2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;48:1489-1497.
To assess airway and lung parenchymal damage noninvasively in cystic fibrosis (CF), chest MRI has been historically out of the scope of routine clinical imaging because of technical difficulties such as low proton density and respiratory and cardiac motion. However, technological breakthroughs have emerged that dramatically improve lung MRI quality (including signal-to-noise ratio, resolution, speed, and contrast). At the same time, novel treatments have changed the landscape of CF clinical care. In this contemporary context, there is now consensus that lung MRI can be used clinically to assess CF in a radiation-free manner and to enable quantification of lung disease severity. MRI can now achieve three-dimensional, high-resolution morphologic imaging, and beyond this morphologic information, MRI may offer the ability to sensitively differentiate active inflammation vs scarring tissue. MRI could also characterize various forms of inflammation for early guidance of treatment. Moreover, functional information from MRI can be used to assess regional, small-airway disease with sensitivity to detect small changes even in patients with mild CF. Finally, automated quantification methods have emerged to support conventional visual analyses for more objective and reproducible assessment of disease severity. This article aims to review the most recent developments of lung MRI, with a focus on practical application and clinical value in CF, and the perspectives on how these modern techniques may converge and impact patient care soon.
Background Little is known about in vivo alterations at bronchial and vascular levels in severe pulmonary hypertension (PH) of different etiologies. We aimed to compare quantitative computed tomography (CT) data from the following three groups of severe precapillary PH patients: COPD, idiopathic pulmonary arterial hypertension (iPAH), and chronic thromboembolic PH (CTEPH). Patients and methods This study was approved by the institutional review board. Severe PH patients (mean pulmonary arterial pressure [mPAP] ≥35 mmHg) with COPD, iPAH, or CTEPH (n=24, 16, or 16, respectively) were included retrospectively between January 2008 and January 2017. Univariate analysis of mPAP was performed in each severe PH group. Bronchial wall thickness (WT) and percentage of cross sectional area of pulmonary vessels less than 5 mm 2 normalized by lung area (%CSA <5 ) were measured and compared using CT, and then combined to arterial partial pressure of oxygen (PaO 2 ) to generate a “paw score” compared within the three groups using Kruskal–Wallis and its sensitivity using Fisher’s exact test. Results WT was higher and %CSA <5 was lower in the COPD group compared to iPAH and CTEPH groups. Mosaic pattern was higher in CTEPH group than in others. In severe PH patients secondary to COPD, mPAP was positively correlated to %CSA <5 . By contrast, in severe iPAH, this correlation was negative, or not correlated in severe CTEPH groups. In the COPD group, “paw score” showed higher sensitivity than in the other two groups. Conclusion Unlike in severe iPAH and CTEPH, severe PH with COPD can be predicted by “paw score” reflecting bronchial and vascular morphological differential alterations.
RationaleChest computed tomography (CT) remains the imaging standard for demonstrating cystic fibrosis airway structural disease in vivo. However, visual scorings as an outcome measure are time-consuming, require training, and lack high reproducibility.ObjectiveTo validate a fully automated artificial intelligence-driven scoring of cystic fibrosis lung disease severity.MethodsData were retrospectively collected in three cystic fibrosis reference centers, between 2008 and 2020, in 184 patients 4 to 54-years-old. An algorithm using three two-dimensional convolutional neural networks was trained with 78 patients’ CTs (23 530 CT slices) for the semantic labeling of bronchiectasis, peribronchial thickening, bronchial mucus, bronchiolar mucus, and collapse/consolidation. 36 patients’ CTs (11 435 CT slices) were used for testing versus ground-truth labels. The method's clinical validity was assessed in an independent group of 70 patients with or without lumacaftor/ivacaftor treatment (n=10 and 60, respectively) with repeat examinations. Similarity and reproducibility were assessed using Dice coefficient, correlations using Spearman test, and paired comparisons using Wilcoxon rank test.Measurement and main resultsThe overall pixelwise similarity of artificial intelligence-driven versus ground-truth labels was good (Dice coefficient=0.71). All artificial intelligence-driven volumetric quantifications had moderate to very good correlations to a visual imaging scoring (p<0.001) and fair to good correlations to FEV1% at pulmonary function test (p<0.001). Significant decreases in peribronchial thickening (p=0.005), bronchial mucus (p=0.005), bronchiolar mucus (p=0.007) volumes were measured in patients with lumacaftor/ivacaftor. Conversely, bronchiectasis (p=0.002) and peribronchial thickening (p=0.008) volumes increased in patients without lumacaftor/ivacaftor. The reproducibility was almost perfect (Dice>0.99).ConclusionArtificial intelligence allows a fully automated volumetric quantification of cystic fibrosis-related modifications over an entire lung. The novel scoring system could provide a robust disease outcome in the era of effective CFTR modulator therapy.
hronic obstructive pulmonary disease (COPD) is characterized by persistent respiratory symptoms and airflow limitation (1) and is the third leading cause of mortality worldwide (2). In the past decades, chest CT has helped to provide valuable insights into COPD. Indeed, CT enables noninvasive, in vivo, and three-dimensional (3D) quantification of the key components of the disease, such as bronchial remodeling (3) and emphysema (4). Evidence of COPD subtypes (1), relationship with genetic variants (5), longitudinal follow-up (6), and mortality (7) have been investigated at CT. More recently, in an era where the advent of artificial intelligence may allow intricate combinations of both morphologic and functional information, lung MRI has emerged as a radiation-free modality for the longitudinal follow-up of lung diseases and (8) to help better identify participant phenotype or predict disease outcomes (9). Studies related to ventilation (10), perfusion (11), dynamic biomarkers (12), and coupling between the cardiovascular and pulmonary systems (13,14) at MRI have been performed. However, there are few studies that evaluated the regional extent of emphysema by using MRI (15-18). Recent advances with ultrashort echo times (UTEs) have been shown to enable submillimeter resolution with good contrast in three dimensions (19,20).
Pulmonary hypertension (PH) is a common complication of chronic obstructive pulmonary disease (COPD) and is associated with increased morbidity and mortality. Reference standard method to diagnose PH is right heart catheterization. Several non-invasive imaging techniques have been employed in the detection of PH. Among them, computed tomography (CT) is the most commonly used for phenotyping and detecting complications of COPD. Several CT findings have also been described in patients with severe PH. Nevertheless, CT analysis is currently based on visual findings which can lead to reproducibility failure. Therefore, there is a need for quantification in order to assess objective criteria. In this review, progresses in automated analyses of CT parameters and their values in predicting PH and COPD outcomes are presented.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.