Capturing COPD heterogeneity: anomaly detection and parametric response mapping comparison for phenotyping on chest computed tomography.

Background: Chronic obstructive pulmonary disease (COPD) poses a substantial global health burden, demanding advanced diagnostic tools for early detection and accurate phenotyping. In this line, this study seeks to enhance COPD characterization on chest computed tomography (CT) by comparing the spatial and quantitative relationships between traditional parametric response mapping (PRM) and a novel self-supervised anomaly detection approach, and to unveil potential additional insights into the dynamic transitional stages of COPD. Methods: Non-contrast inspiratory and expiratory CT of 1,310 never-smoker and GOLD 0 individuals and COPD patients (GOLD 1-4) from the COPDGene dataset were retrospectively evaluated. A novel self-supervised anomaly detection approach was applied to quantify lung abnormalities associated with COPD, as regional deviations. These regional anomaly scores were qualitatively and quantitatively compared, per GOLD class, to PRM volumes (emphysema: PRMEmph, functional small-airway disease: PRMfSAD) and to a Principal Component Analysis (PCA) and Clustering, applied on the self-supervised latent space. Its relationships to pulmonary function tests (PFTs) were also evaluated. Results: Initial t-Distributed Stochastic Neighbor Embedding (t-SNE) visualization of the self-supervised latent space highlighted distinct spatial patterns, revealing clear separations between regions with and without emphysema and air trapping. Four stable clusters were identified among this latent space by the PCA and Cluster Analysis. As the GOLD stage increased, PRMEmph, PRMfSAD, anomaly score, and Cluster 3 volumes exhibited escalating trends, contrasting with a decline in Cluster 2. The patient-wise anomaly scores significantly differed across GOLD stages (p < 0.01), except for never-smokers and GOLD 0 patients. In contrast, PRMEmph, PRMfSAD, and cluster classes showed fewer significant differences. Pearson correlation coefficients revealed moderate anomaly score correlations to PFTs (0.41-0.68), except for the functional residual capacity and smoking duration. The anomaly score was correlated with PRMEmph (r = 0.66, p < 0.01) and PRMfSAD (r = 0.61, p < 0.01). Anomaly scores significantly improved fitting of PRM-adjusted multivariate models for predicting clinical parameters (p < 0.001). Bland-Altman plots revealed that volume agreement between PRM-derived volumes and clusters was not constant across the range of measurements. Conclusion: Our study highlights the synergistic utility of the anomaly detection approach and traditional PRM in capturing the nuanced heterogeneity of COPD. The observed disparities in spatial patterns, cluster dynamics, and correlations with PFTs underscore the distinct - yet complementary - strengths of these methods. Integrating anomaly detection and PRM offers a promising avenue for understanding of COPD pathophysiology, potentially informing more tailored diagnostic and intervention approaches to improve patient outcomes.

Phenotyping of COPD with MRI in comparison to same-day CT in a multi-centre trial.

OBJECTIVES: A prospective, multi-centre study to evaluate concordance of morphologic lung MRI and CT in chronic obstructive pulmonary disease (COPD) phenotyping for airway disease and emphysema. METHODS: A total of 601 participants with COPD from 15 sites underwent same-day morpho-functional chest MRI and paired inspiratory-expiratory CT. Two readers systematically scored bronchial wall thickening, bronchiectasis, centrilobular nodules, air trapping and lung parenchyma defects in each lung lobe and determined COPD phenotype. A third reader acted as adjudicator to establish consensus. Inter-modality and inter-reader agreement were assessed using Cohen's kappa (im-κ and ir-κ). RESULTS: The mean combined MRI score for bronchiectasis/bronchial wall thickening was 4.5/12 (CT scores, 2.2/12 for bronchiectasis and 6/12 for bronchial wall thickening; im-κ, 0.04-0.3). Expiratory right/left bronchial collapse was observed in 51 and 47/583 on MRI (62 and 57/599 on CT; im-κ, 0.49-0.52). Markers of small airways disease on MRI were 0.15/12 for centrilobular nodules (CT, 0.34/12), 0.94/12 for air trapping (CT, 0.9/12) and 7.6/12 for perfusion deficits (CT, 0.37/12 for mosaic attenuation; im-κ, 0.1-0.41). The mean lung defect score on MRI was 1.3/12 (CT emphysema score, 5.8/24; im-κ, 0.18-0.26). Airway-/emphysema/mixed COPD phenotypes were assigned in 370, 218 and 10 of 583 cases on MRI (347, 218 and 34 of 599 cases on CT; im-κ, 0.63). For all examined features, inter-reader agreement on MRI was lower than on CT. CONCLUSION: Concordance of MRI and CT for phenotyping of COPD in a multi-centre setting was substantial with variable inter-modality and inter-reader concordance for single diagnostic key features. CLINICAL RELEVANCE STATEMENT: MRI of lung morphology may well serve as a radiation-free imaging modality for COPD in scientific and clinical settings, given that its potential and limitations as shown here are carefully considered. KEY POINTS: • In a multi-centre setting, MRI and CT showed substantial concordance for phenotyping of COPD (airway-/emphysema-/mixed-type). • Individual features of COPD demonstrated variable inter-modality concordance with features of pulmonary hypertension showing the highest and bronchiectasis showing the lowest concordance. • For all single features of COPD, inter-reader agreement was lower on MRI than on CT.

