A Dual-Channel Deep Learning Approach for Lung Cavity Estimation From Hyperpolarized Gas and Proton MRI.

BACKGROUND: Hyperpolarized gas MRI can quantify regional lung ventilation via biomarkers, including the ventilation defect percentage (VDP). VDP is computed from segmentations derived from spatially co-registered functional hyperpolarized gas and structural proton (1 H)-MRI. Although acquired at similar lung inflation levels, they are frequently misaligned, requiring a lung cavity estimation (LCE). Recently, single-channel, mono-modal deep learning (DL)-based methods have shown promise for pulmonary image segmentation problems. Multichannel, multimodal approaches may outperform single-channel alternatives. PURPOSE: We hypothesized that a DL-based dual-channel approach, leveraging both 1 H-MRI and Xenon-129-MRI (129 Xe-MRI), can generate LCEs more accurately than single-channel alternatives. STUDY TYPE: Retrospective. POPULATION: A total of 480 corresponding 1 H-MRI and 129 Xe-MRI scans from 26 healthy participants (median age [range]: 11 [8-71]; 50% females) and 289 patients with pulmonary pathologies (median age [range]: 47 [6-83]; 51% females) were split into training (422 scans [88%]; 257 participants [82%]) and testing (58 scans [12%]; 58 participants [18%]) sets. FIELD STRENGTH/SEQUENCE: 1.5-T, three-dimensional (3D) spoiled gradient-recalled 1 H-MRI and 3D steady-state free-precession 129 Xe-MRI. ASSESSMENT: We developed a multimodal DL approach, integrating 129 Xe-MRI and 1 H-MRI, in a dual-channel convolutional neural network. We compared this approach to single-channel alternatives using manually edited LCEs as a benchmark. We further assessed a fully automatic DL-based framework to calculate VDPs and compared it to manually generated VDPs. STATISTICAL TESTS: Friedman tests with post hoc Bonferroni correction for multiple comparisons compared single-channel and dual-channel DL approaches using Dice similarity coefficient (DSC), average boundary Hausdorff distance (average HD), and relative error (XOR) metrics. Bland-Altman analysis and paired t-tests compared manual and DL-generated VDPs. A P value < 0.05 was considered statistically significant. RESULTS: The dual-channel approach significantly outperformed single-channel approaches, achieving a median (range) DSC, average HD, and XOR of 0.967 (0.867-0.978), 1.68 mm (37.0-0.778), and 0.066 (0.246-0.045), respectively. DL-generated VDPs were statistically indistinguishable from manually generated VDPs (P = 0.710). DATA CONCLUSION: Our dual-channel approach generated LCEs, which could be integrated with ventilated lung segmentations to produce biomarkers such as the VDP without manual intervention. EVIDENCE LEVEL: 4. TECHNICAL EFFICACY: Stage 1.

Quantification of pulmonary perfusion in idiopathic pulmonary fibrosis with first pass dynamic contrast-enhanced perfusion MRI.

INTRODUCTION: Idiopathic pulmonary fibrosis (IPF) is a fatal disease of lung scarring. Many patients later develop raised pulmonary vascular pressures, sometimes disproportionate to the interstitial disease. Previous therapeutic approaches that have targeted pulmonary vascular changes have not demonstrated clinical efficacy, and quantitative assessment of regional pulmonary vascular involvement using perfusion imaging may provide a biomarker for further therapeutic insights. METHODS: We studied 23 participants with IPF, using dynamic contrast-enhanced MRI (DCE-MRI) and pulmonary function tests, including forced vital capacity (FVC), transfer factor (TLCO) and coefficient (KCO) of the lungs for carbon monoxide. DCE-MRI parametric maps were generated including the full width at half maximum (FWHM) of the bolus transit time through the lungs. Key metrics used were mean (FWHMmean) and heterogeneity (FWHMIQR). Nineteen participants returned at 6 months for repeat assessment. RESULTS: Spearman correlation coefficients were identified between TLCO and FWHMIQR (r=-0.46; p=0.026), KCO and FWHMmean (r=-0.42; p=0.047) and KCO and FWHMIQR (r=-0.51; p=0.013) at baseline. No statistically significant correlations were seen between FVC and DCE-MRI metrics. Follow-up at 6 months demonstrated statistically significant decline in FVC (p=0.040) and KCO (p=0.014), with an increase in FWHMmean (p=0.040), but no significant changes in TLCO (p=0.090) nor FWHMIQR (p=0.821). CONCLUSIONS: DCE-MRI first pass perfusion demonstrates correlations with existing physiological gas exchange metrics, suggesting that capillary perfusion deficit (as well as impaired interstitial diffusion) may contribute to gas exchange limitation in IPF. FWHMmean showed a significant increase over a 6-month period and has potential as a quantitative biomarker of pulmonary vascular disease progression in IPF.

