Emphysema Subtypes Research Articles

Recent advances in airway imaging using micro-computed tomography and computed tomography for chronic obstructive pulmonary disease.

Chronic obstructive pulmonary disease (COPD) is a complex lung disease characterized by a combination of airway disease and emphysema. Emphysema is classified as centrilobular emphysema (CLE), paraseptal emphysema (PSE), or panlobular emphysema (PLE), and airway disease extends from the respiratory, terminal, and preterminal bronchioles to the central segmental airways. Although clinical computed tomography (CT) cannot be used to visualize the small airways, micro-CT has shown that terminal bronchiole disease is more severe in CLE than in PSE and PLE, and micro-CT findings suggest that the loss and luminal narrowing of terminal bronchioles is an early pathological change in CLE. Furthermore, the introduction of ultra-high-resolution CT has enabled direct evaluation of the proximal small (1 to 2-mm diameter) airways, and new CT analytical methods have enabled estimation of small airway disease and prediction of future COPD onset and lung function decline in smokers with and without COPD. This review discusses the literature on micro-CT and the technical advancements in clinical CT analysis for COPD. Hopefully, novel micro-CT findings will improve our understanding of the distinct pathogeneses of the emphysema subtypes to enable exploration of new therapeutic targets, and sophisticated CT imaging methods will be integrated into clinical practice to achieve more personalized management.

Latent traits of lung tissue patterns in former smokers derived by dual channel deep learning in computed tomography images

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease and the traditional variables extracted from computed tomography (CT) images may not be sufficient to describe all the topological features of lung tissues in COPD patients. We employed an unsupervised three-dimensional (3D) convolutional autoencoder (CAE)-feature constructor (FC) deep learning network to learn from CT data and derive tissue pattern-clusters jointly. We then applied exploratory factor analysis (EFA) to discover the unobserved latent traits (factors) among pattern-clusters. CT images at total lung capacity (TLC) and residual volume (RV) of 541 former smokers and 59 healthy non-smokers from the cohort of the SubPopulations and Intermediate Outcome Measures in the COPD Study (SPIROMICS) were analyzed. TLC and RV images were registered to calculate the Jacobian (determinant) values for all the voxels in TLC images. 3D Regions of interest (ROIs) with two data channels of CT intensity and Jacobian value were randomly extracted from training images and were fed to the 3D CAE-FC model. 80 pattern-clusters and 7 factors were identified. Factor scores computed for individual subjects were able to predict spirometry-measured pulmonary functions. Two factors which correlated with various emphysema subtypes, parametric response mapping (PRM) metrics, airway variants, and airway tree to lung volume ratio were discriminants of patients across all severity stages. Our findings suggest the potential of developing factor-based surrogate markers for new COPD phenotypes.

Advances in imaging for lung emphysema.

Lung emphysema represents a major public health burden and still accounts for five percent of all deaths worldwide. Hence, it is essential to further understand this disease in order to develop effective diagnostic and therapeutic strategies. Lung emphysema is an irreversible enlargement of the airways distal to the terminal bronchi (i.e., the alveoli) due to the destruction of the alveolar walls. The two most important causes of emphysema are (I) smoking and (II) α1-antitrypsin-deficiency. In the former lung emphysema is predominant in the upper lung parts, the latter is characterized by a predominance in the basal areas of the lungs. Since quantification and evaluation of the distribution of lung emphysema is crucial in treatment planning, imaging plays a central role. Imaging modalities in lung emphysema are manifold: computed tomography (CT) imaging is nowadays the gold standard. However, emerging imaging techniques like dynamic or functional magnetic resonance imaging (MRI), scintigraphy and lately also the implementation of radiomics and artificial intelligence are more and more diffused in the evaluation, diagnosis and quantification of lung emphysema. The aim of this review is to shortly present the different subtypes of lung emphysema, to give an overview on prediction and risk assessment in emphysematous disease and to discuss not only the traditional, but also the new imaging techniques for diagnosis, quantification and evaluation of lung emphysema.

