| Autor: |
Anywar M; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany., Macedo M; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany., Pazmino S; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany., Bronsch T; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany., Kinast B; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany., Kock-Schoppenhauer AK; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany.; IT Center for Clinical Research, Universität zu Lübeck, Lübeck, Germany., Schreiweis B; Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Germany.; Medical Data Integration Center, University Hospital Schleswig-Holstein, Kiel/Lübeck, Germany. |
| Abstrakt: |
This paper explores the challenges and lessons learned during the mapping of HL7 v2 messages structured using custom schema to openEHR for the Medical Data Integration Center (MeDIC) of the University Hospital, Schleswig-Holstein (UKSH). Missing timestamps in observations, missing units of measurement, inconsistencies in decimal separators and unexpected datatypes were identified as critical inconsistencies in this process. These anomalies highlight the difficulty of automating the transformation of HL7 v2 data to any standard, particularly openEHR, using off-the-shelf tools. Addressing these anomalies is crucial for enhancing data interoperability, supporting evidence-based research, and optimizing clinical decision-making. Implementing proper data quality measures and governance will unlock the potential of integrated clinical data, empowering clinicians and researchers and fostering a robust healthcare ecosystem. |
