11237 research outputs found
Sort by
Automated lightweight model for asthma detection using respiratory and cough sound signals
No full textBackground and objective: Chronic respiratory diseases, such as asthma and COPD, pose significant challenges to human health and global healthcare systems. This pioneering study utilises AI analysis and modelling of cough and respiratory sound signals to classify and differentiate between asthma, COPD, and healthy subjects. The aim is to develop an AI-based diagnostic system capable of accurately distinguishing these conditions, thereby enhancing early detection and clinical management. Our study, therefore, presents the first AI system that leverages dual acoustic signals to enhance the diagnostic ACC of asthma using automated, lightweight deep learning models. Methods: To build an automated, lightweight model for asthma detection, tested separately with respiratory and cough sounds to assess their suitability for detecting asthma and COPD, the proposed AI models integrate the following ML algorithms: RF, SVM, DT, NN, and KNN, with an overall aim to demonstrate the efficacy of the proposed method for future clinical use. Model training and validation were performed using 5-fold cross-validation, wherein the dataset was randomly divided into five folds and the models were trained and tested iteratively to ensure robust performance. We evaluated the model outcomes with several performance metrics: ACC, precision, recall, F1 score, and area under the AUC. Additionally, a majority voting ensemble technique was employed to aggregate the predictions of the various classifiers for improved diagnostic reliability. We applied Gabor time–frequency transformation for feature extraction and NCA) for feature selection to optimise predictive accuracy. Independent comparative experiments were conducted, where cough-sound subsets were used to evaluate asthma detection capabilities, and respiratory-sound subsets were used to evaluate COPD detection capabilities, allowing for targeted model assessment. Results: The proposed ensemble approach, facilitated by a majority voting approach for model efficacy evaluation, achieved acceptable ACC values of 94.05% and 83.31% for differentiating between asthma and normal cases utilising separate respiratory sounds and cough sounds, respectively. The results highlight a substantial benefit in integrating multiple classifier models and sound modalities while demonstrating an unprecedented level of ACC and robustness for future diagnostic predictions of the disease. Conclusions: The present study sets a new benchmark in AI-based detection of respiratory diseases by integrating cough and respiratory sound signals for future diagnostics. The successful implementation of a dual-sound analysis approach promises advancements in the early detection and management of asthma and COPD.We conclude that the proposed model holds strong potential to transform asthma diagnostic practices and support clinicians in their respiratory healthcare practices.</p
Fen Blows: a short story collection and exegesis examining regionalism in 21st century short story collections
No full textMy PhD comprises a collection of original short stories, Fen Blows, and an exegesis in which I examine a range of twenty-first century short story collections, each with a commonality of setting. In identifying a common aesthetic in these contemporary collections, showing how themes of circumscription, marginalised, often youthful, characters and socioeconomic deprivation all predominate, I propose the term ‘millennium regionalism’.The fourteen stories that comprise Fen Blows are all set in the same Cambridgeshire village, close to the river Great Ouse. Influenced by my reading of millennium regionalist collections, they not only share the same aesthetic and many of the tropes I identify and discuss in the first two chapters of the exegesis, but also subvert a number of those tropes by featuring some characters who are not necessarily on the fringes of society, or in straitened economic circumstances, or have no hope of escape. My collection therefore offers a contemporary portrait of a south Cambridgeshire village to sit alongside other millennium regionalist collections that provide aesthetically similar, but nonetheless unique, portraits of places as diverse as Caerphilly, Western Australia, Michigan, Florida and Rio’s favelas.The exegesis opens by suggesting a growth in regionalist short story collections since 2000 and proposing the term ‘millennium regionalism’ to link, what I propose is, their shared aesthetic. I discuss the origins of regionalist literature and its nineteenth-century roots in American ‘local color’ writing, before focussing on, and examining, exemplar stories from a variety of contemporary collections from around the world, highlighting common themes and tropes. I go on to closely analyse three discrete regionalist story collections, selected to illustrate the global spread and variety of the form in the twenty-first century. I conclude by examining my own creative process, contextualising my stories and proposing Fen Blows as an original example of a millennium regionalist short story collection.</p