The Role of Granzyme B Containing Cells in the Progression of Chronic Obstructive Pulmonary Disease.

BackgroundLung inflammation plays a vital role in the pathogenesis of chronic obstructive pulmonary disease (COPD), but the characteristics of the inflammatory process remain unclear. There is growing interest in the role of granzyme B (GzmB) because CD8+ T cells can induce apoptosis of target cells by releasing GzmB, which in turn may cause tissue injury and remodeling. However, GzmB is also expressed by regulatory cells, which are able to suppress CD8+ T cell. The role of GzmB+ cells needs to be defined in COPD.MethodsGzmB+ and CD8+ cells on alveolar wall of surgically resected lungs of microscopically classified 12 nonsmoking control, 12 panlobular emphysema (PLE) and 30 centrilobular emphysema (CLE) subjects were localized by immunohistochemical method. Positively stained cells on alveolar wall were counted and length of corresponding alveolar wall was measured. The results were expressed as mean number of positively stained cells per mm of alveolar wall in each subject.ResultsThe number of GzmB+ and CD8+ cells on alveolar wall of CLE was greater than that of control or PLE subjects (p<0.05 and p<0.001, respectively). There was a positive relationship between the number of alveolar GzmB+ cells and forced expiratory volume in 1 second (FEV1) (r=0.610, p=0.003) in CLE subjects. The number of alveolar GzmB+ cells progressively decreased with decline of FEV1.ConclusionOur finding that number of alveolar GzmB+ cells was associated with FEV1 suggests that GzmB+ cells might have protective role in the progression of lung destruction and airflow limitation in CLE, which is the predominant emphysema subtype of COPD.

CT Assessment of Subtypes of Pulmonary Emphysema in Mongolian Miners

Objectives: To determine the incidence of subtypes of pulmonary emphysema identified by LDCT imaging and the relationship between emphysema subtypes with smoking and pulmonary function in miners. Methods: We reviewed 329 miners (245 men and 84 women). Each miner had a low dose (LD) chest CT. Among them 75 miners had a standard chest CT. The images were reconstructed using contemporary iterative reconstruction. Lung volumes and emphysema severity was calculated using automated densitometry segmentation software and qualitative visual analyses.There were two subtypes of pulmonary emphysema found: centrilobular emphysema (CLE) and paraseptal emphysema (PSE). Based on these subtypes, CLE was divided into the following five categories: trace, mild, moderate, confluent, and advanced CLE. PSE was divided into the following two catergories: mild and substantial PSE. Results: Pulmonary emphysema was found in 89 (27.1%) of 329 miners. According to the survey, 73 (82.0%) miners had centrilobulor emphysema, and 17 (18.0%) had a paraseptal emphysema. The emphysema group was a lower Gensler index than thenon-emphysema group. The smoker group had higher emphysematous changes than the non-smoker groups (p &lt; 0.05). Conclusions: Dose reduced CT isauseful study for the assessment of emphysema.

Exploring the Association Between Emphysema Phenotypes and Low Bone Mineral Density in Smokers with and without COPD.

RationaleEmphysema and osteoporosis are tobacco-related diseases. Many studies have shown that emphysema is a strong and independent predictor of low bone mineral density (BMD) in smokers; however, none of them explored its association with different emphysema subtypes.ObjectiveTo explore the association between the different emphysema subtypes and the presence of low bone mineral density in a population of active or former smokers with and without chronic obstructive pulmonary disease (COPD).MethodsOne hundred and fifty-three active and former smokers from a pulmonary clinic completed clinical questionnaires, pulmonary function tests, a low-dose chest computed tomography (LDCT) and a dual-energy absorptiometry (DXA) scans. Subjects were classified as having normal BMD or low BMD (osteopenia or osteoporosis). Emphysema was classified visually for its subtype and severity. Logistic regression analysis explored the relationship between the different emphysema subtypes and the presence of low BMD adjusting for other important factors.ResultsSeventy-five percent of the patients had low BMD (78 had osteopenia and 37 had osteoporosis). Emphysema was more frequent (66.1 vs 26.3%, p=<0.001) and severe in those with low BMD. Multivariable analysis adjusting for other significant cofactors (age, sex, FEV1, and severity of emphysema) showed that BMI (OR=0.91, 95% CI: 0.76–0.92) and centrilobular emphysema (OR=26.19, 95% CI: 1.71 to 399.44) were associated with low BMD.ConclusionLow BMD is highly prevalent in current and former smokers. BMI and centrilobular emphysema are strong and independent predictors of its presence, which suggests that they should be considered when evaluating smokers at risk for low BMD.

