Volltext-Downloads (blau) und Frontdoor-Views (grau)

Evaluation of Methods for Querying, Filtering and Aggregation Complex Health Care Data with regard to their Applicability in FHIR

  • Objectives FHIR (Fast Healthcare Interoperability Resources) is an emerging Stan¬dard for improving interoperability in the domain of health care. Besides offering features for achieving syntactical, semantical and organizational interoperability, it also specifies a RESTful API for searching purposes. The main objective of the following thesis revolves around investigating open challenges and limitations of the so-called FHIR Search Framework. Methods A variety of operations for searching in FHIR resources, including all search interactions, definitions of search parameters, search parameter types and advanced search concepts are described. Over the course of the thesis, a quality model based on ISO 25010 is established. It serves as the foundation for determining if the FHIR Search Framework is well-suited to cover the Information needs of its users. An analysis of completeness involving the measures defined in the quality model forms the main contribution. The primary discussion of the research questions is concluded by proposing a graph model for determining reachability between FHIR resources, essentially mirroring the chaining and reverse chaining functionality. Using well-known classes for expressiveness in graphs, the thesis assess to which degree a graph search can be formulated with the currently defined capabilities. Results From a functional perspective the FHIR Search Framework can be considered well-suited. Practical limitations should be minimal, grounded on the fact that extensive coverage of the lowest expressiveness classes, RPQs and 2RPQs, can be achieved. Severe gaps where identified only in the support of C(2)RPQs and Data Path Queries. Additionally, ideas for improving non-functional aspects are introduced to support developers in learning the Standard and testing their implementations. Conclusion The evaluation of the FHIR Search Framework showed promising re¬sults in terms of functional completeness. Yet, the Standard is still evolving, and certain parts of the Search API are neither well-known nor implemented widely. A discussion is to be held if the specification should cover more sophisticated aspects that result from the gaps which were identified.
Author:Alexander Zautke
Referee:Heide Faeskorn-Woyke
Document Type:Master's Thesis
Publishing Institution:Hochschulbibliothek der Technischen Hochschule Köln
Granting Institution:Technische Hochschule Köln
Date of Publication (online):2018/11/12
GND-Keyword:Medizinische Informatik
Tag:FHIR; Graph Query Expressiveness; Health Care
Institutes:Informatik und Ingenieurwissenschaften (F10) / Fakultät 10 / Institut für Informatik
CCS-Classification:H. Information Systems
Dewey Decimal Classification:000 Allgemeines, Informatik, Informationswissenschaft
JEL-Classification:L Industrial Organization
Open Access:Open Access
Licence (German):License LogoCreative Commons - CC BY - Namensnennung 4.0 International