Temporal detection and analysis of guideline interactions.

Area of Science:

• Medical Informatics
• Artificial Intelligence
• Clinical Decision Support

Background:

• Clinical practice guidelines (CPGs) ensure quality medical care but struggle with managing patient comorbidities.
• Interactions between CPGs are a challenge, especially when temporal aspects are ignored.
• Existing methods for CPG interaction lack temporal analysis, limiting their practical application.

Purpose of the Study:

• To develop a methodology for detecting and analyzing interactions between CPGs with a focus on the temporal dimension.
• To support physicians in managing complex treatment scenarios involving multiple diseases.

Main Methods:

• Extending an ontological model to include temporal constraints (qualitative and quantitative) for actions, goals, and effects.
• Identifying application scenarios and proposing physician support facilities for temporal interaction detection.
• Employing Artificial Intelligence temporal reasoning techniques, specifically temporal constraint propagation.

Main Results:

• A modular approach integrating temporal reasoning techniques for CPG interaction analysis.
• Successful application of the methodology to two comorbidity cases using simplified CPGs.
• Demonstration of temporal detection of interactions considering various data sources.

Conclusions:

• An innovative, temporally-aware approach to CPG interaction detection and analysis.
• First methodology in the literature to address temporal issues in CPG interactions across diverse scenarios.
• Enhanced support for physicians in managing complex patient care through temporal analysis of guidelines.