Quantitative interstitial abnormalities (QIAs) are early measures of lung injury automatically detected on chest computed tomography scans. QIAs are associated with impaired respiratory health and ...share features with advanced lung diseases, but their biological underpinnings are not well understood.
To identify novel protein biomarkers of QIAs using high-throughput plasma proteomic panels within two multicenter cohorts.
We measured the plasma proteomics of 4,383 participants in an older, ever-smoker cohort (COPDGene Genetic Epidemiology of Chronic Obstructive Pulmonary Disease) and 2,925 participants in a younger population cohort (CARDIA Coronary Artery Disease Risk in Young Adults) using the SomaLogic SomaScan assays. We measured QIAs using a local density histogram method. We assessed the associations between proteomic biomarker concentrations and QIAs using multivariable linear regression models adjusted for age, sex, body mass index, smoking status, and study center (Benjamini-Hochberg false discovery rate-corrected
⩽ 0.05).
In total, 852 proteins were significantly associated with QIAs in COPDGene and 185 in CARDIA. Of the 144 proteins that overlapped between COPDGene and CARDIA, all but one shared directionalities and magnitudes. These proteins were enriched for 49 Gene Ontology pathways, including biological processes in inflammatory response, cell adhesion, immune response, ERK1/2 regulation, and signaling; cellular components in extracellular regions; and molecular functions including calcium ion and heparin binding.
We identified the proteomic biomarkers of QIAs in an older, smoking population with a higher prevalence of pulmonary disease and in a younger, healthier community cohort. These proteomics features may be markers of early precursors of advanced lung diseases.
Bronchiectasis in adults with chronic obstructive pulmonary disease (COPD) is associated with greater mortality. However, whether suspected bronchiectasis-defined as incidental bronchiectasis on ...computed tomography (CT) images plus clinical manifestation-is associated with increased mortality in adults with a history of smoking with normal spirometry and preserved ratio impaired spirometry (PRISm) is unknown.
To determine the association between suspected bronchiectasis and mortality in adults with normal spirometry, PRISm, and obstructive spirometry.
Prospective, observational cohort.
The COPDGene (Genetic Epidemiology of Chronic Obstructive Pulmonary Disease) study.
7662 non-Hispanic Black or White adults, aged 45 to 80 years, with 10 or more pack-years of smoking history. Participants who were former and current smokers were stratified into normal spirometry (
= 3277), PRISm (
= 986), and obstructive spirometry (
= 3399).
Bronchiectasis identified by CT was ascertained using artificial intelligence-based measurements of an airway-to-artery ratio (AAR) greater than 1 (AAR >1), a measure of bronchial dilatation. The primary outcome of "suspected bronchiectasis" was defined as an AAR >1 of greater than 1% plus 2 of the following: cough, phlegm, dyspnea, and history of 2 or more exacerbations.
Among the 7662 participants (mean age, 60 years; 52% women), 1352 (17.6%) had suspected bronchiectasis. During a median follow-up of 11 years, 2095 (27.3%) died. Ten-year mortality risk was higher in participants with suspected bronchiectasis, compared with those without suspected bronchiectasis (normal spirometry: difference in mortality probability Pr, 0.15 95% CI, 0.09 to 0.21; PRISm: Pr, 0.07 CI, -0.003 to 0.15; obstructive spirometry: Pr, 0.06 CI, 0.03 to 0.09). When only CT was used to identify bronchiectasis, the differences were attenuated in the normal spirometry (Pr, 0.04 CI, -0.001 to 0.08).
Only 2 racial groups were studied. Only 1 measurement was used to define bronchiectasis on CT. Symptoms of suspected bronchiectasis were nonspecific.
Suspected bronchiectasis was associated with a heightened risk for mortality in adults with normal and obstructive spirometry.
National Heart, Lung, and Blood Institute.
Background and objectives The scarcity of data for training deep learning models in pediatrics has prompted questions about the feasibility of employing CNNs trained with adult images for pediatric ...populations. In this work, a pneumonia classification CNN was used as an exploratory example to showcase the adaptability and efficacy of such models in pediatric healthcare settings despite the inherent data constraints. Methods To develop a curated training dataset with reduced biases, 46,947 chest X-ray images from various adult datasets were meticulously selected. Two preprocessing approaches were tried to assess the impact of thoracic segmentation on model attention outside the thoracic area. Evaluation of our approach was carried out on a dataset containing 5,856 chest X-rays of children from 1 to 5 years old. Results An analysis of attention maps indicated that networks trained with thorax segmentation placed less attention on regions outside the thorax, thus eliminating potential bias. The ensuing network exhibited impressive performance when evaluated on an adult dataset, achieving a pneumonia discrimination AUC of 0.95. When tested on a pediatric dataset, the pneumonia discrimination AUC reached 0.82. Conclusions The results of this study show that adult-trained CNNs can be effectively applied to pediatric populations. This could potentially shift focus towards validating adult models over pediatric population instead of training new CNNs with limited pediatric data. To ensure the generalizability of deep learning models, it is important to implement techniques aimed at minimizing biases, such as image segmentation or low-quality image exclusion.
