1. Proceedings
  2. I3M
  3. 2023
  4. IWISH
  5. 10.46354/i3m.2023.iwish.001

Enhancing Interoperability of HL7 Resources Using Namespaces in Graph Databases 

  • a Andreas Pointner , 
  • b Christoph Praschl, 
  • c Oliver Krauss
  • a,b,c  University of Applied Sciences Upper Austria Campus Hagenberg, Softwarepark 11, Hagenberg i. M., 4232, Austria *Andreas Pointner. Email address: andreas.pointner@fh-hagenberg.at 
Cite as
Pointner, A., Praschl, C., Krauss, O. (2023). Enhancing Interoperability of HL7 Resources Using Namespaces in Graph Databases. Proceedings of the 12th International Workshop on Innovative Simulation for Healthcare (IWISH 2023).,001. DOI: https://doi.org/10.46354/i3m.2023.iwish.001

Abstract

The adoption of the FHIR (Fast Healthcare Interoperability Resources) standard has led to an exponential growth of modular healthcare data that needs to be managed efficiently. Graph databases such as Neo4j offer an effective way to store and query this data, but can become complex when dealing with FHIR resources that contain numerous extensions. We explore the use of namespaces in Neo4j graph databases to manage FHIR resources and compare it with the existing tool, CyFHIR. We demonstrate that by embedding extensions using the namespace concept, the complexity of the graph can be significantly reduced. Furthermore, we evaluate our approach on a generated dataset and show that the use of namespaces in Neo4j outperforms CyFHIR conventional methods for storing FHIR resources in graph databases. Our findings suggest that the use of namespaces can be a valuable addition to Neo4j graph databases for managing complex FHIR resources.

References

  1. Chen, P. P. S. (1976). The entity relationship model toward a unified view of data. ACM transactions on database systems (TODS), 1(1):9–36.
  2. Fette, G., Kaspar, M., Liman, L., Ertl, M., Krebs, J., and Puppe, F. (2019). Implementation of a hl7-cql engine using the graph database neo4j. Studies in health technology and informatics, 267:46–51.
  3. hapifhir (2023a). hapi-fhir. https://github.com/hapifhir/hapi-fhir. [Online; accessed 3. Apr. 2023].
  4. hapifhir (2023b). hapi-fhir-structures-r4 6.2.2. https://mvnrepository.com/artifact/ca.uhn.hapi.fhir/ hapi-fhir-structures-r4/6.2.2. [Online; accessed 3. Apr. 2023].
  5. HL7 FHIR (2023a). Hl7 core profile austrian patient - fhir v4.0.1. https://fhir.hl7.at/r4 coreballot_2021_01/StructureDefinition-austrianpatient.html. [Online; accessed 3. Apr. 2023].
  6. HL7 FHIR (2023b). Patient - FHIR v4.0.1. https://hl7.org/fhir/R4/patient.html. [Online; accessed 3. Apr. 2023].
  7. Kharwar, R. (2022). Getting FHIR’ed up with a Graph Database(neo4j). Medium.
  8. Kumar, M. (1988). World geodetic system 1984: A modern and accurate global reference frame. Marine Geodesy, 12(2):117–126.
  9. neo4j (2023). Awesome Procedures On Cypher (APOC) - Neo4j Labs. [Online; accessed 21. Jun. 2023].
  10. Optum (2023). CyFHIR. https://github.com/Optum/CyFHIR. [Online; accessed 3. Apr. 2023].
  11. Pointner, A., Krauss, O., Erhard, A., Schuler, A., and Helm, E. (2023). Multi-Perspective Process Mining Interfaces for HL7 AuditEvent Repositories: XES and OCEL. In dHealth. in press
  12. Pointner, A., Praschl, C., and Krauss, O. (2021). AIST Neo4j.
  13. Pointner, A., Praschl, C., and Krauss, O. (2022). Towards modelling namespaces in graph databases. In Proceedings of the European Modeling & Simulation Symposium,
    EMSS. CAL-TEK srl.
  14. synthetichealth (2023). Synthetic Patient Population Simulator. https://github.com/synthetichealth/synthea. [Online; accessed 3. Apr. 2023].
  15. Teorey, T. J., Yang, D., and Fry, J. P. (1986). A logical design methodology for relational databases using the extended entity-relationship model. ACM Comput. Surv.,
    18(2):197–222.