Applying artificial intelligence to clinical guidelines: the GLARE approach.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Clinical guidelines (GL) are essential for evidence-based practice but often lack flexibility and user-friendliness.
• Existing systems struggle with dynamic adaptation and complex temporal reasoning.

Purpose of the Study:

• To present GLARE, a domain-independent system for acquiring, representing, and executing clinical guidelines.
• To leverage Artificial Intelligence (AI) for enhanced flexibility, user-friendliness, and decision support in guideline management.

Main Methods:

• Designed a high-level, user-friendly knowledge representation language for clinical guidelines.
• Developed a user-friendly acquisition tool with physician support and a patient-specific execution tool.
• Employed AI techniques for automatic resource-based adaptation, temporal constraint reasoning, decision support, and model-based verification.

Main Results:

• GLARE facilitates the acquisition, representation, and execution of clinical guidelines.
• AI integration enhances system flexibility, user-friendliness, and provides decision support.
• Developed system-independent methods for guideline adaptation, temporal reasoning, decision making, and verification.

Conclusions:

• The GLARE system effectively integrates AI for clinical guideline management.
• The developed AI-driven methods offer significant advancements in guideline adaptation and execution.
• GLARE's system-independent techniques have broader applicability to other guideline management systems.