Off-campus WSU users: To download campus access dissertations, please use the following link to log into our proxy server with your WSU access ID and password, then click the "Off-campus Download" button below.
Non-WSU users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Access Type
WSU Access
Date of Award
January 2019
Degree Type
Dissertation
Degree Name
Ph.D.
Department
Medical Physics
First Advisor
Thomas Guerrero
Second Advisor
Jacob Burmeister
Abstract
This work evaluates a measureable variation in the pulmonary blood mass during respiration, from four-dimensional computed tomography (4DCT). This phenomenon was comprehensively evaluated at the scale of the entire lung in an 89 patient retrospective analysis. The spatially heterogeneous behavior of this signal was further quantified by breaking the lungs into smaller geometrically distinct sub regions. To evaluate the potential for this signal to be used as a surrogate for local perfusion, a prospective clinical trial was conducted including 15 patients entering the emergency department, suspected of pulmonary embolism (PE). Patients with confirmed PE received additional (cine) 4DCT and SPECT perfusion imaging. The 4DCT images were sorted into full inhale and exhale phases, which were used to generate pulmonary perfusion (ΔPBM) images. The 4DCT images were examined for reproducibility, while the ΔPBM images were evaluated against the SPECT images to determine the efficacy of ΔPBM imaging as a perfusion surrogate.
In the retrospective study, the mean change in pulmonary mass was 50.9 (3.9; -37.7 – 167.4) g, with the maximum occurring at phase 9 and minimum at phase 6, on average. A positive change pulmonary mass during inspiration (exhale to inhale) and negative change during expiration (inhale to exhale) was observed for 86 (96.0%) patients. For each case, a linear regression line was fit from phase 5 to 9, representing inhalation, and phase 0 to 5, representing exhalation. This association was summarized using the Spearman correlation, which was found to be 0.61. A greater ∆PBM is expected in patients with a larger radiographic volume. In the prospective study average variation in radiographic tidal volume (ΔRTV), and parenchymal lung mass (∇PBM) between consecutive 4DCT imaging events was 14.84 (10.58; 2.60-37.63)% and 18.23 (10.4; 0.60-38.32)%, respectively. Four patients (27%) exhibited inter-scan variations in radiographic tidal volume (ΔRTV) larger than 20%. Intra-scan variation in MSE was negligible. Mean SSIM for consecutive imaging events was 0.896(0.050; 0.786-0.972) and 0.897(0.043; 0.795-0.953). Normalized Joint Entropy (ΔJE) was 24.7(10.6; 4.0-39.1) and 7.50(11.1; 6.4-28.6) for the inhale and exhale volumes. Gamma analysis was performed for each patient on the normalized standardized data, using a 12.5%, 10mm search criteria (figure 4.8). Average gamma pass rate was 48.9(7.25; 35.9 – 61.4)%.
Recommended Citation
Myziuk, Nicholas K., "Quantification Of Respiratory Induced Pulmonary Blood Flow From 4dct" (2019). Wayne State University Dissertations. 2241.
https://digitalcommons.wayne.edu/oa_dissertations/2241