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

Re-engineering a monitoring data processing suite for real-time extraction, validation, and utilization of medical documentation from an intensive care patient data management system (PDMS)

Meeting Abstract

  • Henning Begerau - AMP Lab, Department of Anesthesiology & Intensive Care Medicine, University of Bonn Medical Center, Bonn, Germany; Data Integration Center (DIC), Staff Unit for Medical and Scientific Technology Development and Coordination, Commercial Directorate, University of Bonn Medical Center, Bonn, Germany
  • Felix Erdfelder - AMP Lab, Department of Anesthesiology & Intensive Care Medicine, University of Bonn Medical Center, Bonn, Germany; Data Integration Center (DIC), Staff Unit for Medical and Scientific Technology Development and Coordination, Commercial Directorate, University of Bonn Medical Center, Bonn, Germany
  • Daniel Grigutsch - AMP Lab, Department of Anesthesiology & Intensive Care Medicine, University of Bonn Medical Center, Bonn, Germany; Data Integration Center (DIC), Staff Unit for Medical and Scientific Technology Development and Coordination, Commercial Directorate, University of Bonn Medical Center, Bonn, Germany
  • Nils Dittberner - Research IT, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  • Friederike Salman - Department of Intensive Care, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  • Jan Erik Gewehr - Research IT, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  • Matthias Schmid - Institute for Biometrics, Informatics, and Epidemiology, University of Bonn, Bonn, Germany
  • Sven Zenker - AMP Lab, Department of Anesthesiology & Intensive Care Medicine, University of Bonn Medical Center, Bonn, Germany; Data Integration Center (DIC), Staff Unit for Medical and Scientific Technology Development and Coordination, Commercial Directorate, University of Bonn Medical Center, Bonn, Germany; Institute for Biometrics, Informatics, and Epidemiology, University of Bonn, Bonn, 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. 304

doi: 10.3205/19gmds182, urn:nbn:de:0183-19gmds1828

Published: September 6, 2019

© 2019 Begerau 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

Objectives: Secondary use of medical data for scientific purposes typically relies on extraction, transformation, and loading (ETL) processes. If extracted data is to be used for online clinical decision support (CDS) along with post-hoc analysis, as in the Algorithmic Surveillance in Intensive Care (ASIC) use case of the Smart Medical Information Technology for Healthcare (SMITH) consortium (BMBF grant 01ZZ1803Q, medical informatics initiative) [1], the full ETL process either has to be executed continuously or separate data versioning, caching, and change detection mechanisms must be established, requiring development and maintenance effort. The second approach was chosen for Dräger Integrated Care Manager (ICM, Dräger Medical Deutschland GmbH, Lübeck, Germany) PDMS data as an acquisition, processing and visualization suite, AcuWave [2], [3], is already established for high frequency monitoring data and thus redundancies could be minimized.

Methods: First, a requirement and delta analysis was performed. While PDMS data is also largely modeled as timeseries data, challenges requiring low-level re-engineering were identified: unlike monitoring data, being acquired once, PDMS data can be modified retrospectively by clinicians or subsystems with some values (e.g. laboratory results) being routinely assigned timestamps days in the past. Automated detection of changes over time and auditable versioning, providing the complete history of stored data, were identified as key requirements to be implemented.

Online data extraction was achieved via a bidirectional ICM interface (Clinical Application Interface, CLAPP). Each update cycle requires re-extraction and delta detection of the full data range as CLAPP does not provide methods to load changed data only. Revision fields in the persistence layer were added and the retrieval modules modified to accommodate the additional versioning information.

Data extraction was focused on achieving full coverage of the ASIC use case dataset (approximately 100 items), including vital signs, quantitative parenteral medication, and procedural information such as patient positioning.

The setup was tested and evaluated for correctness, stability, and performance. To prove portability, AcuWave was transferred to and tested at a second SMITH site (University Medical Center Hamburg-Eppendorf, UKE).

Results: Results were validated by comparison of extracted data in AcuWave and ICM and proved correct and stable. Using a single CLAPP instance allowed for an update frequency of approximately one patient/min leading to an overall latency of 45 minutes for the Bonn ASIC configuration, additional measurements will be necessary to reduce this to an acceptable level.

Introducing the developed infrastructure at UKE was successful with parameterization and adaptation to the local ICM configuration being currently under way.

Conclusion: ICM integration via CLAPP into AcuWave required significant re-engineering regarding the dynamic nature of medical documentation compared to monitoring data. Existing components could be reused to view and analyze extracted data, including online updates, improving the efficiency of the validation process. Additionally, all data is made available via the state-of-the-art realtime AcuWave REST API to downstream CDS systems. Performance improvements will be required for low-latency, site-wide coverage and are currently discussed with the vendor.

In summary, integration of realtime PDMS data proved efficient and effective suggesting AcuWave usage in even more general settings.

The authors declare that they have no competing interests.

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


References

1.
Winter A, Stäubert S, Ammon D, Aiche S, Beyan O, Bischoff V, Daumke P, Decker S, Funkat G, Gewehr JE, de Greiff A, Haferkamp S, Hahn U, Henkel A, Kirsten T, Klöss T, Lippert J, Löbe M, Lowitsch V, Maassen O, Maschmann J, Meister S, Mikolajczyk R, Nüchter M, Pletz MW, Rahm E, Riedel M, Saleh K, Schuppert A, Smers S, Stollenwerk A, Uhlig S, Wendt T, Zenker S, Fleig W, Marx G, Scherag A, Löffler M. Smart Medical Information Technology for Healthcare (SMITH). Methods Inf Med. 2018 Jul;57(S 01):e92-e105.
2.
Zenker S, Begerau H, Thull B. Modular decoupling of data acquisition from analysis and visualization in acute care monitoring to facilitate evaluation and implementation of new derived parameters and visualization techniques: concept and prototype implementation. In: European Society for Computing and Technology in Anaesthesia and Intensive Care (ESCTAIC) Annual Meeting; 2011; Nürnberg. Amsterdam: Springer Netherlands; 2012; p. 232-233.
3.
Begerau H, Oremek M, Hoeft A, Zenker S. The AcuWave Software Suite: a modular analysis and visualisation tool to facilitate the evaluation of derived parameters for researchers and clinicians in acute care. In: Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie, Hrsg. 62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Oldenburg, 17.-21.09.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocAbstr. 254.