159370 research outputs found
Sort by
MO-SHW: hierarchy-aware multi-objective optimization for open-world segmentation
Get PDFThe exploitation of hierarchical information by vision models has shown significant benefits in various segmentation tasks. However, this remains largely unexplored in open-world scenarios, where models must cope with unknown, evolving, and underrepresented labeled class spaces. Most existing hierarchy-aware segmentation approaches are not readily applicable to open-world settings. This is primarily because they rely on architectural modifications that are incompatible with the design constraints of open-world models. Moreover, hierarchy-aware losses are challenging to integrate into such pipelines, as they often conflict with task-specific objectives and exacerbate optimization complexity in already multi-objective training environments. In this work, we demonstrate that hierarchy-aware losses can be effectively leveraged in open-world models when optimized under a multi-objective learning framework. Specifically, we show that gradient-based multi-objective optimization methods, such as multi-objective gradient descent (MOGD), are well-suited for jointly optimizing hierarchical and task-specific objectives, leading to better overall performance. To support this, we propose SHW, a novel hierarchy-aware loss function based on the Wasserstein distance. SHW is lightweight, model-agnostic, and encourages intra-class compactness and inter-class separation across multiple semantic levels. The integration of SHW with MOGD yields a general, model-agnostic framework that enables the effective exploitation of semantic hierarchies in open-world segmentation tasks, improving the performance of several recent methods
Consensus Gentium on tour: Lessons for touring immersive work beyond arts venues
Get PDFThis report details the findings and offers a
toolkit based on the 2024 tour of Consensus
Gentium, an interactive film by artist Karen Palmer,
to 11 shopping centres across England.
Consensus Gentium, which uses facial detection
and AI to explore themes of state surveillance, was
presented on mobile phones via a specially designed
pop-up installation. The tour, supported by Arts Council
England and delivered by Crossover Labs, aimed to
demonstrate how exhibiting immersive work outside
of arts spaces might attract younger and more diverse
audiences, specifically those aged 14 to 24 from global
majority and working-class backgrounds who are less
likely to engage with conventional arts venues.
The Consensus Gentium tour demonstrates a viable
model for extending the reach of immersive art to
new and diverse audiences by situating experiences
in more accessible environments. This toolkit
underscores the need for thoughtful adaptation
of both the artistic experience and the exhibition
model to suit these unconventional settings
and the specific demographics they attract
Mapping key issues and useful theory to address remaining ethical, practical and political challenges in participatory research in health
Get PDFParticipation is recognised as having a key role in health, for increasing the relevance and effectiveness of health interventions, for the health promoting benefits of community empowerment and as an ethical imperative. Participatory approaches to health research are also increasingly valued for bringing the insights of lived experience, and more relevant research and action. In this paper, we explore key remaining issues in participatory action research highlighted by scholars, practitioners and published literature, and highlight some useful conceptual resources which help to better understand them. We distinguish participatory action research as a paradigm involving those most affected throughout the research process, contrasting it with the more limited use of participatory tools and methods. We outline several aspects of participatory action research in health that would benefit from further theoretical and practical development, including: shifting power in the research process; the compatibility of participatory research with biomedical research; linking local inquiry and action to broader changes in policy and practice; and working with experiential knowledge in a rigorous research process. We highlight useful theory from a range of disciplines (including beyond the participatory research literature) that helps to understand some of the key processes and dynamics implicated in the issues highlighted and how this affects the outcomes achieved. We outline and share these conceptual/theoretical resources, identified as part of preparation for conducting a realist review on participatory action research in health, to contribute to ongoing reflection and development in the field
Student talk as a source for incidental vocabulary learning: what can corpora tell us?
Get PDFWhile experimental studies have confirmed the value of student talk for incidental L2 vocabulary learning, no studies have evaluated it from the corpus linguistics perspective. Taking the Vietnamese EFL university context as a case, this study used corpus linguistics to investigate the value of student talk for incidental vocabulary learning. Three corpora were created: a cross-sectional corpus of 381 student-student interactions in tertiary English language courses, and two longitudinal corpora derived from all lessons in each course. The analyses of words, represented by Nation’s (2012) BNC/COCA lists, and core formulaic sequences, represented by Martínez and Schmitt’s (2012) Phrasal Expressions List, were respectively undertaken with RANGE and Sketch Engine. Results show that knowledge of the most frequent 1000-word families is needed for reasonable comprehension of student talk. 5.23%−23% of the high-frequency word families and 0.59%−13.47% of the core formulaic sequences were encountered multiple times during each course. The results suggest student talk is an excellent source for learning and reinforcing high-frequency vocabulary and a good source for learning and reinforcing core formulaic sequences. Our study highlights the value of student talk as an input source for low-proficiency learners to learn vocabulary incidentally and showcases how corpus linguistics can inform L2 pedagogy
Structural lateral dynamic stability requirements due to pedestrian loading
No full textThe inverted pendulum pedestrian model (IPM) for walking on laterally-oscillating structures, originally proposed by Macdonald [1], has been recently calibrated using data from pedestrians walking on a laterally-oscillating instrumented treadmill and generalised for predictive use in Czaplewski et al. [2]. The former task was accomplished by defining an empirically-derived foot placement control law. The latter task was accomplished by relating the parameters of this law to the basic anthropometric and gait characteristics of the pedestrian. Closed-form solutions for the long-term average lateral forces obtained from the generalised IPM were then derived in Czaplewski & Bocian [3] based on the framework introduced by McRobie [4]. These solutions were used to obtain the probabilistic lateral dynamic (in)stability criteria for structures subjected to pedestrian loading presented in this paper. A framework introduced in Bocian et al. [5] is used in which stability requirements are expressed in terms of the pedestrian Scruton number and the critical number of pedestrians. To achieve this goal it was necessary to propose a framework for defining a statistical model of the anthropometric parameters used within the IPM solutions, relevant for a given population of pedestrians. It was also necessary to define IPM validity criteria enabling spurious solutions to be omitted from the analysis. To make the proposed structural stability criteria applicable in engineering practice, a framework had to be defined enabling simplified envelopes of the self-excited forces to be obtained. Crucially, these simplified envelopes should not penalise structural solutions, rendering them overly conservative. The proposed stability criteria are evaluated based on two case studies of bridges prone to pedestrian-induced lateral dynamic instability. The relatively recent occurrence of instability on the Squibb Park Bridge and its consequences are presented here in detail, as this case is currently little known in the structural engineering community
Cellular responses of Cupriavidus necator H16, a bioplastic producer, to organic acids: a comparative analysis
Get PDFMy husband is my responsibility:” motivations and activities of informal caregiving for patients with chronic diseases in Uganda
Get PDFBackground
Informal care is a vital part of care for many people with chronic diseases in sub-Saharan Africa. Though challenging and demanding, many individuals are involved in the provision of this care. However, it is unclear what motivates people in sub-Saharan Africa to provide informal care. Therefore, this study explored the motivations to becoming informal caregivers of patients with chronic diseases in Uganda.