Chronic obstructive pulmonary disease and related phenotypes: polygenic risk scores in population-based and case-control cohorts

SummaryBackgroundGenetic factors influence chronic obstructive pulmonary disease (COPD) risk, but the individual variants that have been identified have small effects. We hypothesised that a polygenic risk score using additional variants would predict COPD and associated phenotypes.MethodsWe constructed a polygenic risk score using a genome-wide association study of lung function (FEV1 and FEV1/forced vital capacity [FVC]) from the UK Biobank and SpiroMeta. We tested this polygenic risk score in nine cohorts of multiple ethnicities for an association with moderate-to-severe COPD (defined as FEV1/FVC <0·7 and FEV1 <80% of predicted). Associations were tested using logistic regression models, adjusting for age, sex, height, smoking pack-years, and principal components of genetic ancestry. We assessed predictive performance of models by area under the curve. In a subset of studies, we also studied quantitative and qualitative CT imaging phenotypes that reflect parenchymal and airway pathology, and patterns of reduced lung growth.FindingsThe polygenic risk score was associated with COPD in European (odds ratio [OR] per SD 1·81 [95% CI 1·74–1·88] and non-European (1·42 [1·34–1·51]) populations. Compared with the first decile, the tenth decile of the polygenic risk score was associated with COPD, with an OR of 7·99 (6·56–9·72) in European ancestry and 4·83 (3·45–6·77) in non-European ancestry cohorts. The polygenic risk score was superior to previously described genetic risk scores and, when combined with clinical risk factors (ie, age, sex, and smoking pack-years), showed improved prediction for COPD compared with a model comprising clinical risk factors alone (AUC 0·80 [0·79–0·81] vs 0·76 [0·75–0·76]). The polygenic risk score was associated with CT imaging phenotypes, including wall area percent, quantitative and qualitative measures of emphysema, local histogram emphysema patterns, and destructive emphysema subtypes. The polygenic risk score was associated with a reduced lung growth pattern.InterpretationA risk score comprised of genetic variants can identify a small subset of individuals at markedly increased risk for moderate-to-severe COPD, emphysema subtypes associated with cigarette smoking, and patterns of reduced lung growth.FundingUS National Institutes of Health, Wellcome Trust.

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.

Non-inflammatory emphysema induced by NO2 chronic exposure and intervention with demethylation 5-Azacytidine

AimsA rat model of emphysema was established that mimics the features of the human emphysema subtype and explores the effects of demethylation on lung function and blood tests. Materials and methodsRats were randomly assigned to NO2, NO2 + 5-Azacytidine, and normal air groups based on a emphysema rat model induced by chronic NO2 exposure. This study estimates the characteristics of emphysema by conducting an analysis for IL-6 and TNF-α levels in bronchoalveolar lavage fluids (BALF) and plasma. Furthermore, CD68 macrophage immunofluorescent staining and inflammatory cell counts in BALF were compared between rats exposed to NO2 and normal air. Key findings5-Azacytidine treatment led to restored ∆weight at 14 and 75 days of intervention and NO2 + 5-Azacytidine significantly reversed the effect of NO2 exposure on ∆weight. Intervention with 5-Azacytidine alleviated the decline of pulmonary function with a significant increase in FEV100/FVC% at 75 days in NO2 + 5-Azacytidine rats compared to NO2 rats. 5-Azacytidine reduced the counts of white blood cells (WBCs), granulocytes, lymphocytes, and monocytes at 14 days, but increased WBC, granulocyte, and monocyte counts at 45 days. Red blood cell counts, hemoglobin, and hematocrit concentrations were significantly reduced in NO2 + 5-Azacytidine rats. SignificanceThis non-inflammatory rat emphysema model (induced by chronic NO2 exposure with global DNA hypomethylation and demethylation therapy with 5-Azacytidine) effectively improved emphysema by alleviating the decline of lung function and hypoxia, and slightly reinforced immune function. These results indicate the therapeutic potential of demethylation agents for the prevention and treatment of emphysema induced by the air pollutant NO2.

