A Knowledge Authoring Tool for Clinical Decision Support
- 210 Downloads
Anesthesiologists in the operating room are unable to constantly monitor all data generated by physiological monitors. They are further distracted by clinical and educational tasks. An expert system would ideally provide assistance to the anesthesiologist in this data-rich environment. Clinical monitoring expert systems have not been widely adopted, as traditional methods of knowledge encoding require both expert medical and programming skills, making knowledge acquisition difficult. A software application was developed for use as a knowledge authoring tool for physiological monitoring. This application enables clinicians to create knowledge rules without the need of a knowledge engineer or programmer. These rules are designed to provide clinical diagnosis, explanations and treatment advice for optimal patient care to the clinician in real time. By intelligently combining data from physiological monitors and demographical data sources the expert system can use these rules to assist in monitoring the patient. The knowledge authoring process is simplified by limiting connective relationships between rules. The application is designed to allow open collaboration between communities of clinicians to build a library of rules for clinical use. This design provides clinicians with a system for parameter surveillance and expert advice with a transparent pathway of reasoning. A usability evaluation demonstrated that anesthesiologists can rapidly develop useful rules for use in a predefined clinical scenario.
KeywordsDecision support Expert system Situation awareness Usability Anesthesia Monitoring Knowledge resources
This project was supported by the Canadian Institutes of Health Research. We would like to thank all the experts who have assisted and advised us with developing this application. We would also like to thank the volunteers who participated in the usability evaluation for their ideas and feedback and Elizabeth Cheu for reviewing and editing the manuscript.
A technique that is based on the relative probability of an event given the probabilities of associated events in the network; employs Bayes’ theorem.
A significant point of change in a physiological parameter found by using trend detection algorithms.
An expert system that assists and potentially enhances a human’s ability to make decisions.
In a computer graphical user interface, the process of clicking an object and then holding down and dragging it to another location before releasing.
A software-based system that integrates a mass of information based on rules or processing performed within the software program to supply expert knowledge about a specific field.
Reasoning methodology producing a definite conclusion based upon vague, ambiguous, imprecise, noisy or missing input information.
A user interface (part of the program the user interacts with) which contains the graphic elements: icons, text, labels, buttons, etc.
The study of the probability that a human will correctly perform a task and those factors related to this probability.
In terms of Java, a specific object of a Java class.
The ability to communicate and operate with different hardware and software systems.
An object-oriented programming language.
The encoded knowledge for an expert system. In a rule-based expert system, a knowledge base incorporates definitions of attributes and rules along with control information; a store of factual and heuristic data.
The process of building a knowledge base through encoding the human expert knowledge in the computer language being used.
The person encoding the knowledge into the knowledge base.
A rule written in the language of the knowledge base, which is used by a decision support system or expert system to analyze data and make decisions.
A method of artificial intelligence in which patterns are found within the data to enable the application to slowly learn how different pieces of data are interconnected.
A method of artificial intelligence which is used to solve tasks through a network of simple processing units, model similar to biological neuron networks.
Design methodology that breaks down problems into objects rather than procedures.
Any project whose source code is made available for use or modifications as users or developers see fit.
The measure of the spread of a parameter’s values.
Decision making that does not change. Given the same input, the same decision will always be made.
A parameter whose values remains known and unchanged throughout a process.
In computer programming, the conforming rules of the code, which must be followed for the code to be valid in the computer language used.
A computer process which identifies changing trends of the physiological parameters being monitored. Significant trend changes in a parameter are recorded as change points.
A change in patient’s ventilation, outside defined normal limits, within anesthesia.
This is where all the facts in the decision support engine are located.
A general purpose mark-up language used as the format for configuration files.
- 1.Gaba DM, Fish KJ, Howard SK Crisis management in anesthesiology. New York: Churchill Livingstone; 1994Google Scholar
- 2.Ansermino JM. Patient safety: technology and design can help. B C Med J 2006; 48(7): 339–341Google Scholar
- 10.Yang P, Dumont G, Lim J, Ansermino JM. Adaptive change point detection for respiratory variables. In: Proceedings of the EMBS 27th Annual International Conference, Shanghai, China, Sep 1–4, 2005; 780–783Google Scholar
- 11.Yang P, Dumont G, Ansermino JM. An adaptive cusum test based on a hidden semi-Markov model for change detection in non-invasive mean blood pressure trend. In: Proceedings of the EMBS 28th Annual International Conference, New York City, USA, Aug 30–Sep 3, 2006; 3395–3398Google Scholar
- 13.Swain AD, Guttmann HE. Handbook of human-reliability analysis with emphasis on nuclear power plant applications. Albuquerque (US-NM): Sandia National Labs; 1983Google Scholar
- 16.Sukuvaara T, Makivirta A, Kari A, Koski E. An intelligent intensive care alarming system. Proc Comput Cardiol 1989; 225–228Google Scholar
- 17.Mersmann S, Dojat M. SmartCare – automated clinical guidelines in critical care. In: Mantara RL, Saitta L, eds, 16th European Conference on Artificial Intelligence, Valencia, Spain, Aug 22–27, 2004; 745–749Google Scholar
- 18.Knublauch H, Musen MA, Rector AL. Editing description logic ontologies with the protégé OWL plugin. In: Haarslev V, Möller R, eds, Proceedings of the International Workshop on Description Logics, June 6–8, 2004, Whistler, BC, Canada. Central Europe: CEUR Workshop Proceedings; 2004Google Scholar
- 19.IEEE-SA Standards Board. Health informatics—point-of-care medical device communication—part 10201: domain information model. 1st ed. USA: ISO/IEEE; 2004Google Scholar
- 20.IEEE-SA Standards Board. Health informatics—point-of-care medical device communication—part 10101: nomenclature. 1st ed. USA: ISO/IEEE; 2004Google Scholar
- 21.Drools Documentation. Accessed online at http://downloads.jboss.com/drools/docs/220.127.116.1193.GA/html_single/index.html Nov 9, 2007
- 24.Elevitch FR SNOMED CT: electronic health record enhances anesthesia patient safety. J Am Assoc Nurse Anesth 2005; 73(5): 361–366Google Scholar