129Xe MRI Ventilation Defects in Asthma: What is the Upper Limit of Normal and Minimal Clinically Important Difference?

129Xe-MRI ventilation and acinar abnormalities highlight the significance of spirometric dysanapsis: findings from the NOVELTY ADPro UK substudy

RATIONALE

Airways dysanapsis is defined by CT or spirometry as a mismatch between the size of the airways and lung volume and is associated with increased risk of developing chronic obstructive pulmonary disease (COPD). Lung disease in participants with dysanapsis and a label of asthma and/or COPD remains poorly understood.

METHODS

In participants with asthma and/or COPD, we used 129Xe-MRI to assess ventilation, acinar dimensions and gas exchange, and pulmonary function tests, and compared people with spirometric dysanapsis (forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC)<-1.64 z and FEV1>-1.64 z) to those with normal spirometry (FEV1, FVC and FEV1/FVC>-1.64 z).

RESULTS

From 165 participants assessed in the NOVELTY (NOVEL observational longiTudinal studY) ADPro (advanced diagnostic profiling) study with a physician-assigned diagnosis of asthma and/or COPD, 43 had spirometric dysanapsis and were age-matched to 43 participants with normal spirometry. Participants with dysanapsis had significantly increased ventilation defects (median difference (md) (95% CI) = 4.0% (1.42% to 5.38%), p<0.001), ventilation heterogeneity (md (95% CI) = 2.56% (1.31% to 3.56%), p<0.001) and measures of acinar dimensions (md (95% CI) = 0.004 cm2.s-1 (0.0009 to 0.007), p=0.009) from 129Xe-MRI, than those with normal spirometry. At the 1-year follow-up, only participants with dysanapsis had a significant increase in ventilation defects (md (95% CI)=0.45% (0.09% to 2.1%),p=0.016). Lower FEV1/FVC in the dysanapsis cohort was associated with increased ventilation defects (r=-0.64, R2=0.41, p<0.001) and increased acinar dimensions (r=-0.52, R2=0.38, p<0.001), with the highest values seen in those with an FVC above the upper limit of normal.

CONCLUSIONS

Participants with asthma and/or COPD, presenting to primary care with spirometric dysanapsis, exhibited increased lung abnormalities on 129Xe-MRI, when compared with those with normal spirometry. Spirometric dysanapsis in asthma and/or COPD is therefore associated with significant lung disease, and the FEV1/FVC is related to the degree of airways abnormality on 129Xe-MRI.

Quantification of 129Xe MRI Ventilation-defect-percent Using Binary-threshold, Gaussian Linear-Binning and K-means Methods: Differences in Asthma and COPD

RATIONALE AND OBJECTIVES

Hyperpolarized 129Xe magnetic resonance imaging (MRI) provides a way to quantify ventilation heterogeneity as ventilation defect percent (VDP), calculated as the volume of unventilated lung volume normalized to the thoracic cavity volume. Currently used methods for quantifying VDP include (1) binary signal-intensity thresholds (Binary-threshold, BT), (2) Gaussian transformation of signal-intensity histogram with standard deviation thresholds or Gaussian-linear-binning (GLB), and (3) iterative centroid-based clustering of the signal-intensity histogram (k-means). These methods have not been directly compared in patients with asthma and chronic obstructive pulmonary disease (COPD), in whom ventilation defects are hallmark findings. Our objective was to quantify and compare VDP using these four different methods.

PATIENTS AND METHODS

Data from 175 participants (n=42 healthy, n=43 COPD, n=90 asthma) were retrospectively evaluated using a CNN co-registration and segmentation pipeline and GLB, GLBslice, (slice-wise evaluation of GLB) BT and k-means VDP quantification methods. Linear-regression and Bland-Altman plots were used to quantify inter-method correlations and agreement.

RESULTS

VDP was significantly different using GLB (Asthma: 6±9%, COPD: 7±7%, p<.001) and BT (Asthma: 6±7%, COPD: 10±8%, p<.001) methods compared to GLBslice (Asthma: 12±13%, COPD: 16±15%, p<.001) and k-means (Asthma: 12±12%, COPD: 25±17%, p<.001). VDP calculated using GLB (R2=.64, p<.001), GLBslice (R2=.84, p<.001) and BT (R2=.84, p<.001) was significantly correlated with k-means VDP. Bland-Altman plots revealed wide 95% confidence intervals of agreement for k-means with GLB/GLBslice (COPD -6%/-1%: 42%/23%; asthma -5%/-10%:16%/10%) and BT (COPD -4%:36%; asthma -6%:19%).

CONCLUSION

VDP differences in patients with asthma and COPD calculated using four methods are important to consider for multi-center studies.

Pulmonary 129Xe MRI: CNN Registration and Segmentation to Generate Ventilation Defect Percent with Multi-center Validation

RATIONALE AND OBJECTIVES

Hyperpolarized 129Xe MRI quantifies ventilation-defect-percent (VDP), the ratio of 129Xe signal-void to the anatomic 1H MRI thoracic-cavity-volume. VDP is associated with airway inflammation and disease control and serves as a treatable trait in therapy studies. Semi-automated VDP pipelines require time-intensive observer interactions. Current convolutional neural network (CNN) approaches for quantifying VDP lack external validation, which limits multicenter utilization. Our objective was to develop an automated and externally validated deep-learning pipeline to quantify pulmonary 129Xe MRI VDP.

