UCLA

Diseases affecting the respiratory system are among the most common and deadliest worldwide. The ability to characterize the patient-specific, physical properties of the respiratory system could improve treatments and interventions that aim to combat these diseases. With the unique capabilities of modern CT technology, we can formulate novel approaches to capture and characterize breathing motion. The development of fast-helical free-breathing CT (FHFBCT), along with the 5DCT motion modeling protocol, have allowed us to acquire CT images with dynamic information, free of the artifacts typically observed in 4DCT approaches. Though this protocol was developed for radiation oncology, its value stems far beyond. The goal of this dissertation is to use FHFBCT in several applications that provide guidance or improvement to technologies involved in the treatment of patients with respiratory diseases such as lung cancer and chronic obstructive pulmonary disease (COPD). First, we applied FHFBCT and motion modeling within its primary area of CT simulation for radiotherapy. We studied the use of the respiratory bellows and the entire patient skin surface as respiratory surrogates for CT simulation using FHFBCT data and found that the center abdomen would be a reliable skin surface-based surrogate for diaphragm motion. We then investigated an application of FHFBCT in the radiotherapy workflow outside of CT simulation. We demonstrated the feasibility of using FHFBCT and 5DCT motion modeling in cone-beam CT (CBCT) reconstruction to provide accurate motion compensation. Finally, we proposed the use of dynamic breathing information from FHFBCT beyond radiotherapy altogether, into areas such as surgical planning, pulmonology, and diagnostics. We developed an automatic, triangulation meshing technique for airway segmentations extracted from FHFBCT data to enable computational fluid dynamics (CFD) simulations with free-breathing data. We also calculated ventilation from FHFBCTs and characterized its heterogeneity across lobes of patients with COPD of varying severities. We performed a similar study using elasticity, a measure of tissue stiffness obtained through a CT-based elastography approach, instead of ventilation, to propose an additional biomarker for lung function. The work in this dissertation provides a demonstration of the unique advantages of breathing dynamics derived from FHFBCT. We hope that this work encourages others to research the studied topics as well as continue to implement FHFBCT and motion modeling to other areas of medicine to improve diagnostics and treatments for patients suffering from respiratory illnesses.