Explore the vast range of titles available.

The risk factors for development of fibrotic-like radiographic abnormalities after severe COVID-19 are incompletely described and the extent to which CT findings correlate with symptoms and physical function after hospitalisation remains unclear. At 4 months after hospitalisation, fibrotic-like patterns were more common in those who underwent mechanical ventilation (72%) than in those who did not (20%). We demonstrate that severity of initial illness, duration of mechanical ventilation, lactate dehydrogenase on admission and leucocyte telomere length are independent risk factors for fibrotic-like radiographic abnormalities. These fibrotic-like changes correlate with lung function, cough and measures of frailty, but not with dyspnoea.

Electrical impedance tomography (EIT) has undergone 30 years of development. Functional chest examinations with this technology are considered clinically relevant, especially for monitoring regional lung ventilation in mechanically ventilated patients and for regional pulmonary function testing in patients with chronic lung diseases. As EIT becomes an established medical technology, it requires consensus examination, nomenclature, data analysis and interpretation schemes. Such consensus is needed to compare, understand and reproduce study findings from and among different research groups, to enable large clinical trials and, ultimately, routine clinical use. Recommendations of how EIT findings can be applied to generate diagnoses and impact clinical decision-making and therapy planning are required. This consensus paper was prepared by an international working group, collaborating on the clinical promotion of EIT called TRanslational EIT developmeNt stuDy group. It addresses the stated needs by providing (1) a new classification of core processes involved in chest EIT examinations and data analysis, (2) focus on clinical applications with structured reviews and outlooks (separately for adult and neonatal/paediatric patients), (3) a structured framework to categorise and understand the relationships among analysis approaches and their clinical roles, (4) consensus, unified terminology with clinical user-friendly definitions and explanations, (5) a review of all major work in thoracic EIT and (6) recommendations for future development (193 pages of online supplements systematically linked with the chief sections of the main document). We expect this information to be useful for clinicians and researchers working with EIT, as well as for industry producers of this technology.

BackgroundEarly detection of lung cancer saves lives, as demonstrated by the two largest published low-dose CT screening trials. Optimal implementation depends on our ability to identify those most at risk.MethodsVersion 2 of the Liverpool Lung Project risk score (LLPv2) was developed from case-control data in Liverpool and further adapted when applied for selection of subjects for the UK Lung Screening Trial. The objective was to produce version 3 (LLPv3) of the model, by calibration to national figures for 2017. We validated both LLPv2 and LLPv3 using questionnaire data from 75 958 individuals, followed up for lung cancer over 5 years. We validated both discrimination, using receiver operating characteristic (ROC) analysis, and absolute incidence, by comparing deciles of predicted incidence with observed incidence. We calculated proportionate difference as the percentage excess or deficit of observed cancers compared with those predicted. We also carried out Hosmer-Lemeshow tests.ResultsThere were 599 lung cancers diagnosed over 5 years. The discrimination of both LLPv2 and LLPv3 was significant with an area under the ROC curve of 0.81 (95% CI 0.79 to 0.82). However, LLPv2 overestimated absolute risk in the population. The proportionate difference was −58.3% (95% CI −61.6% to −54.8%), that is, the actual number of cancers was only 42% of the number predicted.In LLPv3, calibrated to national 2017 figures, the proportionate difference was −22.0% (95% CI −28.1% to −15.5%).ConclusionsWhile LLPv2 and LLPv3 have the same discriminatory power, LLPv3 improves the absolute lung cancer risk prediction and should be considered for use in further UK implementation studies.

BackgroundThere is no accurate, non-invasive measurement to estimate the degree of pulmonary oedema in acute respiratory distress syndrome (ARDS). We developed the Radiographic Assessment of Lung Oedema (RALE) score to evaluate the extent and density of alveolar opacities on chest radiographs. After first comparing the RALE score to gravimetric assessment of pulmonary oedema in organ donors, we then evaluated the RALE score in patients with ARDS for its relationship to oxygenation and clinical outcomes.MethodsWe compared radiographs with excised lung weights from 72 organ donors (derivation cohort) and radiographs with clinical data from 174 patients with ARDS in the ARDSNet Fluid and Catheter Treatment Trial (validation cohort). To calculate RALE, each radiographic quadrant was scored for extent of consolidation (0–4) and density of opacification (1–3). The product of the consolidation and density scores for each of the four quadrants was summed (maximum score=48).ResultsAgreement between two independent reviewers for RALE score was excellent (intraclass correlation coefficient=0.93, 95% CI 0.91 to 0.95). In donors, pre-procurement RALE score correlated with height-adjusted total lung weight (ρ=0.59, p<0.001). In patients with ARDS, higher RALE scores were independently associated with lower PaO2/fractional inspired oxygen and worse survival. Conservative fluid management significantly decreased RALE score over 3 days compared with liberal fluid management.ConclusionsThe RALE score can be used to assess both the extent of pulmonary oedema and the severity of ARDS, by utilising information that is already obtained routinely, safely and inexpensively in every patient with ARDS. This novel non-invasive measure should be useful for assessing ARDS severity and monitoring response to therapy.

BackgroundLow-dose CT screening can reduce lung cancer-related mortality. However, most screen-detected pulmonary abnormalities do not develop into cancer and it often remains challenging to identify malignant nodules, particularly among indeterminate nodules. We aimed to develop and assess prediction models based on radiological features to discriminate between benign and malignant pulmonary lesions detected on a baseline screen.MethodsUsing four international lung cancer screening studies, we extracted 2060 radiomic features for each of 16 797 nodules (513 malignant) among 6865 participants. After filtering out low-quality radiomic features, 642 radiomic and 9 epidemiological features remained for model development. We used cross-validation and grid search to assess three machine learning (ML) models (eXtreme Gradient Boosted Trees, random forest, least absolute shrinkage and selection operator (LASSO)) for their ability to accurately predict risk of malignancy for pulmonary nodules. We report model performance based on the area under the curve (AUC) and calibration metrics in the held-out test set.ResultsThe LASSO model yielded the best predictive performance in cross-validation and was fit in the full training set based on optimised hyperparameters. Our radiomics model had a test-set AUC of 0.93 (95% CI 0.90 to 0.96) and outperformed the established Pan-Canadian Early Detection of Lung Cancer model (AUC 0.87, 95% CI 0.85 to 0.89) for nodule assessment. Our model performed well among both solid (AUC 0.93, 95% CI 0.89 to 0.97) and subsolid nodules (AUC 0.91, 95% CI 0.85 to 0.95).ConclusionsWe developed highly accurate ML models based on radiomic and epidemiological features from four international lung cancer screening studies that may be suitable for assessing indeterminate screen-detected pulmonary nodules for risk of malignancy.

IntroductionPersisting respiratory symptoms in COVID-19 survivors may be related to development of pulmonary fibrosis. We assessed the proportion of chest CT scans and pulmonary function tests consistent with parenchymal lung disease in the follow-up of people hospitalised with COVID-19 and viral pneumonitis.MethodsSystematic review and random effects meta-analysis of proportions using studies of adults hospitalised with SARS-CoV-2, SARS-CoV, MERS-CoV or influenza pneumonia and followed up within 12 months. Searches performed in MEDLINE and Embase. Primary outcomes were proportion of radiological sequelae on CT scans; restrictive impairment; impaired gas transfer. Heterogeneity was explored in meta-regression.ResultsNinety-five studies (98.9% observational) were included in qualitative synthesis, 70 were suitable for meta-analysis including 60 SARS-CoV-2 studies with a median follow-up of 3 months. In SARS-CoV-2, the overall estimated proportion of inflammatory sequelae was 50% during follow-up (0.50; 95% CI 0.41 to 0.58; I2=95%), fibrotic sequelae were estimated in 29% (0.29; 95% CI 0.22 to 0.37; I2=94.1%). Follow-up time was significantly associated with estimates of inflammatory sequelae (−0.036; 95% CI −0.068 to –0.004; p=0.029), associations with fibrotic sequelae did not reach significance (−0.021; 95% CI −0.051 to 0.009; p=0.176). Impaired gas transfer was estimated at 38% of lung function tests (0.38 95% CI 0.32 to 0.44; I2=92.1%), which was greater than restrictive impairment (0.17; 95% CI 0.13 to 0.23; I2=92.5%), neither were associated with follow-up time (p=0.207; p=0.864).DiscussionSequelae consistent with parenchymal lung disease were observed following COVID-19 and other viral pneumonitis. Estimates should be interpreted with caution due to high heterogeneity, differences in study casemix and initial severity.PROSPERO registration numberCRD42020183139.

129 Xe MRI red blood cell to alveolar tissue plasma ratio (RBC:TP) abnormalities have been observed in ever-hospitalised and never-hospitalised people with postacute COVID-19 syndrome (PACS). But, it is not known if such abnormalities resolve when symptoms and quality-of-life scores improve. We evaluated 21 participants with PACS, 7±4 months (baseline) and 14±4 months (follow-up) postinfection. Significantly improved diffusing capacity of the lung for carbon monoxide (DL CO , Δ=14% pred ;95%CI 7 to 21, p<0.001), postexertional dyspnoea (Δ=−0.7; 95%CI=−0.2 to –1.2, p=0.019), St George’s Respiratory Questionnaire-score (SGRQ Δ=−6; 95% CI=−1 to –11, p=0.044) but not RBC:TP (Δ=0.03; 95% CI=0.01 to 0.05, p=0.051) were observed at 14 months. DL CO correlated with RBC:TP (r=0.60, 95% CI=0.22 to 0.82, p=0.004) at 7 months. While DL CO and SGRQ measurements improved, these values did not normalise 14 months post-infection. ClinicalTrials.gov NCT04584671 .

The past 5 years have seen an explosion of interest in the use of artificial intelligence (AI) and machine learning techniques in medicine. This has been driven by the development of deep neural networks (DNNs)—complex networks residing in silico but loosely modelled on the human brain—that can process complex input data such as a chest radiograph image and output a classification such as ‘normal’ or ‘abnormal’. DNNs are ‘trained’ using large banks of images or other input data that have been assigned the correct labels. DNNs have shown the potential to equal or even surpass the accuracy of human experts in pattern recognition tasks such as interpreting medical images or biosignals. Within respiratory medicine, the main applications of AI and machine learning thus far have been the interpretation of thoracic imaging, lung pathology slides and physiological data such as pulmonary function tests. This article surveys progress in this area over the past 5 years, as well as highlighting the current limitations of AI and machine learning and the potential for future developments.

BackgroundCystic fibrosis (CF) lung disease is characterised by progressive airway wall thickening and widening. We aimed to validate an artificial intelligence-based algorithm to assess dimensions of all visible bronchus-artery (BA) pairs on chest CT scans from patients with CF.MethodsThe algorithm fully automatically segments the bronchial tree; identifies bronchial generations; matches bronchi with the adjacent arteries; measures for each BA-pair bronchial outer diameter (Bout), bronchial lumen diameter (Bin), bronchial wall thickness (Bwt) and adjacent artery diameter (A); and computes Bout/A, Bin/A and Bwt/A for each BA pair from the segmental bronchi to the last visible generation. Three datasets were used to validate the automatic BA analysis. First BA analysis was executed on 23 manually annotated CT scans (11 CF, 12 control subjects) to compare automatic with manual BA-analysis outcomes. Furthermore, the BA analysis was executed on two longitudinal datasets (Copenhagen 111 CTs, ataluren 347 CTs) to assess longitudinal BA changes and compare them with manual scoring results.ResultsThe automatic and manual BA analysis showed no significant differences in quantifying bronchi. For the longitudinal datasets the automatic BA analysis detected 247 and 347 BA pairs/CT in the Copenhagen and ataluren dataset, respectively. A significant increase of 0.02 of Bout/A and Bin/A was detected for Copenhagen dataset over an interval of 2 years, and 0.03 of Bout/A and 0.02 of Bin/A for ataluren dataset over an interval of 48 weeks (all p<0.001). The progression of 0.01 of Bwt/A was detected only in the ataluren dataset (p<0.001). BA-analysis outcomes showed weak to strong correlations (correlation coefficient from 0.29 to 0.84) with manual scoring results for airway disease.ConclusionThe BA analysis can fully automatically analyse a large number of BA pairs on chest CTs to detect and monitor progression of bronchial wall thickening and bronchial widening in patients with CF.