Related Experiment Video
Updated: Apr 26, 2026
06:45
Automated Joint Space Detection Improves Bone Segmentation Accuracy
Published on: November 28, 2025
333
Computer-based diagnostic expert systems in rheumatology: where do we stand in 2014?
Hannes Alder1, Beat A Michel1, Christian Marx2
1Department of Rheumatology, University Hospital Zurich, Gloriastrasse 25, 8091 Zurich, Switzerland.
International Journal of Rheumatology
|August 13, 2014
Summary
Expert systems show promise for diagnosing rheumatic diseases early, but validation and real-world application need improvement. This review highlights key features for future successful development in rheumatology.
Area of Science:
- Rheumatology
- Medical Informatics
- Artificial Intelligence in Medicine
Background:
- Early detection and targeted treatment are crucial for managing rheumatic diseases and improving patient outcomes.
- Diagnosing rheumatic diseases in early stages is difficult due to overlapping symptoms and signs.
- Expert systems offer decision support in medicine, with applications developed across various specialties.
Purpose of the Study:
- To review the development and performance of expert systems in rheumatology.
- To provide a comprehensive overview of the current landscape of diagnostic expert systems in the field.
- To identify characteristics that could enhance the success of expert systems in clinical practice.
Main Methods:
- A systematic literature search was conducted using Medline, Embase, and Cochrane Library databases.
- The review focused on expert systems designed for rheumatology applications.
- Performance evaluations of identified expert systems were analyzed.
Main Results:
- Twenty-five expert systems were identified, with 19 having their performance evaluated.
- Diagnostic accuracy ranged widely, with correctly diagnosed cases between 43.1% and 99.9%.
- Sensitivity and specificity varied from 62-100% and 88-98%, respectively, indicating moderate to excellent performance in some systems.
Conclusions:
- Diagnostic expert systems in rheumatology demonstrate potential with promising performance metrics.
- The validation of these systems was often insufficient.
- Despite good performance, no expert system appears to have been widely adopted in daily clinical practice.
- This review offers insights into optimal system characteristics to guide future development and improve the integration of expert systems in rheumatology.
