Review of Hyperpolarized Pulmonary Functional 129 Xe MR for Long-COVID.

The respiratory consequences of acute COVID-19 infection and related symptoms tend to resolve 4 weeks post-infection. However, for some patients, new, recurrent, or persisting symptoms remain beyond the acute phase and persist for months, post-infection. The symptoms that remain have been referred to as long-COVID. A number of research sites employed 129 Xe magnetic resonance imaging (MRI) during the pandemic and evaluated patients post-infection, months after hospitalization or home-based care as a way to better understand the consequences of infection on 129 Xe MR gas-exchange and ventilation imaging. A systematic review and comprehensive search were employed using MEDLINE via PubMed (April 2023) using the National Library of Medicine's Medical Subject Headings and key words: post-COVID-19, MRI, 129 Xe, long-COVID, COVID pneumonia, and post-acute COVID-19 syndrome. Fifteen peer-reviewed manuscripts were identified including four editorials, a single letter to the editor, one review article, and nine original research manuscripts (2020-2023). MRI and MR spectroscopy results are summarized from these prospective, controlled studies, which involved small sample sizes ranging from 9 to 76 participants. Key findings included: 1) 129 Xe MRI gas-exchange and ventilation abnormalities, 3 months post-COVID-19 infection, and 2) a combination of MRI gas-exchange and ventilation abnormalities alongside persistent symptoms in patients hospitalized and not hospitalized for COVID-19, 1-year post-infection. The persistence of respiratory symptoms and 129 Xe MRI abnormalities in the context of normal or nearly normal pulmonary function test results and chest computed tomography (CT) was consistent. Longitudinal improvements were observed in long-term follow-up of long-COVID patients but mean 129 Xe gas-exchange, ventilation heterogeneity values and symptoms remained abnormal, 1-year post-infection. Pulmonary functional MRI using inhaled hyperpolarized 129 Xe gas has played a role in detecting gas-exchange and ventilation abnormalities providing complementary information that may help develop our understanding of the root causes of long-COVID. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY: Stage 5.

Residual Lung Abnormalities after COVID-19 Hospitalization: Interim Analysis of the UKILD Post-COVID-19 Study.

Rationale: Shared symptoms and genetic architecture between coronavirus disease (COVID-19) and lung fibrosis suggest severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection may lead to progressive lung damage. Objectives: The UK Interstitial Lung Disease Consortium (UKILD) post-COVID-19 study interim analysis was planned to estimate the prevalence of residual lung abnormalities in people hospitalized with COVID-19 on the basis of risk strata. Methods: The PHOSP-COVID-19 (Post-Hospitalization COVID-19) study was used to capture routine and research follow-up within 240 days from discharge. Thoracic computed tomography linked by PHOSP-COVID-19 identifiers was scored for the percentage of residual lung abnormalities (ground-glass opacities and reticulations). Risk factors in linked computed tomography were estimated with Bayesian binomial regression, and risk strata were generated. Numbers within strata were used to estimate posthospitalization prevalence using Bayesian binomial distributions. Sensitivity analysis was restricted to participants with protocol-driven research follow-up. Measurements and Main Results: The interim cohort comprised 3,700 people. Of 209 subjects with linked computed tomography (median, 119 d; interquartile range, 83-155), 166 people (79.4%) had more than 10% involvement of residual lung abnormalities. Risk factors included abnormal chest X-ray (risk ratio [RR], 1.21; 95% credible interval [CrI], 1.05-1.40), percent predicted DlCO less than 80% (RR, 1.25; 95% CrI, 1.00-1.56), and severe admission requiring ventilation support (RR, 1.27; 95% CrI, 1.07-1.55). In the remaining 3,491 people, moderate to very high risk of residual lung abnormalities was classified at 7.8%, and posthospitalization prevalence was estimated at 8.5% (95% CrI, 7.6-9.5), rising to 11.7% (95% CrI, 10.3-13.1) in the sensitivity analysis. Conclusions: Residual lung abnormalities were estimated in up to 11% of people discharged after COVID-19-related hospitalization. Health services should monitor at-risk individuals to elucidate long-term functional implications.

Evaluation of the impact of artificial intelligence-assisted image interpretation on the diagnostic performance of clinicians in identifying pneumothoraces on plain chest X-ray: a multi-case multi-reader study.

