COPD affects up to 24 million people in the United States and it is estimated that 30 to 70% of these patients have clinically significant pulmonary vascular disease (PVD). While the observed increases in mean pulmonary arterial pressure tend to be mild or moderate, they have been shown to be independent predictors of health care utilization and mortality. PVD with secondary pulmonary hypertension, though, is a late stage manifestation of chronic tobacco smoke exposure and there is a continuum of changes to the pulmonary vasculature and heart that reflect the evolution of disease. One of the earliest observable changes in the parenchyma of smokers is pulmonary vascular remodeling. These changes have been described in subjects with normal lung function and are thought to be the precursor to emphysematous destruction of the parenchyma. The subsequent progression of emphysema and hyperinflation then leads to impaired cardiac function. Diastolic volumes of the left ventricle decrease and right ventricle dilates, both of which have been theorized to be due to occult pulmonary vascular disease and elevations in pulmonary vascular resistance. Our knowledge of these processes is limited because it is based upon an aggregation of disparate clinical studies employing an array of investigative techniques such as cardiac catheterization, histopathology, echocardiography, and cardiac magnetic resonance imaging. These tools are largely unable to discriminate the contributions of hemodynamically significant morphologic changes in the lungs of smokers such as vessel elongation in hyperinflated regions of the parenchyma, pruning of the distal small vessels, and widespread absence of a vascular bed due in regions of emphysema. We have previously performed a detailed investigation of these features in the Round 1 COPDGene CT scans and with this renewal application for R01 HL0144624, we are proposing to examine their progression over time, effect on biventricular morphology, and association with clinically relevant outcomes. We are proposing to perform a longitudinal investigation of the clinical relevance of pulmonary vascular and cardiac morphology in 6000 subjects enrolled in the COPDGene Study. To do this we have assembled a multidisciplinary team of pulmonologists, computer scientists, epidemiologists, radiologists, and statisticians. We will refine our surface fitting model of the heart using both 4D contrast enhanced cardiac CT scans and paired cardiac gated CT scans and echocardiograms in the CARDIA Study. We will then deploy these tools as well as our vascular segmentation algorithms in all the Round 1, Round 2 and Round 3 COPDGene CT scans collected at standardized intervals over a 10 year period of observation.
Longitudinal computed tomographic assessments of epicardial and pulmonary vascular morphology in subjects enrolled in the COPDGene Study may provide insight into the burden, severity, and clinical impact of pulmonary vascular remodeling and cardiopulmonary interdependence in smokers. A deeper understanding of these processes may lead to new therapies for patients with COPD.
| González, Germán; Ash, Samuel Y; Vegas-Sánchez-Ferrero, Gonzalo et al. (2018) Disease Staging and Prognosis in Smokers Using Deep Learning in Chest Computed Tomography. Am J Respir Crit Care Med 197:193-203 |
| Miller, Ezra R; Putman, Rachel K; Vivero, Marina et al. (2018) Histopathology of Interstitial Lung Abnormalities in the Context of Lung Nodule Resections. Am J Respir Crit Care Med 197:955-958 |
| Onieva, Jorge Onieva; Serrano, Germán González; Young, Thomas P et al. (2018) Multiorgan structures detection using deep convolutional neural networks. Proc SPIE Int Soc Opt Eng 10574: |
| González, Germán; Washko, George R; Estépar, Raúl San José (2018) Deep learning for biomarker regression: application to osteoporosis and emphysema on chest CT scans. Proc SPIE Int Soc Opt Eng 10574: |
| Cano-Espinosa, Carlos; González, Germán; Washko, George R et al. (2018) Automated Agatston Score Computation in non-ECG Gated CT Scans Using Deep Learning. Proc SPIE Int Soc Opt Eng 10574: |
| Vegas-Sánchez-Ferrero, Gonzalo; Ledesma-Carbayo, Maria J; Washko, George R et al. (2018) Autocalibration method for non-stationary CT bias correction. Med Image Anal 44:115-125 |
| Campo, Mónica Iturrioz; Pascau, Javier; José Estépar, Raúl San (2018) EMPHYSEMA QUANTIFICATION ON SIMULATED X-RAYS THROUGH DEEP LEARNING TECHNIQUES. Proc IEEE Int Symp Biomed Imaging 2018:273-276 |
| Nardelli, Pietro; Jimenez-Carretero, Daniel; Bermejo-Pelaez, David et al. (2018) Pulmonary Artery-Vein Classification in CT Images Using Deep Learning. IEEE Trans Med Imaging 37:2428-2440 |
| Ramos-Llorden, Gabriel; Vegas-Sanchez-Ferrero, Gonzalo; Bjork, Marcus et al. (2018) NOVIFAST: A Fast Algorithm for Accurate and Precise VFA MRI Mapping. IEEE Trans Med Imaging 37:2414-2427 |
| Diaz, Alejandro A; Strand, Matthew; Coxson, Harvey O et al. (2018) Disease Severity Dependence of the Longitudinal Association Between CT Lung Density and Lung Function in Smokers. Chest 153:638-645 |
Showing the most recent 10 out of 57 publications
