Healthc Inform Res.  2012 Dec;18(4):252-258. 10.4258/hir.2012.18.4.252.

Expert System Shells for Rapid Clinical Decision Support Module Development: An ESTA Demonstration of a Simple Rule-Based System for the Diagnosis of Vaginal Discharge

• Kamel Boulos MN

• KMID: 2284559
• DOI: http://doi.org/10.4258/hir.2012.18.4.252

Abstract

OBJECTIVES
This study demonstrates the feasibility of using expert system shells for rapid clinical decision support module development.
METHODS
A readily available expert system shell was used to build a simple rule-based system for the crude diagnosis of vaginal discharge. Pictures and 'canned text explanations' are extensively used throughout the program to enhance its intuitiveness and educational dimension. All the steps involved in developing the system are documented.
RESULTS
The system runs under Microsoft Windows and is available as a free download at http://healthcybermap.org/vagdisch.zip (the distribution archive includes both the program's executable and the commented knowledge base source as a text document). The limitations of the demonstration system, such as the lack of provisions for assessing uncertainty or various degrees of severity of a sign or symptom, are discussed in detail. Ways of improving the system, such as porting it to the Web and packaging it as an app for smartphones and tablets, are also presented.
CONCLUSIONS
An easy-to-use expert system shell enables clinicians to rapidly become their own 'knowledge engineers' and develop concise evidence-based decision support modules of simple to moderate complexity, targeting clinical practitioners, medical and nursing students, as well as patients, their lay carers and the general public (where appropriate). In the spirit of the social Web, it is hoped that an online repository can be created to peer review, share and re-use knowledge base modules covering various clinical problems and algorithms, as a service to the clinical community.

Keyword

Expert Systems; Knowledge Bases; Clinical Decision Support Systems; Computer-Assisted Decision Making; Software Design

MeSH Terms

Archives
Caregivers
Decision Making, Computer-Assisted
Decision Support Systems, Clinical
Expert Systems
Humans
Knowledge Bases
Peer Review
Product Packaging
Software Design
Students, Nursing
Tablets
Uncertainty
Vaginal Discharge
Tablets

Figure

• 

  Figure 1 Mockler chart for diagnosis of vaginal discharge. IUD: intrauterine device.

• 

  Figure 2 User interface enhancement: pictures (attached to parameters) are used in the system's consultation dialogues. The value of a parameter is determined by the user's answer to a question. When such a question is posed during consultation, the user will have the option to ask the Expert System Shell for Text Animation (ESTA) to 'explain the question' (the 'Explain' button) if the knowledge engineer has provided an explanation in the parameter's explanation field. For this system, the author used 'canned text' extensively to enhance ESTA's shallow explanation facility (mere rule trace via the 'Why' button).

Reference

1. Cawsey A. The essence of artificial intelligence. 1998. New York (NY): Prentice Hall.

2. Darlington K. The essence of expert systems. 2000. New York (NY): Prentice Hall.

3. El-Ghamriny M. Ghamriny's manual of clinical dermatology. 2011. 8th ed. Cairo, Egypt: Kasr El Ainy Medical School.

4. Prolog Development Center. ESTA: expert system shell for text animation [Internet]. c2012. cited at 2012 Dec 1. Broendby, Denmark: Prolog Development Center;Available from: http://www.visual-prolog.com/.

5. Kim JA. An implementation of nursing diagnosis expert system using VP-EXPERT. J Korean Soc Med Inform. 1996. 2(1):59–73.

6. Prasad R, Ranjan KR, Sinha AK. AMRAPALIKA: an expert system for the diagnosis of pests, diseases, and disorders in Indian mango. Knowl Based Syst. 2006. 19(1):9–21.

7. Asghar MZ, Khan AR, Asghar MJ. Computer assisted diagnoses for red eye (CADRE). Int J Comput Sci Eng. 2009. 1(3):163–170.

8. Khan FS, Razzaq S, Irfan K, Maqbool F, Farid A, Illahi I, et al. Dr. Wheat: a Web-based expert system for diagnosis of diseases and pests in Pakistani wheat. Proceedings of the World Congress on Engineering. 2008 Jul 2-4; London, UK.