University of the West of Scotland
Research Repository and Portal - University of the West of ScotlandNot a member yet
10867 research outputs found
Sort by
From challenge to consensus:pathways for inclusive growth in Africa
In this white paper, Ephias Ruhode and Kingsley O. Omeihe reflect on the conversations and debates that emerged at the inaugural Atlas Global Academic Conference in Victoria Falls in 2025. They show that Africa’s challenge is not a shortage of ideas or innovation, but the difficulty of turning knowledge into action across institutions, sectors, and communities. The paper highlights how closer collaboration, ethical governance and locally grounded innovation can help build a more inclusive and resilient future for the continent
Accelerated rotation-invariant convolution for UAV image segmentation
Rotation invariance is essential for precise object level segmentation in UAV aerial imagery, where targets can have arbitrary orientations and exhibit fine scale details. Conventional segmentation architectures like UNet rely on convolution operators that are not rotation-invariant, leading to degraded segmentation accuracy across varying viewpoints. Rotation invariance can be achieved by expanding the filter bank across multiple orientations; however, this significantly increases computational cost and memory requirement. In this paper, we introduce a GPU-optimized rotation-invariant convolution framework that eliminates the traditional data lowering (im2col) step required for matrix multiplication based convolution. By exploiting structured data sharing among symmetrically rotated filters, our method achieves multi-orientation convolution with greatly reduced memory requirements and computational redundancy. We further generalize the approach to accelerate convolution with arbitrary (non-symmetric) rotation angles. Integrated into a UNet segmentation model, the framework yields up to a 5.7% improvement in accuracy over the non-rotation-aware baseline. Across extensive benchmarks, the proposed convolution achieves 20–57% faster training and 15–45% lower energy consumption than cuDNN, while maintaining accuracy comparable to state of-the-art rotation-invariant methods. Because the scatter-based operator greatly reduces intermediate feature dimensionality, the efficiency of our design also enables practical sixteen-o rientation convolution and pooling, yielding further accuracy gains that areinfeasible for conventional rotation-invariant implementations. Our sixteen-orientation approach achieves competitive accuracy on multiple datasets compared with state-of-the-art UAV segmentation networks. These results demonstrate that the proposed method provides an effective and efficient alternative to existing rotation-invariant convolution frameworks
Health service contacts for mental health and substance use on release from prison:a retrospective population-based data linkage study
Background Mental health and substance use problems among people released from prison contribute substantially to premature mortality and emergency services demand. Understanding of mental health and substance use-related health service contacts prior to these severe and costly outcomes is limited. We assessed mental health and substance use-related contact with multiple services, comparing rates of contact among people released from prison to a matched general population sample who had not recently been in prison. Objectives To compare rates of health service contacts for mental health and substance use between people released from prison and a matched general population sample. Design We conducted a retrospective cohort study using linked administrative data with nationwide coverage. The cohort contained all people released from any Scottish prison in 2015 (exposed group), and a random general population sample matched (ratio 1:5) on sex, age, postcode and deprivation indices, who had no imprisonment in the 5 years prior (unexposed group). We linked individual-level administrative healthcare (prescriptions, outpatient, inpatient, emergency/ unscheduled care: 2010–2020), prison (admissions, releases: 2010–2020) and deaths records (2015–2020). We estimated adjusted incidence rate ratios (aIRRs) with 95% CIs using fixed-effects Poisson regression with cluster-robust standard errors, controlling for time-in-community, pre-index mental health and substance use-related health service contacts, and comorbidities. We stratified models by mental health (MH), substance use (SU) and dual diagnosis (attributable to both MH and SU). Setting Scotland. Results We linked records for 8313 people released from prison, and 41 213 matched individuals. Mental health and substance use-related contact rates were significantly higher for people released from prison across all services, and particularly for emergency and unscheduled care. aIRRs for ambulance contacts were MH=7.75 (95% CI 5.76 to 10.42), SU=7.58 (95% CI 5.71 to 10.08), dual diagnosis=8.28 (95% CI 6.50 to 10.55); and accident and emergency department contacts were MH=4.88 (95% CI 3.78 to 6.29) and SU=7.98 (95% CI 5.71 to 11.17). aIRRs for community prescriptions were MH=1.80 (95% CI 1.67 to 1.94), SU=5.95 (95% CI 4.83 to 7.32), dual diagnosis=5.33 (95% CI 3.70 to 7.68); drug and alcohol services were 7.13 (95% CI 6.00 to 8.48); and outpatient attendances were 2.61 (95% CI 2.17 to 3.16). aIRRs for 24-hour unscheduled telephone support were MH=7.63 (95% CI 4.93 to 11.83) and SU=8.29 (95% CI 3.99 to 17.22); and out-of-hours general practice were MH=5.14 (95% CI 3.66 to 7.22), SU=5.89 (95% CI 3.11 to 11.14) and dual diagnosis=8.85 (95% CI 2.94 to 26.63). aIRRs for general/acute hospital admissions and day cases were MH=2.97 (95% CI 1.43 to 6.16), SU=7.85 (95% CI 4.42 to 13.91), dual diagnosis=13.11 (95% CI 7.95 to 21.61); and for psychiatric admissions were MH=3.62 (95% CI 2.39 to 5.49), SU=10.74 (95% CI 6.12 to 18.84) and dual diagnosis=7.74 (95% CI 4.30 to 13.94). Conclusions Improved post-release mental health and substance use care is vital for individual and public health. Despite elevated rates of contact with community mental health and substance use services, people released from prison have disproportionately high rates of contact with emergency and unscheduled care services. This suggests that early support is either inadequate or not accessed by those in greatest need.Policymakers and service providers should consider investment in tailored transitional and post-release intervention at individual and population level, to improve health and thus prevent later high-cost service use and avoidable mortality. Our results also suggest high-quality care must be available and accessible beyond the immediate post-release period to permit sustained engagement or engagement at a later date
