U-net convolutional neural network applied to progressive fibrotic interstitial lung disease: Is progression at CT scan associated with a clinical outcome?

BACKGROUND: Computational advances in artificial intelligence have led to the recent emergence of U-Net convolutional neural networks (CNNs) applied to medical imaging. Our objectives were to assess the progression of fibrotic interstitial lung disease (ILD) using routine CT scans processed by a U-Net CNN developed by our research team, and to identify a progression threshold indicative of poor prognosis. METHODS: CT scans and clinical history of 32 patients with idiopathic fibrotic ILDs were retrospectively reviewed. Successive CT scans were processed by the U-Net CNN and ILD quantification was obtained. Correlation between ILD and FVC changes was assessed. ROC curve was used to define a threshold of ILD progression rate (PR) to predict poor prognostic (mortality or lung transplantation). The PR threshold was used to compare the cohort survival with Kaplan Mayer curves and log-rank test. RESULTS: The follow-up was 3.8 ± 1.5 years encompassing 105 CT scans, with 3.3 ± 1.1 CT scans per patient. A significant correlation between ILD and FVC changes was obtained (p = 0.004, ρ = -0.30 [95% CI: -0.16 to -0.45]). Sixteen patients (50%) experienced unfavorable outcome including 13 deaths and 3 lung transplantations. ROC curve analysis showed an aera under curve of 0.83 (p < 0.001), with an optimal cut-off PR value of 4%/year. Patients exhibiting a PR ≥ 4%/year during the first two years had a poorer prognosis (p = 0.001). CONCLUSIONS: Applying a U-Net CNN to routine CT scan allowed identifying patients with a rapid progression and unfavorable outcome.

Comparison between computerised lung SPECT-CT and noncontrast thoracic HRCT for quantitative analysis of post-acute COVID-19 pulmonary vascular pruning.

Computerised processing of images from routine noncontrast HRCT could be an efficient, costless and safe tool to investigate the vascular remodelling that occurs in the months after COVID-19 in a large number of patients https://bit.ly/3qAQZDW.

Comparison of optimization parametrizations for regional lung compliance estimation using personalized pulmonary poromechanical modeling.

Interstitial lung diseases, such as idiopathic pulmonary fibrosis (IPF) or post-COVID-19 pulmonary fibrosis, are progressive and severe diseases characterized by an irreversible scarring of interstitial tissues that affects lung function. Despite many efforts, these diseases remain poorly understood and poorly treated. In this paper, we propose an automated method for the estimation of personalized regional lung compliances based on a poromechanical model of the lung. The model is personalized by integrating routine clinical imaging data - namely computed tomography images taken at two breathing levels in order to reproduce the breathing kinematic-notably through an inverse problem with fully personalized boundary conditions that is solved to estimate patient-specific regional lung compliances. A new parametrization of the inverse problem is introduced in this paper, based on the combined estimation of a personalized breathing pressure in addition to material parameters, improving the robustness and consistency of estimation results. The method is applied to three IPF patients and one post-COVID-19 patient. This personalized model could help better understand the role of mechanics in pulmonary remodeling due to fibrosis; moreover, patient-specific regional lung compliances could be used as an objective and quantitative biomarker for improved diagnosis and treatment follow up for various interstitial lung diseases.

Artificial intelligence workflow quantifying muscle features on Hematoxylin-Eosin stained sections reveals dystrophic phenotype amelioration upon treatment.