Prediction of disease severity in COPD: a deep learning approach for anomaly-based quantitative assessment of chest CT.

OBJECTIVES: To quantify regional manifestations related to COPD as anomalies from a modeled distribution of normal-appearing lung on chest CT using a deep learning (DL) approach, and to assess its potential to predict disease severity. MATERIALS AND METHODS: Paired inspiratory/expiratory CT and clinical data from COPDGene and COSYCONET cohort studies were included. COPDGene data served as training/validation/test data sets (N = 3144/786/1310) and COSYCONET as external test set (N = 446). To differentiate low-risk (healthy/minimal disease, [GOLD 0]) from COPD patients (GOLD 1-4), the self-supervised DL model learned semantic information from 50 × 50 × 50 voxel samples from segmented intact lungs. An anomaly detection approach was trained to quantify lung abnormalities related to COPD, as regional deviations. Four supervised DL models were run for comparison. The clinical and radiological predictive power of the proposed anomaly score was assessed using linear mixed effects models (LMM). RESULTS: The proposed approach achieved an area under the curve of 84.3 ± 0.3 (p < 0.001) for COPDGene and 76.3 ± 0.6 (p < 0.001) for COSYCONET, outperforming supervised models even when including only inspiratory CT. Anomaly scores significantly improved fitting of LMM for predicting lung function, health status, and quantitative CT features (emphysema/air trapping; p < 0.001). Higher anomaly scores were significantly associated with exacerbations for both cohorts (p < 0.001) and greater dyspnea scores for COPDGene (p < 0.001). CONCLUSION: Quantifying heterogeneous COPD manifestations as anomaly offers advantages over supervised methods and was found to be predictive for lung function impairment and morphology deterioration. CLINICAL RELEVANCE STATEMENT: Using deep learning, lung manifestations of COPD can be identified as deviations from normal-appearing chest CT and attributed an anomaly score which is consistent with decreased pulmonary function, emphysema, and air trapping. KEY POINTS: • A self-supervised DL anomaly detection method discriminated low-risk individuals and COPD subjects, outperforming classic DL methods on two datasets (COPDGene AUC = 84.3%, COSYCONET AUC = 76.3%). • Our contrastive task exhibits robust performance even without the inclusion of expiratory images, while voxel-based methods demonstrate significant performance enhancement when incorporating expiratory images, in the COPDGene dataset. • Anomaly scores improved the fitting of linear mixed effects models in predicting clinical parameters and imaging alterations (p < 0.001) and were directly associated with clinical outcomes (p < 0.001).

Qualitative and quantitative evaluation of computed tomography changes in adults with cystic fibrosis treated with elexacaftor-tezacaftor-ivacaftor: a retrospective observational study.

Introduction: The availability of highly effective triple cystic fibrosis transmembrane conductance regulator (CFTR) modulator combination therapy with elexacaftor-tezacaftor-ivacaftor (ETI) has improved pulmonary outcomes and quality of life of people with cystic fibrosis (pwCF). The aim of this study was to assess computed tomography (CT) changes under ETI visually with the Brody score and quantitatively with dedicated software, and to correlate CT measures with parameters of clinical response. Methods: Twenty two adult pwCF with two consecutive CT scans before and after ETI treatment initiation were retrospectively included. CT was assessed visually employing the Brody score and quantitatively by YACTA, a well-evaluated scientific software computing airway dimensions and lung parenchyma with wall percentage (WP), wall thickness (WT), lumen area (LA), bronchiectasis index (BI), lung volume and mean lung density (MLD) as parameters. Changes in CT metrics were evaluated and the visual and quantitative parameters were correlated with each other and with clinical changes in sweat chloride concentration, spirometry [percent predicted of forced expiratory volume in one second (ppFEV1)] and body mass index (BMI). Results: The mean (SD) Brody score improved with ETI [55 (12) vs. 38 (15); p < 0.001], incl. sub-scores for mucus plugging, peribronchial thickening, and parenchymal changes (all p < 0.001), but not for bronchiectasis (p = 0.281). Quantitatve WP (p < 0.001) and WT (p = 0.004) were reduced, conversely LA increased (p = 0.003), and BI improved (p = 0.012). Lung volume increased (p < 0.001), and MLD decreased (p < 0.001) through a reduction of ground glass opacity areas (p < 0.001). Changes of the Brody score correlated with those of quantitative parameters, exemplarily WT with the sub-score for mucus plugging (r = 0.730, p < 0.001) and peribronchial thickening (r = 0.552, p = 0.008). Changes of CT parameters correlated with those of clinical response parameters, in particular ppFEV1 with the Brody score (r = -0.606, p = 0.003) and with WT (r = -0.538, p = 0.010). Discussion: Morphological treatment response to ETI can be assessed using the Brody score as well as quantitative CT parameters. Changes in CT correlated with clinical improvements. The quantitative analysis with YACTA proved to be an objective, reproducible and simple method for monitoring lung disease, particularly with regard to future interventional clinical trials.

GOLD stage-specific phenotyping of emphysema and airway disease using quantitative computed tomography.

