110830 research outputs found
Sort by
Sketching Snapshots : Reflecting on Human-Computer Interaction Education through Sketching
No full textReflection provides students and instructors insight into mindsets, learning progress, and course design. However, most reflective activities in HCI courses focus on specific skills and content. We propose sketching as a flexible and expressive medium for reflecting on larger topics to reveal implicit perspectives about multifaceted concepts of HCI. We developed “Sketching Snapshots,” an activity to elicit student perspectives on high-level concepts. To evaluate this activity, we ran pre- and post-semester workshops in two HCI courses in the US and UK. We analyze 68 sketches from 41 students, 9 student interviews, and 2 instructor interviews, and find that sketching is effective for supporting student and instructor reflection. We then report case studies in two additional contexts and synthesize recommendations for deploying reflective sketching. We propose incorporating Sketching Snapshots in courses can facilitate reflection on complex concepts, improve teaching, and spark conversations about what HCI means to us and our students
Age effects in “returnee” bilingualism : Problematizing their conflation as prima facie evidence for linguistic maturation
Get PDFDifferent responses of African rice cultivars to alternate wetting and drying irrigation and anaerobic digestate application
No full textContext Water and nutrient limitations often restrict African rice production. Scarce fertiliser supplies require suitable alternatives such as anaerobic digestate residues, but few trials have tested rice cultivation with this nutrient source. Applying alternate wetting and drying (AWD) irrigation, that decreases crop water use and maintains yield, can enhance climate resilience of irrigation schemes, but whether African rice (Oryza glaberrima) cultivars are better adapted to AWD is uncertain. Objective We aimed to develop farmer recommendations for optimal water and nutrient use in West African rice cropping systems. We hypothesised that anaerobic digestate was a viable alternative nitrogen source to chemical fertilisers, and that a local African rice cultivar tolerated AWD better than an improved Asian cultivar. Methods Factorial experiments at Fumesua, Ghana during the 2021 and 2022 dry seasons varied irrigation type (alternate wetting and drying, AWD versus continuous flooding, CF), nutrient source (synthetic fertiliser versus anaerobic digestate at the same nitrogen application rate) and cultivars (improved O. sativa cv. CRI-Agrarice versus local O. glaberrima cv. Viowornor short) in a split-plot design. Water- (leaf relative water content and stomatal conductance) and nutrient- (chlorophyll content, leaf nitrogen content) related traits were measured during the irrigation season, with yield and its components (number of effective tillers, grain number per panicle, thousand grain weight) measured at harvest. Results AWD decreased water use by 28 % but maintained grain yield, except synthetic fertiliser-treated local O. glaberrima in which grain yield decreased by 31 %. Harvest index better explained treatment differences in yield than variation in water-related physiological traits. Nutrient source affected the cultivars differently. Synthetic fertiliser increased improved O. sativa grain yield by 18 % compared to digestate coincident with higher leaf chlorophyll content, whereas digestate increased local O. glaberrima grain yield by 66 % despite 20 % lower chlorophyll content. Digestate increased the number of effective tillers of both cultivars by 24 %, but diminished thousand grain weight by 4 %. Conclusions AWD requires cultivars adapted to the prevailing nutrient management to avoid yield losses, while applying digestate boosted grain yields of the local cultivar while maintaining comparable yields to the improved cultivar receiving synthetic fertiliser. Significance Bespoke irrigation × nutrient source × cultivar combinations can boost African rice production while decreasing fertiliser costs and water use
When Visual Privacy Protection Meets Multimodal Large Language Models
No full textThe emergence of Multimodal Large Language Models (MLLMs) and the widespread usage of MLLM cloud services such as GPT-4V raised great concerns about privacy leakage in visual data. As these models are typically deployed in cloud services, users are required to submit their images and videos, posing serious privacy risks. However, how to tackle such privacy concerns is an under-explored problem. Thus, in this paper, we aim to conduct a new investigation to protect visual privacy when enjoying the convenience brought by MLLM services. We address the practical case where the MLLM is a “black box”, i.e., we only have access to its input and output without knowing its internal model information. To tackle such a challenging yet demanding problem, we propose a novel framework, in which we carefully design the learning objective with Pareto optimality to seek a better trade-off between visual privacy and MLLM’s performance, and propose critical-history enhanced optimization to effectively optimize the framework with the black-box MLLM. Our experiments show that our method is effective on different benchmarks
John Flowerdew's essential bookshelf : English for research publication purposes (ERPP)
