Validated machine learning tools to distinguish immune checkpoint inhibitor, radiotherapy, COVID-19 and other infective pneumonitis.

BACKGROUND: Pneumonitis is a well-described, potentially disabling, or fatal adverse effect associated with both immune checkpoint inhibitors (ICI) and thoracic radiotherapy. Accurate differentiation between checkpoint inhibitor pneumonitis (CIP) radiation pneumonitis (RP), and infective pneumonitis (IP) is crucial for swift, appropriate, and tailored management to achieve optimal patient outcomes. However, correct diagnosis is often challenging, owing to overlapping clinical presentations and radiological patterns. METHODS: In this multi-centre study of 455 patients, we used machine learning with radiomic features extracted from chest CT imaging to develop and validate five models to distinguish CIP and RP from COVID-19, non-COVID-19 infective pneumonitis, and each other. Model performance was compared to that of two radiologists. RESULTS: Models to distinguish RP from COVID-19, CIP from COVID-19 and CIP from non-COVID-19 IP out-performed radiologists (test set AUCs of 0.92 vs 0.8 and 0.8; 0.68 vs 0.43 and 0.4; 0.71 vs 0.55 and 0.63 respectively). Models to distinguish RP from non-COVID-19 IP and CIP from RP were not superior to radiologists but demonstrated modest performance, with test set AUCs of 0.81 and 0.8 respectively. The CIP vs RP model performed less well on patients with prior exposure to both ICI and radiotherapy (AUC 0.54), though the radiologists also had difficulty distinguishing this test cohort (AUC values 0.6 and 0.6). CONCLUSION: Our results demonstrate the potential utility of such tools as a second or concurrent reader to support oncologists, radiologists, and chest physicians in cases of diagnostic uncertainty. Further research is required for patients with exposure to both ICI and thoracic radiotherapy.

Market making and the production of nurses for export: a case study of India-UK health worker migration.

BACKGROUND: High-income countries increasingly look to the international recruitment of health workers to address domestic shortages, especially from low-income and middle-income countries. We adapt conceptual frameworks from migration studies to examine the networked and commercialised nature of the Indian market for nurse migration to the UK. METHODS: We draw on data from 27 expert interviews conducted with migration intermediaries, healthcare providers and policymakers in India and the UK. FINDINGS: India-UK nurse migration occurs within a complex and evolving market encompassing ways to educate, train and recruit nursing candidates. For-profit actors shape the international orientation of nursing curricula, broker on-the-job training and offer language, exam and specialised clinical training. Rather than merely facilitate travel, these brokers produce both generic, emigratory nurses as well as more customised nurses ready to meet specific shortages in the UK. DISCUSSION: The dialectic of producing emigratory and customised nurses is similar to that seen in the Post-Fordist manufacturing model characterised by flexible specialisation and a networked structure. As the commodity in this case are people attempting to improve their position in life, these markets require attention from health policy makers. Nurse production regimes based on international market opportunities are liable to change, subjecting nurses to the risk of having trained for a market that can no longer accommodate them. The commercial nature of activities further entrenches existing socioeconomic inequalities in the Indian nurse force. Negative repercussions for the source healthcare system can be anticipated as highly qualified, specialised nurses leave to work in healthcare systems abroad.

'Not in it for huge profits but because it's right': The contested moral economies of UK-India exports in health worker education and training.

While the growth of global markets in health-related services may have significant consequences for healthcare provisioning and training, it has received relatively little attention from the social sciences. This article examines UK-India, and specifically England-India, exports in health worker education and training as one such global market, drawing on sociological scholarship on moral economies to understand how trading in this field is constructed and legitimated by the individuals and organisations involved, what tensions evolve, and what is at stake in them. We employ a qualitative mixed methods approach using publicly available materials on existing UK-India collaborations and primary data from interviews with key stakeholders in India and the UK, including government departments, arms-length bodies, NHS Trusts, trade associations and private providers. Our analysis illustrates the key discursive strategies used to legitimate engagement in these markets, and the complex and contested moral economies unfolding between and across these stakeholders and contexts. Not least, we demonstrate the conflicting moral sentiments and the boundary work required to realise commodification. Situating cross-border trade in health worker education and training in a moral economy framework thus illuminates the social context and moral worlds in which this evolving trade is embedded.