Factorial ANOMC tests:a comparative analysis of type I error and power under multifactor experimental designs
This study introduces a factorial extension of the Analysis of Means with Covariate (ANOMC), building on the classical ANOM framework by incorporating auxiliary covariate information. The proposed factorial ANOMC approach is designed for experiments involving two fixed factors and a continuous covariate, where traditional ANOM or factorial ANOM may lose efficiency. Six variants of the factorial ANOMC test are developed using regression and ratio-type estimators, and their performance is evaluated against the standard factorial ANOM test, which does not utilise covariate information. A comprehensive Monte Carlo simulation study is conducted to assess these tests under diverse conditions, including normal and non-normal error distributions, varying correlation structures, different sample sizes, multiple levels of each factor, and both homogeneous and heterogeneous variances. Performance is examined through empirical Type I error rates and statistical power. The findings show that while factorial ANOM maintains stable Type I error rates under ideal settings, several factorial ANOMC variants (i.e., and ) achieve improved detection capability, especially when regression estimators are used. Some ratio-based versions also perform well under specific correlation and distribution structures. Power increases noticeably for factorial ANOMC when sample sizes or factor-level combinations grow. Overall, the factorial ANOMC framework provides a more adaptable and informative alternative for multifactor experiments involving covariates
‘Build, baby, build’? A critical assessment of housing policy over the first year of the Labour Government in the UK
Housing policy has been a busy area of activity for the Labour government in its first year. In this paper we critically assess the tensions and contradictions within these housing policy changes, examining whether they add up to a coherent, programmatic response to the ‘housing crisis’ which can deliver for individuals and households struggling to access and sustain adequate housing after fourteen years of austerity and neglect. In particular, we question the underlying driver of the housebuilding target and ask whether the Labour government’s apparent desperation for economic growth is subsuming concerns for social justice, despite the increase in support for social housing – a debate with wide international resonance in the current economic context. Finally, we scrutinise whether the rapid start out of the blocks on housing policy can be maintained for the inevitable marathon that is necessary to make significant changes to the UK housing system
Responsibility with the other:plant ethics and education at the end of the world
This paper proposes an ethics that destabilizes the axis of anthropocentrism around which climate change education revolves. Education is future-oriented, narrated as aiming at ‘solving’ the climate crisis, or ‘saving’ the planet. I unsettle established notions of humans’ ethical responsibility for the world, considering instead our capacity to be responsible with non-human others, and in this case, plants. Plants are seen as being beyond ethical considerations; reduced to the level of resources, food sources, or ornaments, we do not feel the need to question or consider our ethical responsibilities to them. The paper proposes how we might begin this ethics, centred on the notion of being responsible as remaining open and attentive to the other, and answering – through meaningful action – in reply. The conversational responsiveness of this ethics does not assume mastery over plants, instead suggesting a new understanding of responsibility that presupposes the agency and awareness of plants.This paper proposes an ethics that destabilizes the axis of anthropocentrism around which climate change education revolves. Education is future-oriented, narrated as aiming at ‘solving’ the climate crisis, or ‘saving’ the planet. I unsettle established notions of humans’ ethical responsibility for the world, considering instead our capacity to be responsible with non-human others, and in this case, plants. Plants are seen as being beyond ethical considerations; reduced to the level of resources, food sources, or ornaments, we do not feel the need to question or consider our ethical responsibilities to them. The paper proposes how we might begin this ethics, centred on the notion of being responsible as remaining open and attentive to the other, and answering – through meaningful action – in reply. The conversational responsiveness of this ethics does not assume mastery over plants, instead suggesting a new understanding of responsibility that presupposes the agency and awareness of plants
Examining Black female narratives in Nollywood films:a feminist standpoint and subjective personal introspection approach to marketing representation
This paper advances Transformative Consumer Research (TCR) by analysing postcolonial media environments through an integration of Feminist Standpoint Theory (FST) and a standpoint-anchored form of Subjective Personal Introspection (SPI). Focussing on Nollywood, the Nigerian movie industry, the study conceptualises media as a symbolic marketplace that produces, circulates, and contests representations of Black Nigerian womanhood. Situating SPI within an FST epistemology provides a reflexive method linking embodied affect with the structural, intersectional hierarchies shaping meaning in Global South media contexts. Drawing on diasporic reflexivity, the study shows how Nollywood narratives function as marketing representational systems that shape consumer identity, emotional well-being, and cultural value. By positioning representation as a site of consumer justice, the paper extends TCR to postcolonial media environments and proposes an ethically oriented, decolonial framework for evaluating implications for marginalised consumers
