Methods Inf Med 2013; 52(01): 18-32
DOI: 10.3414/ME11-01-0099
Review Article
Schattauer GmbH

A Task-based Support Architecture for Developing Point-of-care Clinical Decision Support Systems for the Emergency Department

Authors

  • S. Wilk

    1   Institute of Computing Science, Poznan University of Technology, Poland
    2   Telfer School of Management, University of Ottawa, Canada
  • W. Michalowski

    2   Telfer School of Management, University of Ottawa, Canada
  • D. O’Sullivan

    2   Telfer School of Management, University of Ottawa, Canada
    3   School of Informatics, City University London, United Kingdom
  • K. Farion

    3   School of Informatics, City University London, United Kingdom
  • J. Sayyad-Shirabad

    4   Departments of Pediatrics and Emergency Medicine, University of Ottawa, Canada
  • C. Kuziemsky

    2   Telfer School of Management, University of Ottawa, Canada
  • B. Kukawka

    1   Institute of Computing Science, Poznan University of Technology, Poland
Further Information

Publication History

received: 12 December 2011

accepted: 01 September 2012

Publication Date:
20 January 2018 (online)

Preview

Summary

Objectives: The purpose of this study was to create a task-based support architecture for developing clinical decision support systems (CDSSs) that assist physicians in making decisions at the point-of-care in the emergency department (ED). The backbone of the proposed architecture was established by a task-based emergency workflow model for a patient-physician encounter.

Methods: The architecture was designed according to an agent-oriented paradigm. Specifically, we used the O-MaSE (Organization-based Multi-agent System Engineering) method that allows for iterative translation of functional requirements into architectural components (e.g., agents). The agent-oriented paradigm was extended with ontology-driven design to implement ontological models representing knowledge required by specific agents to operate.

Results: The task-based architecture allows for the creation of a CDSS that is aligned with the task-based emergency workflow model. It facilitates decoupling of executable components (agents) from embedded domain knowledge (ontological models), thus supporting their interoperability, sharing, and reuse. The generic architecture was implemented as a pilot system, MET3-AE – a CDSS to help with the management of pediatric asthma exacerbation in the ED. The system was evaluated in a hospital ED.

Conclusions: The architecture allows for the creation of a CDSS that integrates support for all tasks from the task-based emergency workflow model, and interacts with hospital information systems. Proposed architecture also allows for reusing and sharing system components and knowledge across disease-specific CDSSs.