Explore the vast range of titles available.

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.

This document from the European Society of Radiology (ESR) and the European Society of Thoracic Imaging (ESTI) aims to present the main imaging features, and the role of CT scan in the early diagnosis of COVID-19, describing, in particular, the typical findings which make it possible to identify the disease and distinguish it from bacterial causes of infection, and to define which category of patients may benefit from CT imaging. The precautions that must be taken when performing scans to protect radiologists and technologists from infection will be described. The organisational measures that can be taken within radiology departments in order to cope with the influx of patients, while continuing to manage other emergency and time-sensitive activity (e.g. oncology, other infectious diseases etc.), will be discussed.Key points• Bilateral ground glass opacities are typical CT manifestations of COVID-19.• Crazy paving and organising pneumonia pattern are seen at a later stage.• Extensive consolidation is associated with a poor prognosis.

BackgroundEstimation 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.MethodsA 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.ResultsThe 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.ConclusionThe 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.

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.

Coronavirus disease 2019 (COVID-19) may severely impair pulmonary function and cause hypoxia. However, the association of COVID-19 pneumonia on CT with impaired ventilation remains unexplained. This pilot study aims to demonstrate the relationship between the radiological findings on COVID-19 CT images and ventilation abnormalities simulated in a computational model linked to the patients' symptoms. Twenty-five patients with COVID-19 and four test-negative healthy controls who underwent a baseline non-enhanced CT scan: 7 dyspneic patients, 9 symptomatic patients without dyspnea, and 9 asymptomatic patients were included. A 2D U-Net-based CT segmentation software was used to quantify radiological futures of COVID-19 pneumonia. The CT image-based full-scale airway network (FAN) flow model was employed to assess regional lung ventilation. Functional and radiological features were compared across groups and correlated with the clinical symptoms. Heterogeneity in ventilation distribution and ventilation defects associated with the pneumonia and the patients' symptoms were assessed. Median percentage ventilation defects were 0.2% for healthy controls, 0.7% for asymptomatic patients, 1.2% for symptomatic patients without dyspnea, and 11.3% for dyspneic patients. The median of percentage pneumonia was 13.2% for dyspneic patients and 0% for the other groups. Ventilation defects preferentially affected the posterior lung and worsened with increasing pneumonia linearly (y = 0.91x + 0.99, R2 = 0.73) except for one of the nine dyspneic patients who had disproportionally large ventilation defects (7.8% of the entire lung) despite mild pneumonia (1.2%). The symptomatic and dyspneic patients showed significantly right-skewed ventilation distributions (symptomatic without dyspnea: 0.86 ± 0.61, dyspnea 0.91 ± 0.79) compared to the patients without symptom (0.45 ± 0.35). The ventilation defect analysis with the FAN model provided a comparable diagnostic accuracy to the percentage pneumonia in identifying dyspneic patients (area under the receiver operating characteristic curve, 0.94 versus 0.96). COVID-19 pneumonia segmentations from CT scans are accompanied by impaired pulmonary ventilation preferentially in dyspneic patients. Ventilation analysis with CT image-based computational modelling shows it is able to assess functional impairment in COVID-19 and potentially identify one of the aetiologies of hypoxia in patients with COVID-19 pneumonia.

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 .

Previous preclinical research has shown that extracorporeal devices can be used to enhance the delivery and distribution of systemically administered anticancer drugs, resulting in increased intratumoural concentrations. We aimed to assess the safety and feasibility of targeted release and enhanced delivery of doxorubicin to solid tumours from thermosensitive liposomes triggered by mild hyperthermia, induced non-invasively by focused ultrasound. We did an open-label, single-centre, phase 1 trial in a single UK hospital. Adult patients (aged ≥18 years) with unresectable and non-ablatable primary or secondary liver tumours of any histological subtype were considered for the study. Patients received a single intravenous infusion (50 mg/m2) of lyso-thermosensitive liposomal doxorubicin (LTLD), followed by extracorporeal focused ultrasound exposure of a single target liver tumour. The trial had two parts: in part I, patients had a real-time thermometry device implanted intratumourally, whereas patients in part II proceeded without thermometry and we used a patient-specific model to predict optimal exposure parameters. We assessed tumour biopsies obtained before and after focused ultrasound exposure for doxorubicin concentration and distribution. The primary endpoint was at least a doubling of total intratumoural doxorubicin concentration in at least half of the patients treated, on an intention-to-treat basis. This study is registered with ClinicalTrials.gov, number NCT02181075, and is now closed to recruitment. Between March 13, 2015, and March 27, 2017, ten patients were enrolled in the study (six patients in part I and four in part II), and received a dose of LTLD followed by focused ultrasound exposure. The treatment resulted in an average increase of 3·7 times in intratumoural biopsy doxorubicin concentrations, from an estimate of 2·34 μg/g (SD 0·93) immediately after drug infusion to 8·56 μg/g (5·69) after focused ultrasound. Increases of two to ten times were observed in seven (70%) of ten patients, satisfying the primary endpoint. Serious adverse events registered were expected grade 4 transient neutropenia in five patients and prolonged hospital stay due to unexpected grade 1 confusion in one patient. Grade 3–4 adverse events recorded were neutropenia (grade 3 in one patient and grade 4 in five patients), and grade 3 anaemia in one patient. No treatment-related deaths occurred. The combined treatment of LTLD and non-invasive focused ultrasound hyperthermia in this study seemed to be clinically feasible, safe, and able to enhance intratumoural drug delivery, providing targeted chemo-ablative response in human liver tumours that were refractory to standard chemotherapy. Oxford Biomedical Research Centre, National Institute for Health Research.