Classification and Quantification of Emphysema Using a Multi-Scale Residual Network.

Automated tissue classification is an essential step for quantitative analysis and treatment of emphysema. Although many studies have been conducted in this area, there still remain two major challenges. First, different emphysematous tissue appears in different scales, which we call "inter-class variations." Second, the intensities of CT images acquired from different patients, scanners or scanning protocols may vary, which we call "intra-class variations". In this paper, we present a novel multi-scale residual network with two channels of raw CT image and its differential excitation component. We incorporate multi-scale information into our networks to address the challenge of inter-class variations. In addition to the conventional raw CT image, we use its differential excitation component as a pair of inputs to handle intra-class variations. Experimental results show that our approach has superior performance over the state-of-the- art methods, achieving a classification accuracy of 93.74% on our original emphysema database. Based on the classification results, we also perform the quantitative analysis of emphysema in 50 subjects by correlating the quantitative results (the area percentage of each class) with pulmonary functions. We show that centrilobular emphysema (CLE) and panlobular emphysema (PLE) have strong correlation with the pulmonary functions and the sum of CLE and PLE can be used as a new and accurate measure of emphysema severity instead of the conventional measure (sum of all subtypes of emphysema). The correlations between the new measure and various pulmonary functions are up to |r| = 0.922 (r is correlation coefficient).

Centrilobular emphysema and coronary artery calcification: mediation analysis in the SPIROMICS cohort

BackgroundChronic obstructive pulmonary disease (COPD) is associated with a two-to-five fold increase in the risk of coronary artery disease independent of shared risk factors. This association is hypothesized to be mediated by systemic inflammation but this link has not been established.MethodsWe included 300 participants enrolled in the SPIROMICS cohort, 75 each of lifetime non-smokers, smokers without airflow obstruction, mild-moderate COPD, and severe-very severe COPD. We quantified emphysema and airway disease on computed tomography, characterized visual emphysema subtypes (centrilobular and paraseptal) and airway disease, and used the Weston visual score to quantify coronary artery calcification (CAC). We used the Sobel test to determine whether markers of systemic inflammation mediated a link between spirometric and radiographic features of COPD and CAC.ResultsFEV1/FVC but not quantitative emphysema or airway wall thickening was associated with CAC (p = 0.036), after adjustment for demographics, diabetes mellitus, hypertension, statin use, and CT scanner type. To explain this discordance, we examined visual subtypes of emphysema and airway disease, and found that centrilobular emphysema but not paraseptal emphysema or bronchial thickening was independently associated with CAC (p = 0.019). MMP3, VCAM1, CXCL5 and CXCL9 mediated 8, 8, 7 and 16% of the association between FEV1/FVC and CAC, respectively. Similar biomarkers partially mediated the association between centrilobular emphysema and CAC.ConclusionsThe association between airflow obstruction and coronary calcification is driven primarily by the centrilobular subtype of emphysema, and is linked through bioactive molecules implicated in the pathogenesis of atherosclerosis.Trial RegistrationClinicalTrials.gov: Identifier: NCT01969344.

Effect of the emphysema subtypes of patients with chronic obstructive pulmonary disease on airway inflammation and COTE index.