No full textJohn Flowerdew is a Visiting Chair at the University of Lancaster and a Visiting Research Fellow at Birkbeck, University of London. He was previously a Professor at City University of Hong Kong and at the University of Leeds, UK. His research is in the fields of Applied Linguistics and Discourse Analysis. His books include Academic listening: Research perspectives (CUP); Research perspectives on English for academic purposes (with M. Peacock) (CUP); Second language listening: Theory and practice (with L. Miller) (CUP); Academic discourse (Longman); Discourse in English language education (Routledge); Signalling nouns in discourse: A corpus-based discourse approach (with R. W. Forest) (CUP); Discourse in context (Bloomsbury); The Routledge handbook of critical discourse studies (with J. Richardson); Discipline-specific writing: Theory into practice (with T. Costley) (Routledge), and Introducing English for research publication purposes (with P. Habibie (Routledge)
A Systematic Evaluation of the Potential of Carbon-Aware Execution for Scientific Workflows
No full textScientific workflows are widely used to automate scientific data analysis and often involve computationally intensive processing of large datasets on compute clusters. As such, their execution tends to be long-running and resource-intensive, resulting in substantial energy consumption and, depending on the energy mix, carbon emissions. Meanwhile, a wealth of carbon-aware computing methods have been proposed, yet little work has focused specifically on scientific workflows, even though they present a substantial opportunity for carbon-aware computing because they are often significantly delay tolerant, efficiently interruptible, highly scalable and widely heterogeneous. In this study, we first exemplify the problem of carbon emissions associated with running scientific workflows, and then show the potential for carbon-aware workflow execution. For this, we estimate the carbon footprint of seven real-world Nextflow workflows executed on different cluster infrastructures using both average and marginal carbon intensity data. Furthermore, we systematically evaluate the impact of carbon-aware temporal shifting, and the pausing and resuming of the workflow. Moreover, we apply resource scaling to workflows and workflow tasks. Finally, we report the potential reduction in overall carbon emissions, with temporal shifting capable of decreasing emissions by over 80%, and resource scaling capable of decreasing emissions by 67%
SMART-PIPEBOT : A soft pipe-climbing robot with woodpecker tail-inspired support mechanism for single-sidewall adhesion
No full textSoft crawling robots have attracted substantial attention for their environmental adaptability and compliant locomotion. However, the scalability of many soft in-pipe robots to large diameters remains limited because their locomotion typically relies on multi-point contact with the pipe inner wall. To overcome this limitation, we propose SMART-PIPEBOT, a woodpecker-tail-inspired adhesion-driven soft pipeline climbing robot that achieves single-sidewall attachment and locomotion in large-diameter pipes while markedly reducing dependence on distributed wall support. SMART-PIPEBOT integrates an inflatable silicone bladder, vacuum sucker brackets, silicone vacuum suckers, a constraining layer, and an elastic tail-support frame. Experiments quantified how vacuum-sucker inclination (10°–50°) affects adhesion at 0.05 MPa and evaluated single-sidewall climbing performance in large-diameter pipes (170, 200, and 230 mm). The maximum mean adhesion force was 26.778 N at a 30° inclination. SMART-PIPEBOT achieves maximum climbing speeds of 16.93 mm/s horizontally and 12.63 mm/s vertically under dry conditions and maintains continuous locomotion on wet walls with mean speeds of 4 mm/s (23.6% of the dry horizontal) and 2 mm/s (15.7% of the dry vertical). Payload experiments further demonstrate a pronounced improvement in effective load capacity when inspection equipment is mounted at the rear rather than the front, enabling a twofold increase in horizontal payload (100 g vs. 50 g) and a 1.5× increase in vertical payload (30 g vs. 20 g)
PTAR as a Robust Independent Predictor for Malignant Arrhythmia and Liver-Related Decompensation in Cirrhosis : A Retrospective Cohort Study
No full textBackground The objective of this study was to investigate the correlation between prothrombin time-international normalized ratio to albumin ratio (PTAR) and QTc prolongation and to evaluate the predictive value of PTAR for long-term clinical outcomes in patients with liver cirrhosis. Methods Data from participants with liver cirrhosis admitted to Southwest Hospital between April 2017 and April 2018 were included in this study. The association between PTAR and QTc prolongation was examined using univariate and multivariable logistic regression analyses. The restricted cubic spline (RCS) method was employed for curve fitting and to determine the optimal PTAR threshold for predicting QTc prolongation. The SHapley Additive exPlanations (SHAP) algorithm, based on multivariable logistic regression, was utilized to assess the feature importance of PTAR. Kaplan–Meier curves and Cox proportional hazards models were used to evaluate the prognostic value of PTAR for liver-related decompensation and severe arrhythmia. Results Among the 374 participants (mean age 56.7 years; 66.3% male), 41.4% (155/374) experienced QTc prolongation. The median QT interval was 402.9 ± 41.9 ms. An adverse L-shaped curve association between PTAR and QTc prolongation was observed, with an inflection point at 37.40. The odds ratio (OR) for developing QTc prolongation was 1.17 (95% CI: 1.06–1.28, p = 0.001) as PTAR ranged from 0 to 37.40. Interestingly, a threshold of 37.40 was also identified for predicting QTc prolongation, with PTAR achieving an area under the receiver operating characteristic curve (AUC) of 0.66. According to the feature importance ranking based on the SHAP algorithm, PTAR emerged as the most significant predictor, followed by sex, TBIL, and the etiology of liver cirrhosis. During a median follow-up of 502 days, patients in the high PTAR group had a significantly higher risk of liver-related decompensation (HR 3.51, 95% CI 2.35–5.24) and severe arrhythmia (HR 2.34, 95% CI 1.57–3.48) compared to the low PTAR group. Conclusion There is a significant correlation between PTAR and QTc prolongation in patients with liver cirrhosis. An inflection point at 37.40 was identified, at which PTAR showed strong predictive value for QTc prolongation. Furthermore, elevated PTAR serves as a robust independent predictor for long-term liver-related decompensation and severe arrhythmia