Dissolved 129 Xe lung MRI with four-echo 3D radial spectroscopic imaging: Quantification of regional gas transfer in idiopathic pulmonary fibrosis.

PURPOSE: Imaging of the different resonances of dissolved hyperpolarized xenon-129 (129 Xe) in the lung is performed using a four-echo flyback 3D radial spectroscopic imaging technique and is evaluated in healthy volunteers (HV) and subjects with idiopathic pulmonary fibrosis (IPF). THEORY AND METHODS: 10 HV and 25 subjects with IPF underwent dissolved 129 Xe MRI at 1.5T. IPF subjects underwent same day pulmonary function tests to measure forced vital capacity and the diffusion capacity of the lung for carbon monoxide (DLCO ). A four-point echo time technique with k-space chemical-shift modeling of gas, dissolved 129 Xe in lung tissue/plasma (TP) and red blood cells (RBC) combined with a 3D radial trajectory was implemented within a 14-s breath-hold. RESULTS: Results show an excellent chemical shift separation of the dissolved 129 Xe compartments and gas contamination removal, confirmed by a strong agreement between average imaging and global spectroscopy RBC/TP ratio measurements. Subjects with IPF exhibited reduced imaging gas transfer when compared to HV. A significant increase of the amplitude of RBC signal cardiogenic oscillation was also observed. In IPF subjects, DLCO % predicted was significantly correlated with RBC/TP and RBC/GAS ratios and the correlations were stronger in the inferior and periphery sections of the lungs. CONCLUSION: Lung MRI of dissolved 129 Xe was performed with a four-echo spectroscopic imaging method. Subjects with IPF demonstrated reduced xenon imaging gas transfer and increased cardiogenic modulation of dissolved xenon signal in the RBCs when compared to HV.

Free breathing lung T1 mapping using image registration in patients with idiopathic pulmonary fibrosis.

PURPOSE: To assess the use of image registration for correcting respiratory motion in free breathing lung T1 mapping acquisition in patients with idiopathic pulmonary fibrosis (IPF). THEORY AND METHODS: The method presented used image registration to synthetic images during postprocessing to remove respiratory motion. Synthetic images were generated from a model of the inversion recovery signal of the acquired images that incorporated a periodic lung motion model. Ten healthy volunteers and 19 patients with IPF underwent 2D Look-Locker T1 mapping acquisition at 1.5T during inspiratory breath-hold and free breathing. Eight healthy volunteers and seven patients with IPF underwent T1 mapping acquisition during expiratory breath-hold. Fourteen patients had follow-up scanning at 6 months. Dice similarity coefficient (DSC) was used to evaluate registration efficacy. RESULTS: Image registration increased image DSC (P < .001) in the free breathing inversion recovery images. Lung T1 measured during a free breathing acquisition was lower in patients with IPF when compared with healthy controls (inspiration: P = .238; expiration: P = .261; free breathing: P = .021). Measured lung T1 was higher in expiration breath-hold than inspiration breath-hold in healthy volunteers (P < .001) but not in patients with IPF (P = .645). There were no other significant differences between lung T1 values within subject groups. CONCLUSIONS: The registration technique significantly reduced motion in the Look-Locker images acquired during free breathing and may improve the robustness of lung T1 mapping in patients who struggle to hold their breath. Lung T1 measured during a free breathing acquisition was significantly lower in patients with IPF when compared with healthy controls.