Cell segmentation is a key step for a wide variety of biological investigations, especially in the context of muscle science. Currently, automated methods still struggle to perform skeletal muscle fiber quantification on Hematoxylin-Eosin (HE) stained histopathological whole slide images due to low contrast. On the other hand, the Deep Learning algorithm Cellpose offers new perspectives considering its increasing adoption for segmentation of a wide range of cells. Combining two open-source tools, Cellpose and QuPath, we developed MyoSOTHES, an automated Myofibers Segmentation wOrkflow Tuned for HE Staining. MyoSOTHES enables solving segmentation inconsistencies encountered by default Cellpose model in presence of large range size cells and provides information related to muscle Feret's diameter distribution and Centrally Nucleated Fibers, thus depicting muscle health and treatment effects. MyoSOTHES achieves high quality segmentation compared to baseline workflow with a detection F1-score increasing from 0.801 to 0.919 and a Root Mean Square Error (RMSE) on diameter improved by 31%. MyoSOTHES was validated on an animal study featuring gene transfer in [Formula: see text]-Sarcoglycanopathy, for which dose-response effect is visible and conclusions drawn are consistent with those previously published. MyoSOTHES thus paves the way for wide quantification of HE stained muscle sections and retrospective analysis of HE labeled slices used in laboratories for decades.

Estimation of Regional Pulmonary Compliance in Idiopathic Pulmonary Fibrosis Based on Personalized Lung Poromechanical Modeling.

Pulmonary function is tightly linked to the lung mechanical behavior, especially large deformation during breathing. Interstitial lung diseases, such as idiopathic pulmonary fibrosis (IPF), have an impact on the pulmonary mechanics and consequently alter lung function. However, IPF remains poorly understood, poorly diagnosed, and poorly treated. Currently, the mechanical impact of such diseases is assessed by pressure-volume curves, giving only global information. We developed a poromechanical model of the lung that can be personalized to a patient based on routine clinical data. The personalization pipeline uses clinical data, mainly computed tomography (CT) images at two time steps and involves the formulation of an inverse problem to estimate regional compliances. The estimation problem can be formulated both in terms of "effective", i.e., without considering the mixture porosity, or "rescaled," i.e., where the first-order effect of the porosity has been taken into account, compliances. Regional compliances are estimated for one control subject and three IPF patients, allowing to quantify the IPF-induced tissue stiffening. This personalized model could be used in the clinic as an objective and quantitative tool for IPF diagnosis.

Prone-Positioning for Severe Acute Respiratory Distress Syndrome Requiring Extracorporeal Membrane Oxygenation.

OBJECTIVES: To determine the characteristics and outcomes of patients prone-positioned during extracorporeal membrane oxygenation for severe acute respiratory distress syndrome and lung CT pattern associated with improved respiratory system static compliance after that intervention. DESIGN: Retrospective, single-center study over 8 years. SETTINGS: Twenty-six bed ICU in a tertiary center. MEASUREMENTS AND MAIN RESULTS: A propensity score-matched analysis compared patients with prone-positioning during extracorporeal membrane oxygenation and those without. An increase of the static compliance greater than or equal to 3 mL/cm H2O after 16 hours of prone-positioning defined prone-positioning responders. The primary outcome was the time to successful extracorporeal membrane oxygenation weaning within 90 days of postextracorporeal membrane oxygenation start, with death as a competing risk. Among 298 venovenous extracorporeal membrane oxygenation-treated adults with severe acute respiratory distress syndrome, 64 were prone-positioning extracorporeal membrane oxygenation. Although both propensity score-matched groups had similar extracorporeal membrane oxygenation durations, prone-positioning extracorporeal membrane oxygenation patients' 90-day probability of being weaned-off extracorporeal membrane oxygenation and alive was higher (0.75 vs 0.54, p = 0.03; subdistribution hazard ratio [95% CI], 1.54 [1.05-2.58]) and 90-day mortality was lower (20% vs 42%, p < 0.01) than that for no prone-positioning extracorporeal membrane oxygenation patients. Extracorporeal membrane oxygenation-related complications were comparable for the two groups. Patients without improved static compliance had higher percentages of nonaerated or poorly aerated ventral and medial-ventral lung regions (p = 0.047). CONCLUSIONS: Prone-positioning during venovenous extracorporeal membrane oxygenation was safe and effective and was associated with a higher probability of surviving and being weaned-off extracorporeal membrane oxygenation at 90 days. Patients with greater normally aerated lung tissue in the ventral and medial-ventral regions on quantitative lung CT-scan performed before prone-positioning are more likely to improve their static compliance after that procedure during extracorporeal membrane oxygenation.