Weak Relationships Between Psychological Factors and Experimental Pain Outcomes in Pain-Free Individuals: An Aggregate Analysis of 8 Studies.

Although psychological factors such as anxiety, depression, and pain catastrophizing are known to influence pain outcomes in chronic pain populations, there are mixed results regarding whether they influence experimental pain outcomes in pain-free individuals. The objectives of this study were to determine the associations between psychological factors and experimental pain outcomes in pain-free adolescents and adults. Relationships between anxiety, depression, and pain catastrophizing and experimental pain outcomes across 8 different studies (total N = 595) were examined in different populations of pain-free adult and adolescent participants. Analyses were conducted with and without controlling for sex, age, and race. Studies were analyzed separately and as part of an aggregate analysis. Individual study analyses resulted in 136 regression models. Of these, only 8 models revealed a significant association between psychological factors and pain outcomes. The significant results were small and likely due to Type 1 error. Controlling for demographic factors had minimal effect on the results. The aggregate analyses revealed weak relationships between anxiety and pressure pain threshold (Fisher's z = -.10 [-.19, -.01]), anxiety and cold pain intensity ratings (Fisher's z = .18 [.04, .32]), and pain catastrophizing and pressure pain threshold (Fisher's z = -.14 [-.26, -.02]). Sample size calculations based on the aggregate analyses indicated that several hundred participants would be required to detect true relationships between these psychological factors and pain measures. The overall negative findings suggest that in pain-free individuals, anxiety, depression, and pain catastrophizing are not meaningfully related to experimental pain outcomes. PERSPECTIVE: Psychological variables have been shown to predict pain outcomes in chronic pain populations but these relationships may not generalize to pain-free populations. An analysis of 595 pain-free individuals across 8 studies in our lab revealed that anxiety, depression, and pain catastrophizing were not meaningfully related to experimental pain outcomes.

Radiomics-based decision support tool assists radiologists in small lung nodule classification and improves lung cancer early diagnosis.

BACKGROUND: Methods to improve stratification of small (≤15 mm) lung nodules are needed. We aimed to develop a radiomics model to assist lung cancer diagnosis. METHODS: Patients were retrospectively identified using health records from January 2007 to December 2018. The external test set was obtained from the national LIBRA study and a prospective Lung Cancer Screening programme. Radiomics features were extracted from multi-region CT segmentations using TexLab2.0. LASSO regression generated the 5-feature small nodule radiomics-predictive-vector (SN-RPV). K-means clustering was used to split patients into risk groups according to SN-RPV. Model performance was compared to 6 thoracic radiologists. SN-RPV and radiologist risk groups were combined to generate "Safety-Net" and "Early Diagnosis" decision-support tools. RESULTS: In total, 810 patients with 990 nodules were included. The AUC for malignancy prediction was 0.85 (95% CI: 0.82-0.87), 0.78 (95% CI: 0.70-0.85) and 0.78 (95% CI: 0.59-0.92) for the training, test and external test datasets, respectively. The test set accuracy was 73% (95% CI: 65-81%) and resulted in 66.67% improvements in potentially missed [8/12] or delayed [6/9] cancers, compared to the radiologist with performance closest to the mean of six readers. CONCLUSIONS: SN-RPV may provide net-benefit in terms of earlier cancer diagnosis.

Non-contrast CT synthesis using patch-based cycle-consistent generative adversarial network (Cycle-GAN) for radiomics and deep learning in the era of COVID-19.