MATERIALS AND METHODS

1H and 129Xe MRI data from the primary site (Site1) were used to train and test a CNN segmentation and registration pipeline, while two independent sites (Site2 and Site3) provided external validation. Semi-automated and CNN-based registration error was measured using mean-absolute-error (MAE) while segmentation error was measured using generalized-Dice-similarity coefficient (gDSC). CNN and semi-automated VDP were compared using linear regression and Bland-Altman analysis.

RESULTS

Training/testing used data from 205 participants (healthy volunteers, asthma, COPD, long-COVID; mean age=54 ± 16y; 119 females) from Site1. External validation used data from 71 participants. CNN and semi-automated 1H and 129Xe registrations agreed (MAE=0.3°, R2 =0.95 rotation; 1.1%, R2 =0.79 scaling; 0.2/0.5px, R2 =0.96/0.95, x/y-translation; all p < .001). Thoracic-cavity and ventilation segmentations were also spatially corresponding (gDSC=0.92 and 0.88, respectively). CNN VDP correlated with semi-automated VDP (Site1 R2/ρ = .97/.95, bias=-0.5%; Site2 R2/ρ = .85/.93, bias=-0.9%; Site3 R2/ρ = .95/.89, bias=-0.8%, all p < .001).

CONCLUSION

An externally validated CNN registration/segmentation model demonstrated strong agreement with low error compared to the semi-automated method. CNN and semi-automated registrations, thoracic-cavity-volume and ventilation-volume segmentations were highly correlated with high gDSC for the datasets.

New frontiers in asthma chest imaging

Modern pulmonary imaging can reveal underlying pathologic and pathophysiologic changes in the lungs of people with asthma, with important clinical implications. A multitude of imaging modalities, including computed tomography, magnetic resonance imaging, optical coherence tomography, and endobronchial ultrasound, are now being used to examine underlying structure-function relationships. Imaging-based biomarkers from these techniques, including airway dimensions, blood vessel volumes, mucus scores, extent of ventilation defect, and extent of air trapping, often have increased sensitivity compared with that of traditional lung function measurements and are increasingly being used as end points in clinical trials. Imaging has been crucial to recent improvements in our understanding of the relationships between type 2 inflammation, eosinophilia, and mucus extent. With the advent of effective anti-type 2 biologic therapies, computed tomography and magnetic resonance imaging techniques can identify not just which patients benefit from therapy but why they benefit. Clinical trials have begun to assess the utility of imaging to prospectively plan airway therapy targets in bronchial thermoplasty and have potential to direct future bronchoscopic therapies. Together, imaging techniques provide a diverse set of tools to investigate how spatially distributed airway, blood, and parenchymal abnormalities shape disease heterogeneity in patients with asthma.

Hyperpolarized Gas Imaging in Lung Diseases: Functional and Artificial Intelligence Perspective

Pathophysiologic changes in lung diseases are often accompanied by changes in ventilation and gas exchange. Comprehensive evaluation of lung function cannot be obtained through chest X-ray and computed tomography. Proton-based lung MRI is particularly challenging due to low proton density within the lung tissue. In this review, we discuss an emerging technology--hyperpolarized gas MRI with inhaled 129Xe, which provides functional and microstructural information and has the potential as a clinical tool for detecting the early stage and progression of certain lung diseases. We review the hyperpolarized 129Xe MRI studies in patients with a range of pulmonary diseases, including chronic obstructive pulmonary disease, asthma, cystic fibrosis, pulmonary hypertension, radiation-induced lung injury and interstitial lung disease, and the applications of artificial intelligence were reviewed as well.

129Xe MRI Ventilation Textures and Longitudinal Quality-of-Life Improvements in Long-COVID

RATIONALE AND OBJECTIVES

It remains difficult to predict longitudinal outcomes in long-COVID, even with chest CT and functional MRI. 129Xe MRI reflects airway dysfunction, measured using ventilation defect percent (VDP) and in long-COVID patients, MRI VDP was abnormal, suggestive of airways disease. While MRI VDP and quality-of-life improved 15-month post-COVID infection, both remained abnormal. To better understand the relationship of airways disease and quality-of-life improvements in patients with long-COVID, we extracted 129Xe ventilation MRI textures and generated machine-learning models in an effort to predict improved quality-of-life, 15-month post-infection.

MATERIALS AND METHODS

Long-COVID patients provided written-informed consent to 3-month and 15-month post-infection visits. Pyradiomics was used to extract 129Xe ventilation MRI texture features, which were ranked using a Random-Forest classifier. Top-ranking features were used in classification models to dichotomize patients based on St. George's Respiratory Questionnaire (SGRQ) score improvement greater than the minimal-clinically-important-difference (MCID). Classification performance was evaluated using the area under the receiver-operator-characteristic-curve (AUC), sensitivity, and specificity.