Methods
Focus group discussions (FGDs) were conducted among 33 informal caregivers of patients with chronic diseases at Mbarara Regional Referral Hospital in Uganda. The participants were purposively sampled into one of five FGDs. Discussions were audio-recorded, transcribed verbatim and manually analysed using thematic analysis.
Results
The informal caregivers were aged 18 to 67, with 72.7% being females. They were all related to their respective care recipients. Six themes describing the motivations for becoming a caregiver emerged including reciprocity, love, caring as a responsibility, availability to care, desire to improve the patient’s health and desire to protect others from burden of care. Also, themes for caregiver activities were: approach to caregiving, activities of daily living, activities that require medical training, activities that motivate the patient and activities that support other informal caregivers.
Conclusion
The results indicate that multiple factors motivate people to become informal caregivers and that their activities benefit both the patients and other caregivers. These findings highlight the need for policymakers to design a framework that would enable health workers to collaboratively work with informal caregivers to achieve better experiences and outcomes for both patients and caregivers
Developing prediction models for electrolyte abnormalities in patients indicated for antihypertensive therapy: evidence-based treatment and monitoring recommendations
Get PDFObjectives:
Evidence from clinical trials suggests that antihypertensive treatment is associated with an increased risk of common electrolyte abnormalities. We aimed to develop and validate two clinical prediction models to estimate the risk of hyperkalaemia and hyponatraemia, respectively, to facilitate targeted treatment and monitoring strategies for individuals indicated for antihypertensive therapy.
Design and methods:
Participants aged at least 40 years, registered to an English primary care practice within the Clinical Practice Research Datalink (CPRD), with a systolic blood pressure reading between 130 and 179 mmHg were included the study. The primary outcomes were first hyperkalaemia or hyponatraemia event recorded in primary or secondary care. Model development used a Fine-Gray approach with death from other causes as competing event. Model performance was assessed using C-statistic, D-statistic, and Observed/Expected (O/E) ratio upon external validation.
Results:
The development cohort included 1 773 224 patients (mean age 59 years, median follow-up 6 years). The hyperkalaemia model contained 23 predictors and the hyponatraemia model contained 29 predictors, with all antihypertensive medications associated with the outcomes. Upon external validation in a cohort of 3 805 366 patients, both models calibrated well (O/E ratio: hyperkalaemia 1.16, 95% CI 1.13–1.19; hyponatraemia 1.00, 95% CI 0.98–1.02) and showed good discrimination at 10 years (C-statistic: 0.69, 95% CI 0.69–0.69; 0.80, 95% CI 0.80–0.80, respectively).
Conclusion:
Current clinical guidelines recommend monitoring serum electrolytes after initiating antihypertensive treatment. These clinical prediction models predicted individuals’ risk of electrolyte abnormalities associated with antihypertensive treatment and could be used to target closer monitoring for individuals at a higher risk, where resources are limited
Reorienting frailty in clinical practice, public health, and policy: the Lancet Commission on Frailty
No full textFeature similarity, a sensitive method to capture the functional interaction of brain regions and networks to support flexible behavior
Get PDFThe brain is a dynamic system where complex behaviours emerge from interactions across distributed regions. Accurately linking brain function to cognition requires methods sensitive to these interactions. We introduce Feature Similarity (FS), which integrates a broad set of interpretable time-series features—such as covariance, temporal dependencies, and entropy —to move beyond traditional single-metric approaches. FS captured functional brain organization: regions within the same network showed greater similarity than those in different networks, and FS identified the principal gradient from unimodal to transmodal cortices. Compared with Pearson correlation-based functional connectivity (FC) and 46 out of 49 statistical pairwise interaction metrics (SPIs), FS demonstrated greater sensitivity to task modulation. Critically, FS revealed a task-dependent double dissociation in the Dorsal Attention Network, interacting more strongly with the Visual network during working memory but with the default mode network during long-term memory. FS thus provides a powerful tool for uncovering task-specific brain network interactions
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! 👇