Developing a psychological understanding of students’ perceptions of their school environment and the relationship with academic achievement
Educational research has demonstrated the importance of variables such as Socioeconomic Status (SES), Gender and Attendance in relation to academic achievement. In addition, research has also highlighted the importance of the physical learning environment in relation to academic achievement and in particular objective characteristics such as temperature, air quality and noise. The way in which students subjectively perceive their school environment has received less attention. However, one recent study has demonstrated how students’ subjective perceptions of their physical school environment, along with SES, Gender and Attendance, are all significantly related to academic achievement. The current study applies a new and distinct methodological approach to an existing data set to develop a psychological model of the physical environment to uncover latent factors which may be important in describing the relationship between the environment and academic performance. The study was conducted on data from 387, S5 students in five secondary schools in Scotland. Exploratory Factor Analysis was conducted to identify the underlying factor structure of the original 60 item questionnaire that was used to measure students’ perceptions of their school environment. Multiple regression analyses were then conducted to investigate the relationship between SES, Gender, Attendance and the new factors identified in the factor analysis, with academic achievement. The factor analysis identified a nine-factor model that reflects distinct dimensions of the physical and social aspects of the school and provides a comprehensive understanding of how students experience their surroundings. The multiple regression analyses confirmed that SES, Gender, and Attendance were significant predictors of academic achievement and that the inclusion of the nine factors explained an additional 11% of the variance in academic achievement. The implications of these findings are discussed in terms of how this psychological model of the physical school environment could be used to inform future educational design, policy, and interventions.<br/
A nutritional intervention within highly trained youth soccer players using a COM-B model of behaviour change:a case series approach
Large inter-individual variability in daily energy and carbohydrate intake has been reported in highly trained academy soccer players, potentially impacting training and match-day performance, recovery and physical development. Nutritional interventions in these cohorts typically rely on group education without behaviour change theory, limiting their long-term effectiveness. Consequently, this study aimed to improve the nutritional intake of highly-trained youth soccer players via an individualised, theory-driven dietary behaviour change programme. This mixed method case series utilised the COM-B model and Behaviour change wheel (BCW) to design and implement an eight-week dietary behaviour change programme in three full-time (16.7 ± 0.3 years) academy players (P1; P2; P3). Education and enablement were used to improve players' understanding of their energy and macronutrient requirements, alongside training to develop practical nutrition skills. Environmental restructuring was used to modify player’s physical environment, ensuring they had appropriate food when away from the club. Daily energy intake increased by 55%, 19% and 24% for P1, P2 and P3, respectively, corresponding with DXA-derived fat-free mass gains of 1.4, 0.5 and 0.3kg. Absolute carbohydrate intake increased by 217, 78 and 110g.day-1 respectively, alongside improved fuelling practices around match day. Qualitative insights uncovered improvements in nutritional knowledge and cooking skills (Psychological and Physical Capability), while a desire to improve physical characteristics (Reflective Motivation) was a key adherence factor. The COM-B model and BCW provided a structured framework for designing an effective dietary behaviour change programme that successfully improved dietary behaviours in three highly trained youth soccer players
Preparing graduate entry nursing students to successfully complete research projects:a scoping review
This scoping review investigates how graduate entry nursing (GEN) programs prepare students to successfully complete research projects. For many decades, GEN programs have been offered internationally as an accelerated pathway to nursing registration for graduates. Students who enrol have completed a baccalaureate/bachelor’s degree in a prior discipline, but previous research experience is not a prerequisite. Our review spans GEN programs where English is the language of instruction. With a focus on teaching interventions, the databases included ERIC, Taylor & Francis, Scopus, Wiley Online Library, Sage, CINAHL, Medline and EMcare, with additional citation searching. We followed Arksey and O’Malley’s (2005) framework for scoping reviews. Of the initial 537 studies, five were identified for inclusion. The findings show considerable variation in what aspects of literacy were taught and how literacy teaching was integrated into programs. The reported benefits include improvements to program progression and completion, academic performance, as well as a heightened sense of belonging and positive learning experiences. However, the absence of detail about literacy practices, and reporting on findings only within one course – rather than across a program – poses future difficulties for generating recommendations about program design or refinement. In this regard, specific areas of future research are suggested