Handcrafted and deep learning (DL) radiomics are popular techniques used to develop computed tomography (CT) imaging-based artificial intelligence models for COVID-19 research. However, contrast heterogeneity from real-world datasets may impair model performance. Contrast-homogenous datasets present a potential solution. We developed a 3D patch-based cycle-consistent generative adversarial network (cycle-GAN) to synthesize non-contrast images from contrast CTs, as a data homogenization tool. We used a multi-centre dataset of 2078 scans from 1,650 patients with COVID-19. Few studies have previously evaluated GAN-generated images with handcrafted radiomics, DL and human assessment tasks. We evaluated the performance of our cycle-GAN with these three approaches. In a modified Turing-test, human experts identified synthetic vs acquired images, with a false positive rate of 67% and Fleiss' Kappa 0.06, attesting to the photorealism of the synthetic images. However, on testing performance of machine learning classifiers with radiomic features, performance decreased with use of synthetic images. Marked percentage difference was noted in feature values between pre- and post-GAN non-contrast images. With DL classification, deterioration in performance was observed with synthetic images. Our results show that whilst GANs can produce images sufficient to pass human assessment, caution is advised before GAN-synthesized images are used in medical imaging applications.

A highly sensitive LC-MS/MS method for quantitative determination of 7 vitamin D metabolites in mouse brain tissue.

Despite its critical role in neurodevelopment and brain function, vitamin D (vit-D) homeostasis, metabolism, and kinetics within the central nervous system remain largely undetermined. Thus, it is of critical importance to establish an accurate, highly sensitive, and reproducible method to quantitate vit-D in brain tissue. Here, we present a novel liquid chromatography tandem mass spectrometry (LC-MS/MS) method and for the first time, demonstrate detection of seven major vit-D metabolites in brain tissues of C57BL/6J wild-type mice, namely 1,25(OH)2D3, 3-epi-1,25(OH)2D3, 1,25(OH)2D2, 25(OH)D3, 25(OH)D2, 24,25(OH)2D3, and 24,25(OH)2D2. Chromatographic separation was achieved on a pentaflurophenyl column with 3 mM ammonium formate water/methanol [A] and 3 mM ammonium formate methanol/isopropanol [B] mobile phase components. Detection was by positive ion electrospray tandem mass spectrometry with the EVOQ elite triple quadrupole mass spectrometer with an Advance ultra-high-performance liquid chromatograph and online extraction system. Calibration standards of each metabolite prepared in brain matrices were used to validate the detection range, precision, accuracy, and recovery. Isotopically labelled analogues, 1,25(OH)2D3-d3, 25(OH)D3-c5, and 24,25(OH)2D3-d6, served as the internal standards for the closest molecular-related metabolite in all measurements. Standards between 1 fg/mL and 10 ng/mL were injected with a resulting linear range between 0.001 and 1 ng, with an LLOD and LLOQ of 1 pg/mL and 12.5 pg/mL, respectively. The intra-/inter-day precision and accuracy for measuring brain vit-D metabolites ranged between 0.12-11.53% and 0.28-9.11%, respectively. Recovery in acetonitrile ranged between 99.09 and 106.92% for all metabolites. Collectively, the sensitivity and efficiency of our method supersedes previously reported protocols used to measure vit-D and to our knowledge, the first protocol to reveal the abundance of 25(OH)D2, 1,25(OH)D2, and 24,25(OH)2D2, in brain tissue of any species. This technique may be important in supporting the future advancement of pre-clinical research into the function of vit-D in neurophysiological and neuropsychiatric disorders, and neurodegeneration.

Camera-based automated monitoring of flying insects in the wild (Camfi). II. flight behaviour and long-term population monitoring of migratory Bogong moths in Alpine Australia.

