AUTOMATIC AIRWAY ANALYSIS FOR GENOME-WIDE ASSOCIATION STUDIES IN COPD

Area of Science:

• Medical Imaging
• Pulmonary Medicine
• Computational Biology

Background:

• Chronic Obstructive Pulmonary Disease (COPD) is a major health burden.
• Large-scale genetic studies require robust methods for extracting disease phenotypes.
• Current imaging analysis for airways in COPD is limited for genetic epidemiology.

Purpose of the Study:

• To develop and validate an image analysis pipeline for airway phenotype extraction.
• To enable large-scale genetic and epidemiological studies, including genome-wide association studies (GWAS), in COPD.
• To identify imaging-based phenotypes associated with clinical outcomes in COPD.

Main Methods:

• Utilized scale-space particles for dense sampling of intraparenchymal airway locations in high-resolution CT scans.
• Employed constrained energy minimization and connected components filtering for airway point identification and clustering.
• Applied a second-order derivative edge detector for airway wall detection and computed wall thickening (Pi10,WA%) and luminal remodeling (P%) phenotypes.

Main Results:

• Successfully extracted airway phenotypes from 2,500 high-resolution CT scans.
• Validated the airway extraction technique for accuracy and reproducibility.
• Demonstrated the association of extracted airway phenotypes with clinical outcomes relevant to COPD.

Conclusions:

• The developed image pipeline is suitable for large-scale genetic and epidemiological studies in COPD.
• This method facilitates the identification of novel genetic associations with COPD through GWAS.
• The extracted airway phenotypes provide valuable insights into COPD pathogenesis and progression.