Pulmonary Arterial Pruning and Longitudinal Change in Percent Emphysema and Lung Function: The Genetic Epidemiology of COPD Study.

BACKGROUND: Pulmonary endothelial damage has been shown to precede the development of emphysema in animals, and vascular changes in humans have been observed in COPD and emphysema. RESEARCH QUESTION: Is intraparenchymal vascular pruning associated with longitudinal progression of emphysema on CT imaging or decline in lung function over 5 years? STUDY DESIGN AND METHODS: The Genetic Epidemiology of COPD Study enrolled ever smokers with and without COPD from 2008 through 2011. The percentage of emphysema-like lung, or "percent emphysema," was assessed at baseline and after 5 years on noncontrast CT imaging as the percentage of lung voxels < -950 Hounsfield units. An automated CT imaging-based tool assessed and classified intrapulmonary arteries and veins. Spirometry measures are postbronchodilator. Pulmonary arterial pruning was defined as a lower ratio of small artery volume (< 5 mm2 cross-sectional area) to total lung artery volume. Mixed linear models included demographics, anthropomorphics, smoking, and COPD, with emphysema models also adjusting for CT imaging scanner and lung function models adjusting for clinical center and baseline percent emphysema. RESULTS: At baseline, the 4,227 participants were 60 ± 9 years of age, 50% were women, 28% were Black, 47% were current smokers, and 41% had COPD. Median percent emphysema was 2.1 (interquartile range, 0.6-6.3) and progressed 0.24 percentage points/y (95% CI, 0.22-0.26 percentage points/y) over 5.6 years. Mean FEV1 to FVC ratio was 68.5 ± 14.2% and declined 0.26%/y (95% CI, -0.30 to -0.23%/y). Greater pulmonary arterial pruning was associated with more rapid progression of percent emphysema (0.11 percentage points/y per 1-SD increase in arterial pruning; 95% CI, 0.09-0.16 percentage points/y), including after adjusting for baseline percent emphysema and FEV1. Arterial pruning also was associated with a faster decline in FEV1 to FVC ratio (-0.04%/y per 1-SD increase in arterial pruning; 95% CI, -0.008 to -0.001%/y). INTERPRETATION: Pulmonary arterial pruning was associated with faster progression of percent emphysema and more rapid decline in FEV1 to FVC ratio over 5 years in ever smokers, suggesting that pulmonary vascular differences may be relevant in disease progression. TRIAL REGISTRY: ClinicalTrials.gov; No.: NCT00608764; URL: www.clinicaltrials.gov.

REGRESSION OF THE NAVIER-STOKES EQUATION SOLUTIONS FOR PULMONARY AIRWAY FLOW USING NEURAL NETWORKS.

Computerized fluid dynamics models of particle deposition in the human airways are used to characterize deposition patterns that enable the study of lung diseases like asthma and chronic obstructive pulmonary disease (COPD). Despite this fact, the influence of patient-specific geometry on the deposition efficiency and patterns is not well documented nor modeled. In part, this is due to the complexity of simulating the full Computational Fluid Dynamics (CFD) solution in patient-specific airway geometries, a factor that becomes a major hurdle for patient-specific studies given the complexity of the geometry of human lungs and their related airflow. In this paper, we present an approximation method based on neural networks to the Navier-Stokes equations that govern airway flow in a Physiologically Realistic Bifurcation (PRB) model for the conducting region of a single generation human airway branch. The flow distribution and deposition of tobacco particles have been simulated for the inspiratory regime using ANSYS Fluent and a neural network has been trained to regress the mean velocity and mass flow components. Our results show that the approximation works well under the modeled assumptions and the serial application of this model to a two-generation airway geometry provides reasonable approximations.

A SR-NET 3D-TO-2D ARCHITECTURE FOR PARASEPTAL EMPHYSEMA SEGMENTATION.

Paraseptal emphysema (PSE) is a relatively unexplored emphysema subtype that is usually asymptomatic, but recently associated with interstitial lung abnormalities which are related with clinical outcomes, including mortality. Previous local-based methods for emphysema subtype quantification do not properly characterize PSE. This is in part for their inability to properly capture the global aspect of the disease, as some the PSE lesions can involved large regions along the chest wall. It is our assumption, that path-based approaches are not well-suited to identify this subtype and segmentation is a better paradigm. In this work we propose and introduce the Slice-Recovery network (SR-Net) that leverages 3D contextual information for 2D segmentation of PSE lesions in CT images. For that purpose, a novel convolutional network architecture is presented, which follows an encoding-decoding path that processes a 3D volume to generate a 2D segmentation map. The dataset used for training and testing the method comprised 664 images, coming from 111 CT scans. The results demonstrate the benefit of the proposed approach which incorporate 3D context information to the network and the ability of the proposed method to identify and segment PSE lesions with different sizes even in the presence of other emphysema subtypes in an advanced stage.

DEEP-LEARNING STRATEGY FOR PULMONARY ARTERY-VEIN CLASSIFICATION OF NON-CONTRAST CT IMAGES.

Artery-vein classification on pulmonary computed tomography (CT) images is becoming of high interest in the scientific community due to the prevalence of pulmonary vascular disease that affects arteries and veins through different mechanisms. In this work, we present a novel approach to automatically segment and classify vessels from chest CT images. We use a scale-space particle segmentation to isolate vessels, and combine a convolutional neural network (CNN) to graph-cut (GC) to classify the single particles. Information about proximity of arteries to airways is learned by the network by means of a bronchus enhanced image. The methodology is evaluated on the superior and inferior lobes of the right lung of twenty clinical cases. Comparison with manual classification and a Random Forests (RF) classifier is performed. The algorithm achieves an overall accuracy of 87% when compared to manual reference, which is higher than the 73% accuracy achieved by RF.