Introduction: The Bogong moth Agrotis infusa is well known for its remarkable annual round-trip migration from its breeding grounds across eastern and southern Australia to its aestivation sites in the Australian Alps, to which it provides an important annual influx of nutrients. Over recent years, we have benefited from a growing understanding of the navigational abilities of the Bogong moth. Meanwhile, the population of Bogong moths has been shrinking. Recently, the ecologically and culturally important Bogong moth was listed as endangered by the IUCN Red List, and the establishment of a program for long-term monitoring of its population has been identified as critical for its conservation. Methods: Here, we present the results of two years of monitoring of the Bogong moth population in the Australian Alps using recently developed methods for automated wildlife-camera monitoring of flying insects, named Camfi. While in the Alps, some moths emerge from the caves in the evening to undertake seemingly random flights, filling the air with densities in the dozens per cubic metre. The purpose of these flights is unknown, but they may serve an important role in Bogong moth navigation. Results: We found that these evening flights occur throughout summer and are modulated by daily weather factors. We present a simple heuristic model of the arrival to and departure from aestivation sites by Bogong moths, and confirm results obtained from fox-scat surveys which found that aestivating Bogong moths occupy higher elevations as the summer progresses. Moreover, by placing cameras along two elevational transects below the summit of Mt. Kosciuszko, we found that evening flights were not random, but were systematically oriented in directions relative to the azimuth of the summit of the mountain. Finally, we present the first recorded observations of the impact of bushfire smoke on aestivating Bogong moths - a dramatic reduction in the size of a cluster of aestivating Bogong moths during the fire, and evidence of a large departure from the fire-affected area the day after the fire. Discussion: Our results highlight the challenges of monitoring Bogong moths in the wild and support the continued use of automated camera-based methods for that purpose.

Constructing healthcare services markets: networks, brokers and the China-England engagement.

BACKGROUND: Healthcare services is an expanding international market with which national healthcare systems engage, and from which they benefit, to greater and lesser degrees. This study examines the case of the China-England engagement in healthcare services as a vehicle for illuminating the way in which such market relationships are constructed. FINDINGS: China and England have different approaches to the international healthcare services market. Aware of the knowledge and technology gaps between itself and the leading capitalist nations of the West in healthcare, as in other sectors, the Chinese leadership has encouraged a variety of international engagements to facilitate the bridging of these gaps including accessing new supply and demand relationships in international markets. These engagements are situated within an approach to health system development based on establishing broad policy directions, allowing a degree of local innovation, initiating and evaluating pilot studies, and promulgating new programmatic frameworks at central and local levels. The assumption is that the new knowledge and technologies are integrated into this approach and implemented under the guidance of Chinese experts and leaders. England's healthcare system has the knowledge resources to provide the supply to meet at least some of the China demand but has yet to develop fully the means to enable an efficient market response, though such economic engagement is supported by the UK's trade related departments of state. As a result, the development of China-England commercial relationships in patient care, professional education and hospital and healthcare service development has been led largely by high status NHS Trusts and private sector organisations with the entrepreneurial capacity to exploit their market position. Drawing on their established international clinicians and commercial teams with experience of domestic private sector provision, these institutions have built trust-based collaborations sufficiently robust to facilitate demand-supply relationships in the international healthcare services market. Often key to the development of relations required to make commercial exchange feasible and practicable are a range of international brokers with the skills and capacity to provide the necessary linkage with individual healthcare consumers and institutional clients in China. Integral to the broker role, and often supplied by the broker itself, are the communication technologies of telemedicine to enable the interaction between consumer and healthcare provider, be this in patient care, professional education or healthcare service development. CONCLUSIONS: Although England's healthcare system has the knowledge required to respond to China's market demand and such economic engagement is supported and actively encouraged by the UK's trade related departments of state, the response is constrained by multiple domestic demands on its resources and by the limits of the NHS approach to marketisation in healthcare.

A radiomics-based decision support tool improves lung cancer diagnosis in combination with the Herder score in large lung nodules.

