Bronchiectasis in patients with chronic obstructive pulmonary disease: AI-based CT quantification using the bronchial...

Area of Science:

• Pulmonary Medicine
• Radiology
• Artificial Intelligence in Healthcare

Background:

• Chest CT is standard for bronchiectasis evaluation but lacks precise quantification.
• Accurate measurement of bronchiectasis extent is crucial for predicting clinical outcomes in COPD.

Purpose of the Study:

• To develop and validate an AI-based method for quantifying bronchiectasis using bronchial tapering ratio on chest CT.
• To assess the association between AI-quantified bronchiectasis and clinical outcomes (mortality, acute exacerbations) in COPD patients.

Main Methods:

• AI analysis of bronchial tapering ratio on baseline CT scans from two COPD cohorts.
• Calculation of a whole-lung bronchiectasis score based on abnormal airway tapering.
• Multivariable regression models to assess associations with all-cause mortality and acute exacerbations (AE).

Main Results:

• The AI-derived bronchiectasis score was independently associated with increased all-cause mortality in both discovery and validation cohorts.
• The score also correlated with the risk of any AE, severe AE, and time to first AE (p < 0.05).

Conclusions:

• AI-based quantification of bronchial tapering ratio on CT provides a reliable measure of bronchiectasis extent in COPD.
• This objective, quantitative method shows potential for risk stratification and improved management of COPD patients.