Experimental and quantitative imaging techniques in interstitial lung disease.

Interstitial lung diseases (ILDs) are a heterogeneous group of conditions, with a wide and complex variety of imaging features. Difficulty in monitoring, treating and exploring novel therapies for these conditions is in part due to the lack of robust, readily available biomarkers. Radiological studies are vital in the assessment and follow-up of ILD, but currently CT analysis in clinical practice is qualitative and therefore somewhat subjective. In this article, we report on the role of novel and quantitative imaging techniques across a range of imaging modalities in ILD and consider how they may be applied in the assessment and understanding of ILD. We critically appraised evidence found from searches of Ovid online, PubMed and the TRIP database for novel and quantitative imaging studies in ILD. Recent studies have explored the capability of texture-based lung parenchymal analysis in accurately quantifying several ILD features. Newer techniques are helping to overcome the challenges inherent to such approaches, in particular distinguishing peripheral reticulation of lung parenchyma from pleura and accurately identifying the complex density patterns that accompany honeycombing. Robust and validated texture-based analysis may remove the subjectivity that is inherent to qualitative reporting and allow greater objective measurements of change over time. In addition to lung parenchymal feature quantification, pulmonary vessel volume analysis on CT has demonstrated prognostic value in two retrospective analyses and may be a sign of vascular changes in ILD which, to date, have been difficult to quantify in the absence of overt pulmonary hypertension. Novel applications of existing imaging techniques, such as hyperpolarised gas MRI and positron emission tomography (PET), show promise in combining structural and functional information. Although structural imaging of lung tissue is inherently challenging in terms of conventional proton MRI techniques, inroads are being made with ultrashort echo time, and dynamic contrast-enhanced MRI may be used for lung perfusion assessment. In addition, inhaled hyperpolarised 129Xenon gas MRI may provide multifunctional imaging metrics, including assessment of ventilation, intra-acinar gas diffusion and alveolar-capillary diffusion. PET has demonstrated high standard uptake values (SUVs) of 18F-fluorodeoxyglucose in fibrosed lung tissue, challenging the assumption that these are 'burned out' and metabolically inactive regions. Regions that appear structurally normal also appear to have higher SUV, warranting further exploration with future longitudinal studies to assess if this precedes future regions of macroscopic structural change. Given the subtleties involved in diagnosing, assessing and predicting future deterioration in many forms of ILD, multimodal quantitative lung structure-function imaging may provide the means of identifying novel, sensitive and clinically applicable imaging markers of disease. Such imaging metrics may provide mechanistic and phenotypic information that can help direct appropriate personalised therapy, can be used to predict outcomes and could potentially be more sensitive and specific than global pulmonary function testing. Quantitative assessment may objectively assess subtle change in character or extent of disease that can assist in efficacy of antifibrotic therapy or detecting early changes of potentially pneumotoxic drugs involved in early intervention studies.

Imaging biomarkers of lung ventilation in interstitial lung disease from 129Xe and oxygen enhanced 1H MRI.

