On locating the obstruction in the upper airway via numerical simulation

Area of Science:

• Fluid dynamics
• Biomedical engineering
• Computational modeling

Background:

• Understanding upper airway (UA) fluid dynamics is challenging due to complex 3D geometry and flow variations.
• Experimental measurement of instantaneous airflow velocity and pressure in the UA is difficult.
• Direct numerical simulation (DNS) offers a way to resolve detailed flow properties.

Purpose of the Study:

• To develop and validate a DNS solver using the lattice Boltzmann method (LBM) for analyzing human UA airflow.
• To investigate airflow patterns during inspiration and expiration in patient-specific UAs.
• To identify methods for accurately locating UA obstructions.

Main Methods:

• Developed a DNS solver utilizing the lattice Boltzmann method (LBM).
• Applied the solver to two patient-specific UAs reconstructed from CT scan data.
• Analyzed airflow during both inspiration and expiration phases.
• Utilized time-averaged pressure gradients (∂p/∂z) and second spatial derivatives (∂(2)p/∂z(2)) to identify obstructions.

Main Results:

• The DNS-LBM solver successfully simulated airflow in patient-specific UAs.
• The method accurately captured detailed flow characteristics.
• The second spatial derivative of pressure (∂(2)p/∂z(2)) effectively pinpointed the exact location of UA obstructions.
• The first spatial derivative of pressure (∂p/∂z) helped identify regions of obstruction.

Conclusions:

• The developed DNS-LBM solver is a powerful and accurate tool for studying UA airflow.
• This computational approach can precisely locate upper airway obstructions.
• The findings have implications for diagnosing and understanding airway diseases.