•This study explores the cardiopulmonary relation in patients with atrial fibrillation.•Pulmonary vascular volumes were assessed using automatic computed tomography analysis.•Pulmonary vascular ...remodeling is associated with impaired echocardiographic metrics.•Reduced blood volume in the peripheral pulmonary vessels is linked to persistent atrial fibrillation.
Pulmonary vascular abnormalities, quantified from computed tomography scans, have frequently been observed in patients with pulmonary diseases. However, little is known about pulmonary vascular changes in patients with cardiac disease. Thus, we aimed to examine the cardiopulmonary relation in patients with atrial fibrillation (AF) by comparing pulmonary vascular volume (PVV) to echocardiographic measures and AF severity. A total of 742 patients (median age 63 years, 70% men) who underwent ablation for AF were included. Preprocedural cardiac computed tomography was used to measure the total and small-vessel PVV, along with the pulmonary artery to aorta ratio and the degree of emphysema. The association between PVV and echocardiographic parameters was evaluated using a multivariable linear regression analysis. Lower total and small-vessel PVV were associated with more impaired measures of cardiac structure and function, including but not limited to left ventricular ejection fraction and peak atrial longitudinal strain. Patients with reduced total and small-vessel PVV had higher odds of having persistent AF than paroxysmal AF in the unadjusted logistic regression analyses. However, after clinical and echocardiographic adjustments, only reduced small-vessel PVV remained independently associated with persistent AF (odds ratio 1.90, 95% confidence interval 1.26 to 2.87, p = 0.002). In conclusion, pulmonary vascular remodeling is associated with persistent AF and with more impaired measures of cardiac structure and function, providing further insights into heart-lung interactions in this patient group.
Display omitted
Quantitative interstitial abnormalities (QIA) are an automated computed tomography (CT) finding of early parenchymal lung disease, associated with worse lung function, reduced exercise capacity, ...increased respiratory symptoms, and death. The metabolomic perturbations associated with QIA are not well known. We sought to identify plasma metabolites associated with QIA in smokers. We also sought to identify shared and differentiating metabolomics features between QIA and emphysema, another smoking-related advanced radiographic abnormality.
In 928 former and current smokers in the Genetic Epidemiology of COPD cohort, we measured QIA and emphysema using an automated local density histogram method and generated metabolite profiles from plasma samples using liquid chromatography-mass spectrometry (Metabolon). We assessed the associations between metabolite levels and QIA using multivariable linear regression models adjusted for age, sex, body mass index, smoking status, pack-years, and inhaled corticosteroid use, at a Benjamini-Hochberg False Discovery Rate p-value of ≤ 0.05. Using multinomial regression models adjusted for these covariates, we assessed the associations between metabolite levels and the following CT phenotypes: QIA-predominant, emphysema-predominant, combined-predominant, and neither- predominant. Pathway enrichment analyses were performed using MetaboAnalyst.
We found 85 metabolites significantly associated with QIA, with overrepresentation of the nicotinate and nicotinamide, histidine, starch and sucrose, pyrimidine, phosphatidylcholine, lysophospholipid, and sphingomyelin pathways. These included metabolites involved in inflammation and immune response, extracellular matrix remodeling, surfactant, and muscle cachexia. There were 75 metabolites significantly different between QIA-predominant and emphysema-predominant phenotypes, with overrepresentation of the phosphatidylethanolamine, nicotinate and nicotinamide, aminoacyl-tRNA, arginine, proline, alanine, aspartate, and glutamate pathways.
Metabolomic correlates may lend insight to the biologic perturbations and pathways that underlie clinically meaningful quantitative CT measurements like QIA in smokers.
In recent years, several techniques for image-to-image translation by means of generative adversarial neural networks (GAN) have been proposed to learn mapping characteristics between a source and a ...target domain. In particular, in the medical imaging field conditional GAN frameworks with paired samples (cGAN) and unconditional cycle-consistent GANs with unpaired data (CycleGAN) have been demonstrated as a powerful scheme to model non-linear mappings that produce realistic target images from different modality sources. When proposing the usage and adaptation of these frameworks for medical image synthesis, quantitative and qualitative validation are usually performed by assessing the similarity between synthetic and target images in terms of metrics such as mean absolute error (MAE) or structural similarity (SSIM) index. However, an evaluation of clinically relevant markers showing that diagnostic information is not overlooked in the translation process is often missing. In this work, we aim at demonstrating the importance of validating medical image-to-image translation techniques by assessing their effect on the measurement of clinically relevant metrics and biomarkers. We implemented both a conditional and an unconditional approach to synthesize conventional dose chest CT scans from reduced dose CT and show that while both visually and in terms of traditional metrics the network appears to successfully minimize perceptual discrepancies, these methods are not reliable to systematically reproduce quantitative measurements of various chest biomarkers.