BACKGROUND: Artificial intelligence (AI)-assisted image interpretation is a fast-developing area of clinical innovation. Most research to date has focused on the performance of AI-assisted algorithms in comparison with that of radiologists rather than evaluating the algorithms' impact on the clinicians who often undertake initial image interpretation in routine clinical practice. This study assessed the impact of AI-assisted image interpretation on the diagnostic performance of frontline acute care clinicians for the detection of pneumothoraces (PTX). METHODS: A multicentre blinded multi-case multi-reader study was conducted between October 2021 and January 2022. The online study recruited 18 clinician readers from six different clinical specialties, with differing levels of seniority, across four English hospitals. The study included 395 plain CXR images, 189 positive for PTX and 206 negative. The reference standard was the consensus opinion of two thoracic radiologists with a third acting as arbitrator. General Electric Healthcare Critical Care Suite (GEHC CCS) PTX algorithm was applied to the final dataset. Readers individually interpreted the dataset without AI assistance, recording the presence or absence of a PTX and a confidence rating. Following a 'washout' period, this process was repeated including the AI output. RESULTS: Analysis of the performance of the algorithm for detecting or ruling out a PTX revealed an overall AUROC of 0.939. Overall reader sensitivity increased by 11.4% (95% CI 4.8, 18.0, p=0.002) from 66.8% (95% CI 57.3, 76.2) unaided to 78.1% aided (95% CI 72.2, 84.0, p=0.002), specificity 93.9% (95% CI 90.9, 97.0) without AI to 95.8% (95% CI 93.7, 97.9, p=0.247). The junior reader subgroup showed the largest improvement at 21.7% (95% CI 10.9, 32.6), increasing from 56.0% (95% CI 37.7, 74.3) to 77.7% (95% CI 65.8, 89.7, p<0.01). CONCLUSION: The study indicates that AI-assisted image interpretation significantly enhances the diagnostic accuracy of clinicians in detecting PTX, particularly benefiting less experienced practitioners. While overall interpretation time remained unchanged, the use of AI improved diagnostic confidence and sensitivity, especially among junior clinicians. These findings underscore the potential of AI to support less skilled clinicians in acute care settings.

Reduction in Chest CT Severity and Improved Hospital Outcomes in SARS-CoV-2 Omicron Compared with Delta Variant Infection.

Background The SARS-Cov-2 Omicron variant demonstrates rapid spread but reduced disease severity. Studies evaluating lung imaging findings of Omicron infection versus non-Omicron infection remain lacking. Purpose To compare the Omicron variant with the SARS-CoV-2 Delta variant according to their chest CT radiologic pattern, biochemical parameters, clinical severity, and hospital outcomes after adjusting for vaccination status. Materials and Methods This retrospective study included hospitalized adult patients with reverse transcriptase-polymerase chain reaction test results positive for SARS-CoV-2, with CT pulmonary angiography performed within 7 days of admission between December 1, 2021, and January 14, 2022. Multiple readers performed blinded radiologic analyses that included RSNA CT classification, chest CT severity score (CTSS) (range, 0 [least severe] to 25 [most severe]), and CT imaging features, including bronchial wall thickening. Results A total of 106 patients (Delta group, n = 66; Omicron group, n = 40) were evaluated (overall mean age, 58 years ± 18 [SD]; 58 men). In the Omicron group, 37% of CT pulmonary angiograms (15 of 40 patients) were categorized as normal compared with 15% (10 of 66 patients) of angiograms in the Delta group (P = .016). A generalized linear model was used to control for confounding variables, including vaccination status, and Omicron infection was associated with a CTSS that was 7.2 points lower than that associated with Delta infection (ß = -7.2; 95% CI: -9.9, -4.5; P < .001). Bronchial wall thickening was more common with Omicron infection than with Delta infection (odds ratio [OR], 2.4; 95% CI: 1.01, 5.92; P = .04). A booster shot was associated with a protective effect for chest infection (median CTSS, 5; IQR, 0-11) when compared with unvaccinated individuals (median CTSS, 11; IQR, 7.5-14.0) (P = .03). The Delta variant was associated with a higher OR of severe disease (OR, 4.6; 95% CI: 1.2, 26; P = .01) and admission to a critical care unit (OR, 7.0; 95% CI: 1.5, 66; P = .004) when compared with the Omicron variant. Conclusion The SARS-CoV-2 Omicron variant was associated with fewer and less severe changes on chest CT images compared with the Delta variant. Patients with Omicron infection had greater frequency of bronchial wall thickening but less severe disease and improved hospital outcomes when compared with patients with Delta infection. © RSNA, 2022 Online supplemental material is available for this article.

Gas exchange and ventilation imaging of healthy and COPD subjects using hyperpolarized xenon-129 MRI and a 3D alveolar gas-exchange model.

OBJECTIVES: To investigate the utility of hyperpolarized xenon-129 (HPX) gas-exchange magnetic resonance imaging (MRI) and modeling in a chronic obstructive pulmonary disease (COPD) cohort in comparison to a minimal CT-diagnosed emphysema (MCTE) cohort and a healthy cohort. METHODS: A total of 25 subjects were involved in this study including COPD (n = 8), MCTE (n = 3), and healthy (n = 14) subjects. The COPD subjects were scanned using HPX ventilation, gas-exchange MRI, and volumetric CT. The healthy subjects were scanned using the same HPX gas-exchange MRI protocol with 9 of them scanned twice, 3 weeks apart. The coefficient of variation (CV) was used to quantify image heterogeneities. A three-dimensional computational fluid dynamic (CFD) model of gas exchange was used to derive functional volumes of pulmonary tissue, capillaries, and veins. RESULTS: The CVs of gas distributions in the images showed that there was a statistically significant difference between the COPD and healthy subjects (p < 0.0001). The functional volumes of pulmonary tissue, capillaries, and veins were significantly lower in the subjects with COPD than in the healthy subjects (p < 0.001). The functional volume of pulmonary tissue was found to be (i) statistically different between the healthy and MCTE groups (p = 0.02) and (ii) dependent on the age of the subjects in the healthy group (p = 0.0008) while their CVs (p = 0.13) were not. CONCLUSION: The novel HPX gas-exchange MRI and CFD model distinguished the healthy cohort from the MCTE and COPD cohorts. The proposed technique also showed that the functional volume of pulmonary tissue decreases with aging in the healthy group. KEY POINTS: • The ventilation and gas-exchange imaging with hyperpolarized xenon-129 MRI has enabled the identification of gas-exchange variation between COPD and healthy groups. • This novel technique was promising to be sensitive to minimal CT-diagnosed emphysema and age-related changes in gas-exchange parameter in a small pilot cohort.