Background: In chronic obstructive pulmonary disease (COPD) abnormal lung function is related to emphysema and airway obstruction, but their relative contribution in each GOLD-stage is not fully understood. In this study, we used quantitative computed tomography (QCT) parameters for phenotyping of emphysema and airway abnormalities, and to investigate the relative contribution of QCT emphysema and airway parameters to airflow limitation specifically in each GOLD stage. Methods: Non-contrast computed tomography (CT) of 492 patients with COPD former GOLD 0 COPD and COPD stages GOLD 1-4 were evaluated using fully automated software for quantitative CT. Total lung volume (TLV), emphysema index (EI), mean lung density (MLD), and airway wall thickness (WT), total diameter (TD), lumen area (LA), and wall percentage (WP) were calculated for the entire lung, as well as for all lung lobes separately. Results from the 3rd-8th airway generation were aggregated (WT3-8, TD3-8, LA3-8, WP3-8). All subjects underwent whole-body plethysmography (FEV1%pred, VC, RV, TLC). Results: EI was higher with increasing GOLD stages with 1.0 ± 1.8% in GOLD 0, 4.5 ± 9.9% in GOLD 1, 19.4 ± 15.8% in GOLD 2, 32.7 ± 13.4% in GOLD 3 and 41.4 ± 10.0% in GOLD 4 subjects (p < 0.001). WP3-8 showed no essential differences between GOLD 0 and GOLD 1, tended to be higher in GOLD 2 with 52.4 ± 7.2%, and was lower in GOLD 4 with 50.6 ± 5.9% (p = 0.010 - p = 0.960). In the upper lobes WP3-8 showed no significant differences between the GOLD stages (p = 0.824), while in the lower lobes the lowest WP3-8 was found in GOLD 0/1 with 49.9 ± 6.5%, while higher values were detected in GOLD 2 with 51.9 ± 6.4% and in GOLD 3/4 with 51.0 ± 6.0% (p < 0.05). In a multilinear regression analysis, the dependent variable FEV1%pred can be predicted by a combination of both the independent variables EI (p < 0.001) and WP3-8 (p < 0.001). Conclusion: QCT parameters showed a significant increase of emphysema from GOLD 0-4 COPD. Airway changes showed a different spatial pattern with higher values of relative wall thickness in the lower lobes until GOLD 2 and subsequent lower values in GOLD3/4, whereas there were no significant differences in the upper lobes. Both, EI and WP5-8 are independently correlated with lung function decline.

MRI Compared with Low-Dose CT for Incidental Lung Nodule Detection in COPD: A Multicenter Trial.

Purpose: To investigate morphofunctional chest MRI for the detection and management of incidental pulmonary nodules in participants with chronic obstructive pulmonary disease (COPD). Materials and Methods: In this prospective study, 567 participants (mean age, 66 years ± 9 [SD]; 340 men) underwent same-day contrast-enhanced MRI and nonenhanced low-dose CT (LDCT) in a nationwide multicenter trial (clinicaltrials.gov: NCT01245933). Nodule dimensions, morphologic features, and Lung Imaging Reporting and Data System (Lung-RADS) category were assessed at MRI by two blinded radiologists, and consensual LDCT results served as the reference standard. Comparisons were performed using the Student t test, and agreements were assessed using the Cohen weighted κ. Results: A total of 525 nodules larger than 3 mm in diameter were detected at LDCT in 178 participants, with a mean diameter of 7.2 mm ± 6.1 (range, 3.1-63.1 mm). Nodules were not detected in the remaining 389 participants. Sensitivity and positive predictive values with MRI for readers 1 and 2, respectively, were 63.0% and 84.8% and 60.2% and 83.9% for solid nodules (n = 495), 17.6% and 75.0% and 17.6% and 60.0% for part-solid nodules (n = 17), and 7.7% and 100% and 7.7% and 50.0% for ground-glass nodules (n = 13). For nodules 6 mm or greater in diameter, sensitivity and positive predictive values were 73.3% and 92.2% for reader 1 and 71.4% and 93.2% for reader 2, respectively. Readers underestimated the long-axis diameter at MRI by 0.5 mm ± 1.7 (reader 1) and 0.5 mm ± 1.5 (reader 2) compared with LDCT (P < .001). For Lung-RADS categorization per nodule using MRI, there was substantial to perfect interreader agreement (κ = 0.75-1.00) and intermethod agreement compared with LDCT (κ = 0.70-1.00 and 0.69-1.00). Conclusion: In a multicenter setting, morphofunctional MRI showed moderate sensitivity for detection of incidental pulmonary nodules in participants with COPD but high agreement with LDCT for Lung-RADS classification of nodules.Clinical trial registration no. NCT01245933 and NCT02629432Keywords: MRI, CT, Thorax, Lung, Chronic Obstructive Pulmonary Disease, Screening© RSNA, 2023 Supplemental material is available for this article.

Quantification of airway wall contrast enhancement on virtual monoenergetic images from spectral computed tomography.