Delays in the Stroke Care Pathway in a Low-Income Setting: An Audit Study from Mozambique
No full textBackground: The burden of stroke is on the rise in low-income countries (LICs). Organized stroke care (OSC) is crucial for improving outcomes in LICs and is the very first step to reducing delays in diagnosis and treatment. We aim to evaluate delay times (DT) in accessing OSC at the national reference hospital of Mozambique, a LIC from southern Africa. Methods: An observational study based on consecutive case series of 59 stroke patients confirmed by computed tomography (CT) scans over a period of 3 months (May–July 2023). The total DT (from stroke onset to inward hospitalization) was the main outcome. Other specific DTs were analyzed including initial symptoms to arrival and admission (DT0), arrival to CT scans (DT1), arrival of laboratory results (DT2), and arrival to inward hospitalization (DT3). Results: The mean age was 61.9 (min 30–max 90) and 45.8% were female. The median total DT was 20 h. The median time DT0 was 10.6 h (interquartile range (IQR): 16.48). The median DT1 and DT2 were 4 h (IQR: 3.5) and 5 h (IQR: 2.6), respectively. The median DT3 was 10 h (IQR: 4). None of the patients were treated under a stroke code. Conclusions: This study reveals an unacceptable prehospital and in-hospital DT. Waiting for the CT scan contributed to a large proportion of the total DT, which among other factors can be explained by the absence of a stroke code and limited imaging capacity. These findings mirror disparities in stroke care seen in other LICs, where late presentation, scarce imaging, and limited specialized protocols are common. The urgent implementation of organized prehospital and in-hospital stroke pathways is needed in Maputo to improve outcomes.</p
Sedimentary blue carbon around decommissioned oil and gas platforms in the North Sea
No full textShelf sea habitats contain significant sedimentary carbon stocks, due to their large spatial extent, and act as key ecosystems for global carbon cycling. These shelf sediments are subject to a range of anthropogenic pressures including seabed trawling, climate change and the introduction of man-made structures (MMS), such as oil and gas platforms. In the North Sea, following their production lifespan, MMS are required to be mandatorily decommissioned under OSPAR Decision 98/3. Understanding the impact of decommissioned MMS on sedimentary carbon stores is a significant research gap, as these platforms are widespread throughout many shelf seas, with many now exceeding or approaching the end of their designed lifespan. In this study we quantify carbon stocks, organic and inorganic carbon content, and sediment characteristics from shelf sediment cores taken at increasing distances (50–3200 m) away from two decommissioned North Sea oil and gas platforms, Miller and North West Hutton. Carbon accumulation rates were also determined for sediments collected around North West Hutton. Organic carbon content at Miller was highest within 50 m (24.55 ± 2.49 mg/g of sediment) of the decommissioned platform and significantly lower across all other sampling distances. Conversely, organic carbon content at North West Hutton did not vary greatly with distance; however, a significant decrease was seen at 200 m. These findings highlight that carbon dynamics around decommissioned oil and gas platforms are site-specific. Moreover, the absence of pre-decommissioning data limits our ability or disentangle the effects of decommissioning on sedimentary carbon.</p
What is the relationship between individual and clusters of lifestyle behaviours during early adolescence on future youth mental health? A prospective cohort study
No full textAim: To analyze the gender stratified associations of individual and clustered unhealthy behaviors during early adolescence (i.e., 11 years) with internalizing and externalizing symptoms during mid- and late adolescence (i.e., 14 and 17 years).Methods: We used data from the Millennium Cohort Study including 9127 (4642 girls) participants at 14 years of age and 6970 (3611 girls) at 17 years of age. Internalizing and externalizing symptoms were self-reported by the main caregiver using the Strengths and Difficulties Questionnaire at 11 years, 14 years, and 17 years. Depressive symptoms were estimated using the Mood and Feelings Questionnaire at 14 years. Psychological distress was assessed using the six-item Kessler psychological distress scale at 17 years. Unhealthy behaviors (i.e., low sports practice, elevated TV-viewing, daily soft drink consumption, Results: Clustering three or more unhealthy behaviors at 11 years was associated with higher depressive symptoms (Boys: incidence rate ratio, IRR: 1.35; 95% CI: 1.17–1.56; Girls: 1.23; 1.10–1.38) and peer problems (boys: 1.36; 1.17–1.56; girls: 1.22; 1.08–1.38) at 14 years in both genders, as well as emotional problems among boys. Similarly, clustering three or more unhealthy behaviors was associated with higher peer problems (boys: 1.33; 1.16–1.51; girls: 1.24; 1.08–1.42) and psychological distress (boys:1.14; 1.04–1.25; girls: 1.10; 1.02–1.30) at 17 years among both genders as well as emotional problems among boys. Despite the high prevalence emotional problems among girls in both waves, there was a linear trend of increasing risk of the outcome with the increase in the number of unhealthy behaviors. Low sports practice and TV viewing were the unhealthy behaviors most consistently associated with later mental health-related outcomes.Conclusion: Clustering two or more unhealthy behaviors during early adolescence was associated with higher internalizing and depressive symptoms in mid-adolescence, as well as internalizing symptoms and psychological distress in late adolescence.</p