Implementation of artificial intelligence in thoracic imaging-a what, how, and why guide from the European Society of Thoracic Imaging (ESTI).

This statement from the European Society of Thoracic imaging (ESTI) explains and summarises the essentials for understanding and implementing Artificial intelligence (AI) in clinical practice in thoracic radiology departments. This document discusses the current AI scientific evidence in thoracic imaging, its potential clinical utility, implementation and costs, training requirements and validation, its' effect on the training of new radiologists, post-implementation issues, and medico-legal and ethical issues. All these issues have to be addressed and overcome, for AI to become implemented clinically in thoracic radiology. KEY POINTS: • Assessing the datasets used for training and validation of the AI system is essential. • A departmental strategy and business plan which includes continuing quality assurance of AI system and a sustainable financial plan is important for successful implementation. • Awareness of the negative effect on training of new radiologists is vital.

Comparative accuracy and cost-effectiveness of dynamic contrast-enhanced CT and positron emission tomography in the characterisation of solitary pulmonary nodules.

INTRODUCTION: Dynamic contrast-enhanced CT (DCE-CT) and positron emission tomography/CT (PET/CT) have a high reported accuracy for the diagnosis of malignancy in solitary pulmonary nodules (SPNs). The aim of this study was to compare the accuracy and cost-effectiveness of these. METHODS: In this prospective multicentre trial, 380 participants with an SPN (8-30 mm) and no recent history of malignancy underwent DCE-CT and PET/CT. All patients underwent either biopsy with histological diagnosis or completed CT follow-up. Primary outcome measures were sensitivity, specificity and overall diagnostic accuracy for PET/CT and DCE-CT. Costs and cost-effectiveness were estimated from a healthcare provider perspective using a decision-model. RESULTS: 312 participants (47% female, 68.1±9.0 years) completed the study, with 61% rate of malignancy at 2 years. The sensitivity, specificity, positive predictive value and negative predictive values for DCE-CT were 95.3% (95% CI 91.3 to 97.5), 29.8% (95% CI 22.3 to 38.4), 68.2% (95% CI 62.4% to 73.5%) and 80.0% (95% CI 66.2 to 89.1), respectively, and for PET/CT were 79.1% (95% CI 72.7 to 84.2), 81.8% (95% CI 74.0 to 87.7), 87.3% (95% CI 81.5 to 91.5) and 71.2% (95% CI 63.2 to 78.1). The area under the receiver operator characteristic curve (AUROC) for DCE-CT and PET/CT was 0.62 (95% CI 0.58 to 0.67) and 0.80 (95% CI 0.76 to 0.85), respectively (p<0.001). Combined results significantly increased diagnostic accuracy over PET/CT alone (AUROC=0.90 (95% CI 0.86 to 0.93), p<0.001). DCE-CT was preferred when the willingness to pay per incremental cost per correctly treated malignancy was below £9000. Above £15 500 a combined approach was preferred. CONCLUSIONS: PET/CT has a superior diagnostic accuracy to DCE-CT for the diagnosis of SPNs. Combining both techniques improves the diagnostic accuracy over either test alone and could be cost-effective. TRIAL REGISTRATION NUMBER: NCT02013063.

Artificial Intelligence Tool for Assessment of Indeterminate Pulmonary Nodules Detected with CT.

