Linköping University | Dept. of Biomedical Engineering | Medical Informatics  
   
Medical Informatics Linköping University


Home of MI

Staff

Research

Publications

Tools

Education

MI-seminars

Contact us


Local pages

     

Design, Implementation and Evaluation of a Knowledge-Based System to Support Ventilator Therapy Management

Nosrat Shahsavar
PhD Dissertation
Linköping University
1993

Abstract

[The thesis]

Linköping Studies in Science and Technology. Dissertations
No. 317  

Design, Implementation and Evaluation of a Knowledge-Based System to Support Ventilator Therapy Management

Nosrat Shahsavar  

Akademisk avhandling

som för avläggande av teknisk doktorsexamen vid Tekniska högskolan i Linköping kommer att offentligt försvaras i Aulan i administrationsbyggnaden, Universitetssjukhuset i Linköping, fredag den 26 november 1993. Kl 10.00. 

Abstract

A good deal of research has been directed toward developing medical knowledge-based systems to assist the health care provider in both diagnostic and management decisions. Different studies by various groups have resulted in clinically useful knowledge-based systems with great potential for supporting decision-making, but due to a lack of methodology in the clinical integration and evaluation of these systems, only a few of them are in regular use. This thesis is devoted to the following aspects of knowledge-based system development: 

  • Knowledge Representation: Representing knowledge and mimicking the decision-making behavior of domain experts is a central problem in the development of medical knowledge-based systems. The chosen representation scheme should cover all pieces of the domain knowledge. Reusability and shareability of the knowledge are other desirable features, since the development of useful knowledge bases is a very time and cost consuming process. 
  • Knowledge Acquisition: There are a variety of knowledge acquisition techniques, but all techniques are not suitable for all domains. A successful outcome requires comprehensive knowledge in the knowledge-base. The quality of expert knowledge in the knowledge-base determines the quality of the system. 
  • Knowledge-Base Maintenance: Verification and maintenance of the knowledge-base by the domain experts themselves is the main issue here. Since knowledge develops continuously, it is mandatory that the knowledge-base is updateable. The knowledge-base contents should be correct and free from redundancy and inconsistency. 
  • Integration: Integrating the system into the real environment, particularly with real patient data, constitutes a critical step, because prototype environments often differ from the clinical setting. A knowledge-based system can hardly be useful if it cannot be integrated with other applications in the real environment. 
  • Evaluation: The evaluation of medical decision-support systems is important, and it is also difficult because there is no generally accepted methodology for carrying out this evaluation. The major aspect in the evaluation of a medical knowledge-based system is to find out whether the system is safe and legal, and to study the impact of the system on patients and the organization. 

This thesis examines and discusses the aforementioned factors based on experiences from the design, development, implementation and evaluation of VentEx, a knowledge-based decision-support and monitoring system we have built and applied in ventilator therapy. Our experience covers the whole development process from the prototype to an integrated on-line system. A hybrid knowledge representation has been used and a domain-specific knowledge acquisition tool (KAVE) equipped with a simulator has been developed. Real patient data has been used to validate the knowledge-base and a study to measure the impact of the system is ongoing. Evaluation results indicate a high consensus between the doctors and VentEx according to a "gold" standard. 

  Department of Biomedical Engineering
Medical Informatics Linköping University
S-581 83 Linköping, Sweden. Linköping 1993
ISBN 91-7871-168-1 ISSN 0345-7524


Reference

Shahsavar N. Design, Implementation and Evaluation of a Knowledge-Based System to Support Ventilator Therapy Management. Ph.D. thesis, Linköping University, SE-581 85 Linköping, Sweden, 1993. Dissertation No. 317, ISBN 91-7871-168-1.

Bibtex entry

@PHDTHESIS{shahsavar93,
  author = {Nosrat Shahsavar},
  title = {Design, Implementation and Evaluation of a Knowledge-Based System
	to Support Ventilator Therapy Management},
  school = {Link\"oping University},
  year = {1993},
  address = {SE-581 85 Link\"oping, Sweden},
  note = {Dissertation No. 317, ISBN 91-7871-168-1}
}

 

Editor: Joakim Rydell
Last update: Saturday, 09-Jan-2010 11:59:00 MET
To the top of this page