OBJECTIVES: Quantitative computed tomography (CT) plays an increasingly important role in phenotyping airway diseases. Lung parenchyma and airway inflammation could be quantified by contrast enhancement at CT, but its investigation by multiphasic examinations is limited. We aimed to quantify lung parenchyma and airway wall attenuation in a single contrast-enhanced spectral detector CT acquisition. METHODS: For this cross-sectional retrospective study, 234 lung-healthy patients who underwent spectral CT in four different contrast phases (non-enhanced, pulmonary arterial, systemic arterial, and venous phase) were recruited. Virtual monoenergetic images were reconstructed from 40-160 keV, on which attenuations of segmented lung parenchyma and airway walls combined for 5th-10th subsegmental generations were assessed in Hounsfield Units (HU) by an in-house software. The spectral attenuation curve slope between 40 and 100 keV (λHU) was calculated. RESULTS: Mean lung density was higher at 40 keV compared to that at 100 keV in all groups (p < 0.001). λHU of lung attenuation was significantly higher in the systemic (1.7 HU/keV) and pulmonary arterial phase (1.3 HU/keV) compared to that in the venous phase (0.5 HU/keV) and non-enhanced (0.2 HU/keV) spectral CT (p < 0.001). Wall thickness and wall attenuation were higher at 40 keV compared to those at 100 keV for the pulmonary and systemic arterial phase (p ≤ 0.001). λHU for wall attenuation was significantly higher in the pulmonary arterial (1.8 HU/keV) and systemic arterial (2.0 HU/keV) compared to that in the venous (0.7 HU/keV) and non-enhanced (0.3 HU/keV) phase (p ≤ 0.002). CONCLUSIONS: Spectral CT may quantify lung parenchyma and airway wall enhancement with a single contrast phase acquisition, and may separate arterial and venous enhancement. Further studies are warranted to analyze spectral CT for inflammatory airway diseases. KEY POINTS: • Spectral CT may quantify lung parenchyma and airway wall enhancement with a single contrast phase acquisition. • Spectral CT may separate arterial and venous enhancement of lung parenchyma and airway wall. • The contrast enhancement can be quantified by calculating the spectral attenuation curve slope from virtual monoenergetic images.

MDCT-based longitudinal automated airway and air trapping analysis in school-age children with mild cystic fibrosis lung disease.

Objectives: Quantitative computed tomography (QCT) offers some promising markers to quantify cystic fibrosis (CF)-lung disease. Air trapping may precede irreversible bronchiectasis; therefore, the temporal interdependencies of functional and structural lung disease need to be further investigated. We aim to quantify airway dimensions and air trapping on chest CT of school-age children with mild CF-lung disease over two years. Methods: Fully-automatic software analyzed 144 serial spirometer-controlled chest CT scans of 36 children (median 12.1 (10.2-13.8) years) with mild CF-lung disease (median ppFEV1 98.5 (90.8-103.3) %) at baseline, 3, 12 and 24 months. The airway wall percentage (WP5-10), bronchiectasis index (BEI), as well as severe air trapping (A3) were calculated for the total lung and separately for all lobes. Mixed linear models were calculated, considering the lobar distribution of WP5-10, BEI and A3 cross-sectionally and longitudinally. Results: WP5-10 remained stable (P = 0.248), and BEI changed from 0.41 (0.28-0.7) to 0.54 (0.36-0.88) (P = 0.156) and A3 from 2.26% to 4.35% (P = 0.086) showing variability over two years. ppFEV1 was also stable (P = 0.276). A robust mixed linear model showed a cross-sectional, regional association between WP5-10 and A3 at each timepoint (P < 0.001). Further, BEI showed no cross-sectional, but another mixed model showed short-term longitudinal interdependencies with air trapping (P = 0.003). Conclusions: Robust linear/beta mixed models can still reveal interdependencies in medical data with high variability that remain hidden with simpler statistical methods. We could demonstrate cross-sectional, regional interdependencies between wall thickening and air trapping. Further, we show short-term regional interdependencies between air trapping and an increase in bronchiectasis. The data indicate that regional air trapping may precede the development of bronchiectasis. Quantitative CT may capture subtle disease progression and identify regional and temporal interdependencies of distinct manifestations of CF-lung disease.

Standardized airway wall thickness Pi10 from routine CT scans of COPD patients as imaging biomarker for disease severity, lung function decline, and mortality.

BACKGROUND: Chest computed tomography (CT) is increasingly used for phenotyping and monitoring of patients with COPD. The aim of this work was to evaluate the association of Pi10 as a measure of standardized airway wall thickness on CT with exacerbations, mortality, and response to triple therapy. METHODS: Patients of GOLD grades 1-4 of the COSYCONET cohort with prospective CT scans were included. Pi10 was automatically computed and analyzed for its relationship to COPD severity, comorbidities, lung function, respiratory therapy, and mortality over a 6-year period, using univariate and multivariate comparisons. RESULTS: We included n = 433 patients (61%male). Pi10 was dependent on both GOLD grades 1-4 (p = 0.009) and GOLD groups A-D (p = 0.008); it was particularly elevated in group D, and ROC analysis yielded a cut-off of 0.26 cm. Higher Pi10 was associated to lower FEV1 % predicted and higher RV/TLC, moreover the annual changes of lung function parameters (p < 0.05), as well as to an airway-dominated phenotype and a history of myocardial infarction (p = 0.001). These associations were confirmed in multivariate analyses. Pi10 was lower in patients receiving triple therapy, in particular in patients of GOLD groups C and D. Pi10 was also a significant predictor for mortality (p = 0.006), even after including multiple other predictors. CONCLUSION: In summary, Pi10 was found to be predictive for the course of the disease in COPD, in particular mortality. The fact that Pi10 was lower in patients with severe COPD receiving triple therapy might hint toward additional effects of this functional therapy on airway remodeling. REGISTRATION: ClinicalTrials.gov, Identifier: NCT01245933.