Background Limited data are available regarding whether computer-aided diagnosis (CAD) improves assessment of malignancy risk in indeterminate pulmonary nodules (IPNs). Purpose To evaluate the effect of an artificial intelligence-based CAD tool on clinician IPN diagnostic performance and agreement for both malignancy risk categories and management recommendations. Materials and Methods This was a retrospective multireader multicase study performed in June and July 2020 on chest CT studies of IPNs. Readers used only CT imaging data and provided an estimate of malignancy risk and a management recommendation for each case without and with CAD. The effect of CAD on average reader diagnostic performance was assessed using the Obuchowski-Rockette and Dorfman-Berbaum-Metz method to calculate estimates of area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Multirater Fleiss κ statistics were used to measure interobserver agreement for malignancy risk and management recommendations. Results A total of 300 chest CT scans of IPNs with maximal diameters of 5-30 mm (50.0% malignant) were reviewed by 12 readers (six radiologists, six pulmonologists) (patient median age, 65 years; IQR, 59-71 years; 164 [55%] men). Readers' average AUC improved from 0.82 to 0.89 with CAD (P < .001). At malignancy risk thresholds of 5% and 65%, use of CAD improved average sensitivity from 94.1% to 97.9% (P = .01) and from 52.6% to 63.1% (P < .001), respectively. Average reader specificity improved from 37.4% to 42.3% (P = .03) and from 87.3% to 89.9% (P = .05), respectively. Reader interobserver agreement improved with CAD for both the less than 5% (Fleiss κ, 0.50 vs 0.71; P < .001) and more than 65% (Fleiss κ, 0.54 vs 0.71; P < .001) malignancy risk categories. Overall reader interobserver agreement for management recommendation categories (no action, CT surveillance, diagnostic procedure) also improved with CAD (Fleiss κ, 0.44 vs 0.52; P = .001). Conclusion Use of computer-aided diagnosis improved estimation of indeterminate pulmonary nodule malignancy risk on chest CT scans and improved interobserver agreement for both risk stratification and management recommendations. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Yanagawa in this issue.

Lung Abnormalities Detected with Hyperpolarized 129Xe MRI in Patients with Long COVID.

Background Post-COVID-19 condition encompasses symptoms following COVID-19 infection that linger at least 4 weeks after the end of active infection. Symptoms are wide ranging, but breathlessness is common. Purpose To determine if the previously described lung abnormalities seen on hyperpolarized (HP) pulmonary xenon 129 (129Xe) MRI scans in participants with post-COVID-19 condition who were hospitalized are also present in participants with post-COVID-19 condition who were not hospitalized. Materials and Methods In this prospective study, nonhospitalized participants with post-COVID-19 condition (NHLC) and posthospitalized participants with post-COVID-19 condition (PHC) were enrolled from June 2020 to August 2021. Participants underwent chest CT, HP 129Xe MRI, pulmonary function testing, and the 1-minute sit-to-stand test and completed breathlessness questionnaires. Control subjects underwent HP 129Xe MRI only. CT scans were analyzed for post-COVID-19 interstitial lung disease severity using a previously published scoring system and full-scale airway network (FAN) modeling. Analysis used group and pairwise comparisons between participants and control subjects and correlations between participant clinical and imaging data. Results A total of 11 NHLC participants (four men, seven women; mean age, 44 years ± 11 [SD]; 95% CI: 37, 50) and 12 PHC participants (10 men, two women; mean age, 58 years ±10; 95% CI: 52, 64) were included, with a significant difference in age between groups (P = .05). Mean time from infection was 287 days ± 79 (95% CI: 240, 334) and 143 days ± 72 (95% CI: 105, 190) in NHLC and PHC participants, respectively. NHLC and PHC participants had normal or near normal CT scans (mean, 0.3/25 ± 0.6 [95% CI: 0, 0.63] and 7/25 ± 5 [95% CI: 4, 10], respectively). Gas transfer (Dlco) was different between NHLC and PHC participants (mean Dlco, 76% ± 8 [95% CI: 73, 83] vs 86% ± 8 [95% CI: 80, 91], respectively; P = .04), but there was no evidence of other differences in lung function. Mean red blood cell-to-tissue plasma ratio was different between volunteers (mean, 0.45 ± 0.07; 95% CI: 0.43, 0.47]) and PHC participants (mean, 0.31 ± 0.10; 95% CI: 0.24, 0.37; P = .02) and between volunteers and NHLC participants (mean, 0.37 ± 0.10; 95% CI: 0.31, 0.44; P = .03) but not between NHLC and PHC participants (P = .26). FAN results did not correlate with Dlco) or HP 129Xe MRI results. Conclusion Nonhospitalized participants with post-COVID-19 condition (NHLC) and posthospitalized participants with post-COVID-19 condition (PHC) showed hyperpolarized pulmonary xenon 129 MRI and red blood cell-to-tissue plasma abnormalities, with NHLC participants demonstrating lower gas transfer than PHC participants despite having normal CT findings. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Parraga and Matheson in this issue.

Utility of pre-treatment FDG PET/CT-derived machine learning models for outcome prediction in classical Hodgkin lymphoma.

