Quantitative CT Characteristics of Cluster Phenotypes in the Severe Asthma Research Program Cohorts

CT strain metrics allow for earlier diagnosis of bronchiolitis obliterans syndrome after hematopoietic cell transplant

ABSTRACT

Bronchiolitis obliterans syndrome (BOS) after hematopoietic cell transplantation (HCT) is associated with substantial morbidity and mortality. Quantitative computed tomography (qCT) can help diagnose advanced BOS meeting National Institutes of Health (NIH) criteria (NIH-BOS) but has not been used to diagnose early, often asymptomatic BOS (early BOS), limiting the potential for early intervention and improved outcomes. Using pulmonary function tests (PFTs) to define NIH-BOS, early BOS, and mixed BOS (NIH-BOS with restrictive lung disease) in patients from 2 large cancer centers, we applied qCT to identify early BOS and distinguish between types of BOS. Patients with transient impairment or healthy lungs were included for comparison. PFTs were done at month 0, 6, and 12. Analysis was performed with association statistics, principal component analysis, conditional inference trees (CITs), and machine learning (ML) classifier models. Our cohort included 84 allogeneic HCT recipients, 66 with BOS (NIH-defined, early, or mixed) and 18 without BOS. All qCT metrics had moderate correlation with forced expiratory volume in 1 second, and each qCT metric differentiated BOS from those without BOS (non-BOS; P < .0001). CITs distinguished 94% of participants with BOS vs non-BOS, 85% of early BOS vs non-BOS, 92% of early BOS vs NIH-BOS. ML models diagnosed BOS with area under the curve (AUC) of 0.84 (95% confidence interval [CI], 0.74-0.94) and early BOS with AUC of 0.84 (95% CI, 0.69-0.97). qCT metrics can identify individuals with early BOS, paving the way for closer monitoring and earlier treatment in this vulnerable population.

Discordant definitions of small airway dysfunction between spirometry and parametric response mapping: the HRCT-based study

OBJECTIVES

To analyze the lung structure of small airway dysfunction (SAD) defined by spirometry and parametric response mapping (PRM) using high-resolution computed tomography (HRCT), and to analyze the predictive factors for SAD.

METHODS

A prospective study was conducted with 388 participants undergoing pulmonary function test (PFT) and inspiratory-expiratory chest CT scans. The clinical data and HRCT assessments of SAD patients defined by both methods were compared. A prediction model for SAD was constructed based on logistic regression.

RESULTS

SAD was defined in 122 individuals by spirometry and 158 by PRM. In HRCT visual assessment, emphysema, tree-in-bud sign, and bronchial wall thickening have higher incidence in SAD defined by each method. (p < 0.001). Quantitative CT showed that spirometry-SAD had thicker airway walls (p < 0.001), smaller lumens (p = 0.011), fewer bronchi (p < 0.001), while PRM-SAD had slender blood vessels. Predictive factors for spirometry-SAD were age, male gender, the volume percentage of emphysema in PRM (PRMEmph), tree-in-bud sign, bronchial wall thickening, bronchial count; for PRM-SAD were age, male gender, BMI, tree-in-bud sign, emphysema, the percentage of blood vessel volume with a cross-sectional area less than 1 mm2 (BV1/TBV). The area under curve (AUC) values for the fitted predictive models were 0.855 and 0.808 respectively.

CONCLUSIONS

Compared with PRM, SAD defined by spirometry is more closely related to airway morphology, while PRM is sensitive to early pulmonary dysfunction but may be interfered by pulmonary vessels. Models combining patient information and HRCT assessment have good predictive value for SAD.

CRITICAL RELEVANCE STATEMENT

HRCT reveals lung structural differences in small airway dysfunction defined by spirometry and parametric response mapping. This insight aids in understanding methodological differences and developing radiological tools for small airways that align with pathophysiology.

