The Asgaard project: a task-specific framework for the application and critiquing of time-oriented clinical...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Clinical Decision Support

Background:

• Clinical guidelines are dynamic knowledge schemata refined by healthcare providers.
• Current systems lack robust support for non-designer care providers applying guidelines.
• Dynamic instantiation and refinement of guidelines pose challenges for automated systems.

Purpose of the Study:

• To develop methods supporting care providers in applying clinical guidelines.
• To enable recognition of provider intentions and critique of actions against guidelines and patient records.
• To create a machine-readable guideline representation and an automated acquisition tool.

Main Methods:

• Investigating tasks for guideline application by non-designers.
• Developing methods for intention recognition and action critique.
• Representing guidelines using a task-specific ontology and the Asbru language.
• Utilizing the PROTEGE-II framework for automated guideline acquisition.

Main Results:

• Identified domain-specific knowledge requirements for guideline application methods.
• Presented Asbru, a machine-readable language for guideline representation and annotation.
• Introduced an automated tool for clinical guideline acquisition based on a shared ontology.

Conclusions:

• Asbru facilitates dynamic application and critique of clinical guidelines.
• Automated tools can aid in the acquisition and representation of clinical guidelines.
• The Asgaard project provides methods for supporting care providers in complex clinical scenarios.