OBJECTIVES: Relapse occurs in ~20% of patients with classical Hodgkin lymphoma (cHL) despite treatment adaption based on 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography/computed tomography response. The objective was to evaluate pre-treatment FDG PET/CT-derived machine learning (ML) models for predicting outcome in patients with cHL. METHODS: All cHL patients undergoing pre-treatment PET/CT at our institution between 2008 and 2018 were retrospectively identified. A 1.5 × mean liver standardised uptake value (SUV) and a fixed 4.0 SUV threshold were used to segment PET/CT data. Feature extraction was performed using PyRadiomics with ComBat harmonisation. Training (80%) and test (20%) cohorts stratified around 2-year event-free survival (EFS), age, sex, ethnicity and disease stage were defined. Seven ML models were trained and hyperparameters tuned using stratified 5-fold cross-validation. Area under the curve (AUC) from receiver operator characteristic analysis was used to assess performance. RESULTS: A total of 289 patients (153 males), median age 36 (range 16-88 years), were included. There was no significant difference between training (n = 231) and test cohorts (n = 58) (p value > 0.05). A ridge regression model using a 1.5 × mean liver SUV segmentation had the highest performance, with mean training, validation and test AUCs of 0.82 ± 0.002, 0.79 ± 0.01 and 0.81 ± 0.12. However, there was no significant difference between a logistic model derived from metabolic tumour volume and clinical features or the highest performing radiomic model. CONCLUSIONS: Outcome prediction using pre-treatment FDG PET/CT-derived ML models is feasible in cHL patients. Further work is needed to determine optimum predictive thresholds for clinical use. KEY POINTS: • A fixed threshold segmentation method led to more robust radiomic features. • A radiomic-based model for predicting 2-year event-free survival in classical Hodgkin lymphoma patients is feasible. • A predictive model based on ridge regression was the best performing model on our dataset.

Developing an understanding of artificial intelligence lung nodule risk prediction using insights from the Brock model.

OBJECTIVES: To determine if predictions of the Lung Cancer Prediction convolutional neural network (LCP-CNN) artificial intelligence (AI) model are analogous to the Brock model. METHODS: In total, 10,485 lung nodules in 4660 participants from the National Lung Screening Trial (NLST) were analysed. Both manual and automated nodule measurements were inputted into the Brock model, and this was compared to LCP-CNN. The performance of an experimental AI model was tested after ablating imaging features in a manner analogous to removing predictors from the Brock model. First, the nodule was ablated leaving lung parenchyma only. Second, a sphere of the same size as the nodule was implanted in the parenchyma. Third, internal texture of both nodule and parenchyma was ablated. RESULTS: Automated axial diameter (AUC 0.883) and automated equivalent spherical diameter (AUC 0.896) significantly improved the accuracy of Brock when compared to manual measurement (AUC 0.873), although not to the level of the LCP-CNN (AUC 0.936). Ablating nodule and parenchyma texture (AUC 0.915) led to a small drop in AI predictive accuracy, as did implanting a sphere of the same size as the nodule (AUC 0.889). Ablating the nodule leaving parenchyma only led to a large drop in AI performance (AUC 0.717). CONCLUSIONS: Feature ablation is a feasible technique for understanding AI model predictions. Nodule size and morphology play the largest role in AI prediction, with nodule internal texture and background parenchyma playing a limited role. This is broadly analogous to the relative importance of morphological factors over clinical factors within the Brock model. KEY POINTS: • Brock lung cancer risk prediction accuracy was significantly improved using automated axial or equivalent spherical measurements of lung nodule diameter, when compared to manual measurements. • Predictive accuracy was further improved by using the Lung Cancer Prediction convolutional neural network, an artificial intelligence-based model which obviates the requirement for nodule measurement. • Nodule size and morphology are important factors in artificial intelligence lung cancer risk prediction, with nodule texture and background parenchyma contributing a small, but measurable, role.

Lobar Ventilation in Patients with COPD Assessed with the Full-Scale Airway Network Flow Model and Xenon-enhanced Dual-Energy CT.

Background The full-scale airway network (FAN) flow model shows excellent agreement with limited functional imaging data but requires further validation prior to clinical use. Purpose To validate the ventilation distributions computed with the FAN flow model with xenon ventilation from xenon-enhanced dual-energy (DE) CT in participants with chronic obstructive pulmonary disease (COPD). Materials and Methods In this prospective study, the FAN model extracted structural data from xenon-enhanced DE CT images of men with COPD scanned between June 2012 and July 2013 to compute gas ventilation dynamics. The ventilation distributions on the middle cross-section plane, percentage lobar ventilation, and ventilation heterogeneity quantified by the coefficient of variation (CV) were compared between xenon-enhanced DE CT imaging and the FAN model. The relationship between the ventilation parameters with the densitometry and pulmonary function test results was demonstrated. The agreements and correlations between the parameters were measured using the concordance correlation coefficient and the Pearson correlation coefficient. Results Twenty-two men with COPD (mean age, 67 years ± 7 [standard deviation]) were evaluated. The percentage lobar ventilation computed with FAN showed a strong positive correlation with xenon-enhanced DE CT data (r = 0.7, P < .001). Ninety-five percent of lobar ventilation CV differences lay within 95% confidence intervals. Correlations of the percentage lobar ventilation were negative for percentage emphysema (xenon-enhanced DE CT: r = -0.38, P < .001; FAN: r = -0.23, P = .02) but were positive for percentage normal tissue volume (xenon-enhanced DE CT: r = 0.78, P < .001; FAN: r = 0.45, P < .001). Lung CVs of FAN revealed negative correlations with the spirometry results (CVFAN vs percentage predicted forced expiratory volume in 1 second: r = -0.75, P < .001; CVFAN vs ratio of forced expiratory volume in 1 second to forced vital capacity: r = -0.67, P < .001). Conclusion The full-scale airway network modeled lobar ventilation in patients with chronic obstructive pulmonary disease correlated with the xenon-enhanced dual-energy CT imaging data. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Parraga and Eddy in this issue.

