36750 research outputs found
Sort by
Using logic models in mixed methods research: the example of the Integrating English randomised controlled trial
No full textThis chapter focuses on using logic models as a way of framing mixed methods approaches, which have come to prominence in randomised controlled trials (RCTs). The framework and rationale of logic models motivates a principled eclecticism in our approach to mixed methods research (MMR). It justifies the combination of data collection methods and analyses in measuring and testing the process and outcomes of large-scale research, including RCTs. We summarise an approach to logic models developed in earlier work (Coldwell & Maxwell, 2018) and exemplify its application to a specific RCT, Integrating English, highlighting prompts that can determine an appropriate mix of data collection methods. From these and other applications, we consider issues in the future and challenges involved in applying this approach in MMR. The case is made for MMR exposing the strengths and limitations of logic models and for providing rich, action-oriented data on intervention-based research such as RCTs
Effect of unilateral forearm amputation on fluid torques and body roll in front crawl swimming and the implications for performance
Get PDFPurpose: Rotation of the body about its longitudinal axis or ‘body roll’ in front crawl swimming may reduce injury risk, enhance propulsion and reduce drag. An upper limb amputation may hinder body roll and diminish the benefits associated with this movement. This study examined the external fluid torques (buoyant and hydrodynamic) acting on unilateral upper limb amputee swimmers and their influence on whole-body roll, shoulder roll and hip roll during front crawl.
Methods: Ten Para swimmers with unilateral at-elbow amputation completed front crawl trials at sprinting speed. Three-dimensional motion analysis provided shoulder roll and hip roll angle-time histories. Swimmer’s centre of mass (CM), centre of buoyancy (CB) and whole-body angular momentum (H) were determined relative to the body roll axis. Whole-body roll was calculated by dividing H by the moment of inertia at each time and integrating over the cycle. Buoyant torque was obtained from the cross product of the CM-CB position vector and the buoyant force vector. Net external torque was computed as the time derivative of H and hydrodynamic torque was then found by subtracting buoyant torque from net external torque.
Results: Shoulder roll amplitude, maximum buoyant torque and buoyant torque impulse were greater (p<.01) during recovery (over-water phase) of the non-impaired limb than during recovery of the impaired limb. No significant bilateral differences were found for whole-body roll, hip roll or trunk-twist amplitudes. Mean contributions of buoyant torque and hydrodynamic torque to whole-body roll over the full upper limb cycle were 48% and 52%, respectively.
Conclusions: Swimmers with unilateral forearm amputation experience an asymmetric buoyant torque, requiring them to sacrifice propulsive force to counterbalance the torque asymmetry and maintain symmetric whole-body roll
Robust tumor segmentation in incomplete multi-modal imaging via a synergy of diffusion and Mamba models
Get PDFAutomatic tumor segmentation is a critical task in medical image analysis. Positron emission tomography (PET) and computed tomography (CT) are widely used in early cancer diagnosis because they provide complementary imaging information about anatomical structures. However, obtaining complete images from both modalities in clinical practice is often challenging due to constraints such as cost and physical limitations. Missing modality data can hinder multi-modal understanding and degrade the performance of automatic tumor segmentation models. Existing methods struggle to effectively exploit cross-modal correlations and capture essential semantic information. To address these challenges, we propose DMM-Net, an end-to-end Diffusion Mamba Multi-modal Network for incomplete multi-modal automatic tumor segmentation. DMM-Net optimizes performance by combining generation and segmentation in a synergistic manner. Specifically, DMM-Net consists of two key components:conditional missing modality generation and cross-modal spatial-channel interaction segmentation. In the first component, a score-based diffusion model generates the missing modality, leveraging the available modalities as conditional guidance to reduce semantic ambiguities. In the second component, we propose a cross-modal interaction Mamba and a cross-modal channel attention enhancement module. The interaction Mamba efficiently captures global contextual features from PET and CT images, facilitating cross-modal feature interaction. The channel enhancement module highlights essential channel features within the multi-modal representations, captures inter-modal feature correlations, and suppresses redundant information. Next, we validate the effectiveness of DMM-Net through comprehensive ablation studies and visualization experiments on three multi-modal datasets (STS, LMNCLC, and Hecktor 2022). Quantitative comparisons and segmentation results demonstrate that DMM-Net surpasses state-of-the-art methods, regardless of modality incompleteness
Land-use change undermines the stability of avian functional diversity