Quantitative computed tomography imaging of airway remodeling in severe asthma.

Asthma is a heterogeneous condition and approximately 5-10% of asthmatic subjects have severe disease associated with structure changes of the airways (airway remodeling) that may develop over time or shortly after onset of disease. Quantitative computed tomography (QCT) imaging of the tracheobronchial tree and lung parenchyma has improved during the last 10 years, and has enabled investigators to study the large airway architecture in detail and assess indirectly the small airway structure. In severe asthmatics, morphologic changes in large airways, quantitatively assessed using 2D-3D airway registration and recent algorithms, are characterized by airway wall thickening, luminal narrowing and bronchial stenoses. Extent of expiratory gas trapping, quantitatively assessed using lung densitometry, may be used to assess indirectly small airway remodeling. Investigators have used these quantitative imaging techniques in order to attempt severity grading of asthma, and to identify clusters of asthmatic patients that differ in morphologic and functional characteristics. Although standardization of image analysis procedures needs to be improved, the identification of remodeling pattern in various phenotypes of severe asthma and the ability to relate airway structures to important clinical outcomes should help target treatment more effectively.

Development and Analysis of Patient-Based Complete Conducting Airways Models.

The analysis of high-resolution computed tomography (CT) images of the lung is dependent on inter-subject differences in airway geometry. The application of computational models in understanding the significance of these differences has previously been shown to be a useful tool in biomedical research. Studies using image-based geometries alone are limited to the analysis of the central airways, down to generation 6-10, as other airways are not visible on high-resolution CT. However, airways distal to this, often termed the small airways, are known to play a crucial role in common airway diseases such as asthma and chronic obstructive pulmonary disease (COPD). Other studies have incorporated an algorithmic approach to extrapolate CT segmented airways in order to obtain a complete conducting airway tree down to the level of the acinus. These models have typically been used for mechanistic studies, but also have the potential to be used in a patient-specific setting. In the current study, an image analysis and modelling pipeline was developed and applied to a number of healthy (n = 11) and asthmatic (n = 24) CT patient scans to produce complete patient-based airway models to the acinar level (mean terminal generation 15.8 ± 0.47). The resulting models are analysed in terms of morphometric properties and seen to be consistent with previous work. A number of global clinical lung function measures are compared to resistance predictions in the models to assess their suitability for use in a patient-specific setting. We show a significant difference (p < 0.01) in airways resistance at all tested flow rates in complete airway trees built using CT data from severe asthmatics (GINA 3-5) versus healthy subjects. Further, model predictions of airways resistance at all flow rates are shown to correlate with patient forced expiratory volume in one second (FEV1) (Spearman ρ = -0.65, p < 0.001) and, at low flow rates (0.00017 L/s), FEV1 over forced vital capacity (FEV1/FVC) (ρ = -0.58, p < 0.001). We conclude that the pipeline and anatomical models can be used directly in mechanistic modelling studies and can form the basis for future patient-based modelling studies.

Computed tomography assessment of airways throughout bronchial tree demonstrates airway narrowing in severe asthma.

