Acoustic modeling of lung dynamics using bond graphs

Area of Science:

• Acoustics
• Respiratory Physiology
• Biomedical Engineering

Background:

• Understanding respiratory system acoustics is crucial for diagnosing and treating lung diseases.
• Previous models often simplify complex airway dynamics, limiting their accuracy at higher frequencies.
• Bond graph modeling offers a systematic approach to represent complex physical systems.

Purpose of the Study:

• To develop a bond graph model of the respiratory system's acoustic behavior.
• To accurately represent both upper and lower airway dynamics.
• To predict system eigenvalues as a function of parameters for a non-breathing lung.

Main Methods:

• Utilized bond graphs to model the acoustic properties of the respiratory system.
• Incorporated distributed dynamics for upper airways.
• Modeled lower airway generations using lumped resistance and compliance, terminating in ten lung segments.

Main Results:

• The developed model accurately predicts acoustic behavior for frequencies up to 8500 Hz.
• The model successfully predicts system eigenvalues based on parameters and geometry.
• The model represents a non-breathing lung state.

Conclusions:

• Bond graph modeling provides an accurate method for simulating respiratory acoustics.
• The model's high-frequency accuracy is suitable for detailed respiratory analysis.
• Future enhancements will include dynamic lung segments and turbulence modeling.