Get PDFLand-use change causes widespread shifts in the composition and functional diversity of species assemblages. However, its impact on ecosystem resilience remains uncertain. The stability of ecosystem functioning may increase after land-use change because the most sensitive species are removed, which leaves more resilient survivors1,2,3. Alternatively, ecosystems may be destabilized if land-use change reduces functional redundancy, which accentuates the ecological impacts of further species loss4,5. Current evidence is inconclusive, partly because trait data have not been available to quantify functional stability at sufficient scale. Here we use morphological measurements of 3,696 bird species to estimate shifts in functional redundancy after recent anthropogenic land-use change at 1,281 sites worldwide. We then use extinction simulations to assess the sensitivity of these altered assemblages to future species loss. Although the proportion of disturbance-tolerant species increases after land-use change, we show that this does not increase stability because functional redundancy is reduced. This decline in redundancy destabilizes ecosystem function because relatively few additional extinctions lead to accelerated losses of functional diversity, particularly in trophic groups that deliver important ecological services such as seed dispersal and insect predation. Our analyses indicate that land-use change may have major undetected impacts on the resilience of key ecological functions, hindering the capacity of natural ecosystems to absorb further reductions in functionality caused by ongoing perturbations
Be(a)ware of the dog: the symbolic-disciplinary functions of drug detection dogs in English music festivals
Get PDFThis paper examines the institutionalisation of Drug Detection Dogs (DDDs) at British music festivals, highlighting a paradox at the heart of contemporary festival security practices. While festivals have increasingly embraced harm reduction strategies in response to drug-related deaths, they have simultaneously intensified law enforcement measures at entry points. DDDs are widely deployed despite research demonstrating their ineffectiveness in deterring drug use and their potential to increase drug-related harm. Drawing on ethnographic fieldwork at eight festivals and eleven interviews with police, security personnel, and event managers, this study investigates how DDDs persist as a security measure despite their questionable utility. Using a novel theoretical framework that combines Foucauldian disciplinary power with symbolic policy theory, the paper conceptualises DDDs through a symbolic-disciplinary nexus. It argues that DDDs serve dual functions: at the micro-level, they shape attendee behaviour through visible displays of authority and moral messaging; at the macro-organisational level, they act as symbolic performances of security competence aimed at satisfying external stakeholders such as police and licensing authorities. This dual role contributes to a self-reinforcing security ratchet, where the perceived legitimacy and institutional value of DDDs outweigh their practical effectiveness. The paper concludes by discussing the broader implications of symbolic-disciplinary security practices and offers policy recommendations for moving beyond performative control measures. It advocates for a shift toward evidence-based harm reduction approaches that prioritise public health and civil liberties over symbolic displays of control
Solvent-Assisted Modification of Laser-Induced Graphene for Surface-Enhanced Electrochemical Response
Get PDFThe fabrication and application of laser-induced graphene (LIG) have received significant interest across various research fields, particularly in sensing technologies. Herein, we investigated the surface modification of LIG electrodes with different solvents and demonstrated that dimethyl sulfoxide (DMSO) enhances their electrochemical activity. Importantly, our findings reveal that solvent treatments themselves can induce significant modifications on the electrode surface (e.g., changes in functional groups, wettability, and morphology), which must be carefully considered in sensor design and optimization. The application of a microliter aliquot of DMSO to the LIG surface significantly altered its wettability, promoting a transition to hydrophilic behavior, which was verified by contact-angle measurements. AFM and XRD results indicate that DMSO treatment promotes the rearrangement of disordered carbon atoms, minimizing localized structural defects and enabling more efficient π–π stacking between graphene sheets. The cyclic voltammetric response of the [Fe(CN)6]3-/4- redox probe showed a 3-fold increase in peak current. The enhanced electrochemical effect of DMSO surface changes on LIG electrodes was also confirmed in the presence of several organic species, and a substantial current increase was verified by the effect of DMSO. As proof-of-concept, the antibiotic sulfanilamide was detected in synthetic urine and water samples using differential-pulse voltammetry on DMSO-treated LIG. Under optimized conditions, the sensor exhibited a linear response in the range of 0.3–9.0 μmol L–1 with a limit of detection of 0.09 μmol L–1 and satisfactory recovery values (90–104%)
Learning from Annotator Disagreement via Weighted Ensemble Optimisation for Subjective Text Classification
Get PDFSubjective text classification tasks, such as abuse detection and stance analysis, often suffer from high levels
of annotator disagreement. Conventional approaches typically collapse these disagreements into a single ground
truth, thereby discarding valuable supervision signals. We propose MO-WEL (Multi-Objective Weighted Ensemble Learning), a novel framework that explicitly leverages annotator disagreement by jointly optimising ensemble
weights and size under multiple objectives. Candidate predictors are trained on diverse label projections obtained
through random sampling or annotator-specific selection, and ensemble weights are optimised with respect to
three complementary losses: F1 score, cross-entropy and Manhattan distance, alongside a regularisation term.