Hyperpolarized 129Xe MRI Abnormalities in Dyspneic Patients 3 Months after COVID-19 Pneumonia: Preliminary Results.

Background SARS-CoV-2 targets angiotensin-converting enzyme 2-expressing cells in the respiratory tract. There are reports of breathlessness in patients many months after infection. Purpose To determine whether hyperpolarized xenon 129 MRI (XeMRI) imaging could be used to identify the possible cause of breathlessness in patients at 3 months after hospital discharge following COVID-19 infection. Materials and Methods This prospective study was undertaken between August and December of 2020, with patients and healthy control volunteers being enrolled. All patients underwent lung function tests; ventilation and dissolved-phase XeMRI, with the mean red blood cell (RBC) to tissue or plasma (TP) ratio being calculated; and a low-dose chest CT, with scans being scored for the degree of abnormalities after COVID-19. Healthy control volunteers underwent XeMRI. The intraclass correlation coefficient was calculated for volunteer and patient scans to assess repeatability. A Wilcoxon rank sum test and Cohen effect size calculation were performed to assess differences in the RBC/TP ratio between patients and control volunteers. Results Nine patients (mean age, 57 years ± 7 [standard deviation]; six male patients) and five volunteers (mean age, 29 years ± 3; five female volunteers) were enrolled. The mean time from hospital discharge for patients was 169 days (range, 116-254 days). There was a difference in the RBC/TP ratio between patients and control volunteers (0.3 ± 0.1 vs 0.5 ± 0.1, respectively; P = .001; effect size, 1.36). There was significant difference between the RBC and gas phase spectral full width at half maximum between volunteers and patients (median ± range, 567 ± 1 vs 507 ± 81 [P = .002] and 104 ± 2 vs 122 ± 17 [P = .004], respectively). Results were reproducible, with intraclass correlation coefficients of 0.82 and 0.88 being demonstrated for patients and volunteers, respectively. Participants had normal or nearly normal CT scans (mean, seven of 25; range, zero of 25 to 10 of 25). Conclusion Hyperpolarized xenon 129 MRI results showed alveolar capillary diffusion limitation in all nine patients after COVID-19 pneumonia, despite normal or nearly normal results at CT. © RSNA, 2021 See also the editorial by Dietrich in this issue.

Radiomics in predicting treatment response in non-small-cell lung cancer: current status, challenges and future perspectives.

OBJECTIVES: Radiomics is the extraction of quantitative data from medical imaging, which has the potential to characterise tumour phenotype. The radiomics approach has the capacity to construct predictive models for treatment response, essential for the pursuit of personalised medicine. In this literature review, we summarise the current status and evaluate the scientific and reporting quality of radiomics research in the prediction of treatment response in non-small-cell lung cancer (NSCLC). METHODS: A comprehensive literature search was conducted using the PubMed database. A total of 178 articles were screened for eligibility and 14 peer-reviewed articles were included. The radiomics quality score (RQS), a radiomics-specific quality metric emulating the TRIPOD guidelines, was used to assess scientific and reporting quality. RESULTS: Included studies reported several predictive markers including first-, second- and high-order features, such as kurtosis, grey-level uniformity and wavelet HLL mean respectively, as well as PET-based metabolic parameters. Quality assessment demonstrated a low median score of + 2.5 (range - 5 to + 9), mainly reflecting a lack of reproducibility and clinical evaluation. There was extensive heterogeneity between studies due to differences in patient population, cancer stage, treatment modality, follow-up timescales and radiomics workflow methodology. CONCLUSIONS: Radiomics research has not yet been translated into clinical use. Efforts towards standardisation and collaboration are needed to identify reproducible radiomic predictors of response. Promising radiomic models must be externally validated and their impact evaluated within the clinical pathway before they can be implemented as a clinical decision-making tool to facilitate personalised treatment for patients with NSCLC. KEY POINTS: • The included studies reported several promising radiomic markers of treatment response in lung cancer; however, there was a lack of reproducibility between studies. • Quality assessment using the radiomics quality score (RQS) demonstrated a low median total score of + 2.5 (range - 5 to + 9). • Future radiomics research should focus on implementation of standardised radiomics features and software, together with external validation in a prospective setting.

Dual-energy CT in the diagnosis of occult acute scaphoid injury: a direct comparison with MRI.