Towards more connection in drought and flood management in the transboundary Limpopo basin
No full textImproved drought and flood management in semi-arid transboundary basins requires a better understanding of the connections between dry and wet extremes, surface water and groundwater, upstream and downstream, and local communities and formal governance actors. This study describes a multi-disciplinary and mixed-methods research in the Limpopo River Basin, southern Africa. The methodology included hydrometeorological data analysis to identify drought and flood events, group discussions with 240 local community participants about drought and flood processes, impacts and preparedness, and interviews with 36 (inter)national and regional water managers and policymakers about drought and flood governance, early warning and communication. Additionally, we co-created drought and flood management scenarios through transboundary and national workshops and modelled these with an integrated surface water-groundwater model. We found that floods are crucial for aquifer recharge, providing baseflow during droughts, but also impactful for communities, who receive less training and support for floods than for droughts. Flood early warnings (if provided) are often not acted upon because of cultural values or limited resources. Drought and flood adaptation strategies were simulated to be effective, but factors like investment and maintenance costs, technical capacity and community uptake impact implementation. Furthermore, technical measures alone are inadequate to reduce community risk if underlying vulnerabilities are not addressed. Therefore, strengthening connections between communities and formal governance actors and better transboundary management of surface water and groundwater connections could yield significant benefits.Contribution: This study provides 11 distinct recommendations for managing drought and flood risk, focussing on the four connections analysed.</p
Between policy and practice: a qualitative case study exploring the enactment of healthy public policy in primary school settings
No full textContemporary public health practice for school-aged children in primary schools is a complex system with several different government, public sector, and private sector agencies, organisations, and individuals responsible for developing and enacting healthy public policy. The role and responsibilities of professionals such as school nurses and teachers have adapted and broadened, policy and practice have evolved, approaches and priorities have been re-evaluated, and multi-agency working is challenged. Additionally, policy documents do not clearly link the two sectors of health and education together practically to enable joined-up multi-agency working, creating a separation between policy and practice.Situated within an interpretivist paradigm, a qualitative case study methodology has been taken to explore the system of public health in primary schools in England, and the theoretical lens of policy enactment is used to interpret the perceptions of professionals in the disciplines of school nursing and teaching. Set across two stages of research, individual and group interviews have been used as the data collection method, and reflexive thematic analysis has been used for analysis.Despite having a prominent role in health and education policies, it was found that professionals often work in silos and are not effectively enacting multi-agency working in the way they would like, citing conflicting priorities, overlapping policies, ever-changing expectations, and overstretched services as contributing factors.The development of a conceptual and a theoretical framework in this thesis provides a nuanced understanding of how systems thinking, and policy enactment theory can be combined to understand the enactment practices of professionals in the spheres of health and education. Importantly, the frameworks serve as a tool and an opportunity to facilitate conversations between professionals in and across disciplinary groups and therefore provide important contributions to policy and practice.</p
Global cancer mortality among children, adolescents, and young adults from 77 countries, 1990–2021: a global time series analysis and modelling study
