159370 research outputs found
Sort by
Holocene palaeoenvironmental reconstruction of sea level, coastal and vegetation changes along the southern Solway Firth, United Kingdom
Get PDFWhen means and standard deviations are an incomplete summary of a continuous variable: problems, solutions, and utilising the reference ranges to check normality
Get PDFThe mean and standard deviation are commonly used to summarise continuous data, often without regard for the distribution of the data. For data in a non- normal distribution, the median and interquartile interval are often more appropriate. This article outlines circumstances where the mean and standard deviation alone are insufficient summary variables, and provides a simple appraisal method for reviewers without the raw data
Disciplining sanitation: Interrogating disciplinary narratives of inequalities in access to sanitation
Get PDFPhysical long-term conditions and psychological treatment outcomes in NHS talking therapies: prevalence, impact, and moderators
No full textObjective
This study examined the impact of having one or multiple physical long-term conditions (LTCs) on treatment outcomes and engagement with NHS Talking Therapies (NHS-TT), focusing on clinical and treatment-related moderators.
Methods
A retrospective cohort design was applied to routine data from 44,957 patients discharged from seven NHS-TT services in Northwest England between September 2021 and September 2024. Logistic regression analyses assessed the impact of LTC status (none, one or multiple LTCs) on reliable recovery—defined as clinically significant improvement in both depression and anxiety scores—and treatment engagement, operationalised as planned discharge versus dropout. Moderation analyses explored the influence of provisional diagnosis, baseline functional impairment, treatment intensity and mode of therapy delivery.
Results
LTCs were reported by 47.5% of patients, with 9.4% reporting multiple LTCs. Patients with LTCs had lower odds of reliable recovery (OR = 0.83), with further reductions among those with multimorbidity (OR = 0.71). Conversely, LTC status was associated with higher odds of planned discharge (OR = 1.28). The negative association with reliable recovery was attenuated among patients with depression (OR = 1.21) but exacerbated for those with mixed anxiety and depression (OR = 0.71), who also had lower odds of planned discharge (OR = 0.66).
Conclusion
Multimorbidity is associated with poorer clinical outcomes despite higher treatment engagement. Tailored care pathways are needed to better support patients with LTCs
UK librarians’ views of chatbots: a study based on fictional scenarios
Get PDFThe potential of Artificial Intelligence (AI) in libraries, including use of chatbots, has been widely discussed in recent years. This paper investigates how library professionals see the role of chatbots in the library context. As a data collection method, the study used the arts-based approach of fictional scenarios, which are short, crafted stories representing possible futures. Participants (UK librarians) responded to researcher-authored fictional scenarios and wrote their own. In response to the fictional scenarios authored by the researchers, participants were negative about giving chatbots agency or human-like characteristics. Resource issues and ethics also appeared in their responses as barriers. Participants were more positive about time saving applications and data analysis. In the fictional scenarios participants wrote themselves they tended to imagine rather narrow, low-level functions for AI. The theory of professions is drawn on to interpret the resistance of professionals to more capable AI and the assertion of human agency
Emerging hotspots of agricultural drought under climate change
Get PDFClimate change is intensifying drought risk, yet it is unclear which regions will be most vulnerable in the future. Here we investigate emerging hotspots of agricultural drought across the tropics and Northern Hemisphere extratropics using climate reanalysis and model simulations under a range of Shared Socioeconomic Pathways. Our analysis accounts for soil moisture at the onset of the growing season, as well as variability during the season itself, linking climate change to the land-surface water balance by classifying the dominant controls on evapotranspiration, including a newly defined state governed by plant extraction of water from the root zone. We show that much of Europe, southern Africa, northern South America and western North America are emerging hotspots of agricultural drought, with mechanisms of observed drying consistent with future projections. Drought trends are identified even where precipitation projections diverge. By focusing on growing seasons, our approach captures hotspots overlooked by annual metrics and shows that increasing drought frequency is compounded by shifts towards more severe and intense events. These findings have strong implications for food security and highlight the need for drought-resilient adaptation not only in the global south but also in extratropical regions where risk is already escalating
HD 44892: the youngest (or oldest?) gas-harbouring debris disc around an intermediate-mass star
Get PDF
Explainable AI-driven quality and condition monitoring in smart manufacturing
Get PDFArtificial intelligence (AI) is increasingly adopted in manufacturing for tasks such as automated inspection, predictive maintenance, and condition monitoring. However, the opaque, black-box nature of many AI models remains a major barrier to industrial trust, acceptance, and regulatory compliance. This study investigates how explainable artificial intelligence (XAI) techniques can be used to systematically open and interpret the internal reasoning of AI systems commonly deployed in manufacturing, rather than to optimise or compare model performance. A unified explainability-centred framework is proposed and applied across three representative manufacturing use cases encompassing heterogeneous data modalities and learning paradigms: vision-based classification of casting defects, vision-based localisation of metal surface defects, and unsupervised acoustic anomaly detection for machine condition monitoring. Diverse models are intentionally employed as representative black-box decision-makers to evaluate whether XAI methods can provide consistent, physically meaningful explanations independent of model architecture, task formulation, or supervision strategy. A range of established XAI techniques, including Grad-CAM, Integrated Gradients, Saliency Maps, Occlusion Sensitivity, and SHAP, are applied to expose model attention, feature relevance, and decision drivers across visual and acoustic domains. The results demonstrate that XAI enables alignment between model behaviour and physically interpretable defect and fault mechanisms, supporting transparent, auditable, and human-interpretable decision-making. By positioning explainability as a core operational requirement rather than a post hoc visual aid, this work contributes a cross-modal framework for trustworthy AI in manufacturing, aligned with Industry 5.0 principles, human-in-the-loop oversight, and emerging expectations for transparent and accountable industrial AI systems
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! 👇