RATIONALE AND OBJECTIVES: To analyze airway dimensions throughout the bronchial tree in severe asthmatic patients using multidetector row computed tomography (MDCT) focusing on airway narrowing. MATERIALS AND METHODS: Thirty-two patients with severe asthma underwent automated (BronCare software) analysis of their right lung bronchi, with counts of airways >3 mm long arising from the main bronchi (airway count) and bronchial dimension quantification at segmental and subsegmental levels (lumen area [LA], wall area [WA], and WA%). Focal bronchial stenosis was defined as >50% narrowing of maximal LA on contiguous cross-sectional slices. Severe asthmatics were compared to 13 nonsevere asthmatic patients and nonasthmatic (pooled) subjects (Wilcoxon rank tests, then stepwise logistic regression). Finally, cluster analysis of severe asthmatic patients and stepwise logistic regression identified specific imaging subgroups. RESULTS: The most significant differences between severe asthmatic patients and the pooled subjects were bronchial stenosis (subsegmental and all bronchi: P < .002) and WA% (P < .0003). Stepwise logistic regression retained WA% as the only explanatory covariable (P = .002). Two identified clusters of severe asthmatic patients differed for parameters characterizing airway narrowing (airway count: P = .0002; focal bronchial stenosis: P = .009). Airway count was as discriminant as forced expiratory volume in 1 second/forced vital capacity (P = .01) to identify patients in each cluster, with both variables being correlated (r = 0.59, P = .005). CONCLUSIONS: Severe asthma-associated morphologic changes were characterized by focal bronchial stenoses and diffuse airway narrowing; the latter was associated with airflow obstruction. WA%, dependent on airway caliber, is the best parameter to identify severe asthmatic patients from pooled subjects.

Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: the VESSEL12 study.

The VESSEL12 (VESsel SEgmentation in the Lung) challenge objectively compares the performance of different algorithms to identify vessels in thoracic computed tomography (CT) scans. Vessel segmentation is fundamental in computer aided processing of data generated by 3D imaging modalities. As manual vessel segmentation is prohibitively time consuming, any real world application requires some form of automation. Several approaches exist for automated vessel segmentation, but judging their relative merits is difficult due to a lack of standardized evaluation. We present an annotated reference dataset containing 20 CT scans and propose nine categories to perform a comprehensive evaluation of vessel segmentation algorithms from both academia and industry. Twenty algorithms participated in the VESSEL12 challenge, held at International Symposium on Biomedical Imaging (ISBI) 2012. All results have been published at the VESSEL12 website http://vessel12.grand-challenge.org. The challenge remains ongoing and open to new participants. Our three contributions are: (1) an annotated reference dataset available online for evaluation of new algorithms; (2) a quantitative scoring system for objective comparison of algorithms; and (3) performance analysis of the strengths and weaknesses of the various vessel segmentation methods in the presence of various lung diseases.

Volumetric quantification of airway wall in CT via collision-free active surface model: application to asthma assessment.

Emerging idea in asthma phenotyping, incorporating local morphometric information on the airway wall thickness would be able to better account for the process of airway remodeling as indicator of pathology or therapeutic impact. It is thus important that such information be provided uniformly along the airway tree, not on a sparse (cross-section) sampling basis. The volumetric segmentation of the airway wall from CT data is the issue addressed in this paper by exploiting a patient-specific surface active model. An original aspect taken into account in the proposed deformable model is the management of auto-collisions for this complex morphology. The analysis of several solutions ended up with the design of a motion vector field specific to the patient geometry to guide the deformation. The segmentation result, presented as two embedded inner/outer surfaces of the wall, allows the quantification of the tissue thickness based on a locally-defined measure sensitive to even small surface irregularities. The method is validated with respect to several ground truth simulations of pulmonary CT data with different airway geometries and acquisition protocols showing accuracy within the CT resolution range. Results from an ongoing clinical study on moderate and severe asthma are presented and discussed.

Relationship between the airway wall area and asthma control score in moderate persistent asthma.