BACKGROUND: Large lung nodules (≥15 mm) have the highest risk of malignancy, and may exhibit important differences in phenotypic or clinical characteristics to their smaller counterparts. Existing risk models do not stratify large nodules well. We aimed to develop and validate an integrated segmentation and classification pipeline, incorporating deep-learning and traditional radiomics, to classify large lung nodules according to cancer risk. METHODS: 502 patients from five U.K. centres were recruited to the large-nodule arm of the retrospective LIBRA study between July 2020 and April 2022. 838 CT scans were used for model development, split into training and test sets (70% and 30% respectively). An nnUNet model was trained to automate lung nodule segmentation. A radiomics signature was developed to classify nodules according to malignancy risk. Performance of the radiomics model, termed the large-nodule radiomics predictive vector (LN-RPV), was compared to three radiologists and the Brock and Herder scores. FINDINGS: 499 patients had technically evaluable scans (mean age 69 ± 11, 257 men, 242 women). In the test set of 252 scans, the nnUNet achieved a DICE score of 0.86, and the LN-RPV achieved an AUC of 0.83 (95% CI 0.77-0.88) for malignancy classification. Performance was higher than the median radiologist (AUC 0.75 [95% CI 0.70-0.81], DeLong p = 0.03). LN-RPV was robust to auto-segmentation (ICC 0.94). For baseline solid nodules in the test set (117 patients), LN-RPV had an AUC of 0.87 (95% CI 0.80-0.93) compared to 0.67 (95% CI 0.55-0.76, DeLong p = 0.002) for the Brock score and 0.83 (95% CI 0.75-0.90, DeLong p = 0.4) for the Herder score. In the international external test set (n = 151), LN-RPV maintained an AUC of 0.75 (95% CI 0.63-0.85). 18 out of 22 (82%) malignant nodules in the Herder 10-70% category in the test set were identified as high risk by the decision-support tool, and may have been referred for earlier intervention. INTERPRETATION: The model accurately segments and classifies large lung nodules, and may improve upon existing clinical models. FUNDING: This project represents independent research funded by: 1) Royal Marsden Partners Cancer Alliance, 2) the Royal Marsden Cancer Charity, 3) the National Institute for Health Research (NIHR) Biomedical Research Centre at the Royal Marsden NHS Foundation Trust and The Institute of Cancer Research, London, 4) the National Institute for Health Research (NIHR) Biomedical Research Centre at Imperial College London, 5) Cancer Research UK (C309/A31316).

Gross tumour volume radiomics for prognostication of recurrence & death following radical radiotherapy for NSCLC.

Recurrence occurs in up to 36% of patients treated with curative-intent radiotherapy for NSCLC. Identifying patients at higher risk of recurrence for more intensive surveillance may facilitate the earlier introduction of the next line of treatment. We aimed to use radiotherapy planning CT scans to develop radiomic classification models that predict overall survival (OS), recurrence-free survival (RFS) and recurrence two years post-treatment for risk-stratification. A retrospective multi-centre study of >900 patients receiving curative-intent radiotherapy for stage I-III NSCLC was undertaken. Models using radiomic and/or clinical features were developed, compared with 10-fold cross-validation and an external test set, and benchmarked against TNM-stage. Respective validation and test set AUCs (with 95% confidence intervals) for the radiomic-only models were: (1) OS: 0.712 (0.592-0.832) and 0.685 (0.585-0.784), (2) RFS: 0.825 (0.733-0.916) and 0.750 (0.665-0.835), (3) Recurrence: 0.678 (0.554-0.801) and 0.673 (0.577-0.77). For the combined models: (1) OS: 0.702 (0.583-0.822) and 0.683 (0.586-0.78), (2) RFS: 0.805 (0.707-0.903) and 0·755 (0.672-0.838), (3) Recurrence: 0·637 (0.51-0.·765) and 0·738 (0.649-0.826). Kaplan-Meier analyses demonstrate OS and RFS difference of >300 and >400 days respectively between low and high-risk groups. We have developed validated and externally tested radiomic-based prediction models. Such models could be integrated into the routine radiotherapy workflow, thus informing a personalised surveillance strategy at the point of treatment. Our work lays the foundations for future prospective clinical trials for quantitative personalised risk-stratification for surveillance following curative-intent radiotherapy for NSCLC.