OBJECTIVES: Scaphoid injuries occult on plain radiography often require further imaging for definitive diagnosis. We investigate the utility of dual-energy computed tomography (DECT) for the detection of acute bone marrow oedema and fracture of scaphoid compared to MRI. MATERIALS AND METHODS: Twenty patients who presented acutely (without prior injury) to the emergency department with clinically suspected occult scaphoid fracture and had MRI of the wrist were prospectively recruited to have DECT (GE Revolution CT). Material decomposition images of the water-calcium base pair were generated and assessed in conjunction with the monochromatic images to permit correlation of marrow signal changes with any cortical disruption for fracture confirmation. The assessment was performed by two musculoskeletal radiologists blinded from MRI results. The statistical difference of MRI and reviewers' detection of acute bone oedema (1 = present, 0 = absent) was performed using the Friedman test (SPSS v.16). RESULTS: MRI showed acute scaphoid fracture and/or bone marrow oedema in 14/20 patients of which 6 also had cortical disruption. On DECT, reviewer A identified oedema in 13 and cortical disruption in 10 patients while reviewer B identified oedema in 10 and cortical disruption in seven of the 14 MRI positive patients. No statistically significant difference in oedema detection on MRI and reviewers of DECT (p value 0.61) but DECT was more sensitive at detecting cortical disruption. CONCLUSION: DECT has the capability to detect acute scaphoid oedema in addition to cortical fractures. However, compared to MRI, DECT has lower contrast resolution and less sensitive in the detection of mild oedema. KEY POINTS: • Dual-energy CT is able to detect acute traumatic scaphoid marrow oedema. • Dual-energy CT has greater detection rate of scaphoid fractures than MRI. • Dual-energy CT is an alternative to MRI for occult scaphoid injury.

Evaluation of a novel deep learning-based classifier for perifissural nodules.

OBJECTIVES: To evaluate the performance of a novel convolutional neural network (CNN) for the classification of typical perifissural nodules (PFN). METHODS: Chest CT data from two centers in the UK and The Netherlands (1668 unique nodules, 1260 individuals) were collected. Pulmonary nodules were classified into subtypes, including "typical PFNs" on-site, and were reviewed by a central clinician. The dataset was divided into a training/cross-validation set of 1557 nodules (1103 individuals) and a test set of 196 nodules (158 individuals). For the test set, three radiologically trained readers classified the nodules into three nodule categories: typical PFN, atypical PFN, and non-PFN. The consensus of the three readers was used as reference to evaluate the performance of the PFN-CNN. Typical PFNs were considered as positive results, and atypical PFNs and non-PFNs were grouped as negative results. PFN-CNN performance was evaluated using the ROC curve, confusion matrix, and Cohen's kappa. RESULTS: Internal validation yielded a mean AUC of 91.9% (95% CI 90.6-92.9) with 78.7% sensitivity and 90.4% specificity. For the test set, the reader consensus rated 45/196 (23%) of nodules as typical PFN. The classifier-reader agreement (k = 0.62-0.75) was similar to the inter-reader agreement (k = 0.64-0.79). Area under the ROC curve was 95.8% (95% CI 93.3-98.4), with a sensitivity of 95.6% (95% CI 84.9-99.5), and specificity of 88.1% (95% CI 81.8-92.8). CONCLUSION: The PFN-CNN showed excellent performance in classifying typical PFNs. Its agreement with radiologically trained readers is within the range of inter-reader agreement. Thus, the CNN-based system has potential in clinical and screening settings to rule out perifissural nodules and increase reader efficiency. KEY POINTS: • Agreement between the PFN-CNN and radiologically trained readers is within the range of inter-reader agreement. • The CNN model for the classification of typical PFNs achieved an AUC of 95.8% (95% CI 93.3-98.4) with 95.6% (95% CI 84.9-99.5) sensitivity and 88.1% (95% CI 81.8-92.8) specificity compared to the consensus of three readers.

Assessing the Accuracy of a Deep Learning Method to Risk Stratify Indeterminate Pulmonary Nodules.

Rationale: The management of indeterminate pulmonary nodules (IPNs) remains challenging, resulting in invasive procedures and delays in diagnosis and treatment. Strategies to decrease the rate of unnecessary invasive procedures and optimize surveillance regimens are needed.Objectives: To develop and validate a deep learning method to improve the management of IPNs.Methods: A Lung Cancer Prediction Convolutional Neural Network model was trained using computed tomography images of IPNs from the National Lung Screening Trial, internally validated, and externally tested on cohorts from two academic institutions.Measurements and Main Results: The areas under the receiver operating characteristic curve in the external validation cohorts were 83.5% (95% confidence interval [CI], 75.4-90.7%) and 91.9% (95% CI, 88.7-94.7%), compared with 78.1% (95% CI, 68.7-86.4%) and 81.9 (95% CI, 76.1-87.1%), respectively, for a commonly used clinical risk model for incidental nodules. Using 5% and 65% malignancy thresholds defining low- and high-risk categories, the overall net reclassifications in the validation cohorts for cancers and benign nodules compared with the Mayo model were 0.34 (Vanderbilt) and 0.30 (Oxford) as a rule-in test, and 0.33 (Vanderbilt) and 0.58 (Oxford) as a rule-out test. Compared with traditional risk prediction models, the Lung Cancer Prediction Convolutional Neural Network was associated with improved accuracy in predicting the likelihood of disease at each threshold of management and in our external validation cohorts.Conclusions: This study demonstrates that this deep learning algorithm can correctly reclassify IPNs into low- or high-risk categories in more than a third of cancers and benign nodules when compared with conventional risk models, potentially reducing the number of unnecessary invasive procedures and delays in diagnosis.