KEY POINTS

Spirometry-SAD shows thickened airway walls, narrowed lumen, and reduced branch count, which are closely related to airway morphology. PRM shows good sensitivity to early pulmonary dysfunction, although its assessment of SAD based on gas trapping may be affected by the density of pulmonary vessels and other lung structures. Combining patient information and HRCT features, the fitted model has good predictive performance for SAD defined by both spirometry and PRM (AUC values are 0.855 and 0.808, respectively).

Computed tomography-based measurements associated with rapid lung function decline in severe asthma

BACKGROUND

Patients with severe asthma are susceptible to lung function decline (LFD), but biomarkers that reliably predict an accelerated LFD have not been fully recognized.

OBJECTIVE

To identify variables associated with previous LFD occurrences in patients with severe asthma by exploring the computed tomography (CT) imaging features within predefined LFD groups.

METHODS

We obtained inspiratory and expiratory CT images of 102 patients with severe asthma and derived 2 airway structural parameters (wall thickness [WT] and hydraulic diameter) and 2 parenchymal variables (functional small airway disease and emphysema). We retrospectively calculated the annual changes in forced expiratory volume in 1 second and grouped participants by their values determined. The 4-imaging metrics, along with levels of several biomarkers, were compared among the LFD groups.

RESULTS

Patients with severe asthma with enhanced LFD exhibited significantly lower WT and smaller hydraulic diameter compared with those with minimal change or slight decline in lung function, after an adjustment of smoking status. Conversely, CT-based percentages of emphysema and functional small airway disease did not significantly differ according to LFD. Furthermore, fractional exhaled nitric oxide (FeNO) level and the blood matrix metalloproteinase-9/TIMP metallopeptidase inhibitor 1 ratio were significantly higher in patients with severe asthma with enhanced LFD compared with those in the others.

CONCLUSION

Lower WT on CT scans with increased FeNO that may represent increased airway inflammation significantly correlated with enhanced LFD in patients with severe asthma. Consequently, active management plans may help to attenuate LFD for patients with severe asthma with lower WT and high FeNO.

Novel Machine Learning Identifies 5 Asthma Phenotypes Using Cluster Analysis of Real-World Data

BACKGROUND

Asthma classification into different subphenotypes is important to guide personalized therapy and improve outcomes.

OBJECTIVES

To further explore asthma heterogeneity through determination of multiple patient groups by using novel machine learning (ML) approaches and large-scale real-world data.

METHODS

We used electronic health records of patients with asthma followed at the Cleveland Clinic between 2010 and 2021. We used k-prototype unsupervised ML to develop a clustering model where predictors were age, sex, race, body mass index, prebronchodilator and postbronchodilator spirometry measurements, and the usage of inhaled/systemic steroids. We applied elbow and silhouette plots to select the optimal number of clusters. These clusters were then evaluated through LightGBM's supervised ML approach on their cross-validated F1 score to support their distinctiveness.

RESULTS

Data from 13,498 patients with asthma with available postbronchodilator spirometry measurements were extracted to identify 5 stable clusters. Cluster 1 included a young nonsevere asthma population with normal lung function and higher frequency of acute exacerbation (0.8 /patient-year). Cluster 2 had the highest body mass index (mean ± SD, 44.44 ± 7.83 kg/m2), and the highest proportion of females (77.5%) and Blacks (28.9%). Cluster 3 comprised patients with normal lung function. Cluster 4 included patients with lower percent of predicted FEV1 of 77.03 (12.79) and poor response to bronchodilators. Cluster 5 had the lowest percent of predicted FEV1 of 68.08 (15.02), the highest postbronchodilator reversibility, and the highest proportion of severe asthma (44.9%) and blood eosinophilia (>300 cells/μL) (34.8%).

CONCLUSIONS

Using real-world data and unsupervised ML, we classified asthma into 5 clinically important subphenotypes where group-specific asthma treatment and management strategies can be designed and deployed.

