Related Experiment Videos
Evaluating medical expert systems: what to test and how?
1National Heart and Lung Institute, London, UK.
Medical Informatics = Medecine Et Informatique
|July 1, 1990
Summary
Rigorous evaluation of medical expert systems (MES) is crucial for improving healthcare. A two-stage approach, including laboratory and field testing, is proposed to ensure unbiased assessment of MES impact in clinical practice.
Area of Science:
- Health Informatics
- Medical Artificial Intelligence
- Clinical Decision Support
Background:
- Medical expert systems (MES) show promise for enhancing healthcare delivery.
- Despite their potential, few MES undergo rigorous evaluation, limiting their routine clinical adoption.
Purpose of the Study:
- To propose a structured, two-stage evaluation framework for medical expert systems.
- To emphasize the importance of unbiased assessment for clinical integration and impact analysis.
Main Methods:
- Laboratory testing involving prospective users and implementation experts.
- Field testing designed for unbiased evaluation of system usage and its effects on healthcare encounters.
Main Results:
- User and expert perspectives are vital during laboratory evaluation.
- Field testing must objectively measure MES integration into clinical workflows and its impact on care structure, process, and outcomes.
Conclusions:
- A phased evaluation (laboratory then field) is essential for robust MES assessment.
- Objective evaluation is key to encouraging the development and adoption of effective medical expert systems.