Pulmonary vascular morphology as an imaging biomarker in chronic thromboembolic pulmonary hypertension.

Patients with chronic thromboembolic pulmonary hypertension (CTEPH) have morphologic changes to the pulmonary vasculature. These include pruning of the distal vessels, dilation of the proximal vessels, and increased vascular tortuosity. Advances in image processing and computer vision enable objective detection and quantification of these processes in clinically acquired computed tomographic (CT) scans. Three-dimensional reconstructions of the pulmonary vasculature were created from the CT angiograms of 18 patients with CTEPH diagnosed using imaging and hemodynamics as well as 15 control patients referred to our Dyspnea Clinic and found to have no evidence of pulmonary vascular disease. Compared to controls, CTEPH patients exhibited greater pruning of the distal vasculature (median density of small-vessel volume: 2.7 [interquartile range (IQR): 2.5-3.0] vs. 3.2 [3.0-3.8]; P = 0.008), greater dilation of proximal arteries (median fraction of blood in large arteries: 0.35 [IQR: 0.30-0.41] vs. 0.23 [0.21-0.31]; P = 0.0005), and increased tortuosity in the pulmonary arterial tree (median: 4.92% [IQR: 4.85%-5.21%] vs. 4.63% [4.39%-4.92%]; P = 0.004). CTEPH was not associated with dilation of proximal veins or increased tortuosity in the venous system. Distal pruning of the vasculature was correlated with the cardiac index (R = 0.51, P = 0.04). Quantitative models derived from CT scans can be used to measure changes in vascular morphology previously described subjectively in CTEPH. These measurements are also correlated with invasive metrics of pulmonary hemodynamics, suggesting that they may be used to assess disease severity. Further work in a larger cohort may enable the use of such measures as a biomarker for diagnostic, phenotyping, and prognostic purposes.

Real-time computed tomography-based augmented reality for natural orifice transluminal endoscopic surgery navigation.

BACKGROUND: Natural orifice transluminal endoscopic surgery (NOTES) is technically challenging owing to endoscopic short-sighted visualization, excessive scope flexibility and lack of adequate instrumentation. Augmented reality may overcome these difficulties. This study tested whether an image registration system for NOTES procedures (IR-NOTES) can facilitate navigation. METHODS: In three human cadavers 15 intra-abdominal organs were targeted endoscopically with and without IR-NOTES via both transgastric and transcolonic routes, by three endoscopists with different levels of expertise. Ease of navigation was evaluated objectively by kinematic analysis, and navigation complexity was determined by creating an organ access complexity score based on the same data. RESULTS: Without IR-NOTES, 21 (11·7 per cent) of 180 targets were not reached (expert endoscopist 3, advanced 7, intermediate 11), compared with one (1 per cent) of 90 with IR-NOTES (intermediate endoscopist) (P = 0·002). Endoscope movements were significantly less complex in eight of the 15 listed organs when using IR-NOTES. The most complex areas to access were the pelvis and left upper quadrant, independently of the access route. The most difficult organs to access were the spleen (5 failed attempts; 3 of 7 kinematic variables significantly improved) and rectum (4 failed attempts; 5 of 7 kinematic variables significantly improved). The time needed to access the rectum through a transgastric approach was 206·3 s without and 54·9 s with IR-NOTES (P = 0·027). CONCLUSION: The IR-NOTES system enhanced both navigation efficacy and ease of intra-abdominal NOTES exploration for operators of all levels. The system rendered some organs accessible to non-expert operators, thereby reducing one impediment to NOTES procedures.

The role of a computed tomography-based image registered navigation system for natural orifice transluminal endoscopic surgery: a comparative study in a porcine model.

BACKGROUND AND STUDY AIMS: Most natural orifice transluminal endoscopic surgery (NOTES) procedures have been performed in animal models through the anterior stomach wall, but this approach does not provide efficient access to all anatomic areas of interest. Moreover, injury of the adjacent structures has been reported when using a blind access. The aim of the current study was to assess the utility of a CT-based (CT: computed tomography) image registered navigation system in identifying safe gastrointestinal access sites for NOTES and identifying intraperitoneal structures. METHODS: A total of 30 access procedures were performed in 30 pigs: anterior gastric wall (n = 10), posterior gastric wall (n = 10), and anterior rectal wall (n = 10). Of these, 15 procedures used image registered guidance (IR-NOTES) and 15 procedures used a blind access (NOTES only). Timed abdominal exploration was performed with identification of 11 organs. The location of the endoscopic tip was tracked using an electromagnetic tracking system and was recorded for each case. Necropsy was performed immediately after the procedure. The primary outcome was the rate of complications; secondary outcome variables were number of organs identified and kinematic measurements. RESULTS: A total of 30 animals weighting a mean (± SD) of 30.2 ± 6.8 kg were included in the study. The incision point was correctly placed in 11 out of 15 animals in each group (73.3 %). The mean peritoneoscopy time and the number of properly identified organs were equivalent in the two groups. There were eight minor complications (26.7 %), two (13.3 %) in the IR-NOTES group and six (40.0 %) in the NOTES only group ( P = n. s.). Characteristics of the endoscope tip path showed a statistically significant improvement in trajectory smoothness of motion for all organs in the IR-NOTES group. CONCLUSION: The image registered system appears to be feasible in NOTES procedures and results from this study suggest that image registered guidance might be useful for supporting navigation with an increased smoothness of motion.