LungViT: Ensembling Cascade of Texture Sensitive Hierarchical Vision Transformers for Cross-Volume Chest CT Image-to-Image Translation

Chest computed tomography (CT) at inspiration is often complemented by an expiratory CT to identify peripheral airways disease. Additionally, co-registered inspiratory-expiratory volumes can be used to derive various markers of lung function. Expiratory CT scans, however, may not be acquired due to dose or scan time considerations or may be inadequate due to motion or insufficient exhale; leading to a missed opportunity to evaluate underlying small airways disease. Here, we propose LungViT- a generative adversarial learning approach using hierarchical vision transformers for translating inspiratory CT intensities to corresponding expiratory CT intensities. LungViT addresses several limitations of the traditional generative models including slicewise discontinuities, limited size of generated volumes, and their inability to model texture transfer at volumetric level. We propose a shifted-window hierarchical vision transformer architecture with squeeze-and-excitation decoder blocks for modeling dependencies between features. We also propose a multiview texture similarity distance metric for texture and style transfer in 3D. To incorporate global information into the training process and refine the output of our model, we use ensemble cascading. LungViT is able to generate large 3D volumes of size 320×320×320 . We train and validate our model using a diverse cohort of 1500 subjects with varying disease severity. To assess model generalizability beyond the development set biases, we evaluate our model on an out-of-distribution external validation set of 200 subjects. Clinical validation on internal and external testing sets shows that synthetic volumes could be reliably adopted for deriving clinical endpoints of chronic obstructive pulmonary disease.

Deep learning parametric response mapping from inspiratory chest CT scans: a new approach for small airway disease screening

OBJECTIVES

Parametric response mapping (PRM) enables the evaluation of small airway disease (SAD) at the voxel level, but requires both inspiratory and expiratory chest CT scans. We hypothesize that deep learning PRM from inspiratory chest CT scans can effectively evaluate SAD in individuals with normal spirometry.

METHODS

We included 537 participants with normal spirometry, a history of smoking or secondhand smoke exposure, and divided them into training, tuning, and test sets. A cascaded generative adversarial network generated expiratory CT from inspiratory CT, followed by a UNet-like network predicting PRM using real inspiratory CT and generated expiratory CT. The performance of the prediction is evaluated using SSIM, RMSE and dice coefficients. Pearson correlation evaluated the correlation between predicted and ground truth PRM. ROC curves evaluated predicted PRMfSAD (the volume percentage of functional small airway disease, fSAD) performance in stratifying SAD.

RESULTS

Our method can generate expiratory CT of good quality (SSIM 0.86, RMSE 80.13 HU). The predicted PRM dice coefficients for normal lung, emphysema, and fSAD regions are 0.85, 0.63, and 0.51, respectively. The volume percentages of emphysema and fSAD showed good correlation between predicted and ground truth PRM (|r| were 0.97 and 0.64, respectively, p < 0.05). Predicted PRMfSAD showed good SAD stratification performance with ground truth PRMfSAD at thresholds of 15%, 20% and 25% (AUCs were 0.84, 0.78, and 0.84, respectively, p < 0.001).

CONCLUSION

Our deep learning method generates high-quality PRM using inspiratory chest CT and effectively stratifies SAD in individuals with normal spirometry.

Determining asthma endotypes and outcomes: Complementing existing clinical practice with modern machine learning

There is unprecedented opportunity to use machine learning to integrate high-dimensional molecular data with clinical characteristics to accurately diagnose and manage disease. Asthma is a complex and heterogeneous disease and cannot be solely explained by an aberrant type 2 (T2) immune response. Available and emerging multi-omics datasets of asthma show dysregulation of different biological pathways including those linked to T2 mechanisms. While T2-directed biologics have been life changing for many patients, they have not proven effective for many others despite similar biomarker profiles. Thus, there is a great need to close this gap to understand asthma heterogeneity, which can be achieved by harnessing and integrating the rich multi-omics asthma datasets and the corresponding clinical data. This article presents a compendium of machine learning approaches that can be utilized to bridge the gap between predictive biomarkers and actual causal signatures that are validated in clinical trials to ultimately establish true asthma endotypes.