How to utilise the limited supply of vaccines for Mpox control in Thailand among high-risk GBMSM
No full textMpox re-emerged globally in 2022, disproportionately affecting gay, bisexual, and other men who have sex with men (GBMSM). In 2024, Thailand became the first Asian country to detect Clade Ib Mpox, prompting urgent decisions on deploying a limited supply of 3000 vaccine doses. However, evidence on the comparative effectiveness of different vaccine allocation and behavioural strategies in this context remains scarce. We developed a deterministic compartmental model of Mpox transmission among high- and low-risk GBMSM, calibrated to Thailand's national surveillance data (January 2023–May 2025). The model simulated a range of hypothetical scenarios under a constrained supply of 3000 vaccine doses, distributed either over a short 5-month period or extended across the 28-month epidemic horizon. We evaluated pre-exposure prophylaxis ( P r E P ), post-exposure prophylaxis ( P E P ), dose-sparing regimens, and mixed allocations of the two approaches. Each strategy was examined under alternative rollout timings (early vs. supply-delayed) and in combination with behaviour change, represented as reductions in sexual activity during symptomatic periods. The model reproduced Thailand's epidemic trajectory. Our simulations suggested that early PrEP rollout would have yielded the greatest reduction in incidence, particularly among high-risk GBMSM. PEP strategies would have had a modest impact overall, though single-dose sparing with delayed rollout (months 5–9) would have been notably effective as the epidemic peak occurred during this period. Mixed PrEP and PEP approaches would have produced intermediate benefits, while behaviour change alone significantly would have lowered transmission. Combining PEP with even modest behavioural changes further enhanced prevention and helped reduce spillover into low-risk groups. Under constrained vaccine supply, dose-sparing and mixed vaccination strategies could improve overall coverage and impact, especially when paired with behavioural changes. Integrating flexible and context-specific vaccination approaches with realistic behavioural modifications offers the best potential for Mpox control in Thailand and similar settings
LGBTQ+ experiences of accessing NHS adult mental health services during COVID-19 in an area of North West England : a qualitative interview study
No full textBackground Lesbian, gay, bisexual, transgender, and queer (LGBTQ+) people have a higher prevalence of mental health conditions than heterosexual and cisgender people, and report poor experiences of accessing mental health services. The COVID-19 pandemic may have had detrimental impacts on socially disadvantaged groups, such as LGBTQ+ people, and their mental healthcare access. This study aimed to understand LGBTQ+ experiences of mental health and accessing mental health services during the COVID-19 pandemic. Methods Topic-guided qualitative interviews were conducted with twelve LGBTQ+ people who accessed or tried to access NHS mental health services in an area of North West England between March 2020 and February 2022. Levesque’s Conceptual Framework for Healthcare Access was used as a framework to code the qualitative data. Reflexive thematic analysis was adopted to analyse the qualitative data from a critical realist perspective. Researchers with lived experience of being LGBTQ+ and accessing mental health services were embedded as part of the research team. Results Eight themes across five domains of the framework were identified from the qualitative data. LGBTQ+ participants shared the negative impacts the pandemic had on their mental health (e.g., worsening existing mental health conditions, experiences of isolation, a loss of social connectedness with the LGBTQ+ community). LGBTQ+ participants experienced significant challenges accessing mental health services during COVID-19, associated with experiences of stigma and discrimination, concerns about disclosing their LGBTQ+ identity, living in unsupportive environments, and being unable to access mental health support remotely. Where positive experiences were identified, these highlighted important opportunities for change. Conclusions This study suggests that LGBTQ+ people may have been adversely impacted by the COVID-19 pandemic, with an increased risk of mental ill health and isolation, and poor experiences of accessing mental health services. Opportunities to maximise protective factors and improve timely access to mental health support are needed to mitigate these effects for LGBTQ+ people in the future. Recommendations to improve LGBTQ+ service inclusivity include implementing self-referral options, being more visibly LGBTQ+ inclusive, and improving staff knowledge and training on supporting LGBTQ+ service users