Reproducibility of pulmonary magnetic resonance angiography in adults with muco-obstructive pulmonary disease.

BACKGROUND: Recent studies support magnetic resonance angiography (MRA) as a diagnostic tool for pulmonary arterial disease. PURPOSE: To determine MRA image quality and reproducibility, and the dependence of MRA image quality and reproducibility on disease severity in patients with chronic obstructive pulmonary disease (COPD) and cystic fibrosis (CF). MATERIAL AND METHODS: Twenty patients with COPD (mean age 66.5 ± 8.9 years; FEV1% = 42.0 ± 13.3%) and 15 with CF (mean age 29.3 ± 9.3 years; FEV1% = 66.6 ± 15.8%) underwent morpho-functional chest magnetic resonance imaging (MRI) including time-resolved MRA twice one month apart (MRI1, MRI2), and COPD patients underwent non-contrast computed tomography (CT). Image quality was assessed visually using standardized subjective 5-point scales. Contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) were measured by regions of interest. Disease severity was determined by spirometry, a well-evaluated chest MRI score, and by computational CT emphysema index (EI) for COPD. RESULTS: Subjective image quality was diagnostic for all MRA at MRI1 and MRI2 (mean score = 4.7 ± 0.6). CNR and SNR were 4 43.8 ± 8.7 and 50.5 ± 8.7, respectively. Neither image quality score nor CNR or SNR correlated with FEV1% or chest MRI score for COPD and CF (r = 0.239-0.248). CNR and SNR did not change from MRI1 to MRI2 (P = 0.434-0.995). Further, insignificant differences in CNR and SNR between MRA at MRI1 and MRI2 did not correlate with FEV1% nor chest MRI score in COPD and CF (r = -0.238-0.183), nor with EI in COPD (r = 0.100-0.111). CONCLUSION: MRA achieved diagnostic quality in COPD and CF patients and was highly reproducible irrespective of disease severity. This supports MRA as a robust alternative to CT in patients with underlying muco-obstructive lung disease.

Quantitative CT analysis of lung parenchyma to improve malignancy risk estimation in incidental pulmonary nodules.

OBJECTIVES: To assess the value of quantitative computed tomography (QCT) of the whole lung and nodule-bearing lobe regarding pulmonary nodule malignancy risk estimation. METHODS: A total of 251 subjects (median [IQR] age, 65 (57-73) years; 37% females) with pulmonary nodules on non-enhanced thin-section CT were retrospectively included. Twenty percent of the nodules were malignant, the remainder benign either histologically or at least 1-year follow-up. CT scans were subjected to in-house software, computing parameters such as mean lung density (MLD) or peripheral emphysema index (pEI). QCT variable selection was performed using logistic regression; selected variables were integrated into the Mayo Clinic and the parsimonious Brock Model. RESULTS: Whole-lung analysis revealed differences between benign vs. malignant nodule groups in several parameters, e.g. the MLD (-766 vs. -790 HU) or the pEI (40.1 vs. 44.7 %). The proposed QCT model had an area-under-the-curve (AUC) of 0.69 (95%-CI, 0.62-0.76) based on all available data. After integrating MLD and pEI into the Mayo Clinic and Brock Model, the AUC of both clinical models improved (AUC, 0.91 to 0.93 and 0.88 to 0.91, respectively). The lobe-specific analysis revealed that the nodule-bearing lobes had less emphysema than the rest of the lung regarding benign (EI, 0.5 vs. 0.7 %; p < 0.001) and malignant nodules (EI, 1.2 vs. 1.7 %; p = 0.001). CONCLUSIONS: Nodules in subjects with higher whole-lung metrics of emphysema and less fibrosis are more likely to be malignant; hereby the nodule-bearing lobes have less emphysema. QCT variables could improve the risk assessment of incidental pulmonary nodules. KEY POINTS: • Nodules in subjects with higher whole-lung metrics of emphysema and less fibrosis are more likely to be malignant. • The nodule-bearing lobes have less emphysema compared to the rest of the lung. • QCT variables could improve the risk assessment of incidental pulmonary nodules.

AI-Supported Comprehensive Detection and Quantification of Biomarkers of Subclinical Widespread Diseases at Chest CT for Preventive Medicine.