RESULTS

120 texture features were extracted from 129Xe ventilation MRI in 44 long-COVID participants (54 ± 14 years), including 30 (52 ± 12 years) with ΔSGRQ≥MCID and 14 (58 ± 18 years) with ΔSGRQ

CONCLUSION

A machine learning model exclusively trained on 129Xe MRI ventilation textures explained improved SGRQ-scores 12 months later, and outperformed clinical models. Their unique spatial-intensity information helps build our understanding about long-COVID airway dysfunction.

Collapse

Key Words Collapse

MESH Headings

• Humans
• COVID-19/diagnostic imaging
• COVID-19/complications
• Magnetic Resonance Imaging/methods
• Male
• Female
• Middle Aged
• Quality of Life
• Xenon Isotopes
• Aged
• Machine Learning
• Longitudinal Studies
• SARS-CoV-2
• Lung/diagnostic imaging

Collapse

Grants Collapse

Affiliation(s)

Harkiran K Kooner Robarts Research Institute, Western University, London, Canada; Department of Medical Biophysics, Western University, London, Canada
Maksym Sharma Robarts Research Institute, Western University, London, Canada; Department of Medical Biophysics, Western University, London, Canada
Marrissa J McIntosh Robarts Research Institute, Western University, London, Canada; Department of Medical Biophysics, Western University, London, Canada
Inderdeep Dhaliwal Division of Respirology, Department of Medicine, Western University, London, Canada
J Michael Nicholson Division of Respirology, Department of Medicine, Western University, London, Canada
Miranda Kirby Department of Physics, Toronto Metropolitan University, Toronto, Canada
Sarah Svenningsen Division of Respirology, Department of Medicine, McMaster University and Firestone Institute for Respiratory Health, St. Joseph's Health Care, Hamilton, Canada
Grace Parraga Robarts Research Institute, Western University, London, Canada; Department of Medical Biophysics, Western University, London, Canada; Division of Respirology, Department of Medicine, Western University, London, Canada.

Collapse

Collaborators Collapse

129Xe MRI and Oscillometry of Irritant-Induced Asthma After Bronchial Thermoplasty

Irritant-induced asthma (IIA) may develop after acute inhalational exposure in individuals without preexisting asthma. The effect of bronchial thermoplasty to treat intractable, worsening IIA has not yet been described. We evaluated a previously healthy 52-year-old man after inhalation of an unknown white powder. His pulmonary function and symptoms/quality of life worsened over 4 years, despite maximal guidelines-based asthma therapy. We acquired 129Xe MRI and pulmonary function test measurements on three occasions including before and after bronchial thermoplasty treatment. Seven months after bronchial thermoplasty, improved MRI ventilation and oscillometry small airway resistance were observed. Spirometry and asthma control did not improve until 19 months after bronchial thermoplasty, 5.5 years postexposure. Together, oscillometry measurements of the small airways and 129Xe MRI provided effort-independent, sensitive, and objective measurements of response to therapy. Improved MRI and oscillometry small airway resistance measurements temporally preceded improved airflow obstruction and may be considered for complex asthma cases.

Combining neural networks and image synthesis to enable automatic thoracic cavity segmentation of hyperpolarized 129Xe MRI without proton scans

RATIONALE AND OBJECTIVES

Quantification of 129Xe MRI relies on accurate segmentation of the thoracic cavity, typically performed manually using a combination of 1H and 129Xe scans. This can be accelerated by using Convolutional Neural Networks (CNNs) that segment only the 129Xe scan. However, this task is complicated by peripheral ventilation defects, which requires training CNNs with large, diverse datasets. Here, we accelerate the creation of training data by synthesizing 129Xe images with a variety of defects. We use this to train a 3D model to provide thoracic cavity segmentation from 129Xe ventilation MRI alone.

MATERIALS AND METHODS

Training and testing data consisted of 22 and 33 3D 129Xe ventilation images. Training data were expanded to 484 using Template-based augmentation while an additional 298 images were synthesized using the Pix2Pix model. This data was used to train both a 2D U-net and 3D V-net-based segmentation model using a combination of Dice-Focal and Anatomical Constraint loss functions. Segmentation performance was compared using Dice coefficients calculated over the entire lung and within ventilation defects.

RESULTS

Performance of both U-net and 3D segmentation was improved by including synthetic training data. The 3D models performed significantly better than U-net, and the 3D model trained with synthetic 129Xe images exhibited the highest overall Dice score of 0.929. Moreover, addition of synthetic training data improved the Dice score in ventilation defect regions from 0.545 to 0.588 for U-net and 0.739 to 0.765 for the 3D model.

CONCLUSION

It is feasible to obtain high-quality segmentations from 129Xe scan alone using 3D models trained with additional synthetic images.

Personalized Medicine in Asthma: Current Approach and Future Perspectives

Asthma is one of the most common chronic respiratory diseases, affecting over 300 million people worldwide [...].