Transparent decision support for mechanical ventilation using visualization of clinical preferences

Area of Science:

• Critical Care Medicine
• Biomedical Engineering
• Clinical Decision Support

Background:

• Mechanical ventilation systems require clear visualization of patient data to prevent information overload.
• Existing systems often use metaphor graphics for diagnostic data, not clinical decisions.
• This study explores visualizing conflicting mechanical ventilation goals for better decision-making.

Purpose of the Study:

• To describe a clinical decision support system (CDSS) visualizing mechanical ventilation goals.
• To analyze visual patterns in mechanical ventilation settings from previous studies.
• To assess the utility of these patterns for clinical decision-making.

Main Methods:

• Described the structure and visualizations of a commercial CDSS.
• Visualized conflicting mechanical ventilation goals using a hexagonal representation.
• Performed retrospective analysis on 95 patients from two prior studies using the CDSS.
• Identified recurring patterns of hexagon symbols.

Main Results:

• Identified visual patterns for optimal ventilation, over/under ventilation, pressure support, and over oxygenation.
• These patterns were observed in both control and support modes of mechanical ventilation.
• Presented clinical examples for bedside interpretation of these patterns.

Conclusions:

• Visual patterns effectively represent the trade-offs in mechanical ventilation.
• These patterns can potentially reduce information overload for clinicians.
• Facilitates simple and rapid identification of sub-optimal mechanical ventilation settings.