The Role of Artificial Intelligence in Early Cancer Diagnosis.

Improving the proportion of patients diagnosed with early-stage cancer is a key priority of the World Health Organisation. In many tumour groups, screening programmes have led to improvements in survival, but patient selection and risk stratification are key challenges. In addition, there are concerns about limited diagnostic workforces, particularly in light of the COVID-19 pandemic, placing a strain on pathology and radiology services. In this review, we discuss how artificial intelligence algorithms could assist clinicians in (1) screening asymptomatic patients at risk of cancer, (2) investigating and triaging symptomatic patients, and (3) more effectively diagnosing cancer recurrence. We provide an overview of the main artificial intelligence approaches, including historical models such as logistic regression, as well as deep learning and neural networks, and highlight their early diagnosis applications. Many data types are suitable for computational analysis, including electronic healthcare records, diagnostic images, pathology slides and peripheral blood, and we provide examples of how these data can be utilised to diagnose cancer. We also discuss the potential clinical implications for artificial intelligence algorithms, including an overview of models currently used in clinical practice. Finally, we discuss the potential limitations and pitfalls, including ethical concerns, resource demands, data security and reporting standards.

A comparison of machine learning methods for predicting recurrence and death after curative-intent radiotherapy for non-small cell lung cancer: Development and validation of multivariable clinical prediction models.

BACKGROUND: Surveillance is universally recommended for non-small cell lung cancer (NSCLC) patients treated with curative-intent radiotherapy. High-quality evidence to inform optimal surveillance strategies is lacking. Machine learning demonstrates promise in accurate outcome prediction for a variety of health conditions. The purpose of this study was to utilise readily available patient, tumour, and treatment data to develop, validate and externally test machine learning models for predicting recurrence, recurrence-free survival (RFS) and overall survival (OS) at 2 years from treatment. METHODS: A retrospective, multicentre study of patients receiving curative-intent radiotherapy for NSCLC was undertaken. A total of 657 patients from 5 hospitals were eligible for inclusion. Data pre-processing derived 34 features for predictive modelling. Combinations of 8 feature reduction methods and 10 machine learning classification algorithms were compared, producing risk-stratification models for predicting recurrence, RFS and OS. Models were compared with 10-fold cross validation and an external test set and benchmarked against TNM-stage and performance status. Youden Index was derived from validation set ROC curves to distinguish high and low risk groups and Kaplan-Meier analyses performed. FINDINGS: Median follow-up time was 852 days. Parameters were well matched across training-validation and external test sets: Mean age was 73 and 71 respectively, and recurrence, RFS and OS rates at 2 years were 43% vs 34%, 54% vs 47% and 54% vs 47% respectively. The respective validation and test set AUCs were as follows: 1) RFS: 0·682 (0·575-0·788) and 0·681 (0·597-0·766), 2) Recurrence: 0·687 (0·582-0·793) and 0·722 (0·635-0·81), and 3) OS: 0·759 (0·663-0·855) and 0·717 (0·634-0·8). Our models were superior to TNM stage and performance status in predicting recurrence and OS. INTERPRETATION: This robust and ready to use machine learning method, validated and externally tested, sets the stage for future clinical trials entailing quantitative personalised risk-stratification and surveillance following curative-intent radiotherapy for NSCLC. FUNDING: A full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

Models of cardiovascular surgery biobanking to facilitate translational research and precision medicine.