Automated image analysis plays an increasing role in radiology in detecting and quantifying image features outside of the perception of human eyes. Common AI-based approaches address a single medical problem, although patients often present with multiple interacting, frequently subclinical medical conditions. A holistic imaging diagnostics tool based on artificial intelligence (AI) has the potential of providing an overview of multi-system comorbidities within a single workflow. An interdisciplinary, multicentric team of medical experts and computer scientists designed a pipeline, comprising AI-based tools for the automated detection, quantification and characterization of the most common pulmonary, metabolic, cardiovascular and musculoskeletal comorbidities in chest computed tomography (CT). To provide a comprehensive evaluation of each patient, a multidimensional workflow was established with algorithms operating synchronously on a decentralized Joined Imaging Platform (JIP). The results of each patient are transferred to a dedicated database and summarized as a structured report with reference to available reference values and annotated sample images of detected pathologies. Hence, this tool allows for the comprehensive, large-scale analysis of imaging-biomarkers of comorbidities in chest CT, first in science and then in clinical routine. Moreover, this tool accommodates the quantitative analysis and classification of each pathology, providing integral diagnostic and prognostic value, and subsequently leading to improved preventive patient care and further possibilities for future studies.

COVID-19 pneumonia: Prediction of patient outcome by CT-based quantitative lung parenchyma analysis combined with laboratory parameters.

OBJECTIVES: To evaluate the prognostic value of fully automatic lung quantification based on spectral computed tomography (CT) and laboratory parameters for combined outcome prediction in COVID-19 pneumonia. METHODS: CT images of 53 hospitalized COVID-19 patients including virtual monochromatic reconstructions at 40-140keV were analyzed using a fully automated software system. Quantitative CT (QCT) parameters including mean and percentiles of lung density, fibrosis index (FIBI-700, defined as the percentage of segmented lung voxels ≥-700 HU), quantification of ground-glass opacities and well-aerated lung areas were analyzed. QCT parameters were correlated to laboratory and patient outcome parameters (hospitalization, days on intensive care unit, invasive and non-invasive ventilation). RESULTS: Best correlations were found for laboratory parameters LDH (r = 0.54), CRP (r = 0.49), Procalcitonin (r = 0.37) and partial pressure of oxygen (r = 0.35) with the QCT parameter 75th percentile of lung density. LDH, Procalcitonin, 75th percentile of lung density and FIBI-700 were the strongest independent predictors of patients' outcome in terms of days of invasive ventilation. The combination of LDH and Procalcitonin with either 75th percentile of lung density or FIBI-700 achieved a r2 of 0.84 and 1.0 as well as an area under the receiver operating characteristic curve (AUC) of 0.99 and 1.0 for the prediction of the need of invasive ventilation. CONCLUSIONS: QCT parameters in combination with laboratory parameters could deliver a feasible prognostic tool for the prediction of invasive ventilation in patients with COVID-19 pneumonia.

µCT to quantify muco-obstructive lung disease and effects of neutrophil elastase knockout in mice.

Muco-obstructive lung diseases are characterized by airway obstruction and hyperinflation, which can be quantified by imaging. Our aim was to evaluate µCT for longitudinal quantification of muco-obstructive lung disease in ß-epithelial Na+ channel overexpressing (Scnn1b-TG) mice and of the effects of neutrophil elastase (NE) knockout on its progression. Lungs from wild-type (WT), NE-/-, Scnn1b-TG, and Scnn1b-TG/NE-/- mice were scanned with 9-µm resolution at 0, 5, 14, and 60 days of age, and airway and parenchymal disease was quantified. Mucus adhesion lesions (MAL) were persistently increased in Scnn1b-TG compared with WT mice from 0 days (20.25 ± 6.50 vs. 9.60 ± 2.07, P < 0.05), and this effect was attenuated in Scnn1b-TG/NE-/- mice (5.33 ± 3.67, P < 0.001). Airway wall area percentage (WA%) was increased in Scnn1b-TG mice compared with WT from 14 days onward (59.2 ± 6.3% vs. 49.8 ± 9.0%, P < 0.001) but was similar in Scnn1b-TG/NE-/- compared with WT at 60 days (46.4 ± 9.2% vs. 45.4 ± 11.5%, P = 0.97). Air proportion (Air%) and mean linear intercept (Lm) were persistently increased in Scnn1b-TG compared with WT from 5 days on (53.9 ± 4.5% vs. 30.0 ± 5.5% and 78.82 ± 8.44 µm vs. 65.66 ± 4.15 µm, respectively, P < 0.001), whereas in Scnn1b-TG/NE-/-, Air% and Lm were similar to WT from birth (27.7 ± 5.5% vs. 27.2 ± 5.9%, P = 0.92 and 61.48 ± 9.20 µm vs. 61.70 ± 6.73 µm, P = 0.93, respectively). Our results suggest that µCT is sensitive to detect the onset and progression of muco-obstructive lung disease and effects of genetic deletion of NE on morphology of airways and lung parenchyma in Scnn1b-TG mice, and that it may serve as a sensitive endpoint for preclinical studies of novel therapeutic interventions for muco-obstructive lung diseases.

Quantification of pulmonary perfusion abnormalities using DCE-MRI in COPD: comparison with quantitative CT and pulmonary function.