OBJECTIVES: To assess the association between airway wall area and clinical asthma control, assessed by the Asthma Control Test (ACT). METHODS: This cross-sectional study evaluated 96 adults for asthma control ["at least well controlled" (ACT ≥ 20; n = 52) or "not well controlled" (ACT < 20; n = 44) and airway dimensions: luminal area (LA), wall area (WA) and WA%], obtained using automated dedicated software measurements from volumetric CT images. Results were analysed for segmental bronchi, subsegmental bronchi in the right upper lobe and basilar segments, both uncorrected and corrected for body surface area (BSA). RESULTS: For all bronchi corrected for BSA, there was no correlation between airway wall area and ACT score. There was a weak but statistically significant correlation between uncorrected WA and ACT score (r = -0.203; P = 0.047); WA values were numerically higher in the "not well-controlled" versus the "at least well-controlled asthma" subgroups. For sub-segmental bronchi, there was a correlation between the ACT score and both WA/BSA (r = -0.204; P = 0.047) and WA (r = -0.249; P = 0.014), and for upper lobe bronchi, between the ACT score and WA (r = -0.207; P = 0.044). CONCLUSION: We demonstrated a correlation between subsegmental bronchial airway measurements and clinical control of asthma; this is probably a reflection of airway remodelling and structural changes in chronic poorly controlled asthma. KEY POINTS: • Volumetric computed tomography offers new insights into bronchial morphology. • The relationship between current asthma control and airway wall abnormalities is assessed. • Some relationships between airway wall area and clinical control were demonstrated. • We observed less shape variation of bronchi in "not well-controlled" asthma patients.

Extraction of airways from CT (EXACT'09).

This paper describes a framework for establishing a reference airway tree segmentation, which was used to quantitatively evaluate fifteen different airway tree extraction algorithms in a standardized manner. Because of the sheer difficulty involved in manually constructing a complete reference standard from scratch, we propose to construct the reference using results from all algorithms that are to be evaluated. We start by subdividing each segmented airway tree into its individual branch segments. Each branch segment is then visually scored by trained observers to determine whether or not it is a correctly segmented part of the airway tree. Finally, the reference airway trees are constructed by taking the union of all correctly extracted branch segments. Fifteen airway tree extraction algorithms from different research groups are evaluated on a diverse set of twenty chest computed tomography (CT) scans of subjects ranging from healthy volunteers to patients with severe pathologies, scanned at different sites, with different CT scanner brands, models, and scanning protocols. Three performance measures covering different aspects of segmentation quality were computed for all participating algorithms. Results from the evaluation showed that no single algorithm could extract more than an average of 74% of the total length of all branches in the reference standard, indicating substantial differences between the algorithms. A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.

Multidetector row computed tomography to assess changes in airways linked to asthma control.

BACKGROUND: In asthma, multidetector row computed tomography (MDCT) detects abnormalities that are related to disease severity, including increased bronchial wall thickness. However, whether these abnormalities could be related to asthma control has not been investigated yet. OBJECTIVE: Our goal was to determine which changes in airways could be linked to disease control. METHODS: Twelve patients with poor asthma control were included and received a salmeterol/fluticasone propionate combination daily for 12 weeks. Patients underwent clinical, functional, and MDCT examinations before and after the treatment period. MDCT examinations were performed using a low-dose protocol at a controlled lung volume (65% TLC). Bronchial lumen (LA) and wall areas (WA) were evaluated at a segmental and subsegmental level using BronCare software. Lung density was measured at the base of the lung. Baseline and end-of-treatment data were compared using the Wilcoxon signed-rank test. RESULTS: After the 12-week treatment period, asthma control was achieved. Airflow obstruction and air trapping decreased as assessed by the changes in FEV(1) (p < 0.01) and expiratory reserve volume (p < 0.01). Conversely, LA and WA did not vary significantly. However, a median decrease in LA of >10% was observed in half of the patients with a wide intra- and intersubject response heterogeneity. This was concomitant with a decrease in lung density (p < 0.02 in the anteroinferior areas). CONCLUSIONS: MDCT is insensitive for demonstrating any decrease in bronchial wall thickness. This is mainly due to changes in bronchial caliber which may be linked to modifications of the elastic properties of the bronchopulmonary system under treatment.

Variability of bronchial measurements obtained by sequential CT using two computer-based methods.