Experiments on four benchmark datasets (ConvAbuse, HS-Brexit, MD-Agreement and ArMIS) show that MOWEL consistently outperforms strong baselines in accuracy, calibration, and distributional alignment. A case
study further demonstrates that MO-WEL produces predictions that balance majority correctness with minority
annotator perspectives, yielding interpretable and reliable outputs. Our findings highlight the importance of modelling annotator diversity and suggest ensemble optimisation as a principled means of incorporating disagreement
into subjective NLP tasks
A retrospective service evaluation into non-compliance in patients with mild obstructive sleep apnoea being treated on continuous positive airway pressure (CPAP)
No full textIntroduction: Obstructive sleep apnoea is a sleep breathing disorder. It is characterised by episodes of repetitive airway closure which can either be a partial or full closure of the airway. With the repetitive collapse and opening of the airway, this leads onto oxygen desaturations, excessive daytime sleepiness, snoring and nocturia. The primary treatment according to NICE guidelines for OSA is CPAP therapy. This generates airflow to provide a stream of pressurized room air to splint open the upper airway during sleep via a mask. There are two main factors which measure adherence. AHI must be 4 h a night for at least 70%–80% of the night's. With the change in guidance for mild patients being offered CPAP, the audit will look at the following: Is low compliance seen in those who have a low Epworth sleepiness score and Body mass Index. Is there is a link between adherence data to pre-dict future adherence? If noncompliance is high overall, can the service offer an alternative treatment pathway for Mild OSA patients?
Materials and methods: Data was collected retrospectively from a secure database used by the lung function and sleep department. Patients sleep studies were collected from September 2021 to October 2023. A sample size of 385 was estimated in order to generalise the results to a 95%confidence interval for the Welsh population with OSA. All patient identifiable information was removed. Not all patients were suitable due to the inclusion and exclusion criteria.
Results: Multiple linear regression was used. Initial consultation compliance data showed there was no significant relationship between the ODI and Average Hours (p = 0.499), mask fit (p = 0.843) and%>4 h (p = 0.410). There was no significant relationship between the ODI and Average hours (p = 0.372), mask fit (p = 0.520) and% >4 h(p = 0.65) data after 3 months of treatment.
Conclusions: There is no correlation between in compliance in mild patients from their initial consultation and at 6 months and there was no significant correlation between ESS and BMI against adherence data. However, this does give room to explore and develop the ser-vice with regards to making alternative pathways and referral in non-adhering patients
Enhancing Construction Workers' Safety Behavior in Small- and Medium-Sized Enterprises: Self-Determination Theory Approach
Get PDFLeader autonomy support has been widely recognized for its influence on motivation and subsequent behavior. However, few studies have empirically explored the mediating role of relative self-determined safety motivation in the relationship between leader autonomy support and construction workers' safety behavior, particularly within small- and medium-sized enterprises in the construction industry. To address this gap, the present study applies the self-determination theory model to examine the impact of leader autonomy support on construction workers' safety behavior, with a focus on the motivational processes underlying the relationship between leader autonomy support and construction workers' safety behavior. Data were gathered using a cross-sectional questionnaire from a convenience sample of 283 employees employed by small- and medium-sized enterprises in construction from Anhui, China. The partial least squares structural equation modeling technique was used to analyze the data. The results indicated that leader autonomy support significantly enhances relative self-determined safety motivation, which, in turn, predicts safety compliance and participation. Relative self-determined safety motivation fully mediates the relationship between leader autonomy support and construction workers' safety behavior. Furthermore, the findings have shown that the influence of leader autonomy support on safety participation is greater than that of safety compliance. This study underscores the pivotal role of leader autonomy support in fostering relative self-determined safety motivation, which serves as a key mechanism for improving construction workers' safety behavior. It provides new insights into the motivational mechanisms underlying construction workers' safety behavior. Findings provide practitioners with a guideline to conduct on-site safety supervision and improve the efficiency of leadership in safety management with regard to small- and medium-sized enterprises in construction
Co‐Creative Sustainability: Enacting Ethical Power
Get PDFEntrepreneurship and broader business literature show a growing interest in sustainability issues, illuminating the critical role that entrepreneurship can play in addressing the issues facing economic development in diverse socioeconomic settings. This study undertakes a selective systematic literature review and synthesizes co‐creative entrepreneurship research with Foucault's conceptualizations of ethical power and human relationships to deepen our understanding of the roots of sustainability issues. It offers insights into addressing sustainability challenges in diverse social settings and provides reflections and directions for future research. As such, this study provides a philosophical ground for the further development of sustainable entrepreneurship