This is version 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 . It is not the current version, and thus it cannot be edited.

Back to current version Restore this version

Modeling Clinical Guidelines in KnowWE#

SmartCare - Automated clinical guidelines#

In critical care environments important medical and economical challanges are presented by the enhancement of therapeutic quality and the reduction of costs. It even is expected that these econimical challanges will increase within the the next years. For this purpose, several clinical studies have demonstrated a positive impact of the adoption of clinical guidelines (CG).

SmartCare is a technology framework and at the same time an engineering methodology by Dräger for designing knowledge based systems. The framework’s technology basically comprises a rules engine plus one or more knowledge bases reflecting the corresponding health care process to support. The framework is highly fexible for the execution of CGs in a wide range of medical devices. Only the interface to the user (GUI) and the data interface, i.e. the connection to the host system, have to be implemented. The CG itself is formalized using KnowWE and executed by d3web-core as runtime engine within the context of the medical device.

A large variety guidelines exist in the community of medical experts for almost all areas of health care. They are developed, documented and revised by national and international organizations as there are Deutschen Gesellschaft für Kardiologie – Herz- und Kreislaufforschung e.V., Ärztliches Zentrum für Qualität in der Medizin, National Guideline Clearinghous, et cetera. A systematic methodology has been developed to either augment and utilize existing CGs or to develop a guideline with a medical domain expert.

Freely adopted from:

S Mersmann, M Dojat, SmartCare™ - Automated Clinical Guidelines in Critical Care, 16th European Conference on Artificial Intelligence, Valencia (Spain), pp 745-749, IOS-Press, 2004

Further references:

F Lellouche, J Mancebo, P Jolliet, et al., A Multicenter Randomized Trial of Computer-Driven Protocolized Weaning from Mechanical Ventilation, Am J Respir Crit Care Med, Vol 174, pp 894–900, 2006

SM Burns, S Earven, C Fisher, et al., Implementation of an institutional program to improve clinical and financial outcomes of mechanically ventilated patients: One-year outcomes and lessons learned, Crit Care Med; Vol. 31, No. 12, 2003

RL Chatburn, S Deem, Should Weaning Protocols be used with all Patients who receive Mechanical Ventilation, Respiratory Care; Vol 52 No 5, pp 609 – 621, 2007

D Schaedler, C Engel, G Elke, et al., Automatic Control of Pressure Support for Ventilator Weaning in Surgical Intensive Care Patients, Am J Respir Crit Care Med Vol 185, Iss. 6, pp 637–644, Mar 15, 2012

L Rose, JJ Presneill, L Johnston, JF Cade JF, A randomised, controlled trial of conventional versus automated weaning from mechanical ventilation using SmartCare/PS, Intensive Care Med 2008;34:1788– 1795

WimVent#

Collaborate Guideline Development - Medical treatment at its best#

Within the project CliWE5 (Clinical Wiki Environments), KnowWE is extended by plugins to allow for the collaborative development of Computer-Interpretable Guidelines (CIGs). Clinical guidelines are based on evidence-based medicine and improve patient outcome by providing standardized treatments. Their computerization allows for decision-support systems at the point of care, or even the automated application by closed-loop systems in the setting of Intensive Care Units. The goal of CliWE is to create a platform that supports the engineering of CIGs by spatially distributed domain specialists. Therefore, the graphical CIG language DiaFlux was created. Its focus lies on the direct applicability and understandability by domain specialists. By offering only a small set of intuitive language elements, the guidelines can in the best case be built and maintained by the domain specialists themselves. Currently, the extensions developed within CliWE are used in the project WiM-Vent6. Its goal is to integrate medical expertise concerning mechanical ventilation and physiological models into an automated mechanical ventilator. In the course of this project, one knowledge engineer guides and supports one domain specialists (backed up by a committee of further experts) during the knowledge engineering process. The latest version of the guideline contains 17 DiaFlux modules, that in total contain 295 nodes and 345 edges. During its development, the testing capabilities of KnowWE are extensively used. So far, about 1.100 continuous integration builds were automatically executed. Especially the empirical testing feature is applied to define and process local test cases, as well as ones that are created using external tools, e.g., a Human Patient Simulator. Those simulated patient sessions can then be replayed in KnowWE for introspecting and debugging the guideline execution. A highlighting of the taken paths within the DiaFlux models serves as an accessible means of explanation for the domain specialists.