No full textBackground: The patterns of cancer burden in children, adolescents and young adults are distinct from those in the general adult population, underscoring the importance of developing targeted research and interventions. Thus, we analyzed global, regional, and national cancer mortality trends via the World Health Organization (WHO) Mortality Database among children aged 0–14 years and adolescents and young adults aged 15–34 across 77 WHO Member States.Methods: Age-standardized country-specific cancer mortality rates (deaths per 100,000) and years of life lost between 1990 and 2021 were estimated via a locally weighted scatterplot smooth curve. We analyzed the associations between socioeconomic indices and mortality rates for 21 types of cancer. Furthermore, we conducted a decomposition analysis to understand the factors influencing mortality rates for these 21 types of cancer.Results: Between 1990 and 2021, the age-standardized cancer mortality rate decreased from 12.35% [95% confidence interval (CI) 11.81%–12.88%] to 4.83% (95% CI 4.12%–5.54%) by − 3.29% per year. However, progress in reducing death rates has been heterogeneous in terms of country income, cancer site, and geography. In particular, the decrease in burden was most pronounced in North America; however, cervical and testicular cancer mortality is elevated in Africa, Latin America, and the Caribbean. While the major contributors to children’s cancer mortality (leukemia, brain, and bone tumors) have remained unchanged for the past three decades, a significant decrease in respiratory tract cancer and an increase in colorectal cancer have led to a transition in the cancer profile in adolescents and young adults. Additionally, infection-related cancers are inversely correlated with socioeconomic indices; notably, colorectal cancer appears to have no correlation with these indices. We also revealed significant changes in mortality trends during the COVID-19 pandemic, which were more pronounced in children. Finally, a decomposition analysis revealed that the decrease in the number of cancer deaths worldwide between 1990 and 2021 may be attributed primarily to age-specific mortality rather than population growth or aging.Conclusions: From 1990 to 2021, a significant decreasing trend in cancer mortality in the young population, especially in high-income countries, was observed. However, progress in reducing death rates has been heterogeneous by country income, cancer site, and geography, indicating disparities in control efforts across countries. Future studies are needed to address the exposures responsible for the heterogeneity of cancer burden and the changing cancer profile in this age group.</p
Association of unmet healthcare needs with depressive symptoms and COVID-19 pandemic-related factors among adults, 2009-2022: a nationwide study in South Korea
No full textNo description supplied</p
Exploring the relationship between artificial intelligence and service quality in the United Arab Emirates public sector
No full textThe focus of this thesis is the adoption of Artificial Intelligence (AI) technologies in the United Arab Emirates (UAE) public sector as a means to improve the quality and efficiency of public services. Using a qualitative methodology and the interpretivist research philosophy, the study draws on semi-structured interview data with 29 participants. This approach facilitated an investigation of experiences at both individual and organizational levels with AI.This exploratory work therefore addresses the adoption of AI in the United Arab UAE public sector with implications for service quality, service effectiveness, and ethical governance. In the context of the UAE's strategic aspiration to become a global leader in AI, the study addresses how AI solutions could improve service efficiency and effectiveness, whilst also introducing new ethical, technical and organizational challenges. By employing a qualitative approach, the study highlights the affordances of AI technology to automate activities, facilitate service customization, and enhance responsiveness.Thematic analysis led to three themes: 'Technological Materiality', 'Contextual Relationality', and 'Ethical Affordances', and to a conceptual framework. This framework focuses on balancing AI implementation in the public sector while ensuring human accountability, ethical practices, and inclusion in AI systems.The study also identifies obstacles like digital exclusion, algorithmic black boxes, cultural resistance, and immature ethical guarantees that call the putative ’inclusive’ promise of AI into question. This work contributes to the theory of AI implementation by extending the socio-technical lens used to examine it, and offers practical insights through recommendations on ethical AI governance, cross-departmental collaboration, and inclusive design. . Ultimately, the results make a case for an integrated, citizen-first model for AI adoption in public services — one that weighs the engine of tech innovation against the need for transparency, fairness and public trust. This study provides a useful framework for policy officials and public administrators to use AIs responsibly to offer efficient, accessible and ethically justified public services.</p
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