Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory lung disease that causes obstructed airflow from the lungs. COPD is the fourth leading cause of death in the United States and currently affects 328 million people worldwide with a projected increase in healthcare costs from $32 billion in 2010 to $49 billion in 2020. Quantitative computed tomography (CT)-based characterizations of bronchial metrics, such as airway lumen area and wall thickness, and parenchymal characterizations of emphysema and air trapping, are garnering research interest to help understand the pathophysiology and mechanism of several lung diseases’ occurrence and progression. Airway segmentation and anatomical branch labeling allow spatial matching and referencing across individuals in large multi-center population-based studies such as, COPDGene, SPIROMICS, CanCOLD, MESA Lung, and SARP and has led to novel insights between COPD and airway phenotypes.

New theory and algorithms to automatically segment the human airway tree using thoracic CT imaging and label clinically significant segmental bronchi are developed and evaluated. First, an image processing framework, called freeze-and-grow, is developed for airway tree segmentation which uses a multi-parametric approach to iteratively capture finer details of the airway tree while detection and correction for segmentation leakages. Then, deep learning is used to enhance the airway lumen regions in the chest CT images to address the inherent challenges in CT intensity-based airway tree segmentation while improving efficiency. The new airway segmentation methods outperform an industry standard method requiring manual intervention as well as recently developed automated methods which use deep learning. Lastly, an automated algorithm for anatomical airway branches is developed and evaluated which uses two stage machine learning and hierarchical features to compartmentalize the labeling process based on the airway tree branching anatomy.

The methods developed in this thesis provides a framework to assess the airway tree for large longitudinal or cross-sectional studies and explore the roles played by resulting bronchial metrics in lung pathology. The major contributions of this thesis include: (1) development of a multiparametric freeze-and-grow airway segmentation algorithm; (2) integration of a deep learning-based airway lumen enhancement into the freeze-and-grow airway segmentation algorithm; (3) evaluation of the new airway segmentation methods with industry and recent deep learning-based standard airway segmentation methods on multi-center COPD study data; (4) development of a new airway anatomic branch labeling algorithm using two step machine learning and novel hierarchical topologic and geometric tree features; (5) applying the developed on data from a multi-center COPD study.