PURPOSE: To compare imaging biomarkers from hyperpolarised 129Xe ventilation MRI and dynamic oxygen-enhanced MRI (OE-MRI) with standard pulmonary function tests (PFT) in interstitial lung disease (ILD) patients. To evaluate if biomarkers can separate ILD subtypes and detect early signs of disease resolution or progression. STUDY TYPE: Prospective longitudinal. POPULATION: Forty-one ILD (fourteen idiopathic pulmonary fibrosis (IPF), eleven hypersensitivity pneumonitis (HP), eleven drug-induced ILD (DI-ILD), five connective tissue disease related-ILD (CTD-ILD)) patients and ten healthy volunteers imaged at visit 1. Thirty-four ILD patients completed visit 2 (eleven IPF, eight HP, ten DIILD, five CTD-ILD) after 6 or 26 weeks. FIELD STRENGTH/SEQUENCE: MRI was performed at 1.5 T, including inversion recovery T1 mapping, dynamic MRI acquisition with varying oxygen levels, and hyperpolarised 129Xe ventilation MRI. Subjects underwent standard spirometry and gas transfer testing. ASSESSMENT: Five 1H MRI and two 129Xe MRI ventilation metrics were compared with spirometry and gas transfer measurements. STATISTICAL TEST: To evaluate differences at visit 1 among subgroups: ANOVA or Kruskal-Wallis rank tests with correction for multiple comparisons. To assess the relationships between imaging biomarkers, PFT, age and gender, at visit 1 and for the change between visit 1 and 2: Pearson correlations and multilinear regression models. RESULTS: The global PFT tests could not distinguish ILD subtypes. Percentage ventilated volumes were lower in ILD patients than in HVs when measured with 129Xe MRI (HV 97.4 ± 2.6, CTD-ILD: 91.0 ± 4.8 p = 0.017, DI-ILD 90.1 ± 7.4 p = 0.003, HP 92.6 ± 4.0 p = 0.013, IPF 88.1 ± 6.5 p < 0.001), but not with OE-MRI. 129Xe reported more heterogeneous ventilation in DI-ILD and IPF than in HV, and OE-MRI reported more heterogeneous ventilation in DI-ILD and IPF than in HP or CTD-ILD. The longitudinal changes reported by the imaging biomarkers did not correlate with the PFT changes between visits. DATA CONCLUSION: Neither 129Xe ventilation nor OE-MRI biomarkers investigated in this study were able to differentiate between ILD subtypes, suggesting that ventilation-only biomarkers are not indicated for this task. Limited but progressive loss of ventilated volume as measured by 129Xe-MRI may be present as the biomarker of focal disease progresses. OE-MRI biomarkers are feasible in ILD patients and do not correlate strongly with PFT. Both OE-MRI and 129Xe MRI revealed more spatially heterogeneous ventilation in DI-ILD and IPF.

Hyperpolarised xenon-129 diffusion-weighted magnetic resonance imaging for assessing lung microstructure in idiopathic pulmonary fibrosis.

Background: Hyperpolarised 129-xenon (129Xe) magnetic resonance imaging (MRI) shows promise in monitoring the progression of idiopathic pulmonary fibrosis (IPF) due to the lack of ionising radiation and the ability to quantify functional impairment. Diffusion-weighted (DW)-MRI with hyperpolarised gases can provide information about lung microstructure. The aims were to compare 129Xe DW-MRI measurements with pulmonary function tests (PFTs), and to assess whether they can detect early signs of disease progression in patients with newly diagnosed IPF. Methods: This is a prospective, single-centre, observational imaging study of patients presenting with IPF to Northern General Hospital (Sheffield, UK). Hyperpolarised 129Xe DW-MRI was performed at 1.5 T on a whole-body General Electric HDx scanner and PFTs were performed on the same day as the MRI scan. Results: There was an increase in global 129Xe apparent diffusion coefficient (ADC) between the baseline and 12-month visits (mean 0.043 cm2·s-1, 95% CI 0.040-0.047 cm2·s-1 versus mean 0.045 cm2·s-1, 95% CI 0.040-0.049 cm2·s-1; p=0.044; n=20), with no significant change in PFTs over the same time period. There was also an increase in 129Xe ADC in the lower zone (p=0.027), and an increase in 129Xe mean acinar dimension in the lower zone (p=0.033) between the baseline and 12-month visits. 129Xe DW-MRI measurements correlated strongly with diffusing capacity of the lung for carbon monoxide (% predicted), transfer coefficient of the lung for carbon monoxide (KCO) and KCO (% predicted). Conclusions: 129Xe DW-MRI measurements appear to be sensitive to early changes of microstructural disease that are consistent with progression in IPF at 12 months. As new drug treatments are developed, the ability to quantify subtle changes using 129Xe DW-MRI could be particularly valuable.

Longitudinal Lung Function Assessment of Patients Hospitalized With COVID-19 Using 1H and 129Xe Lung MRI.