A 70-year-old male with sarcomatoid renal carcinoma presented to his general practitioner with worsening breathlessness. He was referred to the radiology department for a radiograph of the chest, which showed recurrence of a known right malignant pleural effusion (MPE) (figure 1A). After 4 days, he underwent an uncomplicated right indwelling pleural catheter (IPC) insertion and drainage. The postprocedure radiograph of the chest (figure 1B) showed improved appearances of the right MPE and a new small left pleural effusion that had been seen on the preprocedure ultrasound and likely progressed since his presenting radiograph. The radiographer’s report raised concerns that the right IPC tip was positioned too far to the left of the midline. A radiology opinion, similar to the impression of the clinical team, suggested that a rotated film may explain the appearance. Subsequently, a CT scan demonstrated that the IPC did cross the midline (figure 1C), but did not clarify whether the left pleural space had been truly entered. The patient was invited to attend clinical review.

Objectives Investigate the effect of a novel Bayesian penalised likelihood (BPL) reconstruction algorithm on analysis of pulmonary nodules examined with 18F-FDG PET/CT, and to determine its effect on small, sub-10-mm nodules. Methods 18F-FDG PET/CTs performed for nodule evaluation in 104 patients (121 nodules) were retrospectively reconstructed using the new algorithm, and compared to time-of-flight ordered subset expectation maximisation (OSEM) reconstruction. Nodule and background parameters were analysed semi-quantitatively and visually. Results BPL compared to OSEM resulted in statistically significant increases in nodule SUV max (mean 5.3 to 8.1, p  < 0.00001), signal-to-background (mean 3.6 to 5.3, p  < 0.00001) and signal-to-noise (mean 24 to 41, p  < 0.00001). Mean percentage increase in SUV max (%ΔSUV max ) was significantly higher in nodules ≤10 mm ( n  = 31, mean 73 %) compared to >10 mm ( n  = 90, mean 42 %) ( p  = 0.025). Increase in signal-to-noise was higher in nodules ≤10 mm (224 %, mean 12 to 27) compared to >10 mm (165 %, mean 28 to 46). When applying optimum SUV max thresholds for detecting malignancy, the sensitivity and accuracy increased using BPL, with the greatest improvements in nodules ≤10 mm. Conclusion BPL results in a significant increase in signal-to-background and signal-to-noise compared to OSEM. When semi-quantitative analyses to diagnose malignancy are applied, higher SUV max thresholds may be warranted owing to the SUV max increase compared to OSEM. Key Points • Novel Bayesian penalised likelihood PET reconstruction was applied for lung nodule evaluation. • This was compared to current standard of care OSEM reconstruction. • The novel reconstruction generated significant increases in lung nodule signal-to-background and signal-to-noise. • These increases were highest in small, sub-10-mm pulmonary nodules. • Higher SUV max thresholds may be warranted when using semi-quantitative analyses to diagnose malignancy.

PurposeTo derive lobar ventilation in patients with chronic obstructive pulmonary disease (COPD) using a rapid time-series hyperpolarized xenon-129 (HPX) magnetic resonance imaging (MRI) technique and compare this to ventilation/perfusion single-photon emission computed tomography (V/Q-SPECT), correlating the results with high-resolution computed tomography (CT) and pulmonary function tests (PFTs).Materials and methodsTwelve COPD subjects (GOLD stages I–IV) participated in this study and underwent HPX-MRI, V/Q-SPECT/CT, high-resolution CT, and PFTs. HPX-MRI was performed using a novel time-series spiral k-space sampling approach. Relative percentage ventilations were calculated for individual lobe for comparison to the relative SPECT lobar ventilation and perfusion. The absolute HPX-MRI percentage ventilation in each lobe was compared to the absolute CT percentage emphysema score calculated using a signal threshold method. Pearson’s correlation and linear regression tests were performed to compare each imaging modality.ResultsStrong correlations were found between the relative lobar percentage ventilation with HPX-MRI and percentage ventilation SPECT (r = 0.644; p < 0.001) and percentage perfusion SPECT (r = 0.767; p < 0.001). The absolute CT percentage emphysema and HPX percentage ventilation correlation was also statistically significant (r = 0.695, p < 0.001). The whole lung HPX percentage ventilation correlated with the PFT measurements (FEV1 with r = − 0.886, p < 0.001*, and FEV1/FVC with r = − 0.861, p < 0.001*) better than the whole lung CT percentage emphysema score (FEV1 with r = − 0.635, p = 0.027; and FEV1/FVC with r = − 0.652, p = 0.021).ConclusionLobar ventilation with HPX-MRI showed a strong correlation with lobar ventilation and perfusion measurements derived from SPECT/CT, and is better than the emphysema score obtained with high-resolution CT.Key Points• The ventilation hyperpolarized xenon-129 MRI correlates well with ventilation and perfusion with SPECT/CT with the advantage of higher temporal and spatial resolution.• The hyperpolarized xenon-129 MRI correlates with the PFT measurements better than the high-resolution CT with the advantage of avoiding the use of ionizing radiation.