Related Experiment Videos
SMR (simulating medical reasoning): an expert shell for non-AI experts
1Faculty of Medicine, Israel Institute of Technology, Haifa.
Computer Methods and Programs in Biomedicine
|January 1, 1988
Summary
This study introduces SMR, an expert system shell empowering domain experts to build medical knowledge bases using free text. It simplifies diagnostic and therapeutic knowledge acquisition and patient data evaluation.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
Background:
- Expert systems offer potential for clinical decision support.
- Knowledge acquisition remains a bottleneck in expert system development.
Purpose of the Study:
- To present the SMR expert system shell from a domain expert's perspective.
- To detail the system's capabilities for knowledge representation, inference, and patient data evaluation.
Main Methods:
- SMR utilizes free-text knowledge bases with simplified syntax for medical terminology.
- The system supports rule formulation for diagnostic and therapeutic knowledge.
- End-users interact by entering patient data for analysis and reporting.
Main Results:
- SMR enables domain experts to directly manage knowledge bases.
- The system provides intelligible reasoning and clear recommendations.
- An expert system for diabetic patient evaluation demonstrates SMR's functionality.
Conclusions:
- SMR facilitates efficient knowledge acquisition and application in medical domains.
- The system enhances clinical decision-making by providing expert-level insights.
- SMR offers a user-friendly approach to expert system development and deployment.