BACKGROUND: Microvascular abnormalities and impaired gas transfer have been observed in patients with COVID-19. The progression of pulmonary changes in these patients remains unclear. RESEARCH QUESTION: Do patients hospitalized with COVID-19 without evidence of architectural distortion on structural imaging exhibit longitudinal improvements in lung function measured by using 1H and 129Xe MRI between 6 and 52 weeks following hospitalization? STUDY DESIGN AND METHODS: Patients who were hospitalized with COVID-19 pneumonia underwent a pulmonary 1H and 129Xe MRI protocol at 6, 12, 25, and 51 weeks following hospital admission in a prospective cohort study between November 2020 and February 2022. The imaging protocol was as follows: 1H ultra-short echo time, contrast-enhanced lung perfusion, 129Xe ventilation, 129Xe diffusion-weighted, and 129Xe spectroscopic imaging of gas exchange. RESULTS: Nine patients were recruited (age 57 ± 14 [median ± interquartile range] years; six of nine patients were male). Patients underwent MRI at 6 (n = 9), 12 (n = 9), 25 (n = 6), and 51 (n = 8) weeks following hospital admission. Patients with signs of interstitial lung damage were excluded. At 6 weeks, patients exhibited impaired 129Xe gas transfer (RBC to membrane fraction), but lung microstructure was not increased (apparent diffusion coefficient and mean acinar airway dimensions). Minor ventilation abnormalities present in four patients were largely resolved in the 6- to 25-week period. At 12 weeks, all patients with lung perfusion data (n = 6) showed an increase in both pulmonary blood volume and flow compared with 6 weeks, although this was not statistically significant. At 12 weeks, significant improvements in 129Xe gas transfer were observed compared with 6-week examinations; however, 129Xe gas transfer remained abnormally low at weeks 12, 25, and 51. INTERPRETATION: 129Xe gas transfer was impaired up to 1 year following hospitalization in patients who were hospitalized with COVID-19 pneumonia, without evidence of architectural distortion on structural imaging, whereas lung ventilation was normal at 52 weeks.

Large-scale investigation of deep learning approaches for ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI.

Respiratory diseases are leading causes of mortality and morbidity worldwide. Pulmonary imaging is an essential component of the diagnosis, treatment planning, monitoring, and treatment assessment of respiratory diseases. Insights into numerous pulmonary pathologies can be gleaned from functional lung MRI techniques. These include hyperpolarized gas ventilation MRI, which enables visualization and quantification of regional lung ventilation with high spatial resolution. Segmentation of the ventilated lung is required to calculate clinically relevant biomarkers. Recent research in deep learning (DL) has shown promising results for numerous segmentation problems. Here, we evaluate several 3D convolutional neural networks to segment ventilated lung regions on hyperpolarized gas MRI scans. The dataset consists of 759 helium-3 (3He) or xenon-129 (129Xe) volumetric scans and corresponding expert segmentations from 341 healthy subjects and patients with a wide range of pathologies. We evaluated segmentation performance for several DL experimental methods via overlap, distance and error metrics and compared them to conventional segmentation methods, namely, spatial fuzzy c-means (SFCM) and K-means clustering. We observed that training on combined 3He and 129Xe MRI scans using a 3D nn-UNet outperformed other DL methods, achieving a mean ± SD Dice coefficient of 0.963 ± 0.018, average boundary Hausdorff distance of 1.505 ± 0.969 mm, Hausdorff 95th percentile of 5.754 ± 6.621 mm and relative error of 0.075 ± 0.039. Moreover, limited differences in performance were observed between 129Xe and 3He scans in the testing set. Combined training on 129Xe and 3He yielded statistically significant improvements over the conventional methods (p < 0.0001). In addition, we observed very strong correlation and agreement between DL and expert segmentations, with Pearson correlation of 0.99 (p < 0.0001) and Bland-Altman bias of - 0.8%. The DL approach evaluated provides accurate, robust and rapid segmentations of ventilated lung regions and successfully excludes non-lung regions such as the airways and artefacts. This approach is expected to eliminate the need for, or significantly reduce, subsequent time-consuming manual editing.

Lung MRI with hyperpolarised gases: current & future clinical perspectives.