This study aimed to evaluate the variability of lumen (LA) and wall area (WA) measurements obtained on two successive MDCT acquisitions using energy-driven contour estimation (EDCE) and full width at half maximum (FWHM) approaches. Both methods were applied to a database of segmental and subsegmental bronchi with LA > 4 mm(2) containing 42 bronchial segments of 10 successive slices that best matched on each acquisition. For both methods, the 95% confidence interval between repeated MDCT was between -1.59 and 1.5 mm(2) for LA, and -3.31 and 2.96 mm(2) for WA. The values of the coefficient of measurement variation (CV(10), i.e., percentage ratio of the standard deviation obtained from the 10 successive slices to their mean value) were strongly correlated between repeated MDCT data acquisitions (r > 0.72; p < 0.0001). Compared with FWHM, LA values obtained using EDCE were higher for LA < 15 mm(2), whereas WA values were lower for bronchi with WA < 13 mm(2); no systematic EDCE underestimation or overestimation was observed for thicker-walled bronchi. In conclusion, variability between CT examinations and assessment techniques may impair measurements. Therefore, new parameters such as CV(10) need to be investigated to study bronchial remodeling. Finally, EDCE and FWHM are not interchangeable in longitudinal studies.

CT imaging of chronic obstructive pulmonary disease: role in phenotyping and interventions.

BACKGROUND: Using CT to define phenotypic abnormalities in patients diagnosed as having chronic obstructive pulmonary disease (COPD) may serve to optimize treatment, guide the prognosis and assess response to potential therapeutic interventions. OBJECTIVE/METHOD: Although the different morphologic abnormalities seen on CT scans of COPD patients often overlap, two separate groups of patients can be identified, those with emphysema predominant disease and those with airway predominant disease due to chronic inflammation with resulting in airway remodelling and narrowing. The former category can be subdivided further based on the type of emphysema present and characterized further by anatomic distribution and severity using visual assessment and volumetric quantitative CT techniques. Patients in the airway predominant category can also be characterized by CT as showing bronchial wall thickening, small airway inflammation, mosaic perfusion and air trapping expressing small airway narrowing, and expiratory bronchial collapse due to cartilage deficiency. Recent advances in automated airway segmentation and quantitative analysis have made measurements of airway dimensions feasible. CONCLUSION: In longitudinal studies, standardization of procedures and quality control are needed, particularly if quantitative CT outcomes are used as end point in clinical trials and ultimately in the clinical management of individual patients.

Investigation of airways using MDCT for visual and quantitative assessment in COPD patients.

Multidetector computed tomography (MDCT) acquisition during a single breath hold using thin collimation provides high resolution volumetric data set permitting multiplanar and three dimensional reconstruction of the proximal airways. In chronic obstructive pulmonary disease (COPD) patients, this technique provides an accurate assessment of bronchial wall thickening, tracheobronchial deformation, outpouchings reflecting dilatation of the submucous glands, tracheobronchomalacia, and expiratory air trapping. New software developed to segment adequately the lumen and walls of the airways on MDCT scans allows quantitative assessment of the airway dimensions which has shown to be reliable in clinical practice. This technique can become important in longitudinal studies of the pathogenesis of COPD, and in the assessment of therapeutic interventions.

Diffuse parenchymal lung diseases: 3D automated detection in MDCT.

Characterization and quantification of diffuse parenchymal lung disease (DPLD) severity using MDCT, mainly in interstitial lung diseases and emphysema, is an important issue in clinical research for the evaluation of new therapies. This paper develops a 3D automated approach for detection and diagnosis of DPLDs (emphysema, fibrosis, honeycombing, ground glass). The proposed methodology combines multi-resolution image decomposition based on 3D morphological filtering, and graph-based classification for a full characterization of the parenchymal tissue. The very promising results obtained on a small patient database are good premises for a near implementation and validation of the proposed approach in clinical routine.

Assessment of airway remodeling in asthma: volumetric versus surface quantification approaches.

This paper develops a volumetric quantification approach of the airway wall in multi-detector computed tomography (MDCT), exploiting a 3D segmentation methodology based on patient-specific deformable mesh model. A comparative study carried out with respect to a reference 2D/3D surface quantification technique underlines the clinical interest of the proposed approach in assessing airway remodeling in asthmatics and in evaluating the efficiency of therapeutic protocols.