Biobanking in health care has evolved over the last few decades from simple biological sample repositories to complex and dynamic units with multi-organizational infrastructure networks and has become an essential tool for modern medical research. Cardiovascular tissue biobanking provides a unique opportunity to utilize cardiac and vascular samples for translational research into heart failure and other related pathologies. Current techniques for diagnosis, classification, and treatment monitoring of cardiac disease relies primarily on interpretation of clinical signs, imaging, and blood biomarkers. Further research at the disease source (i.e. myocardium and blood vessels) has been limited by a relative lack of access to quality human cardiac tissue and the inherent shortcomings of most animal models of heart disease. In this review, we describe a model for cardiovascular tissue biobanking and databasing, and its potential to facilitate basic and translational research. We share techniques to procure endocardial samples from patients with hypertrophic cardiomyopathy, heart failure with reduced ejection fraction, and heart failure with preserved ejection fraction, in addition to aortic disease samples. We discuss some of the issues with respect to data collection, privacy, biobank consent, and the governance of tissue biobanking. The development of tissue biobanks as described here has significant scope to improve and facilitate translational research in multi-omic fields such as genomics, transcriptomics, proteomics, and metabolomics. This research heralds an era of precision medicine, in which patients with cardiovascular pathology can be provided with optimized and personalized medical care for the treatment of their individual phenotype.

Decentred regulation: The case of private healthcare in India.

In order to progress towards more equitable social welfare systems we need an improved understanding of regulation in social sectors such as health and education. However, research to date has tended to focus on roles for governments and professions, overlooking the broader range of regulatory systems that emerge in contexts of market-based provisioning and partial state regulation. In this article we examine the regulation of private healthcare in India using an analytical approach informed by 'decentred' and 'regulatory capitalism' perspectives. We apply these ideas to qualitative data on private healthcare and its regulation in Maharashtra (review of press media, semi-structured interviews with 43 respondents, and three witness seminars), in order to describe the range of state and non-state actors involved in setting rules and norms in this context, whose interests are represented by these activities, and what problems arise. We show an eclectic set of regulatory systems in operation. Government and statutory councils do perform limited and sporadic regulatory roles, typically organised around legislation, licensing and inspections, and often prompted by the judicial arm of the state. But a range of industry-level actors, private organisations and public insurers are involved too, promoting their own interests in the sector via the offices of regulatory capitalism: accreditation companies, insurers, platform operators and consumer courts. Rules and norms are extensive but diffuse. These are produced not just through laws, licensing and professional codes of conduct, but also through industry influence over standards, practices and market organisation, and through individualised attempts to negotiate exceptions and redressal. Our findings demonstrate regulation in a marketised social sector to be partial, disjointed and decentred to multiple loci, actively representing differing interests. Greater understanding of the different actors and processes at play in such contexts can inform future progress towards universal systems for social welfare.

Development of a Structured Query Language and Natural Language Processing Algorithm to Identify Lung Nodules in a Cancer Centre.

Importance: The stratification of indeterminate lung nodules is a growing problem, but the burden of lung nodules on healthcare services is not well-described. Manual service evaluation and research cohort curation can be time-consuming and potentially improved by automation. Objective: To automate lung nodule identification in a tertiary cancer centre. Methods: This retrospective cohort study used Electronic Healthcare Records to identify CT reports generated between 31st October 2011 and 24th July 2020. A structured query language/natural language processing tool was developed to classify reports according to lung nodule status. Performance was externally validated. Sentences were used to train machine-learning classifiers to predict concerning nodule features in 2,000 patients. Results: 14,586 patients with lung nodules were identified. The cancer types most commonly associated with lung nodules were lung (39%), neuro-endocrine (38%), skin (35%), colorectal (33%) and sarcoma (33%). Lung nodule patients had a greater proportion of metastatic diagnoses (45 vs. 23%, p < 0.001), a higher mean post-baseline scan number (6.56 vs. 1.93, p < 0.001), and a shorter mean scan interval (4.1 vs. 5.9 months, p < 0.001) than those without nodules. Inter-observer agreement for sentence classification was 0.94 internally and 0.98 externally. Sensitivity and specificity for nodule identification were 93 and 99% internally, and 100 and 100% at external validation, respectively. A linear-support vector machine model predicted concerning sentence features with 94% accuracy. Conclusion: We have developed and validated an accurate tool for automated lung nodule identification that is valuable for service evaluation and research data acquisition.