The use of pulmonary MRI in a clinical setting has historically been limited. Whilst CT remains the gold-standard for structural lung imaging in many clinical indications, technical developments in ultrashort and zero echo time MRI techniques are beginning to help realise non-ionising structural imaging in certain lung disorders. In this invited review, we discuss a complementary technique - hyperpolarised (HP) gas MRI with inhaled 3He and 129Xe - a method for functional and microstructural imaging of the lung that has great potential as a clinical tool for early detection and improved understanding of pathophysiology in many lung diseases. HP gas MRI now has the potential to make an impact on clinical management by enabling safe, sensitive monitoring of disease progression and response to therapy. With reference to the significant evidence base gathered over the last two decades, we review HP gas MRI studies in patients with a range of pulmonary disorders, including COPD/emphysema, asthma, cystic fibrosis, and interstitial lung disease. We provide several examples of our experience in Sheffield of using these techniques in a diagnostic clinical setting in challenging adult and paediatric lung diseases.

Real world experience of response to pirfenidone in patients with idiopathic pulmonary fibrosis: a two centre retrospective study.

INTRODUCTION: Pirfenidone has been shown to reduce the decline in forced vital capacity (FVC) compared to placebo in patients with idiopathic pulmonary fibrosis (IPF). Previous studies have suggested that patients with a more rapid decline in FVC during the period before starting pirfenidone experience the greatest benefit from treatment. The purpose of this retrospective observational study was to investigate the response to pirfenidone in IPF patients, comparing two groups stratified by the annual rate of decline in FVC % predicted prior to treatment. METHODS: Using the rate of decline in FVC % predicted in the 12 months prior to pirfenidone, patients were stratified into slow (<5%) or rapid (≥5%) decliner groups. Comparisons in the lung function response to pirfenidone in these two groups were performed. RESULTS: Pirfenidone resulted in no statistically significant reduction in the median annual rate of decline in FVC or FVC % predicted. In the rapid decliners, pirfenidone significantly reduced the median (IQR) annual rate of decline in FVC % predicted (-8.7 (-14.2 - -7.0) %/yr vs 2.0 (-7.1 - 6.0) %/yr; n=17; p<0.01). In the slow decliners, pirfenidone did not reduce the median (IQR) annual rate of decline in FVC % predicted (-1.3 (-3.2 - 1.3) %/yr vs -5.0 (-8.3 - -0.35) %/yr; n=17; p=0.028). CONCLUSIONS: We demonstrate the greater net effect of pirfenidone in IPF patients declining rapidly. We suggest that using an annual rate of decline in FVC of <5% and ≥5% may be useful in counselling patients with regard to pirfenidone treatment. (Sarcoidosis Vasc Diffuse Lung Dis 2020; 37 (2): 218-224).

How consistently do physicians diagnose and manage drug-induced interstitial lung disease? Two surveys of European ILD specialist physicians.

INTRODUCTION: Currently there are no general guidelines for diagnosis or management of suspected drug-induced (DI) interstitial lung disease (ILD). The objective was to survey a sample of current European practice in the diagnosis and management of DI-ILD, in the context of the prescribing information approved by regulatory authorities for 28 licenced drugs with a recognised risk of DI-ILD. METHODS: Consultant physicians working in specialist ILD centres across Europe were emailed two surveys via a website link. Initially, opinion was sought regarding various diagnostic and management options based on seven clinical ILD case vignettes and five general questions regarding DI-ILD. The second survey involved 29 statements regarding the diagnosis and management of DI-ILD, derived from the results of the first survey. Consensus agreement was defined as 75% or greater. RESULTS: When making a diagnosis of DI-ILD, the favoured investigations used (other than computed tomography) included pulmonary function tests, bronchoscopy and blood tests. The preferred method used to decide when to stop treatment was a pulmonary function test. In the second survey, the majority of the statements were accepted by the 33 respondents, with only four of 29 statements not achieving consensus when the responses "agree" and "strongly agree" were combined as one answer. CONCLUSION: The two surveys provide guidance for clinicians regarding an approach to the diagnosis and management of DI-ILD in which the current evidence base is severely lacking, as demonstrated by the limited information provided by the manufacturers of the drugs associated with a high risk of DI-ILD that we reviewed.