OBJECTIVES: Pulmonary perfusion abnormalities are prevalent in patients with chronic obstructive pulmonary disease (COPD), are potentially reversible, and may be associated with emphysema development. Therefore, we aimed to evaluate the clinical meaningfulness of perfusion defects in percent (QDP) using DCE-MRI. METHODS: We investigated a subset of baseline DCE-MRIs, paired inspiratory/expiratory CTs, and pulmonary function testing (PFT) of 83 subjects (age = 65.7 ± 9.0 years, patients-at-risk, and all GOLD groups) from one center of the "COSYCONET" COPD cohort. QDP was computed from DCE-MRI using an in-house developed quantification pipeline, including four different approaches: Otsu's method, k-means clustering, texture analysis, and 80th percentile threshold. QDP was compared with visual MRI perfusion scoring, CT parametric response mapping (PRM) indices of emphysema (PRMEmph) and functional small airway disease (PRMfSAD), and FEV1/FVC from PFT. RESULTS: All QDP approaches showed high correlations with the MRI perfusion score (r = 0.67 to 0.72, p < 0.001), with the highest association based on Otsu's method (r = 0.72, p < 0.001). QDP correlated significantly with all PRM indices (p < 0.001), with the strongest correlations with PRMEmph (r = 0.70 to 0.75, p < 0.001). QDP was distinctly higher than PRMEmph (mean difference = 35.85 to 40.40) and PRMfSAD (mean difference = 15.12 to 19.68), but in close agreement when combining both PRM indices (mean difference = 1.47 to 6.03) for all QDP approaches. QDP correlated moderately with FEV1/FVC (r = - 0.54 to - 0.41, p < 0.001). CONCLUSION: QDP is associated with established markers of disease severity and the extent corresponds to the CT-derived combined extent of PRMEmph and PRMfSAD. We propose to use QDP based on Otsu's method for future clinical studies in COPD. KEY POINTS: • QDP quantified from DCE-MRI is associated with visual MRI perfusion score, CT PRM indices, and PFT. • The extent of QDP from DCE-MRI corresponds to the combined extent of PRMEmph and PRMfSAD from CT. • Assessing pulmonary perfusion abnormalities using DCE-MRI with QDP improved the correlations with CT PRM indices and PFT compared to the quantification of pulmonary blood flow and volume.

Fibroblast Activation Protein-Specific PET/CT Imaging in Fibrotic Interstitial Lung Diseases and Lung Cancer: A Translational Exploratory Study.

Interstitial lung diseases (ILDs) comprise over 200 parenchymal lung disorders. Among them, fibrosing ILDs, especially idiopathic pulmonary fibrosis, are associated with a poor prognosis, whereas some other ILDs, such as sarcoidosis, have a much better prognosis. A high proportion manifests as fibrotic ILD (fILD). Lung cancer (LC) is a frequent complication of fILD. Activated fibroblasts are crucial for fibrotic processes in fILD. The aim of this exploratory study was to evaluate the imaging properties of static and dynamic fibroblast activation protein (FAP) inhibitor (FAPI) PET/CT in various types of fILD and to confirm FAP expression in fILD lesions by FAP immunohistochemistry of human fILD biopsy samples and of lung sections of genetically engineered (Nedd4-2-/- ) mice with an idiopathic pulmonary fibrosislike lung disease. Methods: PET scans of 15 patients with fILD and suspected LC were acquired 10, 60, and 180 min after the administration of 150-250 MBq of a 68Ga-labeled FAPI tracer (FAPI-46). In 3 patients, dynamic scans over 40 min were performed instead of imaging after 10 min. The SUVmax and SUVmean of fibrotic lesions and LC were measured and CT-density-corrected. Target-to-background ratios (TBRs) were calculated. PET imaging was correlated with CT-based fibrosis scores. Time-activity curves derived from dynamic imaging were analyzed. FAP immunohistochemistry of 4 human fILD biopsy samples and of fibrotic lungs of Nedd4-2-/- mice was performed. Results: fILD lesions as well as LC showed markedly elevated 68Ga-FAPI uptake (density-corrected SUVmax and SUVmean 60 min after injection: 11.12 ± 6.71 and 4.29 ± 1.61, respectively, for fILD lesions and 16.69 ± 9.35 and 6.44 ± 3.29, respectively, for LC) and high TBR (TBR of density-corrected SUVmax and SUVmean 60 min after injection: 2.30 ± 1.47 and 1.67 ± 0.79, respectively, for fILD and 3.90 ± 2.36 and 2.37 ± 1.14, respectively, for LC). SUVmax and SUVmean decreased over time, with a stable TBR for fILD and a trend toward an increasing TBR in LC. Dynamic imaging showed differing time-activity curves for fILD and LC. 68Ga-FAPI uptake showed a positive correlation with the CT-based fibrosis index. Immunohistochemistry of human biopsy samples and the lungs of Nedd4-2-/- mice showed a patchy expression of FAP in fibrotic lesions, preferentially in the transition zone to healthy lung parenchyma. Conclusion:68Ga-FAPI PET/CT imaging is a promising new imaging modality for fILD and LC. Its potential clinical value for monitoring and therapy evaluation of fILD should be investigated in future studies.

Echo Time-Dependent Observed Lung T1 in Patients With Chronic Obstructive Pulmonary Disease in Correlation With Quantitative Imaging and Clinical Indices.