Immune Checkpoint Inhibitor and Radiotherapy-Related Pneumonitis: An Informatics Approach to Determine Real-World Incidence, Severity, Management, and Resource Implications.

Pneumonitis is a well-described, potentially life-threatening adverse effect of immune checkpoint inhibitors (ICI) and thoracic radiotherapy. It can require additional investigations, treatment, and interruption of cancer therapy. It is important for clinicians to have an awareness of its incidence and severity, however real-world data are lacking and do not always correlate with findings from clinical trials. Similarly, there is a dearth of information on cost impact of symptomatic pneumonitis. Informatics approaches are increasingly being applied to healthcare data for their ability to identify specific patient cohorts efficiently, at scale. We developed a Structured Query Language (SQL)-based informatics algorithm which we applied to CT report text to identify cases of ICI and radiotherapy pneumonitis between 1/1/2015 and 31/12/2020. Further data on severity, investigations, medical management were also acquired from the electronic health record. We identified 248 cases of pneumonitis attributable to ICI and/or radiotherapy, of which 139 were symptomatic with CTCAE severity grade 2 or more. The grade ≥2 ICI pneumonitis incidence in our cohort is 5.43%, greater than the all-grade 1.3-2.7% incidence reported in the literature. Time to onset of ICI pneumonitis was also longer in our cohort (mean 4.5 months, range 4 days-21 months), compared to the median 2.7 months (range 9 days-19.2 months) described in the literature. The estimated average healthcare cost of symptomatic pneumonitis is £3932.33 per patient. In this study we use an informatics approach to present new real-world data on the incidence, severity, management, and resource burden of ICI and radiotherapy pneumonitis. To our knowledge, this is the first study to look at real-world incidence and healthcare resource utilisation at the per-patient level in a UK cancer hospital. Improved management of pneumonitis may facilitate prompt continuation of cancer therapy, and improved outcomes for this not insubstantial cohort of patients.

Optimizing the discovery and assessment of therapeutic targets in heart failure with preserved ejection fraction.

There is an urgent need for models that faithfully replicate heart failure with preserved ejection fraction (HFpEF), now recognized as the most common form of heart failure in the world. In vitro approaches have several shortcomings, most notably the immature nature of stem cell-derived human cardiomyocytes [induced pluripotent stem cells (iPSC)] and the relatively short lifespan of primary cardiomyocytes. Three-dimensional 'organoids' incorporating mature iPSCs with other cell types such as endothelial cells and fibroblasts are a significant advance, but lack the complexity of true myocardium. Animal models can replicate many features of human HFpEF, and rodent models are the most common, and recent attempts to incorporate haemodynamic, metabolic, and ageing contributions are encouraging. Differences relating to species, physiology, heart rate, and heart size are major limitations for rodent models. Porcine models mitigate many of these shortcomings and approximate human physiology more closely, but cost and time considerations limit their potential for widespread use. Ex vivo analysis of failing hearts from animal models offer intriguing possibilities regarding cardiac substrate utilisation, but are ultimately subject to the same constrains as the animal models from which the hearts are obtained. Ex vivo approaches using human myocardial biopsies can uncover new insights into pathobiology leveraging myocardial energetics, substrate turnover, molecular changes, and systolic/diastolic function. In collaboration with a skilled cardiothoracic surgeon, left ventricular endomyocardial biopsies can be obtained at the time of valvular surgery in HFpEF patients. Critically, these tissues maintain their disease phenotype, preserving inter-relationship of myocardial cells and extracellular matrix. This review highlights a novel approach, where ultra-thin myocardial tissue slices from human HFpEF hearts can be used to assess changes in myocardial structure and function. We discuss current approaches to modelling HFpEF, describe in detail the novel tissue slice model, expand on exciting opportunities this model provides, and outline ways to improve this model further.