 16;1:1-8.
2.
Ganslandt T, Boeker M, Loebe M, Prasser F, Schepers J, Semler SC, Thun S, Sax U. Der Kerndatensatz der Medizininformatik-Initiative: Ein Schritt zur Sekundärnutzung von Versorgungsdaten auf nationaler Ebene. Forum der Medizin-Dokumentation und Medizin-Informatik. 2018;20(1):17.
3.
Altmann U, Katz FR, Dudeck J. A reference model for clinical tumour documentation. Studies in health technology and informatics. 2006;124:139.
4.
Arbeitsgemeinschaft Deutscher Tumorzentren. Gemeinsamer einheitlicher onkologischer Basisdatensatz ADT/GEKID. [cited 2019 July 16]. Available from: https://www.tumorzentren.de/onkol-basisdatensatz.html Externer Link
5.
Bender D, Sartipi K. HL7 FHIR: An Agile and RESTful approach to healthcare information exchange. In: Proceedings of the 26th IEEE international symposium on computer-based medical systems 2013; 2013 June 20-22; Porto, Portugal. IEEE; 2013. p. 326-331.
6.
University Health Network. HAPI FHIR - The Open Source FHIR API for Java. [cited 2019 Apr 12]. Available from: http://hapifhir.io Externer Link
7.
KAIROS GmbH. CentraXX – The expert system for treatment and research. [cited 2019 Apr 15]. Available from: https://www.kairos.de/en/ Externer Link
8.
HL7 Deutschland e.V. Basisprofil DE (Draft). [cited 2019 Apr 15]. Available from: http://ig.fhir.de/basisprofile-de/stable/ Externer Link