The aim of the present study was investigate the association of the severity of emphysema of patients with chronic obstructive pulmonary disease (COPD) with airway inflammation and the COPD-specific comorbidity test (COTE) index. A total of 94 patients with COPD were divided into four groups according to the severity of their emphysema; in each patient, comorbidities were recorded and inflammatory biomarkers, including MMP-9 and TIMP-1 were determined in circulating blood. The unbalanced proportion of MMP-9 and its inhibitor, TIMP-1, led to the airway inflammation and lung remodeling in the patients with COPD. A total of 80.85% of the patients had emphysema of different degrees. The quantity of male patients and the smoking index in the three emphysema groups were significantly higher than those in the non-emphysema group (F=7.67 and 5.42, P<0.05). The level of the predicted percent offorced expiratory volume in 1 sec in the non-emphysema group were significantly higher than those in the emphysema group (4.33; P<0.05), and the level of D-dimer in the non-emphysema group was significantly lower than that in the mild and moderate emphysema groups (F=9.38, P<0.05). The low-attenuation area score was negatively correlated with inhaled bronchodilators (r=-0.240, P=0.007) but positively correlated with the frequency of acute exacerbations in the previous year (r=0.211, P=0.001). In terms of treatment, the use of systemic hormone therapy in the emphysema group was more frequent than that in the non-emphysema group (F=6.21, 12.92 and 4.08, P<0.05). The level of MMP-9 was significantly higher in COPD patients with >3 comorbidities, a COTE index of ≥4 and cardiovascular disease as well as coronary heart disease (t=6.40, 2.53, 3.65 and 2.90, P<0.05). The level of MMP-9 was positively correlated with the neutrophilic granulocyte percentage, the number of comorbidities and the COTE index (r=0.193, 0.402 and 0.311, P<0.01). The severity of emphysema in patients with COPD was correlated with the persistence of inflammatory factors in the circulating blood and the frequency of acute exacerbations. It was indicated that MMP-9 has a critical role in numerous comorbidities of COPD.

Coupled Immunological and Biomechanical Model of Emphysema Progression.

Chronic Obstructive Pulmonary Disease (COPD) is a disabling respiratory pathology, with a high prevalence and a significant economic and social cost. It is characterized by different clinical phenotypes with different risk profiles. Detecting the correct phenotype, especially for the emphysema subtype, and predicting the risk of major exacerbations are key elements in order to deliver more effective treatments. However, emphysema onset and progression are influenced by a complex interaction between the immune system and the mechanical properties of biological tissue. The former causes chronic inflammation and tissue remodeling. The latter influences the effective resistance or appropriate mechanical response of the lung tissue to repeated breathing cycles. In this work we present a multi-scale model of both aspects, coupling Finite Element (FE) and Agent Based (AB) techniques that we would like to use to predict the onset and progression of emphysema in patients. The AB part is based on existing biological models of inflammation and immunological response as a set of coupled non-linear differential equations. The FE part simulates the biomechanical effects of repeated strain on the biological tissue. We devise a strategy to couple the discrete biological model at the molecular /cellular level and the biomechanical finite element simulations at the tissue level. We tested our implementation on a public emphysema image database and found that it can indeed simulate the evolution of clinical image biomarkers during disease progression.

Lobar Emphysema Distribution Is Associated With 5-Year Radiological Disease Progression