BACKGROUND: There is a clinical need for imaging-derived biomarkers for the management of chronic obstructive pulmonary disease (COPD). Observed pulmonary T1 (T1 (TE)) depends on the echo-time (TE) and reflects regional pulmonary function. PURPOSE: To investigate the potential diagnostic value of T1 (TE) for the assessment of lung disease in COPD patients by determining correlations with clinical parameters and quantitative CT. STUDY TYPE: Prospective non-randomized diagnostic study. POPULATION: Thirty COPD patients (67.7 ± 6.6 years). Data from a previous study (15 healthy volunteers [26.2 ± 3.9 years) were used as reference. FIELD STRENGTH/SEQUENCE: Study participants were examined at 1.5 T using dynamic contrast-enhanced three-dimensional gradient echo keyhole perfusion sequence and a multi-echo inversion recovery two-dimensional UTE (ultra-short TE) sequence for T1 (TE) mapping at TE1-5  = 70 µsec, 500 µsec, 1200 µsec, 1650 µsec, and 2300 µsec. ASSESSMENT: Perfusion images were scored by three radiologists. T1 (TE) was automatically quantified. Computed tomography (CT) images were quantified in software (qCT). Clinical parameters including pulmonary function testing were also acquired. STATISTICAL TESTS: Spearman rank correlation coefficients (ρ) were calculated between T1 (TE) and perfusion scores, clinical parameters and qCT. A P-value <0.05 was considered statistically significant. RESULTS: Median values were T1 (TE1-5 ) = 644 ± 78 msec, 835 ± 92 msec, 835 ± 87 msec, 831 ± 131 msec, 893 ± 220 msec, all significantly shorter than previously reported in healthy subjects. A significant increase of T1 was observed from TE1 to TE2 , with no changes from TE2 to TE3 (P = 0.48), TE3 to TE4 (P = 0.94) or TE4 to TE5 (P = 0.02) which demonstrates an increase at shorter TEs than in healthy subjects. Moderate to strong Spearman's correlations between T1 and parameters including the predicted diffusing capacity for carbon monoxide (DLCO, ρ < 0.70), mean lung density (MLD, ρ < 0.72) and the perfusion score (ρ > -0.69) were found. Overall, correlations were strongest at TE2 , weaker at TE1 and rarely significant at TE4 -TE5 . DATA CONCLUSION: In COPD patients, the increase of T1 (TE) with TE occurred at shorter TEs than previously found in healthy subjects. Together with the lack of correlation between T1 and clinical parameters of disease at longer TEs, this suggests that T1 (TE) quantification in COPD patients requires shorter TEs. The TE-dependence of correlations implies that T1 (TE) mapping might be developed further to provide diagnostic information beyond T1 at a single TE. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 1.

A calibration CT mini-lung-phantom created by 3-D printing and subtractive manufacturing.

We describe the creation and characterization of a calibration CT mini-lung-phantom incorporating simulated airways and ground-glass densities. Ten duplicate mini-lung-phantoms with Three-Dimensional (3-D) printed tubes simulating airways and gradated density polyurethane foam blocks were designed and built. Dimensional accuracy and CT numbers were measured using micro-CT and clinical CT scanners. Micro-CT images of airway tubes demonstrated an average dimensional variation of 0.038 mm from nominal values. The five different densities of incorporated foam blocks, simulating ground-glass, showed mean CT numbers (±standard deviation) of -897.0 ± 1.5, -844.1 ± 1.5, -774.1 ± 2.6, -695.3 ± 1.6, and -351.0 ± 3.7 HU, respectively. Three-Dimensional printing and subtractive manufacturing enabled rapid, cost-effective production of ground-truth calibration mini-lung-phantoms with low inter-sample variation that can be scanned simultaneously with the patient undergoing lung quantitative CT.

Fully Automated Segmentation of Pulmonary Fibrosis Using Different Software Tools.

OBJECTIVE: Evaluation of software tools for segmentation, quantification, and characterization of fibrotic pulmonary parenchyma changes will strengthen the role of CT as biomarkers of disease extent, evolution, and response to therapy in idiopathic pulmonary fibrosis (IPF) patients. METHODS: 418 nonenhanced thin-section MDCTs of 127 IPF patients and 78 MDCTs of 78 healthy individuals were analyzed through 3 fully automated, completely different software tools: YACTA, LUFIT, and IMBIO. The agreement between YACTA and LUFIT on segmented lung volume and 80th (reflecting fibrosis) and 40th (reflecting ground-glass opacity) percentile of the lung density histogram was analyzed using Bland-Altman plots. The fibrosis and ground-glass opacity segmented by IMBIO (lung texture analysis software tool) were included in specific regression analyses. RESULTS: In the IPF-group, LUFIT outperformed YACTA by segmenting more lung volume (mean difference 242 mL, 95% limits of agreement -54 to 539 mL), as well as quantifying higher 80th (76 HU, -6 to 158 HU) and 40th percentiles (9 HU, -73 to 90 HU). No relevant differences were revealed in the control group. The 80th/40th percentile as quantified by LUFIT correlated positively with the percentage of fibrosis/ground-glass opacity calculated by IMBIO (r = 0.78/r = 0.92). CONCLUSIONS: In terms of segmentation of pulmonary fibrosis, LUFIT as a shape model-based segmentation software tool is superior to the threshold-based YACTA, tool, since the density of (severe) fibrosis is similar to that of the surrounding soft tissues. Therefore, shape modeling as used in LUFIT may serve as a valid tool in the quantification of IPF, since this mainly affects the subpleural space.