Exploring the influence of nasal vestibule structure on nasal obstruction using CFD and Machine Learning method

Area of Science:

• Biomedical Engineering
• Respiratory Medicine
• Computational Science

Background:

• Clinical observations highlight the significance of the nasal vestibule in airflow.
• Understanding nasal airflow dynamics is crucial for diagnosing and treating nasal obstruction.

Purpose of the Study:

• To analyze nasal vestibule aerodynamic characteristics using CFD.
• To identify anatomical features influencing airflow patterns via machine learning.
• To correlate aerodynamic findings with clinical observations of nasal obstruction.

Main Methods:

• Computational Fluid Dynamics (CFD) simulations were employed to detail airflow patterns.
• A novel machine learning model was developed to predict airflow based on anatomical data.
• Feature mining was conducted to pinpoint critical anatomical determinants of airflow.

Main Results:

• CFD analysis revealed two distinct nasal vestibule airflow patterns, aligning with clinical findings.
• The machine learning model successfully predicted airflow patterns from anatomical features.
• Specific anatomical features were identified as having a significant impact on respiratory function.

Conclusions:

• The study successfully integrated CFD and machine learning to elucidate nasal airflow.
• Identified airflow patterns and anatomical influences provide insights into nasal obstruction.
• Validated methods offer a foundation for further research in nasal aerodynamics and clinical applications.