Emphysema has considerable variability in its regional distribution. Craniocaudal emphysema distribution is an important predictor of the response to lung volume reduction. However, there is little consensus regarding how to define upper lobe-predominant and lower lobe-predominant emphysema subtypes. Consequently, the clinical and genetic associations with these subtypes are poorly characterized. We sought to identify subgroups characterized by upper-lobe or lower-lobe emphysema predominance and comparable amounts of total emphysema by analyzing data from 9,210 smokers without alpha-1-antitrypsin deficiency in the Genetic Epidemiology of COPD (COPDGene) cohort. CT densitometric emphysema was measured in each lung lobe. Random forest clustering was applied to lobar emphysema variables after regressing out the effects of total emphysema. Clusters were tested for association with clinical and imaging outcomes at baseline and at 5-year follow-up. Their associations with genetic variants were also compared. Three clusters were identified: minimal emphysema (n= 1,312), upper lobe-predominant emphysema (n= 905), and lower lobe-predominant emphysema (n= 796). Despite a similar amount of total emphysema, the lower-lobe group had more severe airflow obstruction at baseline and higher rates of metabolic syndrome compared with subjects with upper-lobe predominance. The group with upper-lobe predominance had greater 5-year progression of emphysema, gas trapping, and dyspnea. Differential associations with known COPD genetic risk variants were noted. Subgroups of smokers defined by upper-lobe or lower-lobe emphysema predominance exhibit different functional and radiological disease progression rates, and the upper-lobe predominant subtype shows evidence of association with known COPD genetic risk variants. These subgroups may be useful in the development of personalized treatments for COPD.

GENERATIVE METHOD TO DISCOVER EMPHYSEMA SUBTYPES WITH UNSUPERVISED LEARNING USING LUNG MACROSCOPIC PATTERNS (LMPS): THE MESA COPD STUDY.

Pulmonary emphysema overlaps considerably with chronic obstructive pulmonary disease (COPD), and is traditionally subcategorized into three subtypes: centrilobular emphysema (CLE), panlobular emphysema (PLE) and paraseptal emphysema (PSE). Automated classification methods based on supervised learning are generally based upon the current definition of emphysema subtypes, while unsupervised learning of texture patterns enables the objective discovery of possible new radiological emphysema subtypes. In this work, we use a variant of the Latent Dirichlet Allocation (LDA) model to discover lung macroscopic patterns (LMPs) in an unsupervised way from lung regions that encode emphysematous areas. We evaluate the possible utility of the LMPs as potential novel emphysema subtypes via measuring their level of reproducibility when varying the learning set and by their ability to predict traditional radiological emphysema subtypes. Experimental results show that our algorithm can discover highly reproducible LMPs, that predict traditional emphysema subtypes.

A Low Peripheral Blood CD4/CD8 Ratio Is Associated with Pulmonary Emphysema in HIV.

ObjectivesThe prevalence of emphysema is higher among HIV-infected (HIV+) individuals compared to HIV-uninfected persons. While greater tobacco use contributes, HIV-related effects on immunity likely confer additional risk. Low peripheral blood CD4+ to CD8+ T-lymphocyte (CD4/CD8) ratio may reflect chronic inflammation in HIV and may be a marker of chronic lung disease in this population. Therefore, we sought to determine whether the CD4/CD8 ratio was associated with chronic obstructive pulmonary disease (COPD), particularly the emphysema subtype, in a cohort of HIV+ subjects.MethodsWe performed a cross-sectional analysis of 190 HIV+ subjects enrolled in the Examinations of HIV Associated Lung Emphysema (EXHALE) study. Subjects underwent baseline laboratory assessments, pulmonary function testing and chest computed tomography (CT) analyzed for emphysema severity and distribution. We determined the association between CD4/CD8 ratio and emphysema, and the association between CD4/CD8 ratio and pulmonary function markers of COPD.ResultsMild or greater emphysema (>10% lung involvement) was present in 31% of subjects. Low CD4/CD8 ratio was associated with >10% emphysema in multivariable models, adjusting for risk factors including smoking, current and nadir CD4 count and HIV RNA level. Those with CD4/CD8 ratio <0.4 had 6.3 (1.1–39) times the odds of >10% emphysema compared to those with a ratio >1.0 in fully adjusted models. A low CD4/CD8 ratio was also associated with reduced diffusion capacity (DLCO).ConclusionsA low CD4/CD8 ratio was associated with emphysema and low DLCO in HIV+ subjects, independent of other risk factors and clinical markers of HIV. The CD4/CD8 ratio may be a useful, clinically available, marker for risk of emphysema in HIV+ subjects in the antiretroviral therapy (ART) era.

Unsupervised Discovery of Spatially-Informed Lung Texture Patterns for Pulmonary Emphysema: The MESA COPD Study.

Unsupervised discovery of pulmonary emphysema subtypes offers the potential for new definitions of emphysema on lung computed tomography (CT) that go beyond the standard subtypes identified on autopsy. Emphysema subtypes can be defined on CT as a variety of textures with certain spatial prevalence. However, most existing approaches for learning emphysema subtypes on CT are limited to texture features, which are sub-optimal due to the lack of spatial information. In this work, we exploit a standardized spatial mapping of the lung and propose a novel framework for combining spatial and texture information to discover spatially-informed lung texture patterns (sLTPs). Our spatial mapping is demonstrated to be a powerful tool to study emphysema spatial locations over different populations. The discovered sLTPs are shown to have high reproducibility, ability to encode standard emphysema subtypes, and significant associations with clinical characteristics.

Explaining Radiological Emphysema Subtypes with Unsupervised Texture Prototypes: MESA COPD Study.

Pulmonary emphysema is traditionally subcategorized into three subtypes, which have distinct radiological appearances on computed tomography (CT) and can help with the diagnosis of chronic obstructive pulmonary disease (COPD). Automated texture-based quantification of emphysema subtypes has been successfully implemented via supervised learning of these three emphysema subtypes. In this work, we demonstrate that unsupervised learning on a large heterogeneous database of CT scans can generate texture prototypes that are visually homogeneous and distinct, reproducible across subjects, and capable of predicting accurately the three standard radiological subtypes. These texture prototypes enable automated labeling of lung volumes, and open the way to new interpretations of lung CT scans with finer subtyping of emphysema.

A Bayesian Nonparametric Model for Disease Subtyping: Application to Emphysema Phenotypes.

We introduce a novel Bayesian nonparametric model that uses the concept of disease trajectories for disease subtype identification. Although our model is general, we demonstrate that by treating fractions of tissue patterns derived from medical images as compositional data, our model can be applied to study distinct progression trends between population subgroups. Specifically, we apply our algorithm to quantitative emphysema measurements obtained from chest CT scans in the COPDGene Study and show several distinct progression patterns. As emphysema is one of the major components of chronic obstructive pulmonary disease (COPD), the third leading cause of death in the United States [1], an improved definition of emphysema and COPD subtypes is of great interest. We investigate several models with our algorithm, and show that one with age , pack years (a measure of cigarette exposure), and smoking status as predictors gives the best compromise between estimated predictive performance and model complexity. This model identified nine subtypes which showed significant associations to seven single nucleotide polymorphisms (SNPs) known to associate with COPD. Additionally, this model gives better predictive accuracy than multiple, multivariate ordinary least squares regression as demonstrated in a five-fold cross validation analysis. We view our subtyping algorithm as a contribution that can be applied to bridge the gap between CT-level assessment of tissue composition to population-level analysis of compositional trends that vary between disease subtypes.

Texton and sparse representation based texture classification of lung parenchyma in CT images.

Automated texture analysis of lung computed tomography (CT) images is a critical tool in subtyping pulmonary emphysema and diagnosing chronic obstructive pulmonary disease (COPD). Texton-based methods encode lung textures with nearest-texton frequency histograms, and have achieved high performance for supervised classification of emphysema subtypes from annotated lung CT images. In this work, we first explore characterizing lung textures with sparse decomposition from texton dictionaries, using different regularization strategies, and then extend the sparsity-inducing constraint to the construction of the dictionaries. The methods were evaluated on a publicly available lung CT database of annotated emphysema subtypes. We show that enforcing sparse decompositions from texton dictionaries and unsupervised dictionary learning can achieve high classification accuracy (>90%). The flexibility of sparse-inducing models embedded either in the representation stage or dictionary learning stage has potential in providing consistency in classification performance on heterogeneous lung CT datasets with further parameter tuning.