Recent studies show that pulmonary vascular diseases may specifically affect arteries or veins through different physiologic mechanisms. To detect changes in the two vascular trees, physicians manually analyze the chest computed tomography (CT) image of the patients in search of abnormalities. This process is time-consuming, difficult to standardize and thus not feasible for large clinical studies or useful in real-world clinical decision making. Therefore, automatic separation of arteries and veins in CT images is becoming of great interest, as it may help physicians accurately diagnose pathological conditions. In this work, we present a novel, fully automatic approach to classifying vessels from chest CT images into arteries and veins. The algorithm follows three main steps: first, a scale-space particles segmentation to isolate vessels; then a 3D convolutional neural network (CNN) to obtain a first classification of vessels; finally, graph-cuts (GC) optimization to refine the results. To justify the usage of the proposed CNN architecture, we compared different 2D and 3D CNNs that may use local information from bronchus- and vessel-enhanced images provided to the network with different strategies. We also compared the proposed CNN approach with a Random Forests (RF) classifier. The methodology was trained and evaluated on the superior and inferior lobes of the right lung of eighteen clinical cases with non-contrast chest CT scans, in comparison with manual classification. The proposed algorithm achieves an overall accuracy of 94%, which is higher than the accuracy obtained using other CNN architectures and RF. Our method was also validated with contrast-enhanced CT scans of patients with Chronic Thromboembolic Pulmonary Hypertension (CTEPH) to demonstrate that our model generalizes well to contrast-enhanced modalities. The proposed method outperforms state-of-the-art methods, paving the way for future use of 3D CNN for A/V classification in CT images.

Background COPD causes significant morbidity and concomitant pulmonary vascular disease and cardiac dysfunction are associated with poor prognosis. CT-detected relative pulmonary artery enlargement defined as a pulmonary artery to ascending aorta diameter ratio greater than one (PA:A>1) is a marker for pulmonary hypertension and predicts COPD exacerbations. However, little is known about the relationship between the PA:A ratio, pulmonary blood volume, and cardiac function. Methods and Results A single-center prospective cohort study of COPD patients was conducted. Clinical characteristics and CT metrics, including the PA:A and pulmonary blood vessel volume were measured. Ventricular functions, volumes, and dimensions were measured by cine cardiac magnetic resonance imaging (cMRI) with 3D analysis. Linear regression examined the relationships between clinical characteristics, CT and cMRI metrics, and 6-minute walk distance (6MWD). Twenty four patients were evaluated and those with PA:A>1 had higher right ventricular (RV) end-diastolic and end-systolic volume indices accompanied by lower RV ejection fraction (EF) (52±7% vs 60±9%, p=0.04). The PA:A correlated inversely with total intraparenchymal pulmonary blood vessel volume and the volume of distal vessels with a cross sectional area of <5 mm2. Lower forced expiratory volume, PA:A>1, and hyperinflation correlated with reduced RVEF. Both PA diameter and reduced RVEF were independently associated with reduced 6MWD. Conclusions The loss of blood volume in distal pulmonary vessels is associated with PA enlargement on CT. CMRI detects early RV dysfunction and remodeling in non-severe COPD patients with a PA:A>1. Both RV dysfunction and PA enlargement are independently associated with reduced walk distance.

No abstract