Related Experiment Videos
Reuse of knowledge represented in the Arden syntax.
M Shwe1, W Sujansky, B Middleton
1Knowledge Data Systems, Larkspur, CA 94939.
Proceedings. Symposium on Computer Applications in Medical Care
|January 1, 1992
Summary
Knowledge Data Systems identified shortcomings in the Arden syntax for medical expert systems. Separating factual medical knowledge from its application is crucial for knowledge reuse and sharing.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
- Knowledge Representation
Background:
- The Arden Syntax is a standard for representing medical logic in expert systems.
- Knowledge Data Systems developed a medical expert system using Arden Syntax for clinical event monitoring.
- The practical application of Arden Syntax revealed limitations in its knowledge representation capabilities.
Purpose of the Study:
- To identify and analyze the shortcomings of the Arden Syntax for medical knowledge representation.
- To propose a framework that enhances knowledge sharing and reuse in medical logic systems.
- To address the limitations stemming from Arden Syntax's procedural orientation.
Main Methods:
- Analysis of rules encoded in Arden Syntax within a clinical event monitoring system.
- Identification of knowledge representation issues related to the separation of factual knowledge and its application.
- Development of a conceptual framework for improved medical logic representation.
Main Results:
- The procedural nature of Arden Syntax hinders the separation of medical facts from their clinical application.
- This lack of separation leads to knowledge redundancy and difficulties in knowledge reuse.
- Shortcomings were observed in representing diverse medical knowledge types.
Conclusions:
- Standards for medical logic representation should enforce the separation of factual knowledge from its application.
- A proposed framework aims to facilitate greater knowledge sharing and reuse in medical informatics.
- Addressing Arden Syntax's limitations is key to advancing the development of effective medical expert systems.