External validation of a convolutional neural network artificial intelligence tool to predict malignancy in pulmonary nodules.

BACKGROUND: Estimation of the risk of malignancy in pulmonary nodules detected by CT is central in clinical management. The use of artificial intelligence (AI) offers an opportunity to improve risk prediction. Here we compare the performance of an AI algorithm, the lung cancer prediction convolutional neural network (LCP-CNN), with that of the Brock University model, recommended in UK guidelines. METHODS: A dataset of incidentally detected pulmonary nodules measuring 5-15 mm was collected retrospectively from three UK hospitals for use in a validation study. Ground truth diagnosis for each nodule was based on histology (required for any cancer), resolution, stability or (for pulmonary lymph nodes only) expert opinion. There were 1397 nodules in 1187 patients, of which 234 nodules in 229 (19.3%) patients were cancer. Model discrimination and performance statistics at predefined score thresholds were compared between the Brock model and the LCP-CNN. RESULTS: The area under the curve for LCP-CNN was 89.6% (95% CI 87.6 to 91.5), compared with 86.8% (95% CI 84.3 to 89.1) for the Brock model (p≤0.005). Using the LCP-CNN, we found that 24.5% of nodules scored below the lowest cancer nodule score, compared with 10.9% using the Brock score. Using the predefined thresholds, we found that the LCP-CNN gave one false negative (0.4% of cancers), whereas the Brock model gave six (2.5%), while specificity statistics were similar between the two models. CONCLUSION: The LCP-CNN score has better discrimination and allows a larger proportion of benign nodules to be identified without missing cancers than the Brock model. This has the potential to substantially reduce the proportion of surveillance CT scans required and thus save significant resources.

Intercostal vessel screening prior to pleural interventions by the respiratory physician: a prospective study of real world practice.

INTRODUCTION: The rising incidence of pleural disease is seeing an international growth of pleural services, with physicians performing an ever-increasing volume of pleural interventions. These are frequently conducted at sites without immediate access to thoracic surgery or interventional radiology and serious complications such as pleural bleeding are likely to be under-reported. AIM: To assess whether intercostal vessel screening can be performed by respiratory physicians at the time of pleural intervention, as an additional step that could potentially enhance safe practice. METHODS: This was a prospective, observational study of 596 ultrasound-guided pleural procedures conducted by respiratory physicians and trainees in a tertiary centre. Operators did not have additional formal radiology training. Intercostal vessel screening was performed using a low frequency probe and the colour Doppler feature. RESULTS: The intercostal vessels were screened in 95% of procedures and the intercostal artery (ICA) was successfully identified in 53% of cases. Screening resulted in an overall site alteration rate of 16% in all procedures, which increased to 30% when the ICA was successfully identified. This resulted in procedure abandonment in 2% of cases due to absence of a suitable entry site. Intercostal vessel screening was shown to be of particular value in the context of image-guided pleural biopsy. CONCLUSION: Intercostal vessel screening is a simple and potentially important additional step that can be performed by respiratory physicians at the time of pleural intervention without advanced ultrasound expertise. Whether the widespread use of this technique can improve safety requires further evaluation in a multi-centre setting with a robust prospective study.

Delayed ventilation assessment using fast dynamic hyperpolarised Xenon-129 magnetic resonance imaging.

OBJECTIVES: To investigate the use of a fast dynamic hyperpolarised 129Xe ventilation magnetic resonance imaging (DXeV-MRI) method for detecting and quantifying delayed ventilation in patients with chronic obstructive pulmonary disease (COPD). METHODS: Three male participants (age range 31-43) with healthy lungs and 15 patients (M/F = 12:3, age range = 48-73) with COPD (stages II-IV) underwent spirometry tests, quantitative chest computed tomography (QCT), and DXeV-MRI at 1.5-Tesla. Regional delayed ventilation was captured by measuring the temporal signal change in each lung region of interest (ROI) in comparison to that in the trachea. In addition to its qualitative assessment through visual inspection by a clinical radiologist, delayed ventilation was quantitatively captured by calculating a covariance measurement of the lung ROI and trachea signals, and quantified using both the time delay, and the difference between the integrated areas covered by the signal-time curves of the two signals. RESULTS: Regional temporal ventilation, consistent with the expected physiological changes across a free breathing cycle, was demonstrated with DXeV-MRI in all patients. Delayed ventilation was observed in 13 of the 15 COPD patients and involved variable lung ROIs. This was in contrast to the control group, where no delayed ventilation was demonstrated (p = 0.0173). CONCLUSIONS: DXeV-MRI offers a non-invasive way of detecting and quantifying delayed ventilation in patients with COPD, and provides physiological information on regional pulmonary function during a full breathing cycle. KEY POINTS: • Dynamic xenon MRI allows for the non-invasive detection and measurement of delayed ventilation in COPD patients. • Dynamic xenon MRI during a free breathing cycle can provide unique information about pulmonary physiology and pulmonary disease pathophysiology. • With further validation, dynamic xenon MRI could offer a non-invasive way of measuring collateral ventilation which can then be used to guide lung volume reduction therapy (LVRT) for certain COPD patients.