Antje Wulff1, Birger Haarbrandt2, Erik Tute2, Michael Marschollek2, Philipp Beerbaum3, Thomas Jack3. 1. Peter L. Reichertz Institute for Medical Informatics, University of Braunschweig - Institute of Technology and Hannover Medical School, Hannover, Germany. Electronic address: antje.wulff@plri.de. 2. Peter L. Reichertz Institute for Medical Informatics, University of Braunschweig - Institute of Technology and Hannover Medical School, Hannover, Germany. 3. Department of Pediatric Cardiology and Intensive Care Medicine, Hannover Medical School, Germany.
Abstract
BACKGROUND: Clinical decision-support systems (CDSS) are designed to solve knowledge-intensive tasks for supporting decision-making processes. Although many approaches for designing CDSS have been proposed, due to high implementation costs, as well as the lack of interoperability features, current solutions are not well-established across different institutions. Recently, the use of standardized formalisms for knowledge representation as terminologies as well as the integration of semantically enriched clinical information models, as openEHR Archetypes, and their reuse within CDSS are theoretically considered as key factors for reusable CDSS. OBJECTIVE: We aim at developing and evaluating an openEHR based approach to achieve interoperability in CDSS by designing and implementing an exemplary system for automated systemic inflammatory response syndrome (SIRS) detection in pediatric intensive care. METHODS: We designed an interoperable concept, which enables an easy integration of the CDSS across different institutions, by using openEHR Archetypes, terminology bindings and the Archetype Query Language (AQL). The practicability of the approach was tested by (1) implementing a prototype, which is based on an openEHR based data repository of the Hannover Medical School (HaMSTR), and (2) conducting a first pilot study. RESULTS: We successfully designed and implemented a CDSS with interoperable knowledge bases and interfaces by reusing internationally agreed-upon Archetypes, incorporating LOINC terminology and creating AQL queries, which allowed retrieving dynamic facts in a standardized and unambiguous form. The technical capabilities of the system were evaluated by testing the prototype on 16 randomly selected patients with 129 days of stay, and comparing the results with the assessment of clinical experts (leading to a sensitivity of 1.00, a specificity of 0.94 and a Cohen's kappa of 0.92). CONCLUSIONS: We found the use of openEHR Archetypes and AQL a feasible approach to bridge the interoperability gap between local infrastructures and CDSS. The designed concept was successfully transferred into a clinically evaluated openEHR based CDSS. To the authors' knowledge, this is the first openEHR based CDSS, which is technically reliable and capable in a real context, and facilitates clinical decision-support for a complex task. Further activities will comprise enrichments of the knowledge base, the reasoning processes and cross-institutional evaluations.
BACKGROUND: Clinical decision-support systems (CDSS) are designed to solve knowledge-intensive tasks for supporting decision-making processes. Although many approaches for designing CDSS have been proposed, due to high implementation costs, as well as the lack of interoperability features, current solutions are not well-established across different institutions. Recently, the use of standardized formalisms for knowledge representation as terminologies as well as the integration of semantically enriched clinical information models, as openEHR Archetypes, and their reuse within CDSS are theoretically considered as key factors for reusable CDSS. OBJECTIVE: We aim at developing and evaluating an openEHR based approach to achieve interoperability in CDSS by designing and implementing an exemplary system for automated systemic inflammatory response syndrome (SIRS) detection in pediatric intensive care. METHODS: We designed an interoperable concept, which enables an easy integration of the CDSS across different institutions, by using openEHR Archetypes, terminology bindings and the Archetype Query Language (AQL). The practicability of the approach was tested by (1) implementing a prototype, which is based on an openEHR based data repository of the Hannover Medical School (HaMSTR), and (2) conducting a first pilot study. RESULTS: We successfully designed and implemented a CDSS with interoperable knowledge bases and interfaces by reusing internationally agreed-upon Archetypes, incorporating LOINC terminology and creating AQL queries, which allowed retrieving dynamic facts in a standardized and unambiguous form. The technical capabilities of the system were evaluated by testing the prototype on 16 randomly selected patients with 129 days of stay, and comparing the results with the assessment of clinical experts (leading to a sensitivity of 1.00, a specificity of 0.94 and a Cohen's kappa of 0.92). CONCLUSIONS: We found the use of openEHR Archetypes and AQL a feasible approach to bridge the interoperability gap between local infrastructures and CDSS. The designed concept was successfully transferred into a clinically evaluated openEHR based CDSS. To the authors' knowledge, this is the first openEHR based CDSS, which is technically reliable and capable in a real context, and facilitates clinical decision-support for a complex task. Further activities will comprise enrichments of the knowledge base, the reasoning processes and cross-institutional evaluations.
Keywords:
Archetypes; Clinical data repository; Clinical decision support systems; Health information interoperability; Pediatrics; Systemic inflammatory response syndrome; openEHR
Authors: David Rubins; Adam Wright; Tarik Alkasab; M Stephen Ledbetter; Amy Miller; Rajesh Patel; Nancy Wei; Gianna Zuccotti; Adam Landman Journal: J Am Med Inform Assoc Date: 2019-11-01 Impact factor: 4.497
Authors: Antje Wulff; Sara Montag; Bianca Steiner; Michael Marschollek; Philipp Beerbaum; André Karch; Thomas Jack Journal: BMJ Open Date: 2019-06-19 Impact factor: 2.692
Authors: Antje Wulff; Sara Montag; Nicole Rübsamen; Friederike Dziuba; Michael Marschollek; Philipp Beerbaum; André Karch; Thomas Jack Journal: BMC Med Inform Decis Mak Date: 2021-02-18 Impact factor: 2.796
Authors: Konstantin Tachkov; Antal Zemplenyi; Maria Kamusheva; Maria Dimitrova; Pekka Siirtola; Johan Pontén; Bertalan Nemeth; Zoltan Kalo; Guenka Petrova Journal: Front Public Health Date: 2022-07-14
Authors: Roberto Gatta; Mauro Vallati; Carlos Fernandez-Llatas; Antonio Martinez-Millana; Stefania Orini; Lucia Sacchi; Jacopo Lenkowicz; Mar Marcos; Jorge Munoz-Gama; Michel A Cuendet; Berardino de Bari; Luis Marco-Ruiz; Alessandro Stefanini; Zoe Valero-Ramon; Olivier Michielin; Tomas Lapinskas; Antanas Montvila; Niels Martin; Erica Tavazzi; Maurizio Castellano Journal: Int J Environ Res Public Health Date: 2020-09-11 Impact factor: 3.390
Authors: Antje Wulff; Marcel Mast; Marcus Hassler; Sara Montag; Michael Marschollek; Thomas Jack Journal: Methods Inf Med